English

• Recently, porous metal-organic frameworks (MOFs) have attracted much attention in recent two decades because of their potential applications in many areas including gas separation, catalysis, and proton conduction^[1-4]. In particular, the potential utility of MOFs in gas storage and gas separation of important industrial gases such as CO₂/CH₄, C₂, and C₃ hydrocarbons has been demonstrated^[5-8]. This is because of the designability and tunability of functional sites and the pore size/shape of MOF materials^[9]. To efficiently construct MOFs for gas storage and related applications, a better understanding of the function and connectivity of the ligands is important. For instance, Kitagawa et al. developed a solid solution strategy via a fine ligand matrix for gate opening of the flexible MOFs^[10-11]. Bai and co-workers reported a series of highly porous MOFs based on the ligands with inserting amide group for selective CO₂ capture^[12-13]. Notably, some workers prefer to adopt two or more ligands for preparing 3D porous MOFs^[14]. However, understanding the coordination competition between these ligands, particularly for the matrix with sharply different sizes, remains a difficulty since the complex reactions and self-assembly of building blocks occur almost simultaneously.

  To study the coordination competition, the selection of suitable solvents is important as they act as the media to dissolve the ligands and metal salts or as the templates to induce the self-assembly of the metal salts and ligands^[15]. In addition, solvents may also work as a co-ligand to take part in coordination^[16-17]. However, some solvents may be subjected to decompose under hydrothermal conditions^[3]. For example, N, N-dimethylformamide (DMF), a commonly used solvent for hydrothermal reactions, often decomposes to the dimethylamine cation and formate anion under solvothermal conditions. It is worthwhile to note that the dimethylamine cation can be used for the charge balance of the negative MOF network^{[4, 18]}, while the small HCO₂^- anion can be applied as the ligand for the in-situ synthesis of MOFs. Inspired by these observations, herein we report two Mg-MOFs via coordination competitive strategy (Scheme 1). The HCO₂^- anions generated from DMF decomposition reacted directly with Mg²⁺ to form a 3D dia topological MOF, [Mg₃(HCO₂)₆]·DMF (1). However, under the same conditions but with a competing ligand H₄L (1,1': 3',1''-terphenyl-3,3'',5,5''-tetracarboxylic acid), a new 3D sra topological Mg-MOF, [Mg₂(L)(H₂O)₃]·2H₂O·2CH₃CN·DMF (2), was obtained. This result indicates that the short formic acid cannot meet the coordination requirement of Mg²⁺ when a large-sized ligand H₄L is involved. Gas adsorption studies reveal that 1 has a good ability for selective CO₂ capture from CH₄ contained mixture.

  Scheme 1

  

  Scheme 1.  Syntheses of two Mg-MOFs based on coordination competitive strategy

  下载: 全尺寸图片 幻灯片

  1.   Experimental

  1.1   Materials and methods

  All commercially available chemicals were of analytical grade and used without further purification. H₄L was purchased from Shanghai Kaiyulin Pharmaceutical Co., Ltd. The C, H, and N elemental analyses were performed on a PerkinElmer 240 micro analyzer. The FT-IR spectra were performed on a Nicolet 380 FT-IR spectrometer with KBr pellets. Powder X-ray diffraction (PXRD) data were collected on a Bruker D8 Advance diffractometer under Cu Kα radiation (λ = 0.154 06 nm) at 40 kV and 30 mA in a range of 5°-40° Thermogravimetric analysis (TGA) was carried out on a NETZSCH STA 449C thermal analyzer under an N₂ atmosphere with a heating rate of 10 ℃·min^-1.

  1.2   Synthesis of MOF 1

  Mg(NO₃)₂·6H₂O (1.08 g, 4.21 mmol) and HNO₃ (0.25 mL) were added into a mixed solution (7 mL) of DMF and CH₃CN (5:1, V/V) and stirred for ca. 10 min at room temperature (RT). The solution was transferred and sealed in a 20 mL Teflon-lined autoclave, and then heated at 110 ℃ for 48 h. After cooling to RT, colorless crystals of 1 were isolated by filtration, washed with ethanol, and dried in air. Yield: 39.2% (based on Mg²⁺). Anal. Calcd. for C₉H₁₃Mg₃NO₁₃(%): C, 25.98; H, 3.15; N, 3.37. Found(%): C, 25.81; H, 3.01; N, 3.25. FT-IR (KBr discs, cm^-1): 1 674 (s), 1 609 (vs), 1 353 (s), 1 096 (w), 709 (m).

  1.3   Synthesis of MOF 2

  Mg(NO₃)₂·6H₂O (10.8 mg, 0.042 mmol), H₄ L (8.5 mg, 0.021 mmol), and HNO₃ (10 μL) were added into a mixed solution (1.5 mL) of DMF and CH₃CN (5:1, V/V) and stirred for ca. 10 min at RT. The solution was transferred and sealed in a 10 mL Teflon-lined autoclave, and then heated at 110 ℃ for 48 h. After cooling to RT, colorless crystals of 2 were isolated by filtration, washed with DMF, and dried in air. Yield: 15.3% (based on H₄L). Anal. Calcd. for C₂₉H₃₃Mg₂N₃O₁₄(%): C, 50.03; H, 4.78; N, 6.36. Found(%): C, 50.25; H, 4.61; N, 6.22. FT-IR (KBr discs, cm^-1): 3 443 (w), 2 931 (w), 1 663 (s), 1 506 (w), 1 398 (s), 1 252 (w), 1 100 (m), 1 021 (m), 952 (w), 862 (w), 770 (m), 732 (s).

