Qizhao
Xiong
,
Yang
Chen
*,
Dongxiao
Yang
,
Kaihua
Wang
,
Yi
Wang
,
Jiangfeng
Yang
,
Libo
Li
* and
Jinping
Li
College of Chemical Engineering and Technology, Shanxi Key Laboratory of Gas Energy Efficient and Clean Utilization, Taiyuan University of Technology, Taiyuan 030024, China. E-mail: chenyangtyut@163.com; lilibo908@hotmail.com
First published on 10th August 2022
Since metal–organic frameworks, a versatile class of crystalline organic–inorganic hybrid materials featuring well-aligned intrinsic porosity, have come of age, the research focus has shifted from structural considerations toward the many fascinating properties enshrined in their real industrial applications. However, most MOFs reported to date only have microporous structures, which restrict mass transfer and inhibit macromolecules from accessing their pores. Hierarchically porous MOFs have been proposed because mesopores or macropores can alleviate these challenges. The strategies used to construct hierarchically porous MOFs have been discussed in this review article based on the pore size typically generated by each method and several instances of their applications in adsorption diffusion are shown. These applications demonstrate that the mass transport rate of hierarchically porous MOFs is improved when compared to pristine microporous MOFs, which is expected to solve the key problems found in application fields such as adsorption, catalysis, and sensing. Finally, the properties and challenges of hierarchically porous MOFs have been summarized, along with some recommendations for their future development.
International Union of Pure and Applied Chemistry (IUPAC) has defined pores with a pore size of <2 nm as micropores, pores in the range of 2–50 nm as mesopores, and pores >50 nm as macropores. In the field of gas separation, a proven strategy is to use micropores of pore sizes between the dynamic diameters of the gas molecules to sieve the gas mixture. However, in practical applications, the active sites of microporous porous materials are usually confined to the micropores, making the intrinsic mass transfer rates of microporous porous materials severely limited in catalytic and adsorption processes.14 Despite the high selectivity of microporous MOFs for the separation of specific gas components, they are limited by slow mass transfer rates and low separation productivity. Thus, they have not been widely applied in industry.15 To date, the majority of the reported MOFs are microporous and only a few MOFs materials have mesoporous or macroporous structures.16,17
The drawback that microporous MOFs restrict mass transfer and macromolecule passage can be solved by extending the chain length of the ligand to directly enlarge the pore size of MOFs. Therefore, the design and synthesis of long-chain organic ligands have been of great research interest, but the pore size has only been increased to 98 Å, which is probably the limit of what can be achieved by extending the ligand length.18 Moreover, MOFs with extended ligand-constructed mesoporous structures usually collapse rapidly upon removal of the solvent because mesoporous topologies are more fragile than microporous structures. In addition, such extended ligand-constructed mesoporous MOFs are also prone to forming interpenetrating structures, leading to a significant reduction in the specific surface area and pore size. Therefore, the construction of hierarchically porous MOFs has been proposed to solve these problems. Two or more composite pore sizes in the same material is the key characteristic of hierarchically porous MOFs. When compared with conventional MOFs, the introduction of some larger pores can provide two or more pore advantages at the same time, such as the advantages of high specific surface area provided by micropores and the reduction of the molecular mass transfer resistance and accessible entrance of large molecules provided by mesopores and macropores.19 As a result, hierarchically porous MOFs show great promise for various applications in the petrochemical industry.20
In most cases, the formation mechanisms of pores with similar sizes tend to have similarities during the formation of the hierarchical pores in MOFs. Therefore, the construction methods can be effectively classified according to the pore size formed by the hierarchical pores. In this paper, we introduce the methods used to construct hierarchically porous MOFs from micropores (mostly ligands with different functional groups are introduced), mesoporous (mostly pores formed by defects within the crystals), and macropores (predominantly pores formed by crystal stacking), and present the application advantages of hierarchically porous MOFs in adsorption and catalysis, as shown in Scheme 1.21–28
Scheme 1 Constructing strategies of hierarchically porous MOFs with different pore size ranges and their applications in adsorption and catalysis.21–28 |
Fig. 1 (a) Ti-Exchanged UiO-66 enhances CO2 adsorption capacity.47,48 Reprinted with permission from ref. 48. Copyright 2013, The Royal Society of Chemistry. (b) Pure-phase MOF-74 crystals containing 10 different metals.49 Reprinted with permission from ref. 49. Copyright 2014, American Chemical Society. (c) MTV-MOF-5 was constructed simultaneously with eight different ligands.50 Reprinted with permission from ref. 50. Copyright 2010, American Association for the Advancement of Science. |
In some cases, however, the pores introduced by the mixed ligands are mesopores as a result of a lack of ligands, mostly due to the different properties of the ligands used. For example, Jin et al. proposed a mixed ligand strategy to form hierarchically porous MIL-125 based on differences in the electronegativity.51 Hierarchically porous MIL-125 with a series of continuously tunable mesoporous pore sizes can be obtained by simply adjusting the molar ratio of the two organic ligands, BDC-NH2 (2-aminoterephthalic acid) and BDC (terephthalic acid). In addition, trimesic acid (btc) can be mixed with pyridine-3,5-dicarboxylate (pydc) to introduce structural defects to construct hierarchically porous Ru-MOF and reduce the valence of the Ru metal.52 Mixed ligands can also be introduced under mild conditions utilizing a post-synthetic exchange (PSE) strategy,53–55 which was named by Kim et al.56 Cai et al. developed a general post-synthetic ligand substitution (PSLS) strategy to obtain more than 10 g of hierarchically porous UiO-66 and its composites using a simple reflux system.57 This strategy opens new doors for the fast, facile, versatile, and large-scale production of HP-MOF and its related composites to expand the application of conventional microporous MOF-based materials.
