Mesoporous Carbon Production by Nanocasting Technique Using Boehmite as a Template

: A series of mesoporous carbonaceous materials were synthesized by the nanocasting technique using boehmite as a template and glucose as a carbon precursor. After pyrolysis and template removal, the resulting material is a mesoporous carbon that can be additionally doped with N, B and K during prepyrolysis impregnation. In addition, the inﬂuence of doping on the morphology, crystallinity and stability of the synthesized carbons was studied using X-ray diffraction, nitrogen physisorption, thermogravimetry, Raman and IR spectroscopy and transmission electron microscopy. While the nanocasting process is effective for the formation of mesopores, KOH and urea do not modify the textural properties of carbon. The use of H 3 PO 4 as a dopant, however, led to the formation of an AlPO 4 compound and resulted in a solid with a lower speciﬁc surface area and higher microporosity. All doped solids present higher thermal stability as a positive effect of the introduction of heteroatoms to the carbon skeleton. The phosphorus-doped sample has better oxidation resistance, with a combustion temperature 120–150 ◦ C higher than those observed for the other materials.


Introduction
Mesoporous carbons are materials with a continuous carbon skeleton with a pore size within the range of 2 to 50 nm [1]. These materials present very interesting properties such as a large surface area, adjustable framework, tunable pore size, high resistance to corrosion and high mechanical and thermal stability. These features, along with their appreciable electron conductivity and relative chemical inertness, make these materials very useful in various fields such as adsorption, biotechnology, energy storage, sensing applications and in catalysis as catalytic supports or catalysts [2,3]. When obtained from waste biomass (a renewable and low-cost source), mesoporous carbons are considered environmentally benign and cost competitive to other supports or heterogeneous catalysts [4]. Mesoporous carbons can be obtained by different processes such as carbonization and activation of organic precursors or by direct synthesis [5,6]. More complex and interesting carbon-based structures can also be obtained using new methods such as additive manufacturing and high-temperature pyrolysis procedures, which allow the formation of pyrolytic carbon micropillars with nanosized voids and associated nanolattices with improved mechanical properties and large deformability with great potential for energy storage applications [7][8][9].
However, molding or templating is one of the most powerful techniques in which highly structured porous networks are obtained using either solids (exotemplating or hard templating) or molecules (endotemplating or soft templating) as templates to obtain inverse carbon replicas. This method has the advantage of easy control of particle morphology and pore size by selecting the used template [10]. On the other hand, hard templating, also known as nanocasting, is a versatile, cheap and suitable way for industrial applications. The mold that acts as a 'template' is impregnated with a carbon precursor and then processed and submitted to chemical treatments to obtain an inverse replica of the initial mold.
Rigid inorganic solids, such as mesoporous zeolites, aluminas and lamellar solids such as hydrotalcites, can be used as templates. The mesostructure and pore surface of the template are decisive for the infiltration of the precursor and favored by its rigid replica's structure control [5].
The template channels are infiltrated with an appropriate carbon precursor, commonly a sugar, such as sucrose or glucose, that is in situ converted by carbonization or pyrolysis to carbon which within the template structure appears as a continuous carbon network. Once the precursor has replicated the negative of the template, the target materials are received after template removal [2,3,11].
Doping the carbonaceous structure with specific heteroatoms can increase its thermal, mechanical and structural stability while improving the selectivity, activity and durability of these materials in catalytic reactions [12,13]. Heteroatoms such N, B and P have been classically used for this purpose, as they can generate superficial defect sites that break carbon electroneutrality, improving this way the performance of the carbonaceous material. For example, P-or B-doped mesoporous carbons present better catalytic properties in direct dehydrogenation of propane (higher propane conversion and propene selectivity) than undoped carbon due to the increase in the population of the main active sites (carbonyl/quinone groups) [14]. On the other hand, N-doped carbon catalysts speed up the dissociative adsorption of oxygen molecules, decreasing the activation energy for oxidative dehydrogenation [15,16]. Carbon activation achieved by doping with potassium is especially interesting. This element is intercalated between the graphite sheets in their cationic form, K + , increasing the specific surface of these materials. It also manages to increase graphite conductivity by more than one order of magnitude, achieving superconductivity at very low temperatures in crystalline graphite. Finally, and because of its low reduction potential, potassium is easily reduced to its neutral form during all heating stages which is coupled with the oxidation of C [17,18]. The main strategies described in the literature for the production of doped carbons are based on the use of a precursor rich in the desired element, included either directly in the synthesis process or in the post-treatment process after the synthesis of the carbonaceous material [19][20][21].
