Skip to main content
NCBI home page
Royal Society of Chemistry logoLink to Royal Society of Chemistry
. 2026 Sep 25. Online ahead of print. doi: 10.1039/d6tc02199a

Superacid treatment of dielectric-encapsulated monolayer molybdenum disulfide for enhanced photoluminescence

Brendan F M Healy a,✉, Kelsey J Hall a, Sophie L Pain b, Nicholas E Grant a, John D Murphy b
PMCID: PMC13613486  PMID: 42799034

Abstract

With a direct bandgap and strong photoluminescence (PL) emission, monolayer molybdenum disulfide (1L MoS2) is a promising material for next-generation optoelectronic technologies. However, the intrinsically low PL quantum yield (PLQY) and the requirement for dielectric integration present significant challenges in the realisation of practical MoS2-based devices. Chemical treatment with the superacid bis(trifluoromethanesulfonyl)amide (TFSA) is known to enhance dramatically the PL emission of 1L MoS2 through defect passivation and p-type doping. Conversely, atomic layer deposition (ALD) of a dielectric material can introduce defect states and induce n-type doping, thus degrading the MoS2 optical performance. Despite the competing effects of these two processing routes, the combined impact of superacid treatment and ALD dielectric growth on 1L MoS2 has not been thoroughly studied. In this work, we reveal the instability of the TFSA-enhanced MoS2 PL intensity during subsequent ALD of aluminium oxide (Al2O3), due to thermal exposure and precursor interaction. In contrast, we demonstrate that TFSA treatment of thermal Al2O3-encapsulated 1L MoS2 can successfully enhance the PL intensity by up to two orders of magnitude, even for dielectric films of up to ∼96% surface coverage. This enhancement is suppressed only when the Al2O3 layer becomes sufficiently thick (>∼40 nm) and approaches complete coalescence, preventing access of TFSA-derived species to the underlying MoS2. This work establishes an effective encapsulation/passivation strategy for enhancing the optical performance of dielectric-covered 1L MoS2, offering a practical pathway for integrating chemical passivation into device-relevant architectures.

Post-deposition superacid treatment restores and strongly enhances the photoluminescence of dielectric-encapsulated monolayer MoS2, enabling effective chemical passivation in device-relevant structures.graphic file with name d6tc02199a-ga.webp

1. Introduction

Exhibiting novel optical and electrical properties at single-layer thickness, semiconducting transition metal dichalcogenides (TMDCs) have attracted significant research interest for integration in future optoelectronic devices. Among them, monolayer molybdenum disulfide (1L MoS2) is particularly promising due to its direct bandgap of ∼1.8 eV and resulting photoluminescence (PL) emission,1,2 enabling applications in light-emitting devices,3,4 photovoltaics,5–8 photodetectors,9–11 and phototransistors.12–14 Scalable device production requires large-area TMDC growth, and 1L MoS2 is routinely synthesised via chemical vapour deposition (CVD). Compared to mechanically exfoliated monolayers, CVD-grown 1L MoS2 possesses a high defect density,15,16 predominantly sulphur (S) vacancies.17 Such defects trap photoexcited carriers and facilitate non-radiative recombination, with the PL quantum yield (PLQY) from 1L MoS2 reported to be as poor as <1%.1,18 While modifications to the growth conditions can offer modest improvements in the PL emission efficiency from CVD-grown 1L MoS2,19 substantial enhancement of the PLQY is required for efficient, high-performing 1L MoS2-based optoelectronic devices.20,21

Treatment with the organic superacid bis(trifluoromethanesulfonyl)amide (TFSA) is widely recognised as an effective chemical passivation strategy for a range of electronic materials,22 including monolayer TMDCs.17,23–25 Immersion in TFSA solution has been shown to yield ≥95% PLQY in 1L MoS2,26 enhancing the PL intensity by up to two orders of magnitude.17,23–25 We recently revealed the enhanced PL behaviour exhibits long-term stability under ambient conditions.27 TFSA can also significantly improve MoS2-based device performance, with a five-fold increase in electron mobility reported after superacid treatment.28

Many practical optoelectronic devices, particularly those based on the field-effect transistor (FET), require integration of a gate dielectric.29 High dielectric constant (high-κ) materials such as hafnium oxide (HfO2), aluminium oxide (Al2O3), and zinc oxide (ZnO) are commonly grown via atomic layer deposition (ALD). Traditional thermal ALD of high-κ dielectrics on monolayer TMDCs is notoriously challenging. The chemically inert basal plane of 1L MoS2 lacks the dangling bonds required for initial precursor chemisorption.30,31 ALD of dielectrics directly on 1L MoS2 is therefore dominated by precursor physisorption and proceeds via an island growth mode, yielding incomplete layers even at high cycle numbers.31,32 Ultraviolet-ozone (UV-O3) pre-treatment,33–35 plasma-enhanced ALD (PEALD),36,37 and low-temperature ALD30,38 can improve nucleation but may compromise MoS2 material quality and degrade the dielectric/MoS2 interface. Reliable ALD integration of high-κ dielectrics remains an important challenge for realisation of practical 1L MoS2-based optoelectronic devices.

Harnessing the enhanced optoelectronic performance of superacid-treated 1L MoS2 in a practical device architecture may necessitate combining dielectric deposition with TFSA exposure. We previously revealed that superacid treatment and ALD-dielectrics separately induce opposing changes to the CVD-1L MoS2 PL emission, arising from differing charge doping effects.23 TFSA treatment enhances, blueshifts, and sharpens the MoS2 PL emission, consistent with depletion of the electron density – a p-type doping effect. In contrast, thermal ALD of Al2O3 or HfO2 reduces, redshifts, and broadens the PL signal, indicating n-type doping. Therefore, it is important to ascertain the overall effect on the optical behaviour of 1L MoS2 following both ALD-dielectric growth and superacid treatment.

The processing order must be considered when combining superacid and ALD processing of 1L MoS2. ALD of a dielectric on TFSA-enhanced 1L MoS2 may degrade the improved PL behaviour via vacuum exposure,39 elevated temperature, and/or reactive precursors. Alharbi et al. demonstrated ALD of HfO2 on TFSA-treated MoS2 in the fabrication of top-gated FETs,40 citing previous indications that TFSA treatment exhibits thermal stability up to 300 °C. While improved electrical performance was observed, no optical characterisation was reported to verify whether the TFSA-enhanced PL emission survived the ALD process. As such, it remains crucial to determine if the elevated PL character of TFSA-treated 1L MoS2 can survive the processing conditions of subsequent dielectric growth. Reversal of the processing sequence – treating dielectric-capped 1L MoS2 with TFSA – may offer a pathway to enhanced 1L MoS2 optical performance in a device-relevant architecture. While TFSA treatment of uncovered and polymer-capped 1L MoS2 is well-described in the literature,25,41,42 the impact on dielectric-encapsulated 1L MoS2 remains poorly established. Comprehensively evaluating how dielectric morphology, thickness, and coverage influence the optical characteristics of 1L MoS2 after subsequent TFSA treatment is key for translating chemical passivation strategies into practical devices.

