Amination of butenes over protonic zeolites

Article abstract of DOI:10.1006/jcat.1996.0326

The reaction of 1-butene and isobutene with ammonia has been investigated, far from thermodynamic equilibrium, in the pressure range 1-6 MPa over a series of acidic zeolites. The kinetics are compatible with a Langmuir-Hinshelwood mechanism involving adsorbed species. The rates of amination increase with the Si/Al ratio of the solid. A small influence of the zeolite structure is noticed on the relative adsorption coefficients in the case of 1-butene but not in that of isobutene. The catalytic activity calculated per proton is higher on MFI than on BEA or HY zeolites, but this effect of the structure is less than an order of magnitude. Under the conditions of reaction used in this work large pore zeolites show a good resistance to deactivation. It is proposed that deactivation is mainly due to the formation of strongly basic polyalkylamines and not to coke.

Full text of DOI:10.1006/jcat.1996.0326

JOURNAL OF CATALYSIS 163, 255–261 (1996)
ARTICLE NO. 0326
Amination of Butenes over Protonic Zeolites
,1
ꢀ
ꢀ
ꢀ
M. Lequitte, F. Figueras, C. Moreau, and S. Hub†
ꢀ
Laboratoire de Mate´riaux Catalytiques et Catalyse en Chimie Organique, UMR 5618 CNRS/ENSCM, 8 rue de l’Ecole Normale, 34053 Montpellier
Cedex, France; and †Centre de Recherches Rhoˆne Alpes, ATOCHEM, 2 Rue H. Moissan, BP 20, 69310-Pierre Benite, France
Received June 27, 1994; revised June 14, 1996; accepted June 21, 1996
ture is rather scarce in this field, except the pioneer work
of Deeba et al. (7, 8).
The reaction of 1-butene and isobutene with ammonia has been
investigated, far from thermodynamic equilibrium, in the pressure
range 1–6 MPa over a series of acidic zeolites. The kinetics are
compatible with a Langmuir–Hinshelwood mechanism involving
adsorbed species. The rates of amination increase with the Si/Al
ratio of the solid. A small influence of the zeolite structure is noticed
on the relative adsorption coefficients in the case of 1-butene but
not in that of isobutene. The catalytic activity calculated per proton
is higher on MFI than on BEA or HY zeolites, but this effect of the
structure is less than an order of magnitude. Under the conditions of
reaction used in this work large pore zeolites show a good resistance
to deactivation. It is proposed that deactivation is mainly due to
the formation of strongly basic polyalkylamines and not to coke.
Many important points are still obscure, in particular the
influence of zeolite structure and composition on the cata-
lytic properties and deactivation. For ethylene amination
at 643 K under pressure, Deeba et al. (7) reported a lin-
ear correlation between the yield measured for a series of
small-pore zeolites and the number of sites chemisorbing
ammonia above 473 K, thus a constant turnover frequency.
Owing to the larger critical dimensions of propylene and
isobutene and their reaction products, the dependence for
catalytic activity for the production of higher amines on
acidity was difficult to deduce. By contrast, for the am-
ination of isobutene at atmospheric pressure and in the
presence of water, Mizuno et al. (9, 10) recently reported
an increase of the turnover frequency with the Si/Al ratio
for a series of zeolites including MFI, MOR, and FAU, in-
dependently of the zeolite structure. Catalytic activity was
controlled by the chemical composition (Al content of the
solid) with a maximum observed on MFI for a Si/Al ratio of
80, ascribed to an increase of the acid strength at this high
Si/Al ratio.
In the reverse reaction of deamination of sec-butylamine
a small effect of the structure was observed on the rate of
deamination, with a higher activityfor pentasil zeolites(11).
Concerning the kinetics of amination, Deeba et al. (7) pro-
posed a competitive adsorption of the reactants, whereas
Mizuno et al. (9) report that ammonia saturates the sur-
face and isobutene is weakly adsorbed. The real influence
of pore size on deactivation is not very clear: Deeba and
Ford (8) and Ho¨lderich et al. (2–4) hypothesized that de-
activation was related to coke formation from olefins. This
reaction could be shape selective, since in the temperature
range used for the reaction aromatics can be produced by
cyclisation and hydrogen transfer. In the case of small pore
zeolites (MFI, FER) these aromatics are excluded from the
structure as proposed by Rollman and Walsh (12). This hy-
pothesis accounts for the faster deactivation of HY zeolites
compared to pentasils or offretite. However, Deeba (13)
later reported an increased stability of dealuminated FAU
zeolites, which are known to show a significant proportion
of mesopores.
c
ꢁ 1996 Academic Press, Inc.
