We carried out the preparation of large-area GO using our simplified Hummer’s method. This simplified method does not involve controlling temperature during the chemical oxidation of graphite. Unlike the commonly practiced processes where an ice bath is required during the initial addition of KMnO4 and heating during the oxidation stage, our simplified approach is hassle-free. The slight exothermic process, which increases the temperature to 40°C–50°C after the addition of H2O2 solution to terminate the oxidation process, will cool down naturally in a short time. This experiment is relatively safe, and the danger of explosion is reduced significantly, in contrast to the normal route of chemical oxidation of graphite. Furthermore, the mixing and washing steps are simple and straightforward, yielding an almost 100% conversion of large-area GO.
To study the effect of oxidation level we used KMnO4:acid (H2SO4:H3PO4) at the weight ratios of 1:80, 1:40, and 1:20 (hereon denoted as KS-80, KS-40, and KS-20, respectively, where the weight ratio of H2SO4:H3PO4 was fixed at 9:1) and the weight ratio of graphite:sulfuric acid was maintained at 1:100 for all the samples. Oxidation time was fixed at 6 hours, 1 day, 2 days, and 3 days, respectively, to evaluate the oxidation level effects on GO. The samples were gently sonicated for 5 minutes prior to FESEM viewing.
shows the average lateral dimension of the GO produced with different ratios of KMnO4:acids and oxidation time (6 hours to 3 days). After 6 hours of oxidation, GO with an average lateral dimension of 18.7 μm was formed in KS-20, which was larger than that of KS-40 and KS-80. The conversion was very low, as observed from the large amount of precipitate after centrifugation for 5 minutes at 4000 rpm, indicating un exfoliated or unoxidized graphite. The amount of precipitate observed in KS-80 and KS-40 was even greater than KS-20. After 1 day of oxidation, a similar result was observed as compared with an oxidation time of 6 hours. However, the average size of GO had increased for the three ratios. When the oxidation time increased to 2 days, GO in KS-20, KS-40, and KS-80 increased in size as compared with after 1 day of oxidation. The lack of precipitate after centrifugation suggests that KS-20 had the highest yield of GO, even though the oxidation process was not yet optimized. At day 3, KS-20 achieved a 100% conversion of graphite to GO, as precipitate was not found after centrifugation. The size of the GO produced is large, with an average lateral dimension of 58.3 μm.
Average lateral dimension of graphene oxide for different reaction parameters.
show the FESEM micrographs of graphene prepared using three different ratios of KMnO4:acids (KS-80, KS-40, and KS-20) after 3 days of oxidation. The GO samples were gently sonicated for 5 minutes using 80 W power of sonication before being spin-coated on SiO2/Si wafer and viewed under FESEM. From the micrographs, it was found that the average lateral dimension of GO increased as the ratio of KMnO4:acid increased. KS-20 with the highest concentration of KMnO4 produced the largest GO (). However, due to the sonication effect, the average lateral dimension of graphene becomes 58.3 μm. On the contrary, nansonicated GO in KS-20 presents uniform sheets of GO with an average lateral dimension of 95.1 μm (up to 120 um) and an area of ~8000 μm2 ().
Field emission scanning electron microscopy micrographs of graphene oxide prepared in (A) KS-80, (B) KS-40, and (C) KS-20 after 3 days of oxidation and (D) nonsonicated KS-20. All the micrographs were taken at the same magnification.
These observations are different from the reported investigations on the effects of oxidation on the size of GO. It was previously reported that the size of GO reduced with a higher degree of oxidation.17
However, by using our proposed simplified Hummer’s method, GO with a large lateral dimension is produced with a higher degree of oxidation. This can be explained by the fact that the initial graphite used for previous works is smaller in size as compared with that of ours, and the graphite may have overoxidized under elevated temperature (95°C). As for our sample, the optimum oxidation time was 3 days (for KS-20) with the GO produced possessing the largest dimension. However, after 4 days of oxidation, the GO produced showed small lateral dimension in the range of <20 μm (FESEM micrograph not shown).
