*3.1. Organocatalysis*

### 3.1.1. Oxidation Reaction

Oxidation reactions are one of the most elementary reactions, and have already been extensively investigated by various catalytic systems. POM@MOF hybrid materials are considered as potential oxidation catalysts due to the presence of acidic sites within MOFs, along with the strong acidity and redox performance of POMs. Accordingly, some well-known MOFs have been reported to encapsulate POMs for their use in oxidation reactions, including MIL(Cr, Fe, or Al), UiO(Zr), and ZIF series, as well as Cu-BTC and NU-1000 frameworks. Among the different oxidation reactions, oxidative desulfurization (ODS) and the selective oxidation of alcohols and alkenes are the most studied reactions using POM@MOF catalysts.

ODS, as one of the promising methods for removing sulfur-containing compounds from fuels, has significant importance in both academic research and industrial chemistry. Since 2012, several Keggin- and sandwich-type POMs, including [A-PW9O34]<sup>9</sup>− [50], [PW12O40]<sup>3</sup>− [51–54], [PW11Zn(H2O)O39]<sup>5</sup>− [55,56], [Tb(PW11O39)2]<sup>11</sup>− [57], and [Eu(PW11O39)2]<sup>11</sup>− [58], have been incorporated into the cavities of MIL(Cr, Fe, or Al) for the ODS reaction, using H2O2 as the oxidant. The heterogeneous POM@MIL catalysts could not only be easily recycled and reused, but also showed a higher catalytic activity compared to

the homogeneous POM counterparts. For example, Balula's group reported that the heterogeneous Tb(PW11)2@MIL-101 (Tb(PW11)2 =[Tb(PW11O39)2] <sup>11</sup>−) catalyst exhibits 95% conversion of benzothiophene (BT) at 50 ◦C after 2 h, whereas the homogeneous Tb(PW11)2 catalyst affords a conversion of only 32% under the same reaction conditions [57]. Although the synthesized Tb(PW11)2@MIL-101 catalyst showed POM leaching, its structure and morphology remained intact after three consecutive ODS runs. Another study proved that the chemical and thermal stability of POM@MIL-101(Cr) systems could be enhanced compared to individual POMs and MOFs. More specifically, Silva's group demonstrated the high stability of the PW11@MIL-101 (PW11 = [PW11O39]<sup>7</sup>−) catalyst in aqueous H2O2, while PW11 decomposed into peroxo-complexes in the presence of H2O2 [59]. In addition, Naseri and co-workers observed that the thermal stability of the synthesized P2W18Ce3@MIL-101 (P2W18Ce3 = [(OCeIVO)3(PW9O34)2]12−) and P2W18Sn3@MIL-101 (P2W18Sn3 = [(HOSnIVOH)3(PW9O34)2]12−) materials improved in comparison to the single MIL-101(Cr) framework. The thermally stable POM@MOF materials exhibited >95% conversion of diphenyl sulfide after five cycles [29].

In an interesting study by Cao and co-workers, the effect of the window size within MOFs on the catalytic ODS performance of different POM@MOF materials was investigated [60]. In this work, PW12 was encapsulated into three robust MOFs with different window sizes, namely MIL-100(Fe) (8.6 and 5.8 Å), UiO-66 (6 Å), and ZIF-8 (3.4 Å) (see Table 1). Among them, PW12@MIL-100(Fe) exhibited the highest activity (92.8%) for the oxidation of 4,6-dimethyldibenzothiophene (3.62 × 6.17 × 7.86 Å3) compared to UiO-66 (39.1%) and ZIF-8 (9.1%). The observed higher activity was attributed to the large window size of MIL-100(Fe), which enabled a fast diffusion of the substrate into the cages. Another important parameter is the influence of the POM loading on the catalytic performance. The conversion of dibenzothiophene (DBT) was at least two times higher when 16%-PW12@MIL-100(Fe) was used as a catalyst in comparison to the 7%-PW12@MIL-100(Fe) catalyst, owing to the higher POM loading. However, when the loading was increased to 35%, the conversion of DBT decreased a lot due to partial pore blocking, which limited the diffusion of reactants to the active sites.

