Revealing key surface contaminants via stack-flipping strategy: investigating rinsing protocols for clean graphene transfer and enhanced electrical performance

Abstract

Graphene, as a promising 2D material with outstanding thermal, optical, mechanical, and electronic properties, shows great potential for next-generation device applications. During the wet transfer process of CVD monolayer graphene, acid/alkali cleaning agents are gradually used to facilitate the complete removal of polymeric support layers, for example PMMA, enabling a clean transfer of graphene. However, a comprehensive understanding of potential surface contaminants during this process, particularly the key factor influencing subsequent PMMA removal and the ultimate cleanliness of graphene, remains limited. Here, we report a facile stack flipping strategy to directly expose the surface of the etched PMMA/graphene film, enabling in-situ visualization and analysis of the residual contaminants under various rinsing conditions, including deionized (DI) water, acid, and salt solutions. Through systematic characterization and analysis, the key influencing factor and underlying mechanism governing the impact of acid/alkali solution treatments on the surface cleanliness of transferred graphene have been elucidated. Notably, graphene field-effect transistors (GFETs) fabricated using acid rinsing protocols exhibited a distinct negative shift in Dirac point voltage and demonstrated superior carrier mobility compared to devices prepared via the conventional transfer method. The proposed approach demonstrates great promise for enhancing quality control in graphene transfer, thereby facilitating its integration into high performance (opto)electronic device applications.

---

Yifan Yao

Yifan Yao received his PhD degree from the Institute of Chemistry, Chinese Academy of Sciences (ICCAS), in 2016 under the supervision of Prof. Wenping Hu. He then joined the Institute de Science et d’Ingénierie Supramoléculaires (I.S.I.S.) at the University of Strasbourg, France, as a postdoctoral researcher in the group of Prof. Paolo Samorì. In 2021, he joined Hunan University (HNU) as a full professor. His research interests focus on the nanofabrication of organic optoelectronic devices, including organic field-effect transistors, organic photodetectors, and organic electrochemical transistors.

---

Introduction

Graphene, a unique two-dimensional (2D) material consisting of a single layer of sp²-hybridized carbon atoms in a hexagonal lattice, has attracted tremendous attention due to its exceptional intrinsic properties and great potential in various applications.^1–5 Since its discovery through mechanical exfoliation, considerable efforts have been dedicated to the development of efficient and controllable graphene preparation techniques.^6–10 Among these various synthetic routes, chemical vapor deposition (CVD) using catalytically active metal substrates, such as copper (Cu) foils, has emerged as the most reliable and widely adopted method for producing large-area high-quality graphene films. This technique is well-suited for both fundamental experimental research and industrial-scale applications.^11–13 The exceptional properties of CVD-grown graphene make it a highly promising candidate for integration into advanced electronic and photonic devices, state-of-the-art sensing platforms, high-performance energy storage systems, and innovative bioelectronic applications.

However, for most practical applications, CVD-grown graphene films need to be detached from their original growth substrates and transferred onto desired target substrates for device fabrication. Conventionally, poly(methyl methacrylate) (PMMA) serves as a supporting layer during the transfer process and is subsequently removed using organic solvents such as acetone. This widely adopted wet transfer method is favored for its operational simplicity, high transfer yield, good versatility and compatibility with scalable production.^5,11,14 However, complete removal of PMMA residues through solvent-based treatments remains challenging. Such residual polymer contaminants inevitably remain on the graphene surface, significantly degrading its intrinsic properties and thereby compromising the performance and reliability of graphene-based functional devices.^15–18

To address the challenge of polymer residues on the graphene surface, various cleaning strategies have been developed, including high-temperature annealing, electron beam cleaning, supercritical CO₂ treatment, laser cleaning, and plasma or ozone treatments.^15,19–23 Nevertheless, these methods often introduce additional structural defects and modify graphene’s electronic properties, thereby limiting their practical applicability. Consequently, beyond focusing solely on post-transfer residue removal considerable attention has been directed toward optimizing transfer protocols themselves to minimize contamination and obtain cleaner graphene surfaces directly. For instance, using polymers with lower molecular weights or reducing the PMMA layer thickness through decreased polymer concentration has been shown to effectively reduce residual contamination and improve graphene quality.^16,24–26 Moreover, an acid/alkali cleaning step was incorporated into the conventional transfer protocol and successfully demonstrated to achieve a clean and residue-free transfer of graphene.^27,28 This was mainly attributed to the effective elimination of residual metal contaminants (Fe and Cu) by the cleaning agents during the transfer process, thereby promoting the removal of PMMA residues. However, current understanding of the key influencing factor and the mechanism underlying PMMA residue removal facilitated by acid/alkali solution treatments remains limited. Detailed characterization and analysis of the film during the cleaning process, along with the identification of critical influencing factors, have not been adequately addressed. Consequently, further comprehensive investigations are required, including contaminant analysis through surface characterization of PMMA/graphene films, systematic evaluation of cleaning agent efficacy, and advanced predictive modeling of residual polymer elimination.

