Note: Descriptions are shown in the official language in which they were submitted.
-`-` 2 ~ 0
HOE 90/H 030
The invention relates to a process for the simultaneous
prepaxation of acetic acid and acetic anhydride by
reacting mixtures of methanol and methyl acetate and, if
desired, dimethyl ether with carbon monoxide.
Acetic acid and acetic anhydride are important aliphatic
intermediates. Most is used to prepare vinyl acetate or
cellulose acetate.
A process for the simultaneous preparation of acetic acid
and acetic anhydride is disclosed in DE-A-38 23 645. In
this process, the reaction is c~rried out in the presence
of a catalyst system containing carbonyl complexes of
noble metals from Group VIII of the Periodic ~able of the
Elements, in particular of rhodium. The low a~ailability
of rhodium has resulted in an enormously high cost level,
which has an adverse effect on the process. In addition,
the high price requires reco~ery of the rhodium from the
consumed cakalyst solu~ion, which is extremely cost-
inten~ive. The search for a replacement catalyst for
expensive rhodium therefore has a high priority in many
places.
DE-A-26 58 216 describes the preparation of acetic
anhydride by reacting methyl acetate or dimethyl ether
with carbon monoxide in the presence of a catalyst
comprising nickel and chromium in the presence of a
promoter which contains an organonitrogen or
. ' ~ -' ~
~,
-` 2 ~ 7 (~
-- 2
organophosphorus compound containing trivalent nitrogen
or phosphorus. This process gives no acetic acid.
Likewise, the process of DE-C-31 51 371 for the prepara-
tion of acetic anhydride in which nickel i5 employed
together with molybdenum or tungsten as catalyst gives no
acetic acid.
By contrast, DE-C-27 49 954 and 27 49 955 describe
processes for the preparation of acetic acld from methyl
acetate in the presence of nickel as catalyst. These
processes fol~ no acetic anhydride.
The object was to prepare both acetic acid and acetic
anhydride in the same reaction system under anhydrous
conditions. Anhydrous conditions have the advantage of
significantly lower corrosiveness. The process should
provide the opportunity to match the amount ratio of
carboxylic acid to carboxylic anhydride to the particular
economic requirements. Furthermore, the process should
operate at moderate pressures in high space-time yields
and high yields.
The present invention thus relates ko a process for the
si~ultaneous preparation of acetic acid and acetic
anhydride by reacting a starting mixture of methanol and
methyl acetate and, if desired, dimethyl ether with
carbon monoxide in a reaction zone on a catalyst system
under anhydrous conditions, which comprises
: - :
'' ' :- ~, ' :~ '
.
2~a~
-- 3 --
a) working in the presence of a catalyst system com-
prising a non-noble metal from Group VIII of the
Periodic Table, methyl iodide, an alkali metal
compound, organophosphonium compound or organoam-
monium compound as promoter and, if desired, a
compound of a non-noble metal from Groups IV to VII
of the Periodic Table as copromoter;
b) employing a molar ratio between methanol and methyl
acetate and, if used, dimethyl ether of from 10 : 1
~o 1 : 10 as the starting mixture, and
c) carrying out the reaction under a pressure of from
25 to 200 bar and at a temperature of from 150 to
250~C.
The process according to the invention may optionally
have the further features that
aa) the non-noble metal from Group VIII of the Periodic
Table is nickel;
bb~ the catalyst component used is nickel iodide, nickel
chloride, nickel carbonyl, nickel acetylacetonate or
nickel acetate;
cc) the promoter used is an alkali metal compound,
organophosphonium compound or organoammonium com-
pound in the form of its acetate or iodide;
,
." '' ~ '.; ~
~ '
~ .
