3.1. Characterization
Figure 1 shows a SEM-EDX image of typical DT0 samples before and after modification.
The EDX analysis revealed that the pristine sample contained, besides carbon, the following elements: oxygen, sodium, alumina, silicon, potassium, calcium, and iron. After acid treatment, Na, K, Ca, and Fe were completely removed. Al and Si were partially removed. The quantitative results of EDX measurements are presented in
Table 1. The EDX analysis was confirmed by XRF measurements. X-ray fluorescence spectroscopy (XRF) is a powerful analytical technique that provides both qualitative and quantitative information on a wide variety of sample types including solids, liquids, slurries, and loose powders. Theoretically, it can quantify elements from sodium up to uranium. Practically the detection of small amounts of sodium is difficult. This is the reason that there is no carbon, oxygen, and sodium in
Table 2 concerning elemental analysis based on XRF measurements. These elements are listed in
Table 1—with the elemental analysis on the basis of EDX measurements. The quantitative results of XRD and XRF results were comparable, and these different techniques allowed us to draw the same conclusions. The elements present in DT0 were completely or partially removed after acid treatment.
According to XRF measurements, alumina was completely removed after acid treatment. However, EDX investigation show the presence of Al below 1%. Alumina is less soluble in acids than Na, K, Ca, and Fe. It is well known that the solubility of silicon in acids is very low, but we were not able to identify silicon compounds if they were amorphous. Thus, we can say that the silicon and its compounds present in DT0 were very difficult to dissolve in acid, which came as no surprise.
The XPS analysis enables quantitative determination of functional carbon groups [
71]. Depending on the kind of bonding and amount of oxygen atoms attached the shift of carbon signals towards higher binding energies can be observed. The C1s signals presented in
Figure 2 were carefully deconvoluted to components according to the procedure given elsewhere [
72].
Table 3 presents the elemental content of the surface expressed as atomic concentration.
The tail on the left-hand side of the dominant component is related with the presence of carbon-oxygen species. In DT0_HNO
3 and DT0_HCl_HNO
3 samples a hump at about 289 eV is observed. This can be attributed to a significant increase of COOH groups over the carbon surface. It is clearly the oxidation effect of HNO
3.
Table 4 presents surface elemental contents of the samples.
Figure 3 presents the XP spectra of the analyzed samples.
The content of calcium decreases as a result of acid treatment. The higher content of silicon after acid treatment is most probably the effect of removing the calcium screening effect. In the case of DT0_HCl sample, the increase of oxygen content is observed. This is not the result of oxidation but rather an effect of calcium removal as HCl is not an oxidizing agent. Such effects were observed previously [
72].
Figure 4 shows the FTIR spectra of activated carbons.
Figure 4 shows FTIR spectra of DT0 samples. The signal from the carboxylic group (1715 cm
−1) was not observed for DT0 and DT0_HCl. Only samples treated with HNO
3 had COOH groups on the surface. Additionally, a weak signal from the hydrophilic group (1340 cm
−1) was observed only for the carbons treated with HNO
3. The absorption band appeared at 1548 cm
−1 corresponding to the C=O stretching vibration [
73]. This signal was present for all the samples, but for samples treated with HNO
3, it was considerably higher. The presence of C=O groups on the surface of DT0 treated with HNO
3 was enhanced. The intensity of signals attributed to functional groups containing oxygen was as follows: DT0_HCl_HNO
3 > DT0_HNO
3 > DT0_HCl ≈DT0. The broad peak located at about 1070 cm
−1 related to the C–O stretching vibration of various functional groups [
74] and a very small peak located about 1170 cm
−1 related to C-C/C-O [
75] were present on the surface of all samples. The FTIR results were consistent with the XPS results.
The acid-treated DT0 carbons were characterized with Raman spectroscopy and are shown in
Figure 5.
The G peak, centered around 1595 cm
−1, can be assigned to the ordered carbon structure. G-band arises from the stretching of the C-C bond in graphitic materials and is common to all sp
2 carbon systems. In the graphitic disorder of the carbon structure, the band around 1310 cm
−1 named D was observed. The intensity ratio of the G-band and D-band (I
G/I
D) was used to evaluate the quality of carbon materials. The higher the I
G/I
D value, the lower the graphitic disorder observed [
76].
