470
H. Hirao
is interesting in that harvesting light usually requires a less abundant metal, typically
ruthenium or iridium.
DFT calculations were used to understand the mechanism of this photocatalytic
C–C bond cleavage reaction in detail (Fig. 6) [57]. As time-dependent DFT and
correlated ab initio MO methods are deemed too expensive for describing the entire
reaction pathways in the photo-excited states, ordinary DFT calculations were performed to model excited states. It turns out that the first excited singlet state can be
described reasonably well with the broken-symmetry DFT method. Also, the triplet
excited state can be described by simply specifying the spin state as triplet. The
calculations showed that the β-O-4 substrate coordinating to the equatorial site of
the vanadium-oxo complex cannot thermally undergo C–C bond cleavage. However,
the low-lying excited state in the triplet (T 1 ) state, which corresponds to a ligandto-metal charge transfer (LMCT) excited state, has a low barrier to homolytic C–C
bond cleavage. The excited state therefore makes the reaction possible. In this LMCT
transition, an electron is transferred from the hydrazine benzohydroxamate ligand to
the vanadium(V) center, to form an oxidized ligand and a vanadium(IV) state. After
an electronic excitation to a higher-lying excited state, intramolecular relaxation,
internal conversion, and intersystem crossing may occur to access the LMCT-type
T 1 state, in which homolytic C–C bond cleavage is a facile process. The low barrier obtained here is attributed to the fact that one of the two unpaired electrons,
which is produced upon homolytic C–C bond cleavage, is efficiently absorbed into
the low-lying orbital of the oxidized ligand (Fig. 7). Thus, as a result of the C–C
bond cleavage, aryl formate (one of the major products) and a radical species are left
behind. The latter species will rapidly combine with an O 2 molecule that is present in
the system under aerobic conditions. Subsequently, the alkyl peroxy radical binds to
the vanadium center, and the formal oxidation state of vanadium reverts back to +5.
In the second step, aryl formate is formed (Fig. 8). After all, the oxidation state of the
vanadium center is restored to +5, so that the vanadium complex can play a catalytic
role by effecting C–C cleavage in many catalytic rounds. The attachment of electronwithdrawing groups to the hydrazine benzohydroxamate led to the improvement of
the catalytic activity; this can be attributed to the stabilization of the ligand orbital,
which enhances the process illustrated in Fig. 7 [56].
In the photocatalytic reaction under anaerobic conditions, only a small amount
(3%) of the aryl aldehyde product was obtained in experiment. The formation of
the aldehyde product can be understood from the fact that the initial LMCT-driven
homolytic C–C bond cleavage process to form the aldehyde (the process from RC
to Int1 in Fig. 6) is possible even in the absence of O 2 . As for the low yield of
the aldehyde product, it indicates that the vanadium complex is consumed rapidly,
instead of acting as a catalyst. According to our DFT analysis (Fig. 9), the recovery of
the vanadium(V) state under anaerobic conditions is possible after a second LMCT,
but the subsequent barrier is higher than that for the initial C–C bond cleavage
process. Thus, the LMCT-driven C–C bond cleavage is favored over the LMCT-driven
vanadium(V) recovery, which results in the rapid depletion of the catalytically active
vanadium(V) species via preferential C–C bond cleavage. The pathways obtained for
the reaction under anaerobic conditions theoretically lead to aryl formate. However,
H. Hirao
is interesting in that harvesting light usually requires a less abundant metal, typically
ruthenium or iridium.
DFT calculations were used to understand the mechanism of this photocatalytic
C–C bond cleavage reaction in detail (Fig. 6) [57]. As time-dependent DFT and
correlated ab initio MO methods are deemed too expensive for describing the entire
reaction pathways in the photo-excited states, ordinary DFT calculations were performed to model excited states. It turns out that the first excited singlet state can be
described reasonably well with the broken-symmetry DFT method. Also, the triplet
excited state can be described by simply specifying the spin state as triplet. The
calculations showed that the β-O-4 substrate coordinating to the equatorial site of
the vanadium-oxo complex cannot thermally undergo C–C bond cleavage. However,
the low-lying excited state in the triplet (T 1 ) state, which corresponds to a ligandto-metal charge transfer (LMCT) excited state, has a low barrier to homolytic C–C
bond cleavage. The excited state therefore makes the reaction possible. In this LMCT
transition, an electron is transferred from the hydrazine benzohydroxamate ligand to
the vanadium(V) center, to form an oxidized ligand and a vanadium(IV) state. After
an electronic excitation to a higher-lying excited state, intramolecular relaxation,
internal conversion, and intersystem crossing may occur to access the LMCT-type
T 1 state, in which homolytic C–C bond cleavage is a facile process. The low barrier obtained here is attributed to the fact that one of the two unpaired electrons,
which is produced upon homolytic C–C bond cleavage, is efficiently absorbed into
the low-lying orbital of the oxidized ligand (Fig. 7). Thus, as a result of the C–C
bond cleavage, aryl formate (one of the major products) and a radical species are left
behind. The latter species will rapidly combine with an O 2 molecule that is present in
the system under aerobic conditions. Subsequently, the alkyl peroxy radical binds to
the vanadium center, and the formal oxidation state of vanadium reverts back to +5.
In the second step, aryl formate is formed (Fig. 8). After all, the oxidation state of the
vanadium center is restored to +5, so that the vanadium complex can play a catalytic
role by effecting C–C cleavage in many catalytic rounds. The attachment of electronwithdrawing groups to the hydrazine benzohydroxamate led to the improvement of
the catalytic activity; this can be attributed to the stabilization of the ligand orbital,
which enhances the process illustrated in Fig. 7 [56].
In the photocatalytic reaction under anaerobic conditions, only a small amount
(3%) of the aryl aldehyde product was obtained in experiment. The formation of
the aldehyde product can be understood from the fact that the initial LMCT-driven
homolytic C–C bond cleavage process to form the aldehyde (the process from RC
to Int1 in Fig. 6) is possible even in the absence of O 2 . As for the low yield of
the aldehyde product, it indicates that the vanadium complex is consumed rapidly,
instead of acting as a catalyst. According to our DFT analysis (Fig. 9), the recovery of
the vanadium(V) state under anaerobic conditions is possible after a second LMCT,
but the subsequent barrier is higher than that for the initial C–C bond cleavage
process. Thus, the LMCT-driven C–C bond cleavage is favored over the LMCT-driven
vanadium(V) recovery, which results in the rapid depletion of the catalytically active
vanadium(V) species via preferential C–C bond cleavage. The pathways obtained for
the reaction under anaerobic conditions theoretically lead to aryl formate. However,