  1.4   Crystal structure determination

  The crystal data of the MOFs were measured on a Bruker Smart Apex Ⅱ CCD diffractometer at 298 K using graphite monochromated Mo Kα radiation (λ = 0.071 073 nm). Data reduction was made with the Bruker Saint program. The structure was solved by direct methods and refined with the full-matrix least squares technique using the SHELXTL package^[19]. The coordinates of the non-hydrogen atoms were refined anisotropically, and all the hydrogen atoms were put in calculated positions or located from the Fourier maps. DMF molecule was disordered over two sites and refined with an occupancy of 0.685(17) for C7-C9, N1, O13 and 0.315(17) for C7A-C9A, N1A, and O13A. The crystallographic data are listed in Table 1, and selected bond lengths are given in Table 2.

  Table 1

  

  Table 1.  Crystal data and structure refinements for MOFs 1 and 2

  下载: 导出CSV

  Parameter	1	2
  Empirical formula	C9H13Mg3NO13	C29H33Mg2N3O14
  Formula weight	416.13	696.19
  Crystal system	Monoclinic	Monoclinic
  Space group	P21/n	P21/c
  a/nm	1.136 9(7)	1.018 2(7)
  b/nm	0.999 5(5)	1.517 1(10)
  c/nm	1.486 1(7)	2.341 3(16)
  β/(°)	91.433(5)	98.309(11)
  V/nm3	1.688 1(16)	3.579(4)
  Z	4	4
  Dc/(g·cm-3)	1.637	0.926
  μ/mm-1	0.248	0.106
  F(000)	856	1 016
  Crystal size/mm	0.10×0.08×0.08	0.12×0.12×0.12
  θ range/(°)	2.2-25	1.6-28.1
  Reflections collected	8 011	24 662
  Independent reflection	2 972 (Rint=0.108 7)	8 718 (Rint=0.150 7)
  Reflection observed [I > 2σ(I)]	1 991	3 570
  Data, restraint, parameter	2 870, 137, 288	8 718, 316, 210
  Goodness-of-fit on F2	1.055	1.008
  R1, wR2 [I > 2σ(I)]	0.087, 0.229 1	0.078 1, 0.223 5
  R1, wR2 (all data)	0.117 9, 0.249	0.148 1, 0.239 0
  (Δρ)max, (Δρ)min/(e·nm-3)	745, -528	1 026, -388

  Table 2

  

  Table 2.  Selected bond distances (nm) for MOFs 1 and 2

  下载: 导出CSV

  1
  Mg1—O2	0.204 5(5)	Mg3—O6	0.203 2(5)	Mg1⋯Mg3	0.557 4(3)
  Mg1—O3	0.212 7(4)	Mg3—O7	0.208 0(5)	Mg1⋯Mg3i	0.536 1(3)
  Mg1—O8i	0.204 9(5)	Mg3—O9	0.209 9(5)	Mg1⋯Mg4	0.568 5(2)
  Mg2—O1	0.209 6(5)	Mg3—O11	0.211 2(5)	Mg2⋯Mg3	0.317 8(3)
  Mg2—O3	0.206 3(5)	Mg3—O10ii	0.202 1(5)	Mg2⋯Mg3i	0.319 4(3)
  Mg2—O5	0.208 3(4)	Mg4—O4	0.204 2(5)	Mg3⋯Mg3i	0.526 2(3)
  Mg2—O9	0.213 3(5)	Mg4—O5	0.212 2(4)	Mg3⋯Mg4	0.551 9(3)
  Mg2—O7i	0.209 5(5)	Mg4—O12i	0.205 6(5)	Mg3i⋯Mg4	0.560 9(3)
  Mg2—O11i	0.210 2(5)	Mg1⋯Mg2	0.356 2(2)
  Mg3—O1	0.208 3(5)	Mg2⋯Mg4	0.354 7(2)
  2
  Mg1—O1	0.210 3(4)	Mg1—O5i	0.207 3(4)	Mg1—O6	0.215 8(3)
  Mg1—O8ii	0.207 7(3)	Mg1—O7	0.217 2(3)	Mg1—O10iv	0.204 9(3)
  Mg3—O4	0.196 4(4)	Mg3—O8iii	0.228 4(4)	Mg3—O2	0.211 2(5)
  Mg3—O3	0.200 0(5)	Mg3—O9iii	0.210 8(4)	Mg3—O11v	0.201 7(4)
  Symmetry codes: i 3/2-x, y-1/2, 3/2-z; ii 3/2-x, 1/2+y, 3/2-z for 1; i x-1, y, z; ii -x, 2-y, -z; iii 1-x, 2-y, -z; iv x-1, 3/2-y, z-1/2; v x, 3/2-y, z-1/2 for 2.

