Comparison of cresol transformation on USHY and HZSM-5

Article abstract of DOI:10.1006/jcat.2000.2984

In the commercial synthesis of cresols, the most common processes include the alkylation of phenol with methanol. The cresol (methylphenol) transformation on zeolite USHY was studied and compared with HZSM-5 with different pore size, structure, and different acidity in a flow reactor at 380°C and 0.1 bar. The cresol turnover numbers for USHY were pretty similar, whereas on HZSM-5, the cresols reactivities followed the sequence p- ≥ m- > o-cresol. The cresols reacted over the zeolites mainly through two parallel reactions, isomerization and disproportionation, whose relative importance depends on the catalysts characteristics (pore structure and acidity), on the reacting cresol itself, and on the reaction conditions (temperature, total pressure, reactant partial pressure, etc). The USHY selectivity toward the monomolecular isomerization reaction was by far lower than that of HZSM-5, due to its size/shape selectivity. On USHY and HZSM-5 catalysts, the o-cresol showed higher disproportionation selectivity than the other two cresol isomers, this proportionation occurring via diphenylmethane intermediates. The cresol isomerization selectivity was limited by product desorption even on a large pore zeolite like USHY. The cresol isomer composition at high conversions on USHY was 44, 42, and 14% for o-, m-, and p-cresol, respectively. Xylenols underwent a rapid isomerization once formed.

Full text of DOI:10.1006/jcat.2000.2984

Journal of Catalysis 195, 279–286 (2000)
doi:10.1006/jcat.2000.2984, available online at http://www.idealibrary.com on
Comparison of Cresol Transformation on USHY and HZSM-5
F. E. Imbert, ,1 M. Guisnet,† and S. Gnep†
Laboratorio de Cine´tica y Cata´lisis, Departamento de Qu´ımica, Facultad de Ciencias, Universidad de Los Andes, La Hechicera, Me´rida 5101A
Venezuela; and †Laboratoire de catalyse en Chimie Organique, URA CRNS 350, Universite´ de Poitiers, France
Received February 16, 2000; revised June 26, 2000; accepted June 29, 2000
cial chemicals. The choice of zeolite catalysts is based on
their remarkable acid properties and especially to their
well-defined pore structure, with apertures and cages of
approximately the size of organic molecules. Cresol iso-
mers are used in the production of antiseptics, drugs, in-
secticides, dyes, phenolic resins, antioxidants, and polymers
(1). In the commercial synthesis of cresols the most com-
mon processes are alkaline hydrolysis of chlorotoluene, the
cleavage of cymene hydroperoxide, and the alkylation of
phenol with methanol. This alkylation, which is an elec-
trophilicsubstitution, isstronglyortho–para directing(2–6).
The meta isomer can be obtained by subsequent isomer-
ization (7). The isomerization of various methyl aromatics
(xylenes (7–12), halogenotoluenes (13), toluidines (14, 15),
etc.) has been widely studied due to its industrial impor-
tance as well as its scientific interest. Some authors have
reported that their data were consistent with the consecu-
tive reversible 1,2-methyl shift mechanism (7, 8, 16), while
others have proposed a triangular scheme with direct ortho/
para interconversion (16). The interpretation of experi-
mental data is generally complicated, for selectivity de-
pends on various parameters: acid site density and strength
and pore structure. Thus, for large pore zeolites the ortho/
para interconversion is more significant than for amor-
phous silica–alumina due to transalkylation reactions. With
medium pore size zeolites the selectivity is determined by
the product desorption from the pores. We have previously
shown on HZSM-5 (17) and on a series of HY (18) that
cresols are transformed through two main reactions: iso-
merization and disproportionation. For HZSM-5 the iso-
merization rates were always higher than the dispropor-
tionation rates. The initial conversion rates were in the
ratio rp :rm :ro = 20 :3 :1, and the initial I/D selectivi-
ties were (I/D)p : (I/D)m : (I/D)o = 100 : 30 : 8. The isomer
equilibrium composition obtained over HZSM-5 was 36%
o-cresol, 48% m-cresol, and 16% p-cresol. The m- and
o-cresol transformations were investigated at a given con-
tact time in the course of deactivation on the series of HY
(Si/Al = 4.5–55). o-Cresol and m-cresol reactivities were
similar on HY(4.5) but as the Si/Al ratio increased, o-cresol
reactivity increased relative to that of m-cresol; how-
ever, in all cases the catalyst deactivated less for m-cresol
The cresol (methylphenol) transformation on zeolite USHY was
investigated in a flow reactor at 380 C and atmospheric pressure.
