Ultrafast single-step synthesis of graphene flakes using isopropyl alcohol dissociated by microwave plasma generated in resonant cavity

Abstract

We describe an ultrafast single-step method to synthesize few-layer graphene flakes using microwave-generated plasma stored in a resonant cavity. The function of the cavity is to provide standing waves in the TE mode, in different resonance positions, enabling the formation of thermal plasma responsible for the dissociation of isopropyl alcohol, used as carbon precursors, and a galvanized steel wire, used as a catalyst; the graphene synthesis process occurs within 6 min. The results of Raman spectroscopy revealed that the graphene flakes are of good quality, with ID/IG ratios varying approximately in the range of 0.29–0.38, and with the formation of few layers, that is, less than 5. In addition to Raman spectroscopy, other techniques were used for characterization, such as scanning electron microscopy, energy-dispersive spectroscopy, X-ray diffraction, and atomic force microscopy.

Keywords
Graphene flakes; Resonant cavity; Thermal plasma

1. Introduction

GRAPHENE is a two-dimensional (2D) material of sp² hybridization of carbon atoms that has multiple and promising applications in different areas of knowledge, such as electronics, mechanics, optics, magnetic, and others still under study [¹, ², ³, ⁴]. In reality, it is no longer about future applications, but rather about immediate and real applications. Since the discovery of these new nanostructures and their synthetic prospects, there has always been an obstacle between their binomial quantity and quality. For graphene, no significant differences were observed. According to Choi et al., the effective application of graphene and other 2D materials strongly depends on the industrial-scale manufacturing of suitable films and their quality [⁵]. Although there are some well-established synthesis methods for graphene, this issue remains. Other researchers have added that the process must be simple, efficient, cheap, ecological, and of good quality [⁶, ⁷, ⁸]. Therefore, it is necessary to develop additional methods to synthesize this powerful and revolutionary material. Thus, we propose another technique for obtaining graphene, the characteristics of which are: few layers, sub-μm, inexpensive, high quality, security, simple, low cost equipment, ecological, fast. In this study, thermal plasma generated by microwaves, with a frequency of 2.45 GHz, and maintained in a resonant cavity was used. The cavity operates in the TE_01n mode, generating standing waves at different resonance positions. These standing waves promote plasma formation inside a quartz tube. The plasma dissociates the carbon precursors derived from isopropyl alcohol (IPA, C₃H₈O). Once in contact with the catalyst, the dissociated species lead to the formation of graphene.

Some byproducts, such as hydrogen and oxygen, are released into the atmosphere. According to Wang et al. [⁹], H₂ molecules inhibit the rolling and closing of graphene sheets by terminating the dangling carbon bonds. The entire process is performed at atmospheric pressure, and under the following conditions: the plasma assumes high energy levels with excitation temperatures of 3000 K and electronic densities of around 10¹⁷ electrons/cm³. Thus, the plasma generated under these conditions is a strong candidate for the synthesis of nanometric structures because of its high dissociation power [¹⁰]. Zheng et al. [¹¹] also stated that plasma is an excellent alternative to chemical reactions and deposition of thin films, as well as in the preparation of materials and applications in the manufacture of nanostructures [¹¹]. In this article, we report the synthesis of graphene flakes at atmospheric pressure, which is safer and cheaper than H₂/He, according to Wang [⁹]. Several materials with a catalytic influence and low carbon solubility have been tested to obtain a high-quality process [⁷, ¹²], and galvanized steel wire has been used as the metal. The primary objective of this study was to ensure an adequate energy supply for the dissociation of isopropyl alcohol (IPA), the carbon source, while preventing degradation of the metal catalyst. This is important because, under certain conditions, structures other than graphene may form. According to Isoldi et al. [¹³], carbon nanotubes and nanofibers may form when the metal catalyst either melts or undergoes partial surface degradation. In our approach, precise adjustment of the side covers ensures adequate energy input, enabling efficient graphene synthesis.

2. Experimental

2.1. Resonant cavity

The complete system consists of a microwave oven, a cylindrical resonant cavity, and a quartz tube coaxially positioned inside it. The replacement of the rectangular cavity (microwave oven) with a cylindrical cavity is due to the need to adjust multiple resonance positions, which would not be feasible inside a conventional oven because it supports only a single resonance condition.

