Two-dimensional (2D) materials, including graphene [1], WSe₂, ReS₂ [2], and MoS₂, have attracted considerable attention owing to their distinctive electrical, optical, and mechanical characteristics. The- se properties make 2D materials, such as graphene valuable for applications in radio frequency (RF) devices, field-effect transistors (FETs), and photodetectors [3–5]. Moreover, the chemical vapor deposition (CVD) growth method facilitates the large-scale synthesis of 2D materials [6], providing controlled modulation of thickness and electronic characteristics [7–9]. Nevertheless, achieving precise control of these properties remains a persistent challenge. Suitable doping methods are required to control the electrical properties of graphene. However, conventional techniques often face obstacles as they may cause substantial damage to the graphene lattice structure during highenergy processes [10].

The O₂ plasma treatments disrupt the C–C bonds in graphene, leading to the formation of various functional groups such as epoxy (-O-) and carboxyl (-COOH). These functional groups can transform the graphene surface from hydrophobic to hydrophilic because of their stronger interactions with water molecules (hydrogen bonding) than with carbon atoms (van der Waals bonding) [11]. Consequently, the pristine sp² structure of graphene undergoes transformation into the sp³ structure. This structural alteration manifests as observable changes in the lattice structure and electrical properties of graphene [12]. Thermal energy can eliminate functional groups and convert graphene to a hydrophobic state [13].

The primary objective of this study was to elucidate the correlation between this transition and the electrical properties of graphene. We employed low-energy O₂ plasma treatment to facilitate the doping of graphene, enabling control over its electrical properties [14,15]. We aimed to tailor the lattice structure and electrical characteristics of graphene using Raman spectroscopy and current-voltage (I−V) measurements [16,17]. In our initial experiments, we employed a remote plasma treatment method to observe the variations in the Raman spectra and I−V curves over different processing durations [18]. Subsequently, we transitioned to low-energy reactive ion etching (RIE) for comparison with the remote treatment approach. Additionally, we investigated the effect of thermal treatment on the restoration of the plasma-treated graphene to its pristine state. Because the functional groups are created by O₂ plasma, rapid thermal annealing (RTA) was performed through post-treatment to restore the defects and electrical properties.

Our results indicate that the lattice structure of graphene undergoes a gradual transition from sp² to sp³ orbitals as a function of the duration of plasma processing and the specific plasma modes employed. Furthermore, the restorative effects of the heat treatment were confirmed by observing changes in the electrical characteristics of the plasma-treated graphene. This study contributes significant developments and insights into the manipulation of lattice structures and electrical properties of 2D materials for their potential applications.

2.1. Graphene growth

A Cu foil (Alfa Aesar, 99.8 %) was subjected to pre-treatment with an acetic acid solution at 60 °C for 10 min followed by washing with deionized (DI) water. Then, the foil was heated to a graphene synthesis temperature of 1,030 °C and annealed for 3 h at a rate of 100 sccm (standard cubic centimeters per minute) of mixed gases, including 10 % H₂ and 90 % Ar to remove native oxides from the Cu foil and enlarge its grains. For graphene growth, 1 sccm of methane (CH₄) gas and 100 sccm of a mixed gas (10 % H₂ and 90 % Ar) were introduced into a hotwall tubular inductive coupled plasma-CVD reaction chamber. The reactor pressure was maintained at 1 Torr for 20 min at an RF power of 200 W. The graphene synthesized on Cu was transferred to the target substrate using the conventional wet-transfer method, as described in the following section.

2.2. Graphene transfer

(1) The transfer process employed polymethyl methacrylate (PMM-A) [19] as the transfer film, which is a widely used method for transferring 2D materials [20] onto SiO₂/Si substrates, as shown in Fig. 1(a). (2) Initially, PMMA was applied to the CVD-grown graphene on a Cu foil using a spin coater. The spin coating process involved two steps: 400 rpm for 10 s, followed by 6,000 rpm for 1 min. Subsequently, the PMMA-coated graphene/Cu foil underwent heating at 90°C for 1 min, after which it was placed on an etchant (ammonium persulfate) to selectively remove the Cu foil. Throughout this transfer process, the PMMA protected the graphene from physical damage. (3) Following the removal of the Cu foil, the PMMA/graphene film was rinsed four times with DI water to eliminate the residual etchant. The film floating on DI water was then carefully transferred onto a target substrate (SiO₂/Si). Subsequently, the substrate with the transferred film was air dried for more than 12 h. (4) Subsequently, the PMMA/graphene/SiO₂ substrate was immersed in a solvent (acetone) to dissolve and remove the PMMA layer, followed by rinsing with isopropyl alcohol (IPA) and DI water. (5) Finally, the graphene was successfully transferred onto the SiO₂/Si substrate, followed by the formation of silver electrodes at the corners, as shown in Fig. 1(b).

Figure 1. (a) Schematic of the transfer process of graphene on SiO₂/Si substrate using wet transfer method with PMMA as a transfer film. (b) Schematic and photographic images of the transferred graphene on SiO₂ with silver electrodes at the corners.