Boehmite is a mesoporous aluminum oxyhydroxide (AlOOH) that can be used as a template in nanocasting processes. Orthorhombic boehmite minerals (space group Cmcm) crystallize in a two-dimensional structure, with layers of edge-shared AlO 4 (OH) 2 octahedra linked by hydrogen bonds (Figure 1) [22][23][24]. Boehmite is a cheap template and can be prepared using a simple sol-gel method. The pyrolysis treatment to which the template is subjected after infiltration with the carbon source could cause boehmite transformation into different transitional alumina  The pyrolysis treatment to which the template is subjected after infiltration with the carbon source could cause boehmite transformation into different transitional alumina phases, depending on the used temperature and atmosphere. In the temperature range generally used for the pyrolysis of organic matter to obtain carbon materials, a gammaalumina phase is produced. Gamma-alumina, γ-Al 2 O 3 , is transitional metastable alumina in the temperature range of 500-850 • C [25]. The structure of γ-Al 2 O 3 is traditionally considered a cubic defective spinel in which the oxygen atoms are arranged in a cubic close packing (face-centered cubic (fcc)), and the Al atoms occupy the octahedral and tetrahedral sites with existing cationic vacancies that maintain the electric neutrality [26,27]. It is also a mesoporous oxide widely used as catalytic support due to its suitable surface and textural properties.
As far as we know, boehmite has not been used before as a template in the nanocasting process to obtain mesoporous carbon. However, some studies have been published dedicated to obtaining carbon nanotubes using anodic aluminum oxide (AAO) as a template [28][29][30].
Taking into account all described above, the objective of this work is to explore the use of boehmite as a template to obtain mesoporous carbons from biomass by the nanocasting method and to study the influence of different dopants on carbon structure and textural properties. The adopted strategy consists of using boehmite as a template, glucose as a carbon precursor (glucose) and the incorporation of different additives (KOH, H 3 PO 4 and urea as a source of K, P and N, respectively) during the synthesis procedure. Glucose was selected as a carbon precursor, being a molecule representative of renewable vegetal lignocellulosic biomass as its main component [31].

Results and Discussion
A detailed description of doped/undoped mesoporous carbon preparation is provided in Section 3. Figure 2 shows X-ray diffractograms of the pyrolyzed samples before template removal. The pyrolysis process induces the transformation of the boehmite crystalline phase (AlOOH) to γ-Al 2 O 3 (JCPDS: 00-010-0425). This transformation is consistent with the used temperature (700 • C) [25]. Only in the case of the sample impregnated with phosphoric acid does the formation of a new crystalline phase (aluminum phosphate (AlPO 4 ), JCPDS: 00-051-1674) occur after the reaction between additive and template. In all samples, carbon diffractions are not clearly discernible due to their overlap with those of the template. According to the XRD results, the materials generated after pyrolysis can be considered template-carbon hybrids (AlPO4-C in the case of the sample treated with phosphoric acid and γ-Al2O3-C for the rest of the series). However, while not the objective of the present paper, the hybrid materials could be of high potential interest, as carbonaceous hybrid materials with nanocomposites is an emerging field with a large variety of possible applications in different scientific and technical fields, such as catalysis, electrochemistry, According to the XRD results, the materials generated after pyrolysis can be considered template-carbon hybrids (AlPO 4 -C in the case of the sample treated with phosphoric acid and γ-Al 2 O 3 -C for the rest of the series). However, while not the objective of the present paper, the hybrid materials could be of high potential interest, as carbonaceous hybrid materials with nanocomposites is an emerging field with a large variety of possible applications in different scientific and technical fields, such as catalysis, electrochemistry, absorption, sensors, etc. [32,33]. Additionally, our hybrids are obtained in an economical and sustainable way.
The effectiveness of successive template removal treatments, i.e., dissolution stages (up to four), was followed by XRD. As an example, Figure 3A shows an evolution of the X-ray diffractograms of the pyrolyzed B_Glu sample after each dissolution stage. It can be seen how the different dissolution stages lead to more effective removal of the template, confirmed by the appearance of characteristic diffractions of the carbonaceous material visible in the final solid. After the third dissolution stage, a new crystalline phase appears, with a peak at 2θ ≈18 • , which could be assigned to a silicate, perhaps from contamination produced by the glass containers used for the template dilution processes. According to the XRD results, the materials generated after pyrolysis can be considered template-carbon hybrids (AlPO4-C in the case of the sample treated with phosphoric acid and γ-Al2O3-C for the rest of the series). However, while not the objective of the present paper, the hybrid materials could be of high potential interest, as carbonaceous hybrid materials with nanocomposites is an emerging field with a large variety of possible applications in different scientific and technical fields, such as catalysis, electrochemistry, absorption, sensors, etc. [32,33]. Additionally, our hybrids are obtained in an economical and sustainable way.
The effectiveness of successive template removal treatments, i.e., dissolution stages (up to four), was followed by XRD. As an example, Figure 3A shows an evolution of the X-ray diffractograms of the pyrolyzed B_Glu sample after each dissolution stage. It can be seen how the different dissolution stages lead to more effective removal of the template, confirmed by the appearance of characteristic diffractions of the carbonaceous material visible in the final solid. After the third dissolution stage, a new crystalline phase appears, with a peak at 2θ ≈18º, which could be assigned to a silicate, perhaps from contamination produced by the glass containers used for the template dilution processes.   Similar results were obtained for the pyrolyzed B_Glu_K and B_Glu_N samples ( Figures S1 and S2), evidencing that the template removal treatments have a similar effect in such materials.
