Small-Sized Mg–Al LDH Nanosheets Supported on Silica Aerogel with Large Pore Channels: Textural Properties and Basic Catalytic Performance after Activation

Layered double hydroxides (LDHs) have been widely used as an important subset of solid base catalysts. However, developing low-cost, small-sized LDH nanoparticles with enhanced surface catalytic sites remains a challenge. In this work, silica aerogel (SA)-supported, small-sized Mg–Al LDH nanosheets were successfully prepared by one-pot coprecipitation of Mg and Al ions in an alkaline suspension of crushed silica aerogel. The supported LDH nanosheets were uniformly dispersed in the SA substrate with the smallest average radial diameter of 19.2 nm and the thinnest average thickness of 3.2 nm, both dimensions being significantly less than those of the vast majority of LDH nanoparticles reported. The SA/LDH composites also showed large pore volume (up to 1.3 cm3·g) and pore diameter (>9 nm), and therefore allow efficient access of reactants to the edge catalytic sites of LDH nanosheets. In a base-catalyzed Henry reaction of benzaldehyde with nitromethane, the SA/LDH catalysts showed high reactant conversions and favorable stability in 6 successive cycles of reactions. The low cost of the SA carrier and LDH precursors, easy preparation method, and excellent catalytic properties make these SA/LDH composites a competitive example of solid-base catalysts.

For flake-like LDH, the activated catalytic sites are mainly located at their edges [23][24][25][26]. Generally, LDH flakes with a smaller size (especially the radial size) possess higher surface area, corresponding with more edge catalytic sites [27,28]. However, these smaller LDH nanoparticles are readily agglomerated by ab-face stacking in solution due to their high charge density and hydrophilicity [29][30][31]. Furthermore, small LDH nanoparticles are hard to separate after and washed three times with deionized water, and vacuum-dried. The final samples were denoted as SA/LDH-Mg x Al-y, where x and y represent the Mg/Al mole ratio in the precursor solution and the hydrothermal temperature, respectively. For the control sample of unsupported LDH nanosheets (LDH-Mg 2 Al-80), the synthetic method was the same as for the SA/LDH composite except for using a pH of 9.5 instead of 8.8 so as to obtain LDH with an Mg/Al ratio of~2:1. All materials were calcined at 723 K for 2 h and cooled under a flowing stream of nitrogen before catalyst testing.

Characterization
Transmission electron micrographs were taken using a JEOL JEM-2100 Transmission Electron Microscope (TEM) (JEOL Ltd., Tokyo, Japan) operating at an accelerating voltage of 200 kV. Scanning electron micrographs and energy dispersive spectra were obtained using a JEOL JSM-6360LV scanning electron microscope (SEM) equipped with an X-act energy-dispersive X-ray (EDX) analyzer (Oxford INCA, Oxford Instruments, Abingdon, Oxfordshire, UK). N 2 adsorption-desorption isotherms were obtained using a Micromeritics ASAP TriStar II 3020 pore analyzer (Micromeritics Instrument Corp., Norcross, GA, USA) at 77 K under continuous adsorption conditions. The samples were outgassed at 150 • C for 8 h before measurements. The specific surface areas were calculated by the Brunauer-Emmett-Teller (BET) method, and pore size distributions were measured using Barrett-Joyner-Halenda (BJH) analysis from the desorption branches of the nitrogen isotherms. X-ray diffraction (XRD) (Rigaku Corp., Tokyo, Japan) patterns were collected using a Rigaku D/Max-2200 PC X-ray diffractometer with a Cu target (40 kV, 40 mA). CO 2 and NH 3 temperature-programmed desorption (CO 2 -TPD and NH 3 -TPD) profiles of the (supported) LDH samples were obtained using a Quantachrome ChemBET Pulsar equipped with a thermal conductivity detector (TCD) (Quantachrome Instruments, Boynton Beach, FL, USA). A 100 mg sample was calcined for 2 h at 773 K under flowing helium, and then cooled to 323 K to adsorb NH 3 or CO 2 . The physical adsorbed CO 2 was purged by flowing helium at 323 K, and NH 3 -and CO 2 -TPD was subsequently carried out at the rate of 10 K min −1 up to 1073 K. The Mg/Al and Mg/Al/Si mole ratios of the prepared series of SA/LDH composites were assessed by semiquantitative EDS analysis. Mg contents in fresh and used SA/LDH-Mg 2 Al-80 and the Mg/Al mole ratio of unsupported LDH were determined using inductively coupled plasma atomic emission spectroscopy (Leeman Prodigy XP ICP-AES spectrometer) (Teledyne Leeman Labs, Hudson, NH, USA).

