Surface Properties of Synthesized Nanoporous Carbon and Silica Matrices

The critical steps during the preparation of the ordered mesoporous carbon material include the preparation of the ordered mesoporous silica materials as the template with well-defined structural properties that affect the properties of the final materials and a tempering/carbonization step under a nitrogen atmosphere. The modification of the typical method of preparation of the mesoporous ordered silicates with cylindrical pores²⁸ concerns the application of an untypical structure-directing agent that is PE10500 polymer, for the improvement of the structural properties of the material. The three-dimensional, interconnected, and stable porous structure of the template is necessary for the preparation of the mesoporous carbon materials. Moreover, a key disadvantage of the preparation is the essential requirement of the sample treatment for the template removal. The properties of chemicals used in this step can affect the carbon surface and its functionality. According to the presented strategy, the preparation of the negative replication of OMC is based on the ordered mesoporous solid template. The pore size control and symmetric ordering are simply determined using the silica template and are not associated with the interaction between the carbon precursor and the template. The potential of OMC for various electrochemical systems has been indicated in the literature^29,30. The impregnation mechanism presented in this work is responsible for the facile process to precisely replicate the negative structure of the silica template. The nature of the hard template procedure ensures that the pyrolysis phenomena cause less damage to the regular and ordered structure. Moreover, this method allows an easier graphitization of the OMC materials formed within the solid template.

The experimental N₂ adsorption/desorption isotherm (Figure 1A, C) was used to obtain preliminary information about the type and size of porosity. The nanoporous character of the material was clearly confirmed. A small increase in the quantity of nitrogen adsorbed in the initial stage of the process (in the range of relative pressure below 0.03) and its significant increase in the further range of relative pressure (especially in the range of mesopores) indicates the existence of a significant number of mesopores and the lack, or a very small number, of micropores. The adsorption-desorption isotherms obtained for this material are typical of type-IV (IUPAC classification³¹), which is a correct reflection of the carbon mesoporous channel system with a sharp capillary condensation stage at intermediate relative pressure (0.45 – 0.9 P/P₀). The position of the capillary condensation step corresponds to the primary size of mesopores (about 4 – 5 nm). The investigated carbon sample exhibits also a high specific surface area (757 m²·g^-1) and a significant pore volume (0.87 cm³·g^-1).

The critical steps during nitrogen adsorption/desorption measurements are the very precise mass sample determination step and the sufficiently degas step. The measurement procedure was performed according to the appropriate guidelines. Despite the fact that the determination of the surface area and the pore size distribution is based on the physisorption measurements, the interpretation of the experimental isotherms is not always direct.

During the computation of the mesopores' size distribution, by application of the modified Kelvin equation (which is the basis of the theory), it is necessary to accept the assumption of the rigid pores of well-defined shape. Moreover, the range of validity of the Kelvin equation and the interpretation of hysteresis loops on the isotherms remain unresolved problems. The possibility of facilitation is related to the application of empirical methods of isotherm analysis (e.g.,the α_s-method^{29,30,31,32,33,34}). However, this manner requires the use of adsorption data obtained with nonporous reference materials.

Figure 1B, D presents the mesopores' size distributions of studied samples (PSD). The assumption of the BJH theory permitted the determination of the mesopores' size distribution. The PSD curve presented in Figure 1B suggests that the carbon sample contained mesopores of the sizes of 3.5 – 4 nm, whereas SBA-15 possessed pores with a diameter of about 5 nm (see Figure 1D). Interestingly, for the OMC sample, the maxima of the pore size distribution curves calculated from adsorption and desorption data are very similar. Such as accurate agreement between the results, in view of different mechanisms of adsorption and desorption phenomena, evidence a very ordered and uniform nature of the material. The textural properties determined from experimental isotherms are summarized in Table 1.

The properties of the pore network structure based on the physical adsorption/desorption analyses are fundamental to the characterization of nanopowders and nanomaterials. Nitrogen adsorption/desorption methods can be regarded as the first stage in the characterization of microporous and mesoporous solids. The method is, in general, applicable to samples of all classes of porous bodies or materials and allows an estimation of the porous structures on the basis of the shapes of isotherms and hysteresis loops directly from the experimental measurement. The nitrogen adsorption/desorption next to the other methods of porous structure determination (liquid intrusion³⁵ and light, X-ray, and neutron scattering^36,37 and microscopy³²) is the most important and useful technique due to the wide applicability and mutual comparability of the results.

Nitrogen is a standard adsorptive gas molecule for pore characterization by gas adsorption method. It is possible to use other types of molecules (carbon dioxide, krypton, argon) for obtaining new information about the sample and for the characterization of microporous materials.