The cylindrical resonant cavity used to generate and maintain the plasma, Figure 1(A), was made in aluminum, with radius and length of 79.5 mm and 280 mm, respectively. It is worth mentioning that the remaining parts of the oven (also shown in Figure 1A) were kept to control the magnetron. Figure 1(B) Plasma formation using only Ar at atmospheric pressure and Figure 1(C) Color change to turquoise blue owing to the presence of IPA vapors dragged by Ar (In both situations – Figure 1(B) and Figure 1(C) – the resonance regions were already determined, and the flow was adjusted to 1.2 L/min. To ensure proper system operation – specifically, resonance within the cavity and subsequent plasma formation inside the quartz tube, where the steel wire is also placed – the lateral covers must be adjusted slowly and precisely (as indicated in Table 1), while maintaining a continuous argon flow of approximately 1.2 L/min, which is essential for sustaining the plasma inside the quartz tube. To determine the resonance regions of the cavity, the resonant mode chart of the cylindrical cavity (Figure 2 [¹⁴, ¹⁵, ¹⁶]) was first used as a reference, and the results were subsequently confirmed with a frequency generator set to 2.45 GHz. In the mode chart, a denotes the cavity radius, d the length, and f the frequency in MHz, directly obtained from the frequency generator. The oven magnetron was not used for this determination, serving only as the microwave source during plasma generation.

Figure 1
(A) Equipment used in plasma production (cylindrical cavity and magnetron used in a conventional microwave oven); (B) Plasma formation using only Ar at atmospheric pressure; (C) Color change to turquoise blue owing to the presence of IPA vapors dragged by Ar.

Table 1
Values of the distances between the side covers obtained using graphical methods and the frequency generator such that the cavity enters resonance at 2.45 GHz.

		Covers distance-cm
Graphic	Axis-x	Graphic	Generator
m	0.2	35.55	–
n	0.7	19.00	19.63
o	0.8	17.15	17.33
p	1.1	15.62	15.73
q	1.4	13.54	13.72
r	2.7	9.57	9.08
s	4.4	7.58	7.48
t	5.4	6.81	6.61

Figure 2
Resonant Mode Chart used to obtain the distances between the side covers of the cylindrical cavity; the highlighted points, m, n, o, p, q, r, s, and t, are all possible theoretical distances between the side covers, for there to be resonance TE_01n mode. [¹⁴, ¹⁵, ¹⁶]. Used with permission of John Wiley & Sons, from Microwave Engineering, David M. Pozar, 4th Edition, © 2011; permission conveyed through Copyright Clearance Center, Inc.

For a fixed value of the cavity radius, it is possible to find the d values through the graphical relationship, adopting the magnetron frequency of 2.45 GHz. Table 1 lists the internal lengths of the cylindrical cavity, distance between the movable covers, such that there is resonance in the two proposed methods. In the method employing the frequency generator, we used the Agilent E5071C ENA Series Network Analyzer, which allowed us to replace the microwave magnetron. This equipment provides a broad frequency range (100 kHz to 8.5 GHz), enabling precise determination of the cavity dimensions.

Both the graphical method and the frequency generator resulted in close points, although only one point was admitted, that is, the exact distance for the electromagnetic waves to enter into resonance [¹⁴, ¹⁵, ¹⁶, ¹⁷].

2.2. Plasma formation at atmospheric pressure

Plasma formation at atmospheric pressure requires minute adjustments of the internal length of the cavity. The values corresponding to the length of the cavity, that is, the distance between the side covers, according to theory, are just a reference because it is the precise length of the cavity that will enable resonance. At atmospheric pressure, the difficulty becomes even greater, and any variation in the fractions of millimeters in the volume of the cavity is the difference between plasma forming and not forming. Based on the values listed in Table 1 and the system in operation (magnetron 800 W and initial argon flow ∼1,1 L/min), the side cover must be slowly moved until the system enters resonance, resulting in the formation of plasma inside a quartz tube arranged coaxially with the cavity.

2.3. Graphene synthesis

Graphene flakes were formed inside a quartz tube, coaxial to the cavity. Inside the quartz tube was the catalyst metal, which was a galvanized steel wire measuring 130 mm in length and 2 mm in diameter. IPA is released in the form of vapor inside a quartz tube via argon (the carrying gas used for plasma formation). The diagram in Figure 3(A) shows the device used to introduce alcohol vapor into a quartz tube.

Figure 3
Apparatus used to launch IPA vapors into the quartz tube; (A) IPA reservoir; (B) Quartz tube, catalyst, cavity, and magnetron; (C) Argon flow inside the quartz tube performed by the COMSOL multiphysics simulator.

The argon flow rate was 1.3 L/min, and the entire process was performed at room temperature ∼25°C. The IPA vapors reaching the quartz tube were thermally decomposed into carbon species, which, in the presence of the catalyst, contributed to the formation of graphene; the entire process took approximately 6 min. Figure 3(C) shows the result of the simulation, via COMSOL, of the Ar flow distribution, which is laminar inside the quartz tube; the legend in the left corner of Figure 3(C) refers to the initial velocity of the argon flow, 1 L/min, or approximately 0.4 m/s, inside a 7.0 mm diameter quartz tube. The flow must be laminar inside the quartz tube to avoid fluctuations and instabilities in the plasma formation.

