The oxidation with nitric acid was performed using a Soxhlet extraction apparatus. Thus, 200 cm³ of 5 M nitric acid solution was introduced into a 250 cm³ Pyrex round-bottom flask and heated to boiling temperature with a heating mantle. The Soxhlet (250 cm³) with 2 g of carbon xerogels was connected to boiling temperature flask and to the condenser. The reflux was stopped after 6 h. The sample was washed with distilled water to neutral pH and dried in an air convection oven at 110 °C for 24 h. These materials were designated as CX followed by the respective pH, carbonization temperature and HNO₃.

About 3 g of sample was placed in the oven and heated at 10 °C min⁻¹ until 500 °C, passing only a N₂ flow (75 cm³ min⁻¹). When this temperature was reached, 25 cm³ min⁻¹ of air was added. After 3 h the sample was cooled down in an inert atmosphere. These materials were designated as CX followed by the respective pH, carbonization temperature and O₂.

A 2 g of sample was impregnated with urea solution (1M) by stirring it for 24 h. Then it was heated in nitrogen flow by raising the temperature at 10 °C min⁻¹ to 600 °C for 1 h. The samples obtained by this treatment were designated as CX followed by the respective pH, carbonization temperature and U.

The textural characterization of the samples was based on the nitrogen adsorption isotherms at −196 °C, which were determined using a Quantachorome Nova 4200e instrument. The specific volume of micropores (V_micro) and the specific area of mesopores (S_meso) were calculated by the t-method, using the appropriate standard isotherm. The specific surface area of carbon xerogels was calculated by the BET method (S_BET).

Temperature programmed desorption (TPD) was performed in an AMI-200 Catalyst Characterization Instrument (Altamira Instruments), equipped with a quadrupole mass spectrometer (Dymaxion, Ametek). The catalyst sample was placed in a U-shaped quartz tube inside an electrical furnace and heated at 5 °C min⁻¹ up to 1100 °C using a constant flow rate of helium (25 cm³ min⁻¹). The masses monitored for all samples were: 2 (H₂), 18 (H₂O), 28 (CO) and 44 (CO₂) [31].

To determine the point of zero charge of the samples (pH_pzc), 20 cm³ of NaCl 0.01 M were placed in closed vessels and the pH was adjusted to values between 2 and 12, by adding 0.1 M solutions of NaOH or HCl. Then, 0.05 g of sample was added to each vessel. The suspensions were kept under stirring at room temperature for 48 h and then the final pH was recorded. The pH_pzc of each sample was determined by the intersection of the curve pH_finalvs pH_initial with the line pH_final = pH_initial. Blank experiments were carried out, and the pH_initial used in each case was that corresponding to the pH_final of the blank test [32].

The chemical composition (C, H, N, S and O) of the carbon xerogels was analysed using an EA 1108 Elemental Analyser (Carlo Erba instruments).

Thermal analysis (TG) was performed on the xerogel samples in order to obtain information about the loss of water and carbonaceous residues involved in the carbonization step. Thermal analyses were done in a Netzsch STA 409 PC Luxx instrument.

Figure 1 shows the nitrogen adsorption isotherms of carbon xerogels. It is noticeable that the textural properties of carbon xerogels were affected by the synthesis pH and by the carbonization temperature.

The isotherms are typical of materials with micro and mesoporosity, except in the case of xerogels synthesized at pH = 6.9, which show very little or no microporosity. When the pH is high, polymerization is retarded and the formation of a cross-linked network is prevented [33,34], leading to a weak pore structure that cannot withstand carbonization at the higher temperatures (700 and 900 °C), resulting in the collapse of the nanostructure.

The isotherms of the materials prepared at pH = 5.3, independently of the carbonization temperature, exhibit a large increase in the amount of nitrogen adsorbed at relative pressures above 0.9, indicating the presence of very large pores.

Table 1 collects the textural properties of the materials prepared.

In the case of carbon xerogels prepared at pH = 5.3 and pH = 6.0, the maximum BET surface area and micropore volume correspond to the materials carbonized at 700 °C. The largest mesopore surface area corresponds to sample CX-6-900 °C.

These results confirm that it is possible to control the texture of the carbon xerogels by adjusting the pH of the initial solution and the carbonization temperature.

