Note: Descriptions are shown in the official language in which they were submitted.
19
BACKGROUND OF 1~ INVENTION
Field of the Invention
This invention relates to a process of producing
polyamines from polynitriles using a hydrogenation catalyst in
pellet form.
DescriDtion of the Prior Art
It has been found that many hydrogenation catalysts
in pelleted form such as cobalt or nickel when used in the
hydrogenation of polynitriles to corresponding polyamines tend
to disintegrate. During the hydrogenation reaction the
catalyst pellets are swollen or disintegrate into fine
particles or both phenomenon occur. Due to loss of physical
integrity, usefulness of catalyst pellets suffers somewhat in
terms of proper control, particularly in a continuous process
where æuch variables as space velocity, etc. must be carefully
considered and controlled. Specifically, channeling occurs in
the catalyst bed, so there is improper contact of nitrile with
catalyst. Also fine particles sometimes plug the reactor or
reactor lines.
In U. S. Patent No. 3,384,666 a method of inhibiting
catalyst pellet disintegration is set out. Essentially this
method involves use of a sodium, lithium or potassium
hydroxide or alkoxide base. While such expedient use of
caustic stabilizer has been found efficacious, nevertheless
there have been found subsequently to have certain drawbacks
emanating from such use. For example, it has been found that
such a process to be efficiently worked must involve
neutralizing the caustic and filtering off the salt. This, of
course, involves a time consuming, and relatively expensive
additional step. In addition, it was ~iscovered that the
19
caustic reacts with those nitriles which additionally contain
an oxy group in a manner such that the desired amine is not
obtained. That is, undesirable side reactions occur.
Other means of maintaining hydrogenation catalyst
pellet integrity have been discovered. In some cases this
involves use of an amine type stabilizer. However, resort to
such stabilizer while avoiding the above discussed problems
with respect to use of caustic nevertheless still necessarily
adds increased cost to the overall process in terms of
necessary resort to yet another chemical additive in the
catalysis.
Still other proposed solutions to the above problem
are set out in U. S. Patent Nos. 3,427,356; 3,728,284; and
4,007,226.
It would therefore be a considerable advance in the
art if one could discover a specific type of catalyst useful
in hydrogenating polynitriles which did not undergo
disintegration or breakdown during the catalysis whereby the
catalyst pellet form was maintained, and yet no resort to0 extraneous protective chemicals need be sought.
SUMMARY OF T~ INVENTION
In accordance with the invention a method of
hydrogenating polynitriles to polyamines has been discovered.
In its broadest aspect, the invention involves use of a
pelleted co~alt-zinc hydrogenation catalyst usually in oxide
form which is characterized as substantially maintaining its
integral character during said hydrogenation without resort to
inorganic or organic stabilizing agents.
9~19
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
In more detail, the hydrogenation technique of
preparing polyamines from polynitriles via the process of the
invention involves hydrogenation of the polynitrile in
presence of a cobalt-zinc hydrogenation catalyst in pellet
form.
The hydrogenation reaction itself may be run in
presence or in the absence of a solvent. When a solvent is
used, it is preferred that an organic solvent such as an
alcohol be employed. $ypical useful alcohols include
methanol, ethanol, isopropanol, t-butanol, n-propyl alcohol,
and other alcohols, particularly water-miscible alcohols.
The polynitrile to be treated in accordance with the
present invention may be chosen from a wide variety of known
materials of this type. Preferred are di- and tri- nitriles
prepared by reacting acrylonitrile with an amine, polyamine,
polyhydroxy-monoamine, polyhydroxy-polyamine, etc. such as
ammonia, methylamine, piperazine, ethylenediamine,
monoethanolamine, diethylenetriamine, 3-aminopropanol,
methylethanolamine, aminoethylethanolamine, etc. Other
preferred polynitriles are those which additionally contain an
oxy group. Typically these oxynitriles include, for example,
acrylonitrile adducts of polyols such as ethylene glycol, di-
and tri-ethylene glycol, glycerol, trimethylol propane,
butane-1,4-diol, butane-1,3-diol, etc.
The pelleted catalyst used here is a pelleted
cobalt-zinc hydrogenation catalyst which usually consists of
cobalt and zinc in mole ratios ranging from 98:2 to 60:40
expressed as Co and Zn. The metals are usually in oxide form.
-3-
9419
The support or carrier used may be any one inert to
process conditions such as refractory support, charcoal,
silica, alumina and the like which are capable of being
employed with the active hydrogenation catalysts. The methods
S of preparinq such catalysts on supports are well known in the
art.
The hydrogenation reaction itself may be carried out
over a wide range of conditions. Typically, the polynitrile
is hydrogenated in the presence of a catalyst of the class
described at a temperature within the range of from about 70
to about 220C and at a pressure of about 30 to 800 atmospheres
in the additional presence of hydrogen. The reaction
temperature is more preferably 70-150,C with the pressure
more preferably being 500-10,000 psig and most preferably
1000-3000 psig.
