Hierarchical Metal–Organic Frameworks with Macroporosity: Synthesis, Achievements, and Challenges
The advantages of macroporous metal–organic frameworks (MOFs) in comparison with micro- and mesoporous MOFs are discussed.
A range of synthetic methods for the fabrication and characterisation of hierarchical MOFs with macroporosity are reviewed.
The applications, advancements, and challenges of each method are compared and assessed in detail.
KeywordsMetal–organic frameworks Hierarchical Macroporous Composites
The interactions of gases and liquids with porous materials have been a great source of inspiration for scientists tackling problems related to gas separation [1, 2, 3, 4], energy storage [5, 6, 7, 8], drug delivery [9, 10, 11, 12], and catalysis [13, 14, 15, 16]. Pores, which can be discrete or can form complete pathways or channels from one surface to another, play a crucial role in gas adsorption and fluid dynamics . Depending on the size of their pores or channels, materials can be classified as microporous (pore diameters of less than 2 nm), mesoporous (pore diameters between 2 and 50 nm), or macroporous (pore diameters above 50 nm) . There has been a recent trend towards the development of new porous materials incorporating pores with different size regimes to form hierarchical systems with interconnected pores and entirely new properties for desired applications. While the concept of introducing meso- and macropores into microporous materials has been extensively investigated in zeolitic porous systems [19, 20, 21], development of metal–organic frameworks (MOFs) with open framework structures comprised of metal nodes and organic ligands has seen a rapid expansion of recent research interest. Advances in synthetic chemistry have reported numerous MOF structures (> 70,000 structures reported so far ) with potential use in gas storage [23, 24, 25], gas separation [26, 27, 28, 29], catalysis [30, 31], carbon dioxide capture [32, 33, 34], and as semiconductor materials [35, 36]. As the vast majority of MOF materials are microporous, great interest has developed in the creation of multiple porosities in these materials. A number of excellent reviews on hierarchical MOFs have recently been published [37, 38] and focused on MOFs containing micro- and mesoporosity. The current review emphasises the recent achievements in the development of specifically macroporous hierarchical MOF structures which, though important, are under-represented in the MOF field. Macropores are desirable in hierarchical porous materials for enabling faster molecular diffusion and mass transfer, which is especially important for applications involving large molecules, viscous solutions or applications involving high throughput with low pressure gradients.
The benefits of a hierarchical structure for catalytic applications have been well established in purely inorganic systems. For example, hierarchical mesoporous–macroporous structures in high internal phase emulsion (HIPE) template halloysite nanotubes showed greatly improved catalytic activity for the conversion of cellulose to 5-hydroxymethylfurfural because of the catalyst’s increased permeability and mass transfer efficiency . Similarly, the macroporous nature of the support in hierarchical flower-like TiO2 superstructures helps in facilitating more efficient photo-degradation of large dye molecules such as methyl orange . The degradation of Rhodamine B under visible light irradiation using hierarchical core–shell Fe3O4/WO3 was tested by Xi et al. , showing that the combination of mesoporous and macroporous regions enhances the photocatalytic activity with almost 100% decomposition achieved after 90 min compared to < 20% and 50% for the iron(iii) oxide core and tungsten(vi) oxide, respectively. In this case, the large voids between the Fe3O4 and WO3 in the hierarchical structure were deemed to synergistically reduce the recombination rate of photogenerated electron–hole pairs, further improving the photocatalytic degradation. The importance of pore size in directing catalysis was further illustrated by Feng et al.  using a hierarchical MOF composite comprised of PCN-222 (also known as MOF-545) and ZIF-8 to achieve size-selective catalysis. The PCN-222(Fe)@ZIF-8 composite was exposed to two molecules sensitive to oxidation by PCN-222, namely o-phenylenediamine (o-PDA) (0.5 × 0.5 nm) and 2,2′- azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) (0.7 × 1.6 nm). While PCN-222 successfully catalysed the transformation of o-PDA, the catalysis of the larger ABTS molecule was hindered due to constricted diffusion through the small pore size window of ZIF-8. Thus, considering the added value of hierarchical structure in these inorganic systems, the ability to fabricate hierarchically porous MOF structures may be similarly beneficial when considering future practical applications.
The nature of the porosity in a material can be determined by various characterisation techniques, which can be generally grouped into use of fluid penetration, scattering, and imaging. Fluid penetration techniques involve the filling of pores with a liquid or a gas, to obtain information on the pore sizes and volumes. For MOFs (conventionally microporous materials), gas sorption (generally with nitrogen at 77 K up to 1 bar pressure) is the most common characterisation method to establish the specific surface area, pore volume, and pore size distributions. To obtain this information, the gas sorption isotherms are modelled using standard and widely accepted approaches such as the Brunauer–Emmett–Teller (BET) method for calculating surface area , density functional theory (DFT) and the Horvath–Kawazoe (H–K) method for calculation of the pore size distribution in the micropore region, and the Barrett–Joyner–Halenda (BJH) for mesopore distribution [44, 45]. However, nitrogen sorption at 77 K will typically not provide information on macropores larger than 100 nm . For materials containing pores in the mesopore and macropore range (~ 4 nm–60 µm), an alternative method is mercury intrusion porosimetry, which uses non-wetting mercury to penetrate the pores under pressure [44, 46]. While this method can produce a macroporous size distribution in robust porous materials, it has the disadvantages of using a toxic compound (mercury) for the characterisation, cannot be used on soft or deformable structures, and (unlike gas sorption) is destructive to the sample.