To further enhance the reactivity and recyclability of POM@MOF materials in ODS reactions, amine-functionalized MOFs were employed for encapsulating POMs owing to the strong electrostatic interaction between amine groups and POM anions, including NH2-MIL-101(Cr) [33], NH2-MIL-101(Al) [36,56], and NH2-MIL-53(Al) [58]. For instance, Cao and co-workers reported the incorporation of PW12 into NH2-MIL-101(Cr) as a catalyst for the ODS reaction. The obtained material gave a full conversion of DBT at 50 ◦C after 1 h [33]. Notably, the reusability tests indicated that the conversion of DBT remained unchanged during six consecutive catalytic cycles using PW12@NH2-MIL-101(Cr) as a catalyst, due to the strong electrostatic interactions between PW12 and the amine groups. Another report by Su and co-workers showed that the PW12@MIL-101(Cr)-diatomite gave 98.6% conversion of DBT at 60 ◦C for 2 h after three consecutive cycles, which was attributed to the high dispersion of POMs [54].

In addition to ODS reactions, the selective oxidation of alkenes [61–66] and alcohols [34] was evaluated using POM@MIL catalysts. For example, Bo and co-workers synthesized H3<sup>+</sup>xPMo12−<sup>x</sup>VxO40@MIL-100(Fe) (x = 0, 1, 2) materials and their catalytic performance were assessed in the oxidation of cyclohexene, using H2O2 as the oxidant [66]. Among them, the H4PMo11VO40@MIL-100(Fe) material exhibited 83% conversion of cyclohexene, with an excellent selectivity for 2-cyclohexene-1-one (90%) after five successive catalytic cycles. In 2007, our group developed a new POM@MIL-101 catalyst based on dual amino-functionalized ionic liquid (DAIL) [34]. Firstly, DAIL was introduced onto the coordinatively unsaturated chromium sites of MIL-101(Cr) by a post-synthetic strategy, followed by immobilization of the Keggin-type PW12 onto the DAIL-modified MIL-101 through anion exchange (see Figure 2). The PW12/DAIL/MIL-101 catalyst exhibited a very high turnover number (TON: 1900) for the selective oxidation of benzyl alcohol towards benzaldehyde at 100 ◦C for 6 h. The PW12/DAIL/MIL-101 catalyst demonstrated a higher catalytic activity compared to the PW12/MIL-101 catalyst without DAIL functionalities (TON: 1400). The higher activity was due to the presence of remaining free amino groups anchored on the imidazolium moieties of DAIL, which play a

crucial role in enhancing the accessibility of TBHP as the oxidant. Moreover, the PW12/DAIL/MIL-101 catalyst was reused for at least five cycles, with no significant leaching of the tungsten species.

**Figure 2.** (**a**) Schematic illustration of the preparation of PW/dual amino-functionalized ionic liquid (DAIL)/MIL-101(Cr); (**b**) recyclability of the PW/DAIL/MIL-101(Cr) catalyst. Reprinted with permission from [34]. Copyright (2017), Royal Society of Chemistry.

Another type of MOF, namely Cu-BTC, has also been employed to encapsulate POMs. In 2008, six kind of Keggin-type POMs were encapsulated into Cu-BTC (named NENU-n, NENU = Northeast Normal University) using a one-pot hydrothermal method and their crystal structures were determined [43]. Subsequently, various POMs were encapsulated into the Cu-BTC framework and their catalytic performance was examined in ODS reactions [67,68], the oxidation of alcohols [69,70], olefins [39,71–73], benzene, and H2S [41,74]. For example, Zheng et al. prepared different sizes of nanocrystal-based catalysts, [Cu2(BTC)4/3(H2O)2]6[H5PV2Mo10O40] (NENU-9N), by using various copper salts and adjusting the pH of the solution for the ODS reaction (see Figure 3) [75]. They proposed that the reaction kinetics can be facilitated by decreasing the size of the nanocrystals. The 550 nm NENU-9 showed a significantly higher conversion of DBT (~90%) in 60 min compared to 300 μm NENU-9 (41%) and the homogeneous POM (2%) in 90 min. To further improve the stability of POM@MOF materials and their catalytic ODS performance, POM@MOF compounds were confined in other porous materials, e.g., MCM-41 [76,77], carbon nanotubes [78], mesoporous SBA-15 [79], and hollow ZSM-5 zeolite [80]. For example, POM@Cu-BTC was confined in the pores of MCM-41 to prevent deactivation of the catalyst [76]. The POM@Cu-BTC@MCM-41 (POM = Cs2HPMo6W6O40) exhibited almost full conversion (99.6%) of DBT in 180 min under optimal reaction conditions and could be reused more than 15 times without a significant loss of activity. Lu and co-workers prepared a series of POM@Cu-BTC (POM = PW12, [PMo12 − xVxO40](3 + x)− (x = 0, 1, 2, 3)) catalysts and investigated their performance for the oxidation of benzyl alcohol to benzaldehyde, with H2O2 as the oxidant (Figure 4) [70]. The authors observed that the vanadium-containing POMs improved the conversion of benzyl alcohol because of the high redox ability of the POMs. However, when increasing the vanadium content in the POMs, overoxidation to benzoic acid resulted in a lower selectivity towards benzaldehyde. The PMo12@Cu-BTC showed approximately 75% conversion of benzyl alcohol with ~90% selectivity towards benzaldehyde, whereas the PMo9V3@Cu-BTC showed ~98% conversion of benzyl alcohol with ~65% selectivity using the same reaction conditions. In other words, the product distribution could be controlled by adjusting the redox capability of the POMs.