Catalytic Evaluation
The Henry reaction of benzaldehyde with nitromethane was performed in liquid phase under atmospheric pressure to evaluate the catalytic properties of the SA/LDH composite series. Benzaldehyde (1 mmol), o-xylene (0.1 g, internal standard), nitromethane (5 mL), catalyst (50 mg), and a magnetic stir bar were placed into a 50 mL Schott bottle. The bottle was then sealed and put into a magnetic stirring bath (equipped with digital heating). The reaction started when the temperature reached 80 • C. After stirring at 400 rpm for 6 h, the reaction was stopped by cooling the bottle to room temperature. The solid catalyst was separated from the reaction product by centrifugation. The reaction product was analyzed by a Shimadzu GC2010 gas chromatograph (GC) (Shimadzu Corp., Kyoto, Japan) equipped with a flame ignition detector (FID), an AOC-20I autosampler, and a capillary column (Rtx-5, 30 m × 0.25 mm × 0.25 µm, crossbond 5% diphenyl/95% dimethyl polysiloxane). For 6 successive cycles of reactions, the used catalyst was washed with ethanol or water to partly remove the carbonaceous products adsorbed and then vacuum-dried for reuse in the first five cycles and calcined at 823 K to thoroughly remove the carbonaceous products for reuse in the sixth (last) cycle. To evaluate the effect of solvents on the catalytic properties of the SA/LDH catalyst, benzaldehyde (1 mmol), o-xylene (0.1 g, internal standard), nitromethane (5 mmol), solvent (4.73 mL), and catalyst (50 mg) were adopted.

Results and Discussion
The synthesis of SA-supported LDH nanosheets is described in Scheme 1. Briefly, a small amount of mixed solution A (Mg(NO 3 ) 2 /Al(NO 3 ) 3 ) was first added to the SA suspension, in which process the Mg 2+ and Al 3+ were adsorbed onto the negatively charged inner and outer surfaces of the SA. Then solution B (NaOH/NaHCO 3 ) was added to precipitate the Mg 2+ and Al 3+ adsorbed in situ and form LDH precursor (Pre-LDH) uniformly on the surface of the SA. Thereafter, a large amount of mixed solution A and a calculated amount of solution B were added alternately to maintain the solution pH at a certain value. Finally, the SA/LDH composite formed through hydrothermal treatment of the obtained SA-supported Pre-LDH.

