1. Introduction
Proton exchange membrane fuel cells (PEMFCs) represent a significant advancement in hydrogen technology. These fuel cells are characterized by high energy efficiency, easy scalability, long operating times, and low (or zero) emissions. Fuel cell technology has applications in vehicles, stationary power generation, and portable devices [
1].
One key material in PEMFCs is Nafion, a sulfonated fluoropolymer that serves as a state-of-the-art solid electrolyte. Nafion demonstrates an excellent proton conductivity of 10
−1–10
–2 S/cm at temperatures between 333 and 353 K in high humidity conditions [
2,
3]. However, its high production costs and complex synthesis processes have hindered its large-scale application [
4]. Furthermore, Nafion’s conductivity dramatically decreases when temperatures exceed 353 K or drop below 268 K [
3]. This decline is due to the collapse of the proton transport pathway at high temperatures and weak proton transfer in freezing environments [
5]. Most solid electrolytes, including Nafion, suffer from a narrow operating temperature range and a heavy dependence on water for proton conduction, a characteristic explained by the Grotthuss mechanism [
4]. Considering the necessity for PEMFCs to perform in diverse and sometimes extreme conditions, there is an urgent need for the development of stable electrolytes that can deliver high proton conductivity without reliance on water over a broad temperature range from subzero levels to above 373 K [
3].
Water molecules are not the only molecules that can serve as proton carriers, as is the case with Nafion. Another possibility, for example, is to replace water with azoles, which can increase the maximum operating temperature of the fuel cell [
6]. Among the studied alternative solid electrolytes, a very interesting candidate seems to be an inorganic–organic composite based on introduced imidazole (Im). Imidazole is a polar and amphoteric compound with low volatility (boiling point at ca. 523 K). In its solid crystalline form, imidazole exhibits low proton conductivity, which significantly increases upon its melting due to enhanced molecular mobility. Importantly, this increased mobility can also be achieved by dispersing imidazole within porous matrices, where its conductivity is substantially boosted [
7]. So far, numerous materials, including MOFs [
8], mesoporous solids [
9], zeolites [
10], or polymers [
11], have been reported as matrices of imidazole-containing composites, and the number of publications on this subject continues to grow. It is worth noting, however, that among the studied proton conductors, materials containing water molecules still dominate, and the published results may inspire further research on anhydrous materials.
Studies conducted to date suggest that the conductivity of electrolytes, both hydrated and anhydrous, can be influenced by both chemical and structural properties of the porous material used. Among others, electrolytes based on porous silicates (including mesoporous silicas) [
12,
13] and aluminosilicates (e.g., zeolites) [
14,
15] have been studied. Materials with different pore arrangements (2D, 3D) [
16,
17,
18] and different pore sizes (micropores, mesopores) [
19] have been used. The introduction of a significant amount of proton carriers and the maintenance of their mobility may be facilitated by the large specific surface area and pore volume of the porous matrix. In this context, therefore, ordered mesoporous silicas seem attractive [
20,
21].
On the other hand, other studies have shown that the conductivity of electrolytes, e.g., those with imidazole, can be influenced by the parameters of the porous matrix, such as the number, type, and location of different types of active centers [
22,
23,
24]. Mesoporous silicas usually have a significantly lower content of such centers than zeolites [
25], but additional centers can be generated by the appropriate modification of the chemical composition of the silica either during its synthesis or as part of post-synthesis modification [
20]. One of the most commonly used methods of increasing the acidity of mesoporous silicas is modification with sulfonic acid groups. The introduction of -SO
3H generates additional protons, which can significantly increase the proton conductivity of electrolytes [
13].