Fig. 2 (a) Schematic illustration of the process for obtaining hierarchically porous PCN-160 from chemically labile CBAB ligands.58 (b) Schematic illustration of the post-synthesis hydrolysis to obtain the hierarchically porous POST-66(Y).61 Reprinted with permission from ref. 61. Copyright 2015, Wiley. (c) The reduction of HKUST-1 by methanol steam formed hierarchically porous.62 Reprinted with permission from ref. 62. Copyright 2019, Wiley. |
Defect construction can also take advantage of the relative stability of MOFs. This method produces defects in a straightforward way by precisely dissolving a portion of the MOF crystals, forming mesopores with a pore size distribution that is typically broader than that of the mesopores formed using the ligand instability method. According to the different processing methods used,60 most of these methods used to construct mesopores via etching the crystal can be divided into liquid-phase, gas-phase, and metal reduction (the principle is to reduce the coordination number of the metal, resulting in the absence of ligands and the formation of defects) methods. Liquid-phase etching methods include that reported by Kim et al., which obtained hierarchically porous POST-66(Y) via hydrolysis.61 This method generates mesopores in the range of 3 to 20 nm in the MOF, which can be controlled by adjusting the time and temperature of hydrolysis, as shown in Fig. 2b. The resulting mesopore size increases upon increasing the dissolution time or temperature. Since the bridging ligands of some MOFs are carboxylic acid ligands, these MOFs are usually unstable under alkaline conditions. Etching these MOFs in alkaline solutions is likely to cause the rapid collapse of the MOF structure, so that the crystal can be etched with alkaline vapors under relatively mild conditions. For example, Albolkany et al. were the first to propose an ammonia etching strategy for HKUST-1, which can reversibly generate and repair mesopores with tunable pore size.23 This strategy can generate ordered mesopores on specific crystal planes of carboxylate-based microporous MOFs without affecting the crystal morphology. The pore sizes of the mesopores can be controlled by the etching temperature and the porosity of the mesopores can be adjusted by the pressure of etchant. The resulting mesopores can also be modified using the MOF precursor solution for controllable mesopore generation or encapsulation of the adsorbed molecules.
In addition, another method used to construct mesopores utilizing the crystal instability strategy is the metal reduction method. For example, Qi et al. reported an approach to construct hierarchically porous structures in HKUST-1 by tuning the valence state of the metal ions, as shown in Fig. 2c.62 Their experiments showed that a portion of the CuII sites was reduced to CuI by the CH3OH vapor, which destroys some of the coordination bonds, resulting in the formation of mesopores in the microporous framework and the formation of macropores. Moreover, Song et al. used hydroquinone (H2Q) to reduce Cu2+ ions in HKUST-1 for the first time to obtain mesopores.63 Using single-crystal X-ray diffraction, CO adsorption, and 1H NMR spectroscopic analysis, it was demonstrated that the treatment of HKUST-1 with H2Q under anhydrous conditions resulted in the reduction of 33.3% of the Cu2+ sites (half of the framework Cu+ and half of the free Cu+ sites). The reduction reaction will not proceed any further once 33.3% of the Cu2+ sites are reduced to Cu+. This phenomenon is due to the fact that HKUST-1 has a limited number of pore cages. Furthermore, coordination reduction results in a significant enhancement of the hydrolytic stability of HKUST-1, which remains structurally intact even after two years of exposure to humid air.
Fig. 3 (a) HRTEM analysis of missing ligands in UiO-66.64 Reprinted with permission from ref. 64. Copyright 2019, Nature Publishing Group. (b) Schematic diagram of the preparation of hierarchically porous PCN-250 with monocarboxylic acid as modulator.66 Reprinted with permission from ref. 66. Copyright 2019, Wiley. (c) The pore size of mesoporous UiO-66 is systematically adjusted by changing the concentration of acetic acid.67 Reprinted with permission from ref. 67. Copyright 2017, Wiley. |
Fig. 4 (a) Construction of hierarchically porous UiO-66(Ce) by soft template method.24 Reprinted with permission from ref. 24. Copyright 2022, American Chemical Society. (b) Using highly ordered PS to construct hierarchically porous ZIF-8.28 Reprinted with permission from ref. 28. Copyright 2018, American Association for the Advancement of Science. |