In this work, we demonstrate that employing a stack flipping strategy to expose the cleaned surface of PMMA/graphene films not only enables direct observation and characterization of potential contaminants under different rinsing protocols, but also facilitates predictive assessment of the final graphene surface cleanliness upon completion of the entire transfer procedure. Numerous visible contaminants adhering to the PMMA/graphene surface post Cu etching are clearly observed by optical microscopy. Raman spectroscopy indicates that these residues are predominantly carbon-based, likely amorphous carbon, with varying levels of agglomeration. XPS and EDS analyses of the flipped PMMA/graphene films indicate that after rinsing with various rinsing solutions for 2 h, Fe residues from the etchant are completely removed from the film surface, and only trace amounts of Cu elements with similar signal intensity for all samples are detected. However, multiple surface morphology characterizations demonstrate that samples treated with acid rinsing exhibit clean surfaces with almost no visible carbon-based contaminants, while those rinsed with DI water or ammonium persulfate show high concentrations of residual contaminants. Notably, the significant discrepancies in residual contaminant levels, depending on the rinsing method, are highly consistent with the distribution of residual PMMA on the transferred graphene surface under identical rinsing treatments. From these results, the elimination of undesired carbon-based contaminants in advance, resulting in a clean PMMA/graphene surface, is a more critical factor determining whether PMMA can be efficiently and completely removed in the following transfer process. Due to the removal of PMMA residues, graphene transferred by the acid rinsing process exhibits excellent electrical properties, indicated by a substantially increased carrier mobility and more balanced charge-carrier transport compared to those prepared by the conventional transfer method. This work thus provides deeper insights into the mechanisms governing graphene surface contamination during wet transfer, offering valuable guidance for optimizing transfer protocols to advance high-performance graphene-based electronics.

Results and discussion

Fig. 1a illustrates the PMMA-assisted wet transfer process for CVD-grown monolayer graphene onto SiO₂/Si substrates, which consists of four essential steps: (1) spin-coating a PMMA protective layer onto the graphene-covered Cu foil; (2) selective chemical etching to remove the underlying Cu substrate; (3) thorough rinsing of the resultant PMMA/graphene films, and (4) dissolving the PMMA layer to release the graphene films after being transferred onto the target substrates. A detailed description of the transfer protocol can be found in the Experimental section (supplementary information, SI), and corresponding photographs of each transfer step are provided in Fig. S1. Unless otherwise specified, graphene samples used in this study were supplied by LG Electronics. In contrast to previous graphene transfer studies that primarily focused on the initial identification of acid rinsing as a key factor for effective PMMA removal or large-area clean graphene transfer,^27,28 this study evaluates multiple rinsing solutions and assesses their effects on the final surface cleanliness and structural integrity of the transferred graphene. Specifically, these rinsing solutions include DI water, HCl : H₂O₂ : H₂O, HNO₃ : H₂O₂ : H₂O, H₂SO₄ : H₂O₂ : H₂O, and (NH₄)₂S₂O₈, which are subsequently abbreviated as H₂O, HCl, HNO₃, H₂SO₄ and (NH₄)₂S₂O₈, respectively, for simplicity. It is worth noting that multiple critical experimental parameters including the type and concentration of etchants, etching duration, rinsing time, post-transfer annealing temperature and time, and acetone cleaning were rigorously controlled and standardized across all samples to minizine extraneous variability and ensure reliable comparison of transfer outcomes.

Fig. 1 (a) Schematic illustration of the PMMA-assisted wet transfer process used for transferring CVD-grown graphene. (b) Photographs of the H₂O-rinsed PMMA/graphene films on SiO₂/Si substrates (i) before and (iii) after PMMA removal, and HCl-rinsed PMMA/graphene films on SiO₂/Si substrates (ii) before and (iv) after PMMA removal. High-magnification OM images of graphene films on SiO₂/Si substrates, during the transfer process after rinsing treatments with various solutions for 2 h, including (c) H₂O, (d) HCl, (e) HNO₃, (f) H₂SO₄, and (g) (NH₄)₂S₂O₈.

Optical microscopy (OM) offers a straightforward and rapid method to evaluate the uniformity and cleanliness of graphene films. Fig. 1b–g present digital photographs and OM images of graphene films transferred onto SiO₂/Si substrates following treatment with the aforementioned rinsing solutions for 2 h. PMMA/graphene films rinsed exclusively with DI water exhibit significant amounts of dark-colored residues on the graphene surface, indicating the incomplete removal of PMMA. In stark contrast, samples subjected to acid rinsing, including HCl, HNO₃, or H₂SO₄, exhibit intact graphene films with substantially enhanced surface cleanliness, as evidenced by the absence of residual polymer particles in both high- and low-magnification OM images (Fig. S2). Treatment with (NH₄)₂S₂O₈ solution, however, leads to an intermediate level of residual PMMA contamination and a higher tendency for graphene breakage. Besides, graphene wrinkles are consistently observed across all samples, which can be attributed to incomplete stress relaxation and interfacial interactions between graphene and the substrate, originating from the CVD growth and subsequent transfer process.^11,29,30

Building on these findings, we sought to further investigate the driving factors behind the observed outcomes. PMMA/graphene films represent an essential but often overlooked independent intermediate stage during the whole transfer process, existing after the growth substrate etching and before being transferred to target substrates. Fig. 2a illustrates a stamping-based transfer process for flipping PMMA/graphene films onto bare silicon (Si) substrates, inspired by conventional techniques for transferring 2D metal–organic framework (MOF) films at the gas-liquid interface.^31,32 The stamping method offers remarkable advantages such as low substrate requirements, simple operation, high transfer success rate, and good compatibility with large-area film transfer. The detailed transfer procedure is described in the Experimental section (SI), and photographs of the transfer process are also given, as shown in Fig. S3. Briefly, after Cu etching, the resulting PMMA/graphene film was first transferred to DI water for basic rinsing, and then moved to the designated solution for the same rinsing treatment as before the transfer of graphene. Subsequently, a Si substrate was brought into vertical contact with the upper surface of the floating film using a vacuum sucking pen attached to its backside. The film adhered naturally to the substrate surface due to electrostatic interactions (Fig. S4). To ensure film adhesion, Si wafers were used as received without further cleaning, as cleaning renders the surface hydrophilic and impedes adhesion (Fig. S3). Upon completion of the film transfer, the sides that have been cleaned with different solutions are exposed, facilitating subsequent detailed characterization.