The results are compared with those previously reported for
HZSM-5 (J. Catal. 172, 307 (1997)). On USHY, the catalytic stabil-
ity follows the sequence p- m- > o-cresol, the turnover numbers
(TONs) are similar for all three cresol isomers, but slightly higher
for p-cresol. Thus, the access to acid sites is not limited. While
on HZSM-5, the cresol reactivity follows the sequence p- > m- >
o-cresol, with the TON in the ratio p : m : o = 7 :4 :1. The HZSM-5
zeolite catalytic activity is more stable and its acid sites are more
active than those of USHY, for p- and m-, but not for o-cresol, which
means that o-cresol’s access to the active sites in HZSM-5 is limited.
The cresols are transformed through two main routes, a unimolecu-
lar isomerization via a 1,2-methyl shift and a bimolecular dispropor-
tionation or transalkylation reaction. On HZSM-5, isomerization
is the dominant pathway for all cresols, disproportionation being
limited by the space available near the acid sites and by the acid site
density; while on USHY, a large pore zeolite with a higher acid site
density, disproportionation is more significant. On both zeolites, o-
cresol shows the highest disproportionation selectivity, which could
be due to a rapid formation of diphenylmethane intermediates via
a benzylic carbocation of o-cresol. The p/o and m/o ratios obtained
for m- and p-cresol transformation respectively are higher with
HZSM-5 than with USHY, while the p/m ratios obtained for o-
cresol transformation were similar on both catalysts. It is concluded
that not only with HZSM-5 but also with USHY the selectivity
of isomerization is governed by product desorption/diffusion.
The cresol composition at high conversions on USHY is 44%
o-cresol, 42% m-cresol, and 14% p-cresol, which contrast with
the composition found on HZSM-5: 36% o-cresol, 48% m-cresol,
and 16% p-cresol. The xylenol distribution is governed by rapid
interconversion and differences in diffusivities.
Key Words: cresol; isomerization; disproportionation; USHY;
HZSM-5.
c 2000 Academic Press
INTRODUCTION
Zeolites have replaced corrosive and polluting acids as
catalysts in many refining and petrochemical processes and
should substitute for them in the synthesis of fine and spe-
1 To whom correspondence should be addressed. Fax: (58) (74) 401286.
E-mail: imbert@ciens.ula.ve.
279
0021-9517/00 $35.00
c
Copyright
2000 by Academic Press
All rights of reproduction in any form reserved.

280
IMBERT, GUISNET, AND GNEP
TABLE 1
Catalyst Physicochemical Characteristics
Si/Al
overall
Sat
AcT
AcF
Catalyst
Formula
(mmol of H+/g)
(meq H+/g)
(meq H+/g)
HZSM-5
USHY
27
2.5
H2.89Na0.54Al3.43Si92.57O192
0.5
2.96
0.608
1.085
0.141
0.729
H
34.1Na0.8Al34.9Si157.1O384
Note. Sat, theoretical acid sites per gram of catalyst; AcT, total acidity measured by TPD-NH3; and AcF,
strong acidity (amount of NH3 retained at T 350 C).