In the case of the pyrolyzed B_Glu_P sample, complete elimination of the AlPO 4 phase was not achieved after four stages of dissolution, confirmed by the visible diffractions corresponding to that compound ( Figure 3B). Figure 4 shows diffractograms of the final carbons obtained after the full template dissolution treatment. In all cases, reflections at 2θ ≈ 25 • , 44 • and 80 • are observed, corresponding, respectively, to the families of planes (002), (100) and (110) of graphitic carbon. Diffraction of (002) planes is attributed to the structural packing of the aromatic layers through van der Waals forces, while (100) and (110) signals arise from the aromatic structures in the plane [34,35].
The intensity of the (002) peak is related to the degree of graphitic crystallinity in such a way that wide and low-intensity peaks indicate a low order at short range of the graphite carbon sheets organization, while sharp and narrow peaks correspond to highly crystalline carbons with a high degree of order at long range [34][35][36].
To evaluate the degree of crystallinity of the obtained carbons, the crystallite height (Lc) and crystallite diameter (La) parameters were calculated. The magnitudes of these parameters determine the sizes of the crystals within the carbon sample [34,35,37]. An increase in those parameters an increased graphitization of the carbon. Additionally, the interlayer spacing (d 002 ), the number of aromatic layers per carbon crystallite (N ave ) and the R value, considered as the empirical graphical ratio of the (002) XRD peak intensity relative to the background at the same scattering angle, have been estimated [38]. All these values are summarized in Table 1. Figure 4 shows diffractograms of the final carbons obtained after the full template dissolution treatment. In all cases, reflections at 2θ ≈ 25°, 44° and 80° are observed, corresponding, respectively, to the families of planes (002), (100) and (110) of graphitic carbon. Diffraction of (002) planes is attributed to the structural packing of the aromatic layers through van der Waals forces, while (100) and (110) signals arise from the aromatic structures in the plane [34,35]. The intensity of the (002) peak is related to the degree of graphitic crystallinity in such a way that wide and low-intensity peaks indicate a low order at short range of the graphite carbon sheets organization, while sharp and narrow peaks correspond to highly crystalline carbons with a high degree of order at long range [34][35][36].
To evaluate the degree of crystallinity of the obtained carbons, the crystallite height (Lc) and crystallite diameter (La) parameters were calculated. The magnitudes of these parameters determine the sizes of the crystals within the carbon sample [34,35,37]. An increase in those parameters an increased graphitization of the carbon. Additionally, the interlayer spacing (d002), the number of aromatic layers per carbon crystallite (Nave) and the R value, considered as the empirical graphical ratio of the (002) XRD peak intensity relative to the background at the same scattering angle, have been estimated [38]. All these values are summarized in Table 1. (002)

2 angle
Intensity (a.u.)  Insignificant differences are observed within the synthesized series of materials, pointing to a similar degree of crystallinity for all samples. While the presence of dopants such as N and K results in a lower value of the L c parameter and, therefore, a decrease in the average number of graphenic layers (N ave ), the C(B_Glu_P) sample remains similar to the undoped carbon. Figure 5 presents N 2 adsorption isotherms of the obtained carbons. According to IU-PAC [39], all present type IV isotherms with an H3 hysteresis loop in the 0.4 < P/P 0 < 0.9 range, indicating a mesoporous character consisting of aggregates of particles with slit-shaped pores and a nonhomogeneous distribution in shape and/or size [40][41][42]. A low microporosity proportion is suggested by the small adsorption at very low pressures (P/P 0 < 0,1) [41], and the final materials can be described as micromesoporous carbons [39,43].
Catalysts 2021, 11, 1132 6 of 18 Insignificant differences are observed within the synthesized series of materials, pointing to a similar degree of crystallinity for all samples. While the presence of dopants such as N and K results in a lower value of the Lc parameter and, therefore, a decrease in the average number of graphenic layers (Nave), the C(B_Glu_P) sample remains similar to the undoped carbon. Figure 5 presents N2 adsorption isotherms of the obtained carbons. According to IU-PAC [39], all present type IV isotherms with an H3 hysteresis loop in the 0.4 < / 0 < 0.9 range, indicating a mesoporous character consisting of aggregates of particles with slitshaped pores and a nonhomogeneous distribution in shape and/or size [40][41][42]. A low microporosity proportion is suggested by the small adsorption at very low pressures ( ∕ 0 < 0,1) [41], and the final materials can be described as micromesoporous carbons [39,43]. Pore volumes were calculated from the desorption branch using the BJH method. According to the pore size distribution and assuming cylindrical pores (Figure 6), all samples present dominant mesoporosity with the main pores around 4 nm in diameter and a certain total volume fraction due to larger pores (20-50 nm) and micropores.  Pore volumes were calculated from the desorption branch using the BJH method. According to the pore size distribution and assuming cylindrical pores ( Figure 6), all samples present dominant mesoporosity with the main pores around 4 nm in diameter and a certain total volume fraction due to larger pores (20-50 nm) and micropores. Pore volumes were calculated from the desorption branch using the BJH method. According to the pore size distribution and assuming cylindrical pores ( Figure 6), all samples present dominant mesoporosity with the main pores around 4 nm in diameter and a certain total volume fraction due to larger pores (20-50 nm) and micropores.  