N-Alkylation of Aniline with Methanol over CrPO4 and CrPO4-AlPO4 (5-50 wt% AlPO4) Catalysts

Article abstract of DOI:10.1006/jcat.1997.1868

The vapor-phase alkylation of aniline with methanol was carried out at atmospheric pressure and different temperatures (573-673 K) and feed rates over CrPO₄ (CrP) and CrPO₄-AlPO₄ (CrAlP, 5-50 wt% AlPO₄) catalysts. Aniline methylation was a pseudo-first-order process with respect to aniline concentration. The effect of surface acidity on aniline alkylation, as measured by pyridine and 2,6-dimethylpyridine adsorption, is discussed. The influence of feed rate (WHSV), time-on-stream and temperature on both activity and product selectivity was also investigated. Alkylation is a sequential reaction process, in which methylation of aniline produces N-methylaniline (NMA), then N,N-dimethylaniline (NNDMA) and subsequently N,N-dimethyltoluidines (NNDMT, p- > o-). N-methylation products (NMA + NNDMA) are predominant with a selectivity over 90 mol% in the temperature range 573-673 K. Selectivity to N,N-dimethylaniline is enhanced as reaction temperature and/or contact time increases. N,N-dimethyltoluidines are only present in very small amounts at higher temperatures and/or contact times.

Full text of DOI:10.1006/jcat.1997.1868

JOURNAL OF CATALYSIS 172, 103–109 (1997)
ARTICLE NO. CA971868
N-Alkylation of Aniline with Methanol over CrPO and CrPO –AlPO
4
4
4
(
5–50 wt% AlPO ) Catalysts
4
1
F. M. Bautista, J. M. Campelo, A. Garcia, D. Luna, J. M. Marinas, A. A. Romero, and M. R. Urbano
Departamento Qu ´ı mica Org a´ nica, Facultad de Ciencias, Universidad de C o´ rdoba, Avenida San Alberto Magno, s/n, E-14004 C o´ rdoba, Spain
Received February 19, 1997; revised July 7, 1997; accepted August 14, 1997
strength) and shape-selectivity in the solid acid catalyst as
The vapor-phase alkylation of aniline with methanol was car- well as reaction conditions (temperature, composition and
ried out at atmospheric pressure and different temperatures feed rate). Low reaction temperatures favor N-alkylation,
(
(
573–673 K) and feed rates over CrPO4 (CrP) and CrPO4–AlPO4
CrAlP, 5–50 wt% AlPO4) catalysts. Aniline methylation was a
whereas at high temperatures C-alkylation appears. There
are examples in the literature which showed that the forma-
tion of N- and C-methylanilines depend on reaction condi-
tions and catalysts. Zeolites with large pores and a three-
dimensional channel structure (faujasite and beta) favor
polyalkylated phenyl-rings even when no strong Br o¨ nsted
acid sites are present, such as for Mg- or Zn-ion-exchanged
Y zeolites (18). Moreover, the activity of acid catalysts for
pseudo-first-order process with respect to aniline concentration.
The effect of surface acidity on aniline alkylation, as measured by
pyridine and 2,6-dimethylpyridine adsorption, is discussed. The in-
fluence of feed rate (WHSV), time-on-stream and temperature on
both activity and product selectivity was also investigated. Alkyla-
tion is a sequential reaction process, in which methylation of ani-
line produces N-methylaniline (NMA), then N,N-dimethylaniline
(
NNDMA) and subsequently N,N-dimethyltoluidines (NNDMT, aniline alkylation might be suppressed by the adsorption
p- > o-). N-methylation products (NMA + NNDMA) are pre- of aniline since it is a strong base. It therefore seems that
dominant with a selectivity over 90 mol% in the temperature the use of a strong acid catalyst is not suitable for this reac-
range 573–673 K. Selectivity to N,N-dimethylaniline is enhanced
as reaction temperature and/or contact time increases. N,N-
dimethyltoluidines are only present in very small amounts at higher
tion. In this sense, Woo et al. (16) using metallosilicates of
the pentasil family, suggest that strong, medium, and weak
acid sites are active sites to produce, respectively, C-alkylate
and coke, N,N-dimethylaniline and N-methyltoluidine, and
N-methylaniline.
temperatures and/or contact times.
