1 Introduction
While fossil energy brings great convenience to people’s lives, environmental problems such as atmospheric pollution, water pollution, and ecological destruction caused by its development and use are becoming increasingly severe [1]. Hydrogen energy is a new type of clean energy which has such advantages as abundant raw materials, diverse storage methods, high calorific value, and green products [2]. Photocatalytic decomposition of water to produce hydrogen due to its low cost and mild reaction conditions has become a green and environmentally friendly technology for efficient exploiting of solar energy [3].
Metal-organic frameworks (MOFs), also known as porous coordination polymers (PCPs), are new types of porous crystals connected by metal ions/clusters and multidentate organic ligands through coordination bonds. Due to the large specific surface area, diverse structures, adjustable performance, and diverse synthesis methods, MOFs are widely used in adsorption [4,5], catalysis [6,7], sensing [8,9], drug delivery [10,11], and many other scientific research fields. The reaction process of MOFs used in photocatalytic systems is similar to that of semiconductors, as shown in Fig. 1. First, the organic ligands of MOFs absorb photons to generate photo-generated carriers (electrons and holes) under light conditions; Then, the photogenerated carriers migrate and separate from each other. The electrons are transferred to the metal ions/clusters in the MOFs structure through the bridged organic ligands. Finally, the photogenerated electrons or holes are oxidized or reduced with the adsorbed reactants in the active center of the MOFs.
The metal ions and multidentate organic ligands are combined through coordination bonds to form a three-dimensional network structure in most MOFs, which leads to the poor water stability and low thermal stability. In 2008, Cavka of the University of Oslo in Norway [12] and others reported the UiO-66(Zr) material for the first time, which has an octahedral structure. Compared with other MOFs, the biggest advantage is that UiO-66 has an excellent chemical stability, anti-mechanical stability, thermal stability, and water stability [13], which greatly improves the UiO-66 series MOFs in practical applications. This paper systematically reviews the development of UiO-66 (Zr) series MOFs in recent years, in order to provide references for their future development.
2 Defect-engineered UiO-66
In recent years, an increasing number of scientists have studied the relationship between the structure and function of MOFs. The so-called defects are the lack or replacement of atoms, and ions and groups to break the periodic arrangement of the crystals [14]. In the past cognition, it is believed that defects are detrimental to the structure and performance of the material. What is unexpected is that these defect sites in MOFs bring advantages that “perfect” crystals do not have. Defects of materials is a double-edged sword for photocatalytic reactions. On the one hand, they can supply a bigger surface area and aperture, whose benefit offers a more active center for photocatalytic reactions [15]. Moreover, they cut down the carrier diffusion distance and thus enhanced electron hole pair separation could be realized [16]. In addition, they reduce the band gap of the material in order to absorb more photons to improve light absorption [17,18]. On the other hand, too high a concentration of defects will promote the defect site itself to become a carrier recombination center, in addition to reducing the structural stability of the material [19]. Therefore, it attracts a large number of scholars to study the nature of the defect site, including preparation methods, modification methods, characterization, and application, etc., especially the defect-engineered UiO-66 MOFs.
2.1 Type of defect
The defects of defect-engineered MOFs materials can be mainly classified into two major types: metal defect and ligand defect. The metal defect is the lack of metal clusters or the replacement of partial metal ions by other metal ions; while the ligand defect is the lack of organic ligands or the replacement of the ligand position by other small-molecules. It is generally believed that the structure of the crystal determines the function of the crystal, and the change of the microstructure of the defect-engineered MOFs will inevitably affect the original performance or bring new performance.
2.1.1 Metal defects
There are two main ways to generate metal defects. One way is that, in the synthesis process, metal defects are generated by adding foreign metal precursor to compete with the original precursor for coordination. Another way is that, after the original MOFs are synthesized, some metal ions are modified and replaced to become bimetal/multi-metal MOFs.
As shown in Fig. 2, Lee et al. [20] replaced part of the Zr atoms in UiO-66-NH2 with Ti atoms by using the post-synthesis method to prepare bimetallic clusters and mixed ligand UiO-66(Zr/Ti)-NH2. Then, they applied it to the photocatalytic reduction of CO2 to produce formic acid reaction, and study the effect of Ti atoms doping on the photocatalytic activity. The results of this study confirmed that the (Zr/Ti) metal clusters enhance the ability of MOFs to generate photogenerated carriers by light excitation. In addition, the introduction of diamine ligands creates new energy levels, which promotes visible light harvesting and the electron and hole migration, thus significantly improving the photocatalytic activity.