Interestingly, in a few studies, a synergistic effect between the POM and Cu-BTC was observed [41,74,81]. For example, Hill prepared CuPW11@Cu-BTC (CuPW11 = [CuPW11O39]<sup>5</sup>−) for the oxidation of several sulfur compounds and proposed synergistic effects between PW11Cu and Cu-BTC [41]. Not only the hydrolytic stability of the hybrid POM@MOF was improved, but also the

TON (12), as the oxidation of H2S under ambient conditions increased significantly compared to the individual Cu-BTC (0.02) and POM (no production).

**Figure 3.** (a) Field emission SEM of Northeast Normal University (NENU)-9N with (**a**) copper nitrate as the metal source at pH 2.5, (**b**) copper acetate as the metal source at pH 2.5, and c) copper acetate as the metal source at pH 4.0. The percentage of DBT-to-DBTO2 conversion versus reaction time by using a) NENU-9N (average diameter = 550 nm), (**b**) NENU-9 (average diameter = 300 mm), and (**c**) POVM (average diameter = 300 mm) as catalysts. Reaction conditions: catalyst (0.01 mmol), DBT (147 mg, 0.8 mmol), and isobutyraldehyde (0.72 mL, 8 mmol) in decalin (50 mL) at 80 ◦C. Reprinted with permission from [75]. Copyright (2013), John Wiley and Sons.

**Figure 4.** Oxidation of benzyl alcohol by different POM@MOF-199 catalysts. Reprinted with permission from [70]. Copyright (2014), John Wiley and Sons.

The robust Zr-based MOFs have also attracted much attention for hosting POMs for oxidation reactions. The earliest study on the introduction of POMs into a Zr-based MOF was reported by Dolbecq and co-workers in 2015 [82]. Three tungstate POMs ([PW12O40]<sup>3</sup>− (12 Å), [PW11O39]<sup>7</sup><sup>−</sup>, and [P2W18O62]<sup>6</sup>− (14 Å)) were encapsulated into the pores of UiO-67. Subsequently, Dai and co-workers examined the catalytic performance of 35%-PW12@UiO-66 for the selective oxidation of cyclopentene (CPE) to glutaraldehyde (GA) [83]. The catalyst showed ~95% conversion of CPE, with a ~78% yield for GA at 35 ◦C after 24 h of reaction. Unfortunately, the catalyst showed PW12 leaching (~3 wt%) after three catalytic cycles. To address this POM leaching issue, Yu and co-workers used UiO-bpy (bpy = 2,2-bipyridine-5,5-dicarboxylic acid) to encapsulate polyoxomolybdic cobalt (CoPMA) [84]. The bpy sites of the UiO-bpy framework provided an extra interaction with the POM compared to the UiO-67 without bpy moieties. The catalytic activities of CoPMA@UiO-bpy and CoPMA@UiO-67 were assessed in the oxidation of styrene, using O2 as the oxidant. The CoPMA@UiO-bpy exhibited the highest catalytic performance, with 80% conversion of styrene and 59% selectivity towards styrene epoxide.