Results and Discussion
The synthesis of SA-supported LDH nanosheets is described in Scheme 1. Briefly, a small amount of mixed solution A (Mg(NO3)2/Al(NO3)3) was first added to the SA suspension, in which process the Mg 2+ and Al 3+ were adsorbed onto the negatively charged inner and outer surfaces of the SA. Then solution B (NaOH/NaHCO3) was added to precipitate the Mg 2+ and Al 3+ adsorbed in situ and form LDH precursor (Pre-LDH) uniformly on the surface of the SA. Thereafter, a large amount of mixed solution A and a calculated amount of solution B were added alternately to maintain the solution pH at a certain value. Finally, the SA/LDH composite formed through hydrothermal treatment of the obtained SA-supported Pre-LDH. The morphology of initial template SA and the prepared SA/LDH composite were characterized by TEM. Typically, SA shows disordered pore structure ( Figure 1a) and low TEM contrast due to its low density and amorphous structure. In the current SA/LDH composite series (Figure 1b-f), a large amount of LDH nanosheets was observed located at the inner and outer surfaces of the SA substrate. The SA/LDH series synthesized at a low temperature of 80 °C all showed small LDH radial diameters and the average radial diameters were ca 19.2 nm, 31.4 nm, and 35.8 nm for SA/LDH-Mg2Al-80, SA/LDH-Mg5Al2-80, and SA/LDH-Mg3Al-80, respectively (Table 1 and Figure S1). However, the radial diameters of LDH increased to ~53.0 and 57.4 nm for the SA/LDH-Mg2Al-105 and SA/LDH-Mg2Al-150, which were prepared at higher temperatures of 105 °C and 150 °C, respectively. The same trend was found in the thickness change of the SA/LDH series (Table 1 and Figure S2). The thicknesses were 3.2 nm, 3.2 nm, and 3.6 nm for SA/LDH-Mg2Al-80, SA/LDH-Mg5Al2-80, and SA/LDH-Mg3Al-80, respectively, and increased to 4.5 and 4.3 nm, for SA/LDH-Mg2Al-105 and SA/LDH-Mg2Al-150, respectively. It should be noted that the SA/LDH composite synthesized at higher pH values such as 9.5 showed significant destruction of SA structure due to the strong alkali corrosion in the preparation. Therefore, the impact of pH values on the structure of the composite is not discussed in this work. EDS mappings of the prepared SA/LDH composites ( Figure S3) showed an almost complete overlap of the Si, Mg, Al, O elemental mappings, further demonstrating that the supported LDH nanosheets were uniformly dispersed on the SA substrate. However, the Mg/Al ratios in the obtained composites were not consistent with those in precursor solutions (Table 1)  The morphology of initial template SA and the prepared SA/LDH composite were characterized by TEM. Typically, SA shows disordered pore structure ( Figure 1a) and low TEM contrast due to its low density and amorphous structure. In the current SA/LDH composite series (Figure 1b-f), a large amount of LDH nanosheets was observed located at the inner and outer surfaces of the SA substrate. The SA/LDH series synthesized at a low temperature of 80 • C all showed small LDH radial diameters and the average radial diameters were ca 19.2 nm, 31.4 nm, and 35.8 nm for SA/LDH-Mg 2 Al-80, SA/LDH-Mg 5 Al 2 -80, and SA/LDH-Mg 3 Al-80, respectively (Table 1 and Figure S1). However, the radial diameters of LDH increased to~53.0 and 57.4 nm for the SA/LDH-Mg 2 Al-105 and SA/LDH-Mg 2 Al-150, which were prepared at higher temperatures of 105 • C and 150 • C, respectively. The same trend was found in the thickness change of the SA/LDH series (Table 1 and Figure S2). The thicknesses were 3.2 nm, 3.2 nm, and 3.6 nm for SA/LDH-Mg 2 Al-80, SA/LDH-Mg 5 Al 2 -80, and SA/LDH-Mg 3 Al-80, respectively, and increased to 4.5 and 4.3 nm, for SA/LDH-Mg 2 Al-105 and SA/LDH-Mg 2 Al-150, respectively. It should be noted that the SA/LDH composite synthesized at higher pH values such as 9.5 showed significant destruction of SA structure due to the strong alkali corrosion in the preparation. Therefore, the impact of pH values on the structure of the composite is not discussed in this work. EDS mappings of the prepared SA/LDH composites ( Figure S3) showed an almost complete overlap of the Si, Mg, Al, O elemental mappings, further demonstrating that the supported LDH nanosheets were uniformly dispersed on the SA substrate. However, the Mg/Al ratios in the obtained composites were not consistent with those in precursor solutions (Table 1) as only a fraction of the Mg(II) ions precipitated at the synthetic pH of 8.8 due to the high solubility product constant (K sp = 5.1 × 10 −12 ) of Mg(OH) 2 while Al(III) almost completely precipitated in the form of Al(OH) 3 with a low K sp value of 1.3 × 10 −33 under the same conditions. In addition, the Mg/Al mole ratios of SA/LDH-Mg 2 Al-105 and SA/LDH-Mg 2 Al-150 were both higher than that of SA/LDH-Mg 2 Al-80 which is attributed to the formation of magnesium silicate under the high synthetic temperatures as proven by the subsequent XRD analysis.
