Serving as a critical aspect of synthetic chemistry applied in numerous industrial processes, base-catalyzed aldol condensation reactions have caused considerable attention so as to produce various value-added chemicals with all kinds of industrial applications [1-4]. Among these condensation processes, the reaction of isobutyraldehyde (IBD) with formaldehyde (FA) serves as a crucial process to produce hydroxypivalaldehyde (HPA), which is an important intermediate for production of neopentyl glycol (NPG) widely used in the manufacturing of polyesters, lubricants, plasticizers, and synthetic resin coatings [5, 6]. Generally, homogeneous catalysts such as triethylamine, NaOH and KOH have been widely used in this reaction due to cost effectiveness and high activity; whereas, they show disadvantages in separation and recovery, and are unable to meet the increasingly stringent requirements for environmental benignity and sustainability [7-9]. Based on the idea of environmentally friendly catalysis, solid basic catalysts such as metal oxides, metal nitrides and perovskites are extremely desirable in these condensation processes with facile separation and depressed pollution [10-12]. However, compared with homogeneous catalysts (the highest level of HPA yield: 73.5%), the HPA yield of solid basic catalysts is not satisfied (only 4.3%-50%), and the catalytic activity needs to be further improved [10]. Therefore, it is essential to explore and develop solid basic catalysts with high catalytic performance.

Layered double hydroxides (LDHs), owing to their large amount of surface hydroxyl and structural memory effects, have been used in this aldol condensation reaction as solid basic catalysts in our previous work [13, 14]. The prepared rehydrated Ca₄Al-LDHs displays a superior activity (HPA yield: 60%) to those of conventional solid basic catalysts (HPA yield: 22%-50%). Moreover, it has been verified that the weak Brönsted basic site (7-coordinated Ca-OH group) acts as an active site of the aldol condensation reaction and promotes the production desorption [13, 15]. Despite advances in re-Ca₄Al-LDHs catalyst, gaps remain in catalytic activity compared to liquid basic catalysts, which inspires us to take measures to further enhance catalytic performance. Geometric effects by introducing cations with different ionic radius into LDHs laminar lattice play an crucial role on the regulation of the microstructure of active center [16-19]. For instance, geometric effects would impose active-site isolation, distort geometry structure and change adsorption modes, leading to an enhanced activity or selectivity of heterogeneous catalysts [20-24]. Some metal ions as a structural promoter have been investigated in redox, hydrogenation and condensation reactions, such as transition metals (Mn, Zr, etc.) and rare earth elements (La, Y, Ce, etc.) [25-28]. Herein, we hypothesized that a positive geometric effect by insertion of cation with appropriate ionic radius in Ca₄Al-LDHs catalyst will further improve the catalytic performance.

Inspired by the above idea, a series of rehydrated Ca₄Al_1-xGa_x-LDHs and Ca₄Al_1-xIn_x-LDHs (x = 0, 0.05, 0.10, 0.15) catalysts were prepared and used in the reaction of FA with IBD to produce HPA. Among these samples, the re-Ca₄Al_0.90Ga_0.10-LDHs catalyst displays the best catalytic performance with a 91% IBD conversion and 72% HPA yield, which is significantly exceeding the re-Ca₄Al-LDHs catalyst (conversion: 80%; yield: 60%) and even close to the catalytic properties of liquid base catalysts (conversion: 97%; yield: 73.5%). According to our previous work, the weak basic site (7-coordinated Ca-OH group) in re-Ca₄Al-LDHs has been demonstrated to serve as an active site of aldol condensation reaction, which facilitates the desorption of product molecules accompanied with the increment of HPA selectivity. Herein, as a combination study including in situ XANES, CDCl₃-FTIR characterization and structure-property correlation study demonstrated, re-Ca₄Al_0.90Ga_0.10-LDHs with incorporation of Ga into hydrotalcite laminates gives rise to a moderate lattice expansion and thus increases the concentration of weak basic sites, accounting for an extraordinary catalytic performance; whereas, re-Ca₄Al_0.90In_0.10-LDHs possess abundant highly coordinated In-OH group, leads to a decreased 7-coordinated Ca-OH group and poor catalytic properties. Thus, we came to the conclusion that Ca₄Al-LDHs is a structure-sensitive catalyst, whose performance can be optimized by introducing promoter cation with appropriate ionic radius. This work investigates geometry effects on the structure of Ca-based hydrotalcites, and provides a way to attain high performance solid basic catalyst with refined regulation on the structure toward aldol condensation reaction.