Another method of generating acidity is the modification of a mesoporous silica matrix with heteroatoms, including aluminum or cerium atoms. Depending on the method used to introduce cerium atoms, porous materials can vary in the type and number of active centers, among other factors [
26,
27]. Cerium atoms can be incorporated into both microporous materials (such as zeolite Y (FAU) [
28,
29], MOR [
27], SAPO [
26], BEA [
30], MFI [
30], FER [
30], MCM-22 [
31]) and mesoporous materials (e.g., MCM-41 [
32,
33], SBA-15 [
34], MCM-48 [
35], KIT-6 [
36,
37]). For microporous materials, ion exchange [
28,
31,
38] and impregnation [
26,
39] are frequently the preferred modification methods. In contrast, for mesoporous materials, cerium is frequently introduced during one-pot synthesis [
40,
41] and via impregnation [
32,
37]. During ion exchange, cerium atoms can often be introduced as cations located in extra-framework positions [
38], while incorporating cerium precursors into the reaction gel during mesoporous material synthesis can lead to their incorporation into the silica framework [
34,
41]. Conversely, when cerium atoms are introduced via impregnation, their further calcination in air can result in its deposition as an oxide phase (e.g., CeO
2) on a silica surface [
37,
42]. It is also well known that during the thermal treatment of ion-exchanged cerium zeolites, additional Brønsted acid centers may be formed due to the decomposition of hydrated cerium species that increase the activity of the zeolites [
38]. Therefore, the type of modification method chosen makes it possible to obtain materials with the desired properties [
37,
42]. Materials modified with cerium atoms have been used as catalysts [
32,
36] and adsorbents [
33,
43], among other applications.
In our research, KIT-6 mesoporous materials, characterized by a 3D structure with interconnected channels, a very large specific surface area (approximately 800 m
2/g), and significant pore volume (up to 1.05 cm
3/g) [
44], were chosen as components for inorganic–organic composites with imidazole. Catalytic results from KIT-6 modified with cerium atoms incorporated into the mesoporous network confirm the presence of acidic centers in this system [
36]. Therefore, this highly porous 3D system, with a potentially significant number of acidic sites, appears to be a promising candidate for proton conductor applications.
This study focuses on anhydrous proton-conducting materials derived from KIT-6 mesoporous materials and azoles, in which imidazole molecules play the role of the main proton carriers. The aim of the research was to investigate the effect of the acidity of KIT-6 materials on the conductivity of the obtained azole composites. The acid centers in KIT-6 were generated using two modification methods: (1) the introduction of aluminum into the KIT-6 framework at the synthesis stage of this material via the isomorphous substitution of silicon atoms with aluminum atoms; and (2) the introduction of cerium species as part of the post-synthesis modification of KIT-6 (Scheme S1). Both methods were expected to lead to an increase in the number of Brønsted acid centers and consequently an increase in the H+ content in the tested proton conductors. In the case of Al introduction into the framework, protons from the Brønsted acid centers would compensate for the negative lattice charge of KIT-6 generated by the insertion of aluminum atoms into the normally pure silica KIT-6, and the number of these centers could depend on the amount of aluminum in the material. This modification allowed us to achieve porous matrices with varying Si/Al ratios. Additionally, pure silica KIT-6 was also synthesized. Some of the resulting hydrogen forms of KIT-6 were then modified with cerium ions to generate additional acidic centers. It was assumed that Brønsted acid centers in Ce-KIT-6 materials could be formed as a result of the transformation of cerium species introduced into H-KIT-6 by ion exchange.
Both the cerium-modified and -unmodified (hydrogen) forms of KIT-6 materials were subjected to comprehensive physicochemical characterization, including X-ray diffraction (PXRD), transmission electron microscopy (TEM), Fourier-transform infrared spectroscopy (FTIR), UV-visible spectroscopy (UV-Vis), X-ray photoelectron spectroscopy (XPS), and low-temperature nitrogen adsorption. Furthermore, the proton conductivity of the materials with incorporated imidazole was evaluated using impedance spectroscopy (EIS). KIT-6 materials were found to be highly effective porous matrices for imidazole incorporation. The conductivity of the resulting composites ranged from 10−4 to 10−3 S/cm, placing them among the high-performing anhydrous electrolytes containing imidazole.
The studies presented here are an attempt to find a correlation between the acidity of the porous host and the conductivity of anhydrous proton conductors. Although the idea of introducing cerium atoms, e.g., into heterogeneous catalysts, has been known for many years, the attempt to create Bronsted acid centers by introducing cerium species to generate additional H
+ in the proton conductor can definitely be considered as an unconventional approach to such studies. In the case of the presented studies, it should also be emphasized that the introduction of imidazole into silica-based mesoporous materials is still not very common in the literature. For example, the use of KIT-6 materials in proton conductors has so far been mainly limited to the use of materials with sulphonic acid groups, and in all of them, to the best of our knowledge, water molecules have played the role of the main proton carriers [
16]. Therefore, studying anhydrous systems containing Ce-modified KIT-6 in terms of proton conductivity can provide valuable information about these relatively unexplored materials.