The XRD pattern of the fabricated SA/LDH composites ( Figure 2) all showed typical but weak LDH diffraction peaks (Standard PDF card: #89-0460), which can be indexed to the (003), (006), (012), and (110) planes. For the SA/LDH prepared at a high synthetic temperature of 105 • C and 150 • C, the strongest reflection peak of magnesium silicate indexed to the (101) plane (Standard PDF card: #87-2037) appeared, which is probably because SiO 3 2− derived from the partial dissolution of SA in an alkaline preparation solution of SA/LDH tends to react with unprecipitated Mg(II) ions in the LDH precursor at high synthetic temperatures. Figure 3 and Figure S4 show the N 2 adsorption-desorption isotherms of the SA/LDH composites, SA and the unsupported LDH, respectively. All SA/LDH composites displayed a type IV isotherm with H 1 hysteresis and a sharp increase in volume adsorbed at P/P 0 ≈ 0.6 to 0.7, which indicates that SA/LDH composite partially retained the mesoporous structure of the SA. The BET surface area (S BET ), pore volume (V p ), and pore diameter (D p ) of the relative samples determined via nitrogen adsorption-desorption measurements are listed in Table 1. All SA/LDH samples showed a high surface area of more than 400 m 2 ·g, a large pore volume of greater than 0.90 cm 3 ·g, and pore diameters larger than 9 nm. SA/LDH-Mg 2 Al-80 possesses an obviously higher surface area than the other SA/LDH samples, and the surface area of the samples decreased with increasing synthetic temperature. As expected, the surface area, pore volume, and pore diameter of the SA/LDH series all decreased in comparison to the substrate SA due to the partial blockage of SA pore channels by LDH nanosheets and the inherently higher density of the LDH than that of the porous SA in the composite. On the basis of the results in Table 1, it is concluded that the SA/LDH composites with smaller LDH sizes have the higher surface area, which is attributed to the smaller LDH nanosheets minimizing the blocking of SA pore channels. formation of magnesium silicate under the high synthetic temperatures as proven by the subsequent XRD analysis. The XRD pattern of the fabricated SA/LDH composites ( Figure 2) all showed typical but weak LDH diffraction peaks (Standard PDF card: #89-0460), which can be indexed to the (003), (006), (012), and (110) planes. For the SA/LDH prepared at a high synthetic temperature of 105 °C and 150 °C, the strongest reflection peak of magnesium silicate indexed to the (101) plane (Standard PDF card: #87-2037) appeared, which is probably because SiO3 2− derived from the partial dissolution of SA in an alkaline preparation solution of SA/LDH tends to react with unprecipitated Mg(II) ions in the LDH precursor at high synthetic temperatures. Figures 3 and S4 show the N2 adsorption-desorption isotherms of the SA/LDH composites, SA and the unsupported LDH, respectively. All SA/LDH composites displayed a type IV isotherm with H1 hysteresis and a sharp increase in volume adsorbed at P/P0 ≈ 0.6 to 0.7, which indicates that SA/LDH composite partially retained the mesoporous structure of the SA. The BET surface area (SBET), pore volume (Vp), and pore diameter (Dp) of the relative samples determined via nitrogen adsorption-desorption measurements are listed in Table 1. All SA/LDH samples showed a high surface area of more than 400 m 2 ·g, a large pore volume of greater than 0.90 cm 3 ·g, and pore diameters larger than 9 nm. SA/LDH-Mg2Al-80 possesses an obviously higher surface area than the other SA/LDH samples, and the surface area of the samples decreased with increasing synthetic temperature. As expected, the surface area, pore volume, and pore diameter of the SA/LDH series all decreased in comparison to the substrate SA due to the partial blockage of SA pore channels by LDH nanosheets and the inherently higher density of the LDH than that of the porous SA in the composite. On the basis of the results in Table 1, it is concluded that the SA/LDH composites with smaller LDH sizes have the higher surface area, which is attributed to the smaller LDH nanosheets minimizing the blocking of SA pore channels.      