Table 2 summarizes the textural properties of the synthesized carbons. All samples have high BET surface areas (typically around 1000 m 2 /g) and pore volume (higher than 1,2 cm 3 /g) characteristics for mostly mesoporous carbons [41]), which makes them suitable for possible applications as catalytic supports or absorbents. A revision of some uses of templated mesoporous carbons can be found in Ref. [3]. The most suitable application will depend not only on the carbon's pore characteristics but also on the presence of surface functional groups, which, as we will show later, is also dependent on the used dopant agent. In any case, the textural characteristics of our carbons are similar to those reported for mesoporous carbons acting as efficient supercapacitors as catalysts or as adsorbents of organic molecules and metal ions. Saygılı et al. [44] reported that activated carbons produced from tomato wastes in different pyrolysis conditions with a maximal BET area of 1093 m 2 /g, a total pore volume of 1569 cm 3 /g and a mean diameter of 5.92 nm can successfully be used as a sorbent to remove cationic (methylene blue) and anionic (metanil yellow) dyes from aqueous solutions. Santos et al. [45] reported the beneficial effect of the mesoporosity of carbon materials on palladium catalyst activity in the reaction of clean hydrogen production via formic acid decomposition. A pore volume of 1 cm 3 /g and pore sizes of around 2.7 nm are found optimal for maximal hydrogen productivity. The introduction of heteroatoms and potassium does not seem to have a great influence on the textural properties of the materials, except in the case of the C(B_Glu_P) sample. The latter presents different textural properties from the rest, such as a much smaller surface area, lower mesoporosity and a higher percentage of micropores. This is due to the formation of crystalline microporous aluminum phosphate, AlPO 4 , that results in less mesoporous carbon after pyrolysis. This fact confirms the chief role of template porosity for developing carbon replica textural properties. Table 3 shows the most relevant data of the thermogravimetric analysis of the carbonaceous materials after the first template dissolution stage and after its complete elimination. In general, two processes of weight loss are observed: one at low temperature (around 50 • C) ascribed to the loss of adsorbed water and another at higher temperatures due to the combustion of the carbonaceous skeleton. The percentage of weight remaining after combustion (ashes) is related to the mineral content of the sample, indicative of template leftovers. Figure 7 shows, as an example, graphs corresponding to the B_Glu sample, while Figures S3-S5 correspond to the other samples. In general, two processes of weight loss are observed: one at low temperature (around 50 °C) ascribed to the loss of adsorbed water and another at higher temperatures due to the combustion of the carbonaceous skeleton. The percentage of weight remaining after combustion (ashes) is related to the mineral content of the sample, indicative of template leftovers.  After the first dissolution stage, the amount of ashes is much higher than that of the final carbons, confirming the presence of a higher template proportion. Percentages of ashes in the final materials (after the complete dissolution process) are considerably reduced, demonstrating the effectiveness of the template removal process. Only for the B_Glu_P sample is the amount of ashes similar after the first and the final dissolution stages. This must be related to the formation of an AlPO4 phase, not fully removed with the dissolution treatment, as evidenced by XRD (Figures 3B and 4).
Comparing the results for the final carbons ( Figure 8 and Table 3), it is clear that C(B_Glu) carbon has a lower capacity to retain physisorbed water than doped carbons, which is evidenced by the smallest loss of weight at low temperatures (~3 wt% vs. ~13 wt%). This fact points to a higher hydrophilic character for all doped samples, most probably related to a higher functionalization degree. After the first dissolution stage, the amount of ashes is much higher than that of the final carbons, confirming the presence of a higher template proportion. Percentages of ashes in the final materials (after the complete dissolution process) are considerably reduced, demonstrating the effectiveness of the template removal process. Only for the B_Glu_P sample is the amount of ashes similar after the first and the final dissolution stages. This must be related to the formation of an AlPO 4 phase, not fully removed with the dissolution treatment, as evidenced by XRD (Figures 3B and 4).
Comparing the results for the final carbons ( Figure 8 and Table 3), it is clear that C(B_Glu) carbon has a lower capacity to retain physisorbed water than doped carbons, which is evidenced by the smallest loss of weight at low temperatures (~3 wt% vs.~13 wt%). This fact points to a higher hydrophilic character for all doped samples, most probably related to a higher functionalization degree. Taking into account the combustion temperature of the carbonaceous skeleton (Tabl the thermal resistance of the obtained carbons can be deduced. The following order of ther resistance can be established: C(B_Glu_P) > C(B_Glu_N) > C(B_Glu_K) > C(B_Glu). All do solids have higher thermal stability than their undoped homologs, showing the positive e of heteroatom introduction and suggesting that all doped samples will be more difficu oxidize. This must be also related to the differences in crystallinity, structural disorder, centration of defects and edge plane sites exposition [46,47], with all doped samples pres ing less disorder than undoped carbon. In any case, the differences are small among t three samples, as stated for XRD (Table 1 and Figure 4) and Raman (see later) analysis. superior thermal stability of the phosphorus-doped sample in comparison to all other ma als (120-150 °C higher) should also be mentioned, this effect being attributed to the pres of AlPO4.