3. Results and Discussion

3.1. Precise sizing of the resonant cavity

The first point in Table 1 (x-axis = 0.2) corresponds to a cavity with a length of ∼355 mm; this first value is discarded because the cavity has a total length of 280 mm. The following values, both by the resonant mode chart and by the frequency generator, present slight divergences because, in the mode chart, the values are obtained through measurements performed on the graph. In the frequency generator, the values are closer to the experimental values as long as the magnetron used is exactly 2.45 GHz. A 1% fluctuation in the magnetron frequency significantly affects the resonance inside the cavity. This variation can be observed when a burned-out magnetron is replaced with a new one; it is unlikely that the internal length will remain the same.

3.2. Search for greater use of the cavity

The resonant cavity used in this project was a multimode cylindrical cavity, and TE_01n was selected as the working mode because of its high quality factor. To reduce the number of perturbation modes close to the working mode, the cavity dimensions must be carefully designed to move the perturbation modes away from the desired modes as far as possible [¹⁷].

In the graph in Figure 4, the y-axis represents the resonant cavity quality factor and the x-axis represents the 2a/d ratio (cavity sizing). Cavity sizing was significantly optimized in the TE₀₁₁ mode. The length of the cavity to operate in the TE₀₁₁ mode was ∼137 mm.

Figure 4
Merit Index - Normalized unloaded Q for various modes-air filled- [¹⁴, ¹⁵, ¹⁶, ¹⁷].Used with permission of John Wiley & Sons, from Microwave Engineering, David M. Pozar, 4th Edition, © 2011; permission conveyed through Copyright Clearance Center, Inc.

3.3. Characterization of graphene flakes

The material formed inside the quartz tube, shown in Figure 5(A), was extracted and analyzed. Figure 5(B) shows a scanning electron microscopy image of the flakes at 270x magnification. The atomic force microscopy (AFM) image of the same sample, shown in Figure 5(C), has a thickness of ∼50 nm and lateral size in the range of 0.6–0.8 μm. Figure 5(D) corresponds to the Raman spectroscopy performed on a wide area of synthesized material. A Raman spectrometer (Renishaw) equipped with a 532 nm and 25 mW laser was used in conjunction with a 50X objective lens, providing illumination of approximately 2 μm in diameter and highlighting the D, G, and 2D peaks. The peak intensities were obtained after background subtraction and fitted with Gaussian functions

Figure 5
(A) Extracted material; (B) SEM of the sample magnified 270 times; (C) AFM of a ∼0.69μm₂ flake sample; (D) Raman spectrum of graphene, highlighting peaks D, G, and 2D – detail of the right side of the D peak magnification and Origin software processing; (E) SEM with 100,000x magnification; (F) SEM-EDS of the sample.

(1) $$y=y_0+\frac{A}{w\sqrt{\frac{\pi }{2}}}\exp -2{\left(\frac{x-x_c}{w}\right)}^2$$

where y₀ is the baseline offset, A is the integrated area under the peak, x_c is the peak center, and w is related to the full width at half maximum (FWHM). The fitting yielded x_c = 1580.2cm⁻¹, FWHM = 35.6cm⁻¹, and an amplitude of ∼3697 a.u. The ID/IG ratio is used to assess defect density and, consequently, the quality of the produced graphene. While, the I2D/IG ratio can be employed to estimate the crystallite size and, therefore, the number of layers [¹⁸].

The ratio between the intensities of the ID/IG ∼0.38 and I2D/G ∼1.09 peaks corresponds to a sample with few defects and the formation of a double-few layer, according to [¹⁹, ²⁰, ²¹].

The SEM image shown in Figure 5(E) was obtained using JEOL JSM-7200F equipment, and shows graphene sheets with a diameter of ∼165 nm, measured using ImageJ software; Figure 5(F) shows the results of a SEM-EDS measurement of the sample with a predominance of carbon and a low amount of oxygen. All results refer to the sample, Figure 5(A), without any treatment or dispersion of the graphene layers, that is, as obtained. Figure 6(A) shows a diffractogram (DRX-04 -Panalytical-Empyrean) of the sample. When performing the X-ray diffraction (XRD) measurements, the samples were irradiated with a CuK_{α 1} source with a wavelength 1.54059 Å, voltage of 45 kV, current of 40 mA, initial/final angles of 2.5° and 70°, and angular pitch 0.02°.

