3.1 Physical properties of the ASA
The gelation time, bulk density, and pore structure of aerogels prepared by altering the aluminum sol/TEOS molar ratio are presented in Table 1. It can be seen that with an increase in Al/Si molar ratio from 0 to 0.4, the gelation time increases from 46 min to 194 min while the bulk density of the ASA increases from 0.119 g·cm-3 to 0.187 g·cm-3. The significant increase in gelation time is caused by two reasons. Firstly, the aluminum sol used in this work is acidic and its pH value is 4.3. For the hydrolyzed TEOS, the more aluminum sol is used the lower the pH value of the mixture, which is unfavorable to the dehydration condensation reaction of the silicate monomer hydrolysis products and consequently prolongs gelation time. Secondly, for more aluminum sols, the spacing between the active particles after hydrolysis is larger which also leads to an increase in the time to form the gel network. Simultaneously, the bulk density of ASA shows an increasing trend with the increase in Al/Si molar ratio, and this result obtained was consistent with some previous reports on the preparation of aerogels with the addition of inorganic aluminum sources[27]. It was also found that the flowability of the mixed sol decreases when the Al/Si molar ratio exceeds 0.3, and the mixture almost lost its flowability at the Al/Si molar ratio of 0.4, which is unfavorable to the gel reaction.
Al content significantly impacts the pore structure of ASA. The specific surface area, pore volume, and average pore diameter of ASA-0 are 508.266 m2∙g−1, 3.058 cm3∙g−1, and 18.752nm, respectively. As the Al/Si molar ratio escalates from 0 to 0.4, the average pore diameter of ASA decreases first and then increases. This trend indicates that the appropriate amount of aluminum sol can play a role in refining the pore size. Contrarily, the specific area and pore volume display an inverse pattern when the aluminum sol content increases. This phenomenon might stem from the fact that the volume of the composite precursor is not greatly affected by the lesser addition of aluminum sol. Still, the electrostatic attraction between the aluminum sol and particles is intense, thereby shrinking the pore size and enhancing the specific area. However, an increase in the aluminum sol additive amount introduces a complication. An excessive volume of water filling the gel pores becomes detrimental to the solvent replacement and modification process of the wet gel. The capillary forces are amplified by the moisture present within the pores during the APD process, leading to an increase in gel volume shrinkage. When the capillary forces are excessive, they can even trigger a structural collapse, resulting in a deterioration of the pore structure.
Table 1 Density and pore structure characteristics of Al2O3-SiO2 aerogels with different Al to Si molar ratios.
Sample
|
Gelation time
(min)
|
Bulk density(g·cm-3)
|
Specific surface area(m2∙g−1)
|
Pore volume (cm3∙g−1)
|
Average pore diameter
(nm)
|
ASA-0
|
46
|
0.119
|
508.266
|
3.058
|
18.752
|
ASA-0.05
|
63
|
0.124
|
544.149
|
3.122
|
14.624
|
ASA-0.1
|
84
|
0.130
|
571.661
|
3.083
|
13.932
|
ASA-0.2
|
114
|
0.129
|
616.317
|
3.272
|
13.762
|
ASA-0.3
|
147
|
0.141
|
554.842
|
2.996
|
14.170
|
ASA-0.35
|
174
|
0.162
|
527.462
|
2.649
|
14.112
|
ASA-0.4
|
194
|
0.187
|
526.311
|
2.348
|
14.226
|
The N2 adsorption-desorption isotherms and pore diameter distributions of ASAs are shown in Fig. 1. The adsorption-desorption isothermals exhibited in Fig. 1(a) are classified as type IV adsorption curves according to the IUPAC classification, indicating that ASAs are typical mesoporous materials[28]. The curves rise slowly at the relative pressure 0.1-0.8, and the N2 adsorption is mainly monolayer adsorption; Under high pressure (relative pressure at 0.8-1), the adsorption capacity increases rapidly, and the adsorbed gas coalesces into liquid in the pore resulting in capillary condensation; Due to the incomplete desorption, the residual adsorption capacity in the desorption curves is greater than the adsorption capacity at the same pressure, and the desorption curves lag behind the adsorption curves. The adsorption-desorption curve shows an H3 hysteresis loop, indicating that the porous structure is composed of silt-type mesopore pores[29]. Moreover, it can be seen that the ASAs have relatively wide pore size distributions from Fig. 1(b), mainly in the range of 20nm to 50nm.