transformation. The disproportionation was predominant the presence of ultramicropores and mesopores besides
in the case of o-cresol. The object of the present work is to the normal micropores (12). The isotherms were analyzed
complement our two previous reports (17, 18) by studying with the Dubinin–Raduskievich (DR) equation, from
the transformation ofallthree cresolsover USHY, changing which VDR (VDR = Vultra + Vt) can be obtained. The t-plot
both the contact time and the time on stream, and compar- method for the micropore volume and the BJH method
ing its activity and selectivity with HZSM-5 zeolites. The for mesopores were also used. Table 2 gives the total pore
porosity of these catalysts was characterized by nitrogen volume Vtot (measured at P/P0 = 0.97), the micropore vol-
adsorption.
ume given by the t-plot method Vt, that of ultramicropores
(Vultra = VDR Vt), that of mesopores (Vmeso = Vtot DR ),
and specific surface area. The USHY sample shows higher
tot, Vt, and specific surface area, whereas Vultra and Vmeso
V
EXPERIMENTAL
V
are slightly lower. Therefore, this sample offers a higher
accesibility to reactants molecules than the HZSM-5 sam-
ple. On the other hand, the space available for bimolecular
reactions is restricted in the latter, which makes it more
resistant to coke formation, while the former allows coke
formation more easily and, consequently, deactivates more
rapidly.
Two zeolites were tested: a HZSM-5 (PQ zeolite, CBV
5020, Si/Al = 27) and a USHY (Union Carbide, LZY-82,
with a framework Si/Al = 4.5 and overall Si/Al = 2.5). They
were used as powders (200–270 mesh). The physicochemi-
cal characteristics of these catalysts are given in Tables 1
and 2. The nitrogen adsorption–desorption experiments
were carried out at 77 K, using a gas adsorption system
ASAP 2010 (Micromeritics). Prior to the reaction, the cata-
lysts were activated at 500 C overnight under air. The
reagents (Aldrich) were used without further purification.
Reactions were carried out at 380 C and at atmospheric
pressure in fixed bed reactors. A detailed description of the
experimental and analytical procedures have been reported
elsewhere (17, 18).
Activity and Catalytic Stability
Whatever the reactant cresol, both catalysts show a rapid
deactivation (Fig. 1). However, as is generally observed
for high-temperature gas phase reactions (19), HZSM-5 is
found to be more catalytically stable than USHY.
The initial activity and catalytic stability were estimated
by fitting the deactivation data by an exponential equation,
RESULTS AND DISCUSSION
Conversion = C0 exp( kd t),
Adsorption Properties
where C0 represents the initial conversion (at time-on-
The adsorption–desorption nitrogen isotherm shapes
indicate the prevailing microporosity of these catalysts;
however, the isotherms show a hysteresis loop closing at
P/P0 = 0.4, and this irreversibility has been attributed to
TABLE 2
Catalyst Crystallinity and Textural Characteristics
Vtot
,
Vt,
Vultra
,
Vmeso
,
Crist(XRD),
%
Catalyst
cm3/g
cm3/g cm3/g
cm3/g
S, m2/g
HZSM-5
USHY
0.223
0.350
0.118
0.254
0.039
0.034
0.0669
0.0620
82
365
649
FIG. 1. Cresol conversion at 380 C, Pcresol = 0.1 bar, and W/F =
5.15 (gh)/mol as a function of time on stream over (a) USHY and
(b) HZSM-5.
100

CRESOL TRANSFORMATION ON USHY AND HZSM-5
281
TABLE 3
Catalyst Initial Activity and Deactivation Constants for Cresol
Conversion on USHY and HZSM-5 at W/F = 5.15 (gh)/mol
USHY
TON
HZSM-5
TON
1
1
Reactive
C0, %
kd, min
C0, %
kd, min
p-Cresol
m-Cresol
o-Cresol
87.2
73.8
76.8
57.3
48.5
50.5
0.030
0.035
0.064
59.7
35.4
8.04
231.5
137.3
31.2
0.0068
0.024
0.023
FIG. 3. I/D selectivity as a function of conversion for (a) USHY and
(b) HZSM-5. Legend as in Fig. 1.
limited and that their intrinsic molecular property (polarity,
basicity, etc.) differences do not seems to affect cresol reac-
tivities on USHY, whereas on HZSM-5, the TON are quite
different, indicating once again the pronounced diffusion
limitations.
stream, t = 0) and kd a constant that reflects the catalyst
deactivation. As can be seen in Table 3, the USHY zeolite is
more active, though less catalytically stable than HZSM-5.