Another Zr-based MOF, denoted as NU-1000, with small triangular (12 Å) and larger hexagonal (31 Å) channels, has been used to support POMs such as [PW12O40]<sup>3</sup>− and [PMo10V2O40]<sup>5</sup>− [85–87]. For example, Farha's group prepared PW12@NU-1000 through an impregnation method for the oxidation of 2-chloroethyl ethyl sulfide (CEES), using H2O2 as the oxidant. The authors demonstrated that the most likely location for PW12 clusters is in the small triangular channels, which was further confirmed by means of powder X-ray diffraction, scanning transmission electron microscopy, and difference envelope density analysis. At the same time, PW12@NU-1000 showed a higher conversion of CEES (98% after 20 min) compared to the pristine NU-1000 (77% after 90 min) and homogeneous POM (98% after 90 min). However, the PW12@NU-1000 exhibited only 57% selectivity towards 2-chloroethyl ethyl sulfoxide (CEESO). In a subsequent work, the authors demonstrated that the PW12 could migrate from the micropores to the mesopores of NU-1000 under mild thermal activation (see Figure 5). Moreover, the PW12@NU-1000 showed a full conversion of CEES after 5 min, with ~95% selectivity towards CEESO. Recently, this group also prepared the PV2Mo10@NU-1000 catalyst by using the same method and the synthesized material showed a full conversion of CEES, with O2 as the oxidant.

**Figure 5.** Structural representations of the PW12@NU-1000. Reprinted with permission from [86]. Copyright (2018), Royal Society of Chemistry.

In addition to the well-known MOFs, several other POM@MOF hybrid materials, including [Co(BBPTZ)3][HPMo12O40]·24H2O and [CuI6(trz)6(PW12O40)2], have been synthesized and applied for ODS [88], the oxidation of aryl alkenes [89,90], alkylbenzenes [91], and alcohols [92] (see Table 2).

Besides the use of POMs encapsulated in the cages of MOFs, some POMs have been covered on the surface of MOFs to achieve core–shell structured hybrid materials for oxidation reactions [46,93]. For example, PW12 was loaded onto the ZIF-8 surface to obtain a core–shell catalyst for the oxidation of benzyl alcohol. Notably, strong O-N bonding between PW12 and the imidazole group of the ZIF-8 was detected through X-ray photoelectron spectroscopy and X-ray absorption near-edge structure measurements. Accordingly, the ZIF-8@PW12 material was insoluble in hydrophilic solvents. The ZIF-8@PW12 material exhibited a high conversion of benzyl alcohol (>95%), with 90% selectivity towards benzaldehyde, and outperformed the activity of pure PW12 (51%) and ZIF-8 (30%).


**Table 2.** Application of POM@MOF materials in heterogeneous catalysis.

60

 Fe-BTC


**Table 2.** *Cont.*

One-pot fatty acids to biodiesel

[107]

 H3PMo12O40


**Table 2.** *Cont.*


**Table 2.** *Cont.*

BBPTZ = 4,4-bis(1,2,4-triazol-1-ylmethyl)biphenyl]; BBTZ = 1,4-bis(1,2,4-triazol-1-ylmethyl)benzene; trz = 1,2,4-triazole; imi = imidazole; ampyd = 2-aminopyridine; bpy = bipyridine; Phen = 1,10-phenanthroline; bipy = 4,4-bipyridine; Hpzc = pyrazine-2-carboxylic acid, pz = pyrazine; bbi = 1,1-(1,4-butanediyl)bis(imidazole); DTP = dodecatungstophosp; TBA = tetrabutylammonium.

