Influence of activation conditions on textural properties and performance of activated biochars for pyrolysis vapors upgrading

1 Influence of activation conditions on textural 1 properties and performance of activated biochars 2 for pyrolysis vapors upgrading 3 Christian Di Stasi1*, Gianluca Greco1, Rafael L. S. Canevesi2, M. Teresa Izquierdo3, Vanessa 4 Fierro2, Alain Celzard2, Belén González1, Joan J. Manyà1 5 1 Aragón Institute of Engineering Research (I3A), Technological College of Huesca, 6 University of Zaragoza, crta. Cuarte s/n, Huesca, E-22071, Spain 7 2 Université de Lorraine, CNRS, IJL, Épinal, F-88000, France 8 3 Instituto de Carboquímica (ICB-CSIC), Miguel Luesma Castán 4, Zaragoza, E-50018, Spain 9 10 * Corresponding author at: Aragón Institute of Engineering Research (I3A), Technological 11 College of Huesca, University of Zaragoza, crta. Cuarte s/n, Huesca, E-22071, Spain. 12 E-mail address: [email protected]s. 13 2 ABSTRACT 14 The main aim of the present study is to provide a comprehensive assessment of the effects of 15 process activation conditions on the textural properties of the resulting activated carbons, 16 which were produced from wheat straw-derived biochar through chemical activation (with 17 K2CO3 at different pressures and mass impregnation ratios) and physical activation (with CO2 18 at different temperatures and pressures). For chemically activated biochars, it was found that 19 specific surface area and pore size distribution were both only positively affected by increasing 20 the carbonate loading. However, physically activated biochars produced at the highest pressure 21 and lowest temperature (1.0 MPa and 700 °C) had the highest surface areas and widest pore 22 size distributions. The materials with the most appropriate textural properties were then tested 23 as catalysts for steam and dry reforming of the aqueous phase of pyrolysis oil. The best catalytic 24 performance (a total gas yield of 74% and a selectivity toward H2 of almost 40%) was observed 25 for a physically activated biochar. This good performance was ascribed to the high availability 26 of K0 on the catalyst surface, which could effectively promote the reactions involved in the 27 upgrading process. 28 Keywords 29 Biochar; chemical activation; K2CO3; physical activation; pressure; steam and dry reforming 30 of pyrolysis oil 31 32 3 1. Introduction 33 In recent years, the demand for activated carbons has increased due to their usefulness in a 34 wide range of different applications such as gas and liquid purification [1], electrochemistry 35 [2], soil remediation [3] and catalysis [4]. Thanks to their versatility to be adapted to specific 36 applications, these materials are considered promising candidates to address environmental 37 issues related to global warming and pollution [5]. Nevertheless, the current main drawback is 38 that the production of activated carbons is still partly based on fossil fuels, which do not meet 39 sustainability criteria [6,7]. Hence, much more efforts should be made to produce much more 40 carbons from biomass. However, the direct conversion of biomass feedstock into the final 41 product has a low yield. Thus, the production of activated carbons from biochar produced by 42 slow pyrolysis seems to be more appropriate in terms of scalability, while at the same time 43 these value-added materials can strengthen the value chain of existing biochar production 44 systems. 45 In general, pristine biochar has neither a well-developed surface area nor a hierarchical pore 46 size distribution, which is mainly dominated by narrow pores with a diameter (dp) lower than 47 0.7 nm (ultra-micropores). Hence, a subsequent activation step is required to improve the 48 textural properties of the carbon material. The activation process involves the development and 49 opening of the porosity of a char using an activation agent. Depending on the agent used, the 50 procedure can be called chemical or physical activation. Chemical activation usually involves 51 two steps. First, the biochar is impregnated with an aqueous solution of the chemical activation 52 agent, or mixed with it in the dry state, and, in a second step, the blend is heated up to a given 53 temperature at which oxidation, dehydration, aromatization and crosslinking reactions, among 54 others, occur. Although the most widely used chemical reagents are KOH [8] and H3PO4 [9], 55 non-hazardous and relatively cheap alternative compounds such as K2CO3 [10,11] have 56 recently attracted considerable attention. Mai et al. [12] have recently reported that activation 57 4 with potassium carbonate can result in carbons with a high percentage of structural defects and 58 a well-balanced porosity between microand mesopores. 59 When activation is carried out by exposing the precursor to relatively high temperatures 60 under an oxidizing atmosphere (e.g., CO2 [13], H2O [14] and O2 [15]), the process is called 61 physical activation. Through either physical or chemical activation, it is possible to tune the 62 textural features of the starting biochar by properly adjusting the activation conditions, mainly 63 temperature, type and/or concentration of activation agent, as well as pressure. To the best of 64 our knowledge, although both activation procedures are widely reported in the literature, a 65 thorough study on the effects that the activation parameters and their possible interactions have 66 on the textural properties of the resulting activated biochar, has never been reported so far. 67 Such information could be very helpful to properly establish the most appropriate operating 68 conditions to produce engineered carbon materials from biomass. 69 One of the most interesting fields of application of activated carbons is the upgrading of raw 70 pyrolysis vapors, which contain both permanent gases (e.g., CO2, CO, CH4, and H2) and 71 condensable compounds. The condensable fraction, which is generally called pyrolysis oil or 72 bio-oil, is a mixture of hundreds of organic compounds (alcohols, ketones, acids, etc.) and 73 water, and its composition strictly depends on the biomass composition and the pyrolysis 74 operating conditions [16]. For biochar production systems based on slow pyrolysis, 75 downstream processes aimed at upgrading pyrolysis vapors are required to avoid undesirable 76 condensation of organic compounds and to increase simultaneously the quality of the gaseous 77 product. For this purpose, combined steam and dry reforming of pyrolysis oil is a promising 78 option [17,18], due to the presence of a relatively high amount of CO2 and steam in the raw 79 vapor phase. Heterogeneous catalysts based on transition metals such as Ni [19,20], Co [21], 80 Pt and Rh [22] can be used to improve the overall conversion and the selectivity of the products. 