The TEM micrographs of investigated OMC and SBA-15 samples are presented in Figure 2. They confirmed a highly ordered system of mesopores, comprising of parallel and very similar porous channels and a carbon rod-shaped structure with the pore size ~11 nm (Figure 2A-C). TEM micrographs of OMC confirmed a hexagonal symmetry of the pores as a result of the inverse replica of the initial nanoporous silica precursor and a pore diameter between the rods of 4.0 nm. The SBA-15 sample shows the two-dimensional ordered hexagonal structure of the mesopores; the well-ordered structure of SBA-15 is observed in both projections, along with the nanoporous channels and perpendicular to them (Figure 2D, E). These observations are in good agreement with the values obtained from the experimental nitrogen sorption data (see Table 1). Transmission electron microscopy is able to provide images at a higher resolution than light microscopy due to the smaller de Broglie wavelength of the electrons. This enables us to capture the details thousands of times smaller than a resolvable object seen in a light microscopy. In this method, the image comes from the interaction of the electrons with the sample when the beam is transmitted throughout the specimen. Therefore, one of the limitations of the method is that the specimen should be an ultrathin film of less than 100 nm thick or a suspension applied on a grid. TEM can be improved on by a scanning transmission electron microscope (STEM). It should be possible by an addition of a system combined with suitable detectors, which will raster the beam across the sample to form the image.

The pH of the point-of-zero charge of OMC (i.e., the pH value below which the total surface of the carbon particles is positively charged) was found to be above pH = 10 (Figure 3). For comparison, the pH of the point-of-zero charge of the SBA-15 material is equal to about 4.5. Potentiometric titration results of SBA-15 material show a shift of pH_pzc toward lower pH values, evidencing of the existence of some acid centers on the SBA-15 surface. The negatively charged sites increase the Van der Waals interactions between the adsorbent/adsorbate molecules in the SBA-15 matrix, improving the adsorptive properties of the silica matrix.

The critical steps during potentiometric titration measurement include the very precise addition of the titrant to the suspension and continuity of stirring. The potentiometric titration procedure was fully automated to ensure the most reliable results. Another important and unique step was the application of special software for controlling experimental conditions and calculations. The limitation of this method is the calibration step of the pH electrode and the need to ensure a stable atmosphere (for example, nitrogen) and temperature. Potentiometric titration can be categorized as an acid/base titration procedure. This technique requires measurements of the voltage change upon titrant addition steps. It provides an adaptable, affordable, and highly accurate technique to achieve high purity, which is essential in many fields, particularly in pharmaceuticals and functional material studies. In fact, there are a number of types of potentiometric titrations. There are, for instance, acid-base, redox, precipitation, and complexometric techniques. As potentiometric titrations can be realized automatically, it ensures greater capacity for sample characterization.

The pzc is determined as the intersection of the electrolyte titration curve and the titration curve recorded for carbon material, as well as the intersection of the surface charge density curve with the axis. The result suggests that investigated, ordered nanoporous carbon material has basic properties related to the presence of oxygen functional groups (-OH terminal groups of ketonic, pyronic ring, chromenic, and π-electron systems)^38,39,40,41. The basic properties of the carbon surface can be a result of a thermal treatment that causes progressive destruction and removal of acidic oxygen functional groups (for example, carboxyls, lactones, and phenols) and, thus, enriches the carbon surface in basic functional groups⁴². The presence of these basic sites is a consequence of the simultaneous decrease in the number of acidic sites and a reduction of oxygen content associated directly with the acidity. The content of oxygen was verified by the TEM-EDS.

TEM-EDS spectra images of the OMC surface from two different areas of the sample are displayed in Figure 4A, B. Oxygen and silicon atoms from the OMC surface were detected despite the predominant amount of carbon. The atomic and weight percentage of the elements, presented as insets and obtained from different areas of the samples, are similar and indicate about 98% is composed of carbon and only slightly more than 1% of the composition is attributable to oxygen. EDS microanalysis may suggest that the basic character of the OMC surface is associated with very low number of oxygen-containing functional groups, which typically has acidic functionality. Moreover, the basic functional groups can be responsible for the growing of hydrophilicity of the carbon materials. An EDS spectrum of silica matrix confirms the main contribution of oxygen and silicon abundance in the SBA-15 (Figure 4C). EDX is one of the techniques that determine the atomic composition of the specimen. It does not give chemical information (e.g., oxidation state, chemical bonds), like an XPS method. For quantitative analysis, EDX is not suitable for light elements (e.g.,like oxygen) since it can only detect the presence of oxygen but cannot quantitate it. This method works only on the surface of thin layers (of a few microns or less) and is quite sensitive to the contamination in the specimen.

In Figure 5 A, B the results of contact angle measurements for several liquids on silica and carbon substrates are presented; we can see that the systems reveal a wide range of wettabilities. Two of the applied substrates are smooth and planar, namely the silica and carbon surfaces, while the others show roughness and contain mesopores. The measured contact angles are presented vs.the microscopic wetting parameter, α_w. This parameter is the ratio between the liquid-pore wall intermolecular interactions to the interactions of two liquid molecules^22,23,24. Therefore, it measures the wetting properties at the nano- and macroscales. The α_w parameter is shown to be a monotonic function of the contact angle. According to the measurements, the values of contact angles for the surfaces with roughnessare higher than those for the smooth planar surfaces, irrespective of the type of studied liquids, including nonwetting andwell-wetting liquids.