2~57~
-- 4 --
dd) the alkali metal salt used is a lithium salt;
ee) the organophosphonium compound used is methyltri-
butyl-, methyltriphenyl-, tetrabutyl- or dimethyl~
dibutylphosphonium iodide;
ff) the organoammonium compound used is N,N-dLmethyl-
imidazolium iodide, N-methylpyridinium iodide, N-
methyl-3-picolinium iodide or N-methylquinolinium
.iodide;
gg) the concentration of the non-noble metal from
Group VIII of the Periodic Table in the reaction
mixture is set at between 0.01 and 0.75 mol/l, in
particular between 0.05 and 0.5 mol/l;
hh) the concentration of the methyl iodide in the
reaction mixture is set at between 0.1 and
7.5 mol/l, in particular at be~ween 0.5 and 5 mol/l;
ii) the concentration in the reaction mixture of the
alkali metal compound, organophosphoni~m compound or
organoammonium compound employed as promoter is set
at between 0.05 and 4.5 mol/l, in particular at
between 0.25 and 3.0 mol/l;
jj) the copromoter used is a compound of Ti, Zr, V, Nb,
Ta, Cr, Mo, W, Mn or Re;
~ .
2~5~
., .
~ 5 --
kX) the concen~ration of the copromoter in the reaction
mixture is set at between 5.01 and 0.5 mol/l, in
particular at between 0.05 and 0.3 mol/l;
11) up to 15 % by volume of hydrogen are added to the
carbon monoxide employed;
mm) the reaction is carried out under a pressure of from
S0 to 100 bar and at a temperature of from 175 to
225~C;
nn) the carbonylation process is carried out either
continuously or batchwise.
A particular advantage of the process according to the
invention is that variation of the ratio between methanol
and methyl acetate and, if used, dimethyl ether in the
starting mixture allows virtual:Ly any amount ratio
between carbo~ylic acid and carbo~ylic anhydride to be
produced, so that the process can be matched to rapidly
changing requirements. The carbon monoxide employed for
the reaction need not be absolutely pure. Relatively
small amounts of inert gases, such as carbon dioxide,
nitrogen or methane, do not interfere with the carbonyla-
tion reaction if a sufficiently high carbon monoxide
partial pressure is maintained in the reactor. Hydrogen
contents of up ~o 15 % by volume have a positive effect
on the catalyst activity, but reduce the selectivity of
the process by forming hydrogenation products, such as,
; '~; ' ' ' . .,..,: , ,.: ~''
- ' .' , ~ ' ' :
: , ~, ~: ,', ,':
~ 3 7 ~
for example, ethylidene diacetate.
The catalyst employed can be any non-noble metal from
Group VIII of the Periodic Table. Nickel has the highes~
activity. Any nickel salt which is soluble under the
reaction conditions and fonm5 the active carbonyl complex
can be employed.
Of the alkali metal salts which can be employed as
promoter, salts of lithium have higher activity than
those of sodium or potassium.
The non-noble metals of Groups IV, V, VI and VII of the
Periodic Table which can be employed, if desired, as
copromoter are employed in soluble form, in particular as
acetates, acetylacetonates or carbonyls.
Example 1
3 mol of me~hyl acetate, 3 mol of methanol, 1 mol of
met~yl iodide, 0.1 mol of nickel iodide and 0.4 mol of
methyltributylphosphonium iodide were introduced into a
1 1 stirred autoclave made from stainless stee~
(~astelloy B2). A pressure of 25 bar was then established
by injecting carbon monoxide. The mixture was heated to
the reaction temperature of 195C, and an overall pres-
sure of 80 bar was maintained for ao minutes by con-
tinuously re-injecting carbon monoxide. The autoclave was
cooled and decompressed, and 291 g of acetic anhydride
20~0a 70
(2.85 mol) and 176 g of acetic acid (2.94 mol) were
isolated from 590 ml of reaction product. These values
correspond to yields of 95 %, based on the methyl acetate
employed, and of 98 %, based on the methanol employed.
The space~time yield, based on the liquid reaction
volume, is 594 g of acetic acid + acetic anh~dride/l . h.
Example 2
1 mol of methyl acetate, 5 mol of methanol, 2 mol of
methyl iodide, 0.1 mol of nickel iodide and 0.6 mol of
methyltributylphosphonium iodide were introduced into the
1 l stirred autoclave. A pressure of 25 bar was estab-
lished by injecting a mixture of 95 % by volume of carbon
monoxide and 5 % by volume of hydrogen. The mixture was
heated to the reaction temperature of 195C, and an
overall pressure of 80 bar was maintained fox 51 minutes
by continuously re-injecting the same gas mixture. The
autoclave wa~ cooled and decompressedr and 97 g of acetic
anhydride (0.95 mol) and 297 g of acetic acid (4.95 mol~
were isolated from 640 ml of reaction product. These
values give yields of 95 %, based on the methyl acetate
employed, and of 99 %, based o~ the methanol employed.