MOFs can also encapsulate nanoparticles with the desired size and then etch the encapsulated particles to form hierarchically porous materials. The encapsulated particles need to have functional groups that can interact with the MOF precursors. For example, Shen et al. used highly ordered polystyrene microspheres (PS) as the hard template and ZIF-8 as the research object. They reported the world's first ordered macroporous–microporous MOF in the form of a single crystal, as shown in Fig. 4b. This opened up the field of three-dimensional ordered macroporous–microporous materials.28 The ZIF-8 precursor was assembled and grown on the highly ordered monodisperse PS nanospheres. Finally, the PS template was removed using THF selective etching to obtain the highly ordered macroporous–microporous ZIF-8 single crystals. The size of the hard template determined the pore size of the mesopores. However, hard templates are sometimes tricky to remove without destroying the crystallinity of the MOF. Therefore, another approach to the hard template method has been developed, which is to use the reactant as the template (usually a metal or ligand source that is insoluble in the solution), and the formed MOF nanoparticles are stacked and grown around the template. Therefore, the template is involved in the reaction. The mesopores formed using the self-sacrificial template method are constructed via the self-sacrificial formation of defects. Therefore, the pore size formed using this method also depends on the size of the template used, but it is not controllable when compared with the soft template method and other hard template methods. Zhang reported a simple and effective self-sacrificial templating strategy based on the nanoscale Kirkendall effect to form hollow nanorod structures of Co-MOF-74, which can efficiently adsorb/desorb gas molecules. This nanostructured MOF-74 shortens the diffusion distance and increases the rate of CO2 gas adsorption under dynamic adsorption conditions.75 In addition, ionic liquids, CO2-expanded liquid, etc. can also be used as templates to construct hierarchical porous MOFs.76,77
Fig. 5 (a) Nearly linear adjustment of the mesopore pore size of hierarchically porous UiO-66(Hf) by adjusting the molar ratio of monocarboxylic acid/water.80 Reprinted with permission from ref. 80. Copyright 2021, Wiley. (b) Selectively grown to form hierarchical structure MOFs.82 Reprinted with permission from ref. 82. Copyright 2020, American Chemical Society. (c) MOF-74 arrays are closely arranged on the surface of Co/CC rods.85 Reprinted with permission from ref. 85. Copyright 2020, American Chemical Society. |
There are also methods of compounding MOFs with other porous materials, such as mixed matrix membranes. At the same time, there are problems such as the incompatibility between the MOFs and polymer, and clogging of the pores. Wang et al. reported the conversion of an MOF-on-MOF to MOF–polymer composite. This templated synthesis allows the introduction of MOFs of various sizes, distributions, and concentrations into polymeric matrices, which partially form mesopores.84 In the work of Zha et al., a hierarchically structured Co, Fe-MOF-74/Co/CC (carbon layer) hybrid electrode nanorod structure was obtained using a simple solvothermal treatment method.85 MOF-74 arrays are densely arranged on the surface of Co/CC rods, forming a hierarchical structure, as shown in Fig. 5c. McDonald et al. uniformly grafted PMMA (polymethyl methacrylate) on a core–shell MOF, IRMOF-3@MOF-5, maintaining its internal porosity.86 Since the PMMA polymerization initiator can be bound to the amino group using a post-synthesis modification method, the thickness of the polymer film can be controlled by adjusting the thickness of the shell layer carrying the initiator. This strategy to form MOF composites usually enables the materials to exhibit better properties than single pure-phase MOFs. However, the synthesis conditions are relatively harsh, which not only imposes certain restrictions on the MOFs used, but the stability of the MOFs is also required in some cases. In general, this strategy is easier to nucleate on its own than to form composites. Of course, the most considerable difficulty faced by such composite MOF materials with mismatched topologies is how to solve the high interface energy as a result of mismatched topologies.87 To rise above the energy barrier, the use of surfactants or structure-directing agents is an option to simultaneously grow two different MOFs, especially with different topological structures. Otherwise, the crystal growth behaviors need to be adjusted to overcome the surface energy.88
Fig. 6 (a) Schematic representation of MIL-53 (Al) gel formation.22 (b) Mechanical synthesis of HKUST-1.99 Reprinted with permission from ref. 99. Copyright 2018, The Royal Society of Chemistry. (c) Mechanical synthesis of ZIF-8.100 Reprinted with permission from ref. 100. Copyright 2017, American Chemical Society. |
A mixed-ligand strategy has been used to construct a hierarchically porous MOF capable of enhancing CO2 adsorption, which can introduce functional groups that have an effect on CO2 or expose uncoordinated carboxylic acid groups. For example, Park et al. co-assembled TPTC (terphenyl-tetra carboxylate) and R-isoph (5-R isophthalic acid, R represents a functional group) to introduce functionalized mesopores in PCN-125.125 The introduced functional groups enhance the interaction with CO2 and most of the R(N)-PCN-125 materials exhibit higher CO2 uptake and heat of adsorption when compared to pristine PCN-125, although the specific surface area is lower, and NH2(1)-PCN-125 has a higher CO2 uptake at 130 mmHg than pristine PCN-125. The CO2 uptake of NH2(1)-PCN-125 was 30% higher than that of pristine PCN-125, as shown in Fig. 8a. The functionalized MOF obtained using this strategy has a better effect on CO2 adsorption when compared with conventional surface modification methods.