These results suggest the Cassie-Baxter mechanism of wettability on nanorough surfaces of porous walls. Moreover, the contact angles measured for several liquids inside silica and carbon nanopores indicate better wettability of the silica walls than the carbon walls, and the influence of pore roughness on the fluid/wall interactions is more pronounced for silica than for carbon nanopores. The capillary rise method is used to determine the contact angles and wettability of small particles in the powders. For a flat solid substrate, many techniques, such as sessile drop method and Wilhelmy plate method, can be applied to contact angle measurements. The use of capillary rise method assumes the satisfaction of four conditions during the process (i.e., Washburn's equation is derived based on these four assumptions): (1) a constant laminar flow, (2) the absence of an external pressure, (3) negligible gravitational force, and (4) the liquid at the solid-liquid interface does not move. The hydrostatic pressure is much smaller than the capillary pressure; therefore, the capillary pressure causes the liquid to rise upward along the tube. The wettability studies of small particles should always take into account the accuracy and reproducibility of the results.

_Tomeasure accurately the contact angle of small particles, the pressure increments and the hydrostatic effects should be taken into account in Washburn's equation to more accurately describe the relationship between the pressure increment and time.

In Figure 6A, B, we present the fraction f (describing the part of a porous surface which is in direct contact with the liquid interface), defined as f = (1 + cosθ_p) / (1 + cosθ). It takes values from the range of 0.73 (H₂O) to 0.92 (OMCTS) for silica matrix and from the range of 0.82 (H₂O) to 0.93 (OMCTS) for carbon matrix. Moreover, the fraction f increases monotonically with the increasing α_w parameter, which confirms the Cassie-Baxter model of wetting on rough substrates.

These results, discussed within the framework of the Cassie-Baxter model, show that for nanosurfaces, the influence of microroughness significantly affects changes in the fluid-wall interactions.

To investigate the influence of the character of the porous surfaces on the confinement effects of D₂O in SBA-15 and OMC matrices with a comparable pore size of 5 nm, the dielectric method was used. The results of the electric capacity of water placed in OMC and SBA-15 upon heating, as presented in Figure 7A and Figure 7B, respectively, indicate that the temperature dependence of capacitance C shows a sharp increase at T = 260 K, corresponding to the melting of adsorbed D₂O inside the pores of SBA-15, and at T = 246.1 K, which refers to the melting of water adsorbed inside the pores of OMC. For both systems, we observe an increase in the C(T) function at T = 276.5 K, which is referred to the melting point of the bulk deuterated water. The observed signals are related to both the bulk and the confined liquid because the samples are studied as suspension in filled porous matrices with the liquid being in excess. Relative to that of the bulk, the melting temperature of D₂O in SBA-15 pores is depressed by ΔT = T_m,pore− T_m,bulk =-16.5 K, while for OMC, ΔT = -30.4 K.

These results come from various structures of the pore wall. The melting temperature in the pores T_m,pore is dependent on two variables, namely the pore size H and the wettability parameter α_w. For smaller α_w values (α_w < 1), the depression of T_m,pore is expected. If the pore width H is the same, then the change in the α_w value of the system affects the change in T_m,pore. The results of the wettability in pores show that the α_w value in both systems studied decreases due to the roughness effects relative to those of the smooth surface. The work of immersed wettability in pores W_e = γ_lcosθ_p is much less for D₂O in OMC (W_e = 4,2432 [mN/m]) than in SBA-15 (W_e = 20,968 [mN/m]), which leads to a high depression of the melting point in this system, relative to that for D₂O in the SBA-15 system.

The obtained results indicate an improvement in the adhesion effects on the porous silica wall, relative to those on the carbon wall. This method has some limitations when applied to the studied samples. One of them comes from the fact that the record of the C(T) function contains the signal from both the bulk liquid and the liquid confined in the pores. Therefore, the signal from the confined nonpolar liquid is weak, and it is difficult to determine the melting temperature. In the future, it is worth combining some additional methods along with it. (e.g., differential scanning calorimetry on a slow heating rate or centrifuging the sample to get concentrated sample without any liquid). Moreover, for the conductive samples, there is a need to use a polytetrafluoroethylene plate. An advantage of dielectric spectroscopy method is the fact that the method is used a lot, in many areas of research, such as glass transitions and time-scale molecular motions, where the time length is in the tens of femtoseconds to nanoseconds. Therefore, it is important to provide frequencies ranging from megahertz to terahertz. In cases such as decomposition of the obtained spectra or the interpretation and quantitative analysis of the results, the technique resembles more common spectroscopies. However, the dielectric method investigates the collective fluctuations of molecules with a permanent dipole moment (polar liquids). For example, infrared spectroscopy provides additional complementary information, although the sensitivity in the direction of collective modes may impede the interpretation of dielectric spectra.