The space tLme yield, based on the liquid reaction
volume, is 724 g of acetic acid + acetic anhydride/l . h.
Example 3
5 mol of methyl acetate, 1 mol of methanol, 1 mol of
.. . . . . . . . ......
2 ~ 0
-- 8
methyl iodide, 1 mol of acetic acid, 0.1 mol of nickel
iodide and 0.4 mol of methyltributylphosphonium iodide
were introduced into ~he 1 1 stirred autoclave. A pres-
sure of 25 bar was established by injecting a mixture of
S g2 % by volume of carbon monoxide and 8 % by volume of
hydrogen. The mixture was heated to the reaction tempera-
tuxe of 195C, and an overall pressure of 100 bar was
maintained for 83 minutes by continuously re-injecting
the same gas mixture. The autoclave was cooled and
decompressed, and 479 g of acetic anhydride (4.7 mol) and
118 g of acetic acid were isolated from 734 ml of reac-
tion product. Subtraction of the initial amount of acetic
acid gives 58 g of acetic acid (0.97 mol). These values
give yields of 94 ~, based on the methyl acetate em-
ployed, and of 97 %, based on the methanol employed. The
space-time yield, based on the liquid reaction volume, is
5~9 g of acetic acid + acetic anhydride/l . h.
,
Example 4
Example 4 corre~ponds to Example 1, with the difference
that the methyltributylphosphonium iodide was replaced by
0.2 mol of N,N-dimethylim~dazolium iodide. 294 g of
acetic anhydride (2.88 mol) and 178 g of acetic acid
~2.97 mol) were isolated from 590 ml of reaction product.
Th~se values give yields of 96 %, based on the methyl
25 acetate employed, and of 99 ~, based on the methanol
employed. The space-time yield, based on the liquid
reaction volume, is 600 g of acetic acid + acetic
.
7 0
g
anhydride/l . h.
Example 5
3 mol of methyl acetate, 3 mol of methanol, 1 mol of
meth~l iodide, 0.05 mol of nickel iodide and 0.3 mol of
S lithium acetate were introduced into the 1 l stirred
autoclave. A pressure of 30 bar was established by
injecting a mixture of 92 % by vol~e of carbon monoxide
and 8 % by volume of hydrogenO The mixture was heated to
the reaction te~perature of 195C, and an overall pres-
sure of 100 bar was maintained for 90 minutes by con-
tinuously re-injecting the same gas mixture~ The auto-
cla~e was cooled and decompressed, and 291 g of acetic
anhydride (2.85 mol) and 178 g of acetic acid (2.97 mol)
were isolated from 580 ml of reaction product. These
lS values give yields of 95 %, based on the m~thyl acetate
employed, and of 99 %, based on the methanol employ~d.
The space-time yield, based on the liquid reaction
volume, is 539 g of acetic acid + acetic anhydride/l . h.
Example 6
3 mol of methyl acetate, 0.5 mol of methanol, 2 mol of
methyl iodide, 1 mol of acetic acid, 0.2 mol of ~ickel
acetate, 0.4 mol of methyltributylphosphonium iodide and
0.05 mol of zirconium acetylacetonate were in~roduced
into the 1 l stirred autoclave. A pressure of 25 bar was
established by injecting a mixture of 95 % by volume of
: ' `
2~a~0a~7 ~
-- 10 --
carbon monoxide and 5 % by volume of hydrogen. The
mixture was heated to the reaction temperature of 190C,
and an overall pressure of 100 bar was maintained for a
period of 45 minutes by continuously re-injecting the
same gas mixture. 297 g of acetic anhydride (2.91 mol)
and 90 g of acetic acid were isolated fro~l 609 ml of
reaction product. Subtraction of the amount of ace$ic
acid employed leaves 30 g of acetic acid (O.~9S mol).
These values give yields of 97 %, based on the methyl
acetate employed, and 99 %, based on the methanol ~m-
ployed. The space-time yield, based on the liguid reac-
tion volume, is 716 g of acetic acid + acetic an-
hydride~l . h.