3.2 Morphology, crystalline and chemical structure of the ASAs
We performed TEM, EDS, XRD, and FTIR analysis of ASAs to further explore the effect of aluminum sol doping on the samples. APD samples are monolithic as shown in Fig. 2(a). Fig. 2(b-d) presents the TEM images of samples for ASA-0, ASA-0.2 and ASA-0.35. It can be seen that the aerogels possess a typical nano-porous morphology, the size of the continuous network skeleton particles is uniform and the pearl necklace aerogel particles constitute a sophisticated three-dimensional network structure. The ASA-0 is characterized by secondary particles that form coarse spheres, where the connections between these large round particles are notably weak. However, upon the introduction of an aluminum sol, the secondary particles of ASA-0.2 shrink in size, leading to tighter inter-particle connections and increased cross-linking. This alteration enhances the skeletal strength of the aerogel, thus improving its overall structural integrity. It is noteworthy that the pore distribution of ASA-0 and ASA-0.2 is relatively uniform, and the nanopores remain intact. With an increase of aluminum sol, the 3D network skeleton of ASA-0.35 becomes loose and the pores increase obviously. The significant change in pore structure is due to two reasons. Firstly, the hydrolyzed TEOS is known to condense at a pH above the isoelectric point of silicon dioxide under normal conditions. The addition of excessive aluminum sol makes the mixture lower pH value which is unfavorable to the condensation reaction. Secondly, excessive aluminum sol makes the silicon content per unit volume decrease and the spacing of Si-OH in the condensation process increase, leading to increase porosity of the aerogel skeleton. The distributions of Si, O, and Al elements in ASA-0.2 and ASA-0.35 were characterized by EDS, and the results are shown in Fig. 3. The spectrums present that Si, O, and Al elements are uniformly distributed in the aerogel particles. It is worth noting that the Al/Si molar ratios of samples ASA-0.2 and ASA-0.35 are 0.12 and 0.21, respectively. These ratios are much lower than the theoretical values, indicating that the loss of Al occurred in the process of solvent exchange.
Fig. 4(a)displays the XRD presents of ASAs with varying aluminum sol contents. The prepared ASA-0 does not show any distinct crystal structure diffraction peak. Instead, a diffuse diffraction peak package emerges between the corresponding crystalline SiO2 characteristic peak 2θ=20-25°, suggesting that the fabricated SiO2 aerogel is amorphous at room temperature. Although the crystal structure of ASAs does not undergo significant alterations after the aluminum sol addition, the diffraction peaks start appearing as the content increases. The ASA-0.35 sample reveals diffraction peaks at 14.5, 27.8, 38.4, 49.2, and 64.9°, corresponding to the boehmite crystal plane diffraction peaks of (020), (120), (031), (200), and (002), respectively. These are undesirable crystals formed due to their excess water content[30,31], implying a threshold exists for the amount of aluminum sol that can be added, and any excess cannot adequately react with the silicon source.
The acquired FTIR spectra of ASAs are shown in Fig. 4(b). The peaks at 1096 cm-1 and 468 cm-1 originate from the antisymmetric stretching vibration and bending vibration of the Si-O-Si bond, respectively[32]. Absorption peaks near 3437 cm-1 and 1632 cm-1 correspond to the -OH stretching vibration and the H-O-H bending vibration in H2O[33], while The peak at 957 cm-1 is due to the Si-OH stretching [34]. The peaks at 2964 cm-1, 1257 cm-1 and 847 cm-1 are attributable to the Si-CH3 vibration[35]. The hydrophobic modification process eliminates the hydrophilic -OH group from the wet gel skeleton, replacing it with the hydrophobic -CH3. Compared to the spectrum of ASA-0, a new peak emerges at 610cm-1 with the addition of aluminum sol. This peak corresponds to the bending vibration of the Al3+ octahedral structure [36], indicating that not all the Al elements are incorporated into the aerogel skeleton after the aluminum sol addition. Instead, they accumulate in another way, which is consistent with the XRD test results.
Drawing from the aforementioned analyses, we propose a growth mechanism for the composite gel. The precursor TEOS undergoes hydrolysis in an acidic environment to form Si-OH, which is then uniformly blended with the aluminum sol. The introduction of silica sol to the mixture elevates the pH which prompts the gel reaction. The nano-SiO2 particles present in silica sol comprise disordered silica-oxygen tetrahedra, with the internal structure of the particles linked by Si-O-Si bonds. Meanwhile, Si atoms on the surface are connected with numerous -OH groups. The nanoparticles in the aluminum sol exhibit a double-layer structure. The inner layer consists of Al-O bonds, while the surface hydroxyl groups form hydrogen bonds to connect the interlayer. Both the nanoparticles in silica and aluminum sols can serve as nucleation sites. Simultaneously, the hydrolysis product Si-OH reacts with these exposed nucleation sites on the surface and undergoes self-polymerization, as well as mutual polymerization. This sequence of events culminates in the formation of a skeleton possessing a 3D structure.