Fig. 8 (a) Mixed-ligand strategy to construct hierarchically porous MOFs that enhance CO2 adsorption.125 Reprinted with permission from ref. 125. Copyright 2012, American Chemical Society. (b) Modulator strategy to construct hierarchically porous MOFs that enhance CO2 adsorption.21 Reprinted with permission from ref. 21. Copyright 2011, American Chemical Society. (c) MOF composites to enhance SO2 adsorption.136 Reprinted with permission from ref. 136. Copyright 2016, American Chemical Society. |
Introducing defects into MOFs generally improves their ability to adsorb CO2. For example, Wu et al. reported a general gas matching exchange (VPLE) method for the post-synthesis modification of MOFs.126 Using VPLE, ligands with functional groups can be inserted into MOF via ligand exchange. VPLE also allows multi-stage operations to obtain MOF materials with multiple ligands and functional groups. Among them, ZIF-8/Br exhibits a high CO2/N2 adsorption selectivity of 31.1 and CH4/N2 of 10.8, which are higher than the original ZIF-8 material. Mao et al. developed a ligand-assisted etching process for the template-free synthesis of hierarchical single-crystal HKUST-1 at 313 K.127 It was shown that the CO2 adsorption of hierarchically porous HKUST-1 is ∼1.5 times higher than the original HKUST-1 material at 223 K and 20 kPa. Even at 323 K and 20 kPa, the CO2 adsorption was 25% higher than that of pristine HKUST-1. Moreover, the mesoporous structure enhances the CO2 diffusion and mass transfer rate. In addition, modulator and template strategies have been used to enhance the CO2 adsorption capacity.21,128,129 For example, Choi et al. demonstrated the highly crystalline hierarchical structures of spng-MOF-5 (sponge structure) and pmg-MOF-5 (pomegranate structure) and investigated the CO2 adsorption properties of these hierarchically porous MOFs.21 Although the surface area of pmg-MOF-5 (3230 m2 g−1) was smaller than that of MOF-5 (3410 m2 g−1), pmg-MOF-5 (CO2: 2.0 g g−1 at 195 K and 760 torr) adsorbs more CO2 than MOF-5 (CO2: 1.5 g g−1 at 195 K and 760 torr). When comparing the adsorption isotherms of CO2 at 195 K, pmg-MOF-5 forms an adsorption hysteresis loop between 160 torr and 220 torr that was not observed for MOF-5 and the amount of CO2 adsorbed was significantly higher, as shown in Fig. 8b. In situ synchrotron powder X-ray diffraction analysis of the gas adsorption process showed that pmg-MOF-5 has scattered X-ray peaks corresponding to CO2 adsorbed in the micropores and mesopores. In contrast, MOF-5 has scattered X-ray peaks for CO2 adsorbed in the micropore region only. The additional CO2 adsorption in the mesopores and macropores of pmg-MOF-5 was confirmed. Wang et al. obtained hierarchically porous HKUST-1 with micro–meso–macropores upon adding compressed CO2 during the synthesis process.130 The obtained hierarchically porous HKUST-1 can be used for the selective adsorption of CO2 and CH4. The results showed that the adsorption of CH4 increased compared to the original HKUST-1 material, but was much lower than the increased adsorption of CO2. Due to the larger quadrupole moment of CO2 molecules and lower adsorption energy of CH4 molecules at the Cu sites, CO2 has a higher adsorption capacity than CH4.131
In addition, MOFs can also be combined with porous materials to enhance their adsorption capacity for CO2. Rosi et al. reported a core–shell MOF with a hierarchical structure comprised of a bio-MOF-11/14 hybrid core and bio-MOF-14 shell used for CO2 adsorption.132 The core–shell MOF comprised of bio-MOF-11/14@bio-MOF-14 showed 30% higher CO2 adsorption than bio-MOF-14, but lower N2 adsorption than the bio-MOF-11/14 core. When the core–shell structure was disrupted by grinding the fractured microcrystals, the N2 adsorption amount doubled, but the CO2 adsorption capacity remained the same. It was shown that the bio-MOF-14 shell has a significant inhibitory effect on N2 adsorption and no effect on CO2 adsorption, combining the advantages of bio-MOF-11 (high CO2 adsorption capacity) and bio-MOF-14 (high CO2/N2 selectivity and water stability). Singh et al. reported the core–shell synthesis ZIF-8@ZIF-67 and ZIF-67@ZIF-8 with enhanced H2 storage performance (2.03 and 1.69 wt%) at 77 K and 1 bar, which was approximately 41 and 18% higher than that of ZIF-8, respectively. In addition, the CO2 uptake of the core–shell ZIF-8@ZIF-67 structure (1.67 mmol g−1, 7.35 wt%) was two times higher than that of the core ZIF-8 material (0.83 mmol g−1, 3.65 wt%) and also higher than that of the shell ZIF-67 material (1.11 mmol g−1, 4.91 wt%).133 Kim et al. prepared a graphene/ZIF-8 nanocomposite with tunable porosity and surface area, where the distribution of micro–mesopores and the crystal size of ZIF-8 can be controlled by simply varying the annealing temperature of graphene oxide (GO).134 The specific surface area of GO/ZIF-8 was 720.0 m2 g−1 and the CO2 adsorption at 35 bar and 303 K was 17 mmol g−1, which is much higher than the CO2 adsorption measured under the same conditions for ZIF-8 with a specific surface area of 1871 m2 g−1 (8.5 mmol g−1), indicating that the synergistic effect between graphene and ZIF-8 can enhance the CO2 adsorption capacity, while the gas molecules can be rapidly mass transferred to the adsorbent via the mesopores. Pressure swing adsorption (PSA) usually requires relatively high pressure to capture and store CO2, so the GO/ZIF-8 nanocomposite is more favorable for application in PSA than ZIF-8.
Hierarchically porous MOFs with macroporous structures are generally not used for gas adsorption, but are typically used for the adsorption of organic molecules and catalysis. Hierarchically porous MOFs are widely used to enhance the gas adsorption capacity. For example, Wisser et al. used chitinous material obtained from sponges as a carrier for MOFs and this composite has a hierarchically porous structure.135 The specific surface area of the composite prepared with a 53% (w/w) loading of HKUST-1 was up to 800 m2 g−1, where the structure of the mesopores depends on the structure of the chitinous pores. Ammonia penetration experiments showed the good adsorption kinetics of the composites. Moreover, SO2 is one of the major gaseous pollutants harmful to the ecosystem. Zhang et al. adsorbed SO2 by compositing MOFs with polyacrylonitrile (PAN) into hierarchically porous materials.136 Because MOFs have more binding sites than PAN, the MOF-199/PAN and UiO-66-NH2/PAN composites adsorb SO2 more effectively when compared to pure PAN, as shown in Fig. 8c. The N2 isotherms are typical type II curves. Penetration tests indicate that the hierarchically porous structures in the composites enhance the mass transfer rate. In addition to the gases mentioned above, PM (particulate matter) is one of the main gaseous pollutants found in the environment.137 Wang and co-workers reported a roll-to-roll hot-pressing method for the large-scale production of different MOF composites used for PM adsorption.138 The PM removal efficiency of ZIF-8@Plastic with a hierarchically porous structure was improved by a factor of four when compared to the original plastic mesh.