81 However, the main drawback of these catalysts is that the relatively expensive active phase can 82 5 easily be poisoned and/or deactivated by deposition of coke [23]. One possible solution to 83 reduce the overall cost of the upgrading process is to use activated biochar as catalyst. Its 84 porous structure and inherent inorganic contents (especially K, Mg and Ca) could result in a 85 relatively good catalytic activity [24–29]. Furthermore, due to the presence in the reaction 86 system of H2O and CO2, the carbonaceous support, as well as coke deposits, could be 87 continuously partly gasified, thus creating new pores to avoid deactivation. In addition, the 88 spent biochar can be burned to recover energy [30] or, depending on the metal loaded on the 89 support, employed as soil conditioner [31]. 90 Keeping in mind all the above, the main objective of this study was to perform a preliminary 91 investigation about the effects of several activation conditions on the textural properties and 92 the catalytic activity of the resulting activated biochars. To this end, wheat straw-derived 93 biochar was activated chemically and physically with K2CO3 and CO2, respectively, under 94 different operating conditions. To objectively assess the effect of the selected operation 95 conditions (temperature and mass ratio K2CO3/precursor for chemical activation, and 96 temperature and pressure for CO2 activation), we implemented two factorial designs of 97 experiments, with two factors and three replicates at the center point. The specific surface areas 98 and pore size distributions of the resulting activated biochars were then evaluated. The most 99 promising activated biochars were finally tested as catalysts for steam and dry reforming of 100 pyrolysis oil. Their performance was evaluated in terms of conversion of liquid into gas, 101 product selectivity and resistance to deactivation. 102 2. Experimental Section 103 This study was divided into two main steps: in the first, we carried out physical and 104 chemical activations under different process conditions in order to study their influence on 105 the pore size distribution and specific surface area of the resulting activated biochars; in the 106 second part, the most promising materials were then tested as catalysts in pyrolysis vapors 107 6 upgrading experiments. The methodology used in this study is summarized graphically in 108 Fig. A.1. 109 2.1. Biochar production 110 The biochar used in this study was produced by slow pyrolysis of binder-free wheat straw 111 pellets (9 mm OD and 10–13 mm long). Pyrolysis was carried out at atmospheric pressure in a 112 fixed bed reactor, which was placed inside a furnace and heated up to 500 °C at an average 113 heating rate of 5 °C min–1 and using N2 as carrier gas. More details on the pyrolysis device and 114 the experimental procedure are available elsewhere [32]. The resulting biochar (i.e., “pristine”) 115 was ground and then thoroughly sieved to obtain particle sizes in the range of 0.212 to 1.41 116 mm. The pristine biochar was characterized by proximate analysis (performed in quadruplicate 117 according to ASTM standards D3173 for moisture, D3174 for ash, and D3175 for volatile 118 matter) and ultimate analysis by means of an elemental analyzer CHN628 from Leco 119 Corporation (USA). 120 2.2. Activation of pristine biochar 121 2.2.1. Chemical activation 122 For chemical activation, the pristine biochar was first impregnated with a 1 mol L−1 aqueous 123 solution of K2CO3. Three K2CO3: biochar mass impregnation ratios (1:1, 2:1 and 3:1) were 124 achieved by adjusting the volume of solution. The heterogeneous mixture was then stirred for 125 2 h at 50 °C, filtered and dried overnight at 110 °C to remove the residual water. Afterwards, 126 10 g of the impregnated samples were heated up to 700 °C at a heating rate of 10 °C min–1, 127 under an inert atmosphere (N2), and at three different values of absolute pressure (0.10, 0.55 128 and 1.00 MPa). For this purpose, a tubular fixed bed reactor (made of nickel-chromium alloy 129 UNS N06600, 28.1 mm ID and 600 mm long) placed in a vertical furnace (model EVA 12/300 130 from Carbolite Gero, UK) was used. The relatively low activation temperature was chosen to 131 avoid the evaporation of the metallic potassium derived from the decomposition of the 132 7 activation agent and, also, to ensure low activation extensions and clearly see the potential 133 effects of the studied parameters. On the other hand, the range of activating pressures was 134 chosen in order to allow this experimental setup to be easily scaled up, since the involved 135 pressures were not too high (up to 1.0 MPa). A soaking time at the highest temperature of 60 136 min was set. The pressure within the reactor was adjusted using a downstream servo-controlled 137 regulator valve. The gas hourly space velocity (GHSV) at the activation temperature was 138 estimated to be 7000 h–1, considering the pressure applied and a bed void factor of 0.5. Thus, 139 the mass flow rate of the inlet gas stream (N2) was properly adjusted as a function of the 140 selected pressure and the highest temperature to achieve the aforementioned GHSV value. 141 As a last step, the carbons were rinsed to remove the unreacted reagent and other impurities 142 from their surface. In this work, two different washing procedures were adopted to assess 143 possible effects on the resulting surface area and the catalytic activity of the activated carbons 144 produced. For this purpose, among all the unwashed chemically activated biochars, four of 145 them were just washed with hot deionized water (100 °C), while the seven chemically activated 146 biochars adopted in the design of experiments were washed with a 0.25 mol L−1 solution of 147 HCl followed by hot water. Both washing procedures were carried out until neutral pH. The 148 resulting activated biochars were then dried overnight at 110 °C. The efficiency of the washing 149 step was evaluated according to the washing yield, calculated from Eq. 1, in which mc and mw 150 are the masses of activated carbon before and after the washing steps, respectively. 151 𝑌𝑤𝑎𝑠ℎ = (1−𝑚𝑐−𝑚𝑤 𝑚𝑐)100 (1) 152 2.2.2. Physical activation 153 Physically activated biochars were produced under an atmosphere of pure CO2 at three 154 different temperatures(700, 775 and 850 °C), and at three different absolute pressures (0.10, 155 0.55, and 1.00 MPa). The selected activation temperatures are within the range commonly 156 reported in the literature, whereas the pressure values were established with the purpose to 157 8 compare the properties of resulting activated carbons produced under atmospheric and 158 moderate pressures. Using the same device described in the previous section, 10 g of pristine 159 biochar were heated under N2 atmosphere, at a heating rate of 10 °C min–1, until the target 160 temperature was reached. Then, the gas supply was switched from N2 to CO2 at a constant 161 GHSV of 7000 h−1. These conditions were maintained during the time required to reach a 162 degree of burnout (η, defined as the percentage of mass loss) in the range of 30% to 60%. 