Scattering techniques, which involve bombarding the sample with, for example, X-rays, neutrons or electrons to obtain a pattern, can be non-destructive while still enabling characterisation of the porosity. While powder X-ray diffraction (PXRD) and single-crystal X-ray diffraction (SXD) are commonly used to study the microstructures of crystalline porous materials, small angle X-ray scattering (SAXS) can be used to probe the variations in scattering length density which occur over distances exceeding typical interatomic spacing to determine larger particle or pore sizes. Hence, SAXS can be used to calculate pore size distributions in porous materials (1–100 nm range) and can be used for both crystalline and non-crystalline materials, and for materials having regular (but perhaps non-crystalline or disordered) porosity . Both neutron scattering and X-ray scattering can be used to provide information on the pore dimensions in a bulk sample, unlike fluid penetration, which can only give information on interconnected pores that are accessible to penetration by the fluid (“open” porosity), scattering can provide information on isolated or “closed” pores in the material. However, for scattering techniques to be used for characterisation of porosity, there needs to be appropriate scattering length contrast between the pores and the surrounding material, which can be difficult in the case of X-ray scattering from MOFs, where the voids are often filled with air and the materials contain predominantly light organic materials such as C or H. Increased contrast may be achieved by the filling of the pores with a suitable contrast agent , which may have the undesirable effect of distorting the pores, for example, in the case of flexible MOFs, or where there is a soft, compliant matrix, such as a hydrogel. In specific cases, where the scattering length density contrast is too small for X-rays, small angle neutron scattering (SANS) can be used . As neutron scattering length varies independent of atomic number, even light atoms such as hydrogen and carbon can be easily distinguished using neutrons , but the need for a neutron source means this is a far less widely used technique.
As can be seen throughout the literature, imaging is by far the most common technique to visualise the macropores appearing in MOFs due to widespread accessibility of the technique and ease of use. Microscopy can provide valuable insight into the shape and spatial distribution of pores. For small macropores (e.g. below 10 µm), scanning electron microscopy (SEM) and transmission electron microscopy (TEM) can be used to image porosity and can be performed in a very straightforward way on a very small amount of sample. However, these imaging methods are not bulk techniques (with SEM in particular being restricted to definition of surface features) and can only observe a small number of particles and therefore may not be representative of the bulk porosity. For imaging of larger macropores (from a few tens of microns to millimetres in size), optical microscopy can be useful, but cannot provide information on the connectivity or tortuosity of the interior pore network in monolithic structures. The ability of X-rays to penetrate light materials can be used in X-ray computed tomography (CT), whereby a series of 2D X-ray images of a material are computationally compiled and reassembled to obtain a 3D image . X-ray CT can thus be used to image large pores in monolithic structures, though the resolution of such techniques is currently limited to a few tens of microns . Thus, it can be seen that few characterisation techniques can span the full range of pore sizes. Therefore, as will be shown in this review, due to the complexity of such materials, a combination of several techniques may be needed to obtain a full description of the macroporosity in hierarchical MOF structures.
The routes to formation of such complex structures are equally varied. In terms of approaches to increase pore dimensions in MOFs using synthetic chemistry, the typical methods used in MOF synthesis to produce mesopores (such as extending ligand lengths or enlarging building blocks) are of limited use in the creation of much larger macropores. This is due to the difficulties in stabilising these pores against collapse upon desolvation during activation . Hence, the formation of macroporosity (rather than mesoporosity) in MOFs necessitates the use of very different fabrication approaches. In this review, we focus on the synthetic strategies and challenges of creating MOFs with pore structures containing macropores and provide readers with a wider scope of the strategies available for fabrication of hierarchical macroporous MOFs.
A summary of preparation techniques, crystallisation routes, and templates for the formation of hierarchical MOFs with macroporosity
Macropore formation route
Templates, substrates, modulators or reagents used
MOFs prepared using these methods
Hard structural templating
Macroporous carbon derived from kenaf stem, polymer-derived ceramic (PDC) foam, oxide-bonded silicon carbide and alumina (Al2O3), ceramic foams (CF), copper foam
HKUST-1, CAU-10, UiO-66 (Zr), MOF-5
Soft structural templating
Cellulose solution, konjac glucomannan, graphene oxide, melamine foam (MF), three-dimensional polystyrene (PS), N,N-dimethyloctylamine (DMOA), polyacrylamide (PAAm)
ZIF-8, ZIF-9, ZIF-12, PCN-224, HKUST-1
Direct synthesis (linker modulation)
Alkyl chains, monocarboxylic acids
Post-synthetic treatment (acid etching)
Phenolic acid, cyanuric chloride and tetraethylamine (TEA), phosphoric acid, hydroquinone, H3BO3 and NaCl
ZIF-8, IRMOF-3, MIL-101(Fe), HKUST-1
Use of compressed or supercritical CO2
Post-synthetic treatment (scCO2 drying)
HKUST-1, AlBTC, AlBDC
Direct synthesis (expanded solvent)
N-EtFOSA/TMGT solution, DMF, DMSO/MeOH
Zn-BTC, HKUST-1, Co-BTC
Polyvinyl alcohol (PVA), trimethylolpropane propoxylate triacrylate (TMPPTA), polylactic acid (PLA)
MOF-74 (Ni), UTSA-16 (Co), UiO-66, ZIF-8
Anionic 2,2,6,6-tetramethylpiperidine-1-oxylradical-mediated oxidised cellulose nanofibers (TOCNFs)
ZIF-8 and MIL-100 (Fe)
Structural templating involves use of a template containing macroporous voids to produce a macroporous composite structure by directly growing the crystalline MOF either on or around a template. The template can be “hard” (e.g. a foam or porous monolith) or “soft” (e.g. a gel or emulsion) and can be retained or removed after templating.