Fig. 2 (a) Schematic illustration of the transfer procedure used for flipping the PMMA/graphene films onto bare Si substrates. OM, SEM, and AFM images of the flipped PMMA/graphene films after rinsing with (b), (g) and (l) H₂O; (c), (h) and (m) HCl; (d), (i) and (n) HNO₃; (e), (j) and (o) H₂SO₄; and (f), (k) and (p) (NH₄)₂S₂O₈ for 2 h.

To compare the morphology and structural characteristics of the as-flipped PMMA/graphene films obtained using different rinsing solutions, a series of surface characterization techniques, including OM, SEM, and AFM, was employed. It should be noted that due to the poor conductivity of the flipped films, the samples were treated with a gold coating before SEM analysis. As shown in Fig. 2b–p, obvious clumpy contaminant aggregates are observed on H₂O-rinsed flipped samples, whereas films treated with HCl, HNO₃, or H₂SO₄ exhibit negligible residual contamination. Although treatment with (NH₄)₂S₂O₈ effectively reduces surface contaminants, its cleaning efficacy is still inferior to that of the three strong acid solutions mentioned above, possibly because of its significantly weaker acidity. Based on the observation above, it is evident that under identical rinsing conditions, the distribution of residual contaminants on these surfaces closely resembled the pattern of residual PMMA on the graphene surface from the previous transfer. Interestingly, H₂O-rinsed graphene samples occasionally exhibited pronounced regional variations in the distribution of PMMA residues, whereas acid-rinsed samples consistently demonstrated uniform cleanliness across the graphene surfaces (Fig. S5). This phenomenon is likely attributable to the heterogeneous distribution of surface contaminants on the PMMA/graphene film before transfer. Additionally, periodic wavy stripes are visible on all the flipped PMMA/graphene films, reflecting a structural feature inherited from the surface topography of the original Cu foil substrate (Fig. S6). To verify the general applicability and reproducibility of the above experimental results, CVD-grown graphene films from another supplier were subjected to the same transfer and rinsing protocols. A comparative analysis of surface contamination levels and graphene cleanliness under standardized experimental conditions revealed consistent and reproducible results, as illustrated by the surface morphology shown in Fig. S7. Therefore, the correlation established between the degree of surface contamination of PMMA/graphene films and the difficulty of PMMA removal provides predictive insights and practical guidance for developing optimized cleaning protocols to achieve high-quality, residue-free graphene film transfer.

To elucidate the impact of surface contaminants on the ultimate removal of PMMA residues, the chemical composition and formation mechanisms of these contaminants were initially investigated. By tracking the temporal evolution of these contaminants, it was found that a large number of surface contaminants were already present on the PMMA/graphene surface immediately after Cu etching, indicating that their formation occurred during or prior to this processing step, as shown in Fig. S8. As a result, two primary mechanisms can be proposed to rationalize the origin of these contaminants. Firstly, PMMA is susceptible to undergo chemical modification when exposed to the acidic FeCl₃ etchant, resulting in the formation of filamentous or flocculent byproducts that subsequently deposit onto the graphene surface.^23,33 Secondly, the contamination may arise intrinsically from amorphous carbon residues generated during the CVD graphene synthesis process or extrinsically from impurities introduced onto the Cu foil surface before the substrate etching process.^34,35

Raman spectroscopy, a powerful and non-destructive analytical method, was used to characterize the chemical structures of residual contaminants on the surface of the flipped PMMA/graphene films. Fig. 3a–c show representative OM images and Raman spectra collected from the specific regions outlined by white boxes in Fig. 3a and b. The Raman spectrum obtained from a clean area of the flipped film (Position 1, Fig. 3a) exhibits distinct spectral features that are typical of the combination of monolayer graphene and pristine PMMA. In comparison, the spectrum acquired from the contaminated region (Position 2, Fig. 3a) shows a pronounced increase in both the intensity and spectral width of the G band. Also, this is accompanied by a prominent D band with comparable intensity but greater spectral broadening, as well as a noticeably diminished 2D peak. These distinctive Raman signatures unequivocally exclude pristine PMMA as the primary source of these aggregated contaminants, as the spectral features markedly differ from those observed in regions exhibiting clear coexistence of PMMA and graphene (Positions 1 and 3). Instead, the results strongly indicate the presence of carbon-based contaminants, likely amorphous carbon, and align well with previously reported Raman data in the literature.^36–38 These findings provide compelling evidence that aggregated carbon-based species constitute the predominant component of the flocculent contaminants observed on the PMMA/graphene surface following the Cu etching process.

Fig. 3 OM images of (a) a flipped PMMA/graphene film on a Si substrate after rinsing with DI water for 2 h, and (b) a graphene film transferred onto a SiO₂/Si substrate using the same rinsing treatment. (c) Comparison of the Raman spectrum of pristine PMMA with those obtained from the specific regions highlighted in (a) and (b). (d) XPS survey spectra of flipped PMMA/graphene films subjected to different solution treatments for 2 h, including H₂O, HCl, HNO₃, H₂SO₄, and (NH₄)₂S₂O₈. High-resolution C1s XPS spectra (e) of the flipped PMMA/graphene film rinsed with DI water, and Cu2p spectra (f) for flipped PMMA/graphene films rinsed with all the solutions mentioned above.

