Designable assembly of atomically precise Al₄O₄ cubane supported mesoporous heterometallic architectures

Abstract

Heterometallic cluster-based framework materials are of interest in terms of both their porous structures and multi-metallic reactivity. However, such materials have not yet been extensively investigated because of difficulties in their synthesis and structural characterization. Herein, we reported the designable synthesis of atomically precise heterometallic cluster-based framework compounds and their application as catalysts in aldol reactions. By using the synergistic coordination protocol, we successfully isolated a broad range of compounds with the general formula, [Al₄M₄O₄(L)₁₂(DABCO)₂] (L = carboxylates; DABCO = 1,4-diazabicyclo[2.2.2]-octane; M²⁺ = Co²⁺, Mn²⁺, Zn²⁺, Fe²⁺, Cd²⁺). The basic heterometallic building blocks contain unprecedented main-group γ-alumina moieties and surrounding unsaturated transition metal centers. Interestingly, the porosity and interpenetration of these frameworks can be rationally regulated through the unprecedented strategy of increment of the metal radius in addition to general introduction of sterically bulky groups on the ligand. Furthermore, these porous materials are effective catalysts for aldol reactions. This work provides a catalytic molecular model platform with accurate molecular bonding between the supporters and catalytically active metal ions.

---

Introduction

Aluminosilicate zeolites with 4-connected metal centers are a class of inorganic porous solids with a range of applications in catalysis, ion exchange and separation.^1–6 Compared with these traditional 4-connected nodes,^7–18 the use of polynuclear metal clusters as 4-connected nodes is highly desirable because of the availability of large pore sizes and the possibility of combining unique cluster-based properties.^19–22 Such an idea has been realized by Feng et al.^22,23via the assembly of supertetrahedral metal–chalcogenide clusters. Most significantly, elements without tetrahedral geometry can be incorporated, thus providing access to multifunctional framework materials with zeolite type topology.^24–26

Heterometallic oxo clusters may exhibit unique properties resulting from the synergy effect of the two substantially different types of metal ions. Therefore, it is of great interest to incorporate heterometallic clusters as 4-connected centers. However, extensive studies based on 4-connected heterometallic clusters remain scarce because of the lack of a strategy for efficient synthesis. There are two possible synthetic challenges: (1) competitive coordination and integration of two kinds of metal ions, and (2) metal clusters with high nuclearity and positive charge are prone to accepting more organic ligands, thus leading to highly-connected dense topologies. Once the building blocks are established, effective interpenetration regulation is another issue that should be considered.^27,28 In 2019, Zheng et al.²⁹ summarized advances in the design and synthesis of cluster organic frameworks based on 4-connected metal–halide clusters,³⁰ metal–oxygen clusters,³¹ and metal–chalcogen clusters.³² In the domain of metal–oxygen clusters, only transition metal clusters (such as ε-Zn₄PMo₁₂ ²⁶ or Cd₄Cu₆ ³³) have been reported as 4-connected nodes. However, main group element-based examples are challenging and still lacking.

γ-Al₂O₃ is one of the most common kinds of alumina, which has a large surface area, and strong adsorption capacity and has been widely used in adsorbents, catalysts and catalyst supports.³⁴ With our recent research in Al(III) chemistry, we herein demonstrate a synergistic coordination strategy towards the assembly of 4-connected heterometallic cluster-based framework materials. The strategy is based on spontaneous selective coordination between two kinds of metal centers (Al centers and d-transition metals) and two types of ligands (chelating L with O-donors and bridging DABCO with N donors, respectively) (Scheme 1a).³⁵ Such an approach is universal and we successfully isolate a broad range of compounds [Al₄O₄M₄(L)₁₂(DABCO)₂] (denoted as AlOCs; M²⁺ = Mn²⁺, Fe²⁺, Co²⁺, Zn²⁺, Cd²⁺, L = carboxylates; DABCO = 1,4-diazabicyclo[2.2.2]octane) (Table 1). We were surprised to find that the Al₄O₄ cubane (denoted as the γ-alumina moiety) is located in the center of the 4-connected nodes (Scheme 1b).^36,37 In addition, both the inorganic and organic shells of the Al₄O₄ cubane can be modified to a large extent. Interestingly, controllable interpenetration is realized through the introduction of sterically bulky groups, increment of the ionic radius of transition metal ions, and variation of the M²⁺/Al³⁺ concentration. Furthermore, these porous materials are effective catalysts for aldol reactions.

Scheme 1 (a) Scheme demonstrating the synergistic coordination synthetic strategy; (b) comparison between the structures of γ-Al₂O₃ and the Al₄O₄ cubane in a tetrahedral cluster node.