Defect formation describes the disruption of the periodic crystalline structure of the MOF e.g. by incorporating a sizeable ligand during synthesis which introduces a porous defect or by selective removal of sites to produce pores, e.g. via post-synthetic acid etching of MOF single crystals.
Compressed or supercritical CO2 has been used extensively for activation of MOFs  to remove solvents and soft templates to produce porous aerogel structures, but scCO2 has also recently been shown to be useful in expanded solvent systems for producing macroporous voids during direct synthesis of MOFs.
The recent use of 3D printing combines MOFs with binders to create MOF-based inks capable of forming hierarchical monoliths on deposition. This represents a new method with potential to enable the construction of hierarchical monoliths with complex geometries.
Most of these methods can be employed either using MOF precursor solutions (termed “direct synthesis”) or using pre-formed MOFs (“post-synthetic treatment”), as shown in Table 1.
2 Preparation Techniques for Hierarchical MOFs with Macroporosity
2.1 Structural Templating Method
The most common approach to the introduction of macropores into an intrinsically microporous MOF is using structural template to form a hierarchical structure. As demonstrated for a multitude of other porous material systems, the macroscopic shape and the pore size of the resulting hierarchical structures can be well controlled through judicious choice of template [54, 55, 56, 57, 58]. The resulting hierarchical porous MOF structures could potentially exhibit improved properties such as more rapid molecular diffusion compared to the respective non-hierarchical MOFs, better mechanical or thermochemical stability, and immobilisation of nanoparticulates. Thus, they are preferentially used in some applications that require easy integration and regeneration (e.g. for adsorbent systems [59, 60]) or exceptional mass transfer (e.g. for catalyst supports ).
The macroporous MOF structures deriving from the template approach can be divided into two main categories: one is that the template remains in the final MOF structure, forming a composite material, and the other is that the template is sacrificial and removed to obtain a macroporous structure in the pure MOF (often called a MOF aerogel or a foam). Templates can be broadly grouped into “hard” and “soft” structural templates.
2.1.1 Hard Structural Templating
Hard templates may be carbons , ceramics [59, 61, 63, 64, 65], and metals [60, 66, 67, 68]). They are typically in the form of monoliths, porous membranes, or foams [60, 62, 69]. Because hard templates are difficult to be removed, the majority of these templates act as supports, resulting in macroporous composite MOF structures.
2.1.2 Soft Structural Templating
Duan et al.  reported the soft-templated synthesis of a HKUST-1 framework that contained micro-, meso-, and macropores, using a N,N-dimethyloctylamine (DMOA) surfactant as a sacrificial template. The role of the DMOA as a template was demonstrated by mesodynamics simulation in which the introduction of DMOA into the reaction mixture resulted in the formation of supramolecular micelles. Subsequently, the MOF precursors [i.e. a mixture of hydroxy double salt (Zn and Cu)] and benzenetricarboxylic acid (H3BTC) were added and self-assembled on the surface of the template micelles at room temperature and then formed the hierarchical porous HKUST–1 upon the removal of DMOA template. SEM images revealed the hierarchical nature of this MOF whereby continuous pore voids were formed between the nanoparticles. TEM further confirmed the abundance of mesopores and macropores (pore sizes of 40–100 nm) in the final product. Because of the presence of macropores which contribute to high molecular diffusion, the hierarchical MOF exhibited a high toluene storage capacity (646 mg g−1 at 298 K), representing a 25 wt% increase compared to that for the non-hierarchical HKUST-1 (516 mg g−1) . This value is also much higher than other microporous MOFs and zeolites [81, 82, 83]. The use of surfactants in the preparation of such hierarchical MOFs introduces a new way to allow the degree of porosity to be readily tuned by adjusting the amount and type of surfactant used.
The use of structural templating agents remains the most established method in the formation of macroporous MOFs due to the wide range of templates and templating agents that can be employed in this technique. As can be seen by the examples given thus far, the resulting hierarchical structures generally adopt the shape of the templating agents. It was also shown that these templates could be introduced either in the reaction solution during MOF synthesis or in the precursor solution where the prepared MOFs were dispersed. The final materials could be obtained as MOF-based composites with the template retained or as pure MOF superstructures without the template. When the templating agents are incorporated to produce the final composite MOF structures, in addition to the immobilisation of the MOF crystals, they may also contribute to improved mechanical and thermochemical stability of the MOF composites. In the case of MOFs with sacrificial templates, washing processes are required to remove the templates. In some cases (such as coordination polymer frameworks, silicates and MCM-41), it can be detrimental to the stability of the pore structures [84, 85, 86, 87]. Furthermore, incomplete removal can lead to occlusion of the pores and therefore loss of functionality for gas sorption/catalytic reactions. In terms of potential uses, the MOFs formed via this method have been utilised in various applications from catalytic applications, heat and mass transfer to extraction of volatile compounds.