2.1.2 Organic ligand defects
Most MOFs are usually prepared by using the hydrothermal/solvothermal methods. Therefore, coordination errors will inevitably occur during the coordination of metal ions with organic ligands. In addition, water molecules, solvent molecules, and monocarboxylic acid molecules (acetic acid, formic acid, propionic acid, hydrochloric acid, trifluoroacetic acid, etc.) can also participate in coordination with metal ions, causing the ligand defects. Cai and Jiang [21] used carboxylic acid to occupy the position of the organic ligand, and then removed the carboxylic acid molecule by thermal activation, and successfully prepared the pore size of 5.5 nm and the surface area of 980 m2/g high stability MOFs.
2.2 Synthesis method of defect-engineered UiO-66
As defect-engineered MOFs have gradually been widely used in many fields such as catalysis, environmental protection, and sensing, more researchers begin to explore the synthesis methods of defect-engineered MOFs, such as adjusting temperature, adding regulators, modifying metal clusters, and changing the ratio of raw materials, etc.
2.2.1 Adjusting temperature
The reaction temperature has a huge influence on all synthesis reactions. For the synthesis process of MOFs, the temperature directly affects the rate of binding metal ions to multidentate organic ligand molecules and the rate of crystallization. Therefore, controlling the appropriate reaction temperature can produce defects during the synthesis of MOFs. Destefano et al. [22] obtained defect-engineered UiO-66-X (NH2, NO2, OH) by controlling the reaction temperature. The results prove that the reaction temperature is directly related to the concentration of defects: it is beneficial to form perfect crystals under high temperature conditions; while it is not conducive to the combination of metal ions and organic ligands under low temperature conditions, which is easy to produce defects. Decreasing the reaction temperature from 130°C to 25°C is beneficial to increasing the concentration of defects. The most suitable reaction temperature is 45°C, and an average of 1.3 organic ligands are missing for per Zr6O4(OH)4 metal node.
2.2.2 Adding regulators
When synthesizing MOFs, defects may be caused by the coordination errors of metal ions and multidentate organic ligand molecules. Therefore, adding carboxylic acids as regulators to enhance the probability of such coordination errors is a common method to introduce defects in MOFs.
In the synthesizing process of MOFs, Ma et al. [23] controlled the defect concentration of Pt@UiO-66-NH2 photocatalyst by controlling the amount of added acetic acid. To better prove the relationship between defects and carrier separation, they used the femtosecond time-resolved transient absorption (fs-TA) spectroscopy to examine the involved charge separation processes. The average relaxation lifetime can be employed as an indicator to evaluate the charge separation efficiency: the faster the relaxation, the higher the efficiency. The results show that the defect concentration is directly related to the amount of added acetic acid. The more the amount of acetic acid added, the greater the concentration of ligand defects, and as the amount of added acetic acid increases, the photocatalytic activity of Pt@UiO-66-NH2 shows a volcanic distribution. This proves that an appropriate defect concentration is beneficial to carrier transfer and separation, while an excessively high concentration of defects will inhibit the separation of charges. Similarly, Vermoortele et al. [24] used a solvothermal method to add trifluoroacetic acid and hydrochloric acid as regulators to prepare UiO-66 containing ligand defects. The results show that the mixed use of trifluoroacetic acid and hydrochloric acid can produce highly crystalline UiO-66. The subsequent heating activation cannot only dehydroxylate but also remove the trifluoroacetic acid occupying the ligand position, so as to obtain more open sites which can be used as the reactive center to directly contact the reactant for chemical reaction.