Fig. 9 (a) Hierarchically porous UiO-66-NH2 aerogel adsorbs Pb(II).25 Reprinted with permission from ref. 25. Copyright 2020, Elsevier. (b) Core–shell structured MOF adsorbs macromolecules.145 Reprinted with permission from ref. 145. Copyright 2019, Wiley. (c) Hierarchically porous UiO-66-NH2 improves catalytic efficiency.27 Reprinted with permission from ref. 27. Copyright 2019, Wiley. |
Scale | Strategy | MOFs | Metal with linker, additives | Pore diameter (nm) | Control methods | Application | Ref. |
---|---|---|---|---|---|---|---|
BDC = 1,4-benzenedicarboxylate; BTC = 1,3,5-benzenetricarboxylate; DOT = 2,5-dioxidoterephthalate; BPDC = 4,4′-biphenyl-dicarboxylate; TPTC = terphenyltetracarboxylate; isoph = isophthalic acid; MeIM = 2-methylimidazole; IIM = 4-iodoimidazole; BrIM = 4-bromoimidazole; ClBIM = 2-chloromethylbenzimidazole; AZDC = azobenzene-4,4′-dicarboxylate; CBAB = 4-carboxybenzylidene-4-aminobenzate; TCPP = tetrakis(4-carboxyphenyl)porphyrin; CoPc = cobalt tetra(carboxy)phthalocyanine; H3hmtt = methyl substituted truxene tricarboxylic acid; VB12 = Vitamin B12, Cyt c = cytochrome c, HRP = horseradish peroxidase; BTB = 1,3,5-benzenetribenzoic acid; ABTC = 3,3′,5,5′-azobenzenetetracarboxylic; OmimBF4 = 1-octyl-3-methylimidazolium tetrafluoroborate; PS = polystyrene; TPCB-OH = 3,3′,5,5′-tetrakis(4-carboxyphenyl)-4,4′-dihydroxybiphenyl; TBAPy = 1,3,6,8-tetrakis(4-benzoicacid)pyrene; NDC = naphthalene dicarboxylate; PMMA = poly(methyl methacrylate); dabco = 1,4-diazabicyclo[2.2.2]octane; H2BuDC = 5-tert-butylisophthalic acid; FDC = fumaric acid; ADC = 5-aminoisophthalic acid; TMB = 1,3,5-trimethylbenzene; TOCNF = anionic 2,2,6,6-tetramethylpiperidine-1-oxylradical-mediated oxidized cellulose nanofibers. | |||||||
Micro–micro porous | Mix metals | UiO-66 | Zr, Ti with BDC | 0.5–1 | The molar ratio of metals mixture | CO2 absorption | 48 |
MOF-74 | Mg, Ca, Sr, Ba, Mn, Fe, Co, Ni, Zn, Cd with DOT | — | The molar ratio of metals mixture | — | 49 | ||
Mix linkers | MOF-5 | Zn with BDC, BDC-NH2, BDC-Br, BDC-(Cl)2, BDC-NO2, BDC-(CH3)2, BDC-C4H4, BDC-(OC3H5)2, BDC-(OC7H7)2 | 0.5–1.1 | The molar ratio of linkers mixture | CO2, CO absorption | 50 | |
MOF-5 | Zn with BDC, BDC-NH2 | — | The molar ratio of linkers mixture | Benzene, toluene, para-xylene, meta-xylene, ortho-xylene diffusion | 46 | ||
UiO-66 | Zr with BDC, L-serine | — | The molar ratio of linkers mixture | CO2 absorption | 53 | ||
ZIF-8 | Zn with MeIM, IIM, BrIM, ClBIM | — | Type of functional groups | CO2/N2, CH4/N2 adsorption | 126 | ||
Micro–mesoporous | Mix linkers | PCN-125 | Cu with TPTC, H-isoph, CH3-isoph, NH2-isoph, CH2NH2-isoph, NO2-isoph, SO3H-isoph, SO3Na-isoph | 2–30 | The molar ratio of linkers mixture | CO2 absorption | 125 |
Ru3(btc)2Cl1.5 | Ru with BTC, pyridine-3,5-dicarboxylate | — | The molar ratio of linkers mixture | CO2, CO and H2 absorption; catalytic hydrogenation of 1-octene | 52 | ||
MIL-125 | Ti with BDC, BDC-NH2 | 2–50 | The molar ratio of linkers mixture | Toluene adsorption; photodegradation process of toluene | 51 | ||
Post-treatment | PCN-160 | Zr with AZDC, CBAB | 2.5–19 | Exchange ratio and acetic acid concentration | accessibility of enzymes; oxidation of ABTS and o-phenylenediamine (o-PDA) | 58 | |
UiO-66, UiO-67, MOF-5, MIL-125(Ti), MIL-53(Fe) | Zr with BDC; Zr with BPDC; Zn with BDC; Ti with BDC; Fe with BDC | 5.5–13 | Thermolabile linker ratio, temperature and heat time | Meerwein–Ponndorf–Verley (MPV) reaction | 59 | ||
UiO-66, ZIF-8, UiO-67, UiO-66(Hf) | Zr with BDC, TCPP, CoPc; Zr with BPDC; Zn with MeIM; Hf with BDC | 2–4.5 | The time of laser exposure and ratio of photolabile/robust linkers | — | 152 | ||
POST-66(Y), POST-66(Dy), POST-66(Tb), MOF-177(Zn), UiO-67(Zr), MIL-100(Al), | Y with H3hmtt; Dy with H3hmtt; Tb with H3hmtt; Zn with BTB; Zr with BPDC; Al with BTC | 3.8 and 6.2–16 | Etching temperature, time | VB12, Cyt c, myoglobin, HRP absorption; catalytic activity of co-oxidation of 4-aminoantiprine (4-AAP) and phenol | 61 | ||
HKUST-1 | Cu with BTC | 3–30 | Treatment time | Thiophene adsorption | 62 | ||
HKUST-1 | Cu with BTC, hydroquinone | 3.8 and (2–50) | — | CO adsorption | 63 | ||
HKUST-1 | Cu with BTC, NH3 gas | 9.2–38 | Etching temperature and amount of the etchant | Encapsulation of methylene blue | 23 | ||