Raman spectra of the prepared carbons in the first-order region (1800-1100 cm −1 ) shown in Figure 9. Two bands are clearly visible: one more intense around 1600 cm −1 , kn as the G-band, characteristic of sp 2 C hybridization and related to the degree of graphitiza and another wide band around 1300 cm −1 , known as the D-band, attributed to the presenc structural defects, partially disordered structures and sp 3 hybridization of carbon. The esti tion of the intensity ratio between both D-and G-bands (D/G ratio) is generally used to e uate the disordered character of carbonaceous materials [38,48,49]. All prepared samples sent similar D/G ratios ( Table 4), characteristic of medium-grade disordered carbons [50]. tailed analysis of the D/G ratios indicates that, except for the C(B_Glu_P) sample, the do carbons present slightly higher D/G values, indicating mild modifications of the disorder gree, which could be related to the observed changes in the combustion temperature and ter adsorption capacity (Table 3). Taking into account the combustion temperature of the carbonaceous skeleton (Table 3), the thermal resistance of the obtained carbons can be deduced. The following order of thermal resistance can be established: C(B_Glu_P) > C(B_Glu_N) > C(B_Glu_K) > C(B_Glu). All doped solids have higher thermal stability than their undoped homologs, showing the positive effect of heteroatom introduction and suggesting that all doped samples will be more difficult to oxidize. This must be also related to the differences in crystallinity, structural disorder, concentration of defects and edge plane sites exposition [46,47], with all doped samples presenting less disorder than undoped carbon. In any case, the differences are small among these three samples, as stated for XRD (Table 1 and Figure 4) and Raman (see later) analysis. The superior thermal stability of the phosphorus-doped sample in comparison to all other materials (120-150 • C higher) should also be mentioned, this effect being attributed to the presence of AlPO 4 .
Raman spectra of the prepared carbons in the first-order region (1800-1100 cm −1 ) are shown in Figure 9. Two bands are clearly visible: one more intense around 1600 cm −1 , known as the G-band, characteristic of sp 2 C hybridization and related to the degree of graphitization; and another wide band around 1300 cm −1 , known as the D-band, attributed to the presence of structural defects, partially disordered structures and sp 3 hybridization of carbon. The estimation of the intensity ratio between both D-and G-bands (D/G ratio) is generally used to evaluate the disordered character of carbonaceous materials [38,48,49]. All prepared samples present similar D/G ratios ( Table 4), characteristic of medium-grade disordered carbons [50]. Detailed analysis of the D/G ratios indicates that, except for the C(B_Glu_P) sample, the doped carbons present slightly higher D/G values, indicating mild modifications of the disorder degree, which could be related to the observed changes in the combustion temperature and water adsorption capacity (Table 3).
Infrared studies evidence the similarity of the surface functionalities and carbonaceous structure of the obtained carbons ( Figure 10). In addition to aliphatic C-C (1460 cm −1 ), C-H (2800-3000 cm −1 ) and aromatic C=C (1605 cm −1 ) and C-H (3060 cm −1 and group of bands at 660-910 cm −1 ) bonds, bands characteristic of oxygenated functionalities are also visible: C=O (1725 cm −1 ) and C-O (1110-1300 cm −1 ). The functionalization of the surface is also demonstrated by the wide band observed at around 3400 cm −1 due to O-H stretching of surface-adsorbed water. No bands due to the oxidic template are envisaged, demonstrating its complete removal. Only the C(B_Glu-P) sample shows small changes in the 1100-1000 cm −1 region, related to the existence of some traces of AlPO4, as P-O and Al-O stretching vibrations have been described to appear in this region [50].

Sample D/G ratio
Infrared studies evidence the similarity of the surface functionalities and carbonaceous structure of the obtained carbons ( Figure 10). In addition to aliphatic C-C (1460 cm −1 ), C-H (2800-3000 cm −1 ) and aromatic C=C (1605 cm −1 ) and C-H (3060 cm −1 and group of bands at 660-910 cm −1 ) bonds, bands characteristic of oxygenated functionalities are also visible: C=O (1725 cm −1 ) and C-O (1110-1300 cm −1 ). The functionalization of the surface is also demonstrated by the wide band observed at around 3400 cm −1 due to O-H stretching of surface-adsorbed water. No bands due to the oxidic template are envisaged, demonstrating its complete removal. Only the C(B_Glu-P) sample shows small changes in the 1100-1000 cm −1 region, related to the existence of some traces of AlPO 4 , as P-O and Al-O stretching vibrations have been described to appear in this region [50].
Infrared studies evidence the similarity of the surface functionalities and carbonaceous structure of the obtained carbons ( Figure 10). In addition to aliphatic C-C (1460 cm −1 ), C-H (2800-3000 cm −1 ) and aromatic C=C (1605 cm −1 ) and C-H (3060 cm −1 and group of bands at 660-910 cm −1 ) bonds, bands characteristic of oxygenated functionalities are also visible: C=O (1725 cm −1 ) and C-O (1110-1300 cm −1 ). The functionalization of the surface is also demonstrated by the wide band observed at around 3400 cm −1 due to O-H stretching of surface-adsorbed water. No bands due to the oxidic template are envisaged, demonstrating its complete removal. Only the C(B_Glu-P) sample shows small changes in the 1100-1000 cm −1 region, related to the existence of some traces of AlPO4, as P-O and Al-O stretching vibrations have been described to appear in this region [50].  The morphology, homogeneity and elemental chemical composition of the obtained solids were studied by TEM (Figures 11 and 12). The carbons do not present a hierarchical structure but a high degree of porosity, with pores within the mesoporosity order originated by the effective nanocasting process. Again, the C(B_Glu_P) sample presents a different structure, and two regions can be differentiated: one composed of several very thin sheets and more open porosity (Region I, marked in red) and the other more similar to that observed for all other samples (Region II, marked in yellow).