### 3.1.2. Condensation Reaction

POM@MOF has revealed potential applications in a range of condensation reactions for producing value-added cyclic organic compounds. Recently, Malkar et al. compared the catalytic performance of three different catalysts, namely 20%-Cs-DTP-K10, 18%-DTP@ZIF-8, and Al0.66-DTP@ZIF-8 (DTP = dodecatungstophosp), for the aldol condensation of HMF (5-hydroxymethylfurfural), as shown in Figure 6 [95]. It has been proved that the substitution of protons of heteropolyacids with metal ions increases the mobility of protons, which results in an enhancement of the acidity. Based on NH3-TPD analysis, Cs-DTP-K10 displays the highest acidity (1.51 mmol g<sup>−</sup>1), whereas DTP@ZIF-8 and Al-DTP@ZIF-8 possess 0.44 and 0.54 mmol g<sup>−</sup><sup>1</sup> of acidic sites, respectively. Cs-DTP-K10, with the highest acidity, showed the highest activity for the aldol condensation of HMF and acetone to selectively produce the desired C9 product (71.6% after 6 h of reaction), while the selectivity was only 43.1%. Although the total number of acidic sites was much lower in the case of the Al-DTP@ZIF-8 catalyst, a good conversion of 63.1% was still obtained after 6 h of reaction, which is comparable to the former value. Notably, the Al-DTP@ZIF-8 catalyst displayed a much higher selectivity (~92%) towards the C9 product compared to C15. The lowest conversion of HMF was achieved in the case of the 18%-DTP@ZIF-8 material with the lowest acidity. However, a high selectivity towards the C9 product was observed. The higher selectivity towards the C9 adduct, as the desired product in the presence of the DTP@ZIF-8 and Al-DTP@ZIF-8 catalysts, confirms the shape selectivity supplied by the small pore diameter of ZIF-8, which can prevent the production of the C15 adduct.

**Figure 6.** Aldol condensation of 5-hydroxymethylfurfural (HMF) with acetone over Al-DTP@ZIF-8. Reproduced with permission from [95]. Copyright (2019), American Chemical Society.

Another example of the use of MOFs in condensation reactions is the well-known MIL-101. For this purpose, PW12@MIL-101(Cr) composites were synthesized through the direct hydrothermal procedure or post-synthesis modification route [98]. The acidic sites within the MIL-101 and PW12@MIL-101(Cr) materials are desirable for catalyzing the Baeyer condensation of benzaldehyde and 2-naphthol, in the three-component condensation of benzaldehyde, 2-naphthol, and acetamide, as depicted in Figure 7. While no product was produced in the absence of catalysts, a high yield of around 95% was observed for the formation of 1-amidoalkyl-2-naphthol at 130 ◦C using microwave heating for 5 min. Moreover, no leaching of the active sites was observed, and the catalyst could be reused for four cycles without a notable loss in the product yield.

**Figure 7.** Condensation of benzaldehyde, 2-Naphthol, and acetamide. Reprinted with permission from [98]. Copyright (2012), American Chemical Society.

In addition, PW12 clusters were uniformly encapsulated in the cages of MIL-101 as a selective heterogeneous catalyst for the self-condensation of cyclic ketones [97]. As can be observed in Figure 8, the self-condensation of cyclopentanone can result in three different products based on the active sites in the applied catalysts. By using PW12 as the catalyst, a conversion of around 78% could be obtained after 24 h reaction to trindane as the main product. However, PW12@MIL-101 exhibits a considerably higher selectivity (>98%) towards the mono-condensed component (2-cyclopentylidenecyclopentanone) as the desired product due to the possibility of shape-selective catalysis. The PW12@MIL-101 catalyst could be recycled up to five cycles, with no obvious reduction in the conversion and selectivity.

**Figure 8.** Reaction figure of a cyclopentanone self-condensation reaction. Adapted with permission from [97]. Copyright (2015), Royal Society of Chemistry.

#### 3.1.3. Esterification Reaction

Modified MOFs with POMs can be employed as active catalysts for a wide range of esterification reactions. Biodiesel, as a secure and sustainable energy source, is a promising alternative for fossil fuel-based energy systems [156]. Among the various methods for biodiesel production, transesterification is the most common procedure. Recently, Xie et al. investigated the catalytic one-pot transesterification-esterification of acidic vegetable oil transesterification reaction over a functionalized UiO-66-2COOH with a Keggin-type POM, namely, AILs/POM/UiO-66-2COOH (AIL = sulfonated acidic ionic liquid) (see Figure 9) [114]. The prepared catalyst displayed synergistic benefits arising from the introduction of AIL as Brønsted acid sites. The presence of both Brønsted acid sites of ILs and Lewis acid sites of POM promoted the catalytic reaction for green biodiesel production. The control experiments showed that all of the applied POMs (PW12, SiW12, and PMo12) could convert soybean oil to biodiesel with a high catalytic performance (conversion of ~100%). However, challenges associated with the work-up and recyclability of these homogeneous catalysts limit their potential application. The pristine UiO-66-2COOH material presented a poor activity, with an oil conversion of around 8% because of its insufficient acidic properties. In addition, the PW12@UiO-66-2COOH, SiW12@UiO-66-2COOH, and PMo12@UiO-66-2COOH composites suffered from sluggish reaction kinetics with conversions below 30% towards biodiesel production, which could have been due to the lack of enough acidic sites required to advance the catalytic reaction. Another control experiment was performed by using the sulfonic acid-functionalized IL as the homogeneous catalyst, affording a high catalytic activity of around 99% conversion. It is interesting to note that AILs/POM/UiO-66-2COOH catalysts can combine the advantages of POMs, AILs, and porous MOFs and therefore present the highest catalytic performance in the mentioned reaction (conversion > 90%). Furthermore, the strong interaction between the POMs and AILs was able to hinder the leaching of active components into the reaction media, which further resulted in no notable loss in the catalytic conversion of oil to biodiesel after five consecutive catalytic cycles.