2.2 Defect Formation Method
From the point of sustainable processing, the ability to form macroporosity in the absence of sacrificial templates or scaffolds would be preferable. A commonly used strategy for obtaining hierarchical porous MOFs without using templates is through direct introduction of defects into the microporous structures. This process could be carried out using either in situ synthetic methods (e.g. using linker modulation) or using post-synthetic treatment (e.g. via acid etching). Linker modulation and linked labilisation methods predominantly produce mesoporous defects in microporous MOFs. Yuan et al. demonstrated using a pro-labile organic linker for the synthesis of PCN-160 MOF. The defect in the pre-formed MOF was introduced by splitting this pro-labile linker into removable monocarboxylate moieties under acidic conditions, which improved their gas adsorption and catalytic properties . Similarly, Kim et al. created mesoporous defects in microporous HKUST-1 using an acetic acid fragmented linker co-assembly method. A small amount of acetic acid removed the BTC linker to create mesoporous defects. On increasing the amount of acetic acid, metal clusters in the MOF were removed, which resulted in interconnected defect sites and mesoporous channels inside the MOF, which enhanced the methane storage capacity in the defective MOF . These methods have been employed extensively in numerous studies to functionalise meso- and macropores with straightforward synthetic procedures and a number of effective mechanisms have been demonstrated [90, 91], but there are now additional reports of using similar methods for the production of macroporous structures.
An alternative approach to direct synthesis of defective MOFs with macropores is via the post-synthetic etching of prepared microporous MOFs. This can be done either by an acid etching mechanism whereby selected acid molecules diffuse through the MOF structure, forming large voids over time, or by synergistic etching where bulky acid molecules modify and protect the MOF surface to obtain the holey materials.
Koo et al.  investigated a size-selective acid diffusion method for the post-synthetic treatment of MIL-100 (Fe) using phosphoric acid (d = 0.61 nm) as the etching agent. During the etching process, phosphoric acid could specifically diffuse through the hexagonal windows (d = 0.89 nm) but not through the pentagonal windows (d = 0.49 nm) of MIL-101(Fe) to form macropores in the framework, preserving the microporous nature by selective removal of metal nodes and BTC linkers. The framework crystallinity and morphology of the MOF crystals were well maintained even after post-synthetic treatment. Note that this process cannot be achieved by employing smaller acids such as HCl (d = 0.34 nm) in the same manner. This is simply because the small cages could also be attacked during etching, leading to complete collapse of the MOF structure. This etching method seems to be effective for the creation of hierarchical porous MOFs with tuneable pore sizes. However, as mentioned in the literature , these approaches are only applicable for highly water-stable MOFs.
As an extension to the idea of size-selective acid etching, another post-synthetic process, termed “synergistic etching and surface functionalisation”, was reported by Hu et al. . In this study, weak bulky phenolic acids such as gallic acid and tannic acid were used to provide free protons that penetrated into the MOF crystals to form a hollow structure. Interestingly, these acids with large molecular sizes simultaneously blocked the exposed surface of ZIF-8, leading to the preservation of the parent crystalline framework of MOF in the outer shell. Yoo and Jeong  reported that by etching IRMOF-3 with cyanuric chloride and tetraethylamine (TEA), protons released from the modification can create a mesoporous structure or even macroscale defects if the etching process proceeds further. In this case, TEA was used as a proton scavenger to prevent the complete dissolution of IRMOF-3 in cyanuric chloride, thus preserving the MOF crystallinity.
From the above examples, it can be seen that acid etching approaches for the preparation of hierarchical macroporosity in MOFs are a comparatively facile procedure compared to templating methods, as they can be carried out as a post-synthetic step on larger samples of MOFs. Furthermore, it is possible for the defective MOFs resulting from these approaches to possess well-defined crystal structures and preserve their external morphology after acid etching, though judicious choice of the etching agent and etching conditions are essential to obtain controllable macroporosity in MOFs without the loss of crystallinity or complete dissolution of the crystalline structure. While water unstable MOFs were considered as limited candidates in some etching mechanisms, this could be solved by varying the dilute solvents employed. The nature of the interactions of various solvents with MOF frameworks therefore needs to be well understood to generate a method applicable to different MOF systems.
2.3 Compressed or Supercritical Carbon Dioxide Method
In addition to its use in porous material activation and aerogel formation, compressed or scCO2 has been recently investigated as an unconventional solvent for direct MOF synthesis. By varying CO2 pressure in the reactor, particle sizes and porosity can be controlled, which is not possible using conventional solvothermal synthesis. These substantial benefits are reasons why using this alternative solvent in MOF synthesis can be advantageous even though a high pressure is required.
The latest research demonstrates that MOFs with macropores can be synthesised using scCO2. These findings suggest an intelligent method to obtain an open macroporous structure without using templates. It is anticipated that this process may be applicable to other MOF systems. Note that in the case of HKUST-1 metal–ligand coordination forces between Cu2+ and BTC3− in DMSO dominated the MOF aggregation process, resulting in a precursor solution . It was found that the nucleation was only trigged when the introduction of CO2 and a small amount of methanol served to disrupt the H-bonding in the stock solution. Due to critical conditions of CO2 applied in this method (Tc and Pc of CO2 are 31.1 °C and 73.9 bar, respectively ), a high pressure reactor needed to be employed. Also, low temperature synthesis at 40 °C was found to be more favourable, which is not easy to achieve for most MOFs. Another challenge in this synthetic strategy is carrying out the experiment in in situ characterisation to understand the mechanism of the macropore formation during CO2 pressurisation.