On USHY the catalytic stability varies for the different
cresols, following the sequence p- m- > o-cresol, while on
HZSM-5 the catalytic stabilities are p- m- o-cresol.
The conversions (t = 10 min) of the cresols are plotted
in Fig. 2, as a function of contact time taken as W/F, the
ratio between the catalyst weight and the reactant molar
flow rate. On the HZSM-5 catalyst, the reactivity of cresols
clearly follows the sequence p- > m- > o-cresol, while on
USHY the sequence is p- > m- o-cresol, asshown in Fig. 2.
This might be associated with the p-isomer highest diffu-
sivity. USHY is more active than HZSM-5, for all cresols,
which can be related to its acid site higher density and ac-
cessibility (Tables 1 and 2) as well as to the extraframe-
work aluminium species that may interact with framework
protonic sites, generating a very strong acidity (12, 20–23).
However, given the turnover number (TON), as the num-
ber of molecules converted per theoretical acid site (from
Table 1), we find, as expected, that the acid sites on ZSM-
5 are more active than on USHY, for p- and m-cresol,
but not for o-cresol transformation (see Table 3). There-
fore, there must be a strong hindrance to o-cresol enter-
ing the pore of HZSM-5, or if it does enter into the pore,
there is not enough space for the transition state needed
to be formed for its transformation. On the other hand, on
USHY, whatever the reactant cresol isomer, the TONs are
similar, which means that their access to the acid sites is not
Selectivity
The cresols react over the zeolites mainly through two
parallel reactions, isomerization (I) and disproportiona-
tion (D) (17, 18); however, the dealkylation reaction is also
possible, particularly on USHY. The isomerization to dis-
proportionation (I/D) selectivity shows a great difference
between USHY (Fig. 3a) and HZSM-5 (Fig. 3b) (17). In
all cases HZSM-5 is more selective in isomerization than
USHY, as is observed for xylenes transformation (9). This
was quite expected as the formation of the bulky diphenyl-
methane intermediates of disproportionation is strongly
inhibited in the narrow channel intersections of HZSM-5
˚
(
8.5 A in diameter) whereas it is not the case in the su-
˚
percages of USHY (13 A in diameter). However, there is
an important difference between xylene and cresoltransfor-
mations. With both catalysts, the I/D ratio is initially much
lower from o-cresol than from the other two isomers, while
it wasthe reverse in the case ofxylenes(11), where the initial
values were I/D = 1.8 (o-xylene), 0.9 (m-xylene), and 0.3
(p-xylene) on USHY. This could be due to a higher stabil-
ity of the benzylic carbocation formed from the ortho iso-
mer, which is the first intermediate in disproportionation
(11).
It should be remarked that at high conversion of m- and
p-cresols over HZSM-5, there is a pronounced decrease in
the I/D ratio (Fig. 3b). This decrease can be explained by
the secondary disproportionation of o-cresol resulting from
isomerization. This was expected, as the disproportionation
selectivity from o-cresol is much higher than that from p-
or m-cresols (17). No decrease is observed with USHY for
p- and o-cresol, which is not abnormal, as the difference
in the I/D ratio from the cresol isomers is more limited
than that over HZSM-5. The catalysts deactivation affects
both isomerization and disproportionation reactions, but
the latter one is normally the most affected, as can be seen
from the results given in Table 4, where the deactivation
FIG. 2. Cresol conversion (t = 10 min) as a function of contact time
(W/F, (gh)/mol) on (a) USHY and (b) HZSM-5.