In this work, we investigate how the processing order of superacid treatment and dielectric integration impacts the optical properties of 1L MoS2, with the two treatment routes depicted in Fig. 1. We demonstrate the instability of TFSA-enhanced 1L MoS2 PL emission to ALD of a dielectric overlayer. We then reverse the processing order, applying TFSA treatment to dielectric-covered 1L MoS2 to assess whether this encapsulation/passivation strategy can enhance the optical properties of MoS2. Using Al2O3 as a representative ALD-grown high-κ dielectric, we vary the number of ALD cycles prior to TFSA exposure to evaluate how the evolving dielectric morphology influences the overall 1L MoS2 PL character. We demonstrate TFSA treatment is sufficient to overcome the modified PL emission following ALD-Al2O3 growth, with substantial PL enhancement observed even at 96% dielectric coverage. This study provides the first systematic assessment of the overall impact of combined ALD-dielectric growth and superacid treatment on the optical properties of 1L MoS2.

Fig. 1. Simplified schematic illustration of the two different processing sequences investigated in this work: (a) TFSA treatment of 1L MoS2 prior to thermal ALD of Al2O3 and (b) thermal ALD of Al2O3 on 1L MoS2 followed by subsequent TFSA treatment.

Fig. 1

Open in a new tab

2. Methods and materials

Materials

A 1L MoS2 film (>99%) grown on a SiO2/Si substrate via atomic pressure chemical vapour deposition (APCVD) was obtained from 2D Semiconductors. The 1 × 1 cm2 film was received in vacuum-sealed packaging, before being cleaved into nine smaller pieces of approximately equal size, with the cleaved samples stored in a vacuum desiccator prior to treatment.

Chemical treatment

TFSA (≥ 95.0%) and pentane (> 99%) were acquired from Sigma-Aldrich. A solution of TFSA in anhydrous pentane was prepared at a concentration of 2 mg ml−1. Pentane was selected as the solvent as it is known to enable substantial TFSA-induced PL enhancement in TMDCs,27,43 while avoiding the long-term safety concerns associated with more toxic and potentially carcinogenic alternatives, such as dichloromethane (DCM).44,45 TFSA was weighed out in a high-specification filtered and sealed MBRAUN UNIlab modular glovebox with gas purification system, controlled low-humidity atmosphere (< 0.1 ppm of oxygen O2 and water (H2O)), and solvent system. The TFSA–pentane solution preparation was performed in a lower specification glovebox with ambient flowing nitrogen (N2, relative humidity <25%). Each 1L MoS2 sample was immersed separately in a TFSA–pentane solution for 10 min at room temperature before being removed and allowed to dry in ambient air for 24 h prior to measurement or further processing, ensuring evaporation of the pentane solvent and effective interaction between the TFSA molecules and the MoS2.25,27

Atomic layer deposition of Al2O3

To prevent contamination of the ALD system by superacid and solvent species, Al2O3 growth on TFSA-treated 1L MoS2 was performed in a separate ALD system to that used for deposition on untreated samples for subsequent TFSA processing. Thermal ALD of Al2O3 on superacid-treated 1L MoS2 was conducted at 150 °C in an Anric Technologies AT400 system. Trimethylaluminium (TMA) and H2O served as the precursor and co-reactant, respectively, with the delivery lines maintained at 100 °C. Nitrogen (N2) was used as both the carrier and purging gas. Table 1 summarises the reaction conditions used in this ALD process. The vacuum and thermal effects of ALD processing were also investigated using this system.

Table 1. Precursor and co-reactant conditions used in ALD of Al2O3 films on TFSA-treated and untreated 1L MoS2.

Condition TFSA/ALD ALD/TFSA
Deposition temperature (°C) 150 200
Precursor TMA TMA
Number of precursor pulses in half-cycle 3 1
Precursor pulse duration (s) 0.55 0.06
Precursor purge duration (s) 13 8
Co-reactant H2O H2O
Number of co-reactant pulses in half-cycle 2 1
Co-reactant pulse duration (s) 0.55 0.06
Co-reactant purge duration (s) 15 8
Open in a new tab

The Al2O3 films deposited prior to TFSA treatment were grown via thermal ALD at 200 °C in a Veeco Fiji G2 system. Here, TMA and H2O were again used as the precursor and co-reactant, respectively, while argon (Ar) was employed as the carrier and purging gas. The precursor delivery lines were heated to 150 °C. The pulse and purge times associated with this Al2O3 growth process are outlined in Table 1.

Atomic force microscopy

A Bruker Dimension Icon in the PeakForce Tapping Mode with a ScanAsystAir tip (with a nominal tip length of 115 µm, a tip radius of 2 nm and, a spring constant of 0.4 N m−1) was used to acquire atomic force microscopy (AFM) images of surface topographies.46 All AFM images comprised 256 lines per scan and were obtained at a scan rate of 0.5 Hz. Image processing and analysis was performed in the Gwyddion 2.68 software.47 The Fiji distribution of the ImageJ software package was used to estimate the surface coverages of the dielectric films by converting the AFM images to 8-bit grayscale and then binary images via the software's thresholding algorithm.48

Raman and PL spectroscopy

All Raman and PL measurements were acquired at room temperature using a Renishaw inVia Reflex Raman microscope in standard confocal mode with a 50× Leica objective (numerical aperture of 0.75). A 532 nm excitation laser operated at 0.1% of maximum power (18 µW) and an 1800 lines per mm grating was used for all measurements. Single-site Raman spectra were collected as the sum of four 5 s accumulations and fitted with a two-peak Lorentzian function to capture the characteristic MoS2 E12g and A1g Raman features. Corresponding single-point PL data were acquired as the sum of four accumulations of 10 s each and deconvoluted via multipeak Lorentzian fitting, with the exciton and trion peak energies constrained to approximate initial estimates. Raman and PL maps were collected over an area of 12 × 12 µm2, with a step size of 0.5 µm. A superposition of two Lorentzian peaks was used to fit the Raman spectrum at each point in the mapping data, while each mapped PL spectrum was fitted with a single Lorentzian function and the maximum absolute intensity, corresponding peak energy, and spectral linewidth extracted. Identical acquisition conditions were maintained before and after each treatment, with measurements performed at the same location on each sample. For single-spot Raman and PL measurements, the uncertainty in the location of the measurement site was typically smaller than the laser spot size (∼2 µm). The Renishaw WiRE 3.1 software package was used to collect all data and any cosmic-ray artefacts removed from the spectra where necessary.

3. Results and discussion

Characterisation of untreated CVD-1L MoS2

We first characterise the CVD-1L MoS2 film prior to any external treatment. We employ AFM imaging to verify the monolayer nature of MoS2. Fig. 2a presents a relatively low-magnification AFM image of a region of the MoS2 film containing a thin void that reveals the underlying SiO2/Si substrate. A higher-magnification AFM image of the boundary between the MoS2 and SiO2 regions is displayed in Fig. 2b. As shown in Fig. 2c, we estimate the step height between the MoS2 and adjacent SiO2/Si substrate to be 0.67 nm, consistent with the expected thickness of CVD-grown 1L MoS2.49,50 We utilise spatially resolved Raman spectroscopy to further confirm the presence of 1L MoS2. The Raman spectrum of MoS2 comprises two distinct, characteristic peaks: an in-plane E12g vibration at ∼384 cm−1 and out-of-plane A1g feature at ∼403 cm−1.51 The separation between these two modes is related to the number of MoS2 layers, with a value of ∼18–21 cm−1 typically indicative of a single layer. Fig. 2d presents a map of the Raman peak separation obtained from the region of the MoS2 film shown in Fig. 2a. We find a homogenous distribution of the Raman peak separation (mean ∼19.6 ± 0.2 cm−1) across the surface, confirming widespread 1L MoS2.

