Catalytic wet air oxidation of carboxylic acids on TiO2-supported ruthenium catalysts

Article abstract of DOI:10.1006/jcat.1998.2352

The total oxidation of aqueous solutions of carboxylic acids by air was studied in a slurry reactor over the temperature range 180-200°C and oxygen partial pressure of 0.3-1.8 MPa in the presence of a 2.8%Ru/TiO₂ catalyst. The influence of various parameters is presented: the catalytic wet air oxidation of succinic acid is 0 order with respect to succinic acid; the order with respect to oxygen pressure is 0.4, and the activation energy is ca. 125 kJ/mol. It was found that acetic acid, which is one of the intermediates, and CO₂ have no retarding effect on the total organic carbon abatement rate of succinic acid. Substitution of one hydrogen atom of the methyl group in acetic acid by Cl, OH, or NH₂ gives an increase of the oxidation rate. However, it was proposed that the low activity of acetic acid oxidation is due not only to the difficulty to oxidize the methyl group, but also to the low adsorption coefficient of acetic acid on ruthenium surface. Inorganic salts, such as sodium chloride, only slightly decrease the oxidation rate of acetic acid. The absence of metal ions (Ru, Ti) in the effluents after reaction and the absence of particle sintering indicate also a high stability of the catalyst under the conditions employed. The catalyst can be recycled without loss of activity after the second run. The activity becomes stable after the attainment of a steady-state coverage of the Ru particles by oxygen. The study of the effect of reduction-oxidation treatments of the catalyst showed that the activity depends on the oxidation state of the surface.

Full text of DOI:10.1006/jcat.1998.2352

Journal of Catalysis 182, 129–135 (1999)
Article ID jcat.1998.2352, available online at http://www.idealibrary.com on
Catalytic Wet Air Oxidation of Carboxylic Acids
on TiO2-Supported Ruthenium Catalysts
Jean-Christophe Be´ziat, Miche`le Besson, ,1 Pierre Gallezot, and Sylvain Dure´cu†
Institut de Recherches sur la Catalyse-CNRS, 69626 Villeurbanne, France; and †De´partement Recherche, TREDI, 54505 Vandoeuvre-les-Nancy, France
Received June 16, 1998; revised October 25, 1998; accepted November 9, 1998
effluents. Alternatively, industrial aqueous effluents are
generally too diluted to be incinerated economically.
The emergent technology of wet air oxidation (WAO),
which consists of mineralizing the organic carbon, by
oxidation into CO2, is very promising for the treatment and
recycling of industrial waste waters provided investments
and operating cost are reduced significantly. This can be
achieved by employing catalysts designed to accelerate the
rate of oxidation into CO2, thus improving the abatement
of total carbon and enabling operation under temperature
and pressure conditions milder than those required in the
absence of catalysts. Catalytic WAO can be conducted with
catalysts dissolved in the reaction medium (1–7). However,
soluble catalystssuch astransition metalcations(essentially
copper) have to be recovered from the treated water by
additional separation processes. In contrast, heterogeneous
catalysts are easily separated from the liquid phase either
in batch or continuous processes, but they have to be stable
for a long period of time in the acidic, oxidizing, and often
chelating reaction medium; otherwise recovery processes
would also be required. Some leaching of solid catalysts
such as copper oxides on ZnO–Al2O3 spinel supports
(8–10) and manganese–cerium (11) or cobalt–bismuth
composite oxides (10, 12) could not be avoided. In previous
studies, the total oxidation into carbon dioxide of acetic
acid (13) and C6–C4 diacids (14) was achieved at 180–200 C
with air on ruthenium catalysts supported on active carbon
and graphite. No leaching of ruthenium was observed but
the carbon support was partially oxidized, which precludes
any prolonged use of carbon-supported catalysts, at least
under the operating conditions required to oxidize the
most refractory molecules such as acetic acid, i.e., at tem-