2.4 3D Printing Method
The formation of macropores in MOFs by the acid etching and scCO2 methods has been shown to be beneficial for catalytic applications, but, with the exception of the MOF aerogels, most still result in MOF materials consisting of small crystallites, rather than controllable monolithic structures. As discussed in the preceding sections, a key barrier to the implementation of MOFs in industry is their lack of processability owing to their general insolubility and inherently powdered form. While a variety of methods have been attempted to transform MOF powders into appropriate monolithic forms, such as pressing powders into pellets , melting them to form glasses [125, 126, 127], or forming MOF coatings on monolithic supports (such as those mentioned in Sect. 2.1) , typically such techniques result in reduced MOF surface area.
Over the past few decades, 3D printing, also known as additive manufacture or robocasting technology, has been developed and applied successfully to fabricate monoliths with controlled 3D shapes by careful placement of material in space. Currently, printing methods are not limited to utilising inks comprised of plastics for art exhibitions and may also employ inks comprised of ceramics, metals, graphene additives, and even stimuli-responsive hydrogel composites  for prototyping and industrial applications [130, 131, 132]. Recently, there have been a growing number of research papers on developing tailored porous biomedical scaffolds , membranes [134, 135, 136], carbons , zeolites [138, 139, 140], aminosilicate adsorbents , and heterogeneous catalytic systems  by taking advantage of this technique.
In comparison with other processing techniques for making microporous MOFs such as solvothermal and microemulsions, 3D printing promises multiple advantages such as high reproducibility of complex geometries, low cost, scalability, and efficiency. Furthermore, with regards to MOFs specifically, the printing of a MOF composite would represent major advantages over alternative methods of monolith formation, which generally involve the deposition of MOFs onto non-flexible substrates like metals and require complicated techniques for precise MOF positioning [143, 144]. Additionally, a further benefit is the formation of a printed monolith has potential to enable additional porosity to be implemented depending on the additives employed in the ink. Hence, 3D printed monoliths not only offer an implementable contact medium for MOFs, but also enable an increased pore size distribution to be attained compared to parent MOF powders. In addition, polymers with a variety of unique attributes, such as softness, thermal and chemical stability, and optoelectrical properties, can be integrated with MOFs to make hybrids with sophisticated architectures . With numerous potential applications, 3D-printed MOFs have been prepared successfully, though only recently, and therefore have been studied to only a limited extent.
According to Semino et al. , the general method developed by Thakka et al. for the production of hierarchical MOF monoliths is likely governed by the polymer it is embedded in: firstly by the partial blocking of MOF pores by flexible polymers (giving smaller pore sizes) and secondly by the formation of voids within the polymer matrix itself. The influence of polymer characteristics on the fabrication of hierarchical MOF monoliths by 3D printing was in fact investigated by Evans et al. in 2018 . In this study, high loadings of powdered ZIF-8 (up to 50 wt%) were incorporated into a matrix of either polylactic acid (PLA) or thermoplastic polyurethane, representing rigid and flexible polymer matrices, respectively, to produce a filament feedstock for 3D printing. It was revealed that the rigidity of the PLA chains limited their reorganisation around ZIF-8 crystals during extrusion, resulting in various void sizes (micro-, meso-, and macro-voids) and hence giving a hierarchical MOF monolith with preserved MOF crystallinity. Unfortunately, employment of the flexible polymer matrix resulted in almost complete pore occlusion, as concluded from the low specific surface area of the monolith (68 m2 g−1 at 50% MOF loading).
Recently, Young et al.  highlighted a practical solution to the problem of pore occlusion in MOF composite monoliths by 3D-printing UiO-66 in combination with a mixture of flexible acrylates, which exhibited low thermal stability (< 100 °C). As expected, the printed UiO-66 monoliths were non-porous (owing to blockage by flexible acrylate chains); however, on heating to 100 °C, the polymer matrix degraded to recover porosity and reveal UiO-66 sites. Interestingly, although the treated composite displayed physisorption behaviour characteristic of microporous UiO-66, the presence of small mesopores was also identified. This work, carried out by Young et al., promises a method to embed MOFs within 3D printed monoliths and later exposes crystals as required for production of structures with hierarchical porosity. Furthermore, UiO-66 was found to behave as a rheological modifier itself, avoiding the need for additives such as the bentonite clay employed in earlier studies.
Recent developments in the fabrication of 3D-printed MOFs have been successful in both providing a monolith suited for practical applications and adding hierarchical porosity into MOF-based materials. Although 3D printing of MOF monoliths has predominantly required high processing temperatures (up to 230 °C) or ultraviolet curing [150, 151], the work described by Sultan et al. demonstrated the feasibility of room temperature synthesis. However, it should be noted that this technique may not be suited to a variety of MOFs which are unstable in air, since the layer-by-layer printing process may necessitate air exposure for extended periods (15–120 h). For example, Kreider et al.  observed the degradation of water-sensitive MOF-5 during blending of the printable ink with the MOF powder and attributed this to ambient humidity. Interestingly though, irrespective of the observed MOF degradation, the resulting monolith was capable of adsorbing more H2 than the pure polymer counterpart, demonstrating the viability for H2 storage using 3D printed MOFs even in humid conditions.
