Microwave-Assisted Preparation of Activated Carbon from Eupatorium Adenophorum: Effects of Preparation Parameters

Materials

Eupatorium adenophorum was obtained from Kunming, Yunnan Province of China. The raw materials were crushed to a size of 3–6 mm and washed thoroughly with distilled water, then were oven-dried at 80 °C and stored in moisture-free environment for utilization in the experiments. The proximate analysis of Eupatorium adenophorum is presented as follows: volatile 75.96 %, ash 1.86 % and fixed carbon 22.18 %. Compared with fixed carbon content of several precursors (bamboo 16.6 % [33], coconut 18.6 % [34], palm shell 18.7 % [34]), the Eupatorium adenophorum has higher carbon content. Therefore, Eupatorium adenophorum can be a promising raw material for preparation of AC.

Experiment method

Eupatorium adenophorum as raw material was heated up to a carbonized temperature of 500 °C at a heating rate of 10 °C/min with conventional heating and was held for 2 h at the carbonized temperature under the nitrogen flow atmosphere. After carbonization, the material was cooled to room temperature. The proximate analysis of char was as follows: volatile 14.62 %, ash 9.34 % and fixed carbon 76.04 %. Compared with raw material, the carbon content of the char increased significantly after carbonization, and the volatile matter of char decreased. AC was prepared according to the previous reported methods [22, 24, 28]. The char of 8 g was crushed sieved to obtain particles with the mesh size of 600–1,000 µm and mixed with KOH with different IR. The mixtures were heated under purified nitrogen in the microwave heating equipment under different microwave power and activation time. After heating, the mixtures were immediately put into the distilled water. It was beneficial to further form pores of carbon body. The mixtures were washed with 1:1 volume ratio of HCl, followed by rinsing with distilled water until the pH of the washing solution reached 6–7 [22]. The filtered wet samples were dried in a laboratory oven at a temperature of 110 °C for 4 h and stored for further characterization. Microwave furnace was made by the Key Laboratory of Unconventional Metallurgy, Ministry of Education of China with single-mode continuous controllable power (Figure 1).

Figure 1:

Schematic of microwave heating furnace.

Design of experiments

RSM is a collection of mathematical and statistical techniques for modeling and analysis of problems in which a response of interest is influenced by several variables. A standard RSM design, central composite design (CCD), was utilized to optimize the preparation conditions of AC, as well as to analyze the interactions among the parameters [35].

The dependent variables selected for this work were microwave power (X₁), activation time (X₂) and IR (X₃). The iodine number (Y₁) and yield (Y₁) of AC were taken as the two responses of the designed experiments. The upper and lower limits of the process variables are shown in Table 1. For each categorical variable, a full factorial CCD for the three variables, consisting of 8 factorial points, 6 axial points and 6 replicates at the center points, was employed, indicating that altogether 20 experiments were required. It was calculated from the following equation.

N is on behalf of the total number of experiments, and n is the number of dependent variables. Table 2 shows the experimental conditions adopted for preparation of AC as generated by the Design Expert software covering the parameters such as microwave power, activation time and IR, corresponding iodine number (Y₁) and yield (Y₂) of the AC. The results are listed as well in Table 2. Repeating the center point (15–20 runs) is employed to ensure the reproducibility of experimental data, and it could make sure that char has been entirely activated.

Iodine adsorption number and yield

The iodine number of the AC was calculated using the standard testing methods of the People’s Republic of China (GB/T12496.8-1999). The yield was calculated based on the equation: Y=M/M₀×100 %, where M and M₀ were the dry mass of char (g) and dry mass of AC (g), respectively.

Characterization of AC

The BET surface area of AC was measured through nitrogen adsorption isotherm at 77 K (Autosorb-1-C, Quantachrome). The Fourier transform infrared spectroscopy (FTIR) was applied to qualitatively identify the chemical function groups present in the AC. The microstructures were analyzed by the scanning electron microscope (SEM, Philips XL30ESEM-TMP). Transmission electron microscope (TEM, JEM-2100, Japan) was carried out to assess the inner structure of the AC.

Response analysis and verification of the regression model

Verifying the accuracy of the models is crucial part of data analysis; if the model accuracy is not high, they will result in poor results even the wrong conclusion. The final empirical models in terms of coded factors for the iodine number (Y₁) and yield (Y₂) of the AC are listed in eqs. (2) and (3).