Ga(NO₃)₂·6H₂O, Ca(NO₃)₂·4H₂O, In(NO₃)₂·6H₂O, Al(NO₃)₃·9H₂O, FA (37% aqueous solution), IBD, NaOH, 1, 4-dioxane, cetyltrimethylammonium bromide (CTAB) and HPA were purchased from Sigma-Aldrich. Notably, all the reagents were used without further purification. In the total experimental processes, decarbonated and deionized water was also used.

Ca₄Al_1-xGa_x-LDHs precursors with various Ga³⁺/M³⁺ atomic ratios were synthesized by the co-precipitation method. Solution A (a mixture solution) was prepared by dissolving Ca(NO₃)₂·4H₂O, Ga(NO₃)₂·6H₂O, and Al(NO₃)₃·9H₂O with various Ga³⁺/M³⁺ atomic ratios (0, 0.05, 0.10, 0.15) in 100 mL of decarbonated and deionized water. 2.96 g of NaOH was dissolved in the same volume of water to obtain a base solution (solution B). Both solution A and B were simultaneously dropped into a round-bottom flask with three neck under the protection of N₂ stream. Afterwards, the obtained precipitate was stirred vigorously for 10 h; moreover, the product was separated and washed thoroughly, and finally dried at 60 ℃ for another 20 h. Subsequently, the precursors were activated via a two-step approach. Firstly, the LDHs precursor sample was calcined at 500 ℃ for 5 h in a N₂ atmosphere with a heating rate of 5 ℃ min^-1. After cooling to room temperature, the sample was dispersed and stirred in NaOH solution for 3 h. Then the obtained sample was washed thoroughly with deionized water and ethanol; finally, the sample was dried in a vacuum oven at 65 ℃ for 24 h (denoted as re-Ca₄Al_1-xGa_x-LDHs). Rehydrated Ca₄Al_1-xIn_x-LDHs samples were prepared through a similar process and denoted as re-Ca₄Al_1-xIn_x-LDHs.

The condensation reaction was performed on a cylindrical glass tube (batch-type reactor, 50 mL). Dioxane serving as internal standard (0.2 g) and CTAB (a phase transfer catalyst: 0.1 g) were added to FA (6 mL, 37% aqueous solution) and IBD (5 mL), then the catalyst Ca₄Al_1-xGa_x-LDHs (0.2 g) or re-Ca₄Al_1-xIn_x-LDHs was dispersed into the mixture solution and heated to 70 ℃ for 8 h in a N₂ atmosphere whiling stirring. When the reaction started, aliquots were taken every 30 min. Product analysis was carried out using a Shimadzu GC-2014 with an Rtx-5 capillary column (diameter: 0.25 mm; length: 30 m). A flame ionization detector (FID) was used to analyze all the products. GC-MS was used to identify the product composition. Conversion and selectivity were calculated based on the integrated gas-phase peak area of IBD and HPA.