**Figure 9.** Synthesis procedure of the AILs/HPW/UiO-66-2COOH catalyst, and one-pot transesterificationesterification of acidic vegetable oils. Reprinted from [114]. Copyright (2019), with permission from Elsevier.

In 2015, Liu et al. described an effective procedure for designing NENU-3a with different crystal morphologies (cubic and octahedral) comprised of a Cu-BTC skeleton and encapsulated phosphotungstic acid catalyst [113]. The morphology of this framework was generated by applying the method of coordination modulation, using *P*-toluic acid as the modulator. The NENU-3a with cubic crystals ((100) facets) could effectively promote the conversion of long-chain (C12-C22) fatty acids into corresponding monoalkyl esters (>90% yield) compared to the octahedral counterpart (<22% yield). Moreover, the cubic NENU-3a catalyst was highly robust and could be reused for five reaction runs with a preserved structure and catalytic activity. This report confirms the vital impact of morphological control on MOFs for improving the facet exposure of catalytic sites, which accordingly results in an enhancement of the catalytic performance, especially for bulky substrates with limited access to the catalytic active sites within the pores of MOF catalysts. Another important feature of MOFs is the possibility to control the product selectivity arising from the pore size effect of MOFs. Within this context, Zhu et al. studied the selective esterification of glycerol using a MOF-supported POM catalyst [106]. The catalytic performance of the obtained POM@Cu-BTC catalyst was compared to the metal oxide-supported POMs as the reference materials. Since there was no pore limitation impact using the POM@metal-oxide catalyst, the conversion of glycerol stopped at the acid stage without further reaction and was free to be released from the reaction site (Figure 10). However, when the POM@Cu-BTC catalyst was applied, diffusion of the acid product within the MOF pores was limited and further reaction of the acid product produced the corresponding ester compound.

**Figure 10.** Diffusion limited glycerol transformation on MOF-POMs. Reprinted with permission from [106]. Copyright (2015), Royal Society of Chemistry.

#### 3.1.4. Other Organic Transformations

POMs exhibit grea<sup>t</sup> potential as solid acid catalysts because of their strong Brønsted acidity. The first report on a well-defined MOF-supported POM compound, which behaved as a true heterogeneous acid catalyst, was reported by Su et al. [43]. In this work, a series of POM@MOF catalysts were synthesized using a one-pot method. The POM@MOF compound, which contained the strongest Keggin Brønsted acid PW12, was examined in the hydrolysis of ethyl acetate in the presence of an excess amount of water. This catalyst, denoted as NENU-3a, exhibited almost full conversion (>95%) after approximately 7 h of reaction, which is far more superior than most inorganic solid acids and comparable to organic solid acids. More specifically, when the activity was reported per unit of acid, NENU-3a was 3-7 times more active than H2SO4, PW12, nafion-H, and Amberlyst-15. In addition to this, no deactivation of the acid sites by water was observed and no leaching of the POM was noted up to at least 15 cycles. Later on, the same group reported the use of POMs as templates for the construction of novel hybrid compounds, for which the properties of the POM could be tailored towards a specific application [123–125]. One of these targeted applications was the adsorption and subsequent hydrolysis of the nerve gas dimethyl methylphosphate to methyl alcohol, for which the conversion increased up to 93% when the temperature was raised to 50 ◦C [123]. Recycling tests demonstrated that the structural integrity was preserved up until at least 10 cycles. However, it is important to note here that a stabilizing effect of the POM on the MOF will only be obtained when the shape, size, and symmetry of the POM match the MOF host [135]. This stabilizing effect even allowed the application of POM@MOF catalysts in aggressive reactions, as was demonstrated in the very nice work of Hupp, Farha, and Notestein [133]. In this study, the Zr-based MOF, NU-1000, was loaded with H3PW12O40 for its use in the strong acid-catalyzed reaction of o-xylene isomerization/disproportionation at 250 ◦C (see Figure 11). At low POM loadings (0.3 to 0.7 POM per Zr6 node), no activity was observed, which was due to the collapse of the POM and/or MOF structure upon activation or at the start of the reaction. However, when the loading was increased to its maximum, with 1 Keggin unit per unit cell of NU-1000, the hybrid catalyst exhibited an initial reactivity in the examined C-C skeletal rearrangemen<sup>t</sup> reaction which was even higher than that of the reference WOx-ZrO2 catalyst.