Catalysts 2021, 11, 1132 11 of 18 The morphology, homogeneity and elemental chemical composition of the obtained solids were studied by TEM (Figures 11 and 12). The carbons do not present a hierarchical structure but a high degree of porosity, with pores within the mesoporosity order originated by the effective nanocasting process. Again, the C(B_Glu_P) sample presents a different structure, and two regions can be differentiated: one composed of several very thin sheets and more open porosity (Region I, marked in red) and the other more similar to that observed for all other samples (Region II, marked in yellow).   Regarding the elemental composition of the materials, EDS compositional mappings for C(B_Glu_K) and C(B_Glu_P) are compared in Figure 12 (the other samples are summarized in Figures S6 and S7). Average EDS composition values are depicted in Table 5.  Despite the apparently effective template removal evidenced by X-ray diffraction and Raman, aluminum (Al) is detected in low concentrations. Some sodium (Na) is also Regarding the elemental composition of the materials, EDS compositional mappings for C(B_Glu_K) and C(B_Glu_P) are compared in Figure 12 (the other samples are summarized in Figures S6 and S7). Average EDS composition values are depicted in Table 5. Despite the apparently effective template removal evidenced by X-ray diffraction and Raman, aluminum (Al) is detected in low concentrations. Some sodium (Na) is also present and originated from the NaOH used for template dissolution. Additionally, silicium (Si), coming from the beaker where the dilution treatments were carried out, is noticed. The two regions with differentiated morphologies observed for the C(B_Glu_P) sample also have different compositions.

I region
Concerning the incorporation of doping elements, a homogeneous distribution of each heteroelement is detected. K is present in lower concentrations in the C(B_Glu_K) sample due to its partial elimination during the washing process. Nevertheless, higher concentrations of P and N are detected in the C(B_Glu_N) and C(B_Glu_P) samples, respectively, indicating more effective doping. The mappings allow verifying that these elements are homogeneously distributed in the synthesized solids.
For all above, it is evident that boehmite is an adequate candidate to serve as a template for pristine or doped (N, P, K) mesoporous carbon production by nanocasting. Boehmite is a cheap, easily available material with controlled mesoporosity. Although boehmite is transformed at higher temperatures in transitional aluminas, our paper demonstrates that pyrolysis temperatures up to 700 • C can be used without compromising the formation of mesoporous carbons. The proposed method is simple, low cost and very versatile, as the control and modification of the different synthetic parameters allows a large variety of carbon materials. The mesomicroporous character of the final carbon can be modulated by the modification of the textural properties of the starting boehmite and/or by changing the pyrolysis conditions. It is well reported that modifications of pyrolysis process parameters (temperature, heating ramp, number of heating steps, atmosphere flow and nature, residence time, etc.) and biomass activation (prior pyrolysis) with different chemical agents result in carbonaceous materials with different textural, chemical and surface properties [51][52][53]. In fact, one of the main advantages of the pyrolysis procedure is that it can be optimized according to the desired properties of the chars. Therefore, although fixed pyrolysis conditions are used in this paper, this possibility opens the range of properties of the obtained materials. The nature and properties of the produced carbons can also be tailored by changing the carbon precursor and the doping agent. Here, we used glucose as the residual biomass representative, but any carbon-containing molecule can be employed.
Additionally, our method is effective for the production of hybrid carbonaceous materials (without template removal). Again, the control of pyrolysis conditions (temperature, atmosphere, etc.), carbon source and dopant precursor could tailor the properties and nature of the hybrid. The use of H 3 PO 4 as a P source is particularly interesting with the formation of a microporous crystalline AlPO 4 -C hybrid with an ordered microporous structure with potential catalytic properties.

Mesoporous Boehmite Synthesis
The sol-gel process was employed to produce mesoporous boehmite [54]. Aluminum isopropoxide, Al[(CH 3 ) 2 CHO] 3 , dissolved in 2-propanol, was used as a precursor, and the process was catalyzed by ammonia. After cooling to room temperature, a solution of 1.4 mL of NH 3 (Panreac Química S.L.U., Barcelona, Spain), 30% in 10 mL of deionized water was added dropwise to hydrolyze the alkoxide. Under these conditions, the mixture can gel rapidly. The resulting gel was dried in an oven at 85 • C for 2 d and finally calcined at 300 • C for 2 h [55]. The XRD diagram ( Figure S8) corroborates the formation of the boehmite structure (JCPDS: 00-049-0133), and its mesoporous character is confirmed by the N 2 isotherms ( Figure S9), with a BET surface area of 351 m 2 /g, a mean pore diameter of 16.6 nm and a pore volume of 1.74 cm 3 /g.