peratures higher than 170 C. However, most of the oxides
and salts employed as inorganic supports are liable to leach
away under hydrothermal conditions in the acidic, oxidiz-
ing, and chelating medium. Examination of potential-pH
diagrams giving the stability domains of oxides indicates
that TiO2 and ZrO2 are the most stable of readily available
supporting materials (15). Titanium and zirconium oxide
supports were actually employed in Nippon Shokubai (16)
The total oxidation of aqueous solutions of carboxylic acids by
air was studied in a slurry reactor over the temperature range 180–
200 C and oxygen partial pressure of 0.3–1.8 MPa in the presence
of a 2.8%Ru/TiO2 catalyst. The influence of various parameters is
presented: the catalytic wet air oxidation of succinic acid is 0 or-
der with respect to succinic acid; the order with respect to oxygen
pressure is 0.4, and the activation energy is ca. 125 kJ/mol. It was
found that acetic acid, which is one of the intermediates, and CO2
have no retarding effect on the total organic carbon abatement rate
of succinic acid. Substitution of one hydrogen atom of the methyl
group in acetic acid by Cl, OH, or NH2 gives an increase of the oxi-
dation rate. However, it was proposed that the low activity of acetic
acid oxidation is due not only to the difficulty to oxidize the methyl
group, but also to the low adsorption coefficient of acetic acid on
ruthenium surface. Inorganic salts, such as sodium chloride, only
slightly decrease the oxidation rate of acetic acid.
The absence of metal ions (Ru, Ti) in the effluents after reaction
and the absence of particle sintering indicate also a high stability
of the catalyst under the conditions employed. The catalyst can be
recycled without loss of activity after the second run. The activity
becomes stable after the attainment of a steady-state coverage of
the Ru particles by oxygen. The study of the effect of reduction–
oxidation treatments of the catalyst showed that the activity de-
c
pends on the oxidation state of the surface.
1999 Academic Press
INTRODUCTION
Interest in new processes for waste water treatment is
presently stirred by recent and forthcoming environmental
regulations on the disposal of industrial effluents. The main
objective for the future is to treat industrial pollutants at
their source of emission, particularly chemical and paper
pulp plants, and to recycle the treated water in industrial
plants, so as to achieve closed loop operations preventing
the spilling of industrial water in the hydrosphere. Biolog-
ical water treatments are not well adapted for this purpose
and are also limited by the toxicity or concentration of
1 To whom correspondence should be addressed. E-mail: mbesson@
catalyse.univ-lyon1.fr. Fax: 33 (0)4 72 44 53 99.
129
0021-9517/99 $30.00
c
Copyright
1999 by Academic Press
All rights of reproduction in any form reserved.

´
BEZIAT ET AL.
130
and Osaka Gas (17) WAO processes. Ceria, which is also (Sarasep Car-H) using dilute H2SO4 solutions as eluent
a good candidate for stable supporting materials, was used (0.01 N, 0.5 ml min 1). The overall abatement of organic
by Imamura et al. (18) and Duprez et al. (19, 20).
effluents was monitored with a TOC analyzer equipped
In the present work, WAO of aqueous solutions of suc- with an IR detector (Shimadzu 5050A). The TOC was de-
cinic and other carboxylic acids was carried out batchwise termined by subtracting the inorganic carbon (CO2 evolved
in a continuously stirred tank reactor pressurized with air after treatment of the sample in concentrated phosphoric
in the presence of a commercial TiO2-supported ruthenium acid) from the total carbon (CO2 evolved after catalytic
catalyst in powder form.
combustion on platinum at 700 C). The initial reaction
rates were calculated from the curves, giving the concentra-
tion of succinic acid or the TOC as a function of time at low
EXPERIMENTAL
1
conversion; they were expressed either in molacid
h
1 molRu
1
(r0(acid)) or in molC h 1 molRu (r0(TOC)). The percentage of
Materials and Characterization
TOC abatement was defined as 100 (TOC0 TOC)/TOC0.