�c 1997 Academic Press
As far as the aniline alkylation is concerned, in two
previous papers (26, 27) we studied the N-methylation of
aniline over AlPO4, Al2O3, AlPO4–Al2O3, AlPO4–TiO2,
and AlPO4–ZrO2 catalysts. The results showed that
AlPO4–Al2O3 catalysts exhibited significantly higher ac-
INTRODUCTION
Aniline alkylation is an industrially important reaction
due to the fact that many valuable bulk, intermediate and
fine chemicals can be prepared. Thus, alkylanilines form the
basic raw materials for synthesis of organic chemicals and
chemical intermediates or additives in dyes, synthetic rub-
bers, explosives, herbicides, and pharmaceuticals. Vapor-
phase aniline alkylation over environmentally safe solid
catalysts is an answer to the conventional method of pro-
ducing alkylanilines using mineral acids and Friedel–Crafts
type catalysts (1). The traditional route also presents the
disadvantage of high capital cost, reactor corrosion, and
formation of by-products that cannot be recycled. Several
types of catalysts based on oxides and supported oxides
tivity (an order of magnitude) and, besides, there was a
slight increase in activity on increasing Al2O3 loading up to
1
5 wt% (27), thereby showing that AlPO4–Al2O3 catalysts
are suitable for selectively obtaining N-alkylation prod-
ucts. Moreover, a comparison with SAPO-5 (28, 29) and
a commercial SiO2–Al2O3 (13 wt% Al2O3, Si-235 Harshaw
Chemie) catalysts (26), which displayed a greater number
of medium and strong acid sites than AlPO4–Al2O3 cata-
lysts, showed that the former catalysts exhibited a decrease
in aniline methylation activity with respect to the latter
ones. Nevertheless, selectivity to N-alkylation (N-methyl
and N,N-dimethylanilines) attained 100 mol% in all cases.
On the other hand, in previous papers (30–35) we pre-
pared and characterized different CrPO4 (CrP, Cr/P = 1)
and CrPO4–AlPO4 (CrAlP, 5–50 wt% AlPO4) catalysts.
(
2–10), zeolites (11–23), and microporous aluminophos-
phates (23–25) have been tested for aniline methylation.
The main factors influencing activity and selectivity (N-
and/or C-alkylation) are acid-base properties (number and
Their texture, crystal structure, and surface acid–base char-
acter are affected by the preparative method (gellification
1
To whom correspondence should be addressed.
103
0021-9517/97 $25.00
Copyright �c 1997 by Academic Press
All rights of reproduction in any form reserved.

1
04
BAUTISTA ET AL.
agent), AlPO4 loading and calcination temperature. Thus, phoric acid solutions, with aqueous ammonia (CrP–A),
with regard to CrPO4 catalysts, those obtained in propylene propylene oxide (CrP–P), or propylene oxide-aqueous am-
oxide-aqueous ammonia mixed medium exhibited higher monia (CrP–PA). CrPO4–AlPO4 catalysts of varying com-
surface acidity and, hence, improved catalytic activity for position (5–50 wt% AlPO4) were obtained with aqueous
acid catalyzed reactions like cyclohexene (30, 31, 33) and ammonia (CrAlP–A) and CrPO4–AlPO4 catalysts with 5
3
,3-dimethyl-1-butene (35) isomerizations, cumene conver- and 10 wt% AlPO4 were also obtained by precipitation
sion (30, 31, 33), and 2-propanol dehydration (34). Besides, with propylene oxide-aqueous ammonia (CrAlP–PA). De-
this CrPO4 catalyst exhibited an improved activity in rela- tails on the preparation, as well as on the characterization
tion to AlPO4 and AlPO4–Al2O3 catalysts, similarly pre- of all catalysts, have been previously described (30–33). The
pared. Moreover, the modification of CrPO4 with small catalysts are designated by CrP or CrAlP followed by the
amounts of AlPO4 leads to CrPO4–AlPO4 catalysts ex- letters indicating the precipitation medium, a number that
hibiting greater catalytic activity for the above-cited pro- indicates the AlPO4 loading (wt% ) in CrAlP catalysts and
cesses. In all cases, AlPO4, AlPO4–Al2O3, CrPO4, and by another number which refers to the calcination temper-
CrPO4–AlPO4 catalysts, the activities can be well inter- ature in K (CrP–A-923, CrAlP–PA-10–773, and so on).
preted to result from the differences in the number and
strength of acid sites.
The surface area (SBET), pore volume (Vp), and mean
pore radius (rp) are collected in Table 1.
In the present work we have carried out the vapor-phase
N-methylation of aniline over CrPO4 and CrPO4–AlPO4 Surface Acidity
catalysts, prepared in different ways, under identical re-
action conditions that are described elsewhere for AlPO4,
AlPO4–Al2O3, SAPO-5, and SiO2–Al2O3 catalysts (26, 27).
We will focus on the effect of surface acidity upon ac-
tivity and reaction selectivity. In addition, the influences
of methanol/aniline molar ratio, time-on-stream, reaction
temperature, and contact time are also studied.