2.3. Plasma treatments

After graphene transfer, O₂ plasma treatments were performed using two distinct modes, remote and RIE, both employing the same bias power of 20 W. The flow rate and processing duration were set at 10 sccm and between 1 and 4 s, respectively, as shown in Fig. 2. This structural alteration manifested in observable changes in the lattice structure and electrical properties of graphene, as discerned through I_dV_g and I_dV_d characteristics, as well as Raman spectroscopy.

Figure 2. Changes in the lattice structure of graphene after O₂ plasma treatment.

2.4. Electrical measurements

Three electrodes–the source, drain, and gate–are essential to assess the electrical properties of graphene FETs. The Dirac point of graphene was observed by applying a current between the source and drain electrodes while applying gate biases to the SiO₂/Si substrate. Under highvacuum conditions, the Dirac point of graphene is at V_g = 0 V; however, under ambient conditions, it can be observed at approximately V_g ≈+30–60 V [21]. We set the measurement conditions to a gate compliance limit of 10⁻⁹ A, gate bias ranging from −60 to 100 V with a step interval of 0.1 V, drain compliance limit of 0.1 A, and drain bias of 1 V. The source and drain electrodes were formed using a lowresistance silver paste, as depicted in Fig. 1(b), and the gate electrodes were connected to the Cu foil under the graphene FET. Changes in the Dirac points of the graphene FETs were analyzed before and after RIE and remote plasma treatments.

For I_dV_d measurements, we set the measurement conditions with a gate compliance limit of 10−9 A, gate bias ranging from −60 to 60 V with 10 V step intervals, drain compliance limit of 0.1 A, and drain bias ranging from −1 to 1 V with 0.05 V intervals. Changes in the resistance of the graphene FET were analyzed before and after the RIE and remote plasma treatments.

2.5. Raman Spectroscopy

Raman characterization was performed using a Raman spectrometer (UniRAM, Micro Raman-PL sample chamber, grating 1200/500). The Raman spectra were accumulated five times, and the shift center value of the Raman mapping was set to 2,099.77 cm⁻¹. The properties of graphene can be confirmed through Raman spectra involving the analysis of three distinct Raman peaks−the G, 2D, and D bands−to evaluate the characteristics of the material [16]. Alterations in the peak positions can be indicative of the effects of plasma treatments, such as doping. The experimental parameters for the Raman measurements were set as follows: laser power of 100 mW, wavelength of 532 nm, signal accumulation of 4, and exposure time of 1 s.

Raman mapping was performed to assess the uniformity of the plasma-treated graphene surfaces. This involved acquiring mapping data by specifying the measurement area and size for Raman signal collection while the stage gradually adjusted its position during laser irradiation. Additionally, 3D images representing the I_2D/I_G ratio, I_D/I_G ratio, full width at half maximum (FWHM), D peak, and 2D peak were generated. The experimental conditions for Raman mapping were as follows: a wavelength of 531 nm and a measurement area of 30 µm × 30 µm.

2.6. RTA

Plasma-treated graphene films exhibited higher resistance states when subjected to a 4-s treatment duration, in contrast to pristine graphene, as shown in the I_dV_d graph. Consequently, the structural and electrical properties of graphene were restored through RTA, facilitating the removal of the functional groups induced by the plasma treatment, as shown in Fig. 3 [22]. The RTA process was conducted at a temperature of 200 °C for 30 min, with initial and final temperatures set at 10 and 200 °C, respectively, and a heating rate of 50 °C per min.

Figure 3. Restoration of the lattice structure in graphene after RTA treatment by removing the functional groups.

3.1. Electrical characteristics

The electrical characteristics of graphene, both before and after various plasma treatments of different durations, were assessed and compared using I_dV_g and I_dV_d curves. In the context of remote plasma treatment, as shown in Fig. 4(a), the Dirac point of graphene shifts toward higher V_g values, indicating p-type doping [23,24]. This p-type doping results from the withdrawal of electrons due to the involvement of highly electronegative oxygen atoms in the functional groups.

Figure 4. (a) I_dV_g and (b) I_dV_d curves for graphene FETs after remote plasma treatments.

Furthermore, with an increase in the plasma treatment time, a gradual shift in the Dirac point toward higher V_g values was evident, as shown in Fig. 4(a). Figure 4(b) illustrates the alteration in the Dirac point following RIE plasma treatment, wherein a similar shift was observed. However, the reduction in drain current was less pronounced for the RIE plasma treatments, as indicated by the I_dV_d curves shown in Fig. 5. The slope of the I_dV_d curves, which is inversely related to resistance, diminished with the application of both plasma treatments, leading to an overall increase in resistance. Notably, remote plasma treatment induced a more rapid decrease in the slope compared to that by RIE plasma treatment, implying that the plasma treatment is more effective in inducing damage to the graphene structure [25].

Figure 5. (a) I_dV_g and (b) I_dV_d curves for graphene FETs after RIE plasma treatments.

