Note: Descriptions are shown in the official language in which they were submitted.
I
Process for preparing 2-(2-tert-butylaminoethoxy)ethanol (tert-
butylaminodiglycol, TBADG)
Description
The present invention relates to a process for preparing 2-(2-tert-
butylaminoethoxy)ethanol (tert-
butylaminodiglycol, TBADG) by reacting diethylene glycol (DG) with tert-
butylamine (TBA) in the
presence of hydrogen and of a copper catalyst.
One use of the process product is that in gas scrubbing, for example, for the
selective separation of
acidic gases, for example H2S from gas streams which comprise mixtures of one
or more acidic
gases and CO2.
EP 137 478 A2 (BASF AG) relates to a process for preparing N-methylpiperidine
or
N-methylmorpholine by catalytically aminating pentanediols or diethylene
glycol with methylamine
in the gas phase over a copper catalyst which has been obtained by heat
treatment of a basic
copper- and aluminum-comprising carbonate.
EP 235 651 Al (BASF AG) teaches a process for preparing N-methylpiperazine
from
diethanolamine and methylamine over metallic catalysts. The reaction is
performed in the liquid
phase (trickle mode) (page 3 last paragraph). According to the example, a
Cu/A1203catalyst is
used.
EP 816 350 Al (BASF AG) describes processes for preparing N-methylpiperidine
and N-
methylmorpholine by reacting primary amine with a did l over a copper catalyst
which has been
obtained by impregnating SiO2 pellets with basic copper carbonate, in the
liquid or gas phase.
US 4,739,051 A (BASF AG) teaches the preparation of morpholine and piperidine
by reaction of
DEG or pentanediol with ammonia under hydrogenation conditions in the gas
phase at standard
pressure and 200 C over an unsupported Cu/Ni/AI catalyst with yields of 97 and
95%, respectively.
EP 514 692 A2 (BASF AG) discloses processes for preparing amines from alcohols
in the presence
of catalysts comprising copper and nickel and zirconium oxide and/or aluminum
oxide.
DE 198 59 776 Al (BASF AG) relates to the preparation of amines by reacting
alcohols, or
aldehydes or ketones, with amines over a catalyst composed of copper
CA 2798971 2017-06-15
CA 02798971 2012-11-08
PF 70552
2
and Ti02, to which metallic copper has been added before the shaping of the
catalyst
material.
EP 440 829 Al (US 4,910,304) (BASF AG) describes the annination of diols over
copper catalysts, especially the preparation of N-methylpiperidine and
N-methylmorpholine by reaction of pentanediol or diethylene glycol (DEG) with
methylamine and 45% aqueous KOH solution over an unsupported Cu/AI catalyst at
245 C and 250 bar. The reaction is performed in the liquid phase (trickle
mode)
(page 3 last paragraph). Suitable catalysts are the catalysts disclosed in
DE 24 45 303 A (BASF AG), which are obtainable by heat treatment of a basic
copper-
and aluminum-comprising carbonate of the general composition
CumAl6(CO3)0.57,03(OH)m+12, where m is any value, including non-integers, from
2 to 6,
for example the precipitated copper catalyst disclosed in loc. cit., example
1, which is
prepared by treating a solution of copper nitrate and aluminum nitrate with
sodium
bicarbonate and subsequently washing, drying and heat treating the
precipitate.
WO 05/110969 Al (BASF AG) describes a process for continuously preparing an
amine by reacting a primary or secondary alcohol, aldehyde and/or ketone with
hydrogen and a nitrogen compound selected from the group of ammonia, primary
and
secondary amines, at a temperature in the range from 60 to 300 C, in the
presence of
a copper catalyst, wherein the catalytically active material of the catalyst,
before the
reduction thereof with hydrogen, comprises 20 to 85% by weight of aluminum
oxide
(A1203), zirconium dioxide (Zr02), titanium dioxide (Ti02) and/or silicon
dioxide (Si02),
and the reaction is effected in the gas phase isothermally in a tubular
reactor.
WO 2010/031719 Al (BASF SE) relates to a process for continuously preparing an
amine by reacting a primary or secondary alcohol, aldehyde and/or ketone with
hydrogen and a nitrogen compound selected from the group of ammonia, primary
and
secondary amines, at a temperature in the range from 60 to 300 C, in the
presence of
a copper- and aluminum-oxide-containing catalyst, wherein the reaction is
effected in
the gas phase and the catalytically active material of the catalyst, before
the reduction
thereof with hydrogen, comprises aluminum oxide and oxygen compounds of
copper,
and the shaped catalyst body is specified.
US 4,487,967 and US 4,665,195 (both Exxon Res. & Eng. Co.) teach the
preparation
of sterically hindered amino ether alcohols by reaction of corresponding
amines with
diethylene glycol or polyalkenyl ether glycols. The selectivity problem in the
reaction of
TBA with DEG owing to the formation of N-tert-butylmorpholine (TBM) is
described
(US 4,487,967: column 3). The catalysts used are supported and unsupported
metals,
including Ni/A1203/Si02, Ni-Al, Raney Ni, Raney Cu catalysts. In the case of
the copper
catalysts mentioned, the TBADG yield is only 6.4% (US 4,487,967, column 6,
table 1).