163 2.2.3. Design of experiments 164 To objectively assess the effects of the activation conditions on the textural properties of 165 activated biochars, an unreplicated two-level factorial design of experiments (with two factors 166 and three replicates at the center point) was adopted for both chemical and physical activations. 167 For chemical activation, the analyzed factors were the absolute pressure and the impregnation 168 ratio of K2CO3 to raw biochar, whereas in the case of physical activation, the temperature and 169 the absolute pressure were the studied factors. The structure of the regression model used in 170 the statistical analysis was the following: 171 ŷ = 𝛽0+𝛽𝑖𝐴+𝛽𝑗𝐵+𝛽𝑖𝑗𝐴𝐵 (2) 172 where A and B corresponded to the assessed factors (normalized values in the range from –1 to 173 1), whereas β0, βi, and βij were the intercept, linear, and interaction coefficients, respectively. 174 The results obtained for the selected response variables (y) were analyzed using the Minitab 175 v17 statistical package. A significance level of 5% was assumed and the adjusted coefficient 176 of determination (R2adj) was taken as an indicator of the quality of the fit. Table 1 summarizes 177 the adopted designs of experiments and lists the names of the activated biochars produced. 178 Briefly, activated biochar are referred as X_Y_P. X corresponds to the activation procedure (CB 179 or PB for chemically activated biochar and physically activated biochar, respectively); Y is the 180 activation temperature for PBs or the impregnation ratio for CBs; and P is the activation 181 pressure. 182 9 2.3. Catalytic pyrolysis vapors upgrading 183 The liquid feed used here for pyrolysis vapors upgrading tests was the filtered aqueous phase 184 of the pyrolysis oil formed during the production of pristine biochar derived from wheat straw. 185 This resulting liquid product was characterized in terms of elemental composition, using the 186 same CHN analyzer as mentioned in section 2.1, and water content, using a volumetric Karl187 Fischer titrator Titrino plus 870 from Metrohm (Switzerland). 188 A schematic overview of the experimental device used for the reforming tests is shown in 189 Fig. A.2 (Appendix A). The reactor and furnace elements were the same as those previously 190 described in section 2.2. The reforming operating conditions were selected based on the 191 findings of a previous study [33], which aimed to establish the best process conditions to 192 minimize deactivation by coke deposition. 193 Briefly, activated biochar (2–5 g) was loaded into the reactor and heated under N2 194 atmosphere at an absolute pressure slightly above 0.1 MPa. Once the bed temperature reached 195 the target value of 750 °C, CO2 was added to the inlet gas stream at a partial CO2 pressure of 196 0.02 MPa. The liquid feed was then injected into the gas stream at the inlet of the reactor by 197 means of a HPLC pump (model 521 from Analytical Scientific Instruments, USA). The liquid 198 hourly space velocity (LHSV) was kept constant and equal to 2 h–1 for all the experiments 199 conducted. The duration of the experiments was 60 min. The reactor outlet stream, consisting 200 of permanent gases, unreacted pyrolysis oil, and condensable side products, was forced to pass 201 through an ice-bath condensation train. The permanent gases were analyzed using a dual202 channel micro gas chromatograph (μ-GC 490 from Agilent, USA) equipped with TCD 203 detectors and two analytical columns (a Molsieve 5 A and a PolarPlot U). The known amount 204 of N2 fed was used as a tracking compound to calculate the yield of produced gas. 205 16 present study were also found to be strongly dependent on the amount of activating agent (see 299 Fig. 1). The chemical activation process was able to slightly broaden the original pore size 300 distribution of the pristine biochar, leading to a more hierarchical porous structure with a large 301 contribution of ultra-micropores and a slightly increased amount of mesopores. Almost all the 302 carried out N2 isotherms required more than 70 h to be completed, thus confirming the 303 predominant narrow microporous structure of chemically activated biochars (see Fig. A.3a), 304 which hindered the diffusion of N2 within the porosity. For both total and micropore volumes, 305 a statistically significant combined effect of pressure and impregnation ratio (AB) was found. 306 However, these effects were comparatively much weaker than that observed for the main effect 307 of the impregnation ratio (B). From a thermodynamics point of view, an increased pressure 308 shifts the equilibrium of the activation reactions to the left. The fact that the textural properties 309 of the resulting activated carbons were practically independent of pressure could indicate that 310 the chemical activation process was mainly kinetically controlled. 311 It should be pointed out that the overall curvature terms for S2D-NLDFT, Vt and Vmicro were 312 statistically significant (p-values below 0.05, as reported in Table A.2). This indicates that, in 313 further studies, our adopted factorial design should be expanded to a central composite design 314 to be able to evaluate the pure quadratic regression coefficients and then apply Response 315 Surface Methodology (RSM) for optimization purposes. 316 As shown in Table 2, after the carbons were washed only with hot water, there was a slight 317 decrease in both the specific surface area and the micropore volume, compared to those 318 measured for carbons washed with the acidic solution. This can be explained by the fact that 319 water could not completely remove the remaining carbonate (and other chemical species) from 320 the solid surface, thus causing blockage of some micropores. By focusing on the difference 321 between the washing procedures of activated carbons at the same loadings of K2CO3, the results 322 reported in Table 2 indicate that a higher activation pressure led to an increased amount of 323 17 product not soluble in water, i.e., to higher Ywash values. This finding agrees with results 324 reported by Malekshahian et al. [40], who concluded that high gasification pressures can limit 325 the volatilization of potassium. 