In a greatly preferred embodiment, ammonia is also
present during the reaction. The ammonia aids the reaction in
promoting better selectivity to primary amine, and prevents
bimolecular coupling to produce secondary amine formation,
usually unwanted in the reaction. When ammonia is present,
usually there are about from 2 to about 20 moles of ammonia
present per equivalent of nitrile. When hydrogen and ammonia
are used together, the hydrogen partial pressure will usually
amount to from about 60 to about 80 percent of the total
pressure.
The particular space velocity of the hydrogenation
reaction (grams nitrile/hour/cc catalyst) is not critical in
the process. However, we prefer to conduct the hydrogenation
reaction at a velocity of between about 0.5 to about 5 grams
total liguid feed/hour/cc catalyst.
19
The hydrogenation reaction here can be performed in
either a batch or a continous manner, with the latter being
preferred. For this, suitable reactors include either a
closed autoclave resulting in a batch process, or a tubular
S reactor which can be operated in a continuous manner.
The desired polyamine product can then be recovered
from the hydrogenation reaction media by any technique known
in the art, such as by distillation. Thus, usually the
polyamine product must be separated from the amine stabilizer
by distillation when the latter is used in amounts such that
it also acts as a solvent for the hydrogenation reaction.
It is interesting to note that only when polyamines
are derived from polynitriles is the problem of catalyst
instability caused during the hydrogenation step.
Mononitriles when hydrogenated are not subject to this
drawback whatsoever. Even dinitriles such as adiponitrile or
phthlonitrile, when converted to diamines cause only minimal
catalyst problems of stability. However, when di- tri- or
tetranitriles derived from reaction of acrylonitriles and
compounds reactive with said acrylonitrile such as alcohols
and amines are hydrogenated, most catalysts such as of the
cobalt type are physically degrades during the hydrogenation
reaction.
Further, as just noted and as will be more clearly
seen in the examples below fixed bed cobalt catalysts alone,
while useful in converting polynitriles to polyamines, are
distinctly deficient in that the pellets or tablets are
converted to fines when used for a substantial time in the
hydrogenation process. Yet a cobalt-zinc oxide mixture is
stable in terms of being able to be maintained in pellet form,
even in absence of a caustic or amine stabilizing additive.
91~19
While the invention has been described in terms of
using catalysts in pellet form, it is understood that the term
"pellet" also is meant to include tablets with such physical
forms of the cobalt-zinc containing catalyst being
interchangeable for use in the invention.
EXAMPLE
A solution of 488 g. (1.68 moles) of cobalt nitrate
hexahydrate and 36 g. of zinc nitrate hexahydrate (0.12 mole)
in two liters of distilled water was heated to 75C. This
solution was added slowly with stirring to 197 g (1.85
moles) of sodium carbonate in three liters of distilled water,
also at 75C. After addition was complete, the slurry was
stirred one hour longer at 80C, then filtered. Collected
solids were washed eight times with two liters of hot
lS distilled water, dried at 110C and calcined for two hours at
400C. The mixed cobalt and zinc oxides, wt. 142 g.,
contained 66.5% Co, 5.1% Zn and 513 ppm. of sodium. This
product was charged to a glass tube and prereduced at 325C
with hydrogen until no moisture could be detected in the exit
gases. After the catalyst was cooled to room temperature in
situ ~nder nitrogen, it was back-oxidized carefully with
dilute oxygen in nitrogen. The stabilized cobalt-zinc oxide
catalyst was blended with 0.5% graphite before being formed
into 1/8" diameter cylindrical pellets.
To a three liter flask containing 600 g. (10.0
moles) of ethylenediamine was added 1855 g. (35.0 moles) of
acrylonitrile at 35-50C with external cooling. After
addition was complete, the solution was placed in a stirred
autoclave with 60 g. of a commercial silica, alumina catalyst
(Aerocat Silica Alumina TA, American Cyanamid Company) and
19
heated at 160C for four hours. This mixture of
cyanoethylated ethylenediamines was used as a feed in the
experiments described here and in Example II.
Twenty-five ml. of pellets of the cobalt-zinc oxide
catalyst whose preparation is described above was placed in a
small fixed bed reactor. A liquid feed composed of egual
weights of liquid ammonia, methanol and the 3.5/1 mole ratio
cyanoethylated ethylenediamine was passed up through the
catalyst at a rate of 24 m./hour. ~ydrogen was also
introduced at 12 liters/hour STP. ~ressure was maintained at
2500 psig and reactor temperature at 115C. Samples of liquid
product were collected periodically and analyzed for
completeness of reaction by infrared nitrile determination.
