7 Heterogeneous Catalysis by Frustrated Lewis Pairs
257
the Lewis pair within the MOF. To probe the distribution of the DABCO/B(C 6 F 5 ) 3
moieties within the porous framework, they used high-angle annular dark-field scanning tunnelling electron microscopy (HAADF-STEM) and EDX spectroscopy. These
techniques revealed that the Lewis pairs were evenly distributed throughout the
structure, with no localised accumulation in particular regions.
The catalytic performance of 27 towards the hydroboration of imines was subsequently tested, and compared with the homogeneous DABCO/B(C 6 F 5 ) 3 FLP combination previously reported by Crudden et al. [77]. 27 was effective at catalysing
the reduction of N-tert-butyl-1-phenylmethanimine with HBpin, giving quantitative
conversion to the corresponding pinacolboramide after 2 h (Fig. 7.13b). A small
substrate scope for this reaction was examined, and revealed an interesting size
selectivity. The smaller imines, such as N-benzylidene-1-phenylmethanamine, were
reduced as effectively (in some cases more effectively) by 27 compared to the homogeneous system. However, for larger substrates, such as acridine, the catalytic activity
of 27 was significantly worse than the homogeneous FLP, and this was attributed to
the large substrates not being able to enter the pores of the MOF and interact with the
catalyst. In a similar fashion, 27 was also a proficient catalyst for the hydrogenation
of alkylidene malonates directly with H 2 gas (Fig. 7.13c). The long-term stability of
27 was explored. No leaching of the FLP from the MOF into solution was observed
at any point before or after catalysis, as measured by NMR spectroscopy of the supernatant. Furthermore, PXRD and N 2 sorption studies performed on 27 showed that the
material maintained crystallinity and pore structure after catalysis. These findings
prompted the authors to assess the recyclability of the heterogeneous catalyst, and
they showed that 27 could catalyse the reduction reaction in Fig. 7.13b for seven
successive cycles with quantitative conversion achieved after 2 h in all cases.
In a follow-up paper, Ma et al. functionalised the same MIL-101(Cr) MOF with a
slightly modified FLP combination, specifically DABCO and BMes(C 6 F 5 ) 2 [78]. The
synthesis and analysis were analogous to 27 described above. The new MOF-FLP
could activate H 2 at room temperature to afford the [H–DABCO]
+ [H–BMes(C 6 F 5 ) 2 ]
–
ion pair impregnated within the pores of the MOF, which could then be used for effective hydrogenation of imines. Interestingly, the system could carry out the chemoselective reduction of α,β-unsaturated imines to afford the unsaturated amines. This
selectivity is in contrast to the homogeneous FLP combination, which catalyses the
reduction of both the imine and alkene double bonds. The authors propose that the
chemoselectivity of the heterogeneous MOF-FLP system arises from an interaction
of the imine nitrogen lone pair with either an OH group on the chromium cluster
or one of the remaining open metal sites, which activates the C=N bond over the
C=C bond. These results highlight the exciting possibilities for tuning reactivity by
exploiting secondary interactions within the pores of MOFs that is not possible in
simple homogeneous systems.
In a different approach, Stylianou et al. designed a new water-tolerant MOF with
a Lewis acid site in the bridging organic linker [79]. The MOF, named SION105, consists of Eu
III dimers bridged by two water molecules and two tris(paracarboxylate)tridurylborane (TCTB) linkers (Fig. 7.14a). The MOF features zigzagshaped voids that comprise 36.6% of the total unit cell volume. The steric bulk around
257
the Lewis pair within the MOF. To probe the distribution of the DABCO/B(C 6 F 5 ) 3
moieties within the porous framework, they used high-angle annular dark-field scanning tunnelling electron microscopy (HAADF-STEM) and EDX spectroscopy. These
techniques revealed that the Lewis pairs were evenly distributed throughout the
structure, with no localised accumulation in particular regions.
The catalytic performance of 27 towards the hydroboration of imines was subsequently tested, and compared with the homogeneous DABCO/B(C 6 F 5 ) 3 FLP combination previously reported by Crudden et al. [77]. 27 was effective at catalysing
the reduction of N-tert-butyl-1-phenylmethanimine with HBpin, giving quantitative
conversion to the corresponding pinacolboramide after 2 h (Fig. 7.13b). A small
substrate scope for this reaction was examined, and revealed an interesting size
selectivity. The smaller imines, such as N-benzylidene-1-phenylmethanamine, were
reduced as effectively (in some cases more effectively) by 27 compared to the homogeneous system. However, for larger substrates, such as acridine, the catalytic activity
of 27 was significantly worse than the homogeneous FLP, and this was attributed to
the large substrates not being able to enter the pores of the MOF and interact with the
catalyst. In a similar fashion, 27 was also a proficient catalyst for the hydrogenation
of alkylidene malonates directly with H 2 gas (Fig. 7.13c). The long-term stability of
27 was explored. No leaching of the FLP from the MOF into solution was observed
at any point before or after catalysis, as measured by NMR spectroscopy of the supernatant. Furthermore, PXRD and N 2 sorption studies performed on 27 showed that the
material maintained crystallinity and pore structure after catalysis. These findings
prompted the authors to assess the recyclability of the heterogeneous catalyst, and
they showed that 27 could catalyse the reduction reaction in Fig. 7.13b for seven
successive cycles with quantitative conversion achieved after 2 h in all cases.
In a follow-up paper, Ma et al. functionalised the same MIL-101(Cr) MOF with a
slightly modified FLP combination, specifically DABCO and BMes(C 6 F 5 ) 2 [78]. The
synthesis and analysis were analogous to 27 described above. The new MOF-FLP
could activate H 2 at room temperature to afford the [H–DABCO]
+ [H–BMes(C 6 F 5 ) 2 ]
–
ion pair impregnated within the pores of the MOF, which could then be used for effective hydrogenation of imines. Interestingly, the system could carry out the chemoselective reduction of α,β-unsaturated imines to afford the unsaturated amines. This
selectivity is in contrast to the homogeneous FLP combination, which catalyses the
reduction of both the imine and alkene double bonds. The authors propose that the
chemoselectivity of the heterogeneous MOF-FLP system arises from an interaction
of the imine nitrogen lone pair with either an OH group on the chromium cluster
or one of the remaining open metal sites, which activates the C=N bond over the
C=C bond. These results highlight the exciting possibilities for tuning reactivity by
exploiting secondary interactions within the pores of MOFs that is not possible in
simple homogeneous systems.
In a different approach, Stylianou et al. designed a new water-tolerant MOF with
a Lewis acid site in the bridging organic linker [79]. The MOF, named SION105, consists of Eu
III dimers bridged by two water molecules and two tris(paracarboxylate)tridurylborane (TCTB) linkers (Fig. 7.14a). The MOF features zigzagshaped voids that comprise 36.6% of the total unit cell volume. The steric bulk around