Note: Descriptions are shown in the official language in which they were submitted.
~6~
SUPERACIDIC CATALYSTS FOR THE SYNTHESIS
OF MEl'HYI,-TERT-BUIYL l:TH~E:R (MTBE)
FIELD OF THE INVENTION
s
The present invention relates to a novel catalyst for the selective
synthesis of methyl-tert-butyl ether (MTBE). More particularly, the present
invention relates to the acid form of a Y-zeolite onto which triflic acid
(trifluoromethane sulfonic acid, CF3S03H) has been incorporated, and to a process
10 using this catalyst in the production of MTBE.
BACKGROltJND OF THE INVENTION
Synthetic Y-type zeolites are crystalline alumino-silicate zeolites
15 having a tridimentional pore system. These zeolites are used in many industrial
catalytic ~ea.:tions such as catalytic cracking (Fluid Cracking Catalysis or FCC).
Their peculiar catalytic properties are mainly due to their regular framework with
a pore size of app.u.~illlately 7.4 angstroms, and also to their strong acidic sites.
The shape selectivity of the Y-type zeolite is the catalytic expression
of many factors such as:
(i~ the sieving effect, i.e. the capability of the zeolite to allow
or to prevent the access of the reactive molecules into its pore system. This
selection process is based on the critical diameter of the reactive molecules bycol~lpalison with the width of the zeolite pore openings;
(ii) the reverse sieving effect, i.e. the capability of the zeolite
to allow product molecules having a certain critical diameter to diffuse out of its
pores. Thus, in the case of a product molecule having a diameter exceeding the
pore size of the zeolite, this molecule will have to undergo additional cracking into
smaller molecules before diffusing out of the zeolite; and
(iii) the effect on the reaction intermediates, i.e. the capability
of reaction loci to determine the length and geometry of reaction intermediate
species.
9 ~
In the FCC field, the active sites are usually of very acidic nature.
Such acidic sites are generated by eY~h~nging the sodium ions originally present in
the as-synthesized zeolite for protons and/or other cations such as l~n~h~nillm In
any event, it is now also thought that the acidic character of the acid form of Y-
5 zeolite originates from the ~ransted centres created by the tetrahedral alumini--m
sites, although some Lewis acid sites may be formed by activating the zeolite at high
temperature.
Since lead antiknock additives in gasoline will be banned in most
10 industrialized countries by the end of this decade, octane boosters for fuels such as
light alcohols, methyl-tert-butyl ether (MTBE) and ethyl-tert-butyl ether (ETBE)have been incleasin~ used in gasoline blend. MTBE, particularly, is an excellentcandidate for replacing tetraethyl lead in gasoline. When compared to aromatic
hydrocarbons which can be used to upgrade gasoline, MTBE does not evolve any
15 toxic products of incomplete combustion in engines. Furthermore, MTBE does not
provoke any demixing problem when blended with gasoline, which is not the case
with methanol, for instance. MTBE, which is characterized with spe~ifira~ ns close
to those of gasoline, does not require dramatic modifications in the engine
technology. M~E being an oxygenate having a fairly high octane number,
20 accordingly favors the clean combustion of gasolines and other fuels.
MrBE is currently synthesized industrially from mçth lnol and
isobutene over acidic ion-exchange resin, mostly the Amberlyst 15~M which is actually
a macroreticular cation-exchange resin supplied by Aldrich Co. Even though the
25 performance of the Amberlyst catalyst is good, there are several drawbacks to its use
as a commercial catalyst, such as therrnal instability, acid leaching from the resin
surface, and high methanol/isobutene ratio which requires recycling efforts in the
industrial process. Also, the use of Amberlyst results in the production of ~ignifir:lnt
amounts of C8 II~JI~lI,ons as by-products, which reduce the yield of MTBE.
The ZSM-5 zeolite developed by Mobil Oil avoids such
inconveniences. However, this zeolite is not as acidic as the Amberlyst type resin.
In U.S Patent 4,847,223 and U.S. Patent 4,873,392, it has been shown that by coating
the ZSM-5 type zeolite with trifluoromethanesulfonic acid (triflic acid or TFA), it
35 is possible to enhance the surface acidity of the ZSM-5 type zeolites. Such catalysts
7, ~
which are stable to temperatures of up to 240 - 2~0 ~ C, show c ~I,tional catalytic
activity and product selectivity in the ethanol dehydration and acetone cun~ ionto hydrocarbons, in particular to isobutene.