Catalytic studies were carried out with a Ru/TiO2 cata-
lyst loaded with 2.8 wt% metal obtained from Engelhard
(Q500-069). The particle size of the catalyst was less than
0.040 mm. The structure of the catalyst was characterized
by X-ray diffraction (Philips PW1050 diffractometer) and
by high-resolution transmission electron microscopy (Jeol
RESULTS
Catalyst Characterization
Nitrogen adsorption measurementsshowed that the BET
JEM2010). The BET surface area was measured, after des- surface area of the 2.8% Ru/TiO2 catalyst was 50 m2 g
.
1
orption at 200 C under vacuum, by nitrogen adsorption The X-ray diffraction pattern exhibited only the reflec-
with an automatic sorbtometer. The metal dispersion was tions of the anatase form of TiO2, except for a small peak
measured by hydrogen chemisorption with a volumetric ap- corresponding to the main reflection of rutile. Hydrogen
paratus equipped with a Baratron MKS pressure gauge. chemisorption measurements, assuming a H/Ru = 1 stoi-
The catalyst was reduced under hydrogen and outgassed at chiometry, indicated that the ruthenium dispersion in the
300 C; then adsorption isotherm was measured at 45 C. A fresh catalyst was36% . Thisisin reasonable agreement with
second isotherm was measured after desorption at 45 C TEM micrographs showing that Ru particles are in the size
to determine the amount of reversibly adsorbed hydro- range 4–5 nm. These particles are not homogeneously dis-
gen. The treated solutions were analyzed by ICP-AES to tributed on the support but tend to agglomerate in specific
determine the concentration of ruthenium and titanium domains as large as 1 m2. The 2.8% Ru/TiO2 catalyst was
salts.
also studied after complete oxidation of the solution of suc-
cinic acid at 190 C. The TEM view given in Fig. 1 shows
the presence of 4-nm ruthenium particles, indicating that
the ruthenium dispersion was not modified during catalytic
reaction. The solutions analyzed after reaction contained
no ruthenium and titanium within the detection limit of
ICP-AES (0.5 and 0.1 ppm, respectively).
Oxidation Procedure
Oxidation of aqueous solutions of carboxylic acids
(0.5 wt% ) was performed in a 250-ml Hastelloy C22 au-
toclave, connected to an air reserve and equipped with a
magnetically driven turbine. The reactor was loaded with
the carboxylic acid solution and with the catalyst. After
flushing with argon, the temperature of the mixture was
raised to the reaction temperature under stirring. Air was
Preliminary Experiments
It was first verified that TiO2 was not active for the ox-
then admitted until the preset pressure was attained and idation of succinic acid in the absence of ruthenium. The
the reaction was started by adjusting the stirring speed at oxidation reaction was run under standard conditions us-
1800 rpm, which defined time zero. Standard operating con- ing 1 g of TiO2 (Degussa P25) for 4 h. The TOC did not
ditions were 0.15 liter of 5 g liter 1 solution, 1 g of 2.8 wt% change after 4 h, whereas a nearly complete TOC abate-
Ru/TiO2, 190 C, and 5 MPa total pressure (i.e., 0.72 MPa ment was observed during the same period of time in the
of O2).
presence of the 2.8% Ru/TiO2 catalyst (vide infra).
Samples were periodically withdrawn from the reactor
The reaction rates were measured under standard reac-
through stainless-steel 1/1600 tubing, and the pH, total tion conditions as a function of the mass of 2.8% Ru/TiO2
organic carbon (TOC), and product distribution were mea- catalyst. It was verified that the initial reaction rates ex-
sured. The pH measurement gave qualitative information pressed in molacid
h
1 or in molC h 1 were proportional to
on the progress of the reaction, since the pH increased the catalyst mass up to 1.5 g. It can be concluded that the
upon oxidation of the acids into CO2. Analysis of organic reaction rates were not limited by external mass transfer un-
products was performed by HPLC with UV and RID detec- der the standard reaction conditions involving 1 g of cata-
tors in series and separation with an ion-exchange column lyst. It was also verified that the activity did not depend

TiO2-SUPPORTED RUTHENIUM CATALYSTS
131
FIG. 1. TEM view of the 2.8% Ru/TiO2 catalyst after total oxidation of succinic acid.
upon the stirring speed under the standard conditions and centration of acetic acid started to decrease only after suc-
that the adsorption of succinic acid on the support was neg- cinic acid disappeared from the reaction medium. The ini-
tial rate of reaction was 16 molacid h 1 molRu1 or 43 molC h
molR1u, and the TOC abatement was 97.6% after a period of
4 h.
