Verification of experimental results with simulation on the production of few-layered graphene by liquid phase exfoliation using sonication

In this section, the effects of various parameters of power variation and sonication duration on the graphene exfoliation process are presented and discussed.

Effect of radiation power on exfoliation

The graphene dispersion is prepared using the sonicator probe in a water-ethanol solution at different sonicator outputs with 40%, 50%, 60%, 70%, 80% and 90% equivalent to 70, 170, 197 , 225, 245, and 264 watts (we believe at 70-170 wattage the effect of wattage on exfoliation is very small, we gave up testing) in the optimal sonication time of 55 min. From the previous results of experimentation in this research, the best condition is observed to use a solution with a volume of 125 CC. The effect of detailed modification of sonication power output by measuring their UV-Vis absorption spectra and SEM is recorded.

The optical characterization of the graphene sheets is carried out with a Perkin-Elmer model lambda 25 spectrometer. Using the absorption spectrum of the UV-Vis spectrometer, one can estimate the approximate thickness of the layers or also the approximate number of layers. The UV–Vis absorption spectra of graphene samples for 40–90% of the maximum power of the continuous radiation mode are shown in Fig. 4. For the sample sonicated with 40% output power, two distinct peaks at 223 nm and 266 nm are shown which shows that the sample contains a mixture of graphene oxide (223 nm peak) and graphene (266 nm peak). Whereas for cases of 50-90% sonic power, a single peak is observed with increasing absorption. Moreover, from 50 to 80% sonic power, the samples show a single peak at 266 nm, whereas for 90% sonic power the sample shows an absorption peak at 270 nm. It is expected due to the existence of additional oxygen between the layers of graphite oxide, which facilitates exfoliation, thus having more graphene-like flakes but reduced with oxygen. The UV–Vis absorption spectrum of graphene produced by 90% sonicator power is shown in Fig. 4. A prominent peak is found at ≈ 270 nm, corresponding to the π at π* transitions of graphene and graphite34. Figure 5 shows the absorbance at 270 nm and the relative peak intensity increases with increasing output power from 40% power to 90% power. For the sample sonicated at 90% power, a higher amount of graphene at a few layers (1 to 3) compared to 40% power is observed by an increase in peak intensity. As can be seen in Figure 5, there is a continuous increase in absorbance with increasing power.

Figure 4

The UV-visible absorption spectrum for graphene exfoliated in water-ethanol for different sonication powers.

Figure 5
number 5

Measurement of UV-visible absorbance as a function of power variation for graphene samples.

This result is also confirmed by SEM images (ESEM Philips XL30), as shown in Fig. 6. Additionally, FIG. 6 demonstrates that at the power of 264 W, the size and quality of the flakes are increased compared to the previous samples. At 70 W, the flakes are smaller and relatively thicker. At 264 W, the dimensions of the flakes are about twice the size of the exfoliated sample at 70 W with thinner layers.

Figure 6
number 6

SEM micrographs of graphene flakes produced at: (a) 70W, (b) Sonicator power 264 W (scale bars are 2 µm).

Comparison of numerical results and experimentation

From the simulation of the sound pressure distribution, we found that at the same condition32, the change in output power changes the sound pressure range (Figs. 2, 3). We conclude that changing the output power for the exact geometric parameters will significantly change the maximum and minimum sound pressure in the liquid contents of the sono-reactor. When increasing the output power of the sonicator, an increase in overpressure is observed which is always equal to the vacuum in the liquid. The difference between the maximum and minimum simulated pressures ((Delta p)) increases by increasing the output power of the sonicator by 0.91 × 106 Pa at 2.7 × 106 Pa (Table 2). An increase in pressure difference causes more graphene flakes to exfoliate from graphite powder by overcoming the Van der Waals bond between graphite layers. Moreover, at 90% sonicator output, the change in absorption power at 270 nm (Fig. 5) reveals exfoliation of graphene with few layers (1-3 layers), which is confirmed by an earlier report .34. During low power sonication, the multi-layered flakes of graphite in the solvent are produced. With increasing power of the sonicator, the exfoliated species turn into flakes with a few layers. The simulation verifies experimental results which, by a strong increase in ∆p at 90% power reaching few layers, are validated by absorption at 270 nm. The above results of absorption spectra are also approved for few-layered (1-3 layers), multi-layered (4-10 layers) and thick-layered (>10 layers) graphene compared to previous35.