Overall, the ability to tailor both the MOF and binding matrix could render extensive applications viable, from providing photonic platforms to biomedical testing and catalysis. While the precise placement of materials in 3D printing renders more complex macroscale structures achievable, it should be noted that fine control on the micron–nanoscale could be challenging in terms of accurately controlling the size of macropores which form inherently during ink deposition. However, as 3D printing methods become more established, it is anticipated that finer control over macropore sizes will be achievable. Finally, since significant research attention has taken steps towards employing MOFs in biomolecule encapsulation for controlled drug delivery and in biological applications, where macroporosity would be beneficial for facilitating mass transfer of biological fluids, it may be useful to consider the biodegradability of MOFs in low pH environments [153, 154]. To this end, future work could explore the development of ink matrices which could protect embedded MOFs from such degradation.
In this review, a range of fabrication methods for creating hierarchical MOF pore structures with additional macropores were discussed. Introduction of macroporosity has been demonstrated to lead to improvements in mass transfer and catalytic activity, and reductions in pressure drop over traditional powdered or nanocrystalline materials. These macropores can also improve molecular accessibility of reagents to the microporous cavities, which accommodate important functional groups or active sites, enhancing the catalytic performance of the materials. The advantages of macropores in catalytic applications have been well demonstrated for zeolites and metal oxides. For MOFs as microporous crystallites (which are purposely designed to possess high surface areas), introducing additional meso- and macropores to form hierarchical structures without compromising the micropores and the crystallinity, which give them high surface areas, is becoming the subject of much research. While mesopores can be created via numerous methods such as ligand exchange and use of surfactants, further expanding these to the macropore regime remains a challenge due to the ligand size limitation and the post-synthetic activation methods needed to obtain accessible pores.
As shown here, the main techniques reported for fabrication of macroporous materials can be described using four general approaches: use of structural templating, defect formation, use of CO2, and 3D printing. Templating stands out as the most popular route towards obtaining macroporous MOFs with numerous studies reported so far, due to the ease of structural control and the ease with which this method can be adapted for use with a range of MOFs. Deposition of MOFs onto hard macroporous templates and the formation of MOF composites via direct synthesis onto a porous template often do not require significant deviation from the synthetic conditions identified for the MOF itself and may offer an additional advantage in improving mechanical and thermochemical stability and providing immobilisation of nanocrystalline MOF materials. This may improve the outlook for MOFs to be used in industrial catalysis, for example, in alkylation and fluid catalytic cracking in oil refining processes involving bulky hydrocarbons, where harsh conditions require higher catalyst stability. In addition, the use of a hard template means that the active MOF material may only constitute a small proportion of the overall composite by weight, which may be advantageous in the case of more expensive MOF materials. In post-treatment synthetic methods, pre-formed MOFs can be immobilised on macroporous templates such as foams and sponges using dip-coating methods to form the composites. The stability of the MOFs used in this method remains a major challenge due to a long exposure time to the high humidity conditions during the immobilisation process, though this may be addressed through careful choice of solvents. In addition, in these macroporous composite structures, the MOF materials must be suitably bonded to the surface (e.g. via functionalisation of the template surface) to avoid being washed off over time.
If, rather than rigid composites, flexible composites or pure macroporous MOF structures are required, soft macroporous templates such as polymers, emulsions, or gels can be used. Polymer templates can allow for more pliable macroporous structures, while the use of emulsions or gels may allow for easier removal after templating. Use of sacrificial soft templates can result in macroporous stand-alone aerogel monoliths which can lead to easier and safer materials handling for particular applications; however, careful template removal is required to avoid incomplete removal (which could lead to loss of porosity) or collapse of pore networks.
Defect formation can be implemented via direct synthesis using linker modulation techniques or post-synthetic acid etching of prepared MOFs in low pH environments. The former has been widely reported for synthesis of mesoporous MOFs with defects as active adsorption and reaction sites for enhanced catalytic activity and gas storage, with linker modulation now being explored for introduction of larger macroporous voids. Defect formation via acid etching can overcome the limitation of the need for extended ligands and can potentially create macropores via size-selective diffusion processes. In fact, various etching agents such as hydroquinone, boric acid, and phosphoric acid have been studied so far to synthesise some impressive macroporous MOF crystal structures. While this method remains challenging due to the instability of most MOFs under acidic conditions, there have been promising recent developments, such as the discovery that careful selection of solvents can allow large geometrical pores within HKUST to be achieved by acid etching and that the use of synergistic etching to protect surfaces can be used to control etching. These examples indicate that with further investigation of the synthetic conditions and further understanding of the etching mechanisms, this simple approach to the formation of such macropores in MOF crystallites may be more widely applied.