The surface acidity was measured in a dynamic mode
by means of the gas-phase adsorption of probe molecules
by using a pulse-chromatographic technique (36, 37). As
probe molecules we used pyridine (PY), which reacts with
Br o¨ nsted acid sites and Lewis acid sites, and 2,6-dimethyl-
pyridine (DMPY) which titrates only Br o¨ nsted acid sites.
The results are in Table 1.
EXPERIMENTAL
Catalytic Activity Measurements
Materials
The reactions were conducted in a vapor-phase continu-
Methanol, aniline, N-methylaniline, N,N-dimethylani- ous stainless-steel down-flow fixed-bed reactor (6 mm ID)
line, and N,N-dimethyltoluidines (ortho and para) were of surrounded by an electric heater. An iron–constantan ther-
high purity (99+% ) and used without further purification. mocouple was placed in the middle of the catalyst bed and
the unit operated at atmospheric pressure. The substrate
Catalysts
was delivered at a set flow rate using a liquid syringe pump
CrPO4 (CrP, Cr/P = 1) catalysts (Table 1) were obtained (Harward Md. 44) and was vaporized prior to passing it
by precipitation, from chromium nitrate and aqueous phos- through the catalyst bed in the presence ofa flowofnitrogen
TABLE 1
Textural Properties and Surface Acidity (vs PY and DMPY) of CrPO4 and CrPO4–AlPO4 Catalysts
SBET
(m /g)
Vp
(ml/g)
rp
(nm)
PYa
(�mol/g)
PYb
(�mol/g)
DMPYa
(�mol/g)
DMPYb
(�mol/g)
2
Catalysts
CrP–A-923
CrP–P-923
Cr–PA-923
CrAlP–PA-5–923
CrAlP–PA-10–773
CrAlP–PA-10–993
CrAlP–A-5–923
CrAlP–A-10–923
CrAlP–A-20–923
CrAlP–A-30–923
CrAlP–A-50–923
c
8
6
—
—
—
—
15.5
9.3
9.9
14.9
9.9
10.6
18.4
17.6
10.0
18
24
70
91
86
85
25
26
27
38
76
17
17
52
56
63
64
14
26
22
37
26
9
7
—
—
c
23
55
58
38
16
16
9
0.18
0.26
0.29
0.28
0.08
0.09
0.08
0.10
0.20
58
61
60
54
13
8
6
17
15
58
50
54
40
13
8
6
10
15
13
40
a
Adsorption at 573 K.
Adsorption at 673 K.
There is no adsorption of the probe molecule.
b
c

N-ALKYLATION OF ANILINE
105
carrier gas (3 L h 1) regulated by mass-flow controllers. The produces N-methyl (NMA) and N,N-dimethylaniline
catalyst charges (W) were small, usually 0.1–0.3 g, retained (NNDMA) with a selectivity over 90 mol% with respect
by quartz wool at almost the center of the reactor, and reac- to the sum of both products, thus indicating the selective
tion temperature was 673 K, except where otherwise speci- formation of N-alkylated products. N,N-dimethyltoluidines
fied. Standard catalyst pretreatment was carried out in situ (p- > o-) are only present in very small amounts at the
at 673 K for 1 h under a stream of high purity nitrogen highest reaction temperatures.
�
(
99.999% , H2O < 3 ppm). In order to prevent any conden-
In the absence of boundary, inter- and intraparticle dif-
�
6
� 1
sation of reactant and products all connections were heated fusion effects (feed rates over 3 � 10 mol s and catalyst
by heating tapes at 490 K. Blank runs at 673 K showed that particle size lower than 0.149 mm) aniline conversion data
under the experimental conditions used in this work, the (XAN) are fitted in a first-order rate equation:
thermal effects could be neglected.
The reaction products were on-line sampled every 15 min
and analyzed by GLC (Varian Md. 3400) by using a stain-
less-steel column (2 m � 3 mm) of 10% Carbowax 20 M/2%
KOH on Chromosorb W-AW 80/100. Product character-
ization was performed by GC-MS (HP 5890 gas-chro-
atograph coupled with a VG AutoSpec high-resolution
massspectrometer) usingproductscondensed in a cold trap.