Fig. 2. Representative characterisation of untreated CVD-1L MoS2. (a) Low- and (b) high-magnification topographic AFM images of a region of a CVD-1L MoS2 film on SiO2/Si. (c) Height profile measured along the red arrow in (b). (d) Map of the Raman peak separation measured from the region of untreated 1L MoS2 shown in (a). The black area represents the SiO2/Si substrate where no MoS2 Raman peaks are detected.

Fig. 2

Open in a new tab

ALD-dielectric growth on TFSA-treated 1L MoS2

We next evaluate the stability of the enhanced PL character of TFSA-treated CVD-1L MoS2 during ALD of a dielectric. Mapping and single-site PL data measured from 1L MoS2 before and after TFSA treatment and following subsequent ALD of Al2O3 are presented in Fig. 3. A scatter plot of PL intensity as a function of PL energy extracted from the mapping data is displayed in Fig. 3b (full spatial maps of the PL energy and full width at half maximum (FWHM) are provided in Fig. S1 in the SI, with a magnified view of the low-intensity region of Fig. 3b displayed in Fig. S2).

Fig. 3. Effect of thermal ALD of Al2O3 on PL emission from TFSA-treated 1L MoS2. (a) Maps of maximum absolute PL intensity measured from 1L MoS2 before and after TFSA treatment, and after subsequent ALD of Al2O3. The black areas represent the SiO2/Si substrate where no MoS2 PL signal is detected. (b) Peak PL intensity as a function of the corresponding peak PL energy. The small translucent markers arise from every pixel extracted from each relevant PL map, with the larger solid circles indicating each calculated mean. The logarithmic scale to account for the large increase in PL intensity following TFSA treatment should be noted in (a) and (b). (c) PL spectra measured from the same single spot location on 1L MoS2 before and after TFSA treatment, and after subsequent ALD of Al2O3.

Fig. 3

Open in a new tab

Initial TFSA processing yields the expected increased, blueshifted, and sharpened PL emission, with an observed enhancement of one order of magnitude consistent with our recent report of TFSA–pentane treatment.27 The spectral sharpening indicates increased exciton lifetime and reduced non-radiative recombination.52,53 The modified PL character is associated with the synergistic combination of S vacancy passivation (predominantly by dissociated SO2-based species) and p-type doping by electronegative fluorine (F) atoms in TFSA.17,23,24,54,55 The TFSA-MoS2 interaction proceeds via the spontaneous formation of a thin superacid “nanolayer” on the MoS2 surface as the solvent evaporates during the drying process.25,27,56 ALD of Al2O3, however, removes the elevated PL emission, with the 1L MoS2 PL intensity reduced by ∼33% relative to that emitted prior to any treatment. The PL peak energy is also redshifted back to ∼1.86 eV, while the spectral linewidth is broadened to ∼112 meV, comparable to that of untreated MoS2. An AFM image of the Al2O3 film deposited on TFSA-treated 1L MoS2 after 200 ALD cycles is provided in Fig. S3 in the SI, showing the presence of dielectric islands in an incompletely coalesced layer. It is clear that thermal ALD growth of this Al2O3 layer negates the optical improvement in 1L MoS2 induced by TFSA treatment.

To probe changes to the excitonic contributions, we performed multipeak Lorentzian fitting of the single-site PL spectra to separate the constituent spectral features, with full deconvolutions provided in Fig. S4 in the SI. The PL signal from untreated 1L MoS2 comprised a dominant negatively charged trion, A−, feature at ∼1.85 eV, and A and B excitonic transitions at ∼1.87 eV and ∼2.0 eV, respectively. The ratio between the absolute intensities of the A− and A peaks, A−/A, can be related to the electron density in 1L MoS2, with a larger excess of electrons yielding a higher A−/A ratio.57 As expected, the initial TFSA treatment suppressed the trion contribution, significantly reducing the A−/A ratio from ∼1.9 to ∼0.1, evidence of a p-type doping effect.23,24,27 Subsequent ALD growth of Al2O3 then increased the A−/A ratio to ∼1.6, indicating increased trion formation and near-complete reversal of the p-type doping effects of TFSA treatment. The PL spectrum can also reveal information on the abundance of defects, with the ratio of the absolute intensity of the B exciton peak to that of the A exciton known to be a reliable indicator of defect state density.58 We employed a three-peak deconvolution of the PL spectra, and so defined the excitonic intensity ratio as B/(A + A−). In agreement with previous reports,23,27 TFSA treatment induces a significant reduction in the relative contribution of the B exciton, decreasing the B/(A + A−) ratio from ∼0.2 to near zero, consistent with effective defect passivation. After ALD of Al2O3 on TFSA-treated 1L MoS2, the B/(A + A−) ratio increased back to ∼0.2. We therefore suggest that ALD growth of Al2O3 increases the defect density in superacid-treated MoS2, arising either from the reintroduction of previously passivated vacancies or via the formation of new defective sites, possibly at the MoS2/dielectric interface. However, since the FWHM and B/(A + A−) ratio both returned to values comparable to those typical of untreated MoS2, we speculate that reactivation of intrinsic S vacancies is the primary defect creation process that occurs during ALD of Al2O3 on TFSA-treated 1L MoS2. We suggest that Al2O3 growth does not primarily generate new defect types but instead disrupts or removes the adsorbed superacid “nanolayer” that facilitates passivation and charge transfer. Hence, the TFSA-induced improvement in PL behaviour is not stable under ALD processing. The dielectric growth reverses both the defect suppression and p-type doping effects induced by interaction with the superacidic TFSA.

Isolation of ALD processing effects on TFSA-treated 1L MoS2

To ascertain the origin of the reduced PL emission observed following dielectric growth, we subjected separate TFSA-treated 1L MoS2 samples to several modified ALD processes designed to isolate the effects of vacuum exposure, thermal heating, and H2O co-reactant pulsing in the absence of precursor exposure. We have previously reported that the vacuum environment, thermal processing, and pulsed exposure to H2O during ALD-dielectric growth have negligible impacts on the 1L MoS2 optical properties.23 However, the sensitivity of superacid-treated MoS2 to these individual processing steps has not been previously established. Here, we treated a cleaved 1L MoS2 sample with TFSA before placing it in the ALD chamber under vacuum (chamber pressure <15 mTorr) for ∼2 h, a period comparable to the duration of a 200-cycle deposition. A separate TFSA-treated 1L MoS2 sample was placed in the ALD chamber for ∼2 h, with the chamber pumped to vacuum, and all heaters set to the temperatures typical of a 150 °C deposition, but without the introduction of any precursor gases. A third TFSA-enhanced 1L MoS2 sample was subjected to a modified ALD-Al2O3 recipe that cycled only H2O pulses, omitting the TMA precursor injection steps but retaining the vacuum and thermal conditions characteristic of a standard growth process. Single-spot PL spectra and scatter plots associated with each of these processing routes are shown in Fig. 4, with the associated PL mapping data and selected magnified views of the scatter plots provided in Fig. S5–S8 of the SI.