INTRODUCTION
The direct amination of olefins with ammonia in the pres-
ence of zeolites is an attractive route for the synthesis of
primary and secondary alkylamines and is one which has
found industrial application, since BASF uses it to pro-
duce t-butylamine (1). High selectivities, about 95% for
monoalkylamines, have been reported for alumino- (2),
ferro- (3) and boro-pentasils (4), and a series of other zeo-
lites (5–6). The addition of ammonia to olefins is an exother-
mic reaction, and the yield therefore decreases with increas-
ing temperature (7). Moreover, both ammonia and olefins
are nucleophiles, and due to the low reactivity thermody-
namic equilibrium is reached at relatively high tempera-
tures. Both kinetic and thermodynamic considerations then
require high pressure, and in some cases pressures as high
as 30 MPa have been used to reach yields of 13% in the
case of isobutene (2–4). Under these conditions amination
is very fast since thermodynamic equilibrium is reached,
and any kinetic study is then impossible. All these technical
problems probably explain why most of the results are to
be found in the patent literature, and why the open litera-
1
Present address: Institut de Recherches sur la Catalyse, 2 Avenue
Albert Einstein, 69626 Villeurbanne Cedex, France. Tel: (33) 72 44 54 69,
Fax: (33) 72 44 53 99, E-mail: figueras@catalyse.univ-lyon1.fr. Proofs to F.
Figueras at the Villeurbanne address.
255
0021-9517/96 $18.00
c
Copyright ꢁ 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

256
LEQUITTE ET AL.
It appeared then of interest to investigate the reaction of
TABLE 1
amination of isobutene on a series of zeolites and to com-
pare the respective influence of the chemical composition
and structure on activity, selectivity, and deactivation.
Chemical Composition and Pore Volumes of the Catalysts
Used Here
Micropore Mesopore
Acidity
vol.
(ml ꢃ gꢂ1
vol.
(ml ꢃ gꢂ1
>523 K
(meq/g)
EXPERIMENTAL
Sample
Source
Si/Al
)
)
Catalyst Preparation
HY2.5
HY15
MOR6.9 Zeolon 100H
Linde LZ62
ZF515
2.5
15
6.9
14
0.36
0.34
0.22
0.17
0.26
0.04
0.14
0.014
0.03
0.21
1.5
0.48
0.88
0.45
0.85
All the samples used in this work were pure single phase
zeolites, with no extraframework aluminum. Zeolite beta
(BEA) was synthesized using tetraethylammonium hy- BEA
droxide (TEAOH) as organic template, following the pro-
cedure described by Nicolle (14). The crystals appeared as
MFI
CBV3010
Home made
15
operated in the pressure range 1–10 MPa. The reactor was
an inox tube (length 80 mm and diameter 21 mm) fitted with
a fritted inox disk on which the catalyst was laid. A well en-
abled the measurement of the temperature in the center of
the catalyst charge. Preheating was made in a tube 300-mm
long and 8-mm diameter maintained at the reaction tem-
perature. The zeolites were used in their powder form, and
the size of the grains was that of the aggregates of crystals
(a few microns). A quantity of 2 g of catalyst was first reac-
tivated in situ in flowing air at 773 K for 6 h, and then the
temperature was decreased to the desired value. Both reac-
tants were stored as liquids in Grayel bottles and metered to
the reactor as liquids using Gilson HPLC cryostated pumps.
The liquids were vaporised in heated pipes before the re-
actor. The pressure in the reactor was controlled by a back
pressure regulator (Kammer valve).
spheroids with an average size of 0.6 ꢀm. These samples
were calcined at 773 K for 8 h under dry air (120 ml minꢂ1
)
and were then converted to the ammonium form by ion
exchange in a 1 M NH4NO3 solution at 373 K.
A ZSM-5 sample (MFI) with a silicon to aluminum ratio
of 14 was obtained from Conteka. A large port mordenite
(MOR) with a Si/Al ratio of 6.9 was obtained from Norton.