326 3.2. Physical activation 327 It must be pointed out that an increase in absolute pressure also implied an increase in the 328 partial pressure of the reactant (CO2). In the literature, the effects of absolute pressure and 329 partial pressure of reactant are commonly assessed separately. Nevertheless, and as far as we 330 know, no previous studies have addressed the question of whether activation under pressure 331 can cause textural changes in the resulting physically activated biochars. Activation of 332 biomass-derived carbon had only been studied using supercritical water at high pressure [41], 333 but such conditions are too different from the present ones for allowing some comparison to 334 be done. 335 Table 3 lists the textural properties of the activated carbons produced, whereas Fig. 2 shows 336 the normal plots of the standardized effects obtained for each response variable. More detailed 337 statistical outcomes are reported in Table A.3. The most relevant PSDs and N2 adsorption 338 isotherms are shown in Figs. A.3b and A.5, respectively. As previously mentioned in section 339 2.2.2, the activation times (also shown in Table 3) were varied to obtain degrees of burnout in 340 the range of 30% to 60 %, since the gasification rate was strictly dependent on reaction 341 temperature and reactant partial pressure. As expected, the reaction rate was minimal at the 342 lowest temperature (700 °C). In fact, and whatever the pressure applied, an activation time of 343 180 min was required to reach the desired reaction extent. For activation temperatures of 775 344 and 850 °C, an activation time of 60 min was enough, except for the material PB_850_1.00, 345 for which this reaction time led to an almost complete gasification of the sample (η = 85%). 346 Hence, the activation time at the highest levels of temperature and pressure was finally set at 347 30 min to obtain a degree of burnout of 57%. 348 18 Table 3. Degrees of burnout activation times and textural properties for physically activated 349 biochars 350 Material Activation time Burnout Apparent specific surface area Pore volumes tact η ABET S2D-NLDFT V0.97 Vt Vultra Vmicro Vmeso min % m2 g–1 cm3 g–1 PB_700_0.10 180 30.1 552 833 0.237 0.243 0.184 0.227 0.016 (6.77%) c PB_700_1.00 180 52.8 743 1008 0.333 0.333 0.226 0.301 0.032 (9.55%) c PB_850_0.10 60 54.9 600 818 0.263 0.266 0.169 0.240 0.026 (9.64%) c PB_850_1.00 30 56. 7 688 881 0.298 0.296 0.176 0.273 0.022 (7.57%)c PB_775_0.50 60 43.7 719 931 0.314 0.312 0.176 0.285 0.027 (8.62%)c PB_775_0.50 60 42.2 707 936 0.309 0.306 0.171 0.282 0.025 (8.02%) c PB_775_0.50 60 42.4 707 949 0.308 0.306 0.174 0.282 0.023 (7.64%) c PB_650_0.10 d 60 12.5 68.8 482 0.029 0.126 0.100 0.124 0.001 (1.07%) c PB_650_1.00 d 180 25.0 447 750 0.191 0.214 0.154 0.200 0.014 (6.46%) c c Calculated as 𝑉𝑚𝑒𝑠𝑜 𝑉𝑡 100 351 d Additional carbons that were not included in the design of experiments. 352 It was clear that a high pressure accelerated carbon gasification, even at 700 °C. At this 353 temperature, an increase in pressure from 0.1 to 1.0 MPa led to a marked increase in the 354 burnout, from 30.1% to 52.8%, due to the higher reaction rate. In this sense, an increased CO2 355 partial pressure resulted in a higher fraction of reactant adsorbed on the surface of the sample 356 [42]. These outcomes are in agreement with the study conducted by Malekshahian et al. [42], 357 but disagree with the results reported by Fermoso et al. [43], who observed that gasification of 358 heartwood was enhanced using low partial pressures of CO2. 359 19 Fig. 2. Normal plots of standardized effects (α = 0.05) for physically activated biochars: 360 specific surface area (a); total pore volume (b); mesopore volume (c); micropore volume (d); 361 ultra-micropore volume (e). 362 The specific surface areas reported in Table 3 were in line with expectations, given the 363 present activation conditions. Even at the lowest temperatures (700 °C) it was possible to obtain 364 relatively high surface areas, even greater than those measured for activated biochars produced 365 at higher temperatures. As can be seen in Fig. 2a, the specific surface area was positively 366 influenced by the absolute pressure and negatively affected by the activation temperature. 367 Especially at 1.0 MPa, relatively low temperatures and longer activation times allowed the 368 biochar to be gasified more homogeneously, thus resulting in a more developed porous 369 151050-5-10 99 95 90 80 70 60 50 40 30 20 10 5 1 Standardized Effect (%) AB B A S2D-NLDFT (a) 20151050-5-10 99 95 90 80 70 60 50 40 30 20 10 5 1 A Pressure B Impr. Ratio Factor Name Standardized Effect Percent Not Significant Significant Effect Type AB B A Normal Plot of the Standardized Effects (response is Vtot, α = 0.05) 151050-5-10 99 95 90 80 70 60 50 40 30 20 10 5 1 A Temperature B Pressure Factor Name Standardized Effect Percent Significant Effect Type AB B A Normal Plot of the Standardized Effects (response is S2D-NLDFT, α = 0.05) 151050-5-10 99 95 90 80 70 60 50 40 30 20 10 5 1 Standardized Effect (%) AB B A Vt (b) 5.02.50.0-2.5-5.0 99 95 90 80 70 60 50 40 30 20 10 5 1 Standardized Effect (%) AB B A Vmeso (c) 20100-10-20 99 95 90 80 70 60 50 40 30 20 10 5 1 Standardized Effect (%) AB B A Vmicro (d) 1050-5-10 99 95 90 80 70 60 50 40 30 20 10 5 1 Standardized Effect (%) AB B A Vultra (e) 20 structure. Conversely, higher temperatures led to faster reaction rates, which can result in a 370 more severe enlargement of micropores and the subsequent loss of surface area. 371 Fig. 2b shows that the total pore volume increased significantly when the pressure was 372 raised from 0.1 to 1.0 MPa. In general, the total pore volumes for physically activated biochars, 373 which were not significantly affected by the activation temperature, were higher than those 374 measured for chemically activated carbons. As can be seen when comparing the data reported 375 in Tables 2 and 3, physical activation led to porous carbons with slightly higher ultra-micropore 376 volumes. An inspection of Figs. 2d, 2e and A.3b reveals that high temperatures can lead to 377 some widening of the narrowest micropores, since the activation temperature had a negative 378 effect on the ultra-micropore volumes and a positive one on the micropore volumes. With 379 regards to the volume of mesopores, which was mainly affected by pressure (see Fig. 2c), it 380 should be noticed that physical activation under pressure resulted in more hierarchical porous 381 structures with higher contributions of mesopores (see values of Vmeso and relative percentages 382 in Table 3). The statistically significant overall curvature terms reported in Table A.3 for S2D383 NLDFT, Vt, Vmeso, and Vmicro response variables also suggest that a central composite design could 384 be required for optimization purposes. 