Powder XRD measurements were performed on the diffractometer (Rigaku XRD-6000) by using Cu K_a radiation (λ = 0.1542 nm, 40 kV, 40 mA; scanning rate: 10 ℃ min^-1; 2θ angle: 3° to 70°). Scanning electron microscopy (Zeiss Supra 55) was executed to investigate the morphology of the catalyst (accelerating voltage: 20 kV). Low-temperature N₂ adsorption-desorption isotherms of the samples were determined on a Quantachrome Autosorb-1C-VP analyzer and the total specific surface area and pore volume were obtained by the Brunauer-Emmett-Teller (BET) measurement. Elemental composition was obtained with the Shimadzu ICPS-7500 inductively coupled plasma atomic emission spectrometer (ICP-AES).

X-ray photoelectron spectra (XPS) were obtained by using a Thermo VG Escalab X-ray photoelectron spectrometer (pressure, 2 × 10^-9 Pa), using Al K_α X-rays as the excitation source. EXAFS measurements were carried out at the 4B7A of the Beijing Synchrotron Radiation Facility (BSRF), Institute of High Energy Physics (IHEP), Chinese Academy of Sciences (CAS). Additionally, the typical energy (2.2G eV) of the storage ring (maximum current: 100 mA) was used with the Si(111) double crystal monochromator. The data analysis and fitting was conducted on the IFFEFIT (1.2.11) data analysis package including Athena, Atoms, Artemis, and FEFF6.

Fourier-transform infrared spectroscopy (FTIR) was carried out with transmission mode in a Bruker Equinox 55 spectrometer in the range of 4000 and 400 cm^-1 (resolution: 4 cm^-1; per spectrum: 600 scans). FTIR spectroscopy was performed with absorption mode on an instrument spectrophotometer of Nicolet 380. The adsorbed deuterated chloroform CDCl₃-FTIR spectroscopy was used to characterize microstructure of the basic site. The operational processes were obtained according to our previous work [13]. According to the adsorption of CDCl₃, the strength of basic site can be measured via the following equation:

(1)

where ΔνCD represents the shift of νCD (cm^-1) while PA represents the proton affinity (kJ mol^-1).

DFT calculations were performed using the Vienna Ab-initio Simulation Package (VASP) code [29, 30]. Spin-polarized Perdew-Burke-Ernzerhof (PBE) functional was employed to solve the Kohn-Sham equation, and the Grimme's DFT-D3 method was used to consider the effect of van der Waals interaction [31, 32]. The core electrons were performed by the projector augmented wave (PAW) method [33]. A 3 × 3 × 1 k-points was taken in the geometry optimization and a 6 × 6 × 1 k-points was taken in the Bader charge calculations. The convergence criteria was 1.0 × 10^-4 eV for the electronic self-consistent iteration and 0.05 eV/Å for force. Ca₄Al-Cl-LDHs was constructed based on a 3 × 3 supercell of Mg(OH)₂. Ga/In-doped LDHs model was built by replacing one Al atom with one Ga or In atom at (001) surface. Two chloride ions were added in the unit cell of the studied LDHs to keep the charge neutral. A 15 Å vacuum was added to avoid the interaction between the slabs. 7-coordinated Ca was built by adding one OH group on the Ca at (001) surface. HPA desorption energy is calculated as follows:

(2)

where E_surf is the energy of the Ca₄Al-Cl-LDHs (001) surface, E_HPA is the energy of HPA in gas phase, and E_HPA/surf is the energy of HPA adsorbs on the LDHs (001) surface.