2.2.3 Modifying metal clusters
Yang et al. [25] prepared Ge(III)-doped UiO-66 by utilizing the solvothermal method to study the effect of Ge(III) doping on the adsorption performance of UiO-66. The room temperature adsorption capacity of Ge(III)-doped UiO-66 for methylene blue (MB), methyl orange (MO), and Congo red (CR) increased by 490%, 270%, and 70%, respectively, compared with the original UiO-66. The reason for this is that Ge(III) doping increases the electrostatic attraction between the cationic and anionic dyes and increases the adsorption capacity of the adsorbent. In addition, Niu et al. [26] used lithium nitrate ethanol solution instead of organosilicon solution to prepare Li-doped UiO-66 for CO2 absorption. In situ diffuse reflectance Fourier transform infrared spectroscopy was used to study the interaction between adsorbed CO2 and UiO-66. The results prove that doping with Li can increase the number of active metal sites of UiO-66, provide new adsorption sites for CO2, and enhance the adsorption capacity of UiO-66 for CO2. The adsorption capacity of CO2 increases from 20.6 cm3/g to 34.5 cm3/g.
2.2.4 Changing the ratio of raw materials
To synthesize MOFs, metal precursors and organic ligands are usually mixed in organic solvent with a high temperature and high pressure. Shearer et al. [27] adjusted the ratio of metal ions or organic ligands in the reactants to affect the crystallization process, and produced defect-engineered UiO-66. When the ratio of organic ligand to Zr is 2, almost perfect UiO-66 crystal can be obtained; when the ratio gradually increases, the obtained UiO-66 will have defect sites. Therefore, proper control of the ratio of raw materials is also a method to prepare defect-engineered MOFs.
2.3 Modification of UiO-66
In recent years, many scholars have explored UiO-66 modification methods, such as loading noble metals, compound semiconductors, and photosensitization.
2.3.1 Loading noble metals
Noble metals such as Au, Ag, and Pt are the most common metals used in photocatalytic systems [28]. Noble metals mainly improve the photocatalytic efficiency of the system from the following aspects: First, the surface plasmon resonance (SPR) effect of noble metals (Au, Ag, etc.) can absorb photons of a certain frequency, generate additional photo-generated electrons, and promote the photocatalysis. Next, after noble metals are loaded on MOFs, they can capture photo-generated electrons from MOFs, inhibit photo-generated electrons from recombining with holes during the transfer process, thereby increasing the probability of photo-generated carriers participating in the redox reaction. Finally, noble metals can usually be used as active sites on MOFs, such as photocatalytic hydrogen production systems, supporting Pt metals can greatly improve the catalytic efficiency.
As demonstrated in Fig. 3(a), Xiao et al. [29] used in situ synthesis to introduce Pt nanoparticles into the surface and interior of Pt/UiO-66-NH2, and studied the effect of the position of Pt nanoparticles on the photocatalytic hydrogen production. Experiments show that Pt nanoparticles loaded into UiO-66-NH2 have a better photocatalytic activity. The loading of Pt into UiO-66-NH2 can open up new channels for electron transfer, which can reduce the migration distance of photogenerated electrons, so to promote the separation of electron-hole pairs, and inhibit the recombination of carriers. At the same time, Pt nanoparticles loaded inside can be fixed by the pore structure of UiO-66-NH2, preventing Pt nanoparticles from falling off during the reaction process, and improving the life and stability of the catalyst.
Similarly, Gu [30] and others successfully synthesized Au@UiO-66-NH2 heterojunction photocatalyst with a core-shell structure. Au enhances the light absorption of Au@UiO-66-NH2 due to the SPR effect. In addition, the core-shell structure and heterojunction facilitate the transfer of photogenerated electrons from Au nanoparticles to UiO-66-NH2. Zhao et al. [31] designed a simple deposition-photoreduction method to synthesize UiO-66-NH2/Ag/AgCl for photodegradation of RhB. The photodegradation rate of this catalyst for RhB is 10 times higher than that of UiO-66-NH2.
2.3.2 Compound semiconductor
Inspired by the modification strategy of semiconductor photocatalysts, researchers consider combining MOFs with another semiconductor to prepare a new material that simultaneously obtains the advantages of both materials. As exhibited in Fig. 4(a), Wang et al. [32] calcined different proportions of g-C3N4 and UiO-66 mixtures in Ar atmosphere at a high temperature to prepare UiO-66/g-C3N4 heterojunction composite materials. The SEM results indicated that g-C3N4 adheres tightly to the UiO-66 octahedron. The results of time-resolved transient fluorescence spectroscopy suggest that the heterojunction can promote the transfer of carriers on the interface of UiO-66/g-C3N4 heterojunction, improve the separation efficiency of photo-generated carriers, and then improve the photocatalytic activity.