Table 5 shows the determined values of the I
G/I
D ratios.
The values of I
G/I
D for carbons were in the range of 0.65–0.72 and were similar to the previously reported for activated biocarbons [
77]. The highest I
G/I
D ratios were observed in the samples modified with acids in comparison to the unmodified DT0, which correspond to the better ordered structure in modified carbons.
The graphitic structure of DT0 carbons was analyzed with the XRD method. The X-ray diffraction profiles for carbons are shown in
Figure 6.
The diffractograms showed broad peaks and the absence of very sharp peaks. This indicates that the carbon was amorphous in nature [
78]. The amorphous structure of the DT0 was identified by the peak at 2θ = 24° and peak at 2θ = 43°, which referred to the diffraction peaks (002) and (101) [
79,
80,
81].
The carbon morphology of studied samples at 50,000× magnifications is shown in
Figure 7.
SEM images of the DT0 carbons showed that the adsorbent had texture with a heterogeneous surface and a variety of randomly distributed holes. The presence of holes of various diameters and irregular shapes was observed in DT0 HCl and DT0_HCl_HNO
3. Contrary to that, a very smooth surface was identified for DT0. Similar shapes were observed for the other commercial activated carbons [
82,
83,
84].
The proposed methods of DT0 modification with inorganic acids influenced the textural properties.
Figure 8 shows the nitrogen adsorption isotherms at −196 °C of the studied DT0 carbons.
The presented isotherms combine type I, which is characteristic of microporous materials, and type IV, which is characteristic of mesoporous materials [
85]. Nitrogen adsorption increased very quickly at low P/P
0 values, which is a characteristic feature of microporous materials. The formation of hysteresis loops at relative pressure higher than 0.4 indicated the multilayer adsorption process characterizing mesoporous structures. The hysteresis of catalysts was the type H4 associated with narrow slit pores, including pores in the micropore region. These are typical sorption isotherms for activated carbons [
77,
86].
Figure 9 shows the pore size distributions for the ACs determined by means of the N2 adsorption results at −196 °C using the density functional theory (DFT). Overall, it can be seen that the ACs have mainly micropores with a certain amount of mesopores.
It was stated that the acid-treated DT0 carbon samples showed significant differences in pore volume size compared to the pristine sample. In the case of DT0 carbon modified with nitric acid and the two-stage treatment with hydrochloric acid and nitric acid, a significant reduction in the pore volume in the range of 0.4–2.5 nm was observed. In contrast, treatment of DT0 with hydrochloric acid significantly increased the pore volume. This is because the treatment of AC with hydrochloric acid removes minerals, which in turn may increase the porosity. The DT0_HCl carbon had the best-developed micropores in the range of 0.4–2 nm. All the DT0 carbons contained mesopores with a size of 2–3.5 nm as well (
Figure 9).
Table 6 presents the textural parameters of DT0 carbons.
It was found that the treatment of DT0 carbon with HCl, HNO3, HCl, and HNO3 acids affected its porous structure to a greater or lesser extent. Treatment with HCl increased the surface area of the starting DT0 carbon from 1085 to 1267 m2/g and increased the total pore volume from 0.583 to 0.687 cm3/g. Thus, the increase of textural parameters in DT0_HCl may be caused by removing Na, K, and Ca salts by dissolving them in HCl, which promotes the development of the surface and pore structure.
It was evident from the EDX (
Table 1) and XRF (
Table 2) measurements that the minerals were dissolved with HCl and HNO
3. Unfortunately, HNO
3 simultaneously oxidized the carbon structure and damaged the micro- and mezopores. That is why only the S
BET of DT0_HCl increased (
Table 6). The advantage of HNO
3 treatment was the increase of oxygen-containing groups on the surface, especially COOH. The concentration of oxygen on the surface of DT0_HCl_HNO
3 was similar to DT0_HNO
3, but the S
BET and V
tot were higher than those of DT0_HNO
3 because of the simultaneous action of both acids.