X-ray photoelectron spectroscopy (XPS) was used to investigate the surface chemical states and elemental composition of the flipped PMMA/graphene films. As illustrated in Fig. 3d, the survey spectra for all samples exhibit distinct peaks at binding energies of ∼285, 532, 400 and 975 eV due to C1s, O1s, N1s and O KLL, respectively, and the weak Si2p, Si2s, Cl2p and F1s, F KLL and Cu2p peaks are also obtained at binding energies of ∼100, 150, 200, 687, 834 and 933 eV, respectively.^39–41 The detected Si and F species can be attributed to the transferred Si wafers that were used as-received without further cleaning, as mentioned in the previous film transfer process. Additionally, the presence of N element is likely due to residual traces of ammonia remaining after the final DI water rinse, as well as N₂ gas adsorbed on the film surface. In the deconvoluted C1s spectrum for the selected film rinsed with DI water (Fig. 3e), five carbon peaks, at 284.26 (C–C, sp2), 284.86 (C–C/C–H), 285.42 (C*–COO), 286.67 (C–O), and 288.78 eV (O–C=O) are detected, which correspond to the composite characteristics of graphene structure and functional groups of the PMMA molecule.^15,25,42 There are two relatively weak peaks at about 933 and 953 eV in the high-resolution Cu2p XPS spectra, assigned to Cu2p_3/2 and Cu2p_1/2 spin-orbit components, respectively, observed for all samples (Fig. 3f).^43,44 This result further reflects the residual trace amounts of Cu element on the film surface, and it mainly exists in the Cu²⁺ oxidation state, as evidenced by the weak shake-up satellite peaks at binding energies of 937–945 eV. The observed Cu signals likely arise from a small quantity of copper oxide particles on the Cu foil of the growth substrate, as shown in Fig. S6. It is worth noting that no Fe2p signals are detected in any of the films after rinsing with their respective solutions for 2 h. This indicates that under the current rinsing protocols, even with DI water, Fe contamination from the etching (FeCl₃) solution can be effectively removed (Fig. S9), which agrees with the elemental distribution profiles obtained from the energy dispersive X-ray spectroscopy (EDX) mapping (Fig. S11). Minor Cl2p peaks were also consistently detected in the XPS analysis. Given the uniformity of the Cl signal across all samples, regardless of the transfer method or processing conditions, we propose that it may originate from the substrate contribution or reagent residues. The presence of trace Cl does not influence the conclusions regarding the chemical states and composition of the transferred flipped PMMA/graphene layers. Analysis of the experimental data indicates that the surface elemental composition of the flipped films has a limited impact on the effectiveness of PMMA removal, as the elemental contents are both low and largely consistent across samples treated with different rinsing solutions. In contrast, the pronounced differences in the concentration of non-metallic carbon-based contaminants at the PMMA/graphene surface are more likely to play a pivotal role in determining the efficiency of PMMA removal.

To further investigate the potential influence of surface properties regulated by various rinsing solutions on PMMA removal efficiency, contact angle measurements were performed using the sessile drop method. Fig. S12 presents contact angle images of two batches of flipped PMMA/graphene films after rinsing with H₂O, HCl, HNO₃, H₂SO₄, and (NH₄)₂S₂O₈ for 2 h. The measured static contact angle of water on these rinsed films ranged from 60° to 80°, suggesting a moderately hydrophilic surface. Compared with samples rinsed with DI water, those treated with acid solutions and ammonium persulfate showed a slight decrease in contact angle values by 2–6° across two different batches. As mentioned above, the major contaminants initially present on the PMMA/graphene surface were cleansed to different extents by various acidic solutions. The subtle changes in hydrophilic/hydrophobic properties, whether resulting from the cleaned surface condition or other factors, can be considered beneficial for the subsequent removal of PMMA after transfer to the desired substrates. Furthermore, the structural integrity of the PMMA films after exposure to these solutions was evaluated by analyzing chemical bonds using Fourier-transform infrared (FT-IR) spectroscopy. In Fig. S13, the transmission peaks at 2948 and 2841 cm⁻¹ correspond to C–H stretching modes. The bands at 1431 and 1382 cm⁻¹ are attributed to the asymmetric and symmetric bending vibrations of C–H, respectively. The characteristic C=O stretching vibration of the ester group from the PMMA skeleton is observed at 1726 cm⁻¹.⁴⁵ Additionally, the doublet peaks at 1144 and 1189 cm⁻¹ correspond to the C–O stretching vibrations of the ester groups. Under all conditions, the spectra exhibited consistent peak profiles with no noticeable differences among samples subjected to different rinsing treatments, indicating that the overall structural integrity of the PMMA films was maintained throughout the cleaning process.

In consideration of the experimental results and theoretical analysis outlined above, the role of acid/alkali rinsing in enhancing the removal of PMMA residues can be attributed to the enhanced surface cleanliness between the graphene and the substrate by minimizing contaminants, particularly key carbon-based impurities, as well as preventing graphene film breakage. In other words, acid cleaning treatment can effectively promote the elimination of surface contaminants and residual metal impurities from the PMMA/graphene surface, thereby reducing the strong adhesion between PMMA and graphene films, and enabling the complete removal of PMMA residues by organic solvents such as acetone.