The ammonium forms of all samples were activated accord-
ing to the procedure described above. Mordenite consisted
of polycrystalline aggregates, with an average particle size
of about 2 ꢀm.
The Y zeolites (FAU) were Linde LZ-Y62 (Si/Al = 2.5),
a sample obtained from this zeolite by Ce3+ exchange (per-
centage exchange 70% ) and a dealuminated sample ZF515
obtained from Zeocat.
Catalyst Characterization
The composition of the effluent was determined by an
on-line gas chromatograph equipped with two capillary
columns (30 m ꢄ 0.75 mm ID) mounted in series: the first
one was a polar CP Wax 51 for the separation of amines and
the second one an apolar DB1 for the separation of olefins.
The temperature program for chromatographic analysis
was 6 min at 60ꢅC, and then a ramp up to 200ꢅC (10ꢅ/min).
All connecting lines were heated to avoid condensation and
adsorption of the reactants and products.
In a typical experiment, the reactants were introduced at
the reaction temperature. When the desired pressure was
reached the analysis was started, and this instant was con-
sidered as time zero for deactivation.
The chemical composition of the catalysts was deter-
mined by atomic absorption analysis after dissolution of
the sample (SCA-CNRS, Solaize, France). The crystallinity
of zeolites was checked by X-ray powder diffraction on
a CGR Theta 60 instrument using Cu Kꢁ1 filtered radia-
tion. Diffraction patterns showed a good crystallinity of the
samples.
The size and morphology of the crystals were determined
by electron scanning microscopy. The micropore volume
and BET surface area were deduced from the isotherms of
adsorption of nitrogen at 77 K.
Acid properties were determined by stepwise thermal
desorption of ammonia. The catalyst was first activated in
flowing nitrogen in situ, saturated with ammonia at 373 K,
then swept by a flow of dry nitrogen while the tempera-
ture was raised by steps of 50 K. The amount of ammo-
nia evolved by the solid was monitored by conductometry.
The chemical compositions, the pore volumes, and the acid
properties of these catalysts are reported in Table 1.
RESULTS
1. Preliminary studies. Several zeolites were compared in
standard conditions (molar ratio ammonia/olefin = 1, total
pressure 4 MPa) at different temperatures. The results are
reported in Figs. 1 and 2. The major product of the reaction
ofisobutene (Fig. 1) istert-butylamine, which isformed with
high selectivity at low temperature (Fig. 3). In the condi-
tions of reaction (4 MPa), the thermodynamic equilibrium
Catalytic Tests
Isobutene and ammonia of purity more >99.9% were at 523 K is about 20% conversion into tert-butylamine and
supplied by Aga. Tests were performed in a flow reactor therefore does not control the rate. At high temperature

AMINATION OF BUTENES
257
FIG. 3. Amination of isobutene; selectivity to tert-butylamine as a
function of temperature for different zeolites, using a total pressure of
4 MPa and a ratio ammonia/olefin equal to 1, and effect of the ratio
olefin/ammonia on HY2.5.
FIG. 1. Amination of isobutene over protonic zeolites: yield as a
function of temperature for different zeolites. Total pressure 4 MPa;
ammonia/olefin = 1.
a conversion at equilibrium of 3.43% at 593 K, 2.52% at
613 K, and 1.76% at 633 K. The equilibrium is then reached
with many zeolites.
At constant temperature the selectivity decreases with
the conversion on mordenite, but remains high on MFI
(Fig. 5). The decreased selectivity is due to a secondary
reaction of the monoalkylamine, which can be either the
alkylation of 1-butene by sec-butylamine, or the dispropor-
tionation to dialkylamine and ammonia, previously shown
to be fast (9). The former study of the conversion of sec-
butylamine on protonic zeolites showed that both reac-
tions can occur; however, since mordenite and MFI were
very selective for the decomposition of sec-butylamine
into butene and ammonia, the consecutive alkylation of
butene accounts for the decreased selectivity. BEA, on the
other hand, was rather selective for disproportionation of
the selectivity decreases, which is indicative of secondary
reactions.