Table 1 A summary of the 4-connected AlOCs^a

Complex	Composition	Sp. Gr.	R 1 Value	Void ratio
a Abbreviations: BA = benzoate; 2-FBA = 2-fluorobenzoate; 3-FBA = 3-fluorobenzoate; 4-FBA = 4-fluorobenzoate; 2-FA = 2-furan formate; 2-TA = 2-thiophene formate; 3-TA = 3-thiophene formate; 3-MBA = 3-methylbenzoate; 4-MBA = 4-methylbenzoate; 4-CBA = 4-chlorobenzoate; 4-ABA = 4-aminobenzoate.
Two-fold diamond (dia) interpenetration
AlOC-99	Al4Co4O4(BA)12(DABCO)2	Fd3̄c	0.0801	1.2%
AlOC-100	Al4Co4O4(2-FBA)12(DABCO)2	Fd3̄c	0.0759	No
AlOC-101	Al4Co4O4(3-FBA)12(DABCO)2	Fd3̄c	0.1149	7.6%
AlOC-102	Al4Co4O4(4-FBA)12(DABCO)2	Fd3̄c	0.1474	2.5%
AlOC-103	Al4Co4O4(2-FA)12(DABCO)2	Fd3̄c	0.0998	2.8%
AlOC-104	Al4Co4O4(2-TA)12(DABCO)2	Fd3̄c	0.1447	1.6%
AlOC-105	Al4Co4O4(3-TA)12(DABCO)2	Fd3̄c	0.1274	2.4%
AlOC-106	Al4Mn4O4(BA)12(DABCO)2	Fd3̄c	0.0858	No
AlOC-107	Al4Fe4O4(BA)12(DABCO)2	Fd3̄c	0.0566	1.0%
AlOC-108	Al4Zn4O4(BA)12(DABCO)2	Fd3̄c	0.0548	1.0%
Non-interpenetrated diamond (dia) topology
AlOC-109	Al4Co4O4(3-MBA)12(DABCO)2	F4132	0.0736	42.5%
AlOC-110	Al4Fe4O4(BA)12(DABCO)2	F4132	0.0634	52.8%
AlOC-111	Al4Cd4O4(BA)12(DABCO)2	F4132	0.0751	55.6%
Non-interpenetrated lonsdaleite (lon) topology
AlOC-112	Al4Co4O4(BA)12(DABCO)2	P6̄2c	0.0629	53.9%
AlOC-113	Al4Co4O4(3-FBA)12(DABCO)2	P6̄2c	0.0809	47.0%
AlOC-114	Al4Co4O4(4-CBA)12(DABCO)2	P6̄2c	0.0757	44.9%
AlOC-115	Al4Co4O4(4-MBA)12(DABCO)2	P6̄2c	0.0736	44.3%
AlOC-116	Al4Co4O4(4-ABA)12(DABCO)2	P6̄2c	0.0972	48.4%
AlOC-117	Al4Mn4O4(BA)12(DABCO)2	P6̄2c	0.0576	54.3%
AlOC-118	Al4Fe4O4(BA)12(DABCO)2	P6̄2c	0.1081	53.3%
AlOC-119	Al4Zn4O4(BA)12(DABCO)2	P6̄2c	0.0542	53.5%
AlOC-120	Al4Cd4O4(BA)12(DABCO)2	P6̄2c	0.0784	56.4%

---

Experimental

Materials and general methods

All the reagents and solvents employed were commercially available and used without further purification. Al(OⁱPr)₃ was acquired from Aladdin Chemical Reagent Shanghai. Fe(Ac)₂ was obtained from TCI. Acetonitrile (99%), benzoic acid, Co(Ac)₂·4H₂O, MnCl₂·4H₂O, Zn(Ac)₂·2H₂O and Cd(Ac)₂·2H₂O were bought from Sinopharm Chemical Reagent Beijing. Ti(OⁱPr)₄ and other carboxylic acids were purchased from Adamas-beta.

The energy dispersive spectroscopy (EDS) analyses of single crystals were performed on a JEOL JSM6700F field-emission scanning electron microscope equipped with an Oxford INCA system. IR spectra (KBr pellets) were recorded on an ABB Bomem MB102 spectrometer over a range of 400–3900 cm⁻¹. Powder X-ray diffraction (PXRD) data were collected on a Rigaku Mini Flex II diffractometer using CuKα radiation (λ = 1.54056 Å) under ambient conditions. The UV-vis diffuse reflection data were recorded at room temperature using a powder sample with BaSO₄ as a standard (100% reflectance) on a PerkinElmer Lamda-950 UV spectrophotometer and scanned at 200–1200 nm. The absorption data were calculated from the Kubelka–Munk function, (F(R) = (1 − R)²/2R), where R represents the reflectance.³⁸ Surface chemical analyses were performed by X-ray photoelectron spectroscopy (XPS, Thermo Fisher, ESCALAB 250Xi). Metal proportional analyses were performed on an Ultima-2 inductively coupled plasma (ICP) spectrometer. The TGA curves were recorded in the region of 30–800 °C at a heating rate of 10 °C min⁻¹ in a flowing N₂ atmosphere on a Mettler Toledo TGA/SDTA 851^e analyzer.