The structural and surface quality of graphene films transferred onto SiO₂/Si substrates following treatment with various cleaning solutions were systematically evaluated and compared using SEM, AFM, and Raman spectroscopy. Graphene films transferred using acid solutions such as HCl showed a flat and clean surface free of apparent residual polymer particles, in clear contrast to those transferred using DI water rinsing, as revealed by SEM and AFM images in (Fig. 4a–d). A step height of approximately 1.0 nm was observed between the transferred graphene and the underlying substrate (Fig. 4e and f). This thickness value falls within the typical range for CVD monolayer graphene as measured by AFM and agrees well with previously reported results.^28,30,46,47 Fig. 4g displays representative Raman spectra from five randomly selected regions of transferred graphene samples under all transfer conditions, showing well-defined G (∼1587 cm⁻¹) and 2D (∼2682 cm⁻¹) peaks, along with a weak D peak (∼1345 cm⁻¹). Further Raman evidence for the monolayer nature of the transferred graphene comes from the characteristic single Lorentzian shape of the 2D band, along with a G to 2D intensity ratio (I_G/I_2D) typically not exceeding 0.5.^48–50 To evaluate and compare the overall structural disorder and defect level of the graphene film, the integrated intensity ratio of the D to G peaks (I_D/I_G) was quantitatively analyzed, as shown in Fig. 4h. All samples have relatively low I_D/I_G ratios, indicating the high quality of the transferred graphene films. Graphene samples subjected to acid or ammonium persulphate treatments were found to exhibit I_D/I_G ratios comparable to, or even slightly lower than those rinsed with DI water. This result therefore proves that the cleaning protocols employed in this study do not generate significant additional defects into the graphene lattice, and the overall structural quality of the graphene film is maintained. It should be noted that the I_G/I_2D ratio serves as a crucial indicator, not only for verifying the monolayer nature of graphene but also for assessing the degree of doping, which is evidenced by the concomitant increase in the G peak intensity and decrease in the 2D peak intensity.^51,52 Fig. 4i shows higher I_G/I_2D ratios for graphene samples rinsed with HCl, HNO₃, H₂SO₄, and (NH₄)₂S₂O₈ compared to H₂O-rinsed controls, with the most pronounced increase observed in samples treated with (NH₄)₂S₂O₈. Due to the superior surface cleanliness and substantially reduced PMMA residues observed in graphene samples obtained via the latter four transfer methods (Fig. 1), we consider that the differences in Raman spectra among the different samples can primarily be attributed to the adsorption doping by H⁺ ions or other electron acceptors during transfer, rather than p-type doping caused by the adsorbed PMMA. Further evidence is provided by the G band, which is highly sensitive to both electron (n-type) and hole (p-type) doping, typically exhibiting a noticeable blue shift and decreased peak width. The changes in doped graphene are caused by a nonadiabatic removal of the Kohn anomaly at the Γ point and the suppression of phonon decay via electron-hole pair generation.^50,53 A comparison of the statistical data of the G band position and FWHM (G) of these graphene samples is presented in Fig. S14a and b. Across the rinsing sequence from H₂O to (NH₄)₂S₂O₈, the G band position exhibits a systematic blue shift (from an average of 1586 to 1592 cm⁻¹) along with peak narrowing (from an average of 25.5 to 17.9 cm⁻¹), indicating progressively increasing doping levels. Importantly, doping in graphene results in a shift in the Fermi level relative to the Dirac point, directly modulating the carrier concentration and thereby playing a critical role in tuning its electronic properties. Accordingly, for the clean graphene obtained under the three acid washing conditions, samples treated with sulfuric acid are expected to have superior hole mobilities, along with a more positive Dirac point voltage (V_Dirac), as a result of their enhanced carrier concentration under higher levels of p-type doping. Raman spectroscopy serves as one of the most powerful techniques to study both doping and mechanical stress in graphene. In strain studies of graphene, the 2D peak is generally considered more sensitive and reliable than the G peak, owing to its larger shift rate and reduced susceptibility to doping effects.^54,55 The comparison of Fig. S14c and S14d indicates that, compared to DI water-rinsed samples, those treated with acid and ammonium persulfate exhibit a blue shift of the 2D Raman peak from an average position of 2675 cm⁻¹ to 2683 cm⁻¹, with a concurrent decrease in peak width from 48.1 to 42.4 cm⁻¹. During transfer, the high thermal expansion coefficient of PMMA induces tensile strain in graphene upon cooling, resulting in a red-shifted 2D peak, as observed in H₂O-rinsed samples. Complete removal of PMMA, as achieved with acid rinsing, allows the graphene to relax to its intrinsic state, releasing tensile strain and producing a blue shift in the 2D peak. Analysis of the FWHM of the 2G peak further supports this explanation: the H₂O-treated sample exhibits the largest FWHM, indicative of lattice disturbances and non-uniform strain, while the acid-treated samples show a reduced FWHM, consistent with improved lattice uniformity and efficient strain release. Samples treated with (NH₄)₂S₂O₈ exhibit intermediate values, reflecting its moderate efficacy in PMMA removal, as indicated by both the 2D peak position and FWHM. It is concluded that effective removal of PMMA facilitates the relaxation and uniformity of strain in graphene, and the relaxation of strain observed by Raman spectroscopy, in turn, further confirms that the acidic solutions used are favorable for PMMA removal and achieving cleaner graphene films.

Fig. 4 Morphological and spectroscopic characterization of graphene films transferred on SiO₂/Si substrates, after rinsing with different solutions for 2 h during the wet transfer process. SEM and AFM images of graphene films transferred after rinsing with (a) and (b) H₂O, and (c) and (d) HCl solution. (e) AFM image of the graphene film rinsed with HCl, accompanied by the corresponding height profile (f) of the white dotted line in (e). (g) Representative Raman spectra and statistical distribution of (h) integrated I_D/I_G and (i) I_G/I_2D of graphene films rinsed with H₂O, HCl, HNO₃, H₂SO₄, and (NH₄)₂S₂O₈.

