MTBE Reaction in a Boron Pentasil Zeolite
J. Am. Chem. Soc., Vol. 119, No. 18, 1997 4259
as an octane number boosting additive for gasolines. Zeolites
H-Y,8-10 H-ZSM-5,9,10 and H-BZSM-511 are known to be
suitable for the MTBE synthesis. An overview for this kind of
reaction is given in ref 11. In the present study a boron pentasil
zeolite is used, in which a reversible reaction of MTBE takes
place during several heating cycles. The details of MTBE
formation in zeolites are still not well understood, though some
facts seem to be clear: H-ZSM-5 is very suitable for the high-
temperature formation of MTBE because of its excellent
selectivity toward MTBE.10 The efficient methanol adsorption
decreases the probability of oligomerization by effectively
competing for the acid sites.8,9 The present paper is devoted to
the study of the dynamics under nonequilibrium conditions,
because the adsorption/desorption behavior of the reactants
strongly influences the process.
Figure 2. 1H MAS NMR measurements on a sample containing
hydrated H-ZSM-5 zeolite. Left: dependence of the chemical shift on
the static temperature. Right: dynamic experiment using a 20 s laser
pulse.
Experimental Section
Materials and Samples. The synthesis of the boron pentasil zeolite
was described in ref 12. The zeolite (SiO2/B2O3 ) 40) was pretreated
by heating 8 mm deep layers of zeolite in thin glass tubes with 3 mm
outer diameter at a rate of 10 K h-1 under vacuum. After the samples
were maintained at 400 °C and at a pressure of less than 10-2 Pa for
24 h, they were loaded at room temperature with 8-16 molecules of
MTBE per unit cell. A 1:1 mixture of isobutene (IB) and methanol
13C enriched or deuterated was loaded onto the zeolite for the 13C NMR
1
experiments or for the improvement of the resolution of the H NMR
experiments, respectively. Special care was taken, in order to prove
the reversibility of the temperature-jump relaxation experiments:13 10
temperature cycles precede each NMR experiment. Weak changes of
intensities take place during the first two temperature cycles.
NMR Spectroscopy. The resonance frequencies of the Bruker MSL
Figure 3. 1H MAS NMR spectra of the boron pentasil zeolite loaded
with 8 ( 2 MTBE molecules per unit cell.
1
spectrometer were 300 and 75.4 MHz for H and 13C, respectively.
Figure 2 that the final temperature is reached after 6 s and remains
constant until the laser is switched off after 20 s. One computer controls
the start of both the laser heater and the NMR acquisition.
The spinning rate of the samples in the laser-heated high-temperature
MAS probehead is 1-3 kHz.
The heating was performed by a CO2 laser with a maximum power
of 50 W (wavelength 10.6 ( 0.1 µm). The 3 mm glass ampule was
located in a BN container inside the rotor. A more detailed description
of the experimental setup is given in ref 5. The temperature in the
sample could be measured using a 207Pb (lead nitrate) chemical shift
thermometer.4-5,14-16 The maximum deviation from the mean tem-
perature in the sample is (10 K at 120 °C. The lead nitrate chemical
shift thermometer requires a relatively long measuring time, which
makes it less suitable for dynamic temperature measurements; therefore,
a 1H thermometer (hydrated H-ZSM-5 zeolite in a fused glass ampule)
Results and Discussion
1
Figure 3 shows the temperature dependent H MAS NMR
spectra of the MTBE/zeolite system at room temperature (before
heating), at 80 °C, at 110 °C, and at 140 °C. These spectra
were measured in thermal equilibrium after keeping the samples
for more than 10 min at the given temperature. The spectra
consist of four lines due to CH2 groups of isobutene (4.8 ppm),
-OCH3 groups of MTBE (3.4 ppm), tert-butyl CH3 groups of
MTBE (1.4 ppm), and CH3 groups of isobutene (1.8-2.3 ppm).
The CH3OH signals could not be observed at spinning frequen-
cies of 1-3 kHz. This is not surprising, since for adsorbed
methanol8,9 the correlation time of the motion can be on the
order of magnitude of a rotation period, which causes a drastic
broadening of the MAS NMR signal.17 The change of equi-
librium concentrations upon increasing temperature, cf. Figure
3, shows an increasing IB concentration. This is in agreement
with thermal equilibrium values calculated,10,18 which tend
toward IB and methanol, when the temperature increases. At
room temperature a concentration ratio of MTBE to IB of 3:2
in the sealed sample has been obtained; cf. Figure 3 (top).
In order to determine the mole fraction of one species from
the relative intensity of the corresponding signal in the spectrum,
the total intensity of all signals must be constant, except a weak
change due to the Curie factor. Unfortunately, if the sample is
heated to 140 °C or higher for a longer time, a significant loss
of signal intensity is observed. This can be explained by some
immobile byproducts, most probably oligomers of isobutene
such as the isomers of octene. These can be formed in the
1
was used. The temperature dependence of the H MAS signal was
calibrated in the relevant temperature range by means of the lead nitrate
chemical shift thermometer. Figure 2 shows on the left-hand side the
calibration of the 1H chemical shift thermometer, i.e., the temperature
dependence of the chemical shift of the hydrated H-ZSM-5 zeolite.
On the right-hand side of the figure, the chemical shift in a dynamic
experiment with a final temperature of 140 °C is shown. At the
beginning of the measurement the maximum laser power is used for 5
s, and it is then reduced to an appropriate value. It can be seen in
(8) Kogelbauer, A.; Goodwin, J. G., Jr.; Lercher, J. A. J. Phys. Chem.
1995, 99, 8777-8781.
(9) Kogelbauer, A.; Nikolopoulos, A. A.; Goodwin, J. G., Jr.; Marcelin,
G. J. Catal. 1995, 152, 122-129.
(10) Nikolopoulos, A. A.; Oukaci, R.; Goodwin, J. G., Jr.; Marcelin, G.
Catal. Lett. 1994, 27, 149-157.
(11) Ho¨lderich, W. F. In Proceedings of TOCAT 1, Tokyo, Japan;
Yoshida, S., et al., Eds.; 1990; Catalytic Science and Technology, Kodancha
Ltd., 1991; Vol. 1, pp 31-46.
(12) Ho¨lderich, W. F.; Go¨tz, N.; Hupfer, L., BASF AG. DE 3722891 )
US-PS 4891451, July 10, 1987.
(13) Atkins, P. W. Physical Chemistry, 4th ed.; Oxford University
Press: Oxford, 1990.
(14) Mildner, T.; Ernst, H.; Freude, D. Solid State NMR 1995, 5, 269-
271.
(15) Bielecki, A.; Burum, D. P. J. Magn. Reson., A 1995, 116, 215-
220.
(16) Van Gorkom, L. C. M.; Hook, J. M.; Logan, M. B.; Hanna, J. V.;
Wasylishen, R. E. Magn. Reson. Chem. 1995, 33, 791-795.
(17) Fenzke, D.; Gerstein, B. C.; Pfeifer, H. J. Magn. Reson. 1992, 98,
469-474.
(18) Tejero, J.; Cunill, F.; Izquierdo, J. F. Ind. Eng. Chem. Res. 1988,
27, 338-343.