Mesoporous Carbons Synthesis
A 2.5 g amount of glucose (Sigma-Aldrich Merck KGaA, Darstadt, Germany, purity: ≥99.5%) was dissolved in 4 mL of deionized water and added dropwise to 2 g of tem-plate, taking into account the total pore volume of the starting template (approximated to 2 cm 3 /g). The obtained solid was dried in an oven at 100 • C for 16 h.
To prepare the doped samples, three more impregnations were performed following the procedure described above, letting them dry in an oven at 100 • C for 1 h and adding, dropwise, a sufficient amount per gram of the following doping agent solution to ensure that they had a relative influence in the structure and properties of the final solid: (a) 50 mg of potassium hydroxide (KOH, Sigma-Aldrich Merck KGaA, Darstadt, Germany) purity ≥85%) dissolved in 2 mL of deionized water; (b) 100 mg of urea (CH 4 N 2 O, Sigma-Aldrich Merck KGaA, Darstadt, Germany)) dissolved in 2 mL of deionized water; (c) 1 mL of phosphoric acid (H 3 PO 4 , (Panreac Química S.L.U., Barcelona, Spain), 85%) diluted in 1 mL of deionized water. The obtained solids were dried at 100 • C for 16 h. Table 6 shows the used nomenclature for the prepared samples, and Figure 13 summarizes the appearance of the impregnated boehmite samples after drying. The color of the solid reflects a certain degradation (dehydration) degree of glucose as a result of the different compounds used in the impregnation and the thermal dryness treatment. Table 6. Nomenclature of the synthesized samples.

Mesoporous Carbons Synthesis
A 2.5 g amount of glucose (Sigma-Aldrich Merck KGaA, Darstadt, Germany, purity: ≥99.5%) was dissolved in 4 mL of deionized water and added dropwise to 2 g of template, taking into account the total pore volume of the starting template (approximated to 2 cm 3 /g). The obtained solid was dried in an oven at 100 °C for 16 h.
To prepare the doped samples, three more impregnations were performed following the procedure described above, letting them dry in an oven at 100 °C for 1 h and adding, dropwise, a sufficient amount per gram of the following doping agent solution to ensure that they had a relative influence in the structure and properties of the final solid: (a) 50 mg of potassium hydroxide (KOH, Sigma-Aldrich Merck KGaA, Darstadt, Germany) purity ≥85%) dissolved in 2 mL of deionized water; (b) 100 mg of urea (CH4N2O, Sigma-Aldrich Merck KGaA, Darstadt, Germany)) dissolved in 2 mL of deionized water; (c) 1 mL of phosphoric acid (H3PO4, (Panreac Química S.L.U., Barcelona, Spain), 85%) diluted in 1 mL of deionized water. The obtained solids were dried at 100 °C for 16 h. Table 6 shows the used nomenclature for the prepared samples, and Figure 13 summarizes the appearance of the impregnated boehmite samples after drying. The color of the solid reflects a certain degradation (dehydration) degree of glucose as a result of the different compounds used in the impregnation and the thermal dryness treatment. Table 6. Nomenclature of the synthesized samples.
For pyrolysis, the impregnated templates were placed in a ceramic vessel within the controlled temperature zone of the horizontal oven. The pyrolysis process was carried out in two stages: (i) 5 °C/min until 200 °C, maintaining this temperature for 30 min, and (ii) 10 °C/min until 700 °C, remaining for 1 h at this temperature. A constant N2 flow of 100 mL/min was used during the treatment.
These pyrolysis conditions under an inert atmosphere were selected in order to assure the formation of nonactivated and nonmesoporous carbons. As reported, microporous carbons are generally produced under such conditions [53,56]. To demonstrate the decisive role of the boehmite template in producing mesoporous carbons, the utilization of pyrolysis atmosphere and conditions not forming mesoporous carbonaceous materials are mandatory.

Template Removal
Four consecutive dissolution steps were carried out to remove the template after pyrolysis, the first three steps with hot 4M NaOH (NaOH Panreac Química S.L.U., Barcelona, Spain, ≥85%) and the last step with 1M NaOH at room temperature. In the hot steps, the samples were mixed with 50 mL of solvent and left under agitation for 24, 48 and 96 h, respectively, as it was proved that the agitation time influenced the effectiveness of the For pyrolysis, the impregnated templates were placed in a ceramic vessel within the controlled temperature zone of the horizontal oven. The pyrolysis process was carried out in two stages: (i) 5 • C/min until 200 • C, maintaining this temperature for 30 min, and (ii) 10 • C/min until 700 • C, remaining for 1 h at this temperature. A constant N 2 flow of 100 mL/min was used during the treatment.
These pyrolysis conditions under an inert atmosphere were selected in order to assure the formation of nonactivated and nonmesoporous carbons. As reported, microporous carbons are generally produced under such conditions [53,56]. To demonstrate the decisive role of the boehmite template in producing mesoporous carbons, the utilization of pyrolysis atmosphere and conditions not forming mesoporous carbonaceous materials are mandatory.