~xample 7
0.5 mol of methyl acetate, 3.5 mol of methanol, 2 mol of
methyl iodide, 0.1 mol of nickel carbonyl, 0.6 mol of
methyltributylphosphonium iodide and 0.5 mol of vanadium
hexacarhonyl were introduced into the 1 l stirred auto
clave. A pressure of 25 bar was established by injecting
a mixture of 92 % by volume of carbon monoxide and 8 % by
volume of hydrogen. The mixture was heated to the reac-
tion temp~ratur~ of 205C~ and an overall pressure of
80 bar was maintained for a period of 30 minutes by
continuously re-injecting the same gas mixture. 49 ~ of
acetic anhydride (0.475 mol) and 208 g of acetic acid
(3.465 mol) were isolated from 532 ml of reaction pro-
duct. These values give yields of 95 %, based on the
2~3~57~
11 --
methyl acetate employed, and 99 ~, based on the methanol
employed. The space time yield, based on the liquid
reaction volume, is 966 g of acetic acid -~ acetic an-
hydride/l . h.
Example 8
Example 7 was repeated with the difference that the
vanadium hexacarbonyl was replaced by 0~05 mol of chro-
mium acetylacetonate. 48 g of acetic anhydride (0.47 mol)
and 205 g of acetic acid (3.43 mol) were isolated from
535 ml of reaction product. These values give yields of
94 %, based on the methyl acetate employed, and of 98 %,
based on the methanol employed. The space-time yield,
based on the liquid reaction volume, is 946 g of acetic
acid + acetic anhydride/l . h.
Example 9
1 mol of methyl acetate, 3.5 mol of methanol, 1.5 mol of
methyl iodide, 1 mol of acetic acid, 0.2 mol of nickel
acetate, 0.4 mol of N,N-dimethylLmidazolium iodide and
0.05 mol Gf dirhenium decacarbonyl were employed in the
1 1 stirred autoclave. A pressure of 25 bar was estab-
lished by injecting a mixture of ~5 % by volume of carbonmonoxide and 5 % by volume of hydrogen. The mixture was
heated to the react on temperature of 135C, and an
overall pressure of 100 bar was maintained for 35 minutes
by continuously re-injecting the same gas mixture. The
, ~
;:~
. , . ,: .
~0370
- 12 -
autoclave was cooled and decompressed, and ~8 g of acetic
anhydride (0.96 mol) and 268 g of acetic acid were
isolated from 535 ml of reaction product. Subtraction of
the amount of acetic acid employed leaves 208 g of acetic
acid (3.465 mol). These values give yields of 96 ~, based
on methyl acetate, and of 99 %, based on methanol. The
space-time yield, based on the liquid reaction ~olume, is
981 g of acetic acid + acetic anhydride/l . h.
Example 10
2 mol of meth~1 acetate, 2 mol of dimethyl ether, 2 mol
of methanol, 1 mol of methyl iodide, 0.15 mol of nickel
acetylacetonate and 0.4 mol of methyltributylphosphonium
iodide were introduced into the 1 1 stirred autoclave. A
pressure of 30 bar was establishecl by injecting carbon
monoxide. The mixtuxe was heated to a reaction tempera-
ture of 195C, and an overall pressure of 100 bar was
maintained for 120 minutes by continuously re-injecting
carbon monoxide. The au~ocla~e was- cooled and decom-
pressed, and 387 g of acetic anhydride (3.79 mol) and
117 g of acetic acid (1.95 mol) were isolated from 590 ml
of reaction product. These values gi~e yields of 94.75 %,
based on the sum of the methyl acetate/dimethyl ether
employed, and 97.5 %, based on the methanol employed. The
space-time yield, based on the liquid reaction vol~me, is
427 g of acetic acid + acetic anhydride~l . h.