385 Given the significant effects of pressure on the textural properties of carbons activated at 386 700−850 °C, we decided to perform two additional physical activations at 650 °C, leading to 387 the materials called PB_650_0.10 and PB_650_1.00. The results obtained, which are also 388 reported in Table 3, seem to confirm that activation with CO2 under pressure is a very 389 interesting way to produce biomass-derived porous carbons with high specific surface area and 390 wide pore size distributions (including relatively high mesopore volumes), even at relatively 391 low temperatures. 392 393 21 3.3. Catalytic activity 394 Due to their relatively high specific surface area, both PB_700_1.00 and CB_3_0.10 395 activated biochars (one for each activation procedure) were selected as catalysts and tested 396 during the upgrading process of the aqueous phase of a real pyrolysis oil. 397 Elemental and moisture analyses of the aqueous phase of the pyrolysis oil revealed that the 398 liquid sample had an average chemical formula of C5 H6.7 O2 and a water content of 80 wt. %. 399 Since the liquid was fed to the reformer without providing additional water, the steam to carbon 400 molar ratio, S:C, was 4:1. 401 It is generally assumed that deactivation and/or instability of the catalyst can be attributed 402 to two different phenomena: (1) the deposition of carbonaceous material (coke) on the surface 403 of the catalyst, which clogs part of the available active sites; and (2) the extent of gasification 404 reactions (both with steam and CO2) of the carbon-based catalyst, which leads to a loss of mass 405 and a subsequent increase in LHSV (i.e., less contact time). 406 The results obtained in terms of total gas yield, selectivity towards specific gaseous species, 407 hydrogen yield and hydrogen release over time are displayed in Fig. 3a. It should be noted that 408 the poor results measured for the blank test (i.e., empty reactor) suggested that the reactor wall 409 did not play a crucial catalytic role. The physically activated biochar (PB_700_1.00) exhibited 410 performances comparable to those obtained in a previous study [33]. Using this material, which 411 showed a good stability during the whole experiment (see Fig. 3b), a total gas yield of about 412 40% was obtained. Nevertheless, its relatively low hydrogen yield indicates that the extent of 413 steam reforming was modest. 414 Regarding the performance of the chemically activated biochar (CB_3_0.10), a similar total 415 gas yield was measured, compared to the physically activated one. However, selectivity to 416 gaseous products were markedly different, leading to a decrease in the production of H2 (the 417 hydrogen yield was even lower than that of PB_700_1.00) and an increase in that of CO2. This 418 22 result could suggest that the decomposition of pyrolysis oil instead of steam and/or dry 419 reforming, was the main process involved. 420 421 422 Fig. 3. Results obtained from combined steam and dry reforming of pyrolysis oil at 750°C: 423 total and H2 yields (Yg and YH2, respectively) as well as selectivities toward a given gaseous 424 product, Si (a); and evolution of the hydrogen production rate (b). 425 As can be deduced from Fig. 3a, similar results were also obtained for the chemically activated 426 biochar that was washed with just water (CB_3_0.10_W). The differences in the performance 427 Blank PB_700_1.00 CB_3_0.10 CB_3_0.10_W CB_3_0.10_R PB_700_1.00 + K2CO3 0 10 20 30 40 50 60 70 80 90 100 (a) Yg SH2 SCH4 SCO SCO2 SC2H4 SC2H6 YH2 Percentage (%) 010 20 30 40 50 60 0 1 2 H2 evolution (mmol min-1) Time (min) Blank PB_700_1.00 CB_3_0.10 CB_3_0.10_W CB_3_0.10_R PB_700_1.00 + K2CO3 (b) 23 between the physically and chemically activated biochars could be related to their textural 428 properties. In this respect, the physically activated biochar exhibited a higher specific surface 429 area and higher volumes of microand mesopores, thus providing more catalytic active sites 430 of inherent alkali or alkaline earth metal species (AAEMs). The catalytic role of these species 431 (especially K, Mg and Ca) in both steam reforming and gasification of carbon has been widely 432 reported [25,44–46]. Furthermore, the more hierarchical pore size distribution of 433 PB_700_1.00, with relatively high contributions of the mesopores, could shorten the diffusion 434 path to reach the active sites. 435 Fig. 3b clearly shows that H2 release rates fluctuated over time for most of biochars tested. 436 There are at least two possible explanations for this: First, the coke deposits, which gradually 437 covered the catalyst surface, could have ted to large fluctuations in the extent of the reactions 438 involved. Second, the structural modification of the activated carbons (the process temperature 439 was higher than the activation temperature) could also led to unsteady hydrogen flow rates at 440 the outlet. It should be noted that the reaction system studied was certainly complex, due the 441 numerous competitive reactions that can affect the yields of the different gaseous products. 442 Nevertheless, it can be assumed that methane is mainly produced by the cracking of the 443 heaviest fraction of the pyrolysis oil, and that its yield will be correlated to the amount of coke 444 produced [47]. Similarly, Fig. 3a reveals that the selectivity towards methane was almost 445 constant for PB_700_1.00, CB_3_0.10, and CB_3_0.10_W materials. Therefore, it can be 446 concluded that the coke production in the process was quite similar for the above-mentioned 447 carbons, suggesting that some structural modifications in the carbons could be accounted for 448 the unsteady production of hydrogen. These modifications could be ascribed to the reverse 449 Boudouard reaction [48,49], steam gasification and, in the case of the chemical activated 450 biochars, potassium-catalyzed gasification. To support this argument, Fig. 4 shows the textural 451 properties of fresh and spent activated biochars. Contrary to the loss of porosity observed for 452 24 spent PB_700_1.00 and CB_3_0.10 catalysts, the spent CB_3_0.1_W material showed an 453 increase in pore volumes, especially for micropores. In addition, the results from TPD 454 measurements, which are summarized graphically in Figs. 5 and A.6, indicate that the water455 washed chemically activated carbon exhibited a larger mass loss (with an increased release of 456 CO2 and CO, as shown in Fig. A.6) than that of the acid-washed carbon, thus confirming a 457 greater extent of carbon gasification. This fact could be attributed to the residual amount of 458 activation reagent available on the surface of the catalyst CB_3_0.10_W, which might promote 459 further biochar gasification during the experiment, as recently reported by Wang et al. [11]. 