The quality of model developed is always evaluated using the correlation coefficient (R²), which is 0.9556 for eq. (2) and 0.9304 for eq. (3), respectively. Both the R² values of iodine number and yield are close to unity, indicating that there is a good agreement between the experimental and the predicted data.

The ANOVA for the quadratic model of iodine number is presented in Table 3 and carried out to justify the model. The Model F-value of 23.9 and “Prob > F” < 0.001 both imply that the model is significant. If the values of “Prob > F” are less than 0.05, the model terms are significant, or not. In this case, X₁, X₂, X₃, X₁X₂ and X₂X₃ are significant model terms.

Table 4 is analysis of variance for response surface quadratic model for yield. The F value is 14.86, and Prob > F is 0.0001, which means that the model is significant. In this case, X₁, X₂, X₃ along with the interaction parameter X32 are found to be significant, based on “Prob > F” < 0.05. The ANOVA results show that the model is appropriate to predict within the range of the variables studied.

AC iodine number from response surface analysis

The one of most important characteristics of AC is its iodine adsorption number, which has been reported to be close to the BET surface area and pore volume [22]. Figure 2 shows the three-dimensional response surfaces of the combined effect of microwave power and IR on the iodine number when the activation time is 20 min. It can be observed that iodine number increases with increase in the microwave power and IR. Owing to the fact that an increase in the microwave power can reduce the activation energy of the C–KOH reactions, and the rate of reaction is speeded up resulting in developing amounts of pores of the AC. Similarly, an increase in IR can speed up the extent of the C–KOH reaction. The carbons on the active sites react with the KOH and form K₂CO_3, which makes AC form more pores. It is clearly seen that iodine number increases with increasing in IR from 2.0 to 4.0. The mechanism of the C–KOH reactions is complicated, which has been studied by many researchers [22, 23]. However, the mechanism has not been fully understood. The generally accepted view is that a series of chemical reactions and K intercalation form the pores of AC. Furthermore, potassium metal may be liberated as the reaction temperature intercalate and force apart the separate lamellae of crystallite, making new pores [36]. Though the reactions are intricacy, the reactions may be as follows [36, 37]:

Figure 2:

Three-dimensional response surface plot of iodine number: effect of microwave power and IR on iodine number (activation time: 20 min).

Meanwhile, the reactions generate the gas such as H₂, CO and CO₂, which could promote the formation of new pores [22]. The result is similar to Foo and Hameed reporting on the microwave-assisted preparation of AC from biodiesel industry solid residue [38].

Figure 3 shows the three-dimensional response surface of the combined effects of microwave power and activation time on the iodine number at the IR of 3.5. We can obviously discover that the iodine number gradually increases with increase in microwave power as well as activation time. This may be explained by that microwave power and activation time can speed up the C–KOH reaction, and the activation process would be strengthened with more pores formed. Muthanna and Theydan [39] have reported the effect of microwave power and activation time on the iodine number, which is consistent with our experimental results. Table 5 shows the iodine numbers of AC prepared from various raw materials by microwave heating with different activation agents. It can be seen that the iodine number of AC prepared from Eupatorium adenophorum is higher than other raw materials.

Figure 3:

Three-dimensional response surface plot of iodine number: effect of microwave power and activation time on iodine number (KOH/C weight ratio: 3.5).

AC yield from response surface analysis

The yield of AC is also an important index of application as it decides the amount of final product. Figure 4 shows the three-dimensional response surface of the combined effect of microwave power and IR on the yield of AC with the activation duration of 20 min. From Figure 4, the yield of AC is found to decrease with the increase in microwave power and IR. Figure 4 also shows that the minimum of yield corresponds to the maximum of microwave power and IR. An increase in microwave power and IR effectively contributes to an increase in the degree of activation reaction, which will result in the reduction in the yield of AC. The similar results have been reported by Xing et al. [43].

Figure 4:

Three-dimensional response surface plot of yield: effect of microwave power and IR on the yield (activation time: 20 min).

The influence of the microwave power and activation time on the yield of AC is shown in Figure 7 when the IR is 3.5. As shown in Figure 5, the AC yield decreased with gradually increasing microwave power and activation time. An increase in the microwave power and activation time can speed up the C–KOH reaction. As a result, the yield of the AC decreases. In addition to, some components of the AC tar and volatile matter formed in the activation process are easy to get rid from the surface of AC, which will also lead to AC yield reducing. The result is similar with Alslaibi et al. using Olive stone preparation of AC [44].