Synthesis of two-fold interpenetrated diamond (dia) frameworks AlOC-99–AlOC-108

Al(OⁱPr)₃ (0.918 g, 4.5 mmol), Co(Ac)₂·4H₂O (0.75–2.5 g, 3–10 mmol), DABCO (0.224 g, 1 mmol) and benzoic acid (0.978 g, 8 mmol) were dissolved in acetonitrile (10 mL), and then Ti(OⁱPr)₄ (0.5 mL, 1.5 mmol) was added to the bottle and mixed at room temperature. The resultant solution was sealed in a Teflon-lined autoclave (23 mL) and heated at 160 °C for three days. After being cooled to room temperature, red octahedral crystals of AlOC-99 were obtained (maximum yield: ∼64% (1.5 g) for AlOC-99 based on Al(OⁱPr)₃. Synthesis of AlOC-100–AlOC-105 is similar to that of AlOC-99 except that benzoic acid was replaced by 2-FBA, 3-FBA, 4-FBA, 2-FA, 2-TA, and 3-TA. Red octahedral crystals were collected after about three days (maximum yield: ∼10% (260 mg), ∼7% (180 mg), ∼11% (290 mg), ∼1.5% (30 mg), ∼1.6% (40 mg), and ∼1.2% (30 mg) based on Al(OⁱPr)₃ for AlOC-100–AlOC-105, respectively). Synthesis of AlOC-106–AlOC-108 is similar to that of AlOC-99 except that Co(Ac)₂·4H₂O was replaced by MnCl₂·4H₂O, Fe(Ac)₂ and Cd(Ac)₂·2H₂O. Yellow or colorless octahedral crystals were collected after about three days (maximum yield: ∼19% (450 mg), ∼1.9% (50 mg), and ∼4.6% (110 mg) for AlOC-106–AlOC-108.

Synthesis of non-interpenetrated diamond (dia) frameworks AlOC-109–AlOC-111

Synthesis of AlOC-109–AlOC-111 is similar to that of AlOC-99 except that benzoic acid was replaced by 3-methylbenzoic acid or Co(Ac)₂·4H₂O was replaced by Fe(Ac)₂ and Cd(Ac)₂·2H₂O. Red (AlOC-109), yellow (AlOC-110) or colorless (AlOC-111) octahedral crystals were collected after about three days (maximum yield: ∼20.5% (520 mg), ∼3.4% (80 mg) and ∼1.16% (30 mg) for AlOC-109–AlOC-111 based on Al(OⁱPr)₃).

Synthesis of non-interpenetrated lonsdaleite (lon) frameworks AlOC-112–AlOC-120

The synthetic procedure of AlOC-112 is similar to that of AlOC-99, except that the amount of Co(Ac)₂·4H₂O was reduced to 1–2 mmol. Finally, red hexagonal flaky crystals of AlOC-112 were obtained (maximum yield: ∼34% (800 mg) for AlOC-112 based on Al(OⁱPr)₃). Synthesis of AlOC-113–AlOC-116 is similar to that of AlOC-112 except that benzoic acid was replaced by 3-FBA, 4-CBA, 4-MBA and 4-ABA. Red hexagonal flaky crystals were collected after about three days (maximum yield: ∼23% (600 mg), ∼18% (500 mg), ∼32% (800 mg), and ∼8.2% (210 mg) for AlOC-113–AlOC-116 based on Al(OⁱPr)₃). Synthesis of AlOC-117–AlOC-120 is similar to that of AlOC-112 except that Co(Ac)₂·4H₂O was replaced by MnCl₂·4H₂O, Fe(Ac)₂, Zn(Ac)₂·2H₂O and Cd(Ac)₂·2H₂O. Yellow or colorless hexagonal flaky crystals were collected after about three days (maximum yield: ∼51.5% (1.2 g), ∼1.7% (40 mg), ∼12.6% (300 mg) and ∼2.2% (60 mg) for AlOC-117–AlOC-120 based on Al(OⁱPr)₃).

Caution: The synthesis of AlOC-107 should be performed in a glovebox in a vacuum environment and nitrogen should be pumped into the reactor for removing the soluble oxygen from the solvent. Yellow crystals of freshly prepared Fe-containing compounds (AlOC-107, AlOC-110 and AlOC-118) can gradually turn red due to the rapid oxidation when exposed to air.

General methods for X-ray crystallography

Crystallographic data of AlOC-106 and AlOC-119 were collected on a Mercury single crystal diffractometer equipped with graphite-monochromatic Mo Kα radiation (λ = 0.71073 Å). Compounds AlOC-117 and AlOC-118 were collected on a Supernova single crystal diffractometer equipped with graphite-monochromated Cu Kα radiation (λ = 1.54184 Å). Other compounds (AlOC-99–AlOC-105, AlOC-107–AlOC-116, and AlOC-120) were collected on a Hybrid Pixel Array detector equipped with graphite-monochromated Ga Kα radiation (λ = 1.3405 Å). The structures were solved with direct methods using OLEX² and refined by full-matrix least-squares on F² using SHELXTL.³⁹ Contributions to scattering due to disordered solvent molecules were removed using the SQUEEZE routine of PLATON. All hydrogen atoms were theoretically positioned, riding on the concerned atoms and refined with fixed thermal factors. Non-hydrogen atoms were refined anisotropically.