1
ligible.
Kinetics of Succinic Acid Oxidation
Figure 2 gives the composition of the reaction medium
The influence of the partial pressure of oxygen was stud-
as a function of time under standard reaction conditions. ied by measuring the initial reaction rate, r0, at different
Succinic acid was totally converted within 150 min; the total pressures between 3 and 10 MPa, i.e., between 0.32
only reaction products detected were CO2, acetic acid, and and 1.72 MPa of oxygen, taking into account the pres-
acrylic acid, which all appear as initial products. Other ox- sure of nitrogen and the partial pressure of water vapor
idation products may well be formed transiently but are at 190 C. The plot of ln r0(acid) as a function of ln Pox (Fig. 3)
not detected because they are rapidly oxidized at 190 C. gives a straight line from which a 0.4 reaction order can be
Acrylic acid was rapidly converted into CO2 but the con- determined.

´
BEZIAT ET AL.
132
FIG. 4. Arrhenius plot of succinic acid oxidation at 180, 190, and
200 C. (᭺) r0(acid); (᭹) r0(TOC)
.
FIG. 2. Oxidation of succinic acid with air; composition of reaction
effective since the final TOC abatement after 6 h was more
than 99% .
medium and TOC as a function of time under standard reactions con-
1
ditions (1 g of 2.8% Ru/TiO2, 150 ml of 5 g liter aqueous solution of
succinic acid, autoclave pressurized with air at 5 MPa, T = 190 C). TOC
(᭺); succinic acid (᭹); acetic acid ( ); acrylic acid (+).
Influence of Reaction Products
In the course ofsuccinicacid oxidation, the main products
The influence of temperature was studied by running the present in the reaction medium were CO2 and CH3COOH.
reaction at 180, 190, and 200 C. From the Arrhenius plot In order to verify that the rate of succinic acid oxidation
of ln r0(acid) and ln r0(COT) as a function of 1/T (Fig. 4), acti- was not affected by the increasing concentration of these
vation energies of 124 and 129 kJ mol 1, respectively, were
measured.
molecules, acetic acid was initially added in the reaction
1
medium with an initial concentration of 14 mmol liter
.
Another experiment was carried out in the presence of
182 mmol of CO2 added in the form of 8 g of solid CO2;
the total pressure was adjusted to keep the same partial
pressure of oxygen as in other experiments. Figure 5 shows
that the conversion of succinic acid was not affected by the
presence of additional amounts of acetic acid or of carbon
dioxide.
The linearity of the curve giving conversion as a function
of time (Fig. 5) suggests that the reaction order is zero with
respect to succinic acid up to nearly complete conversion;
this indicates a strong adsorption of the molecule on the
surface.
Influence of Sodium Chloride
Since most industrial effluents contain inorganic salts,
particularly chlorides, succinic oxidation was conducted
in the presence of 6.7 g liter of NaCl. The data given in
Table 1 indicate that both activity and TOC abatement
were decreased, the oxidation of acetic acid being the most
affected. However, the oxidation process was still very
Influence of Catalyst Pretreatment
1
The catalytic data reported above were obtained on cata-
lyst 2.8% Ru/TiO2 as received from Engelhard and stocked
in the presence of air. Different reduction or oxidation pre-
treatments were carried out before reaction; passivation
FIG. 3. Influence of the partial pressure of oxygen on the oxidation
of succinic acid with air. Plot of ln r0 vs ln Pox where r0 is the initial reaction
FIG. 5. Influence of additional amount of acetic acid (᭜) and carbon
rate and Pox the partial pressure of oxygen at 190 C calculated from the dioxide ( ) in the reaction medium on succinic acid conversion, (᭺) ref-
total pressure PT (Pox = (PT 0.14)/5). erence experiment.