Similarly, Crake et al. [33] used the in situ growth method to composite TiO2 nanosheets with UiO-66-NH2 for photocatalytic reduction of CO2. The in situ synthesis method makes TiO2 and UiO-66-NH2 form a tight heterojunction, which not only utilizes the high adsorption performance of TiO2 nanosheets for CO2, but also retains the advantages of UiO-66-NH2 with a large specific surface area and multiple pores. The nanocomposites have a better photocatalytic reduction of CO2 than single materials. Transient absorption spectroscopy (TAS) proves that the photogenerated carriers of composite materials have a longer average lifetime, and the heterojunction promotes the charge transfer between interfaces, thus improving the photocatalytic activity.
2.3.3 Photosensitization
Yuan et al. [34] used phycoerythrin B to photosensitize UiO-66, and studied the mechanism of phycoerythrin B as a photosensitizer to improve the photocatalytic hydrogen evolution activity of UiO-66. The attachment of phycoerythrin B on the surface of UiO-66 not only promotes the absorption of visible light, but also quickly transfers the photoelectrons excited on the surface of the dye to the conduction band of UiO-66. Pt nanoparticles act as a co-catalyst to improve the efficiency of electron reduction of hydrogen protons, thus greatly improving the efficiency of photocatalytic hydrogen production.
3 Characterization of “defects” of UiO-66
3.1 Thermogravimetry analysis
Thermogravimetry analysis (TGA) is a characterization method to study the change of material structure with temperature. MOFs crystalline materials are composed of metal clusters and organic ligands. After heating them at a certain heating rate, the structural framework of MOFs will collapse. The entire heating process first loses water molecules, regulator molecules, and solvent molecules. If the temperature continues to rise to a certain degree, the entire framework will collapse, and the organic ligands will oxidize and decompose. Consequently, metal oxides will remain. Therefore, the number of organic ligands can be measured by detecting the weight change of the thermal decomposition of organic ligands, to determine whether there are defect sites in the framework. Taking UiO-66 as an example, the initial weight of the “perfect” UiO-66 molecule after thermal decomposition is 2.2 times the final weight. When the initial weight of UiO-66 is lower than 2.2 times of its final weight, there may be “defects” [35]. However, when TGA is applied to defect analysis, it cannot be judged whether the defect is a metal cluster defect or an organic ligand defect, which has certain limitations.
3.2 X-ray diffraction
X-ray powder diffraction (X-ray diffraction, XRD) is a characterization technique that uses X-rays to penetrate a crystal and then reveal the crystal structure. The crystallinity of MOFs will change after defects are generated. Therefore, the changes in crystallinity can be detected by XRD and mapped to the changes in their structure, but XRD cannot quantitatively test the defects. A single crystal XRD can give more specific information about defects, but the crystal size is required to be 5–100 μm. Trickett et al. [36] successfully prepared defect-containing UiO-66 using formic acid as a regulator, and used single-crystal XRD to determine that the water molecules in UiO-66 were directly coordinated with unsaturated Zr atoms. Similarly, Øien et al. [37] used the single crystal XRD technology to characterize the single crystal UiO-66 and found that the organic ligand only had a 73% occupancy rate, which proved the existence of ligand defects.
3.3 Nuclear magnetic resonance spectroscopy
Nuclear magnetic resonance spectroscopy (NMR) is a method of characterizing the structure of organic and inorganic substances without causing damage to samples. NMR can characterize how regulators, water molecules, and organic ligand molecules are linked or disconnected from metal clusters. Taddei et al. [38] and others used in situ NMR to study the molecular exchange process between terephthalic acid analogs and MOFs. The results prove that when the defect-engineered UiO-66 is immersed in the solution, the terephthalic acid analogs in the solution preferentially exchange with the monocarboxylic acid at the defect site. Nandy et al. [39] used NMR to prove that cyclohexane is preferentially adsorbed on the tetrahedral defect sites of UiO-66(Zr), while acetone and methanol are adsorbed on the Zr-OH sites. In addition, NMR can also reflect the detailed information of defects through chemical shifts of substances.