460 To understand better the effect of the washing procedure on the performance of chemically 461 activated biochars, an unwashed (raw) material (CB_3_0.10_R) was also tested as catalyst. In 462 this case, the total gas yield was 65% with enhanced selectivity towards both H2 and CO (see 463 Fig. 3a), thus indicating a higher extent of the steam reforming reactions (leading to a YH2 of 464 66%). Fig. 6 displays the FTIR spectra obtained for fresh and spent chemically activated 465 biochars as well as for pure K2CO3. In the case of acid and water-washed samples, no evident 466 K2CO3 content was detected on the surface, indicating the effectiveness of both washing 467 procedures. On the other hand, most of the chemical activating agent available on the surface 468 of the fresh CB_3_0.10_R catalyst disappeared after the catalytic test. The potassium carbonate 469 (and other chemical species derived from its partial decomposition) available on the surface of 470 the catalyst at the beginning of the reforming test (reductive environment) could progressively 471 be reduced to metallic potassium (K0), leading to a further promotion of reforming reactions. 472 Furthermore, a process temperature relatively close to the volatilization point of potassium can 473 enhance the mobility and reactivity of the metal [50]. Generally speaking, activated biochars 474 have relatively abundant oxygen and nitrogen-containing functional groups, which are not 475 present in the materials shown in Fig. 6. The reason behind this could be the relatively high 476 activation temperatures, since the majority of functional groups decompose below 800 °C [51]. 477 25 Furthermore, the relatively high amount of K2CO3 loaded on the analyzed samples resulted in 478 a very large peak, which may hide other peaks of interest, such as those related to oxygenated 479 functionalities. 480 481 Fig. 4. Differences in textural properties between the fresh and spent activated biochars used 482 as catalysts in steam and dry reforming tests. 483 484 -0.35 -0.30 -0.25 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 S2D-NLDFT Vt Vmeso Vmicro Vultra Pore Volume difference (cm3g-1) PB_700_1.00 CB_3_0.10 CB_3_0.10_W CB_3_0.10_R PB_700_1.00+K2CO3 -300 -250 -200 -150 -100 -50 0 50 100 S2D-NLDFT difference (m2g-1) 32 Acknowledgments 582 This project received funding from the European Union’s Horizon 2020 research and 583 innovation programme under the Marie Skłodowska-Curie grant agreement No 721991. The 584 authors also acknowledge the funding from the Aragón Government (Ref. T22_20R), co585 funded by FEDER 2014-2020 "Construyendo Europa desde Aragón". ". This study was partly 586 supported by the French PIA project “Lorraine Université d’Excellence”, reference ANR-15587 IDEX-04-LUE, PROMOTEE project and TALiSMAN project, funded by FEDER (2019588 000214). The authors gratefully thank José Antonio Manso (UNIZAR) and Philippe 589 Gadonneix (IJL) for their help in the preparation and characterization of the samples and 590 Sandrine Mathieu (IJL) for the SEM characterization. 591 Appendix A. Supplementary data 592 Further details on characterization of pristine and activated biochars. Statistical outcomes from 593 the adopted designs of experiments. 594 Nomenclature 595 ABET Brunauer-Emmett-Teller area (m2 g–1) 596 dp Pore diameter (nm) 597 FBO Molar flow rate of dry bio oil (mol min–1) 598 FH2 Av Experimental average H2 molar flow rate (mol min–1) 599 FH2 Stoi H2 stoichiometric molar flow rate (mol min–1) 600 m0 Initial mass of biochar before activation (g) 601 mc Biochar mass before the washing step (g) 602 mf Final mass of biochar after activation (g) 603 33 mg Total mass of produced gas during the upgrading process (g) 604 ml Mass of liquid fed into the upgrading reactor (g) 605 mw Biochar mass after the washing step (g) 606 ni Produced amount of a given gaseous specie i (mol) 607 ntot Total amount of produced gas (mol) 608 S2D-NLDFT 2D-NLDFT specific surface area (m2 g–1) 609 Si Selectivity toward a given gaseous specie i (%) 610 V0.97 Gurvitch pore volume (cm3 g–1) 611 Vmeso Volume of mesopores (cm3 g–1) 612 Vmicro Volume of micropores (cm3 g–1) 613 Vt Total pore volume (cm3 g–1) 614 Vultra Volume of ultra-micropores (cm3 g–1) 615 Yg Total gas yield (%) 616 YH2 Hydrogen yield (%) 617 Ywash Washing yield (%) 618 η Degree of burnout (%) 619 Acronyms 620 FTIR Fourier-Transform Infrared spectroscopy 621 GHSV Gas hourly space velocity 622 LHSV Liquid hourly space velocity 623 34 PSD Pore size distribution 624 RSM Response surface methodology 625 S:C Steam to carbon molar ratio 626 STP Standard temperature and pressure 627 TPD Temperature-programmed desorption 628 XPS X-ray photoelectron spectroscopy 629 μ-GC Micro gas chromatograph 630 631 35 References 632 [1] Guedidi H, Lakehal I, Reinert L, Lévêque JM, Bellakhal N, Duclaux L. Removal of 633 ionic liquids and ibuprofen by adsorption on a microporous activated carbon: Kinetics, 634 isotherms, and pore sites. Arab J Chem 2020;13:258–70. 635 https://doi.org/10.1016/j.arabjc.2017.04.006. 636 [2] Santoyo-Cisneros R, Rangel-Mendez JR, Nava JL, Larios-Durán ER, Chazaro-Ruiz 637 LF. Influence of surface chemistry of activated carbon electrodes on electro-assisted 638 adsorption of arsenate. J Hazard Mater 2020;392:122349. 639 https://doi.org/10.1016/j.jhazmat.2020.122349. 640 [3] Acharya SP, Johnson J, Weidhaas J. Adsorption kinetics of the herbicide safeners, 641 benoxacor and furilazole, to activated carbon and agricultural soils. J Environ Sci 642 (China) 2020;89:23–34. https://doi.org/10.1016/j.jes.2019.09.022. 643 [4] Palomo J, Rodríguez-Cano MA, Rodríguez-Mirasol J, Cordero T. On the kinetics of 644 methanol dehydration to dimethyl ether on Zr-loaded P-containing mesoporous 645 activated carbon catalyst. Chem Eng J 2019;378:122198. 646 https://doi.org/10.1016/j.cej.2019.122198. 647 [5] Liu WJ, Jiang H, Yu HQ. Development of Biochar-Based Functional Materials: 648 Toward a Sustainable Platform Carbon Material. Chem Rev 2015;115:12251–85. 649 https://doi.org/10.1021/acs.chemrev.5b00195. 650 [6] Wang Y, Yang H, Jin L, Li Y, Hu H, Ding H, et al. Effect of mineral in coal on 651 preparation of activated carbon for methane decomposition to hydrogen. Fuel 652 2019;258:116138. https://doi.org/10.1016/j.fuel.2019.116138. 653 [7] Li H, Zheng F, Wang J, Zhou J, Huang X, Chen L, et al. Facile preparation of zeolite654 activated carbon composite from coal gangue with enhanced adsorption performance. 655 Chem Eng J 2020;390:124513. https://doi.org/10.1016/j.cej.2020.124513. 656 36 [8] Ding S, Liu Y. Adsorption of CO2 from flue gas by novel seaweed-based KOH657 activated porous biochars. Fuel 2020;260:116382. 658 https://doi.org/10.1016/j.fuel.2019.116382. 659 [9] Iriarte-Velasco U, Sierra I, Zudaire L, Ayastuy JL. Preparation of a porous biochar 660 from the acid activation of pork bones. Food Bioprod Process 2016;98:341–53. 661 https://doi.org/10.1016/j.fbp.2016.03.003. 662 [10] Sayğılı H, Akkaya Sayğılı G. Optimized preparation for bimodal porous carbon from 663 lentil processing waste by microwave-assisted K2 CO3 activation : Spectroscopic 664 characterization and dye decolorization activity. J Clean Prod 2019;226:968–76. 