3. Conclusions
The synthesis of mesoporous molecular sieves with a KIT-6 structure and varying aluminum content demonstrated that the incorporation of aluminum into silica structures generates additional acidic sites, with their number increasing as the aluminum content rises. Additionally, the introduction of cerium cations significantly boosted the number of acidic sites. The catalytic activity of the matrices, confirmed by the esterification reaction of acetic acid with methanol, indicated the presence of Brønsted acidic sites, which are related to the presence of cerium atoms. UV-Vis and XPS analyses verified the presence of Ce3+ and Ce4+ in the KIT-6 materials, and the results suggest that high-temperature treatment after cerium introduction may lead to the partial incorporation of cerium into the framework.
EIS studies showed that dispersing imidazole in the KIT-6 matrices resulted in highly conductive composites, demonstrating that H+ transport in these composites is facilitated by imidazole molecules. All composites, both in their hydrogen and cerium forms, exhibited high conductivity within a wide temperature range (300–393 K). Measurements over several heating and cooling cycles confirmed that the samples maintained high and stable conductivity in subsequent cycles.
The cerium-modified KIT-6 materials provided higher conductivity in composites than their hydrogen counterparts, indicating that cerium incorporation positively impacts composite conductivity by increasing the number of protons. At 393 K, the highest conductivity under anhydrous conditions (1.13 × 10−3 S/cm) was recorded for the Ce-KIT-6 (100) composite, with slightly lower values (1.02 × 10−3 S/cm) observed for Ce-KIT-6 (50) and Ce-KIT-6 (200) composites. Remarkably, their conductivity was even higher than that of molten imidazole.
The presence of weak acidic sites, particularly Brønsted sites, positively affects proton conduction. Using matrices containing such sites leads to composites with high conductivity. Controlling the number, strength, and nature of these sites enables the creation of composites that can effectively compete with other nanoporous matrix-based electrolytes containing imidazole.
4. Materials and Methods
4.1. The Synthesis of KIT-6 Materials
Siliceous and aluminosilicate KIT-6 materials were synthesized by the assembly of polymer surfactant micelles as template under mild acidic conditions. A typical procedure for the synthesis was as follows: 4 g of Pluronic P123 (Sigma Aldrich, Poznań, Poland) was dissolved in 114 mL H
2O and 39.2 mL 2 M HCl (Stanlab, Lublin, Poland). The mixture was stirred at 313 K until a clear solution was obtained (about 4 h). Then, 6.22 mL n-butanol (Eurochem BGD, Tarnów, Poland) was added and the stirring was continued for 1 h. After this time, 12.84 g of TEOS (Sigma Aldrich) and appropriate amounts of aluminum isopropoxide (Al(isop)
3, Acros), given in
Table 7, were added, and stirring was continued for 20 h at 313 K. Subsequently, the mixture was aged at 368 K for 24 h under static hydrothermal condition. The precipitated material was filtered, washed, and calcined in air at 823 K for 8 h to remove the template. The final materials were marked as H-KIT-6 (x) where x denotes Si/Al molar ratio of the gel.
4.2. The Preparation of Ce-KIT-6
For all H-KIT-6 materials, regardless of the Si/Al ratio, the same procedure for introducing cerium ions was used. A quantity of 5× g of the material was immersed in 50 mL of 0.1 M CeCl3 solution and left at room temperature for 2 h. After this time, the solution was centrifuged. The remaining sediment was immersed again in 50 mL of 0.1M CeCl3 solution and left for another 2 h. The sediment was centrifuged again and immersed in a new portion of the cerium chloride solution and then it was left for 20 h. After this time, the sediment was washed with 400 mL of H2O, dried, and then calcined for 2 h at 673 K. The procedure described above was performed three times. The resulting cerium form was labeled as Ce-KIT-6.
4.3. The Preparation of KIT-6/Imidazole Composites
Imidazole was introduced into 1 g of the matrix by the impregnation method. Before impregnation, the matrices were calcined at 623 K to remove water. Imidazole solutions in chloroform (3.4 mL) were added to the activated KIT-6 materials. The weights and mass fractions are listed in
Table 8. The same weights were used for both the hydrogen and cerium forms. The mixture was stirred on a magnetic stirrer at room temperature for 24 h in closed vials. After this time, the solvent was evaporated at room temperature. The mass fraction of imidazole in the composite was labeled as x Im.