In the conversion of 1-butene (Fig. 2), sec-butylamine is
the major product. The selectivity is usually lower in this
case and decreases at the higher temperatures by forma-
tion of di-sec-butylamine and olefinic oligomers (Fig. 4).
1-Butene is also isomerized to 2-butene in the course of the
reaction. The position of the thermodynamic equilibrium
depends on the degree of isomerisation of n-butenes, and
is then more complex to compute: if we assume that the
equilibrium between the n-butenes is reached, we obtain
FIG. 2. Amination of 1-butene; yield as a function of temperature for
several zeolites, at a total pressure of 4 MPa and a stoichiometric ratio
ammonia/olefin = 1.
FIG. 4. Selectivity of the amination of 1-butene into sec-butylamine
as a function of temperature for different zeolites, using a ratio
ammonia/olefin = 1, total pressure = 4 MPa.

258
LEQUITTE ET AL.
TABLE 2
Chemical Analyses, in Terms of C/N Ratios of the Catalysts
after the Conversion of 1-Butene and Isobutene
Zeolite
HY2.5
HY15
MOR6.9
MFI14
BEA15
C/N 1-butene
C/N isobutene
0.3
0.14
5.7
0.8
2
0.2
1.5
0.6
9.2
3.2
3. Kinetics of the reaction. At high ammonia pressure the
rate of amine formation is proportional to the partial pres-
sure of olefin, as illustrated in Fig. 7 for 1-butene and Fig. 8
for isobutene. Since when the pressure is changed, the new
stationary state is reached after a certain time, only the
points corresponding to longer times for the equilibration
were selected in that case. Moreover, at the higher pres-
sures of 1-butene deactivation increases; pressure cannot
then be changed in a large range and it can be expected
that the points at high pressure fall below the line. The se-
lectivity to monoalkylamine increases with the partial pres-
sure of ammonia (Figs. 3 and 9) in the case of HY zeolite,
but it remains relatively constant in the case of MFI. This
difference reflects changes of the adsorption coefficients of
butylamines and ammonia related to the different acidities
of the solids.
FIG. 5. Variations of the selectivity as a function of conversion for the
amination of 1-butene on mordenite at 607 K and MFI at 499 K. The total
pressure was 4 MPa.
sec-butylamine, which could be the secondary reaction ob-
served here.
In these experimental conditions, the rate is independent
of the flow of reactants, and therefore is not controlled by
external mass transfer. Typical values of the rate of amine
formation are 10ꢂ7 mol secꢂ1 ꢃ gꢂ1 over most zeolites, with
average crystal sizes of about 1 ꢀm, using a concentration
of substrate of 10ꢂ4 mol mlꢂ1
.
The criterion of Weisz (15),
dN 1 R2
The mechanism proposed by Deeba et al. (7) involves the
attack of the olefin forming a ꢂ-complex with surface pro-
tons and ammonium ions. In that case the kinetics should
follow a Langmuir–Hinshelwood competitive mechanism,
which can be written
< 0.1,
dt C0 Deff
predictsthat the kineticsare then controlled bythe chemical
process and not by diffusion in the pores.
2. Deactivation. The stability of catalytic activity in
isothermal conditions has been checked for BEA and MFI
zeolites using isobutene as reactant. The results are repre-
sented in Fig. 6. After reaction the catalyst is grey or black
and has, therefore, suffered some coke laydown. The car-
bon and nitrogen analyses of the solids after reaction are
reported in Table 2 in terms of C/N ratios and show the
presence of significant amounts of nitrogen on the solid.
ꢃbut
P
butꢃ
P
NH3 NH3
r = k
.
[1]
2
(1 + ꢃbut Pbut + ꢃNH PNH
)
3
3
In ammonia-rich mixtures, the rate equation is reduced
to
r = k(ꢃbut/ꢃNH (Pbut/PNH )),
which agrees with the first order observed in these experi-
3
3
mental conditions.
FIG. 6. Dependence of catalytic activity as a function of time
for BEA and MFI zeolites of comparable composition using a ratio
FIG. 7. Amination of 1-butene: variation of the rate of formation of
ammonia/olefin = 3, total pressure = 4 MPa. Temperatures of reaction: sec-butylamine as a function of the partial pressure of olefin. Total pres-
543 K for BEA and 525 K for MFI.
sure: 4 MPa, Temperature: 591 K.