S ll~ough the use of ~SM~5 type zeolite, when combined with TFA,
reduces the amount of C8 hydrocarbons as by-products when compared to
Amberlyst, the use of ZSM-5 provides a ~ignifir~ntly lower yield of MTBE.
Accordingly, there is a great need for a catalyst for the preparation
of MTBE, which while providing a high yield of MTBE, would at the same time,
reduce the undesired C8 hydrocarbons as by-products to a minimum.
.
SUMMAKY OF THE INVENTION
In accordance with the present invention, there is now provided a
novel selective catalyst useful for the synthesis of methyl-tert-butyl ether (MI~E)
or ethyl-tert-butyl ether (ETBE) prepared by reacting isobutene with methanol orethanol, respectively. More specifically, the catalyst of the present invention
COIupl i;,es from about 0.5 to about 7% by weigh~ of triflic acid (trifluoro-
methanesulfonic acid, CF3SO3H, hereafter referred to as TFA) inc~l~oldted onto
an acid form Y zeolite. The process for the production of the catalyst of the present
invention is very similar to the one described in U.S. Patents 4,847,223 and
4,873,392.
It is preferred that the concentration of TFA be from about 0.5 to
about 7% by weight in the catalyst and more preferably from about 3 to about 5%
by weight. It is also preferred that the triflic acid incorporation be carried out by
a uniform impregnation of TFA onto the Y zeolite particles in the presence of
organic solvent, such as acetone. The resulting solid is heated at 80 - 150 ~ C and
preferably at 120 - 140 ~ C, for a period of time ranging from 1 to 24 hours. In its
commercial use, the catalyst of the present invention is preferably incull~ola~ed in
an inert filler, such as bentonite.
Also within the scope of the present invention is a process for the
synthesis of MTBE and ElBE in the liquid phase (200 psig) at a temperature
-4 -
ranging from about 80 to about 100~C, preferably from 85 to 90~C, said process
comprising the reaction of isobutene with meth~nnl, or ethanol, in the presence of
an acid form Y zeolite having incorporated therein from about 0.5 to about 7% byweight of TFA, preferably from 3 to S% by weight.
IN THE DRAWINGS
Figure 1 illustrates the thermal analysis curves obtained with the H-
Y/TFA (3%) sample.
Figure 2 illustrates the variation of the MTBE yield with the
temperature observed with the commercial Amberlyst 15, the H-ZSM-5 zeolite and
the H-Y zeolite, respectively.
Figure 3 illustrates the variation of MTBE yield with the temperature
observed with H-ZSM-5/TFA (3%) and H~ZSM-5/TFA (4%).
Figure 4 illustrates the high yields of MTBE obtained with the H-
Y/TFA (3%) of the present invention.
Figure S illustrates the variation of C8 by-products yield with the
temperature observed with Amberbst 15, H-Y/TFA (3%) of the present invention,
H-Y, and H-ZSM-5/TFA (3%).
Figure 6 illustrates the effects of the con~act time and the reaction
temperature on the yields of MTBE.
Figure 7 illustrates the yield of ETBE by reaction of isobutene with
ethanol over the H-Y/TFA (3) catalyst.
DETAILED l~ESCRIPl'IC)N AND PREFERRED EMBODIMENTS
The surprisingly high performance of the catalyst of the present
invention in the synthesis of MTBE and ETBE is due to:
i) the cooperative action of the zeolite sites and the triflic
acid layer which leads to higher acid density and strength. It has been found that
triflic acid strongly reacts with the silica-alumina zeolite surface upon heating at 80 -
150 ~ C. The resulting complexed triflic acid is stable at temperatures of up to 240 -
250OC, although pure TFA boils at 161~C under atmospheric pre~
ii) the larger pore size of the Y zeolite (7.4 Angstroms) when
compared to that of the ZSM-5 zeolite (pore size - 5.5 Angstroms), which allows a
higher loading of triflic acid onto the proton richer Y zeolite surface without
diffusion problems, mostly for the relatively bulky isobutene as co-reactant.