Preparation of active samples for gas adsorption

Crystals were soaked in ethanol and heated at 80 °C for one week. During this time, the ethanol solution should be exchanged several times a day. Subsequently, the sample was degassed under a dynamic vacuum at 100 °C for 6 h to activate the sample. The gas adsorption isotherms of active samples were obtained on a Micromeritics ASAP 2020 volumetric adsorption instrument.

Results and discussion

The tunable shell of Al₄O₄ supported supertetrahedral nodes

The prominent feature of these compounds is the presence of a central Al₄O₄ γ-alumina moiety. The accurate crystal structures of Al₄N₄,⁴⁰ Al₄P₄, Al₄S₄ ⁴¹ and Al₄Se₄ ⁴² have been reported, while the Al₄O₄ cubane has never been reported and has only been studied computationally.⁴³ However, the Al₄O₄ cubane can act as a subunit that exists in high-nuclearity AlOCs (Al₃₀, Cu₂Al₃₀-S, and Zn₂Al₃₂).^44–46 In this work, the Al₄O₄ cubane serves as a cornerstone in supertetrahedron nodes, where both inorganic (M²⁺ = Co, Mn, Zn, Fe, Cd) and organic (a broad range of chelated aromatic carboxylates) compositions are tunable (Fig. 1). And as many as 22 isostructural compounds were successfully isolated (Table 1, AlOC-99–AlOC-120, Fig. S1–S22†). Each Al atom in all compounds adopts a six-coordination geometry, bonding to three μ₄-O atoms and three carboxylate oxygen atoms (O_COO) (Fig. S23†). Every external M ion resided in a trigonal bipyramid environment with three O_COO atoms, one μ₄-O bridge and one N atom from DABCO. Each triangular face of the supertetrahedral core is consolidated by three bridging bidentate chelating ligands (Fig. S24 and S25†). Compared with the charged tetrahedral cages,^47–49 the supertetrahedral cluster [Al₄Co₄O₄(L)₁₂] herein is neutral. Four DABCO linkers on the cluster node allow further connections, which is similar to the 4-connected TO₄ (T = Si, Al, P, etc.) node in inorganic zeolite materials.

Fig. 1 The tunable inorganic metal ion shell and organic chelated ligand shell on tetrahedral cluster nodes. The atom color code: O, red; C, dark grey and black; N, blue; Al, green. Hydrogen atoms are omitted for clarity.

Interpenetration regulation

As shown in Fig. 2, the 4-connected frameworks are categorized into three types. [Al₄M₄O₄(L)₁₂(DABCO)₂] undergoes a change in the degree of interpenetration from a dense doubly interpenetrated framework (AlOC-99–AlOC-108, Fig. S1–S10†) to two kinds of highly porous non-interpenetrated framework with diamond (dia) topology: AlOC-109–AlOC-111, Fig. S11–S13;† and lonsdaleite (lon) topology (AlOC-112–AlOC-120, Fig. S14–S22†). It is reported that factors that affect interpenetration include pressure, temperature, solvent, ligand design, etc.^27,50,51 Herein, precise suppression of interpenetration by an increment of the metal radius is found in addition to the ligand design and concentration ratio of M²⁺/Al³⁺.

Fig. 2 The interpenetration regulation strategies: introduction of sterically bulky groups, increment of the ionic radius and decrease of the M²⁺/Al³⁺ concentration. The mesoporous cavities (a, c, e, and g) and overall frameworks (b, d, f, and h) in two-fold interpenetrated AlOC-99, and non-interpenetrated AlOC-109, AlOC-111 and AlOC-112, respectively. The atom color code: O, red; C, dark grey; N, blue; Co, pink; Cd, turquoise; Al, green. Hydrogen atoms are omitted for clarity. The diamond (dia) frameworks present rhombic windows, while the lonsdaleite (lon) frameworks possess hexagonal channels (diameter: 3.761 nm; interior wall diameter: 0.909 nm) aligned parallel to the c-axis.

Interpenetration regulation through ligand design. The regulation of the topology is greatly related to the types of ligands. For benzoate or fluorobenzoate, both interpenetrated and non-interpenetrated frameworks can be obtained (interpenetrated diamond (dia) topology: AlOC-99–AlOC-102; non-interpenetrated lonsdaleite (lon) topology: AlOC-112 and AlOC-113). However, if large steric groups (–Cl, –CH₃ and –NH₂) are introduced, only non-interpenetrated frameworks can be generated (AlOC-109 and AlOC-114–AlOC-116) (Fig. 2c, d and S11, S16–S18†). Intriguingly, the presence of the substituents can also enhance the intermolecular and intramolecular hydrogen bond interactions (Fig. S26†). For smaller pentacyclic carboxylates, only doubly interpenetrated frameworks (AlOC-103–AlOC-105) were synthesized (Fig. S5–S7†).