Fig. 4. Single-site PL spectra (top) and scatter plots of peak PL intensity as a function of the corresponding peak PL energy extracted from PL mapping data (bottom) measured from 1L MoS2 before and after TFSA treatment, and after subsequent (a) ALD-vacuum exposure, (b) heating at 150 °C in the ALD chamber, and (c) ALD-H2O treatment. No TMA precursor was pulsed during any of these three processes. In the scatter plots, the small translucent markers arise from every pixel extracted from each relevant PL map, with the larger solid circles indicating each calculated mean average. The logarithmic scale to account for the large increase in PL intensity following TFSA treatment should be noted.

Fig. 4

Open in a new tab

In each case, the initial TFSA treatment yielded the expected enhanced MoS2 PL character, and this elevated PL emission survived exposure to vacuum in the ALD chamber for ∼2 h (Fig. 4a), with negligible changes to the PL intensity, energy, and spectral linewidth. However, heating to 150 °C under vacuum for ∼2 h (Fig. 4b) reduced the PL intensity from TFSA-treated 1L MoS2 by ∼4 ×. This heating under vacuum also induced an appreciable redshift of ∼7 meV in the PL peak energy relative to the blueshifted position observed after TFSA treatment, accompanied by noticeable spectral broadening to a FWHM intermediate between that of untreated and TFSA-treated 1L MoS2. Pulsed exposure to H2O (Fig. 4c) produced a more pronounced modification of the TFSA-enhanced MoS2 PL signal, resulting in a five-fold attenuation of the intensity, with a redshift of ∼14 meV and a ∼1.3× increase in the FWHM also observed. Such changes to the PL signal indicate partial reversal of the TFSA-induced p-type doping and defect passivation, suggesting that the thermal effects and exposure to H2O during ALD processing can destabilise or weaken the interaction between TFSA-derived species and the 1L MoS2 surface. Compared to the complete removal of the TFSA-induced PL enhancement observed following ALD of Al2O3, the PL intensity remains partially elevated after either heating under vacuum or H2O pulsing. We therefore conclude that the overall impact of Al2O3 deposition on TFSA-treated 1L MoS2 arises from the combined effects of thermal processing and interaction with reactive precursor and co-reactant species.

TFSA treatment of dielectric-encapsulated 1L MoS2

Having demonstrated superacid treatment of 1L MoS2 prior to ALD of Al2O3 does not yield overall PL enhancement, we now explore exposing Al2O3-encapsulated MoS2 to TFSA in an encapsulation/passivation approach. We grew several Al2O3 films on separate 1L MoS2 samples, varying the number of ALD cycles, from 50 cycles to 500 cycles. Fig. 5 displays selected high-magnification AFM images of the Al2O3 film grown at each number of cycles, with approximated maps of the dielectric surface coverage and representative line profiles also displayed for each image. At least five different regions of each sample were probed, with representative images presented here. While AFM obtains the relative height variation, we offset the height trace for each sample such that the mean value of the line profile is centred on the Al2O3 thickness estimated via the growth characteristics determined in our previous study.32

Fig. 5. High-magnification AFM images (top), maps of estimated surface coverage extracted via ImageJ48 (middle), and representative line profiles (bottom) of the ALD-Al2O3 films grown on 1L MoS2 at cycle numbers (a) 50, (b) 200, (c) 300, (d) 400, and (e) 500. The root mean square (RMS) surface roughness, Rq, is indicated in each image. The line profiles are traced along the red dashed lines in each AFM image and the average island height or estimated dielectric thickness is indicated on the bottom row. The height trace of each AFM image is offset such that the mean value of each line profile is centred on the expected thickness of Al2O3 at each cycle number, using the growth characteristics on CVD-1L MoS2 evaluated in our previous study.32.

Fig. 5

Open in a new tab

Dielectric surface coverage increased systematically with ALD cycle number, from only ∼57% at 50 cycles to ∼96% at 400 cycles. Near-complete surface coverage (>99%), with minimal density of significant voids or pinholes observed in the deposited Al2O3 film, was achieved only after 500 cycles. Each Al2O3-encapsulated 1L MoS2 sample was then submerged separately in a TFSA-pentane solution, before being removed and dried in air. To verify the ALD-grown Al2O3 films survived exposure to TFSA, high-magnification AFM images were acquired from each Al2O3-covered 1L MoS2 sample following superacid treatment, as displayed in Fig. S9 in the SI. No significant changes to the dielectric film morphologies were observed after TFSA processing, demonstrating the robustness of each Al2O3 film to the chemical treatment.

The PL emission from each Al2O3-encapsulated sample was measured before and after ALD of Al2O3, and following TFSA treatment. Fig. 6 presents a comparison of the intensity, energy, and FWHM averaged from the PL mapping data acquired after each processing step, with representative single-site PL spectra displayed in Fig. 7. Full PL maps and deconvoluted single-site spectra are provided in Section S8 (Fig. S10–S14) and Fig. S15 of the SI, respectively. As expected, and as previously reported,23,32 ALD growth of Al2O3 on 1L MoS2 attenuates, redshifts, and broadens the PL signal. The extent of these modifications progresses with ALD cycle number, with accompanying increases in the A−/A (Fig. 7b and B)/(A + A−) (Fig. 7c) ratios consistent with increased interfacial n-type doping and heightened defect-state densities as Al2O3 approaches complete coverage.

Fig. 6. Mean (a) PL intensity, (b) PL energy, and (c) PL FWHM as averaged from mapping data measured from 1L MoS2 before (black) and after (red) ALD of Al2O3, and after subsequent TFSA treatment (blue) for each number of ALD cycles. Full PL mapping data can be found in Section S8 of the SI. The error bars correspond to one standard deviation in each calculated mean value. The logarithmic scale to account for the large increase in PL intensity following TFSA treatment should be noted in (a).

Fig. 6

Open in a new tab

Fig. 7. (a) Single-site PL spectra measured from each 1L MoS2 sample before (black) and after (red) ALD of Al2O3, and after subsequent TFSA treatment (blue) for each number of ALD cycles. A magnified view of the low intensity spectra is displayed in each inset. (b) A−/A and (c) B/(A + A−) intensity ratios extracted from each of the spectra in (a). Full deconvoluted single-site PL spectra are provided in Fig. S15 in the SI.