TiO2-SUPPORTED RUTHENIUM CATALYSTS
133
TABLE 1
Influence of Sodium Chloride on Activity and TOC Abatement
TABLE 3
Activity in TOC Abatement in Succinic Acid Oxidation
after Three Recycling Operations
r0(acid)
r0(TOC)
[Acetic acid]
1
1
[NaCl]
(molacide h
(molC
molRu
h
TOC abatement
at 4 h (% )
at 4 h
(mmol l
Activity
Activity
1
1
1
1
1
1
(g liter
)
molRu
)
)
)
(molSUC
h
)
(molC
molRu)
h
TOC abatement
after 4 h (% )
1
1
molRu
0
6.7
16
10
43
31
97.6
92.3
2.7
7.3
As-received catalyst
First recycle
Second recycle
Third recycle
16
12
12
12
43
33
33
33
97.5
90
90
90
treatment consisted of contacting the freshly reduced sam-
ples with a flowing mixture of 1% O2 in N2 for 6 h at room
temperature. Table 2 gives the corresponding activity data
for TOC abatement in succinic acid oxidation.
Wet Air Oxidation of Other Carboxylic Acids
The data clearly show that the initial activities are sig-
nificantly higher on well-reduced ruthenium particles. The
activity decreased as the surface was covered by oxygen:
the higher the oxygen coverage of the ruthenium surface,
the lower the reaction rate. The rates were particularly slow
as the catalyst was heated in air at 300 C, which probably
resulted in the formation of an overlayer of RuO2 on the
surface and subsurface.
Wet air oxidation of aqueous solutions containing
5 g liter of the following four molecules containing car-
1
boxylic acid functions was carried out under the stan-
dard conditions employed for succinic acid oxidation. The
TOC abatement as a function of time is given in Fig. 7.
1
1
The initial activities expressed in molC h molRu were in
the following series: acetic (17) < succinic (43) < glycine
(64) < chloracetic (151) < glycolic (230). Substitution of
one hydrogen atom of the methyl group in acetic acid by Cl,
hydroxyl, or amino group increased the rate of oxidation.
Catalyst Recycling
Starting from the as-received 2.8% Ru/TiO2 catalyst, four
experiments of succinic oxidation were conducted with the
same catalyst recycled after separation from the oxidized
solutions by filtration. The recovered catalyst was reused as
such for the WAO of fresh solutions. The initial activity and
TOC abatement after 4 h are given in Table 3. The activity
decreased after the first catalyst recycle, then was stable in
the three subsequent oxidation runs. A close examination of
the product distribution showed that the loss of activity was
mainly due to a decrease of the rate of acetic acid oxidation.
This appears clearly in Fig. 6, giving the concentration of
acetic acid in the reaction medium as a function of time.
DISCUSSION
This study shows that succinic acid and other carboxylic
acids, particularly acetic acid, can be completely miner-
alized into CO2 with high specific activity on Ru/TiO2
catalysts using air as an oxidizing agent. Previous studies in
our laboratory showed that similar results, and even better
activities, were obtained on carbon-supported catalysts
(14), but the slow oxidation of carbons limits their practical
application as catalyst support at temperatures higher than
150 C. In contrast, this study demonstrates that titanium
dioxide was stable in both acidic and oxidizing medium
since no titanium leaching was detected. As ruthenium is
TABLE 2
Influence of Catalyst Pretreatment on the Activity for TOC
Abatement in Succinic Acid Oxidation
Initial activities for
TOC abatement
1
Pretreatment of 2.8% Ru/TiO2
(molc h 1 molRu
)
Sample 1: catalyst as received stocked in air
Sample 2: catalyst reduceda under H2 at 300 C,
passivated, kept under Ar
43
75
Sample 3: catalyst reduceda under H2 at 500 C,
passivated, kept under Ar
73
Sample 4: sample 3 heated under air at 300 C
Sample 5: sample 4 reduceda under hydrogen at
300 C, passivated, kept under Ar
13
63
Sample 6: sample 5 kept in air three weeks
31
FIG. 6. Concentration of acetic acid formed by succinic acid oxidation
during four successive run with recycled catalysts. First run (᭺); second
run (᭹); third run ( ); fourth run (+).
a Heating rate 1 K min 1, 2 h at reduction temperature, cooling under
Ar.