  1.5   Sample activation

  The ethanol-exchanged samples were prepared by immersing as-synthesized crystals in ethanol for 3 d to remove the DMF solvent, and the extract was decanted every 8 h and fresh ethanol was replaced. The completely activated sample was obtained by heating the ethanol-exchanged sample at 120 ℃ for 24 h under a dynamic high vacuum.

  1.6   Gas adsorption experiments

  In the gas sorption measurements, ultra-high-purity grade N₂, CH₄ (> 99.999%), and CO₂ gases (99.995%) were used throughout the adsorption experiments. Low-pressure N₂, CO₂, and CH₄ adsorption measurements were performed on Micromeritics ASAP 2020 M+C surface area analyzer. Helium was used for the estimation of the dead volume, assuming that it is not adsorbed at any of the studied temperatures. The pore size distribution was obtained from the DFT method in the Micromeritics ASAP2020 software package based on the N₂ sorption at 77 K.

  1.7   High-pressure gravimetric gas sorption measurements

  High-pressure adsorption of CO₂ and CH₄ was measured using an IGA-003 gravimetric adsorption instrument (Hiden-Isochema, UK) over a pressure range of 0-2 000 kPa at 273 and 298 K, respectively. Before measurements, about 120 mg ethanol-exchanged samples were loaded into the sample basket within the adsorption instrument and then degassed under high vacuum at 130 ℃ for 20 h to obtain about 65 mg fully desolvated samples. At each pressure, the sample mass was monitored until equilibrium was reached (within 25 min).

  1.8   Gas selectivity

  Ideal adsorption solution theory (IAST) was used to predict binary mixture adsorption from the experimental pure-gas isotherms^[20-21]. To perform the integrations required by IAST, the single-component isotherms should be fitted by a proper model. There is no restriction on the choice of the model to fit the adsorption isotherm, but data over the pressure range under study should be fitted very precisely. The dual-site Langmuir-Freundlich equation was used to fit the experimental data:

  $ q=q_{\mathrm{m} 1} \cdot \frac{b_1 p^{1 / n_1}}{1+b_1 p^{1 / n_1}}+q_{\mathrm{m} 2} \cdot \frac{b_2 p^{1 / n_2}}{1+b_2 p^{1 / n_2}} $

  (1)

  where p is the pressure of the bulk gas at equilibrium with the adsorbed phase (kPa); q is the adsorbed amount of the adsorbent (mol·kg^-1); q_m1 and q_m2 are the saturation capacities (mol·kg^-1) of sites 1 and 2, respectively; b₁ and b₂ are the affinity coefficients (kPa^-1) of sites 1 and 2, respectively; and n₁ and n₂ represent the deviations from an ideal homogeneous surface. The R² values for all the fitted isotherms were over 0.999 97. Hence, the fitted isotherm parameters were applied to perform the necessary integrations in IAST.

  1.9   Estimation of the isosteric heats of gas adsorption

  A virial-type expression comprising the temperature-independent parameters a_i and b_i was employed to calculate the enthalpies of adsorption for CH₄ and CO₂ (at 273 and 298 K) on 1. In each case, the data were fitted using the following equation:

  $ \ln p=\ln N+\frac{1}{T} \sum\limits_{i=0}^m a_i N^i+\sum\limits_{i=0}^n b_i N^i $

  (2)

  where p is the pressure (Torr), N is the adsorbed amount (mmol·g^-1), T is the temperature (K), a_i and b_ⁱ are virial coefficients, and m and n represent the number of coefficients required to adequately describe the isotherms (m and n were gradually increased until the contribution of extra added a and b coefficients were deemed to be statistically insignificant towards the overall fit, and the average value of the squared deviations from the experimental ones was minimized).

  $ Q_{\mathrm{st}}=-R \sum\limits_{i=0}^m a_i N^i $

  (3)

  where Q_st is the coverage-dependent isosteric heat of adsorption and R is the universal gas constant.

  2.   Results and discussion

  2.1   Synthesis and structural characterization

  Under solvothermal conditions, MOF 1 was synthesized by adding only Mg(NO₃)₂·6H₂O to DM/CH₃CN solution in the presence of HNO₃. The HCO₂^- ligand comes from the decomposition of DMF at high temperature, autoclave high pressure, and special acid-ic conditions. This simple synthetic approach is quite different from the reported methods earlier for the formate-based MOFs (Mn²⁺, Co²⁺, and Ni²⁺) in which the HCO₂H was directly used as an organic linker^[22-26]. In addition, the present synthetic route can be easily scaled up to prepare MOF 1 in gram grade at a time. Under the same condition, MOF 2 was prepared after adding H₄L and Mg(NO₃)₂·6H₂O to DMF/CH₃CN solu-tion in the presence of HNO₃.