Effect of sonication duration on exfoliation

The influence of effective sonication time for graphene dispersions in water-ethanol solutions at 264 W power is investigated to determine the range of graphene flake production in different sonication times of 25, 35, 45 , 55 and 65 min, the results of which are presented in FIG. 7. From Fig. 8, it can be seen that the average amount of absorbance gradually increases with increasing effective sonication times up to 55 min and then decreases. One explanation for the observed result is that the prolonged exposure time provides more opportunity for delamination of more flakes. However, above 55 min of exposure, there could be a competition between the increasing factors and the tendency of the nanosheets to aggregate with each other. Therefore, it could be seen that after 55 min, the amount of absorbance is reduced.

Picture 7
number 7

The UV-visible absorption spectrum for exfoliated graphene for 264 W power versus different sonication times.

Figure 8
figure 8

Variation of UV-visible absorbance of graphene samples as a function of sonication time.

The SEM micrographs shown in Figure 9 confirm the results mentioned above. As can be seen, the flakes in Fig. 9b for 55 min are more prominent and thinner than the flakes in Fig. 9a for 25 min of sonication. By increasing the effective sonication time from 25 to 55 min, more exfoliation occurs. But, beyond 55 min, the quality of the flakes decreases because of the agglomeration process. Simultaneous agglomeration and crushing is observed in Figure 9c in the flakes.

Figure 9
number 9

SEM micrographs of graphene flakes for different sonication times: (a) 25 minutes, (b) 55 minutes and (vs) 65 min (scale bars are 5 µm).

All of the findings from this research culminate in the fact that our most delicate graphene samples in these experiments are acquired at; 264 W power, 50% pulse duration and 55 min irradiation time which is the “optimal situation” to produce graphene flakes.

The structural characterization of the graphene nano-sheets is then examined by TEM (Zeiss EM10C, 100kv). Figure 10a shows that this suspension contained well-exfoliated graphene sheets, including multilayers (less than 10 layers) and few layers (1–3 layers), with larger sheets (Figure 10b). Figure 10a shows clearly partially exfoliated graphite flakes with some aggregation and also multi-layered graphene sheets that appear darker gray. The few-layered graphene sheets with thin large flat graphene flakes are shown in Fig. 10b (also Fig. 9b). The results indicate that the exfoliated graphene sheets are a few layers away without significant structural defects, so this process could be scaled up successfully.

Picture 10
number 10

TEM micrographs of exfoliated graphene flakes, scale bar: (a) 200 nm, (b) 100nm.

For greater peace of mind, analysis by Raman spectroscopy is used. Figure 11 shows the Raman spectrum of the synthesized FLG. In the Raman spectrum of graphene, three distinct features are prominently named the D peak at ~1350 cm−1peak G at ~1586 cm−1and the 2D peak at ~2655 cm−1 exist. Therefore, the Raman spectrum of the sample, as shown in Figure 11, conforms the formation of graphene to a few layers. ({I}_{2D/{I}_{G}}) Sample ratio can be calculated from observed peaks. Based on ({I}_{2D/{I}_{G}}) The ratio of Raman spectra of the sample is calculated to be approximately 0.95–1. According to the previous report [10]the number of layers can be deduced from the ratio of the peak intensities, ({I}_{2D/{I}_{G}}), as well as the position and shape of these peaks. The ({I}_{2D/{I}_{G}}) the ratio being ~2 to 3 is for monolayer, the ratio being ({2>I}_{2D/{I}_{G}}>1) for bilayer graphene and the ratio being ({I}_{2D/{I}_{G}} for multilayer graphene. Therefore, it can be concluded that the graphene produced in this report has few layers due to ({I}_{2D/{I}_{G}}) ratio being ~ 1.

Picture 11
figure 11

Raman spectrum of few-layered dispersed graphene flakes exfoliated by ultrasonic sonication.