7 Heterogeneous Catalysis by Frustrated Lewis Pairs
255
as long as the reaction is carried out sequentially with H 2 then CO 2 . A wider range
of Lewis acid and base moieties that could be incorporated in MOFs for the hydrogenation of CO 2 were subsequently explored, with phenylene-bridged amine-borane
FLPs being the most promising class of catalyst [71].
Ye and Johnson subsequently targeted the formation of methanol from the reduction of CO 2 via multiple successive hydrogenation steps [72]. To achieve this, they
designed a new catalyst based on UiO-67; this is isostructural to the previously
described UiO-66 and has similar stability and functionality, but has a larger pore
volume due to the larger 4,4
-biphenyldicarboxylate (BPDC; Fig. 7.12b) linker. In
this study, the functionalised linker 25 was used (Fig. 7.12b), and the large pore size
combined with the relatively small FLP moiety enabled the researchers to model four
functionalised linkers per primitive cell (as opposed to 24-UiO-66 that was limited
to one functionalised linker per primitive cell). It was shown that the functionalised
25 4 -UiO-67 featuring four equivalents of 25 per primitive cell provided an energetically feasible pathway for the formation of methanol, with the key step being the
intramolecular release of water from the methanediol (CH 2 (OH) 2 ) intermediate to
afford formaldehyde. Furthermore, in this system, the heterolytic dissociation of H 2
was preferred over the chemisorption of CO 2 , meaning the catalyst is not poisoned
by the latter.
The group also explored the effect of topology on the catalytic ability of the
systems [73]. The class of MOFs denoted as MIL-140 are porous zirconium dicarboxylate structures with triangular channels along the z-axis, composed of zirconium oxide chains connected with different dicarboxylate linkers. They examined
functionalised versions of MIL-140B (featuring the previously described linker 25)
and MIL-140C (featuring linker 26; Fig. 7.12c), and compared them to the aforementioned functionalised 24-UiO-66 and 25-UiO-67 systems. This series of MOFs
differ by both the topology of the pores (triangular channels for MIL-140 and octahedral cavities for UiO-66 and UiO-67); the size of the pores (pore size: MIL-140B
< UiO-66 < MIL-140C < UiO-67); and the orientation of the FLP site within the
pore (thus the steric hindrance within does not necessarily scale with pore size).
As in their previous studies, the authors showed that these systems can catalyse the
hydrogenation of CO 2 . They also showed that the heterolytic dissociation of H 2 is
strongly favoured over chemisorption of CO 2 in the functionalised 25-MIL-140B
and 25-UiO-67 MOFs due to the confinement effects and topological constraints,
providing a route to avoid catalyst poisoning. Furthermore, the small pore size of
functionalised 25-MIL-140B stabilises the formation of a pre-activated CO 2 species,
enabling a mechanistic pathway to formic acid with a lower overall energy barrier,
and highlighting the importance of considering the local environment of the FLP
system.
Finally, from a theoretical standpoint at least, Johnson et al. used the previously described 25-UiO-67 to probe the possibility of synthesising formaldehyde
directly from H 2 and CO [74]. They determined that this system can selectively
form formaldehyde instead of methanol, and do so with significantly lower energy
barriers compared to the gas-phase system. If this reaction can be experimentally
255
as long as the reaction is carried out sequentially with H 2 then CO 2 . A wider range
of Lewis acid and base moieties that could be incorporated in MOFs for the hydrogenation of CO 2 were subsequently explored, with phenylene-bridged amine-borane
FLPs being the most promising class of catalyst [71].
Ye and Johnson subsequently targeted the formation of methanol from the reduction of CO 2 via multiple successive hydrogenation steps [72]. To achieve this, they
designed a new catalyst based on UiO-67; this is isostructural to the previously
described UiO-66 and has similar stability and functionality, but has a larger pore
volume due to the larger 4,4
-biphenyldicarboxylate (BPDC; Fig. 7.12b) linker. In
this study, the functionalised linker 25 was used (Fig. 7.12b), and the large pore size
combined with the relatively small FLP moiety enabled the researchers to model four
functionalised linkers per primitive cell (as opposed to 24-UiO-66 that was limited
to one functionalised linker per primitive cell). It was shown that the functionalised
25 4 -UiO-67 featuring four equivalents of 25 per primitive cell provided an energetically feasible pathway for the formation of methanol, with the key step being the
intramolecular release of water from the methanediol (CH 2 (OH) 2 ) intermediate to
afford formaldehyde. Furthermore, in this system, the heterolytic dissociation of H 2
was preferred over the chemisorption of CO 2 , meaning the catalyst is not poisoned
by the latter.
The group also explored the effect of topology on the catalytic ability of the
systems [73]. The class of MOFs denoted as MIL-140 are porous zirconium dicarboxylate structures with triangular channels along the z-axis, composed of zirconium oxide chains connected with different dicarboxylate linkers. They examined
functionalised versions of MIL-140B (featuring the previously described linker 25)
and MIL-140C (featuring linker 26; Fig. 7.12c), and compared them to the aforementioned functionalised 24-UiO-66 and 25-UiO-67 systems. This series of MOFs
differ by both the topology of the pores (triangular channels for MIL-140 and octahedral cavities for UiO-66 and UiO-67); the size of the pores (pore size: MIL-140B
< UiO-66 < MIL-140C < UiO-67); and the orientation of the FLP site within the
pore (thus the steric hindrance within does not necessarily scale with pore size).
As in their previous studies, the authors showed that these systems can catalyse the
hydrogenation of CO 2 . They also showed that the heterolytic dissociation of H 2 is
strongly favoured over chemisorption of CO 2 in the functionalised 25-MIL-140B
and 25-UiO-67 MOFs due to the confinement effects and topological constraints,
providing a route to avoid catalyst poisoning. Furthermore, the small pore size of
functionalised 25-MIL-140B stabilises the formation of a pre-activated CO 2 species,
enabling a mechanistic pathway to formic acid with a lower overall energy barrier,
and highlighting the importance of considering the local environment of the FLP
system.
Finally, from a theoretical standpoint at least, Johnson et al. used the previously described 25-UiO-67 to probe the possibility of synthesising formaldehyde
directly from H 2 and CO [74]. They determined that this system can selectively
form formaldehyde instead of methanol, and do so with significantly lower energy
barriers compared to the gas-phase system. If this reaction can be experimentally