2~0~7~
- 13 -
Example 11
3 mol of methyl acetate, 3 mol of methanol, 1 mol of
methyl iodide, 0.2 mol of nickel carbonyl and 0.6 mol of
methyltributylphosphonium iodidQ were introduced into the
1 1 stirred autoclave. A pressure of 30 bar was estab-
lished by injecting carbon monoxide. The mixture was
heated to the reaction temperature of lg5C, and an
overall pressure of 80 bar was maintained for 10 minutes
by continuously re-injecting carbon monoxid~. The auto-
clave was cooled and decompressed, and 61 g of acetic
anhydride (0.6 mol), 177 g of acetic acid (2.~5 mol) arld
176 g of methyl acetate (2.38 mol) were isolated from
685 ml of reaction product. These values correspond to
yields of 96. a %, based on the methyl acetate employed,
and of 98.3 %, based on the methanol employed. The space-
time yield, based on the liquid reaction volume, is
1428 g of acetic acid + acetic anhydride/l . h.
~.
Example 12 (continuou~ proce~s)
The carbonylation was carried out at a temperature of
195C under an overall pressure of 80 bar. The reactor
volume used was 4.5 l. The reaction mixture contained the
nickel complex, methyltributylphosphonium iodide and
methyl iodide in the molar ratio 1 : 3.5 : 10. The
concentration of non-noble metal was 0.3 mol of Ni/l, the
concentration of methyl iodide was 3 mol/l and the
concentration of methyltributylphosphonium iodide was
.
.~ , ...
. ~
:. :
'~
.
7 ~
- 14 -
1.05 mol/l of reaction mixture. 1 kg of methyl acetate
(13.5 mol), 1.85 kg of methanol (57.8 mol) and 11.7 l of
low-boiling components ~methyl iodide and methyl acetate)
were fed hourly into the reactor 2 via line 1 (cf.
Figure). 2 kg of carbon monoxide (71.4 mol) were injected
into the reactor via line 3. 10.0 1 of catalyst solution
weremeteredhourlyintothe reactor 2 via line 4. 26.3 l/h
of reaction product were removed from the reactor 2 via
line 5. This gives a mean residence time in the reactor 2
of approximately 10 minutes. With the volume of the
reaction products kept constant, the reaction products
were removed from the reactor 2 via the decompression
valve 6 by decompression to 1 bar and were passed via
line 7 into the separator 8, in which a mean temperature
of 95C became established. 14.7 kg/h of vapor-form
reaction product flowed out of the separator 8 via line 9
into the low~boiling component column 10; liquid reaction
products from the separator 8 flowed via line 11 into the
separation step 12, where a further 7.60 kg/h of reaction
products evaporated under atmospheric pressure and at
1~5C and were passed into the low-boiling component
column 10 via lines 13 and 9. The catalyst-containing
solution separated of f in liquid form in the separation
step 12 was fed back into the reactor 2 via line 4.
In the low-boiling component column 10, the low-boiling
component methyl iodide and methyl acetate were separated
off under atmospheric pressure at a bottom temperature of
126C and a head temperature of 70C and were fed back
, ' ~ . ' ,
2~30a7~
~ 15 -
into ~he reactor 2 via line 1. 0.08 kg/h of noncondens-
able gases (CO, CO2, CH4 and Nz) was puryed from the
condenser 14.
4.765 kg/h of bottom product from the low-boillng com-
S ponent column 10 were remo~ed via line 15, and 3.4 kg/hof pure acetic acid (56.7 mol) were removed via line 20
in the column 16, which was operated under a pressure of
150 mbar at a head temperature of 70C and a bottom
temperature of 99C. This corresponds to a yield of 98 %,
based on the methanol employed.
The bottom product from the column 16 was removed via
line 17 and fractionated in .he column 18, which i.6
operated under a pressure of 150 mbar~ 1.29 kg/h of pure
acetic anhydride (12.65 mol) were obtained via line 21
at a head temperature of 90C and a bottom temperature
of 104C. This corresponds to a yie]d of 93.7 %, based on
the methyl acetate reacted.
0.075 kg/h of high-boiling components was removed via
line 19 as the bottom product from the column 18.
.~
The yield of acetic acid and acetic anhydride corresponds
to 97.1 %, based on the amount of CO employed. The space-
time yield, based on the reactor ~olume utilized, is
1042 g of acetic acid ~ acetic anhydride/l . h.
~. ';' '
., , .
.
~ ' " . '' '""' ' ' . ~ '' ~ '
' ~ '
'"' ~'~ ' ' ' '