To evaluate the basicity of the SA/LDH series, CO2-TPD measurements were conducted. In general, activated LDHs possess weakly basic sites, moderately basic sites, and strongly basic sites, ascribed to the surface hydroxyl group, metal-oxygen pairs (e.g., Mg 2+ -O2), and low-coordination oxygen anions, respectively [54]. For CO2-TPD, the basic sites in the high desorption temperature range are usually designated as strong basic sites, and vice versa [55]. As shown in Figure 4 and Table 2, the SA/LDH series showed a broad CO2 desorption band from 230 to 790 °C with the    To evaluate the basicity of the SA/LDH series, CO2-TPD measurements were conducted. In general, activated LDHs possess weakly basic sites, moderately basic sites, and strongly basic sites, ascribed to the surface hydroxyl group, metal-oxygen pairs (e.g., Mg 2+ -O2), and low-coordination oxygen anions, respectively [54]. For CO2-TPD, the basic sites in the high desorption temperature range are usually designated as strong basic sites, and vice versa [55]. As shown in Figure 4 and Table 2, the SA/LDH series showed a broad CO2 desorption band from 230 to 790 °C with the To evaluate the basicity of the SA/LDH series, CO 2 -TPD measurements were conducted. In general, activated LDHs possess weakly basic sites, moderately basic sites, and strongly basic sites, ascribed to the surface hydroxyl group, metal-oxygen pairs (e.g., Mg 2+ -O 2 ), and low-coordination oxygen anions, respectively [54]. For CO 2 -TPD, the basic sites in the high desorption temperature range are usually designated as strong basic sites, and vice versa [55]. As shown in Figure 4 and Table 2, the SA/LDH series showed a broad CO 2 desorption band from 230 to 790 • C with the highest peak in the high temperature range of 605.7 to 649.3 • C and a broad shoulder peak in the low temperature range of 417.6 to 458.5 • C, demonstrating that the surface basic strength of the composites was widely distributed and the SA/LDH composites possessed mainly moderate and strong base sites originating from the O 2− anions and surface Mg-O 2− pairs of LDH. The highest desorption peak position of SA/LDH-Mg 2 Al-80, SA/LDH-Mg 5 Al 2 -80, and SA/LDH-Mg 3 Al-80 all shifted to higher temperatures compared with SA/LDH-Mg 2 Al-105, SA/LDH-Mg 2 Al-150, and LDH-Mg 2 Al-80, indicating that the SA/LDH samples synthesized at low temperature have more strongly basic sites than those synthesized at high temperatures and the unsupported LDH with large size and some particle aggregation ( Figure S5). In addition, all samples showed a relatively weak peak in the temperature range 121.5 to 160.6 • C, corresponding to weakly basic sites derived from the OHgroups of the LDH. The total peak area of CO 2 -TPD signals is linearly proportional to the amount of CO 2 adsorbed and can thus be used as a semi-quantitative parameter for assessing the number of basic sites of the catalysts. The basic site numbers of the samples are ranked as follows ( Table 2) 2 Al-80, which has a smaller LDH size and a higher surface area, may be due to the higher Mg content of the former than the latter. This discrepancy could also explain the higher observed basic site density of the unsupported LDH in comparison to those of the SA/LDH-Mg 2 Al-105 and SA/LDH-Mg 2 Al-150 with greater surface area and smaller LDH size. From these results, it is concluded that the basicity of LDH in this system is enhanced by multiple factors including small LDH size, high surface area, and high Mg content. LDH-Mg2Al-80, indicating that the SA/LDH samples synthesized at low temperature have more strongly basic sites than those synthesized at high temperatures and the unsupported LDH with large size and some particle aggregation ( Figure S5). In addition, all samples showed a relatively weak peak in the temperature range 121.5 to 160.6 °C, corresponding to weakly basic sites