282
IMBERT, GUISNET, AND GNEP
TABLE 4
Initial Isomerization (I0, %) and Disproportionation (D0, %) and
the Corresponding Deactivation Constants (kd) for USHY(4.5) at
380 C and 5.15 (hg)/mol
o
m
p
a
a
a
X0
kd
X0
kd
X0
kd
I
D
25.8
46.2
0.061
0.062
26.9
43.4
0.0198
0.0503
32.9
54.0
0.017
0.043
a X0 represents I0 in the first row and D0 in the second row.
constant kd is usually much higher for the disproportiona-
tion reaction than for isomerization. This can be attributed
to the deactivation of the stronger acid sites required for a
disproportionation reaction to occur, or to the narrowing
of pore diameter by coke deposition, which favours isomer-
ization products by limiting disproportionation product dif-
fusion, and/or to the xylenols being strongly adsorbed on
the acid sites; once formed, they do not desorb immediately
but continue to react, leading to the formation of coke pr-
ecursors (12).
At low conversion m-cresol leads preferentially to the
para isomer. This was expected over HZSM-5 owing to
the shape selectivity of this medium pore size zeolite; thus,
large values of the p/o ratio are obtained from m-xylene
isomerization (7, 9). On the other hand, the high p/o ratio
(>2 at zero conversion) found for USHY (see Fig. 4b) is
quite unexpected, from the intramolecular mechanism of
isomerization through the methyl shift, in the benzenium
ion intermediates. A value close to 1 should be obtained
as was the case with m-xylene isomerization over HFAU
zeolites (9, 11). However, a high p/o ratio has also been
observed from m-cresol on -alumina and silica-alumina
(24). Therefore, it appears that even with a large pore
zeolites, isomerization is limited by the product desorp-
tion. In agreement with this proposal, there is an appar-
ent direct transformation of o-cresol into p-cresol and vice
versa.
The m/o ratio (from p-cresol) and the m/p ratio (from
o-cresol) are both relatively high at zero conversion (Fig. 4),
as expected for consecutive reactions; however, the m/p
ratios, on both catalysts, have the same values and are lower
than the thermodynamic value. This can be explained by
the p-cresol rapid desorption. In the case of HZSM-5, it
might also be possible that o-cresol is directly transformed
to p-cresol, through consecutive disproportionation and
transalkylation reactions, on the external surface; there are
not space limitations.
FIG. 4. Comparison of isomerization selectivities on both catalysts:
(a) m/o from p-cresol, (b) p/o from m-cresol, and (c) m/p from o-cresol.
comparison, the data for HZSM-5 (17) and the calculated
equilibrium composition. The compositions approached at
high conversions over USHY at 380 C are given in Table 5
together with those previously reported by us for HZSM-5
(17), the thermodynamic calculated values (25), and the
TABLE 5
Equilibrium Composition of Isomer Mixtures
Compounds
ortho
meta
para
Ref.
Cresols
44
42
14
This work on USHY(4.5)
The cresol isomer distribution of the reaction products
for USHY is shown in Fig. 5, in a triangular composition dia-
gram, where each symbol represents an experimental value,
obtained by varying the contact time or reaction time on
stream (catalyst decay). Also on the diagram are shown, for
Cresols
Cresols
Toluidines
Toluidines
Xylenes
36.0
37.0
31.0
33.7
23.0
48.0
58.0
52.0
50.2
53.0
16.0
5.0
17.0
16.1
24.0
(17)
Calculated (25)
(15)
(14)
(26)

CRESOL TRANSFORMATION ON USHY AND HZSM-5
283
should be a function of the activity coefficient characteristic
of the given system (cresol–zeolite).