Reaction products were: N-methyl (NMA) and N,N-di-
methylaniline (NNDMA) and minor amounts of N,N-
dimethyltoluidines (NNDMT, p- > o-). Product such as
diphenylamine was never found by high-resolution mass
spectrometry. Response factors of the reaction products
were determined with respect to aniline from GC analysis
using known compounds in calibration mixtures of spec-
ified compositions. The conversions reported here are on
a methanol-free aniline basis and the selectivities are ex-
pressed as the ratio of moles of a given product to the to-
tal number of moles produced (mol% ). Deactivation by
coke was negligible in all cases except for the CrP–PA-923
ln[1/(1 � XAN)] = k(W/F),
[1]
where W is the catalyst weight and F the feed rate. Runs
were performed with at least three feed rates in the range
�
6
� 1
3
.0–6.7 � 10 mol s (5 M aniline in methanol), and at re-
action temperatures in the range 573–673 K. Under these
conditions, substrate conversion was easily kept well under
2
0 mol% . The ln[1/(1 � XAN)] versus (W/F) plots, accord-
ing to Eq. [1], are linear and also pass through the origin,
indicating a good fit of the data to Eq. [1] and, therefore,
the methylation process is a first-order reaction with respect
to aniline concentration. Moreover, calculations were used
only to compare the reactivities of the different catalysts
and not aimed at finding the detailed rate equations.
The slope of these straight lines yielded the values of the
�
1
� 1
initial reaction rate constants (k) in mol g s . The values
obtained at 673 K for the different catalysts, after 2 h and
h on stream, are collected in Table 2. A Students’ t-test of
significance, using the position of the 98% confidence limit
lines and the determination coefficient value (always over
.99) of the regressions, showed that the activity data are
4
catalyst.
0
significant at levels over 1% . At least three measurements
were used to calculate each value of k and all values are
RESULTS AND DISCUSSION
The alkylation of aniline with methanol on CrPO4 reproducible to within about 8% . The conversion of ani-
and CrPO4–AlPO4 (5–50 wt% AlPO4, CrAlP) catalysts line (XAN) and the reaction selectivities to N-methylaniline
TABLE 2
Aniline Conversion (XAN, mol%), Reaction Rate Constant (k, mol/g s) and Product Selectivities (S, mol%) in Aniline
a
Alkylation with Methanol over CrPO4 and CrPO4–AlPO4 (5–50 wt% AlPO4) Catalysts
2
h on-stream
4 h on-stream
6
6
Catalyst
XAN
k � 10
SNMA
SNNDMA
XAN
k � 10
SNMA
SNNDMA
CrP–A-923
CrP–P-923
CrP–PA-923
CrAlP–PA-5–923
CrAl–PA-10–923
CrAlP–A-5–923
CrAlP–A-10–923
CrAlP–A-20–923
CrAlP–A-30–923
CrAlP–A-50–923
1.0
1.3
10.0
5.9
6.1
2.4
1.5
1.6
1.8
3.4
0.3
0.4
3.5
2.0
2.1
0.8
0.5
0.6
0.6
1.2
100
100
92.7
94.6
94.3
97.7
98.6
98.3
98.1
96.1
—
—
1.0
1.3
8.2
5.9
6.0
2.1
1.5
1.6
1.8
3.3
0.3
0.4
2.8
2.0
2.1
0.7
0.5
0.5
0.6
1.1
100
100
93.6
94.9
94.3
98.4
98.5
98.4
98.0
96.1
—
—
7.3
5.4
5.6
2.3
1.4
1.7
1.9
3.9
6.4
5.1
5.6
1.6
1.5
1.6
2.0
3.9
a
T, 673 K; F = 3.36 � 10� 6 mol/s; WHSV = 9.53 h� 1; 5 M aniline in methanol.

1
06
BAUTISTA ET AL.
(
SNMA) and N,N-dimethylaniline (SNNDMA) are also col-
TABLE 3
lected in Table 2.
Selectivities to N-Methylaniline and N,N-Dimethylaniline (mol%)
On CrPO4 catalysts the aniline conversions are in the
range 1–10 mol% and increase in the order CrP–A <
CrP–P < CrP–PA, i.e., as the surface acidity increases. The
a
at Various Aniline Conversions
XAN = 5 mol% XAN = 10 mol% XAN = 20 mol%
incorporation of AlPO4 into CrPO4 resulted in CrAlP cata-
lysts in which the catalytic performance depends on the pre-
cipitation method. Thus, those CrAlP catalysts obtained in
aqueous ammonia (CrAlP–A) exhibit, irrespective of the
AlPO4 loading, an increase in surface acidity and activ-
ity relative to the CrP–A catalysts; the increase in activ- CrAlP–A-10–923 92.5
ity is more significant in the case of CrAlP–A-50 catalyst. CrAlP–
Catalyst
SNMA SNNDMA SNMA
SNNDMA
SNMA
SNNDMA
CrP–PA-923
CrAlP–PA-5–923 93.6
CrAl–PA-10–923 94.5
93.7
6.3
6.4
5.5
5.4
7.5
8.2
6.1
5.9
91.3
89.1
88.3
88.4
8.7
10.9
11.7
11.6
83.6
87.0
79.9
16.4
13.0
20.1
b
b
b
b
b
CrAlP–A-5–923
94.6
—
—
—
—
—
—b
A-20–923 91.8
CrAlP–A-30–923 93.9
CrAlP–A-50–923 94.1
89.0
88.9
88.6
11.0
11.1
11.4
Moreover, when the CrAlP catalyst is prepared in propy-
lene oxide-aqueous ammonia (CrAlP–PA), aniline methy-
lation activity decreases with respect to the CrP–PA
catalyst. Notwithstanding, CrAlP–PA-5 and CrAlP–PA-10
a
� 1
WHSV = 9.53 h ; 5 M aniline in methanol.