derived from the OHgroups of the LDH. The total peak area of CO2-TPD signals is linearly proportional to the amount of CO2 adsorbed and can thus be used as a semi-quantitative parameter for assessing the number of basic sites of the catalysts. The basic site numbers of the samples are ranked as follows ( Table 2)    The base-catalyzed Henry reaction of conversion of benzaldehyde to trans-1-phenyl-2-nitro-ethanol and trans-1-nitro-2-phenylethylene (Species 1, 2, and 3 in Equation (1), respectively) was taken as a model reaction to evaluate the basic catalytic properties of the SA/LDH composites. The solvent plays a significant role in catalysis reactions, such as dissolution of reactants and products, adjusting the interaction of reactants and products with the surface catalytic sites of catalysts and activating reactants and intermediates. The catalytic properties of SA/LDH-Mg 2 Al-80 with the smallest LDH size in different solvents were given in Table 3. In general, the relatively high conversions were obtained in the solvents with relative weak polarity, such as nitromethane, dichloromethane, and toluene, while the solvents with strong polarity, such as ethanol, THF, DMF, and water, gave low conversions. This trend was similar to that found in the reaction of nitrobenzaldehyde with nitromethane [56] and could be attributed as the reason that protic polar solvents competed with the nitromethane (pKa = 10.21) for the surface basic sites of the catalyst, reducing the number of catalytic sites available for the reaction. When nitromethane was used as both a solvent and a reactant, the highest conversion of 96.8% was obtained owing to the higher concentration of nitromethane and thus a higher contact probability of nitromethane with benzaldehyde. In addition, the adsorption of the yellow product trans-1-nitro-2-phenylethylene on the catalyst was found for the reactions in all solvents except for ethanol ( Figure S6). Based on these results and the low environment hazard, nitromethane was selected as the solvent for the subsequent studies.   (1), respectively) was taken as a model reaction to evaluate the basic catalytic properties of the SA/LDH composites. The solvent plays a significant role in catalysis reactions, such as dissolution of reactants and products, adjusting the interaction of reactants and products with the surface catalytic sites of catalysts and activating reactants and intermediates. The catalytic properties of SA/LDH-Mg2Al-80 with the smallest LDH size in different solvents were given in Table 3. In general, the relatively high conversions were obtained in the solvents with relative weak polarity, such as nitromethane, dichloromethane, and toluene, while the solvents with strong polarity, such as ethanol, THF, DMF, and water, gave low conversions. This trend was similar to that found in the reaction of nitrobenzaldehyde with nitromethane [56] and could be attributed as the reason that protic polar solvents competed with the nitromethane (pKa = 10.21) for the surface basic sites of the catalyst, reducing the number of catalytic sites available for the reaction. When nitromethane was used as both a solvent and a reactant, the highest conversion of 96.8% was obtained owing to the higher concentration of nitromethane and thus a higher contact probability of nitromethane with benzaldehyde. In addition, the adsorption of the yellow product trans-1-nitro-2-phenylethylene on the catalyst was found for the reactions in all solvents except for ethanol ( Figure S6). Based on these results and the low environment hazard, nitromethane was selected as the solvent for the subsequent studies.  The catalytic properties of SA/LDH series catalysts with various textures are given in Table 4. SA/LDH-Mg2Al-80, SA/LDH-Mg5Al2-80, and SA/LDH-Mg3Al-80 synthesized at a low temperature all showed high reaction conversions of 96.8%, 98.2%, and 92.7%, respectively, which are significantly higher than those of SA/LDH-Mg2Al-105 (59.3%) and SA/LDH-Mg2Al-150 (51.9%) Table 3.