**Figure 11.** Phosphotungstic acid encapsulated in NU-1000 for its use in the aggressive hydrocarbon isomerization reaction. Reprinted with permission from [133]. Copyright (2018), American Chemistry Society.

While, in the previously discussed studies, the Keggin ion acted as a template to stabilize the microporous/mesoporous structure of theMOF, the group ofMartens et al. used this templating mechanism to introduce mesopores separated by uniform microporous walls in a single crystal structure [121] (see Figure 12). More specifically, a hierarchical variant of the Cu-based MOF, Cu-BTC, was synthesized using a dual templating approach in which the Keggin ions served as a molecular template for the structural motif of the MOF, while the surfactant cetyltrimethylammonium bromide was used to introduce mesoporosity. The resulting mesoporous MOF, denoted as COK-15, was investigated in the alcoholysis of styrene oxide, which often suffers from a low selectivity. The COK-15 catalyst not only exhibited a remarkable activity (100% conversion), but also achieved 100% selectivity for 2 methoxy-2 phenylethanol after 3 h of reaction at 40 ◦C. For comparison, the microporous POM@Cu-BTC and Cu-BTC material only showed 40% and 2% conversion, respectively. The authors addressed the good activity of the COK-15 to the mesoporous feature, which allowed efficient mass transport. Moreover, the catalyst could be recycled for at least four runs, with a negligible loss in activity and selectivity.

**Figure 12.** A copper benzene tricarboxylate metal–organic framework, COK-15, with a wide permanent mesoporous feature stabilized by Keggin POM ions for the methanolysis of styrene oxide. Adapted with permission from [121]. Copyright (2012), American Chemistry Society.

Besides this increase in stability after the embedding of the POM in a MOF support, several groups have demonstrated the mutual activation of the POM gues<sup>t</sup> and MOF support [32,119,127]. A very special and extreme example of such a synergism was demonstrated in the work of Kögerler et al. [128]. In this work, a POM@MOF composite was prepared through a hydrothermal reaction in which an Mn-based POM was added to the reaction mixture to synthesize MIL-100. The obtained 30 wt% loaded Mn-POM@MIL-100 was evaluated for its catalytic performance in the reduction of p-nitrophenol to p-aminophenol in the presence of NaBH4. While both the individual compounds exhibited no catalytic activity, the composite showed an excellent performance (the activity and rate constant at 50 ◦C were 683 L g<sup>−</sup><sup>1</sup> s<sup>−</sup><sup>1</sup> and 0.23 min−1, respectively), which was even comparable to those observed for noble metal-based catalysts. The authors stated that the high catalytic activity originated from the fact that the Mn-POM facilitated the electron transfer from BH4 − to the Fe3<sup>+</sup> Lewis acid sites of the MOF, as they assumed that the MIL-100 alone could not accept electrons directly from BH4 −. Additionally, the group of Shul observed a distinct acid-base synergy upon examination of the core–shell structured heteropoly acid-functionalized ZIF-8 in the transesterification of rapeseed oil with methanol to produce biodiesel [46]. More than 95% of the rapeseed oil was converted to biodiesel due to the simultaneous presence of the acid functionalities of the POM and the basicity of the imidazolate groups of the MOF, whereas the pure POM and ZIF-8 catalysts showed a catalytic performance of 61% and 32%, respectively. Moreover, the strong chemical O-N bonding between the Keggin and the imidazole units ensured a good recyclability, with no noticeable decrease in the catalytic performance after five cycles and no POM leaching.