´
BEZIAT ET AL.
134
ties were obtained when the competitive adsorption on the
metal surface of oxygen and of the organic substrate was
equilibrated. In the case of WAO on ruthenium at 190 C,
the mechanism of oxidation reaction is probably different
and may involve the formation of radicals. However, the
modeling of the oxidation kinetics of succinic acid in trickle-
bed reactor (21) suggests that adsorption of the organic
molecule on ruthenium surface, whatever its oxidized state,
is still a necessary and rate controlling step. Therefore, the
different activities for TOC abatement as a function of cata-
lyst pretreatment could be attributed to the different oxy-
gen coverage of the surface: lower oxygen coverage gives
higher WAO activities because of easier adsorption of the
molecule or because of easier electron transfer from the
metal to the adsorbed molecule. Note that the deactivation
which occurs during the first recycling operation (Table 3)
can be attributed to a partial oxidation of the surface; the
catalyst did not deactivate in further recycling because the
oxidation state did not change. This means that the oxygen
coverage should stabilize before a steady state was attained;
this point will be better highlighted by the continuous WAO
study in trickle-bed reactor (21).
Literature data on WAO and previous works in this labo-
ratory (13, 19, 20, 25–28) have shown that acetic acid is one
of the most difficult molecules to oxidize into CO2. This
was confirmed by the present study. Thus, the oxidation of
acetic acid started only after succinic acid was consumed
(Fig. 2), and the rate of acetic acid oxidation was much
smaller than the rates of C2 carboxylic acids derived by
substitution of one hydrogen atom of the methyl group by
Cl, NH2, and OH groups. The difficulty in oxidizing acetic
acid could be due to a low adsorption coefficient of the
molecule on the metal surface (effect 1), to a weak reactiv-
ity in the adsorbed state with respect to oxidation (effect 2),
or to the combination of both. Effect 1 is supported by two
experimental observations: acetic acid started to oxidize
only after succinic acid was no longer present in the reac-
tion medium (Fig. 2), and acetic acid added initially did not
modify the kinetics of succinic acid oxidation (Fig. 5). Thus,
it can be inferred that the adsorption coefficient of acetic
acid is much lower than that of succinic acid. This is further
supported by a modeling study of the oxidation kinetics in
trickle-bed reactor, from which a comparatively weak ad-
sorption coefficient of acetic acid was calculated (21). Effect
2, which accounts for the so-called refractory character of
acetic acid, is usually attributed to the difficulty to activate
the C–H bond of the methyl group in -position from the
carboxylic function. This effect is supported by the fact that
by substituting one hydrogen atom with a NH2 or OH group
or Cl, the molecule is then oxidized very rapidly (the rate of
TOC abatement under standard conditions was 17, 64, 151,
and 230 molc h 1 molR1u, for acetic, glycine chloracetic, and
glycolic acid, respectively). With the presently available ex-
perimental data, none of these effects could be definitively
FIG. 7. TOC abatement as a function of time: acetic acid (+); succinic
acid (4); chloroacetic acid (᭡); glycolic acid ( ); glycine (᭹). (1 g of
2.8% Ru/TiO2, 150 ml of 5 g liter 1 aqueous solution, autoclave pressurized
with air at 5 MPa, T = 190 C.)
also stable in the reaction medium, as shown in this and
previous studies (13, 14), Ru/TiO2 catalysts are ideally
suited for heterogeneous WAO processes. Indeed, the
data presented in Fig. 7 indicate that they can be recycled
without loss of activity after the second run so that they
are potentially well adapted to continuous WAO process.