Interpenetration regulation through increasing the ionic radius of TM ions. Cd-containing frameworks (AlOC-111 and AlOC-120) present non-interpenetrated topologies (Fig. 2e, f and S13, S22†), which are different from those of Mn, Fe, Co, and Zn involving compounds. A possible reason is attributed to its larger radius, and the size of Cd₄Al₄O₄ tetrahedral clusters is obviously larger than that of other ones (Fig. S27†). Although a large number of factors influencing interpenetration have been reported, the effect of the ionic radius was discovered for the first time.

Interpenetration regulation through altering the ratio of M²⁺/Al³⁺. The variation of the mole ratio of raw materials has a great impact on the structural types.^8,50 Herein, the M²⁺/Al³⁺ concentration plays an important role in interpenetration regulation. For example, when the ratio of Co²⁺/Al³⁺ is greater than 1.11, the pure octahedral crystals with a doubly interpenetrated diamond (dia) framework can be obtained (Fig. 2a and b, AlOC-99, Fd3̄c; Fig. 3a, left), in which the 6⁶ diamond (dia) cage is exclusively made up of six chair-like six-membered rings (Fig. 3b, left). When the ratio of Co²⁺/Al³⁺ is lower than 0.44, only pure hexagonal flaky crystals with non-interpenetrated lonsdaleite (lon) topologies can be generated (Fig. 2g, h and S14,†AlOC-112, P6̄2c; Fig. 3a, right). Different from the 6⁶ diamond (dia) cage, the 6⁵ lonsdaleite (lon) cage in AlOC-112 is composed of chair-like and boat-like six-membered rings (Fig. 3b, right). However, when the ratio of Co²⁺/Al³⁺ is controlled between 0.44 and 1.11, the product was a mixture phase containing both octahedral crystals and hexagonal flaky crystals (Fig. 3a, middle).

Fig. 3 (a) Photos of reaction products obtained from controlling the ratio of Co²⁺/Al³⁺. (b) The simplified diamond (dia) and lonsdaleite (lon) cages. Chair-like (left) and boat-like (right) six-membered rings have been highlighted with yellow bonds.

Therefore, it is reasonable to infer that the low M²⁺/Al³⁺ ratio favored the non-interpenetrated lonsdaleite (lon) form,⁵² which is supported by the experimental results that come from the isolation of a series of isostructural Mn, Fe, Zn and Cd constructed AlOCs (AlOC-117–AlOC-120; Table 1, Fig. S19–S22†). For Fe constructed frameworks, the control of the Fe²⁺/Al³⁺ ratio is not easy due to the rapid oxidation of Fe²⁺. Thus, the synthesis of the doubly interpenetrated diamond (dia) framework of AlOC-107 with a high Fe²⁺/Al³⁺ ratio should be performed in a glovebox under pumped nitrogen. Instead, non-interpenetrated dia framework AlOC-110 will be generated when synthesized under the same conditions without nitrogen.

Physical performance

The non-interpenetrated framework density (for example, AlOC-112 (0.823 g cm⁻³)) is approximately half that of doubly interpenetrated AlOC-99 (1.651 g cm⁻³). The formation of a non-interpenetrated framework is noteworthy given the mesoporous cavity channels (ca. 2.2–2.3 nm) (Fig. 2c, e, g and S30†). Correspondingly, it also exhibits obvious higher solvent-accessible volumes (e.g., 53.9% for AlOC-112 calculated using PLATON).⁵³ Considering their potential application in gas adsorption and separation, N₂ sorption tests were conducted to evaluate porosity. Based on N₂ sorption isotherms, the BET surface areas were calculated to be 11.48, 779.6 and 797.2 m² g⁻¹ for AlOC-99, AlOC-109 and AlOC-112 (Fig. 4a and S31†). The pronounced H4 hysteresis loops of AlOC-109 and AlOC-112 are located at high P/P₀ (P/P₀ = 0.4–1.0), further confirming the presence of the mesopores in the materials.^54,55 The experimental CO₂ uptake of AlOC-99, AlOC-109 and AlOC-112 was measured to be 10.2, 13.9 and 21.7 cm³ g⁻¹ at 273 K.

The stability of frameworks plays an important role in research and industrial production, because it provides an applicable scope of the material. Their stability is related to interpenetration and topology. For example, the doubly interpenetrated framework is thermodynamically stable when heated to 300 °C, while the non-interpenetrated ones can only tolerate 250 °C (Fig. S32 and S33†), indicating that interpenetration in AlOC-99 obviously enhances the stability of the framework. In addition, AlOC-99 and AlOC-109 were stable after being immersed in an aqueous solution with different pH values (Fig. S34 and S35†). The lonsdaleite (lon) framework of AlOC-112 will collapse gradually in water, indicating that the diamond (dia) type framework is more stable than that of lonsdaleite (lon). Nevertheless, both of them were stable in common organic solvents, including MeOH, EtOH, MeCN, DMF and DMSO (Fig. S36†). Their solution stabilities are related to hydrophobic wettability. And the measured contact angles for AlOC-99, AlOC-109 and AlOC-112 are 160.58°, 140.13° and 153.23°, respectively (Fig. S37 and S38†).⁵⁶ Furthermore, all three materials can remain in the air for more than one year (Fig. S39†), which is related to the presence of the Al₄O₄ core because compounds like Co₂Bdc₂Dabco without aluminium (Fig. S40†) cannot be exposed to air for more than two weeks.^57,58