The XRD patterns of Ca₄Al_1-xGa_x-LDHs and Ca₄Al_1-xIn_x-LDHs precursors (Fig. 1A) display a series of characteristic reflections at 10.5°, 20.8°, 31.4°, 37.8°, and 55.6°, which are indexed to Ca₄Al-LDHs with interlayer NO₃^- anions [34, 35]. The position of the (003) reflection shifts to lower angle after the incorporation of Ga and In compared with Ca₄Al-LDHs, indicating that Ga or In is successfully introduced into the hydrotalcite laminate (Fig. S1). After calcination at 500 ℃ for 4 h, the Ca₄Al_1-xGa_x-LDHs and Ca₄Al_1-xIn_x-LDHs precursors transform to a mixed oxide phase, which are denoted as Ca₄Al_1-xGa_x-MMOs and Ca₄Al_1-xIn_x-MMOs, as shown in Fig. 1B. The diffraction peaks at 37.6°, 43.7°, and 63.8° are indexed to a cubic CaO phase; moreover, the presence of Ga₂O₃ or In₂O₃ phases was not detected, implying a high dispersion with tiny particle size in Ca₄Al-LDHs materials. Afterwards, the reconstructed Ca₄Al_1-xGa_x-LDHs and Ca₄Al_1-xIn_x-LDHs can be obtained through the rehydration process of MMOs samples in an aqueous NaOH. The results (Fig. 1C) show that the rehydrated samples re-express characteristic feature of an LDHs phase and the (003) reflection shifts from 10.5° to 11.5° (2θ) relative to the NO₃^--containing LDHs precursor, indicating a replacement of NO₃^- by OH^- groups in the interlayer region. SEM measurement was performed to explore the morphology of these rehydrated samples (Fig. 2A-2G), from which laminar microcrystals (a lateral diameter: ~2 μm) are observed for all these samples. Nitrogen adsorption-desorption isotherms were performed to study the specific surface area and pore-size distribution of all these rehydrated samples. As summarized in Table 1, with the increase in the amount of Ga or In, the surface area shows an increase trend from 15.1 m²/g to 40.9 m²/g with an average pore diameter ranging in 21.3 nm to 24.9 nm. Furthermore, the molar ratios of Ga³⁺/M³⁺ and In³⁺/M³⁺ determined by ICP-AES are close to nominal values, indicating that Ga or In is successfully introduced into re-Ca₄Al-LDHs.

The catalytic performance of these Ca₄Al_1-xGa_x-LDHs and re-Ca₄Al_1-xIn_x-LDHs samples towards the aldol condensation of FA with IBD to produce HPA was studied. Generally, a low temperature could not meet the requirement of activation of aldol condensation; while a high temperature leads to undesired side reactions [36]. The reaction temperature was studied and the corresponding performances of catalysts are listed in Table S1. The selectivity of HPA in the presence of re-Ca₄Al_0.95Ga_0.05-LDHs shows an obvious change flowing the increase of reaction temperature: it rises from 55.2% (at 65 ℃) to the maximal value of 79.3% (at 70 ℃), followed by a sharp drop to 58.4% (at 75 ℃). Therefore, a reaction temperature of 70 ℃ was chosen. Figure S2 shows the HPA yield over these rehydrated LDHs samples along with reaction time. It is observed that all these catalysts reach reaction equilibrium at 8 h, and the performance of re-Ca₄Al_0.90Ga_0.10-LDHs is far superior to re-Ca₄Al-LDHs and re-Ca₄Al_1-xIn_x-LDHs. Additionally, the catalytic performance results evaluated under optimal conditions are shown in Fig. 3A. For re-Ca₄Al_1-xGa_x-LDHs catalysts, with the improve of Ga³⁺/M³⁺ atomic ratios from 0 to 0.15, the equilibrium conversion enhances obviously at first and then decreases slightly. For the HPA selectivity, a similar trend is found, and the maximal value (conversion: 90%; selectivity: 80.5%) is obtained on the re-Ca₄Al_0.90Ga_0.10-LDHs catalyst with the highest yield of 72%. In the case of re-Ca₄Al_1-xIn_x-LDHs catalysts, both the conversion and selectivity are below those of re-Ca₄Al_1-xGa_x-LDHs samples. In addition, the initial formation rate of HPA at 2% IBD conversion that reflects the intrinsic catalytic property is listed in Table 2. The results show that the formation rate of HPA follows a similar change tendency and re-Ca₄Al_0.90Ga_0.10-LDHs sample exhibits the maximal value (104 mmol g^-1h^-1). A comparison study for the catalytic performances of control samples including conventional solid base (e.g., Ca₄Al_1-xGa_x-LDHs, Ca₄Al_1-xIn_x-LDHs, Ca₄Al_1-xGa_x-MMOs, Ca₄Al_1-xIn_x-MMOs), and liquid alkali catalysts for this reaction were explored (Table S2). The results reveal that re-Ca₄Al_0.90Ga_0.10-LDHs catalyst exhibits a significantly improved HPA yield (72%) than that of LDHs precursors (45%) and conventional solid base catalysts (55%), even close to the level of liquid basic catalysts (73.5%). In addition, the stability test of re-Ca₄Al_0.90Ga_0.10-LDHs catalyst shows a decrease by 10% for conversion and selectivity within five cycles and then remains stable in the sixth cycle (Fig. 3B). XRD pattern of the used catalyst gives the presence of CaCO₃ phase (Fig. S3), which is the reason of partial deactivation.