Utilising supercritical or compressed CO2 in MOF preparation can address some key challenges noted in previous methods including pore collapse during solvent removal steps when using soft structural templating. In fact, this method is well known in MOF synthesis and post-synthetic treatment and has showed some promising results in creating open macroporous aerogel structures. Further adapting the synthetic conditions (e.g. longer reaction times and higher pressures), the use of scCO2 routes was shown to be able to form additional macropores in the direct synthesis of MOFs. In HKUST-1 synthesis, this method was also demonstrated to quickly trigger the reaction with a lowered amount of solvent, and as an added advantage, due to the insolubility of scCO2 in stock solutions, the multiple purification steps required in conventional synthetic methods could be curtailed. Particle sizes and porosity of this MOF can be tuneable via tight control over etching time. This route offers substantial benefits in HKUST-1 preparation compared to conventional solvothermal method. However, the mechanisms of the macroporous HKUST-1 formation need to be fully understood, before generalising this strategy to other MOF systems.
Finally, 3D printing is a very new fabrication technique which shows promise for generalised use in MOF preparation, allowing a very high MOF loading to be incorporated into porous printed structures (~ 85 wt%). The high reproducibility, fabrication of complex geometries, and controlled pore structures of these MOFs have potential to open up new applications for MOF materials in biomedical fields. This method, however, relies on the ability to incorporate MOFs and MOF precursors into inks capable of being printed into stable structures without compromising on the nanoscale structure of the MOF. Enabling higher resolution (sub-mm) printing is also an area for future development, to enable greater control over the macrostructure, for example in biomedical applications such as tailored drug delivery or the size-selective diffusion of proteins to oxidation sites within MOFs facilitates biocatalysis.
As shown by many of the studies mentioned in this review, macroporous MOF structures can show enhanced performance in adsorptive and catalytic applications due to the improved mass transfer and molecular accessibility. It can be predicted that more research will be carried out to produce more useful composites of functional MOFs coordinated with macroporous substrates using reproducible and sustainable synthetic approaches. The range of fabrication approaches presented here indicates the vast variety of different macroporous structures that can be achieved using functional MOF materials. The ability to include structural macroporosity should help to accelerate the practical application of MOFs in such varied fields as large molecule adsorption and separation, water purification, bulky drug delivery, and heterogeneous catalysis. Further development of this toolbox will surely give material scientists greater flexibility in tailoring the porosity and structure of their MOF materials towards particular functions.
This work was financially supported by the Vietnamese Ministry of Education and Training and the UK Engineering and Physical Sciences Research Council (EP/R01650X/1 and EP/L016028/1).
- 17.J. Rouquerol, D. Avnir, C.W. Fairbridge, D.H. Everett, J.M. Haynes et al., Recommendations for the Characterization of Porous Solids (Technical Report) (1994)Google Scholar
- 19.W. Schwieger, A.G. Machoke, T. Weissenberger, A. Inayat, T. Selvam, M. Klumpp, A. Inayat, Hierarchy concepts: classification and preparation strategies for zeolite containing materials with hierarchical porosity. Chem. Soc. Rev. 45, 3353–3376 (2016). https://doi.org/10.1039/c5cs00599j CrossRefGoogle Scholar
- 36.D. Sheberla, L. Sun, M.A. Blood-Forsythe, S. Er, C.R. Wade, C.K. Brozek, A. Aspuru-Guzik, M. Dincǎ, High electrical conductivity in Ni3(2,3,6,7,10,11- hexaiminotriphenylene)2, a semiconducting metal–organic graphene analogue. J. Am. Chem. Soc. 136, 8859–8862 (2014). https://doi.org/10.1021/ja502765n CrossRefGoogle Scholar
- 39.K. Wang, Y. Zhang, J. Zhao, C. Yan, Y. Wei et al., Facile synthesis of hierarchical porous solid catalysts with acid-base bifunctional active sites for the conversion of cellulose to 5-hydroxymethylfurfural. New J. Chem. 42, 18084–18095 (2018). https://doi.org/10.1039/c8nj03812k CrossRefGoogle Scholar
- 44.P. Klobes, K. Meyer, R.G. Munro, Porosity and Specific Surface Area Measurements for Solid Materials (NIST Special Publications, Washington, 2006)Google Scholar
- 59.F. Sandra, M. Depardieu, Z. Mouline, G.L. Vignoles, Y. Iwamoto, P. Miele, R. Backov, S. Bernard, Polymer-derived silicoboron carbonitride foams for CO2 capture: from design to application as scaffolds for the in situ growth of metal–organic frameworks. Chem. A Eur. J. 22, 8346–8357 (2016). https://doi.org/10.1002/chem.201600060 CrossRefGoogle Scholar