b
There is no result at this conversion.
catalysts exhibit higher aniline methylation activity than
those obtained in ammonia (CrAlP–A-5 and CrAlP–A-10).
With respect to the surface acidity, the incorporation of
AlPO4 to the CrP–PA catalyst results in CrAlP–PA cata-
lysts exhibiting higher thermal stability and surface acidity
and, hence, greater catalyst activity in cyclohexene (30, 31,
oxide (Al2O3, TiO2, and ZrO2), SAPO-5, and commercial
SiO2–Al2O3 catalysts (26, 27) provide enough data to cor-
roborate that strong acid sites are not required for aniline
N-methylation.
As far as selectivity of aniline N-methylation is con-
cerned. Table 3 compares product selectivities (SNMA and
SNNDMA) at three similar conversion levels (5–20 mol% )
and at a constant reaction temperature of 673 K. Under
the conditions used, data in Table 3 show that the addition
of AlPO4 to CrPO4 did not cause any significant change
in the selectivity patterns exhibited by CrPO4; thus similar
adsorbed species must be formed on CrPO4 and CrPO4–
AlPO4 catalysts. Moreover, SNNDMA increased with aniline
conversion. As compared to previously studied AlPO4,
AlPO4-metal oxide (Al2O3, TiO2, and ZrO2), SAPO-5,
and commercial SiO2–Al2O3 catalysts (26, 27), CrPO4, and
CrPO4–AlPO4 catalysts are less active but, however, they
3
3) and 3,3-dimethyl-1-butene (35) skeletal isomerizations,
typical acid-catalyzed reactions. However, in aniline methy-
lation this increased acidity not only did not result in in-
creased activity, but also decreased XAN. Thus, the increase
in acidity is not reflected in the aniline methylation activ-
ity. This raises the basic question of the role of acidity in
this aniline alkylation reaction. Comparing the acidity data
in Table 1 and the corresponding aniline methylation in
Table 2, one is tempted to believe that there is not a simple
relationship between aniline alkylation activity and acidity
as measured by pyridine and 2,6-dimethylpyridine adsorp-
tion at 673 K.
are more selective for the N-monomethylation process,
yielding N-methylaniline, when similar conversion level
are attained.
A similar decrease has been reported in aniline alkyla-
tion activity with an increase in acidity (16, 19, 38), so that
it may be suggested that strong acid sites are not required
for aniline N-alkylation. In this sense, when using zeolites
as catalysts, the acidity increases as SiO2/Al2O3 ratio while
the catalytic activity showed a maximum at intermediate
Effect of Feed Rate
Figure 1 shows the influence of weight hourly space
�
1
SiO2/Al2O3 ratio (38). This behavior may be explained be- velocity (WHSV = 1.5–9.5 h ) on aniline conversion and
cause that reaction proceeded to a great degree in the weak the product selectivities for CrAlP–PA-5–923 catalyst after
Br o¨ nsted acid sites because aniline and its derivatives, being 2 h on stream at 673 K. On increasing WHSV, aniline
basic, could be more strongly adsorbed at strong acid sites, conversion decreases. This decrease is from 40 mol%
�
1
� 1
thus deactivating the catalyst. Moreover, in the presence (WHSV = 1.51 h ) to 6 mol% (WHSV = 9.53 h ). The
of strong Br o¨ nsted acid sites, the aniline molecule can be product pattern with respect to WHSV reveals that selec-
protonated, thus reducing its alkylation. Thus, Woo et al. tivity to N-methylaniline continuously increases with the
(
16) suggested that strong, medium, and weak acid sites are feed rate. Moreover, decreasing the WHSV enhances the
active sites to produce C-alkylate and coke, NNDMA, and alkylation rate, leading to the favorable formation of N,N-
NMA, respectively, so that strongly acidic catalysts ought dimethylaniline and N,N-dimethyl-p-toluidine, although
to be modified by the incorporation of alkali metal species the latter appears at low selectivity. There is always more
in order to increase the selectivity to NNDMA.