Henry reaction of benzaldehyde with nitromethane over SA/LDH-Mg 2 Al-80 in different solvents a . The catalytic properties of SA/LDH series catalysts with various textures are given in Table 4. SA/LDH-Mg 2 Al-80, SA/LDH-Mg 5 Al 2 -80, and SA/LDH-Mg 3 Al-80 synthesized at a low temperature all showed high reaction conversions of 96.8%, 98.2%, and 92.7%, respectively, which are significantly higher than those of SA/LDH-Mg 2 Al-105 (59.3%) and SA/LDH-Mg 2 Al-150 (51.9%) synthesized at high temperatures (Table 4), demonstrating that the reactant conversions are positively correlated with the basicity of the catalysts. It should be noted that the SA/LDH-Mg 2 Al-80 has a selectivity of 97.3% to product 3, which is significantly higher than those of the SA/LDH-Mg 5 Al 2 -80 (78.6%) and SA/LDH-Mg 3 Al-80 (87.2%), which could be attributed to the higher Al content of SA/LDH-Mg 2 Al-80, endowing it with higher acidic site density and acidic strength than SA/LDH-Mg 5 Al 2 -80 and SA/LDH-Mg 3 Al-80 as shown by NH 3 -TPD ( Figure S7 and Table S1). This characteristic would promote the dehydration of trans-1-phenyl-2-nitro-ethanol to give trans-1-nitro-2-phenylethylene [57]. This reaction did not proceed when using bare SA as a catalyst, as expected, and the unsupported LDH-Mg 2 Al-80 showed a reactant conversion of 75.9%, lower than SA/LDH-Mg 2 Al-80, which is attributed to the higher basic strength of SA/LDH-Mg 2 Al-80 than LDH-Mg 2 Al-80. Summarizing the results above, the SA/LDH samples with high basicity all showed high reactant conversions (>90%) in the model Henry reaction, indicating the favorable accessibility of the surface basic sites of SA/LDH composite.

Entry Solvent Conversion of 1 (%) Selectivity to 2 (%) Selectivity to 3 (%)
To examine the stability of the SA/LDH catalyst against deactivation, 6 successive cycles of reactions of benzaldehyde with nitromethane over SA/LDH-Mg 2 Al-80 were conducted using nitromethane as a solvent. After each cycle, the used catalyst was washed with ethanol or water to remove the products adsorbed. For the last cycle, the carbonaceous products on the used catalyst were thoroughly removed by calcination before the reaction. In the first five cycles, the conversions decreased continuously from 96.4% to 75% for the catalyst reused by ethanol washing treatment, while those declined sharply to 24.8% for the catalyst reused by water washing treatment ( Figure 5). The contents of the active component Mg are 21.9% and 21.2% in the fresh catalyst and used catalyst after six cycles (ethanol washing), respectively, determined by ICP-MS, demonstrating a negligible leaching of Mg 2+ ions. Therefore, it could be concluded that catalyst deactivation was mainly derived from the covering of carbonaceous products on the surface active sites of the catalyst and ethanol washing could effectively remove the carbonaceous products adsorbed on the catalyst and prevent the deactivation of catalysts, which is consistent with the above observation of a weak adsorption of the yellow product on the catalyst when using ethanol as a solution. In the sixth cycle, after calcining the used catalysts, the conversions restored from 75% to 87.7% and from 24.8% to 84.3% for the ethanol-washing sample and the water-washing sample, respectively. The high selectivities to product trans-1-nitro-2-phenylethylene were kept in all cycles in spite of the obvious changes of the conversions.  Various solid base catalysts have been employed for catalyzing the Henry reaction of benzaldehyde with nitromethane, and different reaction conditions were adopted based on the inherent characteristics of the special catalysts. Table 5 gives a rough catalytic property comparison of the SA/LDH catalysts with other solid base catalysts regardless of the specific reaction conditions. The SA/LDH catalysts showed the highest conversion of 98.2%, which is higher than those of most of the other inorganic catalysts and comparable to those of the catalysts containing organic base groups with high cost and low thermal stability. Furthermore, compared to other inorganic catalysts, the optimized SA/LDH catalyst gave a higher selectivity of 97.3% to the main product trans-1-nitro-2-phenylethylene, which is an important chemical intermediate for slimicides and dyes [57].