3.4 Fourier transform infrared spectroscopy
Fourier transform infrared spectroscopy (FTIR) is a characterization method for detecting the energy level transition generated by the vibration or rotation of the functional group of the substance itself. This characterization method has been widely used in many scientific fields. Driscoll et al. [40] used CO as a highly sensitive probe molecule. By studying the blue shift of the characteristic frequency of the infrared spectrum of CO adsorbed on MOFs, it was determined that MOFs had coordinated unsaturated Zr atoms, and then proved the organic ligands defects.
3.5 Neutron diffraction
Neutron diffraction (ND) is one of the important methods for studying the structure of matter. For MOFs, except for metal nodes, other elements are lighter elements. XRD is not sensitive to light elements, while neutrons are sensitive to both organic ligands and metal ions. Therefore, ND can be used to characterize defect-engineered MOFs. To determine whether the ligand defect actually exists in UiO-66, Wu et al. [41] performed high-resolution ND measurement on the deuterated UiO-66, and proved that the addition of regulators can make the organic ligand missing rate reach 10%.
3.6 High-resolution transmission electron microscope
High-resolution transmission electron microscope (HRTEM) is a common characterization method used in materials science, nanotechnology, and semiconductor research. Liu et al. [42] used a combination of low-dose HRTEM and electronic crystallography to observe the structure of UiO-66 and found that the ligand defects and metal cluster defects coexist in the structure of UiO-66.
4 Application fields of defect-engineered UiO-66
Defect-engineered UiO-66 has been widely used in research fields such as catalysis, adsorption, separation, and conduction.
4.1 Photocatalysis
Defect-engineered UiO-66 not only have a large specific surface area and a unique pore structure, but also have more excellent photoelectric properties, and are widely used in the field of photocatalysis.
4.1.1 Photocatalytic hydrogen production
Peng et al. [43] encapsulated ZnIn2S4 inside the UiO-66 octahedral structure to prepare UiO-66@ZnIn2S4 composite material. Under visible light irradiation, the hydrogen evolution rate of UiO-66@ZnIn2S4 can reach 3061.61 μmol/h. The PL spectrum shows that the encapsulation structure is beneficial to the effective separation and transfer of photo-generated charges, and greatly increases the probability of electron holes participating in the redox reaction. Table 1 summarizes the application of UiO-66 in the field of photocatalytic hydrogen production in recent years.
Tab.1 Application of UiO-66 in the field of photocatalytic hydrogen production |
| MOFs | Photocatalytic hydrogen production activity /(μmol·h−1) | Ref. |
|---|---|---|
| MoS2/UiO-66-NH2/RGO | 25.03 | [44] |
| MoS2/UiO-66/CdS | 650 | [45] |
| UiO-66/CdS/RGO | 13.8 | [46] |
| WP/UiO-66/CdS | 79 | [47] |
| ErB/UiO-66/NiS2 | 18.4 | [48] |
| RhB/Pt@UiO-66 | 116 | [49] |
| Calix[4]arene/Pt@UiO-66-NH2 | 1.53 | [50] |
| CD@NH2-UiO-66/g-C3N4 | 2.93 | [51] |
| GOWPt@UiO-666-NH2 | 18.15 | [52] |
| Ni12P5@UiO-66-NH2 | 293.2 | [53] |
| Cd0.2Zn0.8S@UiO-66-NH2 | 5846.5 | [54] |
| Pt(PTA)@UiO-66-NH2 | 6.22 | [55] |
| TiO2/UiO-66-NH2/GO | 0.27 | [56] |
| Pd/UiO-66 | 9.43 | [57] |
| UiO-66-NH2 | 2.12 × 104 | [58] |
| PW12@UiO-NH2 | 7.27 × 104 | [59] |
4.1.2 Photocatalytic reduction of CO2
Shi et al. [60] used the electrostatic self-assembly method to composite g-C3N4 nanosheets on UiO-66 for photoreduction of CO2. This system not only has an excellent light absorption of g-C3N4, but also retains the huge specific surface area of UiO-66. The contact between UiO-66 and g-C3N4 is close, and the photo-generated electrons are transferred from g-C3N4 to UiO-66, which inhibits the recombination of electron-hole pairs and greatly improves the photocatalytic activity. Under visible light irradiation, the CO evolution rate of UiO-66/CNNS is measured, which achieved 9.9 μmol/h.