The effect of HNO3 had the greatest effect on reducing the specific surface area from 1085 to 532 m2/g, and the total carbon pore volume after modification was the smallest and amounted to 0.311 cm3/g. A fairly significant reduction in SBET resulted from partial oxidation of carbon by strongly oxidizing nitric acid (V).
A significant reduction in textural parameters was also found in the case of carbon modified with HNO
3 after previous modification with non-oxidizing HCl. With this two-step modification, the S
BET of DT0_HCl_HNO
3 decreased from 1085 to 651 cm
3/g and the total pore volume decreased to 0.375 cm
3/g. However, this suggests that the double modification of carbon compared to the treatment of DT0 only with the oxidizing acid HNO
3 had a more favorable effect on the development of porosity of the material. Similar conclusions leading to an increase in the porosity of carbons after their treatment with acids can be found in other studies [
72,
87]. There are also reports showing a decrease in the specific surface area of the carbons after acid treatment [
88,
89].
3.2. Alpha-Pinene Isomerization Process
In the first stage of catalytic tests, studies on alpha-pinene isomerization were carried out on DT0 carbon and on DT0_HCl, DT0_HNO
3, and DT0_HCl_HNO
3 carbons. The aim was to determine the most active catalyst for the process of α-pinene isomerization. At this stage, the following conditions were used in the tests: temperature 160 °C, the content of catalyst at 5 wt% (in relation to the amount of alpha-pinene), and reaction time of 3 h.
Figure 10 shows the results of isomerization of alpha-pinene obtained in the study.
Figure 10 shows that the most active catalyst was DT0_HCl_HNO
3. It allows for achieving the maximum conversion of alpha-pinene (100 mol%). The selectivities of the transformation to camphene and limonene on this catalyst were 41 and 13 mol%, respectively. The formation of alpha-terpinene (selectivity 13 mol%), gamma-terpinene (selectivity 6 mol%), terpinolene (selectivity 11 mol%), and p-cymene (selectivity 5 mol%) could also be observed on this porous material. Similar values of transformation selectivity were observed for the DT0_HNO
3 catalyst, while the alpha-pinene conversion for the process carried out on this catalyst was lower by 21 mol%.
The results obtained on the other two catalysts are much worse, as the conversion on them reaches 16 mol% (DT0 catalyst) and 24 mol% (DT0_HCl catalyst). However, in the case of these two catalysts, attention is drawn to the very high selectivity of the transformation to limonene reaching the value of 36 mol% (DT0 catalyst) and 39 mol% (DT0_HCl catalyst), with the selectivity of transformation to camphene similar to that obtained on the previous two catalysts (44 mol% DT0 catalyst and 41 mol% DT0_HCl catalyst). On the other hand, the transformation selectivities to the remaining products are not significant for these two catalysts.
In the catalytic tests presented in this stage of the studies, the synergistic effect of HCl and HNO3 acids on DT0-activated carbon is also visible. It is visible in the fact that for the DTO_HCl_HNO3 sample, the selectivity of transformation to terpinolene increased compared to the unmodified DT0 sample and the DTO_HCl and DT0_HNO3 samples (more than twofold increase). A similar effect was seen with p-cymene, gamma-terpinene, and alpha-terpinene. The synergistic effect of HCl and HNO3 on the DT0 carbon simultaneously reduced the selectivity of transformation to limonene as compared to the unmodified DT0 sample (by about 20 mol%).
The comparison of the results presented in
Table 3 (tests using the XPS method) shows that the DT0_HNO
3 and DT0_HCl_HNO
3 catalysts differ significantly (almost 3 times) from the other two catalysts in the content of carboxyl groups (COOH group). They also have slightly more C-OH groups (DT0 and DT0_HCl with 12.6 and 12.3, and DT0_HNO
3 and DT0_HCl_HNO
3 with 13.7 and 14.2, respectively). It should also be noted that, in the case of DT0_HNO
3 and DT0_HCl_HNO
3 carbon materials, no keto-enolic groups were observed. This increased content of COOH and C-OH groups in the DT0_HNO
3 and DT0_HCl_HNO
3 carbon materials and the lack of keto-enolic groups in them may be the reason for the greater activity of DT0_HNO
3 and DT0_HCl_HNO
3 in the alpha-pinene isomerization process. It seems more advantageous to wash the DT0 carbon with the aqueous solution containing both nitric acid and hydrochloric acid, and not with the aqueous solution of the nitric acid alone.