Fig. 7

Open in a new tab

Following TFSA treatment of the Al2O3-encapsulated samples, a strong recovery and further substantial enhancement of the PL emission was observed for all ALD cycle numbers up to 400 cycles (Fig. 6a). Each increased PL intensity was accompanied by a blueshift in the PL energy (Fig. 6b) and a narrowing of the PL FWHM (Fig. 6c), consistent with defect passivation and p-type doping.23,27 We define a PL enhancement factor as the ratio of the maximum absolute PL intensity averaged from each TFSA-treated Al2O3/1L MoS2 sample to that obtained from the corresponding encapsulated sample before superacid exposure. An analogous sharpening factor is also introduced to evaluate changes to the PL spectral width. At 50 ALD cycles, TFSA treatment induced a ∼20× PL enhancement, a ∼30 meV blueshift, and a ∼2× narrowing of the PL linewidth relative to the Al2O3-covered state. The degree of PL enhancement increased progressively with ALD cycle number up to ∼100× at 400 cycles. The magnitudes of blueshift and FWHM sharpening also increased systematically with cycle number; at 400 ALD cycles, the relative blueshift in PL energy was ∼100 meV and the FWHM was reduced significantly (by ∼7 ×). The increasing PL modification with cycle number is consistent with the introduction of additional ALD-induced defects that may subsequently be passivated by TFSA. Dielectric deposition can create a variety of defect states in 1L MoS2,32 including S vacancies.59 TFSA is known to repair intrinsic S vacancies effectively and may similarly passivate such process-induced defects. As a result, non-radiative recombination pathways are suppressed and the PL emission becomes exciton-dominated, reversing the ALD-induced n-type doping, as evidenced by the near-zero A−/A and B/(A + A−) intensity ratios observed after TFSA treatment (Fig. 6b and c). At 500 ALD cycles, however, the PL intensity was not enhanced by the TFSA treatment, remaining at a magnitude comparable to that emitted after ALD of Al2O3, with the PL energy and FWHM also unchanged by superacid exposure. In addition, the A−/A and B/(A + A−) ratios were also unaffected, demonstrating the persistence of strong n-type doping and a high defect density.

Using spatially resolved Raman spectroscopy, we further elucidate the underlying mechanisms behind modification of the optical properties of 1L MoS2 following ALD of Al2O3 and subsequent TFSA treatment. Spatial maps of the Raman peak positions are displayed in Section S10 (Fig. S16–S20) in the SI. We performed a correlative analysis of Raman mapping data to separate changes to the mechanical strain, ε, and electron density, n, in 1L MoS2. Such quantitative evaluation of strain and doping is commonly applied to a range of 2D materials,23,32,36,60–67 and full details of this process are provided in Section S11 of the SI. Resulting plots of A1g peak position as a function of E12g wavenumber for each number of ALD cycles are shown in Fig. 8, each overlain with a grid of strain (black) and doping (red dashed) isolines.

Fig. 8. (a) Correlative plots of A1g and E12g peak positions extracted from Raman mapping data acquired from each 1L MoS2 sample before (black) and after (red) ALD of Al2O3, and after subsequent TFSA treatment (blue) for each number of ALD cycles. The small circular translucent markers arise from every pixel extracted from the Raman mapping data, and the large diamond points indicate the corresponding average values. Strain isolines (black solid lines) indicate Δε = ±0.1% variations in the strain and doping isolines (red dashed lines) correspond to relative changes in the electron density of Δn = ± 0.1 × 1013 cm−2. Full Raman mapping data can be found in Section S10 of the SI. Mean relative (b) electron density (Δn) and (c) mechanical biaxial strain (Δε) extracted from (a). The error bars correspond to one standard deviation in each calculated mean value.

Fig. 8

Open in a new tab

The extracted variations in electron density (Fig. 8b) and strain (Fig. 8c) are in excellent agreement with the PL analysis discussed above. At all cycle numbers, ALD of Al2O3 increased the electron density, consistent with the trend in the A−/A intensity ratio and indicating progressive n-type doping. Subsequent TFSA treatment substantially reduced the electron density in Al2O3-encapsulated 1L MoS2 samples for all ALD cycle numbers up to 400 cycles. This confirms the effective reversal of ALD-induced doping via charge transfer to TFSA species and suppression of electron-rich defective sites.68 In contrast, TFSA treatment of the 1L MoS2 sample coated with 500 ALD cycles of Al2O3 had little effect on the electron density, which remained relatively high at ∼2 × 1013 cm−2.

Mechanical strain in 1L MoS2 is known to be sensitive to the presence of S vacancies, with heightened biaxial tensile strain associated with extrinsic occupation of such vacancy sites, corresponding to an increase in the lattice parameters.17,24 Following 50 cycles of Al2O3, we observed no appreciable change in the strain, consistent with low dielectric coverage. At higher cycle numbers, deposition of Al2O3 led to an increase in compressive strain in 1L MoS2, which we attribute to dielectric-induced mechanical interactions and interfacial bonding. Subsequent TFSA treatment of Al2O3-encapsulated 1L MoS2 yielded a progressive increase in mechanical tension in all samples up to 400 cycles, reaching ∼0.6% relative tensile strain at 400 cycles. This increased tensile strain is consistent with the passivation of S vacancies indicated by the changes in the PL emission, namely the narrowed linewidth and reduced B/(A + A−) ratio. However, it is important to note that additional factors, such as changes in the dielectric-induced interfacial strain, may also influence the overall strain in the underlying 1L MoS2 layer. At 500 cycles of Al2O3, however, TFSA treatment gave no increase in mechanical tension, with the small compressive strain (< 0.1%) induced by the dielectric growth maintained after superacid exposure. This absence of strain modification further supports the conclusion that TFSA is unable to influence 1L MoS2 through the Al2O3 layer formed after 500 ALD cycles.

We rationalise the cycle-dependent evolution of the Raman and PL responses in terms of the accessibility of the MoS2 surface to TFSA-derived species. TFSA-induced enhancement of uncovered 1L MoS2 proceeds via the spontaneous formation of a nanoscale superacid layer,25,27,56 enabling adsorption of dissociated TFSA fragments. The elevated PL intensity from TFSA-treated 1L MoS2 is widely attributed to the combined effects of S vacancy passivation and p-type charge transfer.17,23,24,54,55 In this mechanism, dissociated SO2-based species are thought to directly repair vacancy sites,17 with protonic species also able to localise at or close to such defects.69 Electronegative F-containing TFSA species can withdraw electron density from the MoS2 lattice.23,24 Therefore, for TFSA to enhance the PL character of Al2O3-encapsulated 1L MoS2, protonic, S-, and F-containing fragments must be able to reach the basal plane. At low ALD cycle numbers, the Al2O3 film is incompletely coalesced and comprises distinct islands (Fig. 5), leaving the underlying 1L MoS2 partially exposed through pinholes and voids. Here, the dissociated TFSA-derived species can directly access the MoS2 surface, enabling defect passivation and charge transfer to enhance the PL. As the number of ALD cycles increases, the Al2O3 film becomes thicker and more continuous, reducing direct access to the underlying 1L MoS2. TFSA species may instead reach the MoS2 basal plane via defects such as grain boundaries,70,71 or diffusion through the Al2O3 layer. TFSA molecules have been shown to diffuse effectively through fluoropolymeric layers to interact with underlying MoS2.41 While transport through polymers is relatively fast, the diffusion rates of species dissociated from TFSA are expected to be much slower in a higher-density amorphous oxide environment,72 such as that of ALD-grown Al2O3 films.73,74 Previous studies have shown that TFSA-derived species can be detected within oxide layers following superacid treatment,22 while sterically smaller dissociated species, such as protons, have been reported to migrate through ALD-grown dielectric layers.75,76 Since we observe the persistence of TFSA-induced enhancement at high dielectric coverages up to ∼96%, diffusion through the Al2O3 layer cannot be excluded. As the Al2O3 film becomes thicker and approaches complete coalescence, transport pathways for TFSA-derived species to the underlying 1L MoS2 are expected to become increasingly restricted, suppressing the efficacy of TFSA treatment. Once the Al2O3 film is sufficiently thick and near-continuous, no TFSA-induced enhancement of the underlying MoS2 is observed.