HKUST-1, ZIF-67 | Cu with BTC; Co with MeIM | 3.2 and (2–50) | — | Cycloaddition of CO2 with epoxides to produce cyclic carbonates | 149 | ||
PCN-250 | Fe, Ni, Mn, Co, Al, In, Sc with ABTC | 4 and (2–50) | Type of metal clusters and decarboxylation temperature | CH4, CO2 adsorption | 150 | ||
UiO-66, UiO-67, MOF-808, MIL-53 | Zr with BDC, sodium acetate, sodium formate, sodium propionate, sodium benzoate; Zr with BPDC; Zr with BTC; Al with BDC | 2–30 | Type and amount of solution | CO2 absorption; methyl orange absorption; catalytic cycloaddition reactions between CO2 and epoxides with different sizes, one-pot cascade synthesis of secondary arylamines | 57 | ||
Modulator | UiO-66, UiO-66-NH2, UiO-66-NO2, UiO-67, MIL-53, MIL-53-NH2, DUT-5, MOF-808 | Zr with BDC, acetic acid, octanoic acid, dodecanoic acid, palmitic acid; Zr with NH2-BDC; Zr with NO2-BDC; Zr with BPDC; Al with BDC; Al with NH2-BDC; Al with BPDC; Zr with BTC | 2–8 | Molar ratio of reactant and length of modulator | Coomassie brilliant blue R250 absorption; ring opening of styrene oxide with methanol, cycloaddition reactions between 1,3-cyclohexanedione and α,β-unsaturated aldehydes | 67 | |
UiO-66 | Zr with BDC, acetic acid | 1.7 and 16 | Process of crystallization and concentration of modulator | Isomerization of glucose to fructose | 64 | ||
UiO-66 | Zr with BDC, trifluoroacetic acid | 1.6 and >20 | Aging time | Cyclohexanone conversion to cyclohexanol | 69 | ||
UiO-66-NH2 | Zr with NH2-BDC, acetic acid | 1.5–50 | Amount of the modulator | Photocatalytic H2 from water splitting | 27 | ||
HKUST-1 | Cu with BTC, acetic acid | 2–15 and >25 | Amount of the modulator | CH4 absorption | 65 | ||
HKUST-1 | Cu with BTC, benzoic acid | ∼14.8 | — | — | 68 | ||
PCN-250 | Fe with ABTC, propanoic acid, hexanoic acid, nonanoic acid, myristic acid | 4–18 | The length and concentration of the fatty acid | Methylene blue absorption | 66 | ||
Template | HKUST-1 | Cu with BTC, CTAB | 3.8–31 | Molar ratio of template | — | 70 | |
HKUST-1 | Cu with BTC, CTAB, citric acid | 3–30 | Molar ratio of template | — | 72 | ||
HKUST-1 | Cu with BTC, CTAB, 1,3,5-trimethyl benzene | 4–25 | Type of template and molar ratio of template | CO2 adsorption; oxidation of benzyl alcohol to benzaldehyde | 74 | ||
MIL-100, MIL-96, MIL-110 | Al with BTC, CTAB | 3–33 | pH value and ratio of solvent | — | 71 | ||
MFM-100, HKUST-1, MOF-2 | Cu with biphenyl-3,3′,5,5′-tetracarboxylic acid, OmimBF4; Cu with BTC; Zn with BDC | ∼4.7 | Molar ratio of template and electrosynthesis | Catalytic activity of alcohol oxidation | 76 | ||
HKUST-1, Mn-BTC, Mn-BDC | Cu with BTC, CO2-expanded liquid; Mn with BTC; Mn with BDC | 13–23 | Molar ratio of template (CO2 pressure) and type of solvent | Oxidation of benzyl alcohol to benzaldehyde | 77 | ||
UiO-66, HKUST-1 | Zr with BDC, poly(ethylene glycol)-based alkylammonium and bromide ionic liquid; Cu with BTC | 2–50 | Temperature and time of synthesis | Encapsulation of Cyt c | 151 | ||
Nucleation kinetics | UiO-66(Hf), UiO-66, UiO-66-OH2(Hf), UiO-66-NH2(Hf), NUS-6(Hf) | Hf with BDC; Zr with BDC; Hf with OH2-BDC; Hf with NH2-BDC; Hf with SO3Na-BDC | 3.7–16.6 | Molar ratio of solvent | Tetrakis (4-carboxyphenyl)-porphyrin absorption; catalytic methanolysis of styrene oxide | 80 | |
Zn-MOF-74 | Zn with DOT | 5–20 | Time of synthesis | Brilliant Blue R-250 absorption | 78 | ||
MOF composites | PCN-222@PS, PCN-160(Zr)@PS, PCN-224(Zr)@PS, MOF-801(Zr)@PS, PCN-900(Eu)@PS | Zr with TCPP, Zn with BDC, tetrasodium ethylenediaminetetraacetate, PS; Zr with AZDC; Zr with TCPP; Zr with fumaric acid; Eu with TCPP | — | — | — | 84 | |
PCN-222/PCN-608, /NU-1000, /PCN-134, Zr-BTB/PCN-134 | Zr with TCPP, TCPB-OH; Zr with TBAPy; Zr with TCPP, BTB | <5 | — | — | 82 | ||
PCN-222@UiO-67, PCN-134@Zr-BTB, PCN-222@Nu-1000, PCN-222@Zr-AZDC, PCN-222@Zr-NDC, La-TCPP@La-BPDC | Zn with TCPP, Zr with BPDC; Zr with BTB; Zr with TBAPy; Zr with AZDC; Zr with NDC; La with TCPP, La with BPDC | <5 | — | Catalyze epoxidation of alkenes | 88 | ||
Co, Fe-MOF-74/Co/carbon cloth | Co, Fe with DOT, carbon cloth | — | Molar ratio of metal salt | Electrocatalytic splitting water | 85 | ||
PMMA@IRMOF-3@MOF-5 | Zn with BDC, BDC-NH2, PMMA | — | Time of polymerization | — | 86 | ||
UiO-66@ZIF-8, Pd-UiO-66-NH2@ZIF-8 | Zr with BDC, Zn with MeIM, CTAB, PVP, SiO2; Pd, Zr with BDC-NH2, Zn with MeIM | — | — | Ethylene hydrogenation | 83 | ||