Based on the research results presented at this stage, the DT0_HCl_HNO3 catalyst was considered to be the most active.
The research on the effect of temperature on the course of alpha-pinene isomerization on the DT0_HCl_HNO
3 material is shown in
Figure 11. The tests were carried out at the following temperatures: 60, 90, 120, 145, 160, and 175 °C, for 3 h and with catalyst content of 5 wt%.
The comparison of the results obtained at the two lowest temperatures (60 and 90 °C) shows that these temperatures are too low to carry out the alpha-pinene isomerization process, as the conversion of this raw material is only 1 and 14 mol%. In addition, the process temperature increasing from 60 to 90 °C resulted in a decrease of the selectivity of transformation to camphene (from 52 mol% to 44 mol%). The further increase of temperature to 145 °C did not cause any significant changes in the selectivity of the transformation to camphene. Comparing the results obtained for the two lowest temperatures, it can also be noticed that increase of temperature increases the selectivity of transformation to limonene (from 20 to 26 mol%), terpinolene (from 0 to 7 mol%), alpha-terpinene (from 0 to 4 mol%), gamma-terpinene (0 to 3 mol%), tricyclene (1 to 3 mol%), and p-cymene (0 to 2 mol%). The highest alpha-pinene conversions were obtained by carrying out the isomerization process at temperatures of 120 and 145 °C at 29 and 41 mol%, respectively. The highest alpha-pinene conversions were obtained by carrying out the isomerization process at temperatures of 160 and 175 °C (99 and 100 mol%, respectively). At these temperatures, the highest selectivity of transformation to several products was also obtained: tricyclene—3 and 6 mol%, alpha-terpinene—10 and 17 mol%, gamma-terpinene—5 and 8 mol%, and terpinolene—10 and 13 mol%. The highest selectivity of transformation to limonene was obtained at temperatures of 120 and 145 °C, then its value decreased (to 7 mol%). As the process temperature increased, an increase in the selectivity of the transformation to the dehydrogenation product (p-cymene) was also observed, and at the highest temperatures, it was about 4–7 mol%.
To study the effect of reaction time on alpha-pinene isomerization, three temperatures were selected at which the highest values of alpha-pinene conversion were obtained: 145, 160, and 175 °C. The tests were carried out with DT0_HCl_HNO
3 catalyst in the reaction mixture of 5 wt% and in the range of reaction times from 5 to 180 min. The results are presented in
Figure 12,
Figure 13 and
Figure 14.
A comparison of the results obtained at the three tested temperatures shows that the alpha-pinene conversion of about 100 mol% can be achieved by carrying out isomerization at the two highest temperatures, while for the temperature of 160 °C the time required to achieve such a high conversion is 180 min, and for the temperature of 175 °C the time required is 100 min, which is significantly shorter. The maximum selectivities of the transformation to camphene obtained at the three tested temperatures are similar and amount to 41–44 mol%. The highest selectivity of tricyclene was noted for the process carried out at the temperature of 175 °C and for the reaction time of 160–180 min (6 mol%). At the same temperature of 175 °C and for the same reaction time, the highest selectivity of alpha-terpinene (about 17 mol%), gamma-terpinene (about 8 mol%), terpinolene (about 13 mol%), and p-cymene (4 mol%) was also observed.
Figure 12,
Figure 13 and
Figure 14 also show that the highest selectivity of camphene (40–44 mol%) can be achieved by carrying out the isomerization process: longer than 40 min (the most advantageous are 40–50 min) at 145 °C, for 40–50 min at 160 °C, or for 10–50 min at 175°C. The comparison of the results for the selectivity of transformation to limonene shows that the selectivity of this compound decreases with increasing the reaction time and increasing the temperature, and the highest selectivity of the transformation to this compound (about 24–25 mol%) is obtained for short reaction times: for the temperature of 145 °C for a reaction time up to 30 min, for temperatures of 160 °C and 175 °C for a time reaction time not exceeding 10 min.