Figure 6
(A) XRD spectrum high lighting the 2θ point at 26.05Å, (B) Graphic relationship interlayer distance (d₀₀₂) versus number of layers (N_L) [²¹].

The peak referring to the plane (002) 2θ equal to 26.05° and full width at half maximum (FWHM) 1.76° was adjusted using the Gaussian function; the distance d⁰⁰² between the layers was calculated using the Bragg relationship (d = λ/(2 sin θ), and with the diffractogram data equal to 3.42 Å. According to Huh [²²], through the graphic relationship between the interlayer distance (d₀₀₂) and the number of layers (N_L), it is possible to determine the number of layers, which in this case was estimated to be 5, as shown in Figure 6(B).

Although the XRD results differ regarding the number of layers using the Raman technique, the graphene obtained still has a few layers, and the main focus of this study is the investigation of new synthesis routes. Chart 1 summarizes the characteristics of graphene flakes formed by the dissociation of IPA and the catalyst, a galvanized steel wire. These are cheap inputs, easy to obtain, safe, and ecological. The energy source used in graphene synthesis is a magnetron coupled to a cylindrical resonant cavity, without the need for any vacuum system or waveguides (making the equipment also low cost).

Chart 1
Characteristics of graphene flakes produced using the proposed method.

Method	Thickness	Side	Advantage	Disadvantage
Microwave Plasma (Resonant cavity)	Few layers	Sub-μm	Inexpensive, High Quality, Security Simple, Low cost equipment Ecological Fast	*Low yield
				*until now

In addition to the graph presented in Figure 6(B), the number of graphene layers can be determined using the Scherrer equation [²³]:

(2) $$D_{GP}=\frac{k.\lambda }{\beta .cos\theta }$$

where D represents the thickness of the graphene stack; d is the interlayer distance; bis the full width at half maximum (FWHM) of the X-ray diffraction peak (measured in radians); λ is the X-ray wavelength; k is a dimensionless shape factor; and θ is the scattering angle. According to Huh [²²] the value of k depends on the crystallite shape, structure, and experimental conditions. Although conventional k values of 0.89, 1.0, or 1.123 are frequently employed, recent studies on multilayer graphene (MLG) have reported lower values – such as 0.28 or 0.34 – derived from experimental fittings for samples containing 5 to 7 layers. Moreover, according to Huh [²²], the number of layers (N_L) can then be calculated as:

(3) $$N_L=\frac{D}{d}$$

Considering the values obtained from the diffractogram, with b = 1.77°, θ = 26°, and adopting a constant k = 0.34, the calculation yields approximately 5 layers, which is corroborated by the graphical representation shown in the Figure 6(B).

The graphene synthesized in this way is of good quality, with ID/IG < 0.38 and few layers, consistent with reports of few-layer graphene obtained by CVD and plasma methods [⁶, ⁷, ⁸, ⁹, ¹⁹, ²¹]. Moreover, the total synthesis time (6 min) is significantly shorter than in most conventional routes, such as CVD or wet-chemical methods, which typically require tens of minutes to several hours [²⁴, ²⁵]. Although there are isolated reports of faster growth under optimized CVD conditions (≈2 min) [²⁶], these approaches generally demand high temperatures and sophisticated setups. In contrast, the microwave plasma-based method presented here combines short synthesis time with a simpler and efficient route, reinforcing the advantages of the proposed technique.

4. Conclusion

It is possible to quickly obtain good-quality ID/IG ∼0.38 graphene with few layers <5 using energy from a resonant cavity to generate and maintain thermal plasma, which is used to dissociate IPA. The complete graphene synthesis required 6 min. The advantage of using a cylindrical resonant cavity is that the system does not use waveguides, complex refrigeration systems, or even vacuum equipment because the entire process is performed at atmospheric pressure. Furthermore, the inputs are low-cost; only argon is used to form the plasma and entrain the IPA vapors.

Acknowledgment

Evandro Pereira da Silva (Brazilian Research Unity in Astrobiology – NAP/Astrobio); we would like to thank the CAQI/IQSC/USP for Field Emission SEM facilities; we also thank the access to equipment and assistance provided by the Electron Microscope Laboratory (LME/UNICAMP); FAPESP EMU 2019/08178-0.; This research was partially supported by the Technological Characterization Laboratory from USP (LCT-USP); Federal Institute of Education, Science and Technology of São Paulo (IFSP Campus Pirituba); Editage for English language editing.

Data Availability

The entire dataset that supports the results of this study has been made available on SciELO Data and can be accessed at: https://doi.org/10.48331/SCIELODATA.H0MM3F.