AMINATION OF BUTENES
259
FIG. 8. Amination of isobutene: variation of the rate of tert-
butylamine formation as a function of the composition of the feed, at
a constant total pressure of 4 MPa. Temperature: 499 K.
FIG. 10. Linearisation of the Langmuir–Hinshelwood equation as
a function of the ratio olefin/ammonia using the results obtained for
1-butene with MFI at 592 and 569 K, and Y zeolites at 591 K.
In the general case, it can be assumed that the surface is
covered either by ammonia or butene. Eq. [1] then becomes
only the points at low temperature, far from equilibrium.
The values for the two reactions are reported in Table 4.
The activation energy determined for the deamination
of sec-butylamine was 184 kJ/mol on HY2.5 zeolite, and
167 kJ/mol on mordenites (9). In principle the difference
between the activation energies of the two reverse reactions
of amination and deamination should be equal to the en-
thalpy of reaction. The enthalpy of reaction deduced from
thermodynamic tables (16) is ꢂ58.5 kJ/mol and the dif-
ference between the activation energies is ꢂ50 kJ/mol for
MOR and ꢂ59 kJ/mol for HY2.5. A reasonable agreement
is therefore observed, which suggests that these activation
energies are not affected by diffusional limitations.
ꢃbut
P
butꢃ
P
NH3 NH3
r = k
(ꢃbut Pbut + ꢃNH Pbut
)
3
and can be written
s
s
s
ꢀ
ꢁ
1 Pbut
ꢃNH
ꢃbut
Pbut
PNH
3
=
+
.
r PNH
kꢃbut
kꢃNH
3
3
3
A linearized form can then be obtained by plotting
p
(1/r)(Pbut/PNH ) as a function of Pbut/PNH . A reason-
3
3
able fit is then observed for several zeolites, independently
of the total pressure (Fig. 10). From the slope and intercept
at the origin of the linear part, the rate constant k and the
DISCUSSION
ratio of adsorption coefficients ꢃNH /ꢃbut can be obtained
(Table 3).
The apparent activation energies of reaction were de-
duced from the results reported in Figs. 1 and 2 using
3
The results of the kinetic study performed here, far from
equilibrium conditions, support the hypothesis formulated
by Deeba et al. (7) on the mechanism, suggesting a compet-
itive adsorption of the reactants. Since the relative coeffi-
cients of adsorption of the olefin and ammonia are of the
same order of magnitude, the bond between the olefin and
the surface should be rather strong. However, this bond has
TABLE 3
Kinetic Constants Obtained for the Amination of
Butenes of Ammonia
Zeolite
ꢃNH /ꢃbut
k (mol/h ꢃ g)
3
Amination of 1-butene at 591 K
HY2.5
MFI
4.3
1
0.01
0.01
Amination of isobutene at 499 K
1
BEA and MFI
4 ꢄ 10ꢂ3
FIG. 9. Amination of 1-butene. Variation of the selectivity to sec-
butylamine as a function of the composition of the feed at 591 K, total
pressure 4 MPa.
Note. Both reactions were studied at 4 MPa, and a total flow
rate of reactants of 0.133 mol/gcat/h.

260
LEQUITTE ET AL.
TABLE 4
Apparent Activation Energies for the Amination of 1-Butene
and Isobutene on Various Zeolites
Apparent activation energy (kJ/mol)
HY2.5
125
137
CeY
104
129
HY15
67
63
MFI
84
133
BEA
58
220
MOR
117
104
Ea 1-butene
Ea isobutene
to be covalent in order to permit a reaction with ammonium
species. It has been proposed that olefins form not carbe-
nium ions but alkoxide species by adsorption on zeolites
(17–18). The interaction of such a covalent alkoxide with
NH+4 ions is here proposed as the slow step of the reaction.
In that case the reaction requires strong acidity to retain
the alkoxide at the surface.
FIG. 11. Influence of the chemical composition and structure of the
zeolites on the rates of formation of sec-butylamine from 1-butene mea-
sure at 583 K, and tert-butylamine from isobutene measured at 498 K
under a pressure of 4 MPa, olefin/ammonia ratio = 1.