N-methylaniline produced than N,N-dimethylaniline, and
The present results using CrPO4 and CrPO4–AlPO4 cata- the formation of N,N-dimethyltoluidines (p- > o-) is found
lysts and those previously obtained on AlPO4, AlPO4-metal only over the more acidic catalysts and at low WHSV.

N-ALKYLATION OF ANILINE
107
much deactivation of catalysts at 673 K and the selectivity
of N-alkylation remained at 100 mol% during this period
of process time.
Effect of Reaction Temperature
Under the experimental conditions used here (WHSV =
�
1
3
.73 h ), the aniline conversion increased continuously
with reaction temperature (573–673 K), as Fig. 3 shows for
CrAlP–PA-10–773 catalyst after 2 h on-stream. The other
catalysts showed similar behavior.
Although at a given temperature the actual conver-
sion was different for different catalysts, as the tempera-
ture increases from 573 to 673 K, the aniline conversion
always increases. Figure 3 also includes the variation
of selectivity with temperature and, thus, the selectiv-
ity for N-methylaniline decreases, whereas that for N,N-
dimethylaniline increases, as the temperature rises. The se-
lectivity of the former product decreases from 99 mol%
FIG. 1. Effect of feed rate (WHSV, h 1) on aniline conversion and
�
product selectivities over CrAlP–PA-5–923 catalyst (T, 673 K; time-on- (573 K) to 57 mol% (673 K), whereas that of the lat-
stream, 2 h).
ter increases from 1 mol% to 36 mol% . A small in-
crease in the amount of C-alkylated anilines, namely
N,N-dimethyltoluidines (N,N-dimethyl-p-toluidine almost
exclusively), was observed at temperatures above 573 K.
N-alkylation was the predominant reaction with selectivity
nearly 100 mol% up to 623 K, whereas the selectivity to
N,N-dimethyltoluidines in any case attained a high value
The selectivity to N,N-dimethyltoluidines decreases
strongly as WHSV increases.
Effect of Time-on-Stream
A time-on-stream (t.o.s.) study was performed at 673 K at the highest reaction temperature (673 K) even with the
�
1
with a WHSV of 9.53 h and a 5 M solution of aniline in most active catalyst (never surpassing 12 mol% of the total
methanol. A typicalactivityprofile ofaniline methylation as selectivity).
a function of t.o.s. on CrP and CrAlP catalysts is presented
in Fig. 2 for CrP–PA-923 catalyst. Here we can see that
the aniline conversion decreased slightly initially but after
2
h on-stream remained almost unchanged with t.o.s. More-
over, the selectivities to N-methyl and N,N-dimethylaniline
did not show any change within 10 h on-stream. All catalysts
displayed almost the same behavior and so there was not
FIG. 2. Time-on-stream dependence of aniline conversion (XAN) and
product selectivities in aniline alkylation with methanol at 673 K and
WHSV = 9.53 h� (5 M aniline in methanol) over CrP–PA-923 catalyst.
FIG. 3. Effect of reaction temperature on aniline conversion and
�
1
product selectivities over CrAlP–PA-10–773 catalyst (WHSV, 3.73 h
;
1
time-on-stream, 2 h).

1
08
BAUTISTA ET AL.
Reaction Pathway
(p- > o-) at higher temperatures,
In order to better understand the reaction sequence,
we have built the optimum performance envelope (OPE)
curves by plotting fractional conversion to each reac-
tion product against aniline conversion (XAN) for differ-
ent weight ratios of catalyst with respect to introduced
aniline (Fig. 4), such as recently described by Best and
Wojciechowski (39). In these curves we have included ex-
perimental data corresponding to different temperatures
and feed rates on the same diagram. By using such a pro-
cedure, an insignificant scattering of the data is evident in
the selectivity diagrams and clear tendencies can be ob-
served from these curves. Thus, analysis of selectivity data
for aniline methylation through OPE curves (as illustrated
in Fig. 4 for CrAlP–PA-10–773 catalyst), indicated that, in
all cases, N-methylaniline was a primary product coming
from aniline by alkylation, although it is a reactive interme-
diate since a maximum on its OPE curve was found. More-
over, N,N-dimethylaniline is a secondary product since a
zero slope at the origin, together with an upward devia-
tion as XAN increases, was found by plotting their fractional
conversion (XNNDMA) versus XAN. At higher XAN values,
N,N-dimethyltoluidines (p- > o-) also appeared and so they
seem to be secondary reaction products. These secondary
products are formed at the expense of N-methylaniline.