After 53 hours of operation, the catalyst was removed from the
reactor and the pellets were observed to be unchanged in
appearance. A sample of liquid product was passed through a
wiped film evaporator before analysis by gas chromatography
and contained 54% tris(aminopropyl)ethylenediamine and 9%
bis(aminopropyl)ethylenediamine.
EXANPLE II
A commercially manufactured catalyst prepared from a
75:23:2 atomic ratio of cobalt, copper and chromia was used in
this experiment. To the same fixed bed reactor used in
Example I was charged 25 ml. of 1/8" tablets. Using the same
feed, hydrogen flow, pressure and temperature as in Example I,
the run was terminated after 20 hours because the catalyst
pellets had completely disintegrated and plugged the reactor.
EXAMPLE ~II
Commercially produced cobalt, 1/8" pellets was
charged to a 25 ml. capacity fixed bed catalytic reactor. A
-7-
9'~19
3.5/1 mole ratio acrylonitrile-ethylenediamine reaction
product in a 1:1:1 weight ratio methanol-liquid ammonia-
polynitrile solution was passed over the catalyst with
hydrogen and gave relatively high conversion in the
S hydrogenation except near the end of the 55 hour run. ~owever,
the catalyst, when removed from the reactor, was found to be
composed of broken pieces of the tablet, no longer capable of
being used.
EXAMPLE IV
To a solution of 197 g. (1.85 moles) of sodium
carbonate in three liters of distilled water was added 430 g.
(1.48 moles) of cobalt nitrate hexahydrate and 101 g. (0.34
mole) of zinc nitrate hexahydrate in two liters of distilied
water at 75C. Stirring was continued for an hour at that
temperature followed by filtration and washing eight times
with hot distilled water. The oxides were dried at 110C and
calcined for two hours at 400C to provide 144g. of product,
56.9% Co, 12.8% Zn, 142 ppm of sodium. It was prereduced with
hydrogen at 325C, stabilized with dilute oxygen in nitrogen
and pelleted with 0.5% graphite. The tablets were put in the
25 ml. fixed bed reactor and subjected to a 1~1 hour
continuous run. The nitrile employed was made by reaction of
acrylonitrile with ethylenediamine in a 1.5:1 molar ratio. It
was passed over the catalyst in a solution composed of equal
parts by ~eight of nitrile, methanol and anhydrous ammonia at
a rate of 24 ml. per hour accompanied by 12 liters ~STP) per
hour of hydrogen. Gas chromatographic analysis of the
product, free of methanol and ammonia, gave values of 64%
bis(aminopropyl)ethylenediamine and 33%
aminopropylethylenediamine. The catalyst pellets were
physically unchanged.
-8-
9~19
EXA~LE V
Here an experiment was carried out, using the
commercial nitrile hydrogenation catalyst described in Example
II and the same l.S:l mole ratio acrylonitrile-ethylenediamine
addition product feed of Example IV. Normally, the lower the
acrylonitrile-ethylenediamine ratio feed used, the less the
tendency for catalyst tablets to disintegrate. Under the same
conditions as used in Example IV, the cobalt-copper-chromia
catalyst was used for 85 hours, becoming almost completely
converted to a finely divided mud by the termination of the
experiment.
EXAMPLE VI
Two liters of distilled water containing 22.6 g. of
synthetic calcium silicate was heated to 70C. This was
stirred while 244 g. of Co(N03)2.6H20 and 18 g. Zn (N03)2.6~20
in two liters of distilled water was run slowly in
simultaneously with 155 g. of sodium carbonate in another two
liters of distilled water. All solutions were kept at 7~C,
with the additions adjusted to maintain the pH of the reaction
solution at %0. Following an hour digestion period, solids
were collected by filtration, washed eight times, dried at
110C and calcined an hour at 400C. Analysis of the catalyst
before prereduction was Co 51.3%, Zn 4.1%, Ca 3.7%, Si 4.8%
and Na 532 ppm. Following prereduction at 325C and
sta~ilization, pellets were made with the aid of 0.5%
graphite.
The feed for the fixed bed hydrogenation was
prepared by reaction of 1.88 moles of acrylonitrile per mole
of ethylenediamine. The continuous run was made at the same
rates and conditions as were used for Example I. After 22
_g_
~1~9'~19
hours, the gas chromatographic analysis of the product (free
of ammonia and methanol) was 15% aminopropyl-ethylenediamine,
69% bis(aminopropyl)ethylenediamine and 7.6%
tris(aminopropyl)ethylenediamine. The run was terminated
after 251 hours at which time the corresponding analyses were
15%, 69% and 7.3%. The catalyst pellets did not form finely
divided material although a few of the pellets were split.
From the foregoing description and examples of this
invention, those of ordinary skill in the art may make many
modifications and variations therefrom without departing from
the scope of the invention as hereinafter claimed.
--10--