4. Conclusions

We have examined the combination of two processing routes that are critical for realisation of 1L MoS2-based optoelectronic devices: superacid treatment and ALD of a high-κ dielectric material. We have established the instability of TFSA-enhanced 1L MoS2 PL during subsequent ALD-dielectric growth, demonstrating that this degradation arises from thermal exposure and interaction with reactive precursor species during the deposition process. We then investigated the reverse processing sequence, exploring the superacid treatment of Al2O3-coated 1L MoS2 as an encapsulation/passivation approach. We revealed that, when performed after dielectric deposition, TFSA treatment can recover and significantly enhance the optical properties of dielectric-encapsulated 1L MoS2, even at high dielectric coverages. By systematically varying the number of ALD cycles in the growth of Al2O3, we have shown that the effectiveness of post-encapsulation superacid treatment is governed by the ability of TFSA-derived species to access the MoS2 surface. Strong PL enhancement, accompanied by suppressed defect signatures and reversal of n-type doping, is observed for Al2O3 films of up to ∼37 nm thickness and ∼96% surface coverage. At greater thickness and dielectric coverage, however, the Al2O3 layer is sufficiently thick and completely coalesced to eliminate both diffusion and pinhole-mediated access, preventing TFSA-induced enhancement of 1L MoS2. This study provides important insights for the integration of chemical passivation strategies in practical device architectures. The extension of this approach to other high-κ dielectrics, including HfO2, ZnO, and ZrO2, whose growth characteristics and interfacial properties on CVD-1L MoS2 are known to differ from those of Al2O3, represents an important direction for future work.

Author contributions

This study was conceptualised by B. F. M. H. Experimental work was primarily conducted by B. F. M. H., with support from K. J. H. and S. L. P. Data analysis was largely performed by B. F. M. H., with discussions and contributions from S. L. P., N. E. G., and J. D. M. The manuscript was written by B. F. M. H., with input and editing from S. L. P., K. J. H., N. E. G., and J.D.M.

Conflicts of interest

There are no conflicts to declare.

Supplementary Material

TC-OLF-D6TC02199A-s001

Acknowledgments

The authors acknowledge the use of the Raman/PL facilities within the Spectroscopy Research Technology Platform (RTP), ALD facilities within the Nano Fabrication RTP, and AFM facilities within the Electron Microscopy RTP at the University of Warwick, which are supported by EPSRC (No. EP/V007688/1). S. L. P. was supported by a Royal Academy of Engineering Research Fellowship (No. RF-2324-23-197).

Data availability

Data underpinning figures in this paper can be freely downloaded from https://doi.org/10.82444/warw.33952129. Requests for additional data should be made directly to the corresponding author.

Supplementary information (SI): additional PL mapping data, AFM imaging, and single-site PL deconvolution for ALD of Al2O3 on TFSA-treated 1L MoS2; PL mapping data for ALD control processes on TFSA-treated 1L MoS2; AFM images, PL mapping data, single-site PL deconvolution, and Raman mapping data for TFSA treatment of Al2O3-encapsulated 1L MoS2; and correlative analysis of strain and doping in 1L MoS2. See DOI: https://doi.org/10.1039/d6tc02199a.