These compounds were all characterized by powder X-ray diffraction (PXRD), IR spectroscopy, and energy dispersive spectroscopy (EDS) analyses (Fig. S41–S68†). The atomic ratio of M to Al was determined to be approximately 1 : 1 by inductively coupled plasma (ICP) analysis, which is in agreement with the crystallographic results. The valence states of the metal ions were confirmed by bond-valence-sum (BVS) calculations and X-ray photoelectron spectroscopy (XPS) (Table S5, Fig. S69–S90†). Intriguingly, these compounds present a variety of colors (Fig. 4b), that are affected by several factors. For example, Fe-containing compounds are easily oxidized, and their colors can be changed from yellow to red when exposed to air (Fig. S91†). The colors of Co constructed frameworks are related to the framework density and chelated ligands. Doubly interpenetrated AlOC-99 presents darker color when compared with non-interpenetrated AlOC-112. Furthermore, the colors of 4-FBA coordinated frameworks (e.g., AlOC-102) are also different from that of BA modified structures due to the electron-withdrawing effects of F centers (e.g., AlOC-99). Diffuse reflectance spectroscopy measurements were performed on all complexes to better understand their optical absorption properties (Fig. S92–S95†).

Fig. 4 (a) The N₂ sorption isotherms of AlOC-99, AlOC-109 and AlOC-112. (b) Crystal photos of representative samples.

Catalytic performance

The aldol reaction is a basic transformation in creating new carbon–carbon bonds and β-hydroxyl carbonyl compounds.⁵⁹ Considering the higher stability and catalytically active sites (unsaturated transition metals) on the Al₄O₄ support, three representative AlOCs (AlOC-99, AlOC-109 and AlOC-112) were used as catalysts for the direct aldol reaction of acetone and nitro-substituted aromatic aldehyde in DMSO. The catalytic results are summarized in Table S6.†

In the presence of 10 mmol% catalysts, all three AlOCs show high catalytic activities giving desired aldol products in 79–88% yield. Upon completion of the reaction, all of them could be readily recovered by centrifugation and reused for the next cycle without significant loss of activity. The yield after three consecutive runs of AlOC-109 can still remain 88% for isolated products. PXRD exhibited that the recycled sample retained high crystallinity after catalysis (Fig. S96†). In order to explore the impact of penetration on performance, the amount of catalyst is reduced to 5 mmol%, and mesoporous frameworks afford higher yield (67% for AlOC-109, and 68% for AlOC-112) compared with doubly interpenetrated AlOC-99 (48%). This suggests that the catalytic reactions with a low content catalyst are related to the specific surface area. Besides, the control experiment without catalyst loading was also performed, and no condensation product can be observed, indicating that these crystals possess catalytic capability.

The probable catalytic mechanism is shown in Fig. 5.⁶⁰ First, the Co centers with unsaturated coordination sites polarize the carbonyl group of acetone, leading to the dissociation of the α-proton and generating a metal enolate. The oxygen atoms on BAs bond to the α-proton and leave the Co center transitorily. Then, 4-nitrobenzaldehyde combined with the metal enolate through C–C coupling, and condensation products can be obtained.

Fig. 5 Probable catalytic mechanism for the aldol condensation reaction by using AlOC-99, AlOC-109 and AlOC-112 as catalysts.

Conclusions

This is not only the first time that the γ-alumina moiety is found in crystalline porous compounds but it also serves as a support for loading various metal ions. The incorporation of heterometallic ions broadens the scope and potential applications of framework materials. This work provides a typical catalytic molecular model and demonstrates the unambiguous connections between the support (γ-alumina moiety) and catalytically active metal ions at the atomic level. Such γ-alumina moiety-supported materials can provide a reference for the catalytic system to achieve the hetero-disperse requirement meanwhile exposing as many catalytic sites as possible.

Data availability

All data supporting this study are available from article and ESI.† The raw data files are available from author upon reasonable request.

Author contributions

All authors contributed extensively to the work presented in this paper. J. Zhang and W.-H. Fang conceived the research project. Y.-J. Liu performed the synthesis and characterization experiments. Y. Yu performed catalytic experiments. Y.-F. Sun assisted in the data collection. W.-H. Fang and Y.-J. Liu wrote the manuscript and the ESI with input from the other authors.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Key Research & Development Program of China (2021YFA1501500), National Natural Science Foundation of China (92061104, 21935010 and 21973096), Natural Science Foundation of Fujian Province (2021J06035) and Youth Innovation Promotion Association CAS (2021081).