3.2. Raman spectroscopy

Raman spectra were obtained, as shown in Fig. 6, to investigate the material properties of graphene with respect to the plasma treatment duration. Before plasma treatment, the Raman peaks of pristine graphene were identified (black spectra). Subsequently, Raman peak measurements were conducted after the remote and RIE plasma treatments for 1–4 s. With increasing plasma treatment duration, both the remote and RIE plasma treatments led to an increase in the D peak and a decrease in the 2D peak. This phenomenon can be attributed to the bonding of oxygen and hydrogen atoms to dangling bonds resulting from the breaking of C–C bonds during the O₂ plasma treatment. These defect-mediated Raman scattering events gave rise to enhanced D-bands. Moreover, the collapse of sp² bonding orbitals between carbon atoms owing to the O₂ plasma treatment results in the formation of sp³ hybridization bonding orbitals [23,24]. Notably, the tendency for an increase in the D band was more pronounced in the case of the remote plasma treatment than in the RIE plasma treatment, which is consistent with the more rapid decrease in the slope of the I_dV_d curves.

Figure 6. Raman spectra of graphene treated by (a) remote and (b) RIE plasma.

Despite the fewer induced defects, the blue-shift in the G peak was more pronounced in the graphene treated with RIE plasma, with a shift of 10.8 cm⁻¹, compared to that of the remote plasma-treated graphene (8.9 cm⁻¹), indicating stronger doping effects in the case of RIE plasma treatment. Figure 7 shows 2D frequency graphs in relation to the G frequency, demonstrating a linear fit with respect to both doping and strain in graphene subjected to both remote and RIE plasma treatments. The graphs reveal the presence of doping effects along with a slight change in strain. The observed strain is likely attributed to functional groups or defects induced by the plasma treatments.

Figure 7. Raman map of 2D vs G-peak frequencies for pristine graphene and remote and RIE treated graphene for 2 and 4 s, respectively. The dotted lines indicate shift in the G and 2D peaks for different carrier density (grey) and strain (purple) values.

3.3. Raman Mapping

Raman mapping was employed to examine the alterations in the graphene surface following plasma treatment, as illustrated in Figs. 8 and 9. The mapping images depicting the ratio reveal notable changes in the C–C bond structure, which is characterized by a reduction in the I_2D/I_G ratio after plasma treatment. Furthermore, FWHM of the 2D peak exhibited an increase and non-uniformity, indicating the degradation in the crystal quality. This observation aligns with our earlier findings from the Raman peak analysis. As evidenced by the mapping images of the D and 2D peaks and I_D/I_G ratio, the color intensity of the D peak increased, whereas that of the 2D peak deepened following plasma treatment. Furthermore, the mapping of the D peak in graphene indicated a notable change in color after remote plasma treatment compared to that after RIE plasma treatment, suggesting that the defects were predominantly induced by remote plasma treatment.

Figure 8. Raman mapping of pristine and remote plasma-treated graphene.

Figure 9. Raman mapping of pristine and RIE plasma-treated graphene.

3.4. Effect of heat treatment

The use of RTA equipment can facilitate the defect healing process in graphene [26]. The experimental results for Raman spectroscopy and the I_dV_g and I_dV_d curves following the RTA heat treatment are presented in Figs. 10 and 11. After RTA, the intensity of the D band decreased, whereas that of the 2D band increased. Furthermore, the Dirac point of graphene shifts toward the negative V_g region, implying the mitigation of p-doping effects. However, when the electrical and Raman results for plasma-treated graphene were compared with those for pristine graphene, I_D was reduced and the D band became more prominent, as shown in Figs. 12 and 13, suggesting the persistence of residual defects and functional groups even after the RTA treatment. In terms of the Raman spectra and electrical properties, the RTA heat treatment demonstrated a partial recovery of the damaged graphene resulting from the prior plasma treatments, as shown in Fig. 4 [27].

Figure 10. (a) Raman spectra, (b) I_dV_g, and (c) I_dV_d curves for graphene after 4 s remote plasma and RTA treatments.

Figure 11. (a) Raman spectra, (b) I_dV_g, and (c) I_dV_d curves for graphene after 4 s RIE plasma and RTA treatments.

Figure 12. (a) I_dV_g, (b) I_dV_d curves, and (c) Raman spectra for pristine and RTA treated graphene after remote plasma treatment.

Figure 13. (a) I_dV_g, (b)I_dV_d curves, and (c) Raman spectra for pristine and RTA treated graphene after RIE plasma treatment.

In this study, we investigated the changes in the structural and electrical properties of graphene using various O₂ plasma treatment methods and durations [28]. The analysis of the overall characteristics of the material indicated that the remote-mode plasma treatment led to a more pronounced and sudden shift in the Dirac point and resistance, which was attributed to the stronger intensity of remote-mode plasma compared to that of RIE plasma. Furthermore, recovery was observed during the thermal treatment. We observed the electrical characteristics of graphene following O₂ plasma treatment, anticipating that such recovery effects could contribute to the advancement of high-performance 2D FETs [6,12,29], fostering the exploration of alternative materials to Si.

This work was supported by the Sejong Science Fellowship (2022R1- C1C2005607), Regional Innovation Strategy (RIS, 2021RIS-004), and Basic Science Research Program (2021R1I1A1A01057416, 2021R1A6-A1A03043682) funded by the National Research Foundation of Korea (NRF).

The authors declare no conflicts of interest.