Xylenol Distribution
Whatever the catalyst, in the case of o-cresol, at low con-
versions the primary xylenols are as expected: 2,4-, 2,5-,
and 2,6-xylenol, but as shown in Fig. 6, as the conversion
increases, all other xylenols are formed via a rapid intercon-
version, according to the scheme shown in Fig. 7. However,
some differencesmaybe worth noting:on one hand, the per-
centage of 2,4-xylenol is higher on USHY than on HZSM-5,
while the reverse is true for 2,5-xylenol, for a low level of
disproportionation; on the other hand, over HZSM-5 the
percentage of 2,4- and 2,5-xylenol initially are similar and
from about 2.5% disproportionation, 2,5-xylenol predom-
inates over 2,4-xylenol, whereas on USHY the percentage
of 2,4- is clearly higher than that of 2,5-xylenol up to about
15% disproportionation. Since 2,4-xylenol should be kinet-
ically favoured due to the stronger para-directing effect of
FIG. 5. Cresol triangular composition diagram for USHY and
HZSM-5.
equilibrium composition obtained on HZSM-5 for xylenes
(26) and toluidines (14, 15). As can be seen, our results
substantially differ from the calculated values using the
thermodynamic free energies of formation (25). Our ex-
perimentally found composition at high conversions was
further tested using binary and ternary cresol mixtures of
different compositions and in all cases the product compo-
sition shifted toward 44% o-cresol, 42% m-cresol, and 14%
p-cresol on USHY, which are quite different from the calcu-
lated 37, 58, and 5% , respectively. Although for xylenes the
experimentally found equilibrium composition agrees with
the thermodynamically calculated value (26), in the case of
toluidines it does not (14, 15), nor in the case of cresols,
as reported for HZSM-5 in (17) and in the present work
for USHY. Hardy and Davis (14), based on the similar-
ity between their experimental equilibrium compositions
for toluidines and those experimentally found for xylenes
have suggested that the equilibrium value calculated from
the API (27) data is incorrect. In our case on both cata-
lysts, para-selective—HZSM-5—and non-para-selective—
USHY—, the experimentally found amount of p-cresol is
substantially higher than that expected from the calculated
equilibrium composition. Beyond what we have shown in
previous works (17, 18), Fig. 5 clearly illustrates the shape-
selectivitypath toward the equilibrium from p and m-cresol,
but not from o-cresol; here, the two catalysts show diffusion
disguise kineticssince apparentlyp-cresolisformed directly
from ortho from the very beginning. The difference in the
composition at high conversions between the two catalysts
and that calculated using the thermodynamic data seems to
imply that the composition is sensitive to the catalyst struc-
ture; otherwise, to obtain the true equilibrium values, the
FIG. 6. Xylenols distribution from o-cresol disproportionation (a) on
experimental ones have to be corrected by a factor which USHY and (b) on HZSM-5.

284
IMBERT, GUISNET, AND GNEP
2,6- 3,5-xylenol (see Fig. 8b), while on USHY they fol-
low2,4- 3,5- > 2,6-xylenol(see Fig. 8a). The rapid xylenol
isomerization explains why the 2,4-xylenol, a secondary-
formed product, is found in greater proportion than the
primary-formed 3,4- and 2,3-xylenol.
For p-cresol, the main product of disproportionation ex-
pected is 2,4-xylenol, which should be formed by inter-
molecular transfer of a methyl group to a position ortho to
the –OH. On USHY (see Fig. 9a) both 2,4- and 3,4-xylenol
decrease with conversion, while the rest (secondary-formed
xylenols) increase. In contrast, on HZSM-5 at low con-
versions the only xylenols observed were the 2,4- and
2,5-xylenol isomer, as shown in Fig. 9b. Therefore, 2,4-
xylenol rapidly isomerizes to 2,5-xylenol, which desorbs
very rapidly as soon as it is formed; this result once again
confirms our previous statement in the sense that 2,5-
xylenol should diffuse faster than 2,4-xylenol within the
FIG. 7. Xylenols interconversion scheme.