  Although the crystal structure of MOF 1 is known^[25], the synthetic methods are quite different. 1 crystallizes in the monoclinic P2₁/n space group (Fig. 1, Table 1), which is also different from another formate-based Mg-MOF with the Pbcn space group^[26]. Of particular interest is that there is a pentanuclear Mg₅ cluster consisting of Mg1, Mg2, Mg3, Mg3ⁱ, and Mg4 ions, which can be viewed as a [Mg₄@Mg2] tetrahedron with the Mg2 ion in the center to act as a secondary building unit (SBU). The SBUs are further connected by the formate anions to form a neutral 3D dia net topology (Fig. 1e). 1 possesses 1D channels along the b-axis with a diameter of about 0.44 nm (Fig. 1d). The channels are filled with DMF molecules, which form two intermolecular hydrogen bonds with the H atoms of HCO₂^- anions (C2⋯O13 0.354 6(2) nm, C5ⁱⁱⁱ⋯O13 0.313 1(2) nm, Fig.S1, Supporting information). Interestingly, all the H atoms of HCO₂^- anions point to the channels in 1 (Fig. S2), which may also provide strong interactions with CO₂ after removing the DMF, reflecting high selective CO₂ capture.

  Figure 1

  

  Figure 1.  Structure of MOF 1: (a) OPTEP drawing of the asymmetric unit with 50% thermal ellipsoids probability; (b) a pentanuclear Mg₅ cluster consisting of Mg1-Mg3, Mg3ⁱ, and Mg4 ions; (c) a [Mg₄@Mg2] tetrahedron with the Mg2 ion in the center; (d) 1D channels along the b-axis with a diameter of about 0.44 nm; (e) corresponding dia topology

  All H atoms are omitted for clarity; Symmetry code: ⁱ 3/2-x, y-1/2, 3/2-z

  下载: 全尺寸图片 幻灯片

  MOF 2 crystallizes in the monoclinic P2₁/c space group with relatively large unit cell parameters. In this asymmetric unit, two Mg²⁺ ions, one L^4- ligand, and three coordinated water molecules are observed (Fig. 2a, Table 1). However, the HCO₂^- anion was not observed in 2, despite that the synthetic condition was the same as that of 1. This result demonstrates that there is a coordination competition between H₄L and formate acid. The small-sized formate ions cannot meet the coordination requirements of Mg²⁺ in the presence of a large-sized H₄L ligand. Further analysis shows that the Mg-O distances in both 1 and 2 are all in a normal range (0.196 4(4)-0.228 4(4) nm). In MOF 2, each L^4- ligand is connected by six Mg²⁺ ions with a distorted [MgO₆] octahedron. Due to the chelate coordination nature of two carboxylate groups in L^4-, two Mg²⁺ ions can be viewed as a binuclear cluster, which is bridged by four different L^4- ions. Interestingly, this connection mode makes 2 show the obvious 1D channels with dumbbell window aperture along the a-axis. The window size is 1.42 nm (Fig. 2d). Further packing of these channels forms a 3D porous framework (Fig. 2e). To better understand this structure, topology analysis was performed. Each L^4- linker can be viewed as a 4-connected node (Fig. 2b) and the binuclear Mg₂ cluster can be described as another 4-connected node (Fig. 2c). Thus, 2 shows a 3D sra topology^[27-29]. In addition, the ideal porosity of 2 is as high as 49.2%, making it a highly porous MOF material.

  Figure 2

  

  Figure 2.  Structure of MOF 2: (a) OPTEP drawing of the asymmetric unit with 30% thermal ellipsoids probability; (b) connection of L^4-; (c) connection of Mg₂ cluster; (d) a twisted window aperture along the a-axis with a size of 1.42 nm; (e) packing view of the 3D framework; (f) corresponding sra topology

  All H atoms are omitted for clarity

  下载: 全尺寸图片 幻灯片

  PXRD patterns of as-synthesized samples were in good agreement with their simulated results, revealing the high purity of the bulk products (Fig. 3a). Additionally, activated 1 possessed identical PXRD peaks to the simulated ones, indicating good framework stability of activated 1. However, after guest removal, nearly no diffraction peaks were observed on activated 2 (Fig. 3b), reflecting that the framework of 2 collapses. In addition, the TGA curve of 1 shows that the weight loss of 18.0% between RT and 200 ℃ can be assigned to the removal of one DMF molecule (Calcd. 17.6%, Fig. 3c). For 2, the first weight loss of 28.2% from RT to 145 ℃ is ascribed to the removal of two CH₃CN molecules, two lattice water molecules and one DMF molecule (Calcd. 27.5%). The second weight loss of 7.1% until 245 ℃ is ascribed to the loss of three coordinated water molecules (Calcd. 7.8%, Fig. 3d). Compared with the decomposition temperatures of 390 ℃ for 1 and 300 ℃ for 2, it is worthwhile to note that the short linker prefers to form a more stable porous MOF material.

  Figure 3

  

  Figure 3.  PXRD patterns (a, b) and TGA curves (c, d) of MOFs 1 and 2

  下载: 全尺寸图片 幻灯片

  2.2   Pore evaluation and single gas adsorption

  The permanent micro-porosity of MOFs 1 and 2 was evaluated by N₂ adsorption isotherm at 77 K (Fig. 4a and S3). The N₂ adsorption isotherm of 1 shows a quick uptake with a type-Ⅰ behavior at low pressure and a total uptake of 104.5 cm³·g^-1 at p/p₀=1. The Brunauer-Emmett-Teller (BET) and Langmuir surface areas were calculated to be 342 and 378 m²·g^-1, respectively. As shown in Fig. 4b, the pore size centered at about 0.40 nm, which was very close to the value determined from the crystal structure (Fig. 1d). However, nearly negligible uptake was found in 2, which agrees with the decomposition of the framework after the removal of guest (Fig. 3b).

