Synthesis of 2-hydroxybenzoic acid from CO and phenol in aqueous
2
heterogeneous photocatalytic systems
Antonino Sclafani, Leonardo Palmisano and Gianluca Farneti
Dipartimento di Ingegneria Chimica dei Processi e dei Materiali, Universit a` di Palermo, Viale delle Scienze, 90128, Palermo,
Italy
The photocatalytic reaction occurring in aqueous deaereated
suspensions saturated with CO and containing polycrystal-
line semiconductors, bare or loaded with Pt or Cu, affords
-hydroxybenzoic acid and catechol as the main products.
using anatase TiO
the formation of catechol TiO
and 20% rutile) was found to be less photoactive than anatase
TiO (Merck) which could be due to the higher oxidant
properties of TiO (Degussa P25) with respect to TiO (Merck).
The high oxidant properties of TiO (Degussa) when used as
photocatalyst for degradation reactions of organic molecules
carried out in aqueous media are well known. The runs carried
out using the other solids showed the formation of significantly
smaller amounts of the above compounds. As expected, no
2-hydroxybenzoic acid, only traces of catechol, were found
2
(Merck), while Pt(2 mass%)–WO
3
favoured
2
2
(Degussa P25, ca. 80% anatase
2
2
2
2
Many studies1–6 have reported the photocatalytic reduction of
CO in both gas–solid and liquid–solid heterogeneous systems
using mild experimental conditions and in the presence of H O.
The solids used as photocatalysts were polycrystalline semi-
conductors, and bare Cu-doped TiO (anatase) proved to be the
most suitable. Light organic compounds, such as for instance
HCO H, MeOH and CH , were obtained as the main products
2
2
2
2
6
2
4
when the insulator oxide SiO
photoactivity of bare WO should be noted in comparison with
the corresponding sample Pt(2 mass%)–WO in the presence of
which the biggest production of catechol was observed. The
presence of Pt on the surface of WO particles modifies the
2
was used. Moreover, the poor
and their relative ratios depended on the type of photocatalyst
and experimental conditions used.
3
3
However, very few studies7,8 can be found on the insertion of
CO
2
in organic aromatic substrates using photochemical
3
methods. The heterogeneous photocatalytic insertion of CO
into aromatic molecules, to the best of our knowledge, has not
yet been reported.
2
adsorption of the reagents, but in this case the major role
consists in enhancing the lifetime of the photoproduced
electron–hole pairs, due to its high work function. The
9
Here preliminary results are reported on the production of
-hydroxybenzoic acid (salicylic acid) by the photocatalytic
3
electrons migrate from WO to Pt, giving rise to reduction
2
reactions. The trend of the photoactivity of the samples is not
much different if the BET specific surface areas of the powders
reported in Table 1 are considered, i.e. if the specific
photoactivity of the samples is calculated.
reaction between CO and phenol in aqueous deaerated
2
suspensions containing a polycrystalline semiconductor, bare or
loaded with Pt or Cu. Catechol was produced together with
2
-hydroxybenzoic acid and its amount depended on the catalyst
used.
The experimental runs were performed in a Solarbox
manufactured by CO.FO.ME.GRA., Milan, equipped with a
500 W Xe lamp, simulating the solar spectrum. Pyrex
photoreactors (50 ml) containing 25 ml of aqueous phenol (8.2
The results suggest that the photoreactivity is strongly
influenced by both the surface physico-chemical and the
electronic properties of the catalysts. The adsorption phe-
2
nomena, indeed, are of primary importance for CO photo-
1
activation and they are related to several factors such as, for
instance, the presence of catalytic metals, the surface acid–base
properties, the zero charge potential and the surface hydroxy-
lation.
2
1
2
g l ) solutions at pH = 5.85 saturated with CO were used for
each run. The photon flux provided by the lamp was measured
in the locations of the photoreactors by means of the standard
potassium ferrioxalate actinometric method which gave 6.6 3
In Fig. 1 the concentrations of 2-hydroxybenzoic acid and
catechol vs. illumination time are reported when anatase TiO
2
2
7
21
1
0
Einstein s
with oscillations of only ca. 3%. The
(Merck) is used as photocatalyst.
It can be seen that the production of 2-hydroxybenzoic acid
approaches the maximum value asymptotically. This could be
2
1
photocatalyst (2 g l ) was suspended in the phenol solution and
magnetically stirred.
The solids used as photocatalysts were: anatase TiO
2
(
Merck), TiO
2
(P25 Degussa), Cu-doped TiO
2
(P25 Degussa),
Table 1 Average 2-hydroxybenzoic acid and catechol concentrations per
hour and BET specific surface areas (SA) of the photocatalysts used
Rutile TiO (Tioxide), WO
2
3
(Carlo Erba, RPE), Pt-doped WO
3
(
2
Carlo Erba, RPE), ZnO (BDH) and SiO (BDH). The runs
2
-Hydroxybenzoic
Catechol/
lasted 4 h and samples were withdrawn for HPLC analyses. The
analyses were carried out after filtration through 0.45 mm
cellulose acetate membrane (HA, Millipore) and subsequent
Surface area/ acid/mmol
mmol
2
21
21 21
21 21
Catalyst
m g
l
h
l
h
2
centrifugation. One run lasted 8 h (anatase TiO Merck) and it
Anatase TiO
2
was contemporaneously performed using five identical photo-
reactors: two of them were used for analyses carried out after 1
and 2 h of illumination, the other three for analyses carried out
after 4, 6 and 8 h.
(
Merck)
2
10.5
44
371.2
64.2
44.31
3.5
TiO (P25 Degussa)
Cu (0.1 mass%)–TiO
(P25 Degussa)
Rutile TiO
WO (Carlo Erba,
RPE)
Pt (2.0 mass%)–WO
2
50
27.9
13.6
22.7
8.5
2
(Tioxide) 20
Typically the filtered samples were injected into a chromato-
graph composed of a Varian model 9010 pump, a Varian 9050
UV–VIS detector and a 10 mm Econosil C18 (Alltech) 250 3
3
15.5
18.5
—
3
4
0
3 4
.6 mm i.d. column. Isocratic elution with methanol–H PO
(Carlo Erba, RPE)
16.4
4
117.7
9.4
—
1118.2
—
traces
.05 mass% aqueous solution was performed at 1 ml min21.
ZnO (BDH)
SiO
The products were monitored at l = 254 nm.
a
2
(BDH)
n.d.
The results are shown in Table 1. It can be seen that the most
significant formation of 2-hydroxybenzoic acid was obtained by
a
N.d. = not determined.
Chem. Commun., 1997
529
Sclafani, Antonino