Therefore, the 3 wt~o TFA/Y zeolite is much more active than 3 vt% 1 FA/ZSM-5
zeolite. These are the optimum triflic acid loading for these two zeolites in terms
of catalytic performance; and
iii) the shape selectivity of the Y zeolite, which results in the
production of the same amount of MTBE as the Amberlyst 15 but negligible amount
of C8 hydrocarbons as by-products. Even at 65OC, the Amberlyst 15 produces a
signifir~nt amount of C8 hydrocarbons as by-products, thus revealing a lack of shape
selectivity due to the certain openness of the resin structure.
The TFA incorporated Y zeolite catalyst of the present invention
possesses the combined catalytic properties of the two current commercially
available catalysts, Amberlyst 15 and ZSM-5/TFA. It is at least as active as theAmberlyst-type resin, at a relativeb low reaction temperature (~5 - 90 oC), and as
selective as the ZSM-5 zeolite: the production of C8 hydrocarbons as by-productsis extremely low when compared to that of the Amberlyst resin at 65 ~ C and the
catalyst is thermally and chemically very stable.
The H-Y zeolite used in the preparation of the catalyst of the
invention has a pore size of at least 7.4 an~tlul-.s.
The basic system can be a pure Y zeolite in acid form (H-Y~ or a
composite Y zeolite-asbestos material, also called Y chrysozeolite. The latter
composite material is prepared by a multi-step process which includes the partial
leaching of the metallic components, such as magnesium and iron, from the zeolite
chrysotile asbestos fibers, followed by the in situ zeolite crystallization, and ended
by the incorporation of the acid sites into the zeolite lattices (see, for example, U.S.
' ' '
'
5~
Patent 4,511,667, Le Van Mao et al.). The acid form of the Y zeolite may ~e
obtained by heating the LZY-82~M (Linde, a division of Union Carbide) in air at
550 o C for 10 hours.
The preferred catalyst of the present invention is prepared by
intimately mixing a H-Y zeolite having a TFA content of from about O.S to about
7% by weight, with 5 to 50% by weight of an inert filler such as bentonite or
colloidal silica, for example, LUDOX~M, manufactured and sold by DuPont & Co.
Preferably, 20% by weight of inert filler is used. The resulting mixture is made into
a paste by adding a sufficient amount of distilled water. Although the amount ofwater incorporated may range ~rom about 0.5 to about 3 ml per gram of solid used,
1 ml of water per gram of the solid is preferred. The resulting product is then dried
in air at a temperature ranging from 1~0~C to 150~C for 1 to 24 hours. The
resulting solid is then ready to be used as a catalyst for the synthesis of MTBE or
ETBE.
SYNTHESIS OF MTBE AND ETBE
Although the commercial process for synthesizing MTBE is carried
out at 60 - 70OC in liquid phase (pressure ~ 200 psig), gas phase reaction at
atmospheric pressure is currently used for the catalyst development, in order toeasily assess the catalyti- properties of the prepared samples.
For the catalytic testing, a vertically mounted stainless-steel fixed-bed
reactor of 2.5 cm diameter and 30 cm in length was used. The reactor had a
preheating and a reaction zone monitored by two chromel-alumel thermocouples in
a thermocouple well positioned at the center of the reactor. The temperature
controller was a potentiometric, time-proportioning controller, with temperaturecontrol achieved by adjusting power input to each zone. Gaseous isobutene was fed
to the reactor, and its flow rate was monitored by using a gas transducer connected
to a digital mass flowmeter and a gas volume totalizer. Liquid methanol was
pumped, by means of a Milton Roy pump, from a graduated reservoir to the
preheating zone of the reactor currently kept at about 100 ~ C to vapori~e methanol.
Condensable reaction products were collected in a flask maintained at -5 ~ C. The
gaseous products were analysed on-line with a Hewlett-Packard 5790 gas
- 7 -
chromatograph equipped with a FID and a reporting integrator Model HP-3392.
Separation of the gaseous phase was achieved on series-packed columns of 5 m of
squalane on Chromosorb P and 2.5 m of Ca~l~pacl~ (Supelco.), graphite coated with
picric acid. Analysis of the liquid phase was done by using a 50 m PONA HP
capiUary column and a FID using 2,2-dimethylbutane as external standard.
