10.1002/asia.201700658
Chemistry - An Asian Journal
COMMUNICATION
Figure 1 ESR signal of DMPO-OH after illumination with TiO2 aqueous
solution under different conditions. a) different concentrations of sodium
benzoate (BA, 1 mM to 1M) and 1M of NaClO4 and Na2SO4; b) different
concentrations of toluene. The concentration of DMPO in all cases was 0.1M.
easily adsorbed on the surface and can complete to react with
hvb+/OHad, the signal of DMPO-OH adducts should diminish with
the increase of substrates in the solution, and might disappear
as the concentration is high enough. For the third case, no
matter how much weakly polar substrates was added, the signal
would not be obviously influenced because the capturing of
Next, why oxidations of polar substrates by direct hvb+/OHad
were difficult to become r.d.s, while the oxidations of weakly
either OHdiff radicals or hvb+/OHad by DMPO should be easier
than that of OHdiff radicals by general weakly polar benzenes.
polar substrates by the indirect OHdiff -initiated oxidation can?
.
-
As expected, the strongest DMPO-OH radical signal was
observed when the TiO2-H2O-DMPO-O2 system without any
substrates was excited by 355 nm laser as shown in Fig. 1a.
With the addition of benzoic acid and the increase of its
concentration, the signal was strongly inhibited and when the
concentration of benzoic acid reached 1M, almost no DMPO-OH
radical signal was observed. This inhibition of DMPO-OH
formation may result from either that the adsorbed benzoic acid
consume most hvb+/OHad prior to DMPO, which forbids the
formation of OHdiff radicals or that little OHdiff radicals diffuse into
solution to react with DMPO as captured by benzoic acid. In
order to exclude the interference caused by the adsorption
capacity of benzoic acids, we added 1M NaClO4 and Na2SO4 to
the TiO2-H2O-DMPO-O2 system to make the surface of TiO2
Even for the same weakly polar substrate, why BQ as ecb
eliminator OHdiff -initiated oxidation can be r.d.s., while dioxygen
O2 as ecb eliminator cannot? For the latter case, it was
.
-
reminiscent of our previous work that the aqueous TiO2
photocatalytic hydroxylation of weakly polar substrates toluene
in water will access to Wagner-Meerwein rearrangement of
carbocation to exhibit NIH shifts under anaerobic conditions[11]
.
Conversely, the NIH shift was significantly diminished and even
completely disappeared under aerobic condition. Beyond that,
the formation and rearrangement of cyclohexadienyl carbocation
(Scheme 1) should involve a hybridization change of aryl carbon
from sp2 to sp3, which tends to exhibit an obvious inverse KIE if
it is the r.d.s. This corresponding relationship between 1,2-NIH
shift and inverse H/D KIE was indeed proved to be tenable in
hydroxylation of weakly polar aromatics. However, for the polar
aromatics, initial substrates and theirs intermediates can hardly
be desorbed from the surface and are oxidized by hvb+/OH until
thoroughly converted to phenols. As a result, no Wagner-
Meerwein rearrangement was supposed to exist during the
hydroxylation. Hence, no obvious NIH shift should be observed
as inverse H/D KIE never occurred. To support this assumption,
we designed and specially prepared partly deuterated benzoic
acid, 3,5-D2-benzoic acid, to observe its 1,2 NIH shift during the
aqueous TiO2 photocatalytic hydroxylation. If the hydroxylation
go through the 1,2 NIH shift pathway shown in Scheme 3, the
deuterium atom should shift to the adjacent carbon and maintain
in the formed m-hydroxyl benzoic acid, representing
intermediate III of Wagner-Meerwein rearrangement must be
formed during the hydroxylation. Indeed, a very low NIH shift
values for hydroxylation of 3,5-D2-benzoic acid were observed in
both aerobic condition and anaerobic condition relative to that of
toluene hydroxylation (Table 2). It indicates that the aqueous
TiO2 photocatalytic hydroxylation of polar aromatics involves no
Wagner-Meerwein rearrangement but a direct breakage of the
C-H/D bond in the ring, which only generates a normal H/D KIE.
-
2-
covered with ClO4 or SO4
,
[10]and still a same strong DMPO-OH
radical signal appeared. This clearly indicated that the
adsorption of anions did not influence neither the generation of
+
OH radicals by the reaction of hvb and H2O nor the subsequent
capture of OHdiff radicals by DMPO. Conversely, when toluene
was added to the TiO2-H2O-DMPO-O2 system, toluene reduced
the signal of DMPO-OH radicals very slightly as shown in Fig.
1b and the signal was rarely changed when the toluene
concentration increased significantly. This illustrated that the
weakly polar toluene, which significantly differs from the polar
benzoic acid, never directly reacts with hvb+/OHad on the surface
and seldom inhibited DMPO-OH formation in the bulk solution.
Moreover, when anisole is used as a substrate (notice that its
H/D KIE was 0.98) and added to the TiO2-H2O-DMPO-O2
system, the DMPO-OH signal obviously decreased and the
signal continued to decrease when the concentration of anisole
reached 1M (see Fig.S2). The inhibition degree of DMPO-OH
signal among three substrates was highly consistent with the
polarity order benzoic acid > anisole > toluene, which was also
in accordance with H/D KIEs of their hydroxylation. That is, the
hydroxylation of polar benzoic acid is initiated directly by
hvb+/OHad on the surface and the hydroxylation of weakly polar
toluene was initiated by diffusive OH radicals in the bulk solution,
while the hydroxylation of anisole might go through both
mechanisms. Therefore, they should have, of course, different
reaction pathways and r.d.s in the hydroxylation process.
Table 2. NIH shift during the hydroxylation of aromatics in TiO2 photocatalysis
Entry
substrates
reagent
NIH shift values[a]
1
2
3
4
3,5-D2-benzoic acid
hv/TiO2/BQ
hv/TiO2/O2
hv/TiO2/BQ
hv/TiO2/O2
10.2
4.4
4-D-toluene
27.7
9.0
[a] The calculation of NIH shift values for each substrate see SI. (±0.5%)
This article is protected by copyright. All rights reserved.
Gong, Yuanzheng