Notes and references

1. P. Feng, X. Bu and G. D. Stucky, Nature, 1997, 388, 735.
2. R. Banerjee, H. Furukawa, D. Britt, C. Knobler, M. O'Keeffe and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 3875.
3. R. Simancas, D. Dari, N. Velamazán, M. T. Navarro, A. Cantín, J. L. Jordá, G. Sastre, A. Corma and F. Rey, Science, 2010, 330, 1219.
4. Z. Wang, J. Yu and R. Xu, Chem. Soc. Rev., 2012, 41, 1729.
5. J. Cejka, A. Corma and S. Zones, Zeolites and Catalysis. Synthesis, Reactions and Applications, Wiley, Weinheim, 2010.
6. C. Sun, Z. Liu, S. Wang, H. Pang, R. Bai, Q. Wang, W. Chen, A. Zheng, W. Yan and J. Yu, CCS Chem., 2021, 3, 189.
7. S.-T. Zheng, T. Wu, F. Zuo, C. Chou, P. Feng and X. Bu, J. Am. Chem. Soc., 2012, 134, 1934.
8. E. E. Knyazeva and I. I. Ivanova, Pet. Chem., 2019, 59, 262.
9. X. Luo, Y. Cao, T. Wang, G. Li, J. Li, Y. Yang, Z. Xu, J. Zhang, Q. Huo, Y. Liu and M. Eddaoudi, J. Am. Chem. Soc., 2016, 138, 786.
10. H. Wang, L. Han, D. Zheng, M. Yang, Y. H. Andaloussi, P. Cheng, Z. Zhang, S. Ma, M. J. Zaworotko, Y. Feng and Y. Chen, Angew. Chem., Int. Ed., 2020, 59, 6263.
11. K. Williams, L. Meng, S. Lee, L. Lux, W. Gao and S. Ma, Inorg. Chem. Front., 2016, 3, 393.
12. S.-Y. Zhang, Z.-Y. Wang, J. Gao, K. Wang, E. Gianolio, S. Aime, W. Shi, Z. Zhou, P. Cheng and M. J. Zaworotko, Chem, 2019, 5, 1609.
13. X.-C. Huang, Y.-Y. Lin, J.-P. Zhang and X.-M. Chen, Angew. Chem., Int. Ed., 2006, 45, 1557.
14. F. Wang, Z.-S. Liu, H. Yang, Y.-X. Tan and J. Zhang, Angew. Chem., Int. Ed., 2011, 50, 450.
15. K. S. Park, Z. Ni, A. P. Côté, J. Y. Choi, R. Huang, F. J. Uribe-Romo, H. K. Chae, M. O'Keeffe and O. M. Yaghi, Proc. Natl. Acad. Sci. U.S.A., 2006, 103, 10186.
16. N. T. T. Nguyen, H. Furukawa, F. Gándara, H. T. Nguyen, K. E. Cordova and O. M. Yaghi, Angew. Chem., Int. Ed., 2014, 53, 10645.
17. J. Zhang, T. Wu, C. Zhou, S. M. Chen, P. Y. Feng and X. H. Bu, Angew. Chem., Int. Ed., 2009, 48, 2542.
18. H.-X. Zhang, F. Wang, H. Yang, Y.-X. Tan, J. Zhang and X. Bu, J. Am. Chem. Soc., 2011, 133, 11884.
19. Y. Kang, F. Wang, J. Zhang and X. Bu, J. Am. Chem. Soc., 2012, 134, 17881.
20. H. Li, A. Laine, M. O'Keeffe and O. M. Yaghi, Science, 1999, 283, 1145.
21. P. Feng, X. Bu and N. Zheng, Acc. Chem. Res., 2005, 38, 293.
22. J. Zhang, X. Bu, P. Feng and T. Wu, Acc. Chem. Res., 2020, 53, 2261.
23. N. F. Zheng, X. H. Bu and P. Y. Feng, Nature, 2003, 426, 428.
24. K. S. Lokare, N. Frank, B. Braun-Cula, I. Goikoetxea, J. Sauer and C. Limberg, Angew. Chem., Int. Ed., 2016, 55, 12325.
25. B. Nohra, H. El Moll, L. M. Rodriguez Albelo, P. Mialane, J. Marrot, C. Mellot-Draznieks, M. O'Keeffe, R. N. Biboum, J. Lemaire, B. Keita, L. Nadjo and A. Dolbecq, J. Am. Chem. Soc., 2011, 133, 13363.
26. L. M. Rodriguez-Albelo, A. R. Ruiz-Salvador, A. Sampieri, D. W. Lewis, A. Gómez, B. Nohra, P. Mialane, J. Marrot, F. Sécheresse, C. Mellot-Draznieks, R. Ngo Biboum, B. Keita, L. Nadjo and A. Dolbecq, J. Am. Chem. Soc., 2009, 131, 16078.
27. G. Verma, S. Butikofer, S. Kumar and S. Ma, Top. Curr. Chem., 2020, 378, 4.
28. B. Hou, C. Qin, C. Sun, X. Wang and Z. Su, CCS Chem., 2021, 3, 287.
29. L.-D. Lin, D. Zhao, X.-X. Li and S.-T. Zheng, Chem.–Eur. J., 2019, 25, 442.
30. D. Braga, L. Maini, P. P. Mazzeo and B. Ventura, Chem.–Eur. J., 2010, 16, 1553.
31. L. Chong, H. G. Mao, Y. Q. Sun and J. L. Song, Inorg. Chem., 2004, 43, 1964.