Results from the gas phase synthesis of MTBE were computed as
follows. The two rnain by-products of the reaction were C8 diisobutene (or 2,4,4-
trirnethyl-1-pentene) and its isomer 2,4,4-trimethyl-2-pentene. Trace amounts oftrimers were also c1etecte-1 The production of terbutanol was negligible. The yield
of product(s) i was calculated as ~ollows:
Yj = (Ni/Njso) x lOO(C atom%)
wherein N; and Niso are the number of carbon atoms of product(s) i and of feed
isobutene, respectively.
Finally, the molar ration R (methanol/isobutene) was determined as
the ratio of number of moles of methanol to the number of moles of isobutene in
the feed. The reaction parameters were as follows:
- weight of catalyst = 12 g;
- flow rate of feed isobutene = 0.06 to 0.08 C atom/h;
- R = 0.8 to 1.2;
- temperature = 55 -100 ~ C (+ /-1 ~ C).
For the study of the influence of the contact time on the MTBE (or
ETBE) yield, R = 1.5 - 2.0 and the flow rate of feed isobutene was varied from 0.06
to 1.07 mole/h. The contact tirne, ~ressed in hours, is defined as the reciprocal
30 of the hourly weight space velocity (W.H.S.V.). The latter parameter is the ratio of
the total weight (in g) of reactants (methanol or ethanol ~ isobutene) injected per
hour, to the weight (in g) of catalyst used.
The Amberlyst 15 was purchased from Aldrich Co. and used without
35 any further purification.
2 ~
- 8 -
The following examples are pro~/ided to illustrate the present
invention rather than limiting its scope.
EXAMPLE 1
S H-Y zeolite catalyst
The acid form of the Y zeolite (H-Y) was obtained by heat activation
of the LZY-82~M (ammonium form, from Linde) in air at 550 ~ C for 10 hours.
The final catalyst was prepared according to the following procedure:
the powder acid ~orm of the Y zeolite synthesized above, was mixed with bentonite
(20 wt%) and made into a paste with distilled water, 1 ml of water was used for each
gram of zeolite. Finally, the extrudates were dried at 120 ~ C for 12 hours. Talole 1
reports some physico-chemical and adsorptive properties of H-Y zeolite, as well as
for the catalysts pre~ared in Examples 2-4.
3 :XAMPLE 2
H-ZSM-5 ~eolite catalyst
The ZSM-S zeolite was synthesized in autoclave according to the
known method of Argauer and Landolt (U.S. Patent 3,702,886). The as-synthesized
solid was washed with distilled water until the washing liquid had reached a pH
lower than 9, dried at 120~C for 12 hours, and activated at 550~C in air for 12
hours. The acid forrn (H-ZSM-5) was prepared according to the method described
in Fxample 1 of U.S. Patent 4,847,223. The final catalyst was obtained according to
the method described in Example 1.
EXAMPLE 3
H-Y/TFA catalyst
l he H-Y zeolite (powder ~orm) of Example 1 was treated with triflic
acid according to the following procedure: 0.3 g of tri~ic acid (TFA, 98% from
Fluka Chemie AG) were dissolved in 15 ml of pure acetone. This solution was thenslowly added to 10 g of H-Y powder. The resulting suspension was allowed to stand
and dried in the air at roorn temperature. The solid was washed quickly with S ml
2 ~
9.
of acetone and then heated at 120~ - 140~C in air for 12 hours, resulting in an H-
Y/TFA zeolite having a TFA content of 3%. The final catalyst was obtained by
using the method of preparation of extrudates as described in Example 1.
S EXAMP~E 4
H-ZSM-5 zeollte cat~lyst
The tril1ic acid addition (3 or 4 wt%) onto the H-ZSM-5 zeolite
powder was done according to the procedure of ~xample 3. 0.3 g and 0.4 g of triflic
10 acid were used, respectively. The final catalyst extrudates were obtained by using
the method of preparation as described in Example 1.