  Figure 4

  

  Figure 4.  (a) N₂ adsorption isotherms of MOF 1 at 77 K; (b) Pore size distribution of 1; (c) Single gas adsorption isotherms of 1; (d) IAST selectivity of 1; (e) Q_st of 1 for CO₂ and CH₄; (f) PXRD patterns of treated 1

  下载: 全尺寸图片 幻灯片

  Due to the micro-porosity of MOF 1, pure gascomponent sorption isotherms of CO₂ and CH₄ were collected at 273 and 298 K, respectively (Fig. 4c). With reversible type-Ⅰ isotherms, 1 exhibited a higher CO₂ uptake (mass fraction) of 2.4% (0.53 mmol·g^-1) at 298 K and 15 kPa, the partial pressure of CO₂ in the flue gas. This value was higher than that of NJU-Bai50 (2.11%) ^[27], FZU (2.01%) ^[30] and approaching to that of ZIF-78 (3.3%)^[31]. Interestingly, by reducing the adsorption temperature to 273 K, the uptake at 15 kPa increased by more than two times (5.3%), which makes 1 a good CO₂ collector. In addition, the excess CO₂ uptake reached 11.7% (2.6 mmol·g^-1) at 273 K and 100 kPa, while the unsaturation CO₂ uptake was as high as 17.2% (3.9 mmol·g^-1) at 2 000 kPa. With a nearly similar adsorption trend, the maximum CO₂ uptake was about 14.7% at 298 K and 2 000 kPa. Although the CO₂ uptake at 2 000 kPa was limited by the volume of the micropore, the uptake value of 1 at 100 kPa was higher than those of the known microporous MOFs^[32]. However, corresponding CH₄ uptakes of 1 at 2 000 kPa were only 4.4% at 273 K and 4.0% at 298 K. This adsorption difference indicates the high potential of 1 for selective CO₂ capture from CH₄-contained mixture.

  The unique CO₂ adsorption isotherms encouraged us to further examine the capacity of MOF 1 for the selective capture of CO₂/CH₄ at 298 K. IAST was employed to predict multi-component adsorption behaviors from the experimental pure-gas isotherms. The predicted adsorption selectivity in 1 as a function of bulk pressure is presented in Fig. 4d, S4, and S5. The equimolar selectivity of CO₂ over CH₄ was very sensitive to the loading, which showed two steps in the changes of selectivity: a quick decrease from 11 to 5.2 at the low-pressure region and a slow increase from 5.2 to 6.6 following the increased pressure. Interestingly, the CO₂/CH₄ selectivity was also sensitive to the gas ratio, particularly at high pressure. The higher the CO₂ concentration was, the higher selectivity was. To understand these results, the adsorption enthalpies were calculated by the virial method (Fig. S6 and S7). 1 exhibited a strong binding affinity (33.5 kJ·mol^-1) for CO₂ at zero coverage, and the enthalpy of adsorption increased to 36.5 kJ·mol^-1 at about 500 kPa. The initial high value indicates that there are interactions between the H atom of the HCO₂^- ion and CO₂ molecule, while the increased values stem from pressure-driven CO₂⋯CO₂ interactions. However, 1 had a relatively low enthalpy of adsorption (21.5 kJ·mol^-1) for CH₄.

  Moreover, the framework structure of MOF 1 after the adsorption measurements and water treatment for one month was still kept, confirmed by the PXRD patterns (Fig. 4f). The convenient synthesis, high stability towards the water, good selectivity, and facile regeneration make 1 a promising porous MOF material for the separation of CO₂ and CH₄ for long-term use.

  3.   Conclusions

  In summary, two Mg-based MOFs 1 and 2 were prepared by using a coordination competition strategy between formic acid generated from the decomposition of DMF and 1,1': 3',1''-terphenyl-3,3'', 5,5''-tetracarboxylic acid. MOF 1 possesses a 3D dia topological network and has a 1D channel, while MOF 2 has a unique binuclear Mg₂ cluster, yielding a 3D sra topology network. These results demonstrate that ligands with the same coordinating groups and different sizes are difficult to be compatible with when reacting with Mg²⁺ ions. In addition, with good water stability, 1 exhibited quick CO₂ uptake and good selectivity for CO₂/CH₄ separation in a wide pressure range at 298 K. This work permits us to envision that coordination competition strategy may be an important method for the design and preparation of MOF materials in the future.