The effect of pH on the surface area and micropore volume of carbon xerogels carbonized at 700 °C is shown in Figure 2.

It is clear that the micropore volume and the surface area of carbon xerogels are very low at high pH. When the pH decreases from 6.9 to 6.0, there is a large increase of micropore volume and surface area. With further decrease to pH = 5.3, there is not much change in the textural properties.

The effect of pH on the porosity of carbon xerogels can be explained by the polymerization mechanism between resorcinol and formaldehyde, as described in the literature [34].

The subsequent treatments performed on the carbon xerogels were intended to modify their surface chemical properties, but additional changes in the textural properties also occur [31]. In general, the BET surface area increases upon treatment with nitric acid and with oxygen in the gas phase. The development of the porosity of the materials probably occurs by the widening of existing pores and/or creation of new pores by selective gasification of structural components, or by the opening of some of the previously inaccessible pores.

Figure 3 shows the TG curves for the organic xerogels (RF) synthesized at different pHs. All the TG curves have a common behavior; the different initial pH values do not lead to any major difference.

Two peaks of mass loss were detected during the carbonization process: one at 200 °C and another at 300 °C. The first peak probably corresponds to the extraction of the remaining solvent and/or the elimination of H₂O formed by the condensation of –OH groups [34]. The second peak is observed when the remaining hydrogen and oxygen atoms included in the polymer network are eliminated as CO₂, CH₄ or other organic molecules [34].

The results of elemental analysis of the original and modified carbon xerogels are shown in Table 2.

The process of pyrolysis, development of the surface area and generation of new pores, leads to an increase in the content of carbon, and a significant decrease in the content of hydrogen and oxygen.

The oxidative treatments with nitric acid in liquid phase and with 5% of oxygen in gas phase introduce a large amount of oxygen groups on the surface of carbon xerogels. For example, sample CX-6-700°C has 7% of oxygen and after the treatment in the gas phase presents 16% of oxygen.

The treatment with nitric acid, in addition to introducing large amounts of oxygen groups in the carbon xerogels, also introduces some nitrogen groups. Larger amounts of nitrogen groups are introduced by the treatment with urea. The incorporation of nitrogen from urea occurs by derivatization, predominantly of the carboxylic oxygen species, and in general leads to a decrease of the oxygen content.

The total amount of the various oxygen surface groups present on the carbon xerogels (carboxylic acids, carboxylic anhydrides, lactones, phenols and carbonyls or quinones) can be determined by the TPD spectra, since these groups are decomposed into CO and/or CO₂ upon heating.

The total amounts of CO and CO₂ released are presented in Table 3.

The ratio of CO/CO₂ released, included in Table 3, can be regarded as a measure of the surface acidity and basicity. Low values of the ratio correspond to strong surface acidity while high values of this parameter correspond to strong surface basicity [31]. The samples treated with HNO₃ show the lowest CO/CO₂ ratio, indicating that the surface chemistry of these samples is acidic in character, since this treatment introduces large amounts of acid groups (carboxylic acids) onto the surface of carbon materials. This result is consistent with the lowest values of pH_pzc exhibited by these samples.

All the other materials showed neutral or basic properties (pH_pzc ~ 6.9–8.1) independently of the pH used in their preparation.

The carbon xerogels carbonized at 500 °C present large amounts of surface oxygenated groups, which are partially explained by the relatively low temperature used during their preparation. Most of the CO and CO₂ released above 500 °C is still the result of the xerogel matrix carbonization. The amount of oxygen-containing surface groups decreases as the carbonization temperature increases.

In general, the samples treated with urea have higher values of pH_pzc than the samples treated with oxygen. This is due to the introduction of nitrogen groups, which increase the basicity of the carbon materials, namely pyridine and pyrrole groups which are Lewis bases.

Figure 4 shows the influence of O₂ concentration on NO conversion at 25 °C.

In the presence of oxygen, the carbon materials catalytically oxidize NO with rates larger than those of the homogeneous oxidation (X_NO: 36% for 20% O₂).

The oxidation of NO to NO₂ increases with the increase of O₂ concentration from 2% to 10%. Further increase of oxygen concentration from 10% to 20% does not affect so much the conversion. Even in the presence of 2% oxygen, the NO conversion can reach 86% with sample CX-5.3-900 °C.

