Adsorption performance and mechanisms of mercaptans removal from gasoline oil using core-shell 1 AC-based adsorbents : Role of intra-particles space 2


 Sulfur compound detection such as mercaptans in liquid fuels is undesirable because sulfur is the main sourcing emission of sulfur oxide (SOx) into the air. The use of activated carbon (AC) has proven to efficiently remove mercaptans. In the meantime, it is limited by the generation of the second pollution in oil, and the difficulties of recovery and regeneration. To address these issues, a core-shell structured AC with high mechanical strength and big intra-particles space were synthesized and demonstrated to efficiently remove organic pollutants from an aqueous solution without generation of the second pollution in our previous work. However, the performance, characteristics, and mechanism of mercaptans adsorption from gasoline oil by core-shell structured AC was still unclear. In this study, the mercaptans adsorption behaviors using core-shell powdered activated carbon (CSAC) and core-shell granulated activated carbon (CSGAC), along with raw PAC, PAC-core, raw GAC, and GAC-core were carried out. The results showed that both the CSAC and CSGAC adsorbents effectively removed sulfur-based pollutants and were provided with good recovery and recyclability without second pollution in gasoline oil. The CSGAC exhibited a higher mercaptans removal efficiency compared to those of CSAC as a result of the bigger intra-particles space. PAC-based adsorbents, presented the shrinking of removal efficiency after regeneration. The Pseudo-second-order kinetics and Langmuir isotherms models were dominated for mercaptans adsorption by both CSAC and CSGAC. Furthermore, the interactions between mercaptans and the composites were probably ascribed to the Van der Waals force, hydrophobic compatibility, pore texture, and π-π dispersion interaction.

based adsorbents (PAC and PAC-core), and 0.03 g of GAC-based adsorbents (GAC and GAC-core). Each 117 adsorbent was taken into a bottle of 100 mL and then sealed with rubber caps. Then, 50 mL of each sulfide 118 compound solution was introduced into the bottles which contain the adsorbents. The cellulose membrane 119 with 0.45 μm pore size was used for filtration of the supernatant after shaking the bottle samples with a 120 where q e and q t (mgS g -1 ) are the adsorption capacity at equilibrium and time t, R (%) is the removal 144 percentage, and C 0 and C e are the liquid-phase concentrations of mercaptans at initial time and equilibrium, 145 respectively. V (L) is the volume of the solution and M (g) is the mass of the CSAC. 146

Adsorption isotherms and kinetics 147
The capacity of mercaptans adsorption and isotherms parameters were calculated in virtue of the 148 adsorption isotherms model. The details for the calculation of the isotherm parameters were presented in 149 studies (Barzamini et al., 2014). 150 The Langmuir model and Freundlich isotherm are expressed as follows: equation (4) and (5) The corresponding introductions of the parameters were found in lots of studies (Ndagijimana et al., 2019a). 154 For further studying the adsorption characteristics, three kinetics models were applied, including pseudo-155 first-order, pseudo-second-order, and intra-particle models. The equations (6), (7), and (8) below were 156 applied to fit the kinetics data. 157 Pseudo-first order model: ln (q e − q t ) = lnq t − k 1 t (6) 158 Pseudo-second order model: Similarly, the introductions of the corresponding parameters were described in the study (Liu et al., 2019). 161

Characterizations 163
The physical image, the morphological information of CSAC, and its core and shell are presented in 164 Fig.1. Fig. 1a shows that the shell of the materials is porous, and favorable for adsorbate solution to diffuse mass diffusion which is good evidence for increasing removal efficiency for CSGAC. Figs. 1d-g present 170 the physical images of the CSAC, CSGAC, GAC-core, and PAC-core, respectively. As shown in the figure, 171 the GAC-core presented the larger intra-particles space compared to that of PAC, which is owing to the 172 larger size in GAC (2 mm). Figs. 1h and I show that the pore structure of the PAC core and GAC core after 173 the sintering and the detail is illustrated in our recently reported work (Ndagijimana et al., 2020). The 174 structure of the materials indicates the presence of pores which is beneficial for the adsorption of the 175 aforementioned pollutants in gasoline oil. 176 The textural structure such as specific surface area calculated by BET equation (S BET ) and pore Table 2, the CSAC presented a higher S BET compared to that of CSGAC. Moreover, S BET of the PAC-core 179 and GAC-core detached from shell were higher than those of PAC and GAC, respectively, because of 180 sintering of the sample at high temperature (1250 o C).