NH2-UiO-66@NH2-MIL-125, MOF-76@NH2-MIL-125, MIL-101@NH2-MIL-125, NH2-UiO-66 & MOF-76@NH2-MIL-125 | Zr, Ti with BDC-NH2, PVP; Tb with BTC; Cr with BDC | — | — | CrVI adsorption; photocatalytic CrVI reduction | 26 | ||
ZIF-67@ZIF-8 | Zn, Co with MeIM | — | — | H2 storage, CO2 adsorption | 133 | ||
M2(ndc)2(dabco) (M = Co, Ni, Cu, Zn) | Co, Ni, Cu, Zn with NDC, dabco | 5–>20 | — | Methane, ethane, ethene, propane, propene adsorption | 119 | ||
ZIF-67@Yeast | Co with MeIM, yeast | — | — | Pb2+ adsorption | 143 | ||
Polyacrylonitrile/UiO-66 | Zr with BDC, polyacrylonitrile | 2.5–10 | — | H2 storage | 122 | ||
Chitin fibers/HKUST-1 | Cu with BTC, chitin fibers | — | — | NH3 adsorption | 135 | ||
ZIF-8/PAN, Mg-MOF-74/PAN, UiO-66-NH2/PAN, MOF-199/PAN | Zn with MeIM, PAN; Mg with DOT; Zr with H2N-BDC; Cu with BTC | — | — | Particulate matter, SO2 adsorption | 136 | ||
Gelation | Cr, Fe-BTC, BDC, NDC, ADC, FDC, BuDC, BTB | Cr, Fe with BTC, BDC, NDC, ADC, FDC, BuDC, BTB, BPDC | 3–33 | Molar ratio of reactant, temperature | Methyl orange, dimethyl phthalate, methylene blue adsorption | 90 | |
Micro–macroporous | Template | ZIF-8 | Zn with MeIM, PS | 190–470 | Particle size of template | Knoevenagel reaction between benzaldehydes and malononitriles | 28 |
3D printing | MOF-5 | Cu with BTC, acrylonitrile butadiene styrene | (>50) | — | H2 storage | 112 | |
Micro–meso–macroporous | Post-treatment | UiO-66 | Zr with BDC, acetate, formate, propionate, benzoate | 3–70 | The length of the monocarboxylic acid, etching temperature and amount of the etchant | Cytochrome c, horse radish peroxidase absorption | 60 |
HKUST-1, MOF-5 | Cu with BTC; Zn with BDC | 30–260 | Etching temperature, time, molar ratio of solvent | CO2 adsorption | 127 | ||
Modulator | MOF-5 | Zn with BDC, 4-(dodecyloxy)benzoic acid | 10–100 | Molar ratio of reactant | CO2 absorption | 21 | |
Template | HKUST-1, Al-MIL-96 | Cu with BTC, CO2-expanded liquid; Al with BTC, F127 | ∼25 and 68–106 | Molar ratio of template (CO2 pressure), molar ratio of solvent | CH4, CO2 adsorption; methylene blue adsorption | 130 | |
HKUST-1, CuBDC | Cu with BTC, PS, citrate/CTAB; Cu with BDC | ∼20 and 80–400 | Particle size of template | Methylene blue adsorption; accelerate Friedländer reaction between 2-aminobenzophenones and acetylacetone, hydrolyse the ester in water | 147 | ||
MIL-101 | Cr with BDC, CTAB | 5–100 | Molar ratio of template | Methene blue adsorption | 123 | ||
ZIF-8, ZIF-90 | Zn with MeIM, N,N-diethylethanolamine; Zn with imidazole-2-carboxyaldehyde | 5–100 | Molar ratio of template | CO2 adsorption | 128 | ||
UiO-66(Ce) | Ce with BDC, P123/F127 | 20–110 and >1000 | Molar ratio of template | DNA adsorption; catalyze the hydrolysis of the high-energy phosphate bonds | 24 | ||
Cu-BDC | Cu with BDC, PS, P123 | 3.9 and 455 | Particle size of template | CO2 adsorption; CO2 carbonylative coupling reaction with 4-methylbenzyl chloride | 129 | ||
Nucleation kinetics | ZIF-8 | Zn with MeIM | 10–60 | Molar ratio of reactant | CO2 adsorption; methylene blue, methyl blue, rhodamine B absorption | 124 | |
HKUST-1 | Cu with BTC | 26–72 | Temperature of synthesis | Electrocatalytic oxygen reduction reaction | 79 | ||
MOF composites | Graphene/ZIF-8 | Zn with MeIM, graphene | 3.7 and 10–100 | Temperature of thermally annealed | CO2 adsorption | 134 | |
Gelation | MIL-100(Fe) | Fe with BTC | 3–4.5 and >50 | Molar ratio of linkers | CH4 adsorption | 89 | |
MIL-53(Al), Al-BTC | Al with BDC, CTAB, TMB; Al with BTC | 5.4–87.8 | Gelation temperature, reactant concentration | H2 storage, CO2 adsorption; benzene, n-hexane, methanol adsorption; congo red, brilliant blue R-250 adsorption | 22 | ||
UiO-66-NH2 | Zr with BDC-NH2 | — | — | Pb2+ | 25 | ||
Mechano-synthesis | MOF-5 | Zn with BDC | <200 | Molar ratio of reactant, grinding speed and time | Methane, ethane, propane, n-butane, n-pentane, n-hexane, n-heptane absorption | 92 | |
HKUST-1 | Cu with BTC | — | Molar ratio of solvent, treatment time | Benzene absorption | 93 | ||
HKUST-1 | Cu with BTC, NaCl, KCl, triethylenediamine | 2–100 | Molar ratio of reactant | I2 adsorption | 99 | ||
ZIF-8 | Zn with MeIM | 10–100 | Molar ratio of reactant | n-Butanol and hexane adsorption; rhodamine B adsorption | 100 | ||
ZIF-1, ZIF-3, ZIF-4, ZIF-8 | Zn with imidazole; Zn with MeIM | 10–100 | — | — | 96 | ||