- 63.U. Betke, S. Proemmel, S. Rannabauer, A. Lieb, M. Scheffler, F. Scheffler, Silane functionalized open-celled ceramic foams as support structure in metal organic framework composite materials. Microporous Mesoporous Mater. 239, 209–220 (2017). https://doi.org/10.1016/j.micromeso.2016.10.011 CrossRefGoogle Scholar
- 64.U. Betke, S. Proemmel, J.G. Eggebrecht, S. Rannabauer, A. Lieb, M. Scheffler, F. Scheffler, Micro-macroporous composite materials: SiC ceramic foams functionalized with the metal organic framework HKUST-1. Chem.-Ing.-Tech. 88(3), 264–273 (2016). https://doi.org/10.1002/cite.201500141 CrossRefGoogle Scholar
- 65.U. Betke, M. Klaus, J.G. Eggebrecht, M. Scheffler, A. Lieb, MOFs meet macropores: dynamic direct crystallization of the microporous aluminum isophthalate CAU-10 on reticulated open-cellular alumina foams. Microporous Mesoporous Mater. 265, 43–56 (2018). https://doi.org/10.1016/j.micromeso.2018.01.020 CrossRefGoogle Scholar
- 76.Y. Yuan, D. Yang, G. Mei, X. Hong, J. Wu, J. Zheng, J. Pang, Z. Yan, Preparation of konjac glucomannan-based zeolitic imidazolate framework-8 composite aerogels with high adsorptive capacity of ciprofloxacin from water. Colloids Surf. A 544, 187–195 (2018). https://doi.org/10.1016/j.colsurfa.2018.01.042 CrossRefGoogle Scholar
- 89.S.Y. Kim, A.R. Kim, J.W. Yoon, H.J. Kim, Y.S. Bae, Creation of mesoporous defects in a microporous metal–organic framework by an acetic acid-fragmented linker co-assembly and its remarkable effects on methane uptake. Chem. Eng. J. 335, 94–100 (2018). https://doi.org/10.1016/j.cej.2017.10.078 CrossRefGoogle Scholar
- 96.H.V. Doan, A. Sartbaeva, J.-C. Eloi, S. Davis, V.P. Ting, Defective hierarchical porous copper-based metal–organic frameworks synthesised via facile acid etching strategy. ArXiv preprint (2019). https://arxiv.org/abs/1904.10524v1
- 97.F. Vermoortele, B. Bueken, G. Le Bars, B. Van De Voorde, M. Vandichel et al., Synthesis modulation as a tool to increase the catalytic activity of metal–organic frameworks: the unique case of UiO-66(Zr). J. Am. Chem. Soc. 135, 11465–11468 (2013). https://doi.org/10.1021/ja405078u CrossRefGoogle Scholar
- 102.N. Portolés-Gil, A. Lanza, N. Aliaga-Alcalde, J.A. Ayllón, M. Gemmi et al., Crystalline curcumin bioMOF obtained by precipitation in supercritical CO2 and structural determination by electron diffraction tomography. ACS Sustain. Chem. Eng. 6, 12309–12319 (2018). https://doi.org/10.1021/acssuschemeng.8b02738 CrossRefGoogle Scholar
- 111.N. Hüsing, U. Schubert, Aerogels—airy materials: chemistry, structure, and properties. Angew. Chem. Int. Ed. 37, 22–45 (2005). https://doi.org/10.1002/(sici)1521-3773(19980202)37:1/2%3c22:aid-anie22%3e3.0.co;2-i CrossRefGoogle Scholar
- 117.P. López-Domínguez, A.M. López-Periago, F.J. Fernández-Porras, J. Fraile, G. Tobias, C. Domingo, Supercritical CO2 for the synthesis of nanometric ZIF-8 and loading with hyperbranched aminopolymers. Applications in CO2 capture. J. CO2 Util. 18, 147–155 (2017). https://doi.org/10.1016/j.jcou.2017.01.019 CrossRefGoogle Scholar
- 118.A.M. López-Periago, N. Portoles-Gil, P. López-Domínguez, J. Fraile, J. Saurina, N. Aliaga-Alcalde, G. Tobias, J.A. Ayllón, C. Domingo, Metal–organic frameworks precipitated by reactive crystallization in supercritical CO2. Cryst. Growth Des. 17, 2864–2872 (2017). https://doi.org/10.1021/acs.cgd.7b00378 CrossRefGoogle Scholar
- 119.N. Portolés-Gil, S. Gowing, O. Vallcorba, C. Domingo, A.M. López-Periago, J.A. Ayllón, Supercritical CO2 utilization for the crystallization of 2D metal–organic frameworks using tert-butylpyridine additive. J. CO2 Util. 24, 444–453 (2018). https://doi.org/10.1016/j.jcou.2018.02.004 CrossRefGoogle Scholar
- 120.H.V. Doan, Y. Fang, B. Yao, Z. Dong, T.J. White, A. Sartbaeva, U. Hintermair, V.P. Ting, Controlled formation of hierarchical metal–organic frameworks using CO2-expanded solvent systems. ACS Sustain. Chem. Eng. 5, 7887–7893 (2017). https://doi.org/10.1021/acssuschemeng.7b01429 CrossRefGoogle Scholar
- 122.R. Ameloot, E. Cobechiya, H. Uji-i, J.A. Martens, J. Hofkens, L. Alaerts, B.F. Sels, D.E. De Vos, Direct patterning of oriented metal–organic framework crystals via control over crystallization kinetics in clear precursor solutions. Adv. Mater. 22, 2685–2688 (2010). https://doi.org/10.1002/adma.200903867 CrossRefGoogle Scholar
- 131.J.C. Ruiz-Morales, A. Tarancón, J. Canales-Vázquez, J. Méndez-Ramos, L. Hernández-Afonso, P. Acosta-Mora, J.R. Marín Rueda, R. Fernández-González, Three dimensional printing of components and functional devices for energy and environmental applications. Energy Environ. Sci. 10, 846–859 (2017). https://doi.org/10.1039/c6ee03526d CrossRefGoogle Scholar
- 141.H. Thakkar, S. Eastman, A. Al-Mamoori, A. Hajari, A.A. Rownaghi, F. Rezaei, Formulation of aminosilica adsorbents into 3D-printed monoliths and evaluation of their CO2 capture performance. ACS Appl. Mater. Interfaces. 9, 7489–7498 (2017). https://doi.org/10.1021/acsami.6b16732 CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.