The electrical properties of the graphene films were studied by fabricating graphene field-effect transistors (GFETs) to evaluate the effects of different rinsing solutions used during the transfer process. Fig. 5a presents the schematic device structure of GFETs with a bottom-gate, top-contact configuration, and the corresponding fabrication process is schematically illustrated in Fig. S15a. Firstly, the transferred graphene on the SiO₂/Si substrate was patterned into a 10 × 10 array through photolithography using custom-designed photomasks (Fig. S15b). Next, the samples were transferred to an N₂-filled glove box, where the fabrication was completed by depositing gold electrodes onto the graphene layer via thermal evaporation using a shadow mask (Fig. S15c). The digital photograph and OM image of as-fabricated GFETs are presented in Fig. S15d and e, respectively. Fig. 5b shows the transfer characteristics of GFETs fabricated using graphene prepared by the conventional transfer method, as well as by the acid-rinsed and ammonium persulfate-rinsed methods. A gate voltage (V_GS) sweep was performed from −30 V to +100 V with a constant drain-source bias (V_DS) of 0.1 V. Graphene films treated with acid solutions exhibited relatively superior electrical performance, characterized by a more symmetric transfer curve, indicative of enhanced and balanced electron and hole transport. Furthermore, the corresponding output characteristics and the quantitative ratio of hole to electron mobilities (μ_h/μ_e) of the GFETs provided additional confirmation of this observation, as given in Fig. S16 and S17. This can be explained by the impurity doping, local charge inhomogeneity from significant residual PMMA on H₂O and (NH₄)₂S₂O₈-rinsed graphene samples (Fig. 1), and enhanced electrical asymmetry due to charge scattering by residues and the varying effects of surface potential traps on carrier mobility. Statistical comparisons and analyses of hole mobility and Dirac point voltage, extracted from 45 individual devices under each rinsing condition, were conducted. As expected from the surface morphology and Raman analyses, most devices fabricated using the acid cleaning method including HCl, HNO₃ and H₂SO₄ exhibit room-temperature mobilities in the range of 1200–1700 cm² V⁻¹ s⁻¹, approximately 2–3 times higher than those of devices fabricated with graphene transferred via the conventional H₂O rinsing method as shown in Fig. 5c. Among these, higher overall device mobility and a maximum value of 1772.1 cm² V⁻¹ s⁻¹ were obtained in samples cleaned with H₂SO₄, which could be due to the increased p-type doping concentration as evidenced by a large average Dirac voltage in Fig. 5d. This explanation is consistent with the Raman analysis in Fig. 3, which implies that in addition to visible PMMA residues, the doping effects caused by invisible and even inevitable impurities should not be overlooked during the transfer process of CVD graphene. Further, this measured mobility represents intermediate performance, remaining lower than some previously reported values for high-mobility graphene devices.^16,20,28,46 The primary limiting factors likely include: (i) high contact resistance due to mismatches in work function between graphene and gold, (ii) charge carrier scattering arising from wrinkles, cracks, and other defects in graphene and (iii) substrate-mediated charge impurity scattering. It is believed that by addressing the above factors, the carrier mobility of transferred CVD graphene devices can be significantly improved. Moreover, a moderate enhancement in the device performance was observed in the GFETs fabricated from graphene rinsed with (NH₄)₂S₂O₈ in terms of mobility, Dirac voltage and symmetry in charge carrier transport. This is attributed to moderate polymer residues and surface breakage in the transferred graphene. Thus, the acid rinsing step proves to be a superior approach, markedly improving both the surface quality of graphene and the overall performance of the resulting devices.

Fig. 5 Electrical characterization of GFETs fabricated from graphene films subjected to various rinsing treatments, including H₂O, HCl, HNO₃, H₂SO₄, and (NH₄)₂S₂O₈ solutions. (a) Schematic diagram of the device structure of the GFETs. (b) Transfer characteristic curves of GFETs. Statistical distribution of (c) field-effect mobility and (d) Dirac point voltage (V_Dirac) extracted from 45 GFET devices.

Conclusions

In summary, we have systematically examined the surface conditions on the backside of the PMMA/graphene films using a straightforward stamping-transfer technique. A significant number of visible flocculent contaminants remaining on the PMMA/graphene surface after Cu etching are identified using optical microscopy. Raman spectroscopic analysis reveals that carbon-based impurities, primarily in the form of aggregated amorphous carbon, are the main components of the residual contaminants observed on the PMMA/graphene surface. The introduction of an acid/alkali cleaning step has proven effective in improving surface cleanliness between the graphene and the substrate by removing these key residual contaminants, resulting in a clean graphene surface free of polymer residues. The efficacy of the surface cleanliness evaluation approach, based on examining the backside surface conditions of the flipped PMMA/graphene films after treatment with different rinsing solutions, is further validated across multiple graphene samples. Impressively, graphene films treated with acid solutions exhibit significantly enhanced electrical properties, achieving 2–3 times higher carrier mobility and improved symmetry in charge-carrier transport compared to control samples prepared using the conventional transfer method. Among graphene samples subjected to acid solution treatments, those rinsed with sulfuric acid demonstrate superior hole mobilities and a more positive Dirac point voltage, attributed to increased carrier concentrations induced by enhanced p-type doping. These findings provide robust experimental and theoretical validation for the effectiveness of the acid-assisted graphene transfer process in achieving an intact, clean graphene surface free of PMMA residues. Consequently, this study offers a promising and practical strategy for optimizing the fabrication process and enhancing the performance of graphene-based electronic and optoelectronic devices.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article are included in the SI. See DOI: https://doi.org/10.1039/d5tc02259b.