ANALYSIS OF ll'HE RESULTS
As shown in Table 1, the TFA loading onto the H-Y and H-ZSM-5
zeolites did not lead to 5ignifi~nt changes in the degree of crystallinity. However,
the surface area and the water/n-hexane adsorption capacities of the resulting
materials were ~ignifi~n!ly decreased. This might be attributed to some reduction
in the zeolite pore size due to the presence of the TFA layer on the zeolite surface.
20 Tablc 1
Some physico chemical and adsolptive properlies of the zeoli~e c31alysts
Zeolite Si/AI TFADegree of Surrace area Adsorplion
loading crysta~linily (Langmuir lvol %)
twt~O)(%) m2/g) n-hexanc waler
H-ZSM-S 18 0 100 360 14.3 10.6
- 2 H-ZSM-5/TFA l3) 18 3 99 209 91 10.9
H-Y 2.5 0 100 576 22.8 35.8
H-Y/TFA (3) 2.5 3 9~ 300 14.9 24.2
The PYictçncp of the two exothermic peaks at temperatures higher
than 240~ - 260~C, as illustrated in Figure 1, indicates that the complexed TFA
species decompose only at temperatures higher than those reported previously.
Referring to l'able 2, and Figure 2, it can be seen that there is a clear
sequence in the catalyst activity for production of MTBE as follows:
- 35 Amberlyst > > H-ZSM-5 > H-Y
5 ~ l~
- 10-
~ABLE 2 Yield of MTBE versus reaction temperature (R = 1.0 and contact
time = 2.5 h). (Maximum) are the ma~dimum MTBE yields as derived from curves
of Figure 2.
C~T~LY8T TEMPER~TURE (oC) MT13E YiELD ~C atom X.)
~mberly~t 1~5 ~7 38.1
hq 4~ . 9
51.8
71 4~.1
74 4~.2
78
~o 28. 8
(Maxlmum) 70 48.B
H- Z 8M~ . 2
19.4
2 . 5
82 ~4 . 4
83 4~. 0
84 40.
8S 3BD 1
34. 4
9 1 ;!S2 ~ S
92 ~3 . 1
tM~x 1 mum) 84 3Ei. 8
~t-Y 70 Q. 8
11.3
19.4
~35 ~. 9
8~1 24.4
9;~ 10.
t Max 1 rnum ~ 86 ~6 .
,
Figure 3 and Table 3 clearly demonstrate that though some increase
in the MTBE was obtained by incorporating TFA onto the H-ZSM-5 zeolite, the
resulting MTBE yield was still much lower than that of the Amberlyst 15.
S Also, increasing the TFA loading to 4 wt% did not ~nprove the
production of MTBE, as shown in Figure 3 and Table 3. On the contrary, a
decrease in the MTBE yield indicates the ~Yi~tence of diffusional problems due to
the TFA layer on the zeolite surface.
?, ~ r3 ~
TABLE 3 Yield of M'l'BE versus reaction temperature (R = 1.0 and contact
time = 2.5 h). (Maximum) are the maximum MTBE yields as derived from cur-~es
of Figures 3 and 4.
C~T~LY8T TEMPER~TURE (oC) MT~E YIELD tC atom '~.)
H-ZSM-5/TF~ t~) 70 19.4
7S 33.5
81 ~7.1
~5 41.2
32.9
100 30.0
tMaxlmum) 8S 41.2
H~Z5M-5/TF~ t4~ 82 9.1
16.5
94 21.B
9~ 20.0
~Maxlmum) 95 21.5
H-YZTF~ t~) 71 18.8
79 ~2.9
4~.5
~1 4~.
~4 46.5
~7 ~9.0
88 41.2
89 ~.4
91 29.4
98 9.4
99 1~.9
t MAX I mUm ) B5 47 . 9
': ' , .
.: .
,
The level of MTBE obtained when H-Y/TFA (3%) used is as high
as tha~ of ~he Amberlyst 15. (Table 3 and Figure 4)
Finally, the production of the diusobutene by-products when H-
S Y/TFA (3%) is used, is much smaller than that of the Amberlyst 15 (Figure S and
Table 4). This is explained by the advantageous shape selectivity of the zeoliteframework of the invention.
3 ~ ~
- 14
TABLE ~ Yield of C8 (dii~obutenes) by-products versus reaction temperature
(R = t.0 and contact ~ime - 2.S h). (Maximum) are the C8 yields as derived from
curves of ~igure 5, at temperatures which proYide the ma7umum MTBE yields
(Tables 2 and 3).