According to Rabo and Gajda (19), the rigidity of the lat-
tice can affect the acid strength since it is well known that
acidity is controlled by the Si–O–Al bond angle (20). Re-
cent quantum chemical calculations demonstrate that de-
protonation of the zeolite lattice leads to large local changes
in geometry that change acidity (21). These calculations
account for the modifications of the infrared spectrum of
NH4Y and HY zeolites related to the relaxation of the lat-
tice when the proton is lost and then for the higher acidity,
at the same Al content, of MFI zeolites which possess a
rigid lattice, compared to more flexible systems composed
of large cages such as BEA or FAU.
Some effect of the zeolite structure on the catalytic prop-
erties is then expected for those reactions catalyzed by pro-
tons, and so it was interesting to investigate the influence
of the structure of the zeolite on the kinetic parameters of
the reaction. Some differences could be expected when the
substrate is changed since the coverage of NH+4 ions could
then change and, thus, the acidity.
This kinetic study therefore suggests a small influence
of the structure of the zeolite when considering k, which
is an average over a large number of experiments on the
same sample, and can then be affected by slow deactivation
with time. The comparison of several zeolites of different
structures with a ratio ammonia/olefin = 1 is reported in
Fig. 11. It appears here also that the chemical composition
of the solid is more important than the structure: at similar
Si/Al ratios MFI and BEA show comparable activities for
the amination of 1-butene or isobutene. Dealuminated Y
has a lower activity for the formation of monoalkylamines
because the secondary reaction of disproportionation is fast
in that case. For the amination of isobutene, HY shows a
low activity in spite of a great number of sites. If we consider
turnover numbers, MFI is the most active zeolite for both
reactions, but the difference with BEA or MOR is within a
factor of 2.
By increasing the butene/ammonia ratio and contact
time, conversions close to the thermodynamic equilibrium
at 548 K can be reached (Table 5). These yields can be com-
pared to those reported by Ho¨lderich et al. (2–4) who used
much higher pressures (30 MPa) and obtained yields of
about 12–13% , corresponding also to the thermodynamic
equilibrium in the same temperature range.
1-Butene is difficult to react and requires temperatures
above 573 K. From the kinetic analysis it appears that
ꢃNH /ꢃbut changes when the zeolite is changed from FAU to
3
MFI. The change of the relative adsorption coefficient must
reflect the higher acidity of MFI, which permits retention
of the olefin at the high temperature required for the reac-
tion. The reaction rate determined from this analysis is the
same for the two zeolites. However, if we take into account
the different numbers of acid sites, the rate constant per
proton (turnover frequency) is higher on MFI, compared
to HY zeolite. The difference, a factor of about 3, is con-
sistent with the results previously reported for the reverse
reaction of deamination.
It is interesting to notice that the rate of amination is
not proportional to the rate of deamination, when these
TABLE 5
Amination of Isobutene: Comparison of the Yields (%) Observed
at 548 K, 4 MPa in the Presence of Different Zeolites
The reaction of isobutene is much easier and proceeds
at a lower temperature. In this case the relative adsorption
Zeolite
HY2.5
BEA15
MFI14
MOR6.9
Ratio NH3/olefin = 1
Ratio NH3/olefin = 3
Selectivity
2.3
3.2
100
4.6
8.2a
90
4.5
8.9
100
4.3
5.6
97
coefficients ꢃNH /ꢃbut are comparable for MFI and BEA
3
zeolites. The specific activities are similar, and MFI appears
to be 2 times more active than BEA when the rate constant
is calculated per proton.
a Temperature 532 K, pressure 4.5 MPa.

AMINATION OF BUTENES
261
tive reactions. In that respect the stability of BEA would be
accounted for by the small size of the particles which favors
desorption of monoalkylamines before their reaction.
REFERENCES
1. Chauvel, A., Delmon, B., and Ho¨lderich, W. H., Appl. Catal. 115, 173
(1994).
2. Taglieber, V., Ho¨lderich, W., Kummer, R., Mross, W. D., and Saladin,
G., DBP 33 26 579 (1985), assigned to BASF AG.
3. Ho¨lderich, W., Taglieber, V., Henning, H., Kummer, R., and Gerhard,
G., DBP 36 34 247 (1987), assigned to BASF AG.
4. Taglieber, V., Ho¨lderich, W., Kummer, R., Mross, W. D., and Saladin,
G., DBP 3 327 000 (Feb. 7, 1985), assigned to BASF AG.
