For the determination of the reaction order of the decom-
position of the aminal, a defined amount of catalyst (10 mg
H-CBV 760 or 200 mg of H-BEA 25) was contacted with 5.0 g
of a solution containing different concentrations of aminal in
aniline at 70 ◦C (80 ◦C in the case of H-BEA 25). The amount
of catalyst was chosen to adjust the residence time to achieve a
comparable conversion with the different catalysts. In order to
obtain the reaction order of the PABA decomposition, a defined
amount of catalyst (20 mg CBV 760 or 200 mg of H-BEA 25) was
added to 4.0 g of a solution containing different conc◦entrations
◦
of PABA in aniline at a reaction temperature of 70 C (80 C
in case of H-BEA 25). The initial decomposition rates of
aminal and PABA conversion were derived at conversions below
10%.
In order to investigate the influence of the aniline concen-
tration, reaction mixtures with aniline to aminal ratios ranging
from 3 to 20 have been prepared. 15.0 g of these mixtures were
reacted with 0.79 g of catalyst CBV 760 at 100 ◦C. The compo-
sition of the reaction mixture was monitored by GC. When the
reaction mixture reached a constant composition after 2–4 h,
the respective product distributions were determined.
Scheme 2 Proposed reaction network for the MDA synthesis over
zeolite catalysts.3,6
a beta zeolite (Si/Al 15) with different crystallite sizes to the
reaction. It was shown, that smaller crystallites displayed higher
activity in the reaction, as more acid sites close to the particle
surface are accessible to the reacting molecules.4
Therefore, we decided to explore the kinetics and the reaction
network of MDA synthesis starting from the aminal on various
well characterized zeolite catalysts in order to deduce a reliable
reaction network and establish the overall reaction mechanism
as a first step towards new catalysts.
For GC analysis a Shimadzu GC 2010, equipped with an
Optima 35 MS column (length = 30 m, ID = 0.32 mm, film
thickness 0.25 mm), a FID detector and an auto sampler was
available. A temperature profile beginning at 60 ◦C, hold for
5 min, heating withan incrementof 15K min-1 to◦170 ◦C, holding
for 40 min, heating with 25 K min-1 up to 300 C, holding for
15 min and heating with 25 K min-1 up to 350 ◦C and holding this
temperature for 2 min, was applied. The injection volume was set
to 1 mL with the injection port heated to 280 ◦C and a split ratio
of 50. The instrument was calibrated to 4,4¢-MDA, 2,4¢-MDA,
PABA, OABA and aminal, the response factors for heavier
products were assumed to be identical in first approximation
and were estimated by closing the mass balance of the reaction.
As catalysts, a dealuminated Y-type zeolite (CBV 760,
Zeolyst), a parent Beta-type zeolite (H-BEA 25, Su¨dchemie)
and a set of Na+-exchanged CBV 760 have been tested. The
Na+-exchanged CBV 760 samples were prepared from H-CBV-
760 by ion exchange with NaNO3. Three partially exchanged
samples were prepared by dispersing 6.0 g of CBV 760 in
90 mL of distilled water, containing 0.14 g (1.65 mmol), 0.54 g
2. Methods
The aminal solution, which was used as the starting material
for all further reactions, was prepared as follows. In a 1 L round
bottom flask, 600 mL of aniline (6.58 mol, Sigma, purity ≥99.5%)
was heated to 50 ◦C under vigorous stirring. 100 mL of formalin
(1.32 mol formaldehyde, Sigma, 37 wt% of formaldehyde in
water, stabilized with methanol) was added dropwise. After
addition, stirring was continued at 50 ◦C for 1 h. Water and
methanol were removed in a Rotavapor. Concentration and
purity of the resulting solution of 1 equivalent of aminal in
3 equivalents of aniline, which is ready for use in the test
1
reactions, was verified by H- and 13C-NMR, as well as gas
chromatography.
For the test reaction, 10 mL of the aminal solution was placed
in a three necked round bottom flask equipped with a reflux
condenser and heated to the desired temperature. After the set
temperature was reached, 0.50 g catalyst was added. 100 mL
of sample was taken from the reaction mixture after defined
time intervals and diluted with 0.9 mL of acetonitrile (Sigma,
purity ≥99.5%), containing 1 mL of diphenylmethane (Fluka,
purity ≥99%) per 100 mL of acetonitrile as internal standard.
After removal of the catalyst by filtration through a syringe filter
(Minisart SRC; 0.20 mm, d = 4 mm) the sample was analyzed by
GC.
In order to calculate the activation energies for the key
steps of the reaction network the test reaction was carried
out at temperatures between 60 and 150 ◦C. The initial rates
for the decomposition of the aminal, the formation of PABA
and OABA and the formation of 4,4¢-MDA were determined.
The rate of PABA decomposition at a PABA concentration of
0.10 mol/mol of aniline was applied for the calculation of the
apparent activation energy of the PABA decomposition.
◦
(6.35 mmol) or 0.90 g (10.6 mmol) NaNO3, at 80 C for 16 h
under vigorous stirring. One completely exchanged sample was
prepared according to the same procedure using a 0.2 M NaNO3
solution and repeating the treatment two times. All samples were
washed with 200 mL distilled water and dried over night at 80 ◦C
prior to use.
Nitrogen physisorption isotherms were measured using a PMI
automated sorptometer at liquid nitrogen temperature (77 K),
after outgassing under vacuum at 623 K for 4 h. The apparent
surface area was calculated by applying the Brunauer–Emmett–
Teller (BET) theory to the adsorption isotherms over a relative
pressure range from 0.01 to 0.09. The micropore volumes were
evaluated using the t-plot method9 according to Halsey.10 The
mesopore volumes were determined by the cumulative pore
volume of pores with diameters ranging from 2–50 nm according
to the BJH method.11 Because of the limitations of the PMI
instrument, the isotherms were measured at relative partial
pressures higher than 10-5 p/p0.
150 | Green Chem., 2011, 13, 149–155
This journal is
The Royal Society of Chemistry 2011
©