Accordingly, subsequent studies conducted in trickle-bed
reactor on the same catalyst in pellet form have shown
that its activity in succinic oxidation was stable during
several weeks on stream (21). The fact that the presence of
sodium chloride in solution moderately affects the rate of
TOC abatement (Table 1) is also encouraging. Discussion
on the potential interest of Ru/TiO2 catalysts in industrial
application will await forthcoming results on the WAO of
real effluents. The present discussion will focus on more
fundamental aspects such as the effect of oxygen coverage
on the ruthenium surface and why carboxylic acids exibit
different reactivities toward oxidation.
The activity of Ru/TiO2 catalysts was shown to de-
pend upon reduction or oxidation pretreatments (Table 2).
Ruthenium was most active for succinic acid oxidation
when it was freshly reduced and kept under argon before
use; the activity became significantly lower when it was
stocked in air, and ruthenium heated in air at 300 C ex-
hibited a weak activity. These simple facts point to a detri-
mental effect on the WAO of increasingly higher oxygen
coverages of the surface or of the incorporation of oxygen
in subsurface. Assuming that the organic molecules need to
be adsorbed on the surface of the metal to be oxidized, the
activity pattern could be easily rationalized: higher oxygen
coverage hampers the adsorption of the molecule. It was
shown that this type of interpretation holds for the selec-
tive oxidation of oxygenates at low temperatures on plat-
inum and palladium catalysts which occurs via a mechanism
of oxidative dehydrogenation. The total or partial poison-
ing of the catalysts was interpreted by an overoxidation of
the metal surface (22–24), whereas high oxidation activi-

TiO2-SUPPORTED RUTHENIUM CATALYSTS
135
ruled out, and a negative synergy due to the combination
REFERENCES
of the two effects is quite possible.
1. Mantzavinos, D., Hellenbrand, R., Livingston, A. G., and Metcalfe,
I. S., Appl. Catal. B 7, 379 (1996).
2. Chang, C. J., and Lin, J. C., J. Chem. Technol. Biotechnol. 57, 355
(1993).
3. Lin, S. H., and Ho, S. J., Appl. Catal. B 9, 133 (1996).
4. Phull, K. K., and Hao, O. J., Ind. Eng. Chem. Res. 32, 1772 (1993).
5. Hao, O. J., Phull, K. K., Chen, J. M., Davis, A. P., and Maloney, S. W.,
J. Hazardous Mater. 34, 51 (1993).
6. Mishra, V. S., Joshi, J. B., and Mahajani, V. V., Ind. Chem. Eng. 35, 211
(1993).
CONCLUSION
This investigation highlights the following points:
(i) Complete mineralization into CO2 of aqueous solu-
tions of carboxylic acid (5 g liter 1) was obtained by oxida-
tion with air at 190 C on a commercially available Ru/TiO2
catalyst. This was supported by the fact that the treated
water did not contain any organic products (100% TOC
abatement) and that the catalyst can be recycled without
loss of activity after the first recycling operation, which rules
out formation of polymeric deposit as observed in the WAO
of phenol on CuO–ZnO–Al2O3 or Mn/Ce oxide catalysts
(9, 29).
7. Shende, R. V., and Mahajani, V. V., Ind. Eng. Chem. Res. 33, 3125
(1994).
8. Pintar, A., and Levec, J., Chem. Eng. Sci. 47, 2395 (1992).
9. Levec, J., and Pintar, A., Catal. Today 24, 51 (1995).
10. Hellenbrand, R., Mantzanivos, R., Livingston, A. G., and Metcalfe,
I. S., in “Environmental Catalysis—For a Better World and Life” (G.
Centi, C. Cristiani, P. Forzatti, and S. Perathoner, Eds.), p. 487. Societa`
Chimica Italiana, Rome, Italy, 1995.
(ii) No leaching of ruthenium or titanium was detected.
The stability of heterogeneous catalysts in the acidic and
oxidizing reaction medium is a key issue in WAO processes.