References

  1. Mak K. F. Lee C. Hone J. Shan J. Heinz T. F. Phys. Rev. Lett. 2010;105:136805. doi: 10.1103/PhysRevLett.105.136805. [DOI] [PubMed] [Google Scholar]
  2. Splendiani A. Sun L. Zhang Y. Li T. Kim J. Chim C.-Y. Galli G. Wang F. Nano Lett. 2010;10:1271–1275. doi: 10.1021/nl903868w. [DOI] [PubMed] [Google Scholar]
  3. Lien D.-H. Amani M. Desai S. B. Ahn G. H. Han K. He J.-H. Ager J. W. Wu M. C. Javey A. Nat. Commun. 2018;9:1229. doi: 10.1038/s41467-018-03218-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Kwon H. Garg S. Park J. H. Jeong Y. Yu S. Kim S. M. Kung P. Im S. npj 2D Mater. Appl. 2019;3:9. [Google Scholar]
  5. Fontana M. Deppe T. Boyd A. K. Rinzan M. Liu A. Y. Paranjape M. Barbara P. Sci. Rep. 2013;3:1634. doi: 10.1038/srep01634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Pospischil A. Furchi M. M. Mueller T. Nat. Nanotechnol. 2014;9:257–261. doi: 10.1038/nnano.2014.14. [DOI] [PubMed] [Google Scholar]
  7. Furchi M. M. Pospischil A. Libisch F. Burgdörfer J. Mueller T. Nano Lett. 2014;14:4785–4791. doi: 10.1021/nl501962c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Zhang Y. Li H. Wang L. Wang H. Xie X. Zhang S.-L. Liu R. Qiu Z.-J. Sci. Rep. 2015;5:7938. doi: 10.1038/srep07938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Kufer D. Konstantatos G. Nano Lett. 2015;15:7307–7313. doi: 10.1021/acs.nanolett.5b02559. [DOI] [PubMed] [Google Scholar]
  10. Lopez-Sanchez O. Lembke D. Kayci M. Radenovic A. Kis A. Nat. Nanotechnol. 2013;8:497–501. doi: 10.1038/nnano.2013.100. [DOI] [PubMed] [Google Scholar]
  11. Yang Y. Huo N. Li J. J. Mater. Chem. C. 2017;5:11614–11619. [Google Scholar]
  12. Yin Z. Li H. Li H. Jiang L. Shi Y. Sun Y. Lu G. Zhang Q. Chen X. Zhang H. ACS Nano. 2012;6:74–80. doi: 10.1021/nn2024557. [DOI] [PubMed] [Google Scholar]
  13. Lee H. S. Min S.-W. Chang Y.-G. Park M. K. Nam T. Kim H. Kim J. H. Ryu S. Im S. Nano Lett. 2012;12:3695–3700. doi: 10.1021/nl301485q. [DOI] [PubMed] [Google Scholar]
  14. Park H. Liu N. Kim B. H. Kwon S. H. Baek S. Kim S. Lee H.-K. Yoon Y. J. Kim S. ACS Appl. Mater. Interfaces. 2020;12:20645–20652. doi: 10.1021/acsami.0c02393. [DOI] [PubMed] [Google Scholar]
  15. Hong J. Hu Z. Probert M. Li K. Lv D. Yang X. Gu L. Mao N. Feng Q. Xie L. Zhang J. Wu D. Zhang Z. Jin C. Ji W. Zhang X. Yuan J. Zhang Z. Nat. Commun. 2015;6:6293. doi: 10.1038/ncomms7293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Plechinger G. Mann J. Preciado E. Barroso D. Nguyen A. Eroms J. Schüller C. Bartels L. Korn T. Semicond. Sci. Technol. 2014;29:064008. [Google Scholar]
  17. Roy S. Choi W. Jeon S. Kim D.-H. Kim H. Yun S. J. Lee Y. Lee J. Kim Y.-M. Kim J. Nano Lett. 2018;18:4523–4530. doi: 10.1021/acs.nanolett.8b01714. [DOI] [PubMed] [Google Scholar]
  18. Wang H. Zhang C. Rana F. Nano Lett. 2015;15:339–345. doi: 10.1021/nl503636c. [DOI] [PubMed] [Google Scholar]
  19. Madapu K. K. Bhuyan C. A. Srivastava S. K. Dhara S. J. Mater. Chem. C. 2021;9:3578–3588. [Google Scholar]
  20. Ahmed T. Zha J. Lin K. K. H. Kuo H.-C. Tan C. Lien D.-H. Adv. Mater. 2023;35:2208054. doi: 10.1002/adma.202208054. [DOI] [PubMed] [Google Scholar]
  21. Nalwa H. S. RSC Adv. 2020;10:30529–30602. doi: 10.1039/d0ra03183f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grant N. E. Pain S. L. White J. T. Walker M. Prokes I. Murphy J. D. ACS Appl. Energy Mater. 2022;5:1542–1550. [Google Scholar]
  23. Healy B. F. M. Pain S. L. Lloyd-Hughes J. Grant N. E. Murphy J. D. Adv. Mater. Interfaces. 2024;11:2400305. [Google Scholar]
  24. Kim H. Lee T. Ko H. Kim S. Rho H. Appl. Phys. Lett. 2020;117:202104. [Google Scholar]
  25. Yamada Y. Zhang Y. Ikeno H. Shinokita K. Yoshimura T. Ashida A. Fujimura N. Matsuda K. Kiriya D. ACS Appl. Mater. Interfaces. 2021;13:25280–25289. doi: 10.1021/acsami.1c04980. [DOI] [PubMed] [Google Scholar]
  26. Amani M. Lien D.-H. Kiriya D. Xiao J. Azcatl A. Noh J. Madhvapathy S. R. Addou R. Kc S. Dubey M. Cho K. Wallace R. M. Lee S.-C. He J.-H. Ager J. W. Zhang X. Yablonovitch E. Javey A. Science. 2015;350:1065–1068. doi: 10.1126/science.aad2114. [DOI] [PubMed] [Google Scholar]
  27. Healy B. F. M. Pain S. L. Grant N. E. Murphy J. D. J. Vac. Sci. Technol., A. 2026;44:032201. [Google Scholar]
  28. Lin G. Zhao M.-Q. Jia M. Cui P. Zhao H. Zhang J. Gundlach L. Liu X. Johnson A. T. C. Zeng Y. J. Phys. D: Appl. Phys. 2020;53:415106. [Google Scholar]
  29. Wang J. Lai H. Huang X. Liu J. Lu Y. Liu P. Xie W. Materials. 2022;15:5859. doi: 10.3390/ma15175859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu H. Xu K. Zhang X. Ye P. D. Appl. Phys. Lett. 2012;100:152115. [Google Scholar]
  31. Park T. Kim H. Leem M. Ahn W. Choi S. Kim J. Uh J. Kwon K. Jeong S.-J. Park S. Kim Y. Kim H. RSC Adv. 2017;7:884–889. [Google Scholar]
  32. Healy B. F. M. Pain S. L. Grant N. E. Murphy J. D. Nanoscale. 2025;17:15436–15447. doi: 10.1039/d5nr01144b. [DOI] [PubMed] [Google Scholar]
  33. Jung C. Yang H. I. Choi W. Nanoscale Res. Lett. 2019;14:278. doi: 10.1186/s11671-019-3119-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Park S. Kim S. Y. Choi Y. Kim M. Shin H. Kim J. Choi W. ACS Appl. Mater. Interfaces. 2016;8:11189–11193. doi: 10.1021/acsami.6b01568. [DOI] [PubMed] [Google Scholar]
  35. Azcatl A. McDonnell S. Santosh K. C. Peng X. Dong H. Qin X. Addou R. Mordi G. I. Lu N. Kim J. Kim M. J. Cho K. Wallace R. M. Appl. Phys. Lett. 2014;104:111601. [Google Scholar]
  36. Healy B. F. M. Pain S. L. Walker M. Grant N. E. Murphy J. D. ACS Appl. Nano Mater. 2025;8:7334–7346. doi: 10.1021/acsanm.5c00901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Price K. M. Najmaei S. Ekuma C. E. Burke R. A. Dubey M. Franklin A. D. ACS Appl. Nano Mater. 2019;2:4085–4094. [Google Scholar]
  38. McDonnell S. Brennan B. Azcatl A. Lu N. Dong H. Buie C. Kim J. Hinkle C. L. Kim M. J. Wallace R. M. ACS Nano. 2013;7:10354–10361. doi: 10.1021/nn404775u. [DOI] [PubMed] [Google Scholar]
  39. Pointon A. I. Grant N. E. Bonilla R. S. Wheeler-Jones E. C. Walker M. Wilshaw P. R. Dancer C. E. J. Murphy J. D. ACS Appl. Electron. Mater. 2019;1:1322–1329. [Google Scholar]
  40. Alharbi A. Zahl P. Shahrjerdi D. Appl. Phys. Lett. 2017;110:033503. [Google Scholar]
  41. Kim H. Lien D.-H. Amani M. Ager J. W. Javey A. ACS Nano. 2017;11:5179–5185. doi: 10.1021/acsnano.7b02521. [DOI] [PubMed] [Google Scholar]
  42. Tyson K. H., Godfrey J. R., Fraser J. M. and Knobel R. G., 2020 Photonics North, 2020, p. 1 [Google Scholar]