To identify the nature of basic site, in situ DRIFTS spectra, with a suitable probe molecule (CDCl₃, a weak acid molecule) under mild conditions, were performed over these rehydrated samples [37]. As shown in Fig. 4, a wide FTIR band can be observed in the range 2180-2300 cm^-1, and further deconvoluted to two peaks by a Gaussian peak fitting approach in order to obtain the distribution of basic strength and semi-quantitative data. The two bands, with maximum at 2254 and 2214 cm^-1, are defined as the C-D stretching vibration of CDCl₃ adsorbed on the weak and medium basic sites, respectively. In addition, in Ca₄Al_0.85Ga_0.15-LDHs, the wide FTIR band can deconvoluted to four peaks by a Gaussian peak fitting approach, the two peaks at relatively high wavenumber (2275 and 2264 cm^-1) are assigned to the C-D stretching vibration of CDCl₃ gas and physically adsorbed CDCl₃, respectively. With the increase of Ga³⁺/M³⁺ atomic ratios from 0 to 0.15, the peak area of v_C-D on weak alkaline sites enhances obviously and re-Ca₄Al_0.90Ga_0.10-LDHs catalyst gives the maximum value. This indicates that the incorporation of Ga element enhances the concentration of 7-coordinated Ca-OH group (weak basic site). However, in the case of re-Ca₄Al_1-xIn_x-LDHs catalysts, the normalized peak area of weak basic site decreases significantly along with the increase of In content. Moreover, compared with the re-Ca₄Al-LDHs sample, no obvious band shift is observed for re-Ca₄Al_1-xGa_x and re-Ca₄Al_1-xGa_x system, indicating that the strength of alkaline sites remains unchanged with the introduction of Ga or In. On the other hand, the desorption energies of HPA from three kinds of samples were studied (Fig. S4) by DFT calculations. The results show that HPA desorption energy from re-Ca₄Al-LDHs (1.10 eV) is close to re-Ca₄Al_1-xGa_x-LDHs (0.94 eV) and re-Ca₄Al_1-xIn_x-LDHs (1.19 eV), in accordance with their close catalytic selectivity towards HPA. Our previous work has demonstrated that the weak basic site is assigned to the active 7-coordinated Ca-OH group and the medium basic site is considered as the physically-adsorbed surface OH^- group in the re-Ca₄Al-LDHs catalyst [13]. To give a deep insight into the relationship between the catalytic performance and the quantity of basic sites for these rehydrated samples, the initial formation rate of HPA vs. the relative concentration of weak or medium basic site is represented in Fig. 5A and 5B, respectively. A positive correlation between formation rate of HPA and the relative concentration of weak basic site is obtained, demonstrating that 7-coordinated Ca-OH site performs as active site for this reaction. In general, the incorporation of Ga leads to an increase in weak basic site in re-Ca₄Al-LDHs, accounting for its high catalytic activity. In contrast, the proportion of active site in re-Ca₄Al_1-xIn_x-LDHs decreases significantly, resulting in the activity decline toward aldol condensation.