Template Removal
Four consecutive dissolution steps were carried out to remove the template after pyrolysis, the first three steps with hot 4M NaOH (NaOH Panreac Química S.L.U., Barcelona, Spain, ≥85%) and the last step with 1M NaOH at room temperature. In the hot steps, the samples were mixed with 50 mL of solvent and left under agitation for 24, 48 and 96 h, respectively, as it was proved that the agitation time influenced the effectiveness of the dissolution process. After this, the mixtures were heated to 60 • C for 1 h and filtered while still hot. In the last step, the mixtures were directly filtered after the agitation process. This template removal procedure was established from our previous studies on the effect of the NaOH concentration, time and temperature of treatment, on the dissolution of boehmite and alumina pure solids. Once filtered, the samples were subjected to several washings, an initial wash with a portion of the corresponding NaOH solvent and then with water up to neutral pH. Finally, the resulting products were dried in an oven at 100 • C for 2 h.

Characterization Techniques
Conventional XRD measurements were performed on an X'Pert Pro PANalytical diffractometer (Malvern Panalytical Ltd., Malvern, UK) equipped with a Cu anode and a graphite monochromator diffracted beam, working at 45 kV and 40 mA. Diffractograms were obtained in a continuous scan mode from 10 to 90 • 2θ using a 0.05 • step size and an acquisition time of 300 s. For each diffraction pattern, structural determination was performed by comparison with the database PDF2 ICDD2000 (Powder Diffraction File 2 International Center for Diffraction Data, 2000), PA, USA).
From the XRD analysis, quantitative information of carbonaceous structure could be obtained following the approach of Alexander et al. [57]. According to these authors, based on the reduced intensity, several structural parameters are calculated: crystallite diameter along basal planes, which is the length of the crystalline domain in the plane (100) (L a ); the average stacking height or length of the crystalline domain in the plane (002) (L c ); the interlayer spacing (d 002 ); and the R value, defined as the ratio of the (002) peak intensity relative to the background at the same scattering angle. This parameter is related to the concentration of single graphene layers in a sample in such a way that the lower the R value, the higher the concentration of single layers [58].
Bragg's equation was applied to calculate the interlayer spacing: (1) L a and L c parameters were calculated applying the Scherrer equation: L a = K a λ β 100 cosθ 100 (3) where λ corresponds to the X-ray wavelength used (1.18 Å, copper Kα), K is a dimensionless shape factor (0.89 for the reflection (002), K c , and 1.84 for the (100) one, K a ) [28], β is the full width at half maximum (FWHM) of the considered diffraction peak and θ is the peak position. Finally, the average number of aromatic layers per carbon crystallite (N ave ) has been also calculated as: (L c /d002) + 1.
The textural properties of synthesized solids were determined from the nitrogen adsorption-desorption isotherms obtained at 77 K using Micromeritics Tristar II equipment. (Micromeritics, Norcross, GA, USA). Prior to the measurement, samples were degassed under vacuum at 350 • C for 12 h using a vacuum degasser system 061 VacPrep of Micromeritics. Brunauer-Emmett-Teller (BET) method, t-plot and Barrett-Joyner-Halenda (BJH) method were used to calculate the specific surface area, the micropore area and the pore size distribution.
Thermogravimetric analysis was performed using TA Instruments Q600 equipment (TA Instruments, New Castle, NC, USA). Samples were heated from room temperature to 900 • C with a heating rate of 10 • C/min under 100 mL/min flow of pure synthetic air.
Raman spectroscopy measurements were carried out on an HR800 Horiba Jobin Yvon dispersive microscope with a 600 grooves/mm diffraction grating, using a green laser (λ = 532.14 nm) working at 5 mW power, a 50 × objective and a confocal aperture of 1000 microns. TEM micrographs were acquired with a FEI Talos electron microscope (Hillsboro, OR, USA) working at 200 kV and equipped with a field emission filament. The compositional analyses of the samples were performed by energy dispersive X-Ray spectroscopy (EDS). For TEM preparation, a few milligrams of each sample was deposited directly on a 300mesh holey carbon-coated copper TEM grid.
Infrared spectra of the samples were collected in diffuse reflectance mode (DRIFTS) on a JASCO FT/IR-6200 6200 (JASCO International Co., Easton, MD, USA) unit to which a PIKE reflector accessory model EASI-DIFF was attached. The spectra were acquired on the pure samples (without dilution) in the range of 600 to 4000 cm −1 and with a resolution of 4 cm −1 , accumulating 100 scans.

Conclusions
Boehmite is a very effective template to obtain mesoporous carbonaceous materials by the nanocasting method. The incorporation of different doping agents during the synthetic process is also satisfactory, resulting in a homogeneous distribution of the doping element. Nevertheless, the K doping procedure results in a lower doping element concentration in comparison to the rest of the doped carbons.
Doping with KOH and urea insignificantly modifies the textural properties of the resulting carbon. However, the use of H 3 PO 4 as a dopant leads to a chemical reaction with the boehmite template and the formation of AlPO 4 , an indirect template, resulting in solids with lower specific surface values and a higher percentage of microporosity. This fact clearly demonstrates the significant influence of the used template on the textural properties of the final carbon.
All doped carbons show better thermal resistance than the undoped carbons, particularly the P-doped sample, with a combustion temperature 120 • C higher than the rest of the samples.
The described process is an effective, easy, versatile and economical method to obtain mesoporous carbons from renewable biomass sources.