4.1.3 Photocatalytic degradation of pollutants
Xu et al. [61] mixed FeUiO-66 and polyaniline (PANI) heat treatment to obtain the PANI/FeUiO-66 nanocomposite material for aromatic alcohol oxidation. Under visible light irradiation, the catalytic activity of the modified PANI/FeUiO-66 was significantly increased. The enhancement of the photocatalytic activity is due to the doping of Fe element inhibiting the recombination of photogenerated carriers. The heterostructure of PANI and FeUiO-66 enhances the separation of photogenerated carriers. Similarly, Chen et al. [62] used the in situ reduction method to introduce Cu element into UiO-66-NH2(Zr), and found that the average lifetime of photogenerated carriers of modified UiO-66-NH2(Zr) was 4 times higher than the original UiO-66-NH2, greatly improving the utilization efficiency of photogenerated carriers.
4.2 Adsorption and membrane separation
MOFs has a specific pore structure similar to zeolite. Therefore, it is widely used in the field of adsorption. Zhang et al. [63] used cetyltrimethylammonium bromide (CTAB) as a modified material to synthesize a defect-engineered UiO-66 for toluene vapor adsorption. Experiments show that the concentration of CTAB affects the number of defect sites, which are the main adsorption sites for toluene vapor, and the improvement of adsorption performance is caused by the introduction of defects. Peterson et al. [64] used a hydrothermal method to prepare UiO-66-NH2 for NO2 removal experiments, and found that the adsorption capacity of UiO-66-NH2 for NO2 reached 30.4 μmol/g.
Zhang et al. [65] used methacrylic anhydride to modify UiO-66-NH2, and then polymerized it with butyl methacrylate (BMA) to prepare a UiO-66-NH2/BMA composite film, and used the composite film for efficient separation of Cr(VI) ions in water. The results prove that the UiO-66-NH2/BMA composite membrane can separate Cr(VI) ions from water by up to 8 mg/g. Compared with the traditional composite membrane, the composite membrane not only has a stronger separation ability, but also has a better compatibility.
4.3 Conduction
Phang et al. [66] prepared UiO-66-SH by utilizing the hydrothermal method, and then oxidized the ligand functional group to modify UiO-66-SO3H. The result demonstrated that the conductivity of UiO-66-SO3H reached 8.4 × 10−2 S/cm. Pu et al. [67] used the solvothermal method to prepare defect-engineered UiO-66/CNT (carbon nanotube) composite materials. They found that, due to the existence of defect sites, CNT can penetrate the entire UiO-66 particles to form a complete three-dimensional conductive network. The composite material not only has an excellent mechanical stability, but also has a good electrical conductivity.
5 Summary and outlook
In recent years, an increasing number of MOFs have been synthesized by hydrothermal synthesis or other methods of metal precursors and organic ligands. Defect-engineered UiO-66 have shown great advantages in various fields such as adsorption, separation and conduction, especially photocatalysis. However, the research on the defects of MOFs is still in its infancy. Therefore, the following prospects are made for the future defect-engineered MOFs.
At this stage, when monocarboxylic acid regulators are used to prepare defect-engineered MOFs, most studies focus on the influence of regulator concentration and acidity on the properties of MOFs. There are few studies involving carboxylic acid chain lengths (steric effect) on the properties of MOFs. The steric effect of carboxylic acid regulators may become a hot spot in the next stage.
For doped defect-engineered MOFs, the doping elements are mainly metal elements such as Ti, Ge, Co, and there are few non-metallic elements (C, N, O, P, etc.) for doping. Whether the doping of non-metallic elements can generate new energy levels and whether they can affect the separation efficiency of carriers may become the focus of the next stage of research.
When synthesizing defect-engineered MOFs, the formed defects may be an unfavorable factor for the stability of MOFs, which may limit the practical application of MOFs. For example, with regard to the use of carboxylic acid regulators, when the acidity is too strong, or the number of defective sites is too large, it will reduce the stability of MOFs. Further research is needed to improve the stability of defect-engineered MOFs.
The characterization methods that can fully and directly observe the local structure of defects are still lacking, and it is difficult to determine the location of the defect.
In summary, the paper reviews the synthetic methods, characterization methods, and application fields of the defect-engineered UiO-66 involving the intersection of materials science, catalysis science, and environmental science. The research results of the defect-engineered UiO-66 can be applied to the research fields of photocatalysis, adsorption, and conduction and so on. Therefore, it can be concluded that it has a promising prospect.