  43. Pain S. L. Grant N. E. Murphy J. D. ACS Nano. 2022;16:1260–1270. doi: 10.1021/acsnano.1c09085. [DOI] [PubMed] [Google Scholar]
  44. Singh O. and Singh V. K., in Hazardous Chemicals, ed. M. Chawla, J. Singh and R. D. Kaushik, Academic Press, 2025, pp. 305–313 10.1016/B978-0-323-95235-4.00005-0 [DOI] [Google Scholar]
  45. Schlosser P. Bale A. Gibbons C. Wilkins A. Cooper G. Environ. Health Perspect. 2014;123:114–119. doi: 10.1289/ehp.1308030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Xu K. Sun W. Shao Y. Wei F. Zhang X. Wang W. Li P. Nanotechnol. Rev. 2018;7:605–621. [Google Scholar]
  47. Nečas D. Klapetek P. Open Phys. 2012;10:181–188. [Google Scholar]
  48. Schindelin J. Arganda-Carreras I. Frise E. Kaynig V. Longair M. Pietzsch T. Preibisch S. Rueden C. Saalfeld S. Schmid B. Tinevez J.-Y. White D. J. Hartenstein V. Eliceiri K. Tomancak P. Cardona A. Nat. Methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Jeon J. Jang S. K. Jeon S. M. Yoo G. Jang Y. H. Park J.-H. Lee S. Nanoscale. 2015;7:1688–1695. doi: 10.1039/c4nr04532g. [DOI] [PubMed] [Google Scholar]
  50. Radisavljevic B. Radenovic A. Brivio J. Giacometti V. Kis A. Nat. Nanotechnol. 2011;6:147–150. doi: 10.1038/nnano.2010.279. [DOI] [PubMed] [Google Scholar]
  51. Zhang X. Qiao X.-F. Shi W. Wu J.-B. Jiang D.-S. Tan P.-H. Chem. Soc. Rev. 2015;44:2757–2785. doi: 10.1039/c4cs00282b. [DOI] [PubMed] [Google Scholar]
  52. Li R. Li Y. Tian H. Liao P. Wang H. Zhang S. Yao Z. Wang H. Liu S. Chen G. Yu S. Li Z. Liu J. Xu Z. Mei F. Liu P. Guo J. Liu K. Li X. Liu L. ACS Appl. Electron. Mater. 2020;2:1981–1988. [Google Scholar]
  53. Lin Y. Ling X. Yu L. Huang S. Hsu A. L. Lee Y.-H. Kong J. Dresselhaus M. S. Palacios T. Nano Lett. 2014;14:5569–5576. doi: 10.1021/nl501988y. [DOI] [PubMed] [Google Scholar]
  54. Lu H. Kummel A. Robertson J. APL Mater. 2018;6:066104. [Google Scholar]
  55. Feng Q. Sun Y. Li Y. Yan J. Zhong W. Yang G. Liu W. Xu H. Liu Y. Laser Photonics Rev. 2021;15:2100104. [Google Scholar]
  56. Tyson K. H. Godfrey J. R. Fraser J. M. Knobel R. G. ACS Appl. Mater. Interfaces. 2024;16:53186–53194. doi: 10.1021/acsami.4c00836. [DOI] [PubMed] [Google Scholar]
  57. Mak K. F. He K. Lee C. Lee G. H. Hone J. Heinz T. F. Shan J. Nat. Mater. 2013;12:207–211. doi: 10.1038/nmat3505. [DOI] [PubMed] [Google Scholar]
  58. McCreary K. M. Hanbicki A. T. Sivaram S. V. Jonker B. T. APL Mater. 2018;6:111106. doi: 10.1021/acsami.8b01224. [DOI] [PubMed] [Google Scholar]
  59. Jung H. Kim M. Lee Y. Sim G. B. Gu H. Hong S. Lee S. Lee J. Lee D. Zou T. Kang K. Myung C. W. Noh Y.-Y. Kwon J. ACS Nano. 2025;19:6069–6078. doi: 10.1021/acsnano.4c12927. [DOI] [PubMed] [Google Scholar]
  60. Mohiuddin T. M. G. Lombardo A. Nair R. R. Bonetti A. Savini G. Jalil R. Bonini N. Basko D. M. Galiotis C. Marzari N. Novoselov K. S. Geim A. K. Ferrari A. C. Phys. Rev. B. 2009;79:205433. [Google Scholar]
  61. Lee J. E. Ahn G. Shim J. Lee Y. S. Ryu S. Nat. Commun. 2012;3:1024. doi: 10.1038/ncomms2022. [DOI] [PubMed] [Google Scholar]
  62. Schilirò E. Lo Nigro R. Panasci S. E. Gelardi F. M. Agnello S. Yakimova R. Roccaforte F. Giannazzo F. Carbon. 2020;169:172–181. [Google Scholar]
  63. Schilirò E. Panasci S. E. Mio A. M. Nicotra G. Agnello S. Pecz B. Radnoczi G. Z. Deretzis I. La Magna A. Roccaforte F. Lo Nigro R. Giannazzo F. Appl. Surf. Sci. 2023;630:157476. [Google Scholar]
  64. Michail A. Anestopoulos D. Delikoukos N. Parthenios J. Grammatikopoulos S. Tsirkas S. A. Lathiotakis N. N. Frank O. Filintoglou K. Papagelis K. 2D Mater. 2021;8:015023. [Google Scholar]
  65. Chae W. H. Cain J. D. Hanson E. D. Murthy A. A. Dravid V. P. Appl. Phys. Lett. 2017;111:143106. [Google Scholar]
  66. Panasci S. E. Schilirò E. Greco G. Cannas M. Gelardi F. M. Agnello S. Roccaforte F. Giannazzo F. ACS Appl. Mater. Interfaces. 2021;13:31248–31259. doi: 10.1021/acsami.1c05185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Schilirò E. Nigro R. L. Panasci S. E. Agnello S. Cannas M. Gelardi F. M. Roccaforte F. Giannazzo F. Adv. Mater. Interfaces. 2021;8:2101117. doi: 10.1021/acsami.1c05185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Tan A. M. Z. Freysoldt C. Hennig R. G. Phys. Rev. Mater. 2020;4:064004. [Google Scholar]
  69. Ahn B. Ahn J. Kim M. Lim S. C. Ji H. ACS Appl. Mater. Interfaces. 2024;16:68553–68561. doi: 10.1021/acsami.4c14152. [DOI] [PubMed] [Google Scholar]
  70. Sahadeo E. Rubloff G. Lee S. B. Lin C.-F. Front. Energy Res. 2021;9:618368. [Google Scholar]
  71. Tautschnig M. P. Harrison N. M. Finnis M. W. Acta Mater. 2017;132:503–516. [Google Scholar]
  72. Suarez S. N. Rúa A. Cuffari D. Pilar K. Hatcher J. L. Ramati S. Wishart J. F. J. Phys. Chem. B. 2015;119:14756–14765. doi: 10.1021/acs.jpcb.5b08658. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Oh H. Wu Q. Park S. Kisslinger K. Nam C.-Y. ACS Appl. Mater. Interfaces. 2026;18:22552–22564. doi: 10.1021/acsami.5c25914. [DOI] [PubMed] [Google Scholar]
  74. Li Y. Li X. Chen H. Shi J. Shang Q. Zhang S. Qiu X. Liu Z. Zhang Q. Xu H. Liu W. Liu X. Liu Y. ACS Appl. Mater. Interfaces. 2017;9:27402–27408. doi: 10.1021/acsami.7b08893. [DOI] [PubMed] [Google Scholar]
  75. Zeng Y. Khandelwal S. Shariar K. F. Wang Z. Lin G. Cheng Q. Cui P. Opila R. Balakrishnan G. Addamane S. Taheri P. Kiriya D. Hettick M. Javey A. IEEE Trans. Electron Devices. 2019;66:1856–1861. [Google Scholar]
  76. Muta H. Etoh Y. Ohishi Y. Kurosaki K. Yamanaka S. J. Nucl. Sci. Technol. 2012;49:544–550. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

TC-OLF-D6TC02199A-s001
TC-OLF-D6TC02199A-s001.pdf (5.8MB, pdf)

Data Availability Statement

Data underpinning figures in this paper can be freely downloaded from https://doi.org/10.82444/warw.33952129. Requests for additional data should be made directly to the corresponding author.

Supplementary information (SI): additional PL mapping data, AFM imaging, and single-site PL deconvolution for ALD of Al2O3 on TFSA-treated 1L MoS2; PL mapping data for ALD control processes on TFSA-treated 1L MoS2; AFM images, PL mapping data, single-site PL deconvolution, and Raman mapping data for TFSA treatment of Al2O3-encapsulated 1L MoS2; and correlative analysis of strain and doping in 1L MoS2. See DOI: https://doi.org/10.1039/d6tc02199a.

Articles from Journal of Materials Chemistry. C are provided here courtesy of Royal Society of Chemistry

RESOURCES