Figure 5:

Three-dimensional response surface plot of yield: effect of microwave power and activation time on the yield (KOH/C weight ratio: 3.5).

Process optimization

To sum up, industrial production of AC augurs a high iodine number and yield, which is beneficial to the industrialized production, but the adsorption performance and yield respond in opposite ways to the variation of process parameters. Since it must determine the optimum experiment parameters to maximize the yield and iodine number, the most desirable experimental parameters were identified with the help of the Design Expert software. The optimum experimental conditions for the preparation of AC are listed in Table 6. The experiments are repeated three times to verify the accuracy of the predicted result at the optimum experimental conditions, and the results are also presented in Table 6. Compared with the AC prepared by Fierro et al. [45], the utilized method of microwave heating and KOH activation makes activation time of 1.5 h shorten more than 83 % and improves significantly the surface area of AC.

Characterizations of pore structure

Figure 6 shows that the nitrogen adsorption isotherm of AC and char. N₂ adsorption quantity is very small and increases slowly for the char, indicating that the char don’t have many pores. However, the AC is microporous material, evidenced by the type IV of the referred IUPAC classification [46]. No obvious hysteresis loop can be seen in the adsorption isotherm. When the relative pressure (P/P₀) is below 0.1, N₂ adsorption increases quickly for the AC. When the relative pressure reaches 0.1, the adsorption capacity reaches 57 % of saturation adsorption capacity. The adsorption quantity increases slowly and is upwardly convex as the relative pressure increases, showing that the adsorption transits from the single molecular layer to the multimolecular layer [47].

Figure 6:

Nitrogen adsorption isotherm of the AC and char.

Moreover, the details of pore structure of Char and AC derived from Eupatorium adenophorum are shown in Table 7. As can be seen from Table 7, the BET surface area, total pore volume and average pore diameter have a large increase. The pore structures of the AC prepared from various waste materials by microwave heating and conventional heating using KOH activation are listed in Table 8. This proves that using microwave heating and a chemical activation agent for the preparation of AC are feasible. The surface area of the AC is 3,918 m²/g which is very rarely reported in literature, which can be used as an energy storage, a catalyst support or a super capacitor, for example.

Characterizations of function groups

The Fourier transformed infrared spectra of char and AC are shown in Figure 7. The FTIR spectrometer provides some information about chemical structure of the materials. As shown in Figure 7, some peaks were shifted, disappeared and new peaks were also detected in char and AC. The spectrum of char has peaks at 3,440, 1,692, 1,590, 1,375, 880, 812, 748 and 610 cm⁻¹ and the AC has peaks at 3,435, 1,710, 1,625, 1,360, 1,080 and 913–780 cm⁻¹. The band at 3,435 and 3,440 cm⁻¹ can be assigned to the O–H stretching vibration mode of hydroxyl functional groups and the band at 1,590,1,625 and 1,692 cm⁻¹ can be assigned to C=C symmetrical stretching of pyrone found on the lignin structure and C=O of carboxylic groups [51]. The peak at 1,080 cm⁻¹ for the AC is caused by stretching vibration of ester (–C–O) [52]. And the bands at 748, 812, 880 and 913–780 cm⁻¹ are produced by plane external bending of –C–H for different substituted benzene ring [53]. On comparison of the FTIR spectrum of char samples with AC, significant changes in the spectra are observed that the bands located in a range between 1,692 and 1,375 cm⁻¹ are, however, not present for the AC. This is probably due to the two different mechanism of heating methods [54].

Figure 7:

Fourier transform infrared spectroscopy (FTIR) spectra of the AC and char.

Microscopic structure analysis

The SEM images of char (a), AC (b) and TEM image of AC are shown Figure 8. It can be found from Figure 8(a) that the surface of char is relatively rough, containing little pores. The fine particles are mainly coal tar produced during carbonization. However, as shown in Figure 8(b), the impurities on the surface are removed with well-developed pore structure. Compared with Figure 8(a), the surface of AC becomes more smooth and aperture of pore looks large relatively. These observations are consistent with the previous characterization and analysis. Figure 8(c) shows that the inner structure of the AC under the optimum preparation conditions is viewed by the TEM. The carbon particles on the surface of AC are extremely small, which could easily reach nanoscale level. Moreover, the surface of AC seems to be covered with a layer of thin film, indicating lots of pore formed, which is in good agreement with SEM results.

Figure 8:

Scanning electron microscopy (SEM) images of char (a) and AC (b), TEM images of AC (c).