  Supporting information is available at http://www.wjhxxb.cn

  
  
• 1. [1]

     Behera N, Duan J G, Jin W Q, Kitagawa S. The chemistry and applications of flexible porous coordination polymers[J]. Energy Chem., 2021, 3:  100067. doi: 10.1016/j.enchem.2021.100067

  2. [2]

     Nam D H, De Luna P, Rosas-Hernandez A, Thevenon A, Li F W, Agapie T, Peters J C, Shekhah O, Eddaoudi M, Sargent E H. Molecular enhancement of heterogeneous CO₂ reduction[J]. Nat. Mater., 2020, 19:  266-276. doi: 10.1038/s41563-020-0610-2

  3. [3]

     Wang X, Qin T, Bao S S, Zhang Y C, Shen X, Zheng L M, Zhu D R. Facile synthesis of a water stable 3D Eu-MOF showing high proton conductivity and its application as a sensitive luminescent sensor for Cu²⁺ ions[J]. J. Mater. Chem. A, 2016, 4:  16484-16489. doi: 10.1039/C6TA06792A

  4. [4]

     Zhang S L, Xie Y X, Yang M R, Zhu D R. Porosity regulation of metalorganic frameworks for high proton conductivity by rational ligand design: Mono-versus disulfonyl-4, 4'-biphenyldicarboxylic acid[J]. Inorg. Chem. Front., 2022, 9:  1134-1142. doi: 10.1039/D1QI01610E

  5. [5]

     Zeng H, Xie M, Wang T, Wei R J, Xie X J, Zhao Y F, Lu W G, Li D. Orthogonal-array dynamic molecular sieving of propylene/propane mixtures[J]. Nature, 2021, 595:  542-548. doi: 10.1038/s41586-021-03627-8

  6. [6]

     Dong Q B, Huang Y H, Hyeon-Deuk K, Chang I Y, Wan J M, Chen C L, Duan J G, Jin W N, Kitagawa S. Shape- and size-dependent kinetic ethylene sieving from a ternary mixture by a trap-and-flow channel crystal[J]. Adv. Funct. Mater., 2022, 32:  2203745. doi: 10.1002/adfm.202203745

  7. [7]

     Dong Q B, Zhang X, Liu S, Lin R B, Guo Y N, Ma Y S, Yonezu A, Krishna R, Liu G P, Duan J G, Matsuda R, Jin W Q, Chen B L. Tuning gate-opening of a flexible metal-organic framework for ternary gas sieving separation[J]. Angew. Chem. Int. Ed., 2020, 59:  22756-22762. doi: 10.1002/anie.202011802

  8. [8]

     Lin R B, Li L B, Zhou H L, Wu H, He C H, Li S, Krishna R, Li J P, Zhou W, Chen B L. Molecular sieving of ethylene from ethane using a rigid metal-organic framework[J]. Nat. Mater., 2018, 17:  1128-1133. doi: 10.1038/s41563-018-0206-2

  9. [9]

     Furukawa H, Cordova K E, O'Keeffe M, Yaghi O M. The chemistry and applications of metal-organic frameworks[J]. Science, 2013, 341:  1230444. doi: 10.1126/science.1230444

  10. [10]

      Gu C, Hosono N, Zheng J J, Sato Y, Kusaka S, Sakaki S, Kitagawa S. Design and control of gas diffusion process in a nanoporous soft crystal[J]. Science, 2019, 363:  387-391. doi: 10.1126/science.aar6833

  11. [11]

      Fukushima T, Horike S, Inubushi Y, Nakagawa K, Kubota Y, Takata M, Kitagawa S. Solid solutions of soft porous coordination polymers: Fine-tuning of gas adsorption properties[J]. Angew. Chem. Int. Ed., 2010, 49:  4820-4824. doi: 10.1002/anie.201000989

  12. [12]

      Zhang Y F, Zhang Z H, Ritter L, Fang H, Wang Q, Space B, Zhang Y B, Xue D X, Bai J F. New reticular chemistry of the rod secondary building unit: Synthesis, structure, and natural gas storage of a series of three-way rod amide-functionalized metal-organic frameworks[J]. J. Am. Chem. Soc., 2021, 143:  12202-12211. doi: 10.1021/jacs.1c04946

  13. [13]

      Xue D X, Wang Q, Bai J F. Amide-functionalized metal-organic frameworks: Syntheses, structures and improved gas storage and separation properties[J]. Coord. Chem. Rev., 2019, 378:  2-16. doi: 10.1016/j.ccr.2017.10.026

  14. [14]

      Farha O K, Yazaydin A O, Eryazici I, Malliakas C D, Hauser B G, Kanatzidis M G, Nguyen S T, Snurr R Q, Hupp J T. De novo synthesis of a metal-organic framework material featuring ultrahigh surface area and gas storage capacities[J]. Nat. Chem., 2010, 2:  944-948. doi: 10.1038/nchem.834

  15. [15]

      Takaishi S, DeMarco E J, Pellin M J, Farha O K, Hupp J T. Solventassisted linker exchange (SALE) and post-assembly metallation in porphyrinic metal-organic framework materials[J]. Chem. Sci., 2013, 4:  1509-1513. doi: 10.1039/c2sc21516k

  16. [16]

      Howlader P, Mukherjee P S. Solvent directed synthesis of molecular cage and metal organic framework of copper(Ⅱ) paddlewheel cluster[J]. Isr. J. Chem., 2019, 59:  292-298. doi: 10.1002/ijch.201800155

  17. [17]

      Guo X G, Yang W B, Wu X Y, Zhang K, Lin L, Yu R M, Lu C Z. Stoichiometry, temperature, solvent, metal-directed syntheses of metal-organic frameworks based on flexible V-shaped methylenebis (3, 5-dimethylpyrazole) and various aromatic dicarboxylate acids[J]. CrystEngComm, 2013, 15:  3654-3663. doi: 10.1039/c3ce00048f

  18. [18]

      Duan J G, Jin W Q, Krishna R. Natural gas purification using a porous coordination polymer with water and chemical stability[J]. Inorg. Chem., 2015, 54:  4279-4284. doi: 10.1021/ic5030058

  19. [19]

      Sheldrick G M. A short history of SHELX[J]. Acta Crystallogr. Sect. A, 2008, A64:  112-122.