665 https://doi.org/10.1016/j.jclepro.2019.04.121. 666 [11] Wang L, Sun F, Hao F, Qu Z, Gao J, Liu M, et al. A green trace K2CO3 induced 667 catalytic activation strategy for developing coal-converted activated carbon as 668 advanced candidate for CO2 adsorption and supercapacitors. Chem Eng J 669 2020;383:123205. https://doi.org/10.1016/j.cej.2019.123205. 670 [12] Mai TT, Vu DL, Huynh DC, Wu NL, Le AT. Cost-effective porous carbon materials 671 synthesized by carbonizing rice husk and K2CO3 activation and their application for 672 lithium-sulfur batteries. J Sci Adv Mater Devices 2019;4:223–9. 673 https://doi.org/10.1016/j.jsamd.2019.04.009. 674 [13] Lahijani P, Zainal ZA, Mohammadi M, Mohamed AR. Conversion of the greenhouse 675 gas CO2 to the fuel gas CO via the Boudouard reaction: A review. Renew Sustain 676 Energy Rev 2015;41:615–32. https://doi.org/10.1016/j.rser.2014.08.034. 677 [14] Şahin Ö, Saka C. Preparation and characterization of activated carbon from acorn shell 678 by physical activation with H2O – CO2 in two-step pretreatment. Bioresour Technol 679 2013;136:163–8. https://doi.org/10.1016/j.biortech.2013.02.074. 680 [15] Guo Y, Tan C, Sun J, Li W, Zhang J, Zhao C. Porous activated carbons derived from 681 37 waste sugarcane bagasse for CO2 adsorption. Chem Eng J 2020;381:122736. 682 https://doi.org/10.1016/j.cej.2019.122736. 683 [16] Bertero M, Puente G De, Sedran U. Fuels from bio-oils : Bio-oil production from 684 different residual sources , characterization and thermal conditioning. Fuel 685 2012;95:263–71. https://doi.org/10.1016/j.fuel.2011.08.041. 686 [17] Fu M, Qi W, Xu Q, Zhang S, Yan Y. Hydrogen production from bio-oil model 687 compounds dry ( CO2) reforming over Ni/Al2O3 catalyst. Int J Hydrogen Energy 688 2015;41:1494–501. https://doi.org/10.1016/j.ijhydene.2015.11.104. 689 [18] Santamaria L, Artetxe M, Lopez G, Cortazar M, Amutio M, Bilbao J, et al. Effect of 690 CeO2 and MgO promoters on the performance of a Ni/Al2O3 catalyst in the steam 691 reforming of biomass pyrolysis volatiles. Fuel Process Technol 2020;198:106223. 692 https://doi.org/10.1016/j.fuproc.2019.106223. 693 [19] Santamaria L, Lopez G, Arregi A, Amutio M, Artetxe M, Bilbao J, et al. Stability of 694 different Ni supported catalysts in the in-line steam reforming of biomass fast 695 pyrolysis volatiles. Appl Catal B Environ 2019;242:109–20. 696 https://doi.org/10.1016/j.apcatb.2018.09.081. 697 [20] Bimbela F, Ábrego J, Puerta R, García L, Arauzo J. Catalytic steam reforming of the 698 aqueous fraction of bio-oil using Ni-Ce / Mg-Al catalysts. Appl Catal B Environ 699 2017;209:346–57. https://doi.org/10.1016/j.apcatb.2017.03.009. 700 [21] Phongprueksathat N, Meeyoo V, Rirksomboon T. Steam reforming of acetic acid for 701 hydrogenproduction : Catalytic performances of Ni and Co supported on 702 Ce0·75Zr0·25O2 catalysts. Int J Hydrogen Energy 2019;44:9359–67. 703 https://doi.org/10.1016/j.ijhydene.2019.02.085. 704 [22] Domine ME, Iojoiu EE, Davidian T, Guilhaume N, Mirodatos C. Hydrogen production 705 from biomass-derived oil over monolithic Ptand Rh-based catalysts using steam 706 38 reforming and sequential cracking processes. Catal Today 2008;133–135:565–73. 707 https://doi.org/10.1016/j.cattod.2007.12.062. 708 [23] Li X, Zhang Z, Zhang L, Fan H, Li X, Liu Q, et al. Investigation of coking behaviors 709 of model compounds in bio-oil during steam reforming. Fuel 2020;265:116961. 710 https://doi.org/10.1016/j.fuel.2019.116961. 711 [24] Ma Z, Xiao R, Zhang H. Catalytic steam reforming of bio-oil model compounds for 712 hydrogen-rich gas production using bio-char as catalyst. Int J Hydrogen Energy 713 2016;42:3579–85. https://doi.org/10.1016/j.ijhydene.2016.11.107. 714 [25] Feng D, Zhao Y, Zhang Y, Sun S, Meng S, Guo Y, et al. Effects of K and Ca on 715 reforming of model tar compounds with pyrolysis biochars under H2O or CO2. Chem 716 Eng J 2016;306:422–32. https://doi.org/10.1016/j.cej.2016.07.065. 717 [26] Buentello-Montoya D, Zhang X, Li J, Ranade V, Marques S, Geron M. Performance 718 of biochar as a catalyst for tar steam reforming: Effect of the porous structure. Appl 719 Energy 2020;259:114176. https://doi.org/10.1016/j.apenergy.2019.114176. 720 [27] Feng D, Zhao Y, Zhang Y, Zhang Z, Sun S. Roles and fates of K and Ca species on 721 biochar structure during in-situ tar H2O reforming over nascent biochar. Int J 722 Hydrogen Energy 2017;42:21686–96. https://doi.org/10.1016/j.ijhydene.2017.07.096. 723 [28] Zhang Z, Ou Z, Qin C, Ran J, Wu C. Roles of alkali/alkaline earth metals in steam 724 reforming of biomass tar for hydrogen production over perovskite supported Ni 725 catalysts. Fuel 2019;257:116032. https://doi.org/10.1016/j.fuel.2019.116032. 726 [29] Megaritis A, Zhuo Y, Messenböck R, Dugwell DR, Kandiyoti R. Pyrolysis and 727 gasification in a bench-scale high-pressure fluidized-bed reactor. Energy Fuels 728 1998;12:144–51. https://doi.org/10.1021/ef970115x. 729 [30] Dufour A, Celzard A, Fierro V, Martin E, Broust F, Zoulalian A. Catalytic 730 decomposition of methane over a wood char concurrently activated by a pyrolysis gas. 731 39 Appl Catal A Gen 2008;346:164–73. https://doi.org/10.1016/j.apcata.2008.05.023. 732 [31] Ranguin R, Jean-Marius C, Yacou C, Gaspard S, Feidt C, Rychen G, et al. Reduction 733 of chlordecone environmental availability by soil amendment of biochars and activated 734 carbons from lignocellulosic biomass. Environ Sci Pollut Res 2020. 735 https://doi.org/10.1007/s11356-019-07366-2. 736 [32] Greco G, Videgain M, Di Stasi C, González B, Manyà JJ. Evolution of the mass-loss 737 rate during atmospheric and pressurized slow pyrolysis of wheat straw in a bench-scale 738 reactor. J Anal Appl Pyrolysis 2018;136:18–26. 739 https://doi.org/10.1016/j.jaap.2018.11.007. 740 [33] Di Stasi C, Alvira D, Greco G, González B, Manyà JJ. Physically activated wheat 741 straw-derived biochar for biomass pyrolysis vapors upgrading with high resistance 742 against coke deactivation. Fuel 2019;255:115807. 743 https://doi.org/10.1016/j.fuel.2019.115807. 744 [34] Jagiello J, Kenvin J, Celzard A, Fierro V. Enhanced resolution of ultra micropore size 745 determination of biochars and activated carbons by dual gas analysis using N2 and 746 CO2 with 2D-NLDFT adsorption models. Carbon 2019;144:206–15. 747 https://doi.org/10.1016/j.carbon.2018.12.028. 748 [35] Sanchez-Sanchez A, Izquierdo MT, Medjahdi G, Ghanbaja J, Celzard A, Fierro V. 749 Ordered mesoporous carbons obtained by soft-templating of tannin in mild conditions. 750 Microporous Mesoporous Mater 2018;270:127–39. 751 https://doi.org/10.1016/j.micromeso.2018.05.017. 752 [36] Matos J, Labady M, Albornoz A, Laine J, Brito JL. Catalytic effect of KOH on 753 textural changes of carbon macro-networks by physical activation. J Mol Catal A 754 Chem 2005;228:189–94. https://doi.org/10.1016/j.molcata.2004.09.039. 