Based on the FESEM images, a plausible mechanism for the formation of large-area GO is a complete oxidation of graphite but not overly oxidized, as it could lead to tearing of the GO. The complete oxidation condition resulted in graphite oxide to be exfoliated easily in large pieces. Another contributing factor that leads to the formation of large GO is that the reactions were carried out at room temperature. It has been reported that high temperature during the oxidation process reduces the size of the GO produced.17
The large-area graphene for sample KS-20 after 3 days of oxidation was measured using atomic force microscopy to determine the thickness of the GO produced. A typical tapping mode AFM image and the corresponding height cross-section profile of the GO sheets deposited on a mica substrate are shown in . Based on the AFM height profile analysis, the thickness of the KS-20 GO sample is around 1.2 nm. The result is consistent with the thickness of the characteristic single-layer GO reported.17
Figure 2 Atomic force microscopy measurement of graphene oxide (scan area: 1 μm × 1 μm). Tapping mode atomic force microscopy image of single-layer graphene oxide (KS-20) with corresponding height cross-sectional profile with an average (more ...)
Ultraviolet-visible spectroscopy measurement was carried out to monitor the degree of oxidation for the graphene samples. The spectra for KS-80, KS-40, and KS-20 were recorded after 3 days of oxidation, as shown in , with absorption peaks at 233 nm, 232 nm, and 229 nm, respectively. These peaks are due to the π → π* transition for the C=C bonding, which is similar to the reported value in the literature.25
The peak position for KS-20 sample exhibits the lowest wavelength, indicating that the sample has a lower amount of remaining conjugation (thus requires higher energy for the electronic transition) as compared with the other two samples (KS-40 and KS-80), which is due to the higher degree of oxidation with more functional groups on the GO basal plane. Also, there is a similar shoulder around 300 nm observed for all the three samples, which is attributed to n → π* transition of the carbonyl groups. Another observation on the ratio of the intensity of C=C bonding peak (~230 nm) to the intensity of 300 nm peak finds that the higher the degree of oxidation, the greater the ratio. This is because a more oxidized graphene basal plane yields a greater amount of isolated aromatic rings that increase the intensity of C=C bonding peak. In contrast, a less oxidized graphene basal plane gives rise to a lesser amount of isolated aromatic rings and remained in the extended conjugated aromatic rings form, and therefore the intensity of C=C bonding peak is relatively low.19
In this scenario, the peak intensity ratio for KS-20 is 3.2, and 2.1 for both KS-80 and KS-40, respectively. To the best of our knowledge, the reported ultraviolet-visible spectra for GO in the literature have ratios that fall within the range of <2, in which the lateral dimension of the GO produced is significantly smaller.
(A) Ultraviolet-visible absorbance spectra of graphene oxide after 3 days of oxidation time, and (B) X-ray diffraction patterns of graphite flakes and graphene oxide.
XRD patterns in show a distinct diffraction peak at 10.24° for GO and 26.56° for graphite flakes. The interlayer spacing values for graphite flakes and GO are 0.34 nm and 0.87 nm, respectively. The distance between consecutive carbon layers was increased for GO due to the introduction of oxide functional groups to the carbon basal plane via chemical oxidation reaction. This increased interlayer spacing for GO is in line with the results measured from AFM.
In , the C1 s band of the GO samples (KS-20, KS-40, and KS-80 after 3 days of oxidation) can be fitted to four deconvulated components, located at 284.3 eV, 286.6 eV, 287 eV, and 288.6 eV. These components can be assigned to the nonoxygenated ring carbon, C in C–O, C in C=O, and C in C(O)O, respectively, which are separated by ~2.0 eV, 1.0 eV, 1.0 eV, and 1.0 eV.26
All the oxygenated features increased in intensity as the concentration of KMnO4
increased from KS-80 to KS-20. The components that experienced the highest degree of oxidation is the C in C=O for KS-20, as can be seen from the high-intensity peak at around 287 eV. The components are in agreement with the results in the literature for similar C1 s band shapes.13
These X-ray photoelectron spectrometry results are in accordance with the ultraviolet-visible analysis, revealing that the concentration of KMnO4
directly influenced the degree of oxidation.
X-ray photoelectron spectrometry analysis of graphene oxide with different degrees of oxidation illustrating the deconvoluted X-ray photoelectron spectra in the C1s peak region for the KS-20, KS-40, and KS-80 samples.