  20. [20]

      Bae T H, Lee J S, Qiu W L, Koros W J, Jones C W, Nair S. Highperformance gas-separation membrane containing submicrometer-sized metal-organic framework crystals[J]. Angew. Chem. Int. Ed., 2010, 49:  9863-9866. doi: 10.1002/anie.201006141

  21. [21]

      Bae Y S, Mulfort K L, Frost H, Ryan P, Punnathanam S, Broadbelt L J, Hupp J T, Snurr R Q. Separation of CO₂ from CH₄ using mixed-ligand metal-organic frameworks[J]. Langmuir, 2008, 24:  8592-8598. doi: 10.1021/la800555x

  22. [22]

      Li K H, Olsan D H, Lee J Y, Bi W H, Wu K, Yuen T, Xu Q, Li J. Multifunctional microporous MOFs exhibiting gas/hydrocarbon adsorption selectivity, separation capability and three-dimensional magnetic ordering[J]. Adv. Funct. Mater., 2008, 18:  2205-2214. doi: 10.1002/adfm.200800058

  23. [23]

      Wang Z M, Zhang B, Fujiwara H, Kobayashi H, Kurmoo M. Mn₃(HCOO)₆: A 3D porous magnet of diamond framework with nodes of Mn-centered MnMn₄ tetrahedron and guest-modulated ordering temperature[J]. Chem. Commun., 2004, :  416-417.

  24. [24]

      Dybtsev D N, Chun H, Yoon S H, Kim D, Kim K. Microporous manganese formate: A simple metal-organic porous material with high framework stability and highly selective gas sorption properties[J]. J. Am. Chem. Soc., 2004, 126:  32-33. doi: 10.1021/ja038678c

  25. [25]

      Rood J A, Noll B C, Henderson K W. Synthesis, structural characterization, gas sorption and guest-exchange studies of the lightweight, porous metal-organic framework α-[Mg₃(O₂CH)₆][J]. Inorg. Chem., 2006, 45:  5521-5528. doi: 10.1021/ic060543v

  26. [26]

      Mallick A, Saha S, Pachfule P, Roy S, Banerjee R. Structure and gas sorption behavior of a new three dimensional porous magnesium formate[J]. Inorg. Chem., 2011, 50:  1392-1401. doi: 10.1021/ic102057p

  27. [27]

      Song X H, Zhang M X, Duan J G, Bai J F. Constructing and finely tuning the CO₂ traps of stable and various-pore-containing MOFs towards highly selective CO₂ capture[J]. Chem. Commun., 2019, 55:  3477-3480. doi: 10.1039/C8CC10116G

  28. [28]

      Wang X Z, Zhu D R, Xu Y, Yang J, Shen X, Zhou J, Fei N, Ke X K, Peng L M. Three novel metal-organic frameworks with different topologies based on 3, 3'-dimethoxy-4, 4'-biphenyldicarboxylic acid: Syntheses, structures, and properties[J]. Cryst. Growth Des., 2010, 10:  887-894. doi: 10.1021/cg9012262

  29. [29]

      Luo R, Xu H, Gu H X, Wang X, Xu Y, Shen X, Bao W W, Zhu D R. Four MOFs with 2, 2'-dimethoxy-4, 4'-biphenyldicarboxylic acid: Syntheses, structures, topologies and properties[J]. CrystEngComm, 2014, 16:  784-796. doi: 10.1039/C3CE41428K

  30. [30]

      Jiang J J, Lu Z Y, Zhang M X, Duan J G, Zhang W W, Pan Y, Bai J F. Higher symmetry multinuclear clusters of MOFs for highly selective CO₂ capture[J]. J. Am. Chem. Soc., 2018, 140:  17825-17829. doi: 10.1021/jacs.8b07589

  31. [31]

      Banerjee R, Furukawa H, Britt D, Knobler C, O'Keeffe M, Yaghi O M. Control of pore size and functionality in isoreticular zeolitic imidazolate frameworks and their carbon dioxide selective capture properties[J]. J. Am. Chem. Soc., 2009, 131:  3875-3877. doi: 10.1021/ja809459e

  32. [32]

      Sumida K, Rogow D L, Mason J A, McDonald T M, Bloch E D, Herm Z R, Bae T H, Long J R. Carbon dioxide capture in metal-organic frameworks[J]. Chem. Rev., 2012, 112:  724-781. doi: 10.1021/cr2003272