Acknowledgements

The authors thank the National Natural Science Foundation of China (52273174, 52350058 and 22375059), Special Funds for the Construction of the Science and Technology Project of Natural Science Foundation of Hunan Province (2023JJ10014, 2024RC1027, 2024JJ6330, and 2024JJ4013), Key Laboratory of Optoelectronic Information Technology, Ministry of Education, Tianjin University (2024KFKT012) for financial support. We are also grateful to the Analytical Instrumentation Center of Hunan University for their support in sample preparation, data acquisition, and analysis of SEM and AFM measurements.

References

1. L. Liao and X. Duan, Mater. Sci. Eng., R, 2010, 70, 354–370.
2. K. S. Novoselov, V. I. Fal’ko, L. Colombo, P. R. Gellert, M. G. Schwab and K. Kim, Nature, 2012, 490, 192–200.
3. M. Wang, M. Huang, D. Luo, Y. Li, M. Choe, W. K. Seong, M. Kim, S. Jin, M. Wang, S. Chatterjee, Y. Kwon, Z. Lee and R. S. Ruoff, Nature, 2021, 596, 519–524.
4. W. Yin, C. Liu and J. Li, Energy Environ. Mater., 2025, 8, e70009.
5. M. Chen, R. C. Haddon, R. Yan and E. Bekyarova, Mater. Horiz., 2017, 4, 1054–1063.
6. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
7. C. Berger, Z. Song, X. Li, X. Wu, N. Brown, C. Naud, D. Mayou, T. Li, J. Hass, A. N. Marchenkov, E. H. Conrad, P. N. First and W. A. de Heer, Science, 2006, 312, 1191–1196.
8. S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen and R. S. Ruoff, Carbon, 2007, 45, 1558–1565.
9. K. V. Emtsev, A. Bostwick, K. Horn, J. Jobst, G. L. Kellogg, L. Ley, J. L. McChesney, T. Ohta, S. A. Reshanov, J. Röhrl, E. Rotenberg, A. K. Schmid, D. Waldmann, H. B. Weber and T. Seyller, Nat. Mater., 2009, 8, 203–207.
10. W. Yang, G. Chen, Z. Shi, C. C. Liu, L. Zhang, G. Xie, M. Cheng, D. Wang, R. Yang, D. Shi, K. Watanabe, T. Taniguchi, Y. Yao, Y. Zhang and G. Zhang, Nat. Mater., 2013, 12, 792–797.
11. X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo and R. S. Ruoff, Science, 2009, 324, 1312–1314.
12. X. Li, L. Colombo and R. S. Ruoff, Adv. Mater., 2016, 28, 6247–6252.
13. J. Plutnar, M. Pumera and Z. Sofer, J. Mater. Chem. C, 2018, 6, 6082–6101.
14. J. W. Suk, A. Kitt, C. W. Magnuson, Y. Hao, S. Ahmed, J. An, A. K. Swan, B. B. Goldberg and R. S. Ruoff, ACS Nano, 2011, 5, 6916–6924.
15. A. Pirkle, J. Chan, A. Venugopal, D. Hinojos, C. W. Magnuson, S. McDonnell, L. Colombo, E. M. Vogel, R. S. Ruoff and R. M. Wallace, Appl. Phys. Lett., 2011, 99, 122108.
16. J. W. Suk, W. H. Lee, J. Lee, H. Chou, R. D. Piner, Y. Hao, D. Akinwande and R. S. Ruoff, Nano Lett., 2013, 13, 1462–1467.
17. S. Ullah, X. Yang, H. Q. Ta, M. Hasan, A. Bachmatiuk, K. Tokarska, B. Trzebicka, L. Fu and M. H. Rummeli, Nano Res., 2021, 14, 3756–3772.
18. L. P. Ma, W. Ren and H. M. Cheng, Small Methods, 2019, 3, 1900049.
19. N. Peltekis, S. Kumar, N. McEvoy, K. Lee, A. Weidlich and G. S. Duesberg, Carbon, 2012, 50, 395–403.
20. Y. Jia, X. Gong, P. Peng, Z. Wang, Z. Tian, L. Ren, Y. Fu and H. Zhang, Nano-Micro Lett., 2016, 8, 336–346.
21. B. H. Son, H. S. Kim, H. Jeong, J. Y. Park, S. Lee and Y. H. Ahn, Sci. Rep., 2017, 7, 18058.
22. T. T. Pham, Q. H. Do, T. K. V. Ngo and R. Sporken, Mater. Today Commun., 2019, 18, 184–190.
23. B. Zhuang, S. Li, S. Li and J. Yin, Carbon, 2021, 173, 609–636.
24. G. Borin Barin, Y. Song, I. de Fátima Gimenez, A. G. Souza Filho, L. S. Barreto and J. Kong, Carbon, 2015, 84, 82–90.
25. S. Kim, S. Shin, T. Kim, H. Du, M. Song, C. Lee, K. Kim, S. Cho, D. H. Seo and S. Seo, Carbon, 2016, 98, 352–357.
26. C. D. Liao, A. Capasso, T. Queirós, T. Domingues, F. Cerqueira, N. Nicoara, J. Borme, P. Freitas and P. Alpuim, Beilstein J. Nanotechnol., 2022, 13, 796–806.