In some cases, the maximum NO conversion was obtained when 10% of oxygen was used. It seems that there is an optimum oxygen concentration above which the NO conversion levels off due to saturation of the adsorption sites with atomic oxygen. The highest NO conversion was obtained on sample CX-5.3-900 °C with 10% of O₂ (X_NO: 98%).

In the experiments with carbon xerogels that have micropores, it was observed that no NO₂ is detected at the reactor outlet during the initial stages of the experiment. The NO₂ produced during the first minutes of reaction is retained in the micropores of the carbon xerogel. Thereafter, the concentration of NO₂ in the gas phase increases and reaches a plateau (steady state conversion). In the case of sample CX-6.9-900 °C, NO₂ is detected from the beginning of the reaction, because this sample has no micropores.

At steady state, the micropores are occupied with NO₂ adsorbed, so only the mesopore surface area is available for reaction. S_meso will thus subsequently normalize the rates.

The effect of oxygen in the removal of NO has been reported in many studies [7,8,9,10,12]. In the absence of oxygen, NO passes through the carbon xerogel bed unreacted.

There is virtually no NO adsorption when oxygen is absent, indicating that oxygen is needed for the process. These observations suggest that NO₂, rather than NO, is the adsorbed molecule, possibly because of its higher polarity.

The strong dependence of NO conversion on the oxygen concentration leads to the conclusion that oxygen is first adsorbed on the surface of carbon xerogels, and then it reacts with NO to form adsorbed NO₂. Then, NO₂ can desorb to the gas phase. A simplified scheme of this process is represented in Figure 5.

The highest NO conversions were obtained with the samples which presented the highest surface areas (CX-5.3-500 °C, CX-5.3-700°C, CX-5.3-900 °C, CX-6-700 °C and CX-6-900 °C). In addition, the lowest NO conversions were obtained with the samples which presented the lowest surface areas (CX-6-500 °C, CX-6.9-500 °C, CX-6.9-700 °C and CX-6.9-900 °C).

Therefore, the surface area determines the number of sites available for NO oxidation, showing that the surface of carbon xerogels is fully accessible and has a uniform distribution of active sites.

There are many contradictions about the influence of the texture of the carbon materials on NO oxidation. Shimizu et al. [35], Illan-Gómez et al. [36] and Jan et al. [37] reported a correlation between the activity of carbon materials for NO reduction and their surface area. Mochida et al. [38] found no relationship between surface area of the carbon materials and the removal of NO; however, the carbon materials used in this case were treated with sulphuric acid, and it is difficult to separate the effect of surface area and the effect of oxygen groups in the catalytic activity of carbon materials. Using nitrogen-doped carbon materials, the present authors did not find any correlation between NO conversion and surface area [8,39]. The carbon materials used in these works contained large amounts of nitrogen groups, which are known to enhance the activity of carbon materials in oxidation reactions [40,41,42].

Thus, the catalytic activity of carbon materials is conditioned both by their textural and chemical properties. Chemical functionalities existing on the carbon surface, such as oxygen groups with an acid or basic character, and heteroatoms (mainly nitrogen functionalities) are responsible for the catalytic activity of the carbon materials. In this sense, the textural properties contribute to the catalytic activity only as long as the surface of carbon materials does not suffer any modification, and in this case the most important textural property is the surface area.

Therefore, the surface area of the carbon xerogels is a key factor in determining their activity for NO oxidation.

In each range of pH, the carbon xerogels that presented the highest surface area were subjected to three different treatments (oxidation with 5% of oxygen in gas phase, oxidation with nitric acid 5 M in liquid phase and treatment with a 1 M urea solution at room temperature).

In the previous section, it was noted that the increase of oxygen concentration in the gas phase increases NO conversion. The treatments with nitric acid in liquid phase and with oxygen in gas phase were intended to introduce oxygen groups on the surface of the carbon xerogels and to evaluate their effect on the catalytic activity for NO oxidation. The oxygen groups introduced by these treatments are different. The treatment with nitric acid introduces mainly carboxylic acid groups, while the treatment with oxygen in gas phase introduces mainly phenol and carbonyl/quinone surface groups.