32. J. P. Lang, Q. F. Xu, R. X. Yuan and B. F. Abrahams, Angew. Chem., Int. Ed., 2004, 43, 4741.
33. L.-D. Lin, C.-C. Deng, D. Zhao, X.-X. Li and S.-T. Zheng, Chem.–Eur. J., 2018, 24, 251.
34. Y. Xie, D. Kocaefe, Y. Kocaefe, J. Cheng and W. Liu, Nanoscale Res. Lett., 2016, 11, 259.
35. R. G. Pearson, J. Am. Chem. Soc., 1963, 85, 3533.
36. G. Paglia, C. E. Buckley, A. L. Rohl, B. A. Hunter, R. D. Hart, J. V. Hanna and L. T. Byrne, Phys. Rev. B, 2003, 68, 144110.
37. B. C. Lippens and J. H. De Boer, Acta Crystallogr., 1964, 17, 1312.
38. W. W. Wendlandt and H. G. Hecht, Reflectance Spectroscopy, Interscience, New York, 1966.
39. G. M. Sheldrick, SHELXL-2014 Program for Crystal Structure Solution and Refinement, Univerity of Göttingen, Göttingen, Germany, 2014.
40. G. D. Piero, M. Cesari, G. Dozzi and A. Mazzei, J. Organomet. Chem., 1977, 129, 281.
41. C. Schnitter, A. Klemp, H. W. Roesky, H.-G. Schmidt, C. Röpken, R. Herbst-Irmer and M. Noltemeyer, Eur. J. Inorg. Chem., 1998, 2033.
42. C. Jeff Harlan, E. G. Gillan, S. G. Bott and A. R. Barron, Organometallics, 1996, 15, 5479.
43. A. C. Stelzer, P. Hrobárik, T. Braun, M. Kaupp and B. Braun-Cula, Inorg. Chem., 2016, 55, 4915.
44. S. Abeysinghe, K. W. Corum, D. L. Neff, S. E. Mason and T. Z. Forbes, Langmuir, 2013, 29, 14124.
45. S. Abeysinghe, D. K. Unruh and T. Z. Forbes, Inorg. Chem., 2013, 52, 5991.
46. K. W. Corum, M. Fairley, D. K. Unruh, M. K. Payne, T. Z. Forbes and S. E. Mason, Inorg. Chem., 2015, 54, 8367.
47. C. M. Hong, R. G. Bergman, K. N. Raymond and F. D. Toste, Acc. Chem. Res., 2018, 51, 2447.
48. D. Zhang, T. K. Ronson, S. Güryel, J. D. Thoburn, D. J. Wales and J. R. Nitschke, J. Am. Chem. Soc., 2019, 141, 14534.
49. Y.-P. He, L.-B. Yuan, G.-H. Chen, Q.-P. Lin, F. Wang, L. Zhang and J. Zhang, J. Am. Chem. Soc., 2017, 139, 16845.
50. T. Yamada and H. Kitagawa, J. Am. Chem. Soc., 2009, 131, 6312.
51. J. Zhang, L. Wojtas, R. W. Larsen, M. Eddaoudi and M. J. Zaworotko, J. Am. Chem. Soc., 2009, 131, 17040.
52. Q. Lin, X. Bu, C. Mao, X. Zhao, K. Sasan and P. Feng, J. Am. Chem. Soc., 2015, 137, 6184.
53. A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.
54. M. Kruk and M. Jaroniec, Chem. Mater., 2001, 13, 3169.
55. S. Yang, X. Song, P. Zhang and L. Gao, ACS Appl. Mater. Interfaces, 2015, 7, 75.
56. X. Fan, F. Yuan, D. Li, S. Chen, Z. Cheng, Z. Zhang, S. Xiang, S.-Q. Zang, J. Zhang and L. Zhang, Angew. Chem., Int. Ed., 2021, 60, 12949.
57. H. Wang, J. Getzschmann, I. Senkovska and S. Kaskel, Microporous Mesoporous Mater., 2008, 116, 653.
58. M. Andrzejewski and A. Katrusiak, J. Phys. Chem. Lett., 2017, 8, 279.
59. X. Yang, P. K. Majhi, H. Chai, B. Liu, J. Sun, T. Liu, Y. Liu, L. Zhou, J. Xu, J. Liu, D. Wang, Y. Zhao, Z. Jin and Y. R. Chi, Angew. Chem., Int. Ed., 2021, 60, 159.
60. S. Rojas-Buzo, P. García-García and A. Corma, Green Chem., 2018, 20, 3081.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section, detailed structure charts, stability analysis, sorption isotherms, characterization, catalytic mechanism, and tables. The structures reported in this article have been deposited at the Cambridge Crystallographic Data Centre (CCDC). CCDC 2056099–2056118, 2056127, 2156657 (AlOC-99–AlOC-120). For ESI and crystallographic data in CIF or other electronic format see https://doi.org/10.1039/d2sc00526c

---