3.3 Thermal stability of ASAs
Table 2 presents the specific surface areas of the ASAs after being calcined at 600, 800, and 1000 ℃ for 2h, respectively. As the calcination temperature rises, the specific surface area of ASA-0 decreases significantly, with a retention rate is merely 4.7% at 1000 ℃. An increase in calcination temperature after alumina sol addition similarly reduces the specific surface area of aerogel. However, the specific surface area remains significantly larger than that of ASA-0 at the same temperature. Notably, ASA-0.1, ASA-0.3, and ASA-0.5 show an increase in specific surface area at 600 ℃, while the other three samples remain almost unchanged. ASA-0.2 exhibits the highest specific surface area of 290.04 m2·g and a retention rate of 47.1% after being calcined at 1000 ℃ for 2h. As a result, the subsequent analysis will mainly focus on the ASA-0.2 sample because it exhibits the best thermal stability at 1000 ℃ among all the samples.
Sample
|
Specific surface area (25 ℃)/m2·g
|
Specific surface area (600 ℃)/m2·g
|
Specific surface area (800 ℃)/m2·g
|
Specific
surface area (1000 ℃)/m2·g
|
Retention rate of specific surface area (1000 ℃)/%
|
ASA-0
|
508.266
|
349.676
|
35.347
|
23.912
|
4.7
|
ASA-0.05
|
544.149
|
537.651
|
371.912
|
224.332
|
41.2
|
ASA-0.1
|
574.982
|
591.871
|
371.769
|
248.147
|
43.2
|
ASA-0.2
|
616.317
|
598.781
|
363.463
|
290.040
|
47.1
|
ASA-0.3
|
530.567
|
544.044
|
338.579
|
248.719
|
46.9
|
ASA-0.35
|
527.462
|
579.419
|
331.782
|
224.332
|
42.5
|
ASA-0.4
|
523.587
|
514.432
|
329.291
|
216.577
|
41.4
|
Table 2 The specific surface area of Al2O3-SiO2 aerogels with different Al/Si molar ratios at high temperature
Fig. 5(a)-(d) presents the N2 adsorption-desorption isotherms, pore diameter distribution curves, average pore diameters, and the pore volume curve of ASA-0.2 after being calcined at 1000 ℃ for 2h. All ASA-0.2 samples at different temperatures exhibit type Ⅳ isotherms with an H3 hysteresis loop, indicating that the porous is primarily composed of slit-type mesoporous. The N2 adsorption capacity of ASA-0.2 weakens as the calcination temperature increases and the hysteresis loop decreases significantly. In Fig. 5(b), the number of pores decreases after heat treatment, and the pore diameter tends to increase. This change is mainly attributed to the deformation of the aerogel skeleton under thermal stress, resulting in the formation of larger pores by interconnecting some of the smaller pores. However, even after treatment at 1000℃, the pore size of ASA-0.2 remains in the range of 20 to 50nm. This observation is consistent with Fig. 5(c), which shows that the average pore diameters of the samples range from 13 nm to 25 nm, indicating that ASA-0.2 can maintain a stable mesoporous structure up to 1000 ℃. Fig. 5(d) shows that the higher temperatures reduce the pore volume, but it still retains a relatively high-value 1.615cm3/g after calcination 1000 ℃.
Fig. 6(a-d) presents the TEM images of ASA-0.2 before and after calcination. At room temperature, the nano-secondary particles comprising the ASA-0.2 skeleton structure appear spherical, and they aggregate to form particle clusters, creating a loose 3D network structure with uniform pore size and distribution. Remarkably, the morphology of the sample has no obvious change after calcination at 600 and 1000 °C, and the skeleton structure remains similar to that before calcination. At 1000 ℃, a certain degree of sintering occurs in the sample, resulting in the transformation of the aerogel skeleton from loose to dense. The spherical particles are observed to stack closer together, forming larger particle clusters, but the 3D nano-porous structure of ASA-0.2 is preserved. In comparison, the TEM image of ASA-0 calcined at 1000 ℃ exhibits severe sintering, volume shrinkage, and destruction of the 3D network structure. These findings indicate that the addition of aluminum sol can enhance the high-temperature resistance of silica aerogel.
3.4 Thermal, crystalline, and chemical structure of ASA after calcination
To further investigate the impact of high temperatures on ASA, samples calcined at various temperatures were subjected to TG, XRD, and FTIR analysis. Fig. 7(a-c) shows the TG and DTA curves of ASA-0, ASA-0.2, and ASA-0.35, respectively. The TG curve of ASA-0 reveals three distinct stages. In the first stage, ranging from room temperature to 250 ℃, a total weight loss of 0.95% occurs, attributed to the evaporation of a small amount of organic solvent and water from the pores and surface of the sample. The second stage shows an exothermic peak at 353.3 ℃, corresponding to a mass loss of 4.04% within the temperature range of 250 ℃ to 420 ℃. This loss is due to the partial oxidation of the hydrophobic group Si-CH3 to Si-OH[37]. As the temperature rises further, the Si-CH3 oxidation reaction tends to complete in the third stage, spanning from 420 ℃ to 1200 ℃. At this point, the generated Si-OH undergoes a condensation reaction at high temperatures, forming Si-O-Si bonds[38], leading to a mass loss of 8.83%.