The role of nitrogen in catalysis is usually linked to its basicity and ability to activate oxygen via formation of O₂⁻ superoxide, which has tremendous effects on oxidation reactions [40,41,42].

This section is also intended to evaluate the effect of nitrogen groups introduced by post-treatment and to compare with N-doped carbon xerogels (where nitrogen is incorporated during the synthesis) [39]. Treatment with a urea solution at room temperature was performed in order to introduce nitrogen groups on the surface of the carbon xerogels. According to the literature [42,43,44], such treatment should introduce amide groups on the surface. Upon heating to 600 °C, these are converted into more stable nitrogen groups, such as pyridine and pyrrole [45].

The NO conversions obtained with the original materials and with the materials chemically modified are shown in Table 4.

Since these treatments also modified the textural properties of the carbon xerogels, Figure 6 shows the rate of NO converted per unit of (mesopore) surface area as a function of the amount of nitrogen, for the urea treated samples.

Although there are only three data points, there is a clear relationship between the rate of NO conversion and the amount of nitrogen. This is in agreement with our previous report [39], where the presence of nitrogen groups on carbon xerogels was shown to enhance the oxidation of NO to NO₂.

To assess the influence of surface oxygen groups on NO oxidation, the rate of NO converted per unit surface area was plotted as a function of the oxygen content (Figure 7).

It appears that the oxidative treatment of carbon xerogels, which produced a carbon surface largely covered by oxygen groups (Table 3), inhibits the oxidation of NO at low temperatures. The oxygen complexes formed during the treatments with nitric acid and with oxygen in gas phase are not active sites for the NO oxidation. These oxygen groups are stable under the reaction conditions, and so they occupy active centers of the carbon xerogels, decreasing the activity for NO oxidation.

The treatment with nitric acid introduces small amounts of nitrogen, but not in the form of pyrrole or pyridine (as does the urea treatment). On the other hand, it introduces very large amounts of oxygen groups, especially carboxylic acids. Therefore, the inhibitive effect of the oxygen groups offsets any beneficial effect that might be due to the presence of nitrogen.

In the context of the practical application of this process for the abatement of NO_x in flue gases of combustion systems, it is important to know the influence of other compounds such as CO and CO₂. The experiments were conducted with 1000 ppm NO, 20% O₂, 4% CO (CO₂) and at room temperature over samples CX-5.3-900 °C and CX-6-900 °C.

The NO conversions obtained over these samples in the presence of CO or CO₂ are equal to those obtained when the reaction was realized only with NO and O₂, so it may be concluded that the presence of CO (or CO₂) does not affect NO oxidation. It appears that CO and CO₂ do not compete with oxygen for the catalyst active sites.

The stability of a catalyst is an important aspect in the global evaluation of its performance. So, an experiment was carried out during one week with the purpose of evaluating the stability of carbon xerogels as catalysts for NO oxidation. The reaction was carried out with 1000 ppm of NO and 10% of oxygen over the most efficient catalyst (CX-5.3-900 °C). The results are shown in Figure 8. The steady-state conversion is reached after 26 h of reaction (X_NO: 98%), and does not change thereafter.

The dependence of the NO conversion on the reaction temperature was studied with sample CX-5.3-900 °C. The experiments were carried out with 1000 ppm of NO, 10% of O₂ and in the temperature range 25–150 °C. Figure 9 shows the profiles of NO oxidation at different reaction temperatures.

A conversion of 98% was obtained at 25 °C, but decreased to 86% at 50 °C, to 77% at 100 °C, and to 53% at 150 °C. Thus, reaction at room temperature was found to be favorable for the maximum oxidation of NO.

The carbon surface can act as a catalyst for oxidation of NO to NO₂, which then may remain adsorbed. Inspection of Figure 9 shows that no NO₂ is detected at the outlet during 523 min, at 25 °C. The NO₂ produced is completely retained on the carbon xerogel in this period. By increasing the reaction temperature, the induction time of NO₂ (time interval during which the NO₂ produced is retained) is shortened, because the adsorption of NO₂ decreases. So, at room temperature there are two phenomena, reaction and adsorption, but the latter becomes negligible at higher reaction temperatures.