• 

  Scheme 1  Syntheses of two Mg-MOFs based on coordination competitive strategy

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Structure of MOF 1: (a) OPTEP drawing of the asymmetric unit with 50% thermal ellipsoids probability; (b) a pentanuclear Mg₅ cluster consisting of Mg1-Mg3, Mg3ⁱ, and Mg4 ions; (c) a [Mg₄@Mg2] tetrahedron with the Mg2 ion in the center; (d) 1D channels along the b-axis with a diameter of about 0.44 nm; (e) corresponding dia topology

  All H atoms are omitted for clarity; Symmetry code: ⁱ 3/2-x, y-1/2, 3/2-z

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Structure of MOF 2: (a) OPTEP drawing of the asymmetric unit with 30% thermal ellipsoids probability; (b) connection of L^4-; (c) connection of Mg₂ cluster; (d) a twisted window aperture along the a-axis with a size of 1.42 nm; (e) packing view of the 3D framework; (f) corresponding sra topology

  All H atoms are omitted for clarity

  下载: 全尺寸图片 幻灯片
  

  Figure 3  PXRD patterns (a, b) and TGA curves (c, d) of MOFs 1 and 2

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) N₂ adsorption isotherms of MOF 1 at 77 K; (b) Pore size distribution of 1; (c) Single gas adsorption isotherms of 1; (d) IAST selectivity of 1; (e) Q_st of 1 for CO₂ and CH₄; (f) PXRD patterns of treated 1

  下载: 全尺寸图片 幻灯片

  Table 1.  Crystal data and structure refinements for MOFs 1 and 2

  Parameter	1	2
  Empirical formula	C9H13Mg3NO13	C29H33Mg2N3O14
  Formula weight	416.13	696.19
  Crystal system	Monoclinic	Monoclinic
  Space group	P21/n	P21/c
  a/nm	1.136 9(7)	1.018 2(7)
  b/nm	0.999 5(5)	1.517 1(10)
  c/nm	1.486 1(7)	2.341 3(16)
  β/(°)	91.433(5)	98.309(11)
  V/nm3	1.688 1(16)	3.579(4)
  Z	4	4
  Dc/(g·cm-3)	1.637	0.926
  μ/mm-1	0.248	0.106
  F(000)	856	1 016
  Crystal size/mm	0.10×0.08×0.08	0.12×0.12×0.12
  θ range/(°)	2.2-25	1.6-28.1
  Reflections collected	8 011	24 662
  Independent reflection	2 972 (Rint=0.108 7)	8 718 (Rint=0.150 7)
  Reflection observed [I > 2σ(I)]	1 991	3 570
  Data, restraint, parameter	2 870, 137, 288	8 718, 316, 210
  Goodness-of-fit on F2	1.055	1.008
  R1, wR2 [I > 2σ(I)]	0.087, 0.229 1	0.078 1, 0.223 5
  R1, wR2 (all data)	0.117 9, 0.249	0.148 1, 0.239 0
  (Δρ)max, (Δρ)min/(e·nm-3)	745, -528	1 026, -388

  下载: 导出CSV

  Table 2.  Selected bond distances (nm) for MOFs 1 and 2

  1
  Mg1—O2	0.204 5(5)	Mg3—O6	0.203 2(5)	Mg1⋯Mg3	0.557 4(3)
  Mg1—O3	0.212 7(4)	Mg3—O7	0.208 0(5)	Mg1⋯Mg3i	0.536 1(3)
  Mg1—O8i	0.204 9(5)	Mg3—O9	0.209 9(5)	Mg1⋯Mg4	0.568 5(2)
  Mg2—O1	0.209 6(5)	Mg3—O11	0.211 2(5)	Mg2⋯Mg3	0.317 8(3)
  Mg2—O3	0.206 3(5)	Mg3—O10ii	0.202 1(5)	Mg2⋯Mg3i	0.319 4(3)
  Mg2—O5	0.208 3(4)	Mg4—O4	0.204 2(5)	Mg3⋯Mg3i	0.526 2(3)
  Mg2—O9	0.213 3(5)	Mg4—O5	0.212 2(4)	Mg3⋯Mg4	0.551 9(3)
  Mg2—O7i	0.209 5(5)	Mg4—O12i	0.205 6(5)	Mg3i⋯Mg4	0.560 9(3)
  Mg2—O11i	0.210 2(5)	Mg1⋯Mg2	0.356 2(2)
  Mg3—O1	0.208 3(5)	Mg2⋯Mg4	0.354 7(2)
  2
  Mg1—O1	0.210 3(4)	Mg1—O5i	0.207 3(4)	Mg1—O6	0.215 8(3)
  Mg1—O8ii	0.207 7(3)	Mg1—O7	0.217 2(3)	Mg1—O10iv	0.204 9(3)
  Mg3—O4	0.196 4(4)	Mg3—O8iii	0.228 4(4)	Mg3—O2	0.211 2(5)
  Mg3—O3	0.200 0(5)	Mg3—O9iii	0.210 8(4)	Mg3—O11v	0.201 7(4)
  Symmetry codes: i 3/2-x, y-1/2, 3/2-z; ii 3/2-x, 1/2+y, 3/2-z for 1; i x-1, y, z; ii -x, 2-y, -z; iii 1-x, 2-y, -z; iv x-1, 3/2-y, z-1/2; v x, 3/2-y, z-1/2 for 2.

  下载: 导出CSV
•