Effect of adsorbent dosage 187
One, two, and three balls of CSAC and CSGAC were used to investigate the influence of adsorbent 188 dosage (Fig. 3a). As shown in the figure, the removal efficiency of mercaptans and the dosage of adsorbents 189 were positively correlated. For instance, the removal efficiency was 33, 67, and 100% for ethanethiol and core inside the shell. Despite a low removal mercaptans efficiency by CSAC, it should be noted that as-196 synthesized protected AC materials showed a promising future for removing mercaptans from the gasoline 197 oil. As illustrated in Figs. 3b and c, the dosages used in this experiment were 0.03, 0.05, and 0.10 g for 198 PAC, PAC-core, GAC, and GAC-core, respectively. The results also showed that the removal efficiency 199 increased with adding more adsorbents as a result of more active sites. In addition, the performance of 200 mercaptans adsorption by PAC-core and GAC-core were better than those of their counterparts (PAC and 201 GAC). This phenomenon is likely due to the higher S BET after sintering treatment at high temperatures. 202 Accordingly, the large intra-particles space in shell and AC-core and high S BET are key factors enhancing 203 the adsorption of mercaptans using CSGAC. 204

Effect of contact time 205
The effect of contact time on mercaptan adsorption is shown in Fig. 4. It is observed from Figs. 4a and 206 b that the equilibrium of ethanethiol adsorption attained within 420-480 min and 600 min for CSGAC and 207 CSAC, respectively. The results demonstrate a faster adsorption rate of ethanethiol on CSGAC compared 208 to that of CSAC. Concerning the 1-butanethiol, the adsorption equilibrium reached 720 min for CSAC and 209 around 540-600 min for CSGAC, respectively. This phenomenon similarly indicates a higher removal 210 efficiency of mercaptans using CSGAC. These outcomes could be resulted from the large porous channels 211 in a shell for CSGAC and in GAC-core. Furthermore, the removal rate of 1-butanethiol is slower than 212 ethanethiol by both adsorbents, which is likely due to the short-chain of ethanethiol molecular. This was 213 evidenced by (Barzamini et al., 2014), that the adsorption of mercaptans with long-chain would take a where and T are presenting the universal gas constant (8.314 J mol -1 K -1 ) and the absolute temperature. 277 Plotting ln (q e /C e ) versus 1/T shows a linear line with slope and intercept similar to -ΔH°/R and ΔS°/R, 278 The positive value of ΔH° in Table 6, Table S3, and Table S4 indicate that ethanethiol and 1-butanethiol 280 adsorption by CSAC, CSGAC, PAC-core, GAC, and GAC-core are endothermic processes. The data for 281 PAC, PAC-core, GAC, and GAC-core were presented in supporting information Table S3 and Table S4. 282 The negative value of ΔH° (Table S3)

Adsorption competition 289
Except for the ethanethiol and 1-butanethiol, 1-propanethiol was used as a coexisting organics to 290 investigate adsorption competence between these molecular by CSAC and CSGAC. In this study, the 291 solution of ethanethiol, 1-propanethiol, and 1-butanethiol with a concentration of 1000 ppb in a bottle of 292 100 mL was used to investigate the capability of the material in the competition of adsorption to different 293 kinds of mercaptans. Fig. 8 depicts that the ethanethiol is highly adsorbed while the 1-butanethiol is weakly 294

Availability of data and materials 397
Regarding the data and material in this manuscript, all authors concluded that, the data available online 398 (DOI link provided by Journal after publication) will be shared according to the requirement of this 399 Barzamini R, Falamaki C, Mahmoudi R (2014) Adsorption of ethyl, iso-propyl, n-butyl, and iso-butyl mercaptans on       and PAC) and 0.03 g (GAC-core and GAC). 597 Fig. 9. Thermal regeneration of mercaptans by core-shell AC (a) and PAC and GAC based adsorbents (b, c). 598