Zn-MOF-74 | Zn with DOT | — | Type of solvent | — | 101 | ||
3D printing | HKUST-1 | Cu with BTC | (2–50 and >50) | — | CH4 adsorption | 120 | |
HKUST-1 | Cu with BTC, PVA, bentonite clay, methyl-cellulose | 2.9–9.1 and (50) | Synthesis temperature and activation solvent | CO2 adsorption | 108 | ||
MOF-74(Ni), UTSA-16(Co) | Ni with DOT, bentonite clay, poly(vinyl alcohol); Co with citric acid | 10–50 and (>50) | — | CO2 adsorption | 106 | ||
ZIF-8 (MIL-100(Fe)) | Zn with MeIM, TOCNF, curcumin, alginate | 2–50 and (>50) | MOF contents | Curcumin release | 110 |
Methods that introduce pores whose pore sizes are usually micropores mainly replacing part of the metal or ligand in the MOF. Usually, the effect of replacing the metal on the MOF pore size is minimal. Still, this approach changes the active sites and has an impact on the adsorption performance. Under the condition of ensuring the same topology, the choice of the replacement ligand mostly involves the use of a derivatized ligand bearing different functional groups, so the change in the pore size is usually smaller than the original pore. However, mixed ligands sometimes have the phenomenon of missing ligands, which cannot only improve diffusion, but also introduce different functional groups to change the active sites. There is still a lack of sufficient knowledge to apply this approach to construct mesopores. There are three mechanisms in the methods used to introduce mesopores, including the introduction of defects, nucleation kinetics, and combination of other porous materials (including MOFs). Constructing mesoporous MOFs by introducing defects usually requires the MOFs to be structurally stable before and after the introduction of the defects in order to obtain homogeneous pore sizes. The strategies used to form mesopores via introducing defects include post-treatment, modulator, and template strategies. Among them, post-treatment can easily construct hierarchically porous structures in different MOFs. The modulator strategy is often applied to partial MOFs bearing carboxylic acid ligands. Although the template strategy usually forms relatively uniform mesopores, the problem of repelling the templates during the crystal growth process of MOFs and the additional steps required to remove the template need to be addressed (the removal of the template may lead to the collapse of the uniform mesopores formed during this process). The nucleation kinetics strategy enables the formation of tunable mesopores by changing the synthesis conditions, but at the expense of the MOF crystallinity. In addition, MOFs composites can combine the characteristics of two porous materials in addition to introducing different pore sizes, e.g., MOF composites with carbon materials can significantly enhance the CO2 adsorption performance. Methods that usually introduce macropores mainly include mechanosynthesis and gelation. The mechanosynthesis strategy is a green and economic approach that can synthesize MOFs in large quantities. MOFs gelation and 3D printed MOFs are both molding processes. In addition, the 3D printing strategy can fabricate MOFs monolithic materials and precisely control the pore size of the macropores. It is expected that after 3D printing technology is upgraded, work on 3D printed and accurately customized MOFs structures will be seen.
Despite the extensive work fabricating hierarchically porous MOFs, three main challenges still need to be addressed. One is to meet the requirements of their industrial production and find a general method for the green synthesis and reproducible construction of hierarchically porous MOFs. The second is to synthesize hierarchically porous MOFs with uniform and highly ordered porous structures. The third is the effect of activation on the structural stability of the hierarchically porous MOFs. In addition, there are few reports on the construction of hierarchically porous MOFs with multiple strategies and it is speculated that micro–meso–macroporous MOF materials can be constructed through the combination of multiple strategies. In conclusion, the extensive application of hierarchically porous MOFs in adsorption-diffusion validates that the strategy of constructing hierarchically porous MOFs to improve the mass transfer rate of microporous MOFs is feasible.
This journal is © the Partner Organisations 2022 |