From the data on feed rate, temperature, and OPE curves,
it can be concluded that aniline methylation on CrPO4 and
CrPO4–AlPO4 catalysts follows a sequential reaction path-
MeOH
MeOH
MeOH
AN ���→ NMA ���→ NNDMA ���→ NNDMT (p- > o-),
as occurs on AlPO4, AlPO4–Al2O3, and other acid cata-
lysts (10, 16, 19, 23, 26, 27, 38). The driving force here is
the greater nucleophylicity of the N atom over the C atom.
Besides, N-methylaniline reacts faster than aniline due to
its greater negative charge on the N-atom associated with
the electron donating effect on the methyl group that fa-
cilitates its alkylation. Moreover, dimethyl substitution on
N-atom makes easy ring alkylation at o- or p-position by
electron donation to the benzene ring, while the steric hin-
drance in the o-isomer is responsible for the larger forma-
tion of p-isomer. Notwithstanding, N-methylation always
predominates. The formation of N,N-dimethylaniline via
disproportionation is less favorable than its formation via
successive alkylation (16).
CONCLUSIONS
In the present reaction conditions, alkylation to N-methyl
and N,N-dimethylaniline is the major reaction path in
the alkylation of aniline with methanol over CrPO4 and
CrPO4–AlPO4 catalysts prepared in different ways. Thus,
N-alkylated products result in selectivity over 90 mol%
throughout the 573–673 K temperature range. Varying the
feed rates indicates that low contact time and low tempera-
ture promote the N-monomethylation reaction. Moreover,
N,N-dimethyltoluidines (p- > o-) are only present in very
small quantities at the highest temperatures and/or contact
times. On the other hand, the additon of AlPO4 to CrPO4
did not cause any significant change in the selectivity pat-
tern exhibited by the CrPO4 catalyst. This behavior could
indicate that similar adsorbed active species are formed
on CrPO4 and CrPO4–AlPO4 catalysts. Furthermore, there
way in the formation of N-methylaniline, which is further
converted to N,N-dimethylaniline and, then, this under-
goes further C-methylation to give N,N-dimethyltoluidines
is no simple relationship between aniline alkylation ac-
tivity and surface acidity measured by pyridine and 2,6-
dimethylpyridine adsorption at 573–673 K. As compared
to previously studied AlPO4, AlPO4-metal oxide (Al2O3,
TiO2, and ZrO2), SAPO-5, and commercial SiO2–Al2O3
catalysts (26, 27), CrPO4 and CrPO4–AlPO4 catalysts are
less active, but, however, they are more selective for the
N-monomethylation process, yielding N-methylaniline,
when similar conversion levels are attained.
ACKNOWLEDGMENT
This research was subsidized by grants from the DGICYT (Project
PB92/0816), Ministerio de Educacion y Cultura, and from the Consejeria
de Educacion y Ciencia (Junta de Andalucia).
REFERENCES
FIG. 4. OPE curves for aniline conversion over CrAlP–PA-10–773
catalyst. Fractional conversion at a particular reaction product (X) versus
aniline conversion (XAN).
1. Bhattacharya, A. K., and Nandi, D. K., Ind. Eng. Chem. Prod. Res.
Dev. 14, 162 (1975). [And references cited therein]

N-ALKYLATION OF ANILINE
109
2
. Hill, A. G., Shipp, J. H., and Hill, A. J., Ind. Eng. Chem. 43, 1579 22. Rao, P. R. H. P., Massiani, P., and Barthomeuf, D., in “Zeolite Science
(
1951).
1994: Recent Progress and Discussions (H. G. Karge and J. Weitkamp,
Eds.), p. 287. Elsevier, Amsterdam, 1995.
3
4
. Evans, T. H., and Bourns, A. N., Can. J. Tech. 29, 1 (1951).
. Parera, J. M., Gonzalez, A., and Barral, A., Ind. Eng. Chem. Prod. 23. Sing, P. S., Bandyopadhyay, R., and Rao, B. S., Appl. Catal. A 136, 177
Res. Dev. 7, 259 (1968).
(1996).
5
6
. Naccache, C. M., and Taarit, Y. B., J. Catal. 22, 171 (1971).
. Takamiya, N., Koinuma, Y., Ando, K., and Murai, S., Nippon Kagaku
Kaishi, 1452 (1979).
. Takamiya, N., Yamabe, H., Ando, K., and Murai, S., Nippon Kagaku
Kaishi, 1316 (1980).