References

• ^[1] M.M.S.A. Bhuyan, M.N. Uddin, Md.M. Islam, F.A. Bipasha and S.S. Hossain, International Nano Letters 6, 65 (2016).
• ^[2] K.E. Whitener Jr. and P.E. Sheehan, Diamond and Related Materials 46, 25 (2014).
• ^[3] V.B. Mbayachi, E. Ndayiragije, T. Sammani, S. Taj, E.R. Mbuta and A.U. Khan, Results in Chemistry 3, 100163 (2021).
• ^[4] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva and A.A. Firsov, Science 306, 666 (2004).
• ^[5] S.H. Choi, S.J. Yun, Y.S. Won, C.S. Oh, S.M. Kim, K.K. Kim and Y.H. Lee, Nature Communications 13, 1484 (2022).
• ^[6] N. Woehrl, O. Ochedowski, S. Gottlieb, K. Shibasaki and S. Schulz, AIP Advances 4, 047128 (2014).
• ^[7] F. Zhang, K. Yang, G. Liu, Y. Chen, M. Wang, S. Li and R. Li, Composites Part A 160, 107051 (2022).
• ^[8] T. Yusaf, Sustainability 14, 12336 (2022).
• ^[9] Z. Wang, N. Li, Z. Shi and Z. Gu, Nanotechnology 21, 75602 (2010).
• ^[10] A. Sola, M.D. Calzada and A. Gamero, Journal of Physics D: Applied Physics 28, 1099 (1995).
• ^[11] J. Zheng, R. Yang, L. Xie, J. Qu, Y. Liu and X. Li, Advanced Materials 22, 2010 (2010).
• ^[12] A.N. Obraztsov, A.A. Zolotukhin, A.O. Ustinov, A.P. Svirko and Y.K. Jefimovs, Diamond and Related Materials 12, 917 (2003).
• ^[13] M. Isoldi, E.M. Ozono, R.D. Mansano and D.L. Rodrigues Junior, IEEE Transactions on Nanotechnology 19, 760 (2020).
• ^[14] D.M. Pozar, Microwave Engineering (John Wiley & Sons, New York, 2005).
• ^[15] P.F. Combes, Micro-ondes, Lignes, guides et cavités (Dunod, Paris, 1996), v. 1.
• ^[16] MASSACHUSETTS INSTITUTE OF TECHNOLOGY, Resonant Cavities and Waveguides, available in: https://web.mit.edu/22.09/ClassHandouts/Charged%20Particle%20Accel/CHAP12.PDF, acessed available in:12/06/2023.
  » https://web.mit.edu/22.09/ClassHandouts/Charged%20Particle%20Accel/CHAP12.PDF
• ^[17] E. Li, Z. Nie, G. Guo and Q. Zhang, Progress in Electromagnetic Research 92, 103 (2009).
• ^[18] R. Saito, M. Hofmann, G. Dresselhaus, A. Jorio and M.S. Dresselhaus, Adv. Phys. 60, 3 (2011).
• ^[19] A.A. Mostafa, M. El-Bahy, M.A. Ahmed and H.M.A. Abdallah, Carbon 31, 109 (2023).
• ^[20] INSTANANO, Graphene Number of Layers Calculator From ID/IG and I2D/IG Ratio via Raman Spectroscopy, available in: https://instanano.com/all/characterization/raman/graphene-layers/, acessed available in: 05/09/2023.
  » https://instanano.com/all/characterization/raman/graphene-layers/
• ^[21] Y. Hwangbo, C.K. Lee, A.E. Mag-isa, J.W. Jang, H.J. Lee, S.B. Lee, S.S. Kim and J.H. Kim, Carbon 77, 454 (2014).
• ^[22] S.H. Huh, Carbon 78, 617 (2014).
• ^[23] P. Scherrer, Göttinger Nachrichten Math. Phys. 2, 98 (1918).
• ^[24] F. Ghaemi, A. Abbasi, M. Shamsaie and A. Moshfegh, Nanoscale Res. Lett. 11, 2016 (2016).
• ^[25] H. Bohr-Ran, Y. Chen, T. Li and X. Zhang, Nanomaterials 13, 2023 (2023).
• ^[26] H.C. Lee, S. Kim, J. Park and K. Choi, RSC Adv. 7, 2017 (2017).

*Correspondence email address:

[e-mail]

Editor-in-Cheif:

Marcello Ferreira https://orcid.org/0000-0003-4945-3169

Publication Dates

• Publication in this collection
  28 Nov 2025
• Date of issue
  2025

History

• Received
  26 July 2025
• Reviewed
  14 Sept 2025
• Accepted
  20 Oct 2025