XANES (Fig. 6A) and EXAFS spectroscopy (Fig. 6B) were used to study the detailed structure of these samples. We selected re-Ca₄Al_0.90Ga_0.10-LDHs, which showed the best performance, as the optimal catalyst; while re-Ca₄Al-LDHs and re-Ca₄Al_0.90In_0.10-LDHs were chosen as reference samples catalysts. As shown in Fig. 6A, compared with Ca(OH)₂ sample, these three catalysts manifest the pre-peak observed at 4043 eV corresponding to transitions state (1s to 3d), indicating a distorted 7-coordinated Ca²⁺-OH with a low structural symmetry existing in Ca-based LDHs [13, 38]. Moreover, according to the Ca K-edge EXAFS spectra (Fig. 6B), the bond length of Ca-O of re-Ca₄Al-LDHs increases from 1.719 Å to 1.769 Å for re-Ca₄Al_0.90Ga_0.10-LDHs and 1.781 Å for re-Ca₄Al_0.90In_0.10-LDHs, respectively, indicating a lattice expansion of re-Ca₄Al-LDHs with the incorporation of Ga or In. It is interesting that the Ca-O shell coordination number in re-Ca₄Al_0.90Ga_0.10-LDHs is larger than that of re-Ca₄Al sample; in contrast, the Ca-O shell coordination number in re-Ca₄Al_0.90In_0.10-LDHs is smaller than that of re-Ca₄Al sample. In order to explore this unusual phenomenon, we used In(OH)₃ as a reference sample to investigate the coordination number of In-O shell in re-Ca₄Al_0.90In_0.10-LDHs catalyst by using In L-edge EXAFS (Fig. 6C) and In 3d XPS characterization. Figure 6C shows that the coordination number of In-O shell increases obviously compared with In(OH)_3, suggesting the existence of highly coordinated In-OH group in re-Ca₄Al_0.90In_0.10-LDHs [39, 40]. Moreover, the binding energy of In in the XPS spectrum (Fig. S5) moves toward high energy relative to In(OH)₃ sample, indicating the existence of electron transfer in re-Ca₄Al_0.90In_0.10-LDHs catalyst. Therefore, according to results of EXAFS spectra and catalytic evaluation, Ga as a structural promoter introduced into the laminate causes a moderate lattice expansion, which results in a certain degree of distortion of the hydrotalcite laminates with more exposed active sites. This is in accordance with the results of FTIR, in which the introduction of Ga induces an increase in quantity of weak active sites (7-coordinated Ca-OH) and largely promoted IBD conversion. On the contrary, for re-Ca₄Al_0.90In_0.10-LDHs system, the IBD conversion decreases remarkably. Based on the Ca K-edge and In L-edge EXAFS analysis, the introduction of In with a large ionic radius prefers to induce a significant expansion of the lattice framework, which leads to a high In-OH coordination and a decreased number of 7-coordinated Ca-OH group. Therefore, a depressed catalytic activity is obtained in the re-Ca₄Al_1-xIn_x-LDHs system, which is consistent with the CDCl₃-FTIR and catalytic evaluation results.

X-ray photoelectron spectroscopy (XPS) was carried out to investigate the electronic structure of O species. For all samples, the broad asymmetric O 1s peak can be deconvoluted into three peaks by the Gaussian peak fitting approach at 532.4, 531.4 and 530.4 eV, which are attributed to the surface OH^- in 7-coordinated Ca-OH group, lattice oxygen species of Ca-O-Al in Ca(OH)₆ octahedron, and defective O^2- species with low coordination, respectively [41, 42]. Figures 6D and S6 show that the binding energy of OH^- oxygen species in these samples gives inconspicuous change (between 532.5 eV and 532.3 eV), indicating that the incorporation of Ga or In has little impact on the electron density of oxygen in LDHs catalysts. Moreover, Bader charge analysis by DFT calculation shows that 7-coordinated OH^- in three samples possesses close electron density (Fig. S7: Ca₄Al-LDHs, 1.45 eV; re-Ca₄Al_1-xGa_x-LDHs, 1.41 eV; re-Ca₄Al_1-xIn_x-LDHs, 1.47 eV), which is in agreement with the results obtained from XPS. This does not induce significant change in the strength of basic sites, consistent with the CDCl₃-FTIR result.