C~T~LYST TEMPER~qTUP~ ~oC) C8 yl ~l d ~C ~,tom %)
~mberly~t 15 ~1 1.9
5.~
7i 4.7
74 6.7
~8 9.~5
81 11.1
~Max I mum) 70 4 .
I~-Y 70 O.b
0.9
1,5
2 . 1
8~ 2.4
9~ 4.7
~Max i mum) 8~ 2. 5
~-Y/TF~ ~3) 71 0.
~'9
82 1.2
8;5 0.9
87 0.
92 i.2
~Max 1 mum) 85 0. 9
It-Z6M-5/TFA ~3) ~5 0, 2
0.;5
B0 0.4
~ 0.:2
83 C).4
0.5
~;2 C) . 3
93 0.;5
~M~x i m~lm) 85 0. 4
2 ~
~ 1s-
Figure 6 and corresponding Table S illustrate the effects of thecontact time and the reaction temperature on Ihe yields of MTBE in the presence
of the H-Y/TFA (3%) catalyst. Therefore, to achieve high yield of hlTBE,
temperature of 85 - 90 ~ C and contact time of at least 2.5 h, must be used.
TABLE 5 Influence of temperature and contact time on the yield of MTBE
(Catalyst = H-Y/TFA (3); R = 1.5 -2.0).
TEMPER~TURE ~oc~ CONT~C:T TIME (h) MTBE YIEL~ ~C a~om ~/.)
0.~ 0.9
0.90 1.4
~.24 1.4
1.8~ 2.5
2.70 ~2,9
~.24 ~9.3
87 0. 20 ~ .
n.30 1.4
0.40 1.~3
0.4~5 ~.5
O.S~O 4.
.~o ~.~
1 ~ 9:5 10. 7
2.40 29.~
2.4~ 4~.9
2.~2 50.7
2. 6~ . 8
~.05 ~59.6
10~ O.;~i7 2.
3. ~
0.70 3.9
0.92 ~,1
.as ~9.3
.~0 ~.2
;!;. OS ~. 2
2 ~
- 16 -
Table 6, which col I e~lJonds to Figure 7, illustrates the yield of ETBE
by reaction of isobutene with ethanol over H-Y/TFA (3%) catalyst of the present
invention.
TABLE 6 Intluence of temperature and contact time on the production of
ETBE (Catalyst = H-Y/ I~A (3); R (ethanol/isobutene = 1.5 - 2.0).
TEMPE~RP~TURE ~oC) Ct)NT~CT TIME ~h) ET~E YIELD ~C atom %)
ql 0.52 l.~i
0.~0 ~.7
~.00 13.2
1. 10 13.0
1. 15 1~.
1.20 12.;5
1.25 11.~5
.~0 1~.0
1.~0 18.7
2.00 18.5
:2.70 2S.
0 2 1 . ;~
;5. 10 ;S~.S
;5. 2S 25. 7
:5.28 22.4
0 22. 0
3.40 21.5
3.S0 ~1,;5
:~5.80 20.4
3.9C~ 22.
105 0. 25 1 . 1
0.60 2.;~
0.85 2.5
1 . 1 0 7 . 0
2.20 19.4
;~. 7~ . 4
2,~35 lS.1
;2.95 12.b
.
- . ~
.
.
- 17-
ln summary, the H-Y/TFA is as active as the Amberlyst 15 (Table
7). However, it is more shape sele~tive, and more thermally and chemically
resistant than the commercial ion-~Y~h~npe resin catalyst.
TABLE 7 Catalyt;c resul~s (MTBE s~nthesis in gaseous phase) obtained with
the commercial Amberlyst 15 and the H-Y/TFA (3) catalyst of the present
invention. (R = 1.0; contact time - 2.5 h).
CAT~LY8T MAX I MUM MTBE Y ~ ELû C8 ~Y-PRODUCT8
~tom '~,) TEMP~R~TURE ~oc) PRODUCED ~T
M~X I MUM
MTBF Y~ELD
~C Qtom %)
~mbsrly~~ 1~ 48,8 70 4.b
H-Y/TF~ ~) 47, 9 85 O. q