Conclusions
In this work, small-sized Mg-Al LDH nanosheets supported on silica aerogel with a large average pore diameter and large pore volume were successfully prepared by a one-pot coprecipitation method. The impact of synthesis parameters on the textural properties of the obtained composite was evaluated. Low Mg/Al precursor ratios and low hydrothermal treatment temperatures result in small LDH sizes and high surface area of the SA /LDH composite. The Various solid base catalysts have been employed for catalyzing the Henry reaction of benzaldehyde with nitromethane, and different reaction conditions were adopted based on the inherent characteristics of the special catalysts. Table 5 gives a rough catalytic property comparison of the SA/LDH catalysts with other solid base catalysts regardless of the specific reaction conditions. The SA/LDH catalysts showed the highest conversion of 98.2%, which is higher than those of most of the other inorganic catalysts and comparable to those of the catalysts containing organic base groups with high cost and low thermal stability. Furthermore, compared to other inorganic catalysts, the optimized SA/LDH catalyst gave a higher selectivity of 97.3% to the main product trans-1-nitro-2-phenylethylene, which is an important chemical intermediate for slimicides and dyes [57].

Conclusions
In this work, small-sized Mg-Al LDH nanosheets supported on silica aerogel with a large average pore diameter and large pore volume were successfully prepared by a one-pot coprecipitation method. The impact of synthesis parameters on the textural properties of the obtained composite was evaluated. Low Mg/Al precursor ratios and low hydrothermal treatment temperatures result in small LDH sizes and high surface area of the SA /LDH composite. The SA/LDH composite synthesized at a hydrothermal treatment temperature of 80 • C using an Mg/Al precursor ratio of 2:1 showed the highest surface area of 587.4 m 2 ·g and the smallest LDH size with an average lateral dimension of ca 19.2 nm and an average thickness of ca 3.2 nm. These dimensions are significantly lower than those of the vast majority of (supported) LDH nanoparticles reported. The CO 2 -TPD results revealed that the basicity of LDH in this system is enhanced by multiple factors, including small LDH sizes and high Mg contents of the catalysts. However, these factors may conflict. For example, the sample with the highest Mg content did not exhibit the smallest size of LDH and careful consideration of the implications of optimizing one parameter over the other would be needed. In the base-catalyzed model reaction of benzaldehyde with nitromethane, the activated SA/LDH samples with smaller LDH sizes showed higher reactant conversions (>90%) than those with larger LDH sizes and unsupported LDH nanosheets with some particle aggregation. The used SA/LDH catalysts could be refreshed effectively through ethanol washing or calcination to partly or thoroughly remove the carbonaceous products adsorbed. All these results demonstrate high surface basic site strength and favorable accessibility of the surface catalytic sites of the silica-aerogel-supported, ultra-small LDH nanosheets. The low cost of commercial SA, easy preparation, and excellent properties make the silica-aerogel-supported LDH composite a competitive candidate for solid-base catalysts.  Figure S5: TEM image of unsupported LDH nanosheets (LDH-Mg 2 Al-80) using as a contrast sample. The inset is the corresponding radial diameter distribution, Figure S6: The used catalyst SA/LDH-Mg 2 Al-80 collected by centrifugation after the reaction of benzaldehyde with nitromethane in different solvents such as nitromethane, ethanol, dichloromethane, toluene, DMF, Water, and THF, Figure S7: NH 3 -TPD profiles of SA/LDH series synthesized at a temperature of 80 • C, Table S1: The semi-quantitative results of NH 3 -TPD measurements.