755 [37] Lozano-Castelló D, Calo JM, Cazorla-Amorós D, Linares-Solano A. Carbon activation 756 40 with KOH as explored by temperature programmed techniques, and the effects of 757 hydrogen. Carbon 2007;45:2529–36. https://doi.org/10.1016/j.carbon.2007.08.021. 758 [38] Ganga Devi T, Kannan MP. Calcium catalysis in air gasification of cellulosic chars. 759 Fuel 1998;77:1825–30. https://doi.org/10.1016/S0016-2361(98)00038-6. 760 [39] Dehkhoda AM, Gyenge E, Ellis N. A novel method to tailor the porous structure of 761 KOH-activated biochar and its application in capacitive deionization and energy 762 storage. Biomass Bioenergy 2016;87:107–21. 763 https://doi.org/10.1016/j.biombioe.2016.02.023. 764 [40] Malekshahian M, Hill JM. Effect of pyrolysis and CO2 gasification pressure on the 765 surface area and pore size distribution of petroleum coke. Energy Fuels 2011;25:5250– 766 6. https://doi.org/10.1021/ef201231w. 767 [41] Montané D, Fierro V, Marêché JF, Aranda L, Celzard A. Activation of biomass768 derived charcoal with supercritical water. Microporous Mesoporous Mater 769 2009;119:53–9. https://doi.org/10.1016/j.micromeso.2008.09.040. 770 [42] Malekshahian M, Hill JM. Kinetic Analysis of CO2 Gasification of Petroleum Coke at 771 High Pressures. Energy Fuels 2011;25:4043–8. https://doi.org/10.1021/ef2009259. 772 [43] Fermoso J, Stevanov C, Moghtaderi B, Arias B, Pevida C, Plaza MG, et al. High773 pressure gasification reactivity of biomass chars produced at different temperatures. J 774 Anal Appl Pyrolysis 2009;85:287–93. https://doi.org/10.1016/j.jaap.2008.09.017. 775 [44] Yip K, Tian F, Hayashi JI, Wu H. Effect of alkali and alkaline earth metallic species 776 on biochar reactivity and syngas compositions during steam gasification. Energy Fuels 777 2010;24:173–81. https://doi.org/10.1021/ef900534n. 778 [45] Sueyasu T, Oike T, Mori A, Kudo S, Norinaga K, Hayashi JI. Simultaneous steam 779 reforming of tar and steam gasification of char from the pyrolysis of potassium-loaded 780 woody biomass. Energy Fuels 2012;26:199–208. https://doi.org/10.1021/ef201166a. 781 41 [46] Jiang L, Hu S, Wang Y, Su S, Sun L, Xu B, et al. Catalytic effects of inherent alkali 782 and alkaline earth metallic species on steam gasification of biomass. Int J Hydrogen 783 Energy 2015;40:15460–9. https://doi.org/10.1016/j.ijhydene.2015.08.111. 784 [47] Matos J, Laine J. Ethylene conversion on activated carbon-supported NiMo catalysts: 785 Effect of the support. Appl Catal A Gen 2003;241:25–38. 786 https://doi.org/10.1016/S0926-860X(02)00427-1. 787 [48] Díaz K, García V, Matos J. Activated carbon supported Ni – Ca: Influence of reaction 788 parameters on activity and stability of catalyst on methane reformation. Fuel 789 2007;86:1337–44. https://doi.org/10.1016/j.fuel.2006.05.011. 790 [49] Matos J, Díaz K, García V, Cordero TC, Brito JL. Methane transformation in presence 791 of carbon dioxide on activated carbon supported nickel – calcium catalysts. Catalysis 792 Lett. 2006;109:163–169. https://doi.org/10.1007/s10562-006-0073-3. 793 [50] Ducousso M, Weiss-Hortala E, Lyczko N, Nzihou A, Castaldi MJ. 110th Anniversary: 794 Syngas Production Enhancement Using CalciumAnd Potassium-Impregnated Chars. 795 Ind Eng Chem Res 2019;58:15134–41. https://doi.org/10.1021/acs.iecr.9b02238. 796 [51] Figueiredo JL, Pereira MFR. The role of surface chemistry in catalysis with carbons. 797 Catal Today 2010;150:2–7. https://doi.org/10.1016/j.cattod.2009.04.010. 798 [52] Kopyscinski J, Rahman M, Gupta R, Mims CA, Hill JM. K2CO3 catalyzed CO2 799 gasification of ash-free coal. Interactions of the catalyst with carbon in N2 and CO2 800 atmosphere. Fuel 2014;117:1181–9. https://doi.org/10.1016/j.fuel.2013.07.030. 801 [53] Li X, Liu J, Kong F, Liu X, Xu J, Chen H. Potassium-doped graphene for 802 simultaneous determination of nitrite and sul fi te in polluted water. Electrochem 803 Commun 2012;20:109–12. https://doi.org/10.1016/j.elecom.2012.04.014. 804 805 7 Fig. A.4. N2 adsorption/desorption isotherms of the following chemically activated biochars: CB_3_0.10 (a); CB_2_0.55 (b); and CB_3_0.10 (c). 0.0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 80 100 120 140 Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption (a) 0.0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 80 100 120 140 (b) Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption 0.0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 80 100 120 140 (c) Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption 8 Fig. A.5. N2 adsorption/desorption isotherms of the following physically activated biochars: PB_700_1.00 (a); PB_700_0.10 (b); PB_850_1.00 (c); and PB_850_0.10 (d). 0.0 0.2 0.4 0.6 0.8 1.0 100 120 140 160 180 200 220 Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption (a) 0.0 0.2 0.4 0.6 0.8 1.0 100 120 140 160 180 200 220 (b) Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption 0.0 0.2 0.4 0.6 0.8 1.0 100 120 140 160 180 200 220 (c) Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption 0.0 0.2 0.4 0.6 0.8 1.0 100 120 140 160 180 200 220 (d) Quantity Adsorbed (cm3STP g-1) Relative Pressure Adsorption Desorption 9 Fig. A.6. Differential thermogravimetric curves (in % of mass min−1) and profiles of released species (H2O, CO2, and CO) from the TPD measurements conducted for the following materials: CB_3_0.10 (a); CB_3_0.10_W (b); CB_3_0.10_R (c); and PB_700_1.00+K2CO3 (d). 0100 200 300 400 500 600 700 1E-12 1E-11 1E-10 1E-9 1E-8 Ion Current (A) Temperature (°C) H2O CO CO2 Derivative mass loss -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 (a) d(mass loss) d(t)-1 (% min-1) 0100 200 300 400 500 600 700 1E-12 1E-11 1E-10 1E-9 1E-8 (b) Ion Current (A) H2O CO CO2 Derivative mass loss -0.8 -0.6 -0.4 -0.2 0.0 d(mass loss) d(t)-1 (% min-1) Temperature (°C) 0100 200 300 400 500 600 700 1E-12 1E-11 1E-10 1E-9 1E-8 Temperature (°C) Ion Current (A) H2O CO CO2 Derivative mass loss -0.8 -0.6 -0.4 -0.2 0.0 d(mass loss) d(t)-1 (% min-1) (c) 0100 200 300 400 500 600 700 1E-12 1E-11 1E-10 1E-9 1E-8 (d) Ion Current (A) Temperature (°C) H2O CO CO2 Mass loss derivative -0.8 -0.6 -0.4 -0.2 0.0 d(mass loss) d(t)-1 (% min-1) 10 Fig. A.7. C1s, O1s and K2p spectra of fresh and spent PB_700_1.00 and PB_700_1.00+K2CO3. 538 537 536 535 534 533 532 531 530 529 528 O1s Metal Oxide PB_700_1.00+K2CO3 (Fresh) PB_700_1.00 (Spent) PB_700_1.00 (Fresh) PB_700_1.00+K2CO3 (Spent) BE (eV) Intensity (a.u.) 300 299 298 297 296 295 294 293 292 291 290 BE (eV) K2p PB_700_1.00+K2CO3 (Fresh) PB_700_1.00 (Spent) PB_700_1.00 (Fresh) PB_700_1.00+K2CO3 (Spent) Intensity (a.u.) 290 289 288 287 286 285 284 283 282 281 280 Intensity (a.u.) BE (eV) Carbide C1s PB_700_1.00+K2CO3 (Fresh) PB_700_1.00 (Spent) PB_700_1.00 (Fresh) PB_700_1.00+K2CO3 (Spent)