. Doraiswamy, L. K., Krishnan, G. R. V., and Mukerjee, S. P., Chem.
Eng. 88, 78 (1981).
. Matsuhashi, H., and Arata, K., Bull. Chem. Soc. Jpn. 64, 2605 (1991).
24. Prasad, S., and Rao, B. S., J. Mol. Catal. 62, L17 (1990).
25. Yang, S. M., and Pan, T. W., J. Chin. Chem. Soc. 42, 935 (1995).
26. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
and Romero, A. A., in “Heterogeneous Catalysis and Fine Chem-
icals IV,” Stud. Surf. Sci. Catal., Vol. 108, p. 123, Elsevier, New
York/Amsterdam, 1997.
7
8
9
27. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
and Romero, A. A., Appl. Catal. A, in press.
1
0. Ko, A. N., Yang, C. L., Zhu, W., and Lin, H., Appl. Catal. A 134, 53 28. Campelo, J. M., Lafont, F., and Marinas, J. M., Zeolites 15, 97 (1995).
(
1996).
29. Campelo, J. M., Lafont, F., and Marinas, J. M., in “13th Iberoamerican
1
1
1
1. Onaka, M., Ishikawa, K., and Izumi, Y., Chem. Lett. 1783 (1982).
2. Mokoto, O., Koji, I., and Yusuke, I., J. Inclusion Phenom. 2, 359 (1984). 30. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
3. Chivadze, G. O., and Chkheidze, L. Z., Iz. Akad. Nauk. Gruz. SSR,
Ser. Khim. 10, 232 (1984). [C. A. 103, 7725u (1985)]
4. Chen, P. Y., Chen, M. C., Chu, H. Y., Chang, N. S., and Chuang,
T. K., in “New developments in Zeolite Science and Technology”
Symposium on Catalysis, Segovia, Spain, 1992,” p. 1031.
and Urbano, M. R., J. Mater. Chem. 3, 975 (1993).
31. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
and Urbano, M. R., J. Mater. Chem. 4, 311 (1994).
32. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
Romero, A. A., and Urbano, M. R., React. Kinet. Catal. Lett. 53, 45
(1994).
1
(
Y. Murakami, A. Iijima, and J. W. Ward, Eds.), p. 739. Elsevier, Am-
sterdam, 1986.
1
1
1
1
5. Ione, K. G., and Kikhtyanin, O. V., in “Zeolites: Facts, Figure, Future”
33. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
Romero, A. A., and Urbano, M. R., React. Kinet. Catal. Lett. 53, 55
(1994).
34. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
Romero, A. A., and Urbano, M. R., React. Kinet. Catal. Lett. 55, 133
(1995).
(
P. A. Jacobs and R. A. van Santen, Eds.), p. 1073. Elsevier, Amster-
dam, 1989.
6. Woo, S. I., Lee, J. K., Hong, S. B., Park, Y. K., and Hu, Y. S., in
“
Zeolites: Facts, Figure, Future” (P. A. Jacobs and R. A. van Santen,
Eds.), p. 1095. Elsevier, Amsterdam, 1989.
7. Chem, P. Y., Chu, S. J., Chang, N. S., and Chuang, T. K., in “Zeolites: 35. Bautista, F. M., Campelo, J. M., Garcia, A., Luna, D., Marinas, J. M.,
Facts, Figure, Future” (P. A. Jacobsand R. A. van Santen, Eds.), p. 1105.
Elsevier, Amsterdam, 1989.
8. Kikhtyanin, O. V., Ione, K. G., Malysheva, L. V., and Toktarev, A. V.,
Romero, A. A., and Urbano, M. R., React. Kinet. Catal. Lett., in press.
36. Ghosh, A. K., and Curthoys, G., J. Chem. Soc. Faraday Trans. I 79,
2569 (1983).
in “Chemistry of Microcrystals” (T. Inui, S. Namba, and T. Tstsumi, 37. Campelo, J. M., Garcia, A., Luna, D., and Marinas, J. M., J. Mater. Sci.
Eds.), p. 319. Elsevier, Amsterdam, 1991.
25, 2513 (1990).
19. Park, I. K., Park, K. Y., and Woo, S. I., Catal. Lett. 26, 169 (1994).
20. Su, B. L., and Barthomeuf, D., Appl. Catal. A 124, 73 (1995).
21. Su, B. L., and Barthomeuf, D., Appl. Catal. A 124, 81 (1995).
38. Narayanan, S., Kumari, V. D., and Rao, A. S., Appl. Catal. A 111, 133
(1994).
39. Best, D. A., and Wojciechowski, B. W., J. Catal. 47, 343 (1977).