5. Peterson, J., and Fales, H. S., U.S. Pat. 4,375,002 (Feb. 22, 1983) and
EP 0 039 918 (1981), assigned to Air Products.
FIG. 12. Correlation between the rates of amination of 1-butene and
deamination of sec-butylamine on a series of zeolites. Both rates are ex-
pressed in mmol hꢂ1 ꢃ gꢂ1
.
6. Deeba, M., and Ambs, W. J., EP 0 077 016 (1982) and EP 0 101 921
(1983), assigned to Air Products.
7. Deeba, M., Ford, M. E., and Johnson, T. A., “Catalysis 1987”
(J. W. Ward, Ed.), Elsevier, Amsterdam, 1988, p. 221; “Catalysis of Or-
ganic Reactions” (D. W. Blackburn, Ed.), p. 241. Dekker New York,
1990.
8. Deeba, M., and Ford, M. E., Zeolites 10, 794 (1990).
9. Mizuno, N., Tabata, M., Uematsu, T., and Iwamoto, M., J. Catal. 146,
249 (1994).
10. Mizuno, N., Tabata, M., Uematsu, T., and Iwamoto, M., Stud. Surf. Sci.
Catal. 90, 71 (1994).
two rates are measured in the kinetic regime (Fig. 12). At
equilibrium these two rates must be equal, and the princi-
ple of microreversibility imposes that the reaction path is
the same for the two reverse reactions. The mechanism of
Hofmann degradation ofaminescan account for the reverse
reaction of deamination and then for the reaction of amina-
tion. Far from equilibrium, the kinetics should be related to
the reactivity of the two substrates, so this nonproportion-
ality does not contradict the principle of microreversibil-
ity, since it simply implies that the different reactivities of
11. Lequitte, M., Figueras, F., Moreau, C., and Hub, S., Appl. Catal. 84,
155 (1992).
sec-butylamine and 1-butene control the respective rates of 12. Rollman, L. D. and Walsh, D. E., J. Catal. 56, 139 (1979).
13. Deeba, M., EP 0 305 564 (1987), assigned to Air Products.
14. Nicolle, M. A., Ph.D. thesis, University of Montpellier, France,
reaction.
A very important point is the stability of catalytic activ-
ity. In the present work, a reasonable stability was obtained
1991.
15. Weisz, P. B., Adv. Catal. 13, 137 (1962).
for both wide and small pore zeolites, suggesting that coke
formation from olefins is not the main source of deactiva-
tion. Indeed the C/N ratios after reaction suggest that the
surface is not covered by coke only, but at least partly by
N-containing compounds. It is difficult to determine the
chemical nature of the surface species since the solid is
black and opaque in the infra-red. We can note, however,
that the secondary reactions of the monoalkylamines yield
dialkylamines. Secondary alkyalmines are stronger bases
than ammonia and can then block the active sites. In this
hypothesis diffusional limitations would play a decisive role
in the stability of zeolites, since they favor these consecu-
16. Stull, D., Sinke, G., and Westrum, E., “The Chemical Thermodynamics
of Organic Compounds,” Wiley, New York, 1969.
17. Medin, A. S., Borovkov, V. Y., Kazansky, V. B., Pelmenshchikov,
A. G., and Zhidomirov, G. M., Zeolites 10, 668 (1990).
18. Kazansky, V. B., and Senchenya, I. N., J. Catal. 119, 108 (1989).
19. Rabo, J. A., and Gajda, G. J., in “Guidelines for Mastering the Prop-
erties of Molecular Sieves” (D. Barthomeuf, E. Derouane, and W.
Ho¨lderich, Eds.), p. 273–297. NATO ASI Ser. B, Vol. 221, Plenum
Press, New York, 1989.
20. Pelmenshchikov, A. G., Paukshtis, E. A., Stepanov, V. G., Pavlov, V. I.,
Yurchenko, E. N., Ione, K. G., Zhidomirov, G. M., and Beran, S., J.
Phys. Chem. 93, 6725 (1989).
21. van Santen, R. A., de Man, A. J. M., Jacobs, W. P. J. H., Teunissen,
E. H., and Kramer, G. J., Catal. Lett. 9, 273 (1991).