the –OH group in relation to that of the –CH3 group, the
fact that it is not so in the case of HZSM-5 has to be associ-
ated with a diffusion disguised kinetics, which implies that
2,5-xylenol diffuses faster than 2,4-xylenol. In the case of
secondary-formed xylenols, 2,3- and 3,4-xylenols can only
be directly formed from 2,4-xylenol, and their formation in-
volves a –CH3 shift to a position which is ortho to a methyl
group and meta to the –OH group, so they are formed
in approximately the same proportion on HZSM-5. The
third secondary xylenol is the 3,5-xylenol, which can only
be formed directly from the 2,5-xylenol by a methyl shift
to the unfavoured meta position to both –OH and –CH3
groups, and whose formation is further limited, due to its
size and shape, inside the HZSM- 5 channels. Therefore,
the secondary-formed xylenols follow the sequence 2,3-
3,4- > 3,5-xylenol, as shown in Fig. 6b; in contrast, on USHY
the order is 2,3- > 3,5- > 3,4-xylenol (see Fig. 6a).
For m-cresol, at low conversions the observed isomers co-
incide with the expected ones; among the primary-formed
xylenols 2,5-xylenol is the most favored on both zeolites
because it is formed by alkylation in a position, which
is ortho to the –OH group and para to the –CH3 group,
for meta-cresol disproportionation. The other two primary
xylenols, although favored because one (3,4-xylenol) is or-
tho to the methyl group and para to the –OH group and the
other (2,3-xylenol) is ortho to both, face sterical hindrance,
particularly the latter one. Consequently, their sequence
is 2,5-xylenol > 3,4-xylenol 2,3-xylenol (see Fig. 8). The
FIG. 8. Xylenols distribution from m-cresol disproportionation. Leg-
secondary-formed xylenols on HZSM-5 follow 2,4- > end as in Fig. 6.

CRESOL TRANSFORMATION ON USHY AND HZSM-5
285
tive importance depends on the catalysts characteristics
(pore structure and acidity), on the reacting cresol itself,
and on the reaction conditions (temperature, total pres-
sure, reactant partial pressure, etc.). The USHY selectivity
toward the monomolecular isomerization reaction is by far
lower than that of HZSM-5, due to its size/shape selec-
tivity. On both USHY and HZSM-5 catalysts, the o-cresol
shows higher disproportionation selectivity than the other
two cresol isomers, this disproportionation occurring via
diphenylmethane intermediates. On USHY the I/D selec-
tivities are similar for all cresols at medium- and high-
conversion levels, but for low conversions m-cresol shows
higher isomerization selectivity than o-cresol. Moreover,
even on a large pore zeolite like USHY, the cresol iso-
merization selectivity is limited by product desorption. The
cresol isomer composition at high conversions on USHY
is 44% o-cresol, 42% m-cresol, and 14% p-cresol. Once
formed, xylenols undergo a rapid isomerization.
ACKNOWLEDGMENTS
The authors wish to acknowledge the PCP-France, CONICIT (Project
S1-95000715) and CDCHT–ULA (Projects C-667-94-8-B and C- 665-94-
8-A) for their financial support.
REFERENCES
1. Franck, H. G., and Stadelhofer, J./W., in “Industrial Aromatic Chem-
istry.” Springer, Berlin, 1987. Wu, M. M., U.S. Patent 4,391,998, Mobil
Oil Co. (1983). Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
and Moreno, M. S., Stud. Surf. Sci. Catal. 41, 249 (1988). Campelo,
J. M., Garcia, A., Luna, D., Marinas, J. M., and Moreno, M. S., Bull.
Soc. Chim. France 283 (1988).
FIG. 9. Xylenols distribution from p-cresol disproportionation. Leg-
end as in Fig. 6.
2. Marczewski, M., Bodibo, J. P., Perot, G., and Guisnet, M., J. Mol. Catal.
50, 211 (1989).
3. Marczewski, M., Perot, G., and Guisnet, M., Stud. Surf. Sci. Catal. 41,
273 (1988).
4. Chang, N. S., Cheng, C. C., Chu, S. J., Chen, P. Y., and Chuang, T. K.,
Stud. Surf. Sci. Catal. 46, 223 (1989).