The TG curves of ASA-0.2 and ASA-0.35 also exhibit three stages. In contrast to ASA-0, the mass loss of these samples gradually increased with the addition of aluminum sol. Specifically, the total weight loss in the first stage increases, which may be attributed to the enhanced adsorption of water and organic solvents after the addition of aluminum sol. ASA-0.2 and ASA-0 show similar mass loss in the second stage, and the heat absorption peak corresponding to the Si-OH dehydration condensation reaction is more obvious in ASA-0.35, reaching a mass loss of 10.07%. The XRD analysis in section 3.2 indicates that the ASA-0.35 aerogel at room temperature contains a certain amount of boehmite(γ-AlOOH). Consequently, the dehydration reaction of γ-AlOOH begins to transform into γ-Al2O3, leading to an increase in mass loss in the second stage [39]. When the temperature continues to rise, the γ-AlOOH dehydration reaction tends to be complete. When the temperature exceeds 800 ℃, the mass of ASA-0.2 and ASA-0.35 decreases slightly, and the TG curves become smooth, indicating better thermal stability compared to ASA-0.
Fig. 7(d) illustrates the XRD patterns of ASA-0.2 at different temperatures. ASA-0.2 exhibits an amorphous structure at room temperature, and its crystal structure remains unchanged even after high-temperature calcination. Interestingly, unlike some previous studies on SiO2-Al2O3 composite aerogels, there is no evidence of γ-Al2O3 crystallization in the curve at 1000 ℃. To further investigate the crystal structure of ASA-0.2 at high temperatures, we conducted an additional XRD test after calcination at 1100°C for 2 hours. The results reveal that ASA-0.2 maintains its original structure, and no mullite crystals formed between 1000 ℃ to 1100 ℃, as reported in other studies[40-42]. We hypothesize that this phenomenon is attributed to two reasons. Firstly, the Al content in ASA-0.2 is relatively low, inhibiting the formation of γ-Al2O3 and mullite crystalline phase. Secondly, the Al is uniformly dispersed at the nanoscale throughout the aerogel, and integrates into the SiO2 lattice, effectively suppressing the crystal phase change after high-temperature heat treatment.
Fig .7(e) displays the FTIR spectrum of ASA-0.2 calcined at different temperatures for 2h. The Si-CH3 groups (2964cm-1, 1257cm-1, and 847cm-1) in the skeleton of ASA-0.2 are destroyed during the high-temperature calcination process, leading to the complete disappearance of the absorption peaks corresponding to C-H bonds (2964cm-1 and 1257cm-1) from the graph at 600 ℃. Moreover, the content of Si-OH (957cm-1) in the sample increases above 600 ℃ which is attributed to the oxidation of a portion of Si-CH3 to form Si-OH. A Si-OH condensation reaction occurs at the heat treatment temperature is further raised, resulting in the formation of Si-O-Si bonds, which weakens the characteristic peaks of Si-OH. Upon comparison, it is observed that the absorption peak near 1632cm-1 is enhanced in the heat-treated sample. This increase is attributed to the adsorption of water by the sample, making it completely hydrophilic after high-temperature heat treatment, leading to moisture absorption from the air after a period of time. Additionally, we note that the absorption peak at 610cm-1 of the sample calcined at 600 °C almost disappears due to the thermal decomposition of γ-AlOOH, which is consistent with the TG analysis.
According to the sintering mechanism, high-temperature sintering of aerogels is driven by two main factors: (1) Thermodynamically, phase transition plays a significant role; (2) Kinetically, diffusion of aerogel particles at high temperatures is an important driving force. The influence of high temperature on ASA-0 can be attributed to the following two reasons. Firstly, Si-CH3 on the surface of the aerogel is oxidized to form Si-OH, which subsequently undergoes condensation to form Si-O-Si bonds at high temperatures. This process leads to the sintering of the skeleton particles and disruption of the 3D network structure. Secondly, due to the aerogel's high surface energy, surface diffusion of particles occurs at high temperatures. Continuous particle diffusion increases the particle diameter and contact area, resulting in severe necking and collapse of the pore structure. However, when aluminum sol is added, the nanoparticles provide numerous nucleation sites and enhance the crosslinking degree of the aerogel, making the composite aerogel skeleton more stable at high temperatures. Consequently, the composite aerogel exhibits higher heat resistance compared to pure SiO2 aerogel. The addition of aluminum sol serves two purposes. Firstly, it inhibits the surface driving force between particles by promoting aerogel particle growth and reinforcing the skeleton. Secondly, the uniform distribution of Al in the micron-sized SiO2 particles effectively inhibits phase transition.