4.4. Characterization
PXRD measurements were performed using powder X-ray diffraction on a BRUKER D8 ADVANCE diffractometer (Bruker, Billerica, MA, USA) equipped with a Cu lamp (Cu Kα1) emitting radiation with a wavelength of λ = 0.15406 nm. The analysis was conducted in the small-angle (0.6–8°) and the wide-angle (6–60°) ranges.
N2 adsorption/desorption studies were conducted using a Quantachrome NOVA 1000 instrument (Anton Paar, Warsaw, Poland). Prior to measurement, the samples were degassed at a temperature of 393 K for 16 h. The surface area was calculated using the Brunauer–Emmet–Teller (BET) method. The total pore volume and the average pore diameter were determined by the Barrett–Joyner–Halenda (BJH) method.
The FTIR spectra were recorded using a Bruker Tensor 27 spectrophotometer (Bruker Polska, Poznan, Poland). Measurements were performed using the transmission technique in the wavenumber range of 4000–400 cm−1 with a resolution of 1 cm−1. The sample was mixed with 200 mg of KBr and formed into pellets using a press (150 MPa).
The temperature-programmed desorption of ammonia (TPD-NH3) was conducted using a PulseChemiSorb 2705 apparatus (Micromeritics, Norcross, GA, USA) with a flow system. The samples were pretreated in helium at a temperature of 823 K for 0.5 h. Gaseous ammonia was adsorbed at 393 K. The physically adsorbed NH3 was removed by purging with helium flow at 393 K for 1 h. Then, thermal desorption was carried out up to 923 K with a heating rate of 10 K/min. The amount of desorbed ammonia was measured using a TCD detector and recalculated per 1 g of sample.
The esterification reaction of acetic acid with methanol was performed in sealed glass vials. For this reaction, 0.05 g of catalyst, which had been pre-activated in an oven at 623 K for 1 h to remove water from the catalyst’s pores, was used. To the weighed catalyst, 1.501 g of glacial acetic acid (Carlo Erb, Cornaredo, Italy) and 1.602 g of methanol (Eurochem) were added, maintaining a molar ratio of acetic acid to methanol of 1:2. The reaction was conducted in a thermostatic bath with continuous stirring at 343 K for 4 h. The activity of the catalyst was estimated by titrating the post-reaction mixtures with 0.1 M sodium hydroxide solution to calculate the amount of unreacted acetic acid. Prior to analysis, the obtained mixtures were centrifuged to separate the catalyst from the reagent mixture. Subsequently, 0.5 cm
3 of the resulting solution was transferred to a volumetric flask and diluted with 10 cm
3 of water. Phenolphthalein was used as the indicator for the alkalimetric titration. The conversion of acetic acid
CHAc (%) was calculated using the following equation:
where
V0 and
V are the volumes of NaOH solutions used for titration of the solution of substrates mixture before and after the reaction (mL), respectively.
The UV-Vis studies were conducted using the reflectance method on a Jasco V670 apparatus (ABL&E-JASCO Polska, Krakow, Poland). BaSO4 was used as an internal standard for recording UV-Vis DRS spectra. Measurements were performed at room temperature in the range 190–900 nm.
Transmission electron microscopy (TEM) images were recorded on a JEOL 2000 microscope (JEOL, Tokyo, Japan) operating at an accelerating voltage of 80 kV.
X-ray photoelectron spectroscopy (XPS) measurements were conducted using an Ultra High Vacuum (UHV) System (Specs, Berlin, Germany). The examined materials were irradiated with a monochromatic Al Kα radiation (1486.6 eV). The operating pressure in the chamber was close to 2 × 10−9 mbar. Binding energies (BE) were calibrated against the C1s peak originating from the carbon surface layer, set at 284.6 eV. Spectroscopic data were processed through CasaXPS software (version 2.3.22PR1.0) developed by Casa Software Ltd., Teignmouth, UK, employing a peak-fitting algorithm with a linear background correction.
The proton conductivity of the obtained composites was measured using impedance spectroscopy (EIS) in the temperature range of 300–393 K. Powder samples were pressed in a Teflon vessel between electrodes under a pressure of 200 MPa. The measurements were carried out in a nitrogen atmosphere using a Hewlett Packard 4284A precision LCR meter (20 Hz–1 MHz) (Keysight Technologies UK, Wokingham, UK). The measurement temperature was controlled and stabilized with a LakeShore 340 temperature controller (Lake Shore Cryotronics, Inc., Westerville, OH, USA).