(iii) Different reduction and oxidation treatments of the
Ru/TiO2 catalyst prior to WAO of succinic acid resulted
in different initial activities: high oxygen coverages of the
ruthenium surface result in low activities. This can be ratio-
nalized by assuming that the adsorption coefficient of the
organic molecule is lower on surface highly covered by oxy-
gen or that the interaction of the adsorbed molecule with
the metal protected by an oxygen layer is weaker. A steady
state of activity is attained only after the metal surface is
equilibrated and reaches a constant oxygen coverage. This
accounts for the fact that the catalyst deactivated during a
first reaction run, but its activity was stable in subsequent
recycling operations.
(iv) This investigation, supported by the modeling study
reported elsewhere (21), brings a new interpretation for the
different rates of oxidation of carboxylic acids, and particu-
larly for the difficulty in oxidizing acetic acid in WAO pro-
cesses. The first step of the reaction on heterogeneous cata-
lysts involves the adsorption of the molecule on the metal
surface. Therefore the respective adsorption coefficients of
carboxylic acids have to be taken into account to interpret
the different rates of TOC abatement. However, this does
not rule out the influence of other factors such as those
associated with the electronic structure of the molecules.
11. Imamura, S., Nishimura, H., and Ishida, S., Sekiyu Gakkaishi 30, 199
(1987).
12. Imamura, S., Matsushige, H., and Kawabata, N., J. Catal. 78, 217(1982).
13. Gallezot, P., Chaumet, S., Perrard, A., and Isnard, P., J. Catal. 168, 104
(1997).
14. Be´ziat, J. C., Besson, M., Gallezot, P., Juif, S., and Durecu, S.,
in “3rd World Congress on Oxidation Catalysis” (R. K. Grasselli,
S. T. Oyama, A. M. Gaffney, and J. E. Lyons, Eds.), p. 615, Studies in
Surface Science and Catalysts, Vol. 110. Elsevier, Amsterdam, 1997.
15. Pourbaix, M., “Atlas d’e´quilibres e´lectrochimiques,” Gauthier–Villars,
Paris, 1963.
16. Ishii, T., Mitsui, K., Sano, K., and Inoue, A., EP 0431 932A1 to Nippon
Shokubai (1990).
17. Harada, Y., and Yamasaki, K., Desalination 98, 27 (1994).
18. Imamura, S., Fukuda, I., and Ishida, S., Ind. Eng. Chem. Res. 27, 718
(1988).
19. Duprez, D., Delanoe¨, F., Barbier, J., Jr., Isnard, P., and Blanchard, G.,
Catal. Today 29, 317 (1996).
20. Duprez, D., Delanoe¨, F., Barbier, J., Jr., Isnard, P., and Blanchard, G., in
“Environmental Catalysis—For a Better World and Life” (G. Centi, C.
Cristiani, P. Forzatti, and S. Perathoner, Eds.), p. 495. Societa` Chimica
Italiana, Rome, Italy 1995.
21. Be´ziat, J. C., Besson, M., Gallezot, P., and Durecu, S., submitted for
publication.
22. Mallat, T., and Baiker, A., Catal. Today 19, 247 (1994).
23. Imamura, I., Nishimura, H., and Ishida, S., Sekiyu Gakkaishi 30, 199
(1987).
24. Besson, M., Lahmer, F., Gallezot, P., Fuertes, P., and Fle`che, G.,
J. Catal. 152, 116 (1995).
25. Gallezot, P., Besson, M., and Fache, F., in “Chem. Ind. 62: Catalysis
of Organic Reactions” (M. G. Scaros and M. L. Prunier, Eds.), p. 331.
Dekker, New York, 1995.
26. Mishra, V. S., Mahajani, V. V., and Joshi, J. B., Ind. Eng. Chem. Res.
34, 2 (1995).
ACKNOWLEDGMENTS
27. Krajnc, M., and Levec, J., Appl. Catal. B 13, 93 (1997).
We gratefully acknowledge Engelhard for providing the catalyst sam- 28. Mantzavinos, D., Hellenbrand, R., Livingston, A. G., and Metcalfe,
ple. We thank Marion Sergent for her assistance with the TOC and HPLC
measurements.
I. S., Wat. Sci. Technol. 35, 119 (1997).
29. Hamoudi, S., Larachi, F., and Sayari, A., J. Catal. 177, 247 (1998).