To further understand the geometric effect of Ga or In on rehydrated Ca₄Al-LDHs catalysts, DFT calculations were performed to investigate detailed structural information of these three samples, and the lattice parameters of crystals are listed in Table S3. According to the geometric optimized results, the Ca-O bond length in Ca-O-Ga octahedron for Ca₄Al_0.90Ga_0.10-LDHs (Fig. 7A₂) is 2.348 Å while that in Ca-O-In system (Fig. 7A₃) is 2.417 Å, both of which are larger than that in Ca-O-Al structure (Fig. 7A₁, 2.315 Å) for re-Ca₄Al-LDHs. This indicates that Ga or In atoms introduced into Ca₄Al-LDHs laminate cause a lattice expansion, which is consist with Ca K-edge EXAFS results. From CDCl₃-FTIR spectra, we concluded that adding metal elements with different ionic radius has different effects on the quantity of active sites. Figure 7B₁ shows the normal Ca-Al hydrotalcites model with a 7-coordinated Ca-OH group (weak active basic site). Interestingly, with the incorporation of Ga into the Ca-Al hydrotalcites, the number of 7-coordinated Ca-OH group increases, as shown in Fig. 7B₂ (i.e., more weak basic sites are formed with the insertion of Ga atoms), which is agreement with the FTIR results. However, with the incorporation of In atoms, the number of 7-coordinated Ca-OH group decreases accompanied with the formation of a new highly coordinated In-OH group (Fig. 7B₃). According to the result of EXAFS and FTIR, a reasonable explanation is proposed. The incorporation of Ga or In atoms with larger ion radius induces a lattice expansion of hydrotalcites laminates, which is attributed to geometric effect. A moderate lattice expansion with the introduction of Ga atoms with appropriate ion radius (0.47 Å) results in an enhancement of weak basic sites, accounting for the improved catalytic activity. However, the incorporation of In (ion radius: 0.79 Å) causes an excessive lattice expansion by a stronger geometric effect, accompanied with the formation of undesirable In-OH with medium basic sites. Therefore, the introduction of a cation as a structure promoter with appropriate ionic radius is essential to achieve optimized catalytic performance towards aldol condensation between IBD and FA.

In summary, rehydrated Ca₄Al_1-xGa_x-LDHs and Ca₄Al_1-xIn_x-LDHs were fabricated by virtue of the structural topotactic transformation and memory effect of LDHs precursors. The obtained re-Ca₄Al_0.90Ga_0.10-LDHs exhibits the best catalytic activity toward the aldol condensation reaction that isobutyraldehyde (IBD) reacts with formaldehyde (FA) to produce hydroxypivalaldehyde (HPA) (HPA yield: 72%), which is comparable to the level of liquid basic catalysts (HPA yield: 73.5%). A combination study including in situ XANES and CDCl₃-FTIR spectra reveals that incorporation of Ga with moderate ionic radii leads to the lattice expansion of hydrotalcite laminates (geometric effect), which significantly influences the concentration of weak basic sites. Studies on the structure-property correlation indicate that the optimal re-Ca₄Al_0.90Ga_0.10-LDHs catalyst possesses the most abundant 7-coordinated Ca-OH group (weak basic site), accounting for its excellent catalytic activity. This work systematically investigates the geometric effects of doping elements with different ionic radii on the structure of Ca-based hydrotalcites, which can be used as a prospective candidate in green catalysis of aldol condensation reactions.