27. X. Liang, B. A. Sperling, I. Calizo, G. Cheng, C. A. Hacker, Q. Zhang, Y. Obeng, K. Yan, H. Peng, Q. Li, X. Zhu, H. Yuan, A. R. Hight Walker, Z. Liu, L. M. Peng and C. A. Richter, ACS Nano, 2011, 5, 9144–9153.
28. W. Hou, Y. Xu, Y. Zhang, X. H. Yao and Y. Xu, Carbon, 2023, 204, 427–433.
29. H. Hattab, A. T. N′Diaye, D. Wall, C. Klein, G. Jnawali, J. Coraux, C. Busse, R. van Gastel, B. Poelsema, T. Michely, F. J. Meyer zu Heringdorf and M. Horn-von Hoegen, Nano Lett., 2012, 12, 678–682.
30. Y. Zhao, Y. Song, Z. Hu, W. Wang, Z. Chang, Y. Zhang, Q. Lu, H. Wu, J. Liao, W. Zou, X. Gao, K. Jia, L. Zhuo, J. Hu, Q. Xie, R. Zhang, X. Wang, L. Sun, F. Li, L. Zheng, M. Wang, J. Yang, B. Mao, T. Fang, F. Wang, H. Zhong, W. Liu, R. Yan, J. Yin, Y. Zhang, Y. Wei, H. Peng, L. Lin and Z. Liu, Nat. Commun., 2022, 13, 4409.
31. G. Wu, J. Huang, Y. Zang, J. He and G. Xu, J. Am. Chem. Soc., 2017, 139, 1360–1363.
32. G. Xu, T. Yamada, K. Otsubo, S. Sakaida and H. Kitagawa, J. Am. Chem. Soc., 2012, 134, 16524–16527.
33. J. Song, F. Y. Kam, R. Q. Png, W. L. Seah, J. M. Zhuo, G. K. Lim, P. K. H. Ho and L. L. Chua, Nat. Nanotechnol., 2013, 8, 356–362.
34. K. Jia, J. Zhang, L. Lin, Z. Li, J. Gao, L. Sun, R. Xue, J. Li, N. Kang, Z. Luo, M. H. Rummeli, H. Peng and Z. Liu, J. Am. Chem. Soc., 2019, 141, 7670–7674.
35. L. Lin, J. Zhang, H. Su, J. Li, L. Sun, Z. Wang, F. Xu, C. Liu, S. Lopatin, Y. Zhu, K. Jia, S. Chen, D. Rui, J. Sun, R. Xue, P. Gao, N. Kang, Y. Han, H. Q. Xu, Y. Cao, K. S. Novoselov, Z. Tian, B. Ren, H. Peng and Z. Liu, Nat. Commun., 2019, 10, 1912.
36. J. Robertson, Adv. Phys., 1986, 35, 317–374.
37. C. Schünemann, F. Schäffel, A. Bachmatiuk, U. Queitsch, M. Sparing, B. Rellinghaus, K. Lafdi, L. Schultz, B. Büchner and M. H. Rümmeli, ACS Nano, 2011, 5, 8928–8934.
38. C. H. An Wong, A. Ambrosi and M. Pumera, Nanoscale, 2012, 4, 4972–4977.
39. M. Mathews, A. Graves, D. Boris, S. Walton and C. Stinespring, J. Appl. Phys., 2024, 135, 153301.
40. S. Feng, Z. Mu, H. Liu, J. Huang, X. Li and Y. Yang, J. Fluoresc., 2019, 29, 1133–1141.
41. M. P. Tran, P. Gonzalez-Aguirre, C. Beitia, J. Lundgren, S. I. Moon and H. Fontaine, Microelectron. Eng., 2019, 207, 1–6.
42. H. Piao, N. Fairley and J. Walton, Surf. Interface Anal., 2013, 45, 1742–1750.
43. F. Zhou, Z. Li, G. J. Shenoy, L. Li and H. Liu, ACS Nano, 2013, 7, 6939–6947.
44. S. Y. Huang, B. Zhao, K. Zhang, M. M. F. Yuen, J. B. Xu, X. Z. Fu, R. Sun and C. P. Wong, Sci. Rep., 2015, 5, 14260.
45. C. Zhang, Y. Zhou, T. Shao, Q. Xie, J. Xu and W. Yang, Appl. Surf.Sci., 2014, 311, 468–477.
46. B. Wang, M. Huang, L. Tao, S. H. Lee, A. R. Jang, B. W. Li, H. S. Shin, D. Akinwande and R. S. Ruoff, ACS Nano, 2016, 10, 1404–1410.
47. M. Wang, L. Fu, L. Gan, C. Zhang, M. Rümmeli, A. Bachmatiuk, K. Huang, Y. Fang and Z. Liu, Sci. Rep., 2013, 3, 1238.
48. A. C. Ferrari and D. M. Basko, Nat. Nanotechnol., 2013, 8, 235–246.
49. S. Dai, M. Xie, C. Wang, Y. Wang, B. Han, S. Xu, K. Wang, B. A.Zhuravlova, L. Chi, W. Tian, P. Samorì and Z. Liu, J. Mater. Chem. C, 2024, 12, 14667–14674.
50. M. Kalbac, A. Reina-Cecco, H. Farhat, J. Kong, L. Kavan and M. S.Dresselhaus, ACS Nano, 2010, 4, 6055–6063.
51. D. J. Jang, M. M. Haidari, J. H. Kim, J. Y. Ko, Y. Yi and J. S. Choi, Nanomaterials, 2023, 13, 1494.
52. H. Van Ngoc, Y. Qian, S. K. Han and D. J. Kang, Sci. Rep., 2016, 6, 33096.
53. A. Das, S. Pisana, B. Chakraborty, S. Piscanec, S. K. Saha, U. V. Waghmare, K. S. Novoselov, H. R. Krishnamurthy, A. K. Geim, A. C. Ferrari and A. K. Sood, Nat. Nanotechnol., 2008, 3, 210–215.
54. E. del Corro, L. Kavan, M. Kalbac and O. Frank, J. Phys. Chem. C, 2015, 119, 25651–25656.
55. M. Huang, H. Yan, T. F. Heinz and J. Hone, Nano Lett., 2010, 10, 4074–4079.

---