5. Balsama, S., Beltrame, P., Beltrame, P. L., Carniti, P., Forni, L., and
Zuretti, G., Appl. Catal. 13, 161 (1984).
HZSM-5 channel system. As conversion increases, the rest
of the xylenols are formed in approximately the same pro-
portion, while 2,5-xylenol decreases.
6. Parton, R. F., Jacobs, J. M., van Ooteghem, H., and Jacobs, P. A., Stud.
Surf. Sci. Catal. 46, 211 (1989).
7. Chen, N. Y., Garwood, W. E., and Dwyer, F. G., “Shape Selective
Catalysis in Industrial Applications, Chemical Industries.” Dekker,
New York, 1989.
8. Corte´s, A., and Corma, A., J. Catal. 51, 338 (1978).
9. Martens, J. A., Pe´rez-Pariente, J., Sastre, E., Corma, A., and Jacobs,
P. A., Appl. Catal. 45, 85 (1988).
10. Corma, A., Llopis, F., and Monton, J. B., J. Catal. 140, 384 (1993).
11. Morin, S., Gnep, N. S., and Guisnet, M., J. Catal. 159, 296 (1996).
12. Morin, S., Ayrault, P., Gnep, N. S., and Guisnet, M., Appl. Catal. 166,
281 (1998).
13. Toray Industries, Japan Kokai Tkkyo Koho, JP 82 85 330 (1982). Baltes,
H., Leupold, E. I. (Hoechst A. G.) Ger. Offen. DE 3 334 674, 1985.
14. Hardy, R. H., and Davis, B. H., J. Catal. 111, 146 (1988).
15. Weigert, F. J., J. Org. Chem. 51, 3296 (1987).
16. Young, L. B., Butter, S. A., and Kaeding, W. W., J. Catal. 76, 418
(1982).
CONCLUSIONS
The transformation of cresols has been studied over
USHY and compared with HZSM-5, with different pore
size, structure, and different acidity; their catalytic stability,
activity, and selectivity were compared at 380 C and a re-
actant partial pressure of 0.1 bar. Under the experimental
conditions investigated HZSM-5 is more catalytically sta-
ble, though less active than USHY, in terms of moles con-
verted per gram of catalyst; however, in terms of molecules
converted per acid site, the HZSM-5 is more active than
USHY, except for o-cresol. The cresol TON for USHY
are pretty similar, whereas on HZSM-5 the cresol reac-
tivities follow the sequence p- > m- > o-cresol. The cresols
react over the zeolites mainly through two parallel reac-
tions, isomerization and disproportionation, whose rela-
17. Imbert, F. E., Gnep, N. S., and Guisnet, M., J. Catal. 172, 307
(1997).

286
IMBERT, GUISNET, AND GNEP
18. Imbert, F. E., Gnep, N. S., and Guisnet, M., Catal. Lett. 49, 121 24. Herna´ndez, J. C., and Imbert, F. E., 5th Euroworkshop “Microporous
(1997).
Materials: their Application in Catalysis,” Gandı´a, Spain, 1998.
25. Stull, D. R., Westrum, E. F., Jr., and Sinke, G. C., “The Chemical
Thermodynamics of Organic Compounds.” Wiley, New York, 1969.
26. Collins, D. J., Medina, R. J., and Davis, B. H., Can. J. Chem. Eng. 61,
29 (1983).
19. Guisnet, M., and Magnoux, P., Catal. Today 36, 477 (1997).
20. Khabtou, S., Chevreau, T., and Lavalley, J. C., Microporous Mater. 3,
133 (1994).
21. Corma, A., Fornes, V., and Rey, F., Appl. Catal. 59, 267 (1990).
22. Fritz, P. O., and Lunsford, J. H., J. Catal. 118, 85 (1989).
23. Lonyi, F., and Lunsford, J. H., J. Catal. 136, 566 (1992).
27. Kudchadker, A. P., Kudchadker, S. A., and Wilhoit, R. C., API Publ.
720 1982. Cited in (14).