In the case of the Ni/A1203/Si02 catalyst, the isolated TBADG yield is only
54% (US
4,487,967, column 5, example 1).
PF 70552 CA 02798971 2012-11-08
=
3
WO 07/021462 A2 (Exxon-Mobil Res. & Eng. Comp.) relates to the use of
dialkylamine
glycols or monoalkylamine glycol ethers in acid gas scrubbing, and the
preparation
thereof by aminating corresponding glycols. Page 15 mentions the preparation
of
TBADG from DEG and TBA over a nickel catalyst in only 30% yield, and also
discusses the selectivity problem (cf. the scheme on page 15).
WO 05/081778 A2 (Exxon-Mobil Res. & Eng. Comp.) describes, inter alia, the
synthesis of TBADG from DEG and TBA over supported metal catalysts, the
support
having specific pore sizes, pore distributions and surface areas (BET).
Preference is
given to using nickel catalysts (page 3, paragraph [0009]). In all examples,
unsupported nickel catalysts are used. Illustrative results are:
Example 6c, page 28, run 27: DEG conversion = 72%, molar TBADG:TBM ratio = 13,
Example 9, page 31,8 h: DEG conversion = 62.5%, molar TBADG:TBM ratio = 15,
and
Example 12, page 37, #170-8: DEG conversion = 51.9%, TBADG:TBM mass
ratio = 15.8, i.e. molar TBADG:TBM ratio = 14.
Since no selectivities are reported here, no yield can be calculated. At a
TBADG
selectivity of 80% (based on DEG), the TBADG yield in the best example
(example Sc,
page 28, run 27) would be approx. 57%.
US 4,405,585 (Exxon Res. & Eng. Comp.) describes the use of strongly
sterically
hindered secondary amino ether alcohols for selective removal of H2S from a
gas
comprising CO2 and H2S. Example 1, in column 9, describes the preparation of
TBADG
from TBA and 2-chloroethoxyethanol.
WO 05/082834 (Exxon-Mobil Res. & Eng. Comp.) describes a process for preparing
sterically strongly hindered amino ether alcohols and diaminopolyalkenyl
ethers by
reaction of a primary amine with polyalkylene glycol at elevated temperature
and
pressure in the presence of a specific catalyst. The catalyst is characterized
in that its
preparation involved decomposition of organic metal complexes on a support.
It has been recognized in accordance with the invention that the reaction of
DG with
TBA over nickel catalysts has the considerable disadvantage from a safety
point of
view that decomposition products of DG form, which cause a critical situation,
for
example, in the case of disrupted operation of the reactor (especially power
failure).
In the amination of DG, for example as a result of decarbonylation, there is
enhanced
formation of undesired components such as methoxyethanol, methoxyethylamine,
methanol, methane (see scheme below). Methoxyethanol is toxic, can be removed
from TBADG only with difficulty owing to its physical properties, and can thus
lead to
problems with regard to specification and product quality.
4
NABu
DG -CO TBADG
'1g
methanol '4¨
methoxyethanol
methane TBA
0 N¨tBu tBuNONABu
TBM
methoxyethyl-tert-butylamine
In order to solve this problem, complex specialty reactors are used in some
cases; cf.,
for example WO 2009/092724 Al (BASF SE), especially page 10, lines 14-21.
It was an object of the invention to remedy the disadvantages of the prior art
and to
discover an improved economically viable process for preparing TBADG. More
particularly, the process should firstly enable high yields, space-time yields
(STY) and
selectivities and secondly suppress decarbonylation of DG and the associated
disadvantages, and hence enable a safe process regime.
[Space-time yields are reported in 'amount of product/(volume of catalyst =
time)'
(kg/(lcat. = h)) and/or 'amount of product/(reactor volume = time)'
(kg/(lreactor = h)].
Accordingly, a process has been found for preparing 2-(2-tert-butylamino-
ethoxy)ethanol (tert-butylaminodiglycol, TBADG) by reacting diethylene glycol
(DG)
with tert-butylamine (TBA) in the presence of hydrogen and of a copper
catalyst, which
comprises effecting the reaction at a temperature in the range from 160 to 220
C in the
presence of the copper catalyst, where the catalytically active material of
the catalyst,
before the reduction thereof with hydrogen, comprises
20 to 75% by weight of aluminum oxide (A1203),
20 to 75% by weight of oxygen compounds of copper, calculated as CuO, and
.µ 5% by weight of oxygen compounds of nickel, calculated as NiO.
In the process, the reaction is preferably conducted only up to a DG
conversion in the
range from 20 to 80%, particularly a DG conversion in the range from 30 to
70%.
Preference is given to effecting the reaction not in purely liquid phase, but
in the gas
phase or gas/liquid mixed phase. Particular preference is given to effecting
the reaction
in the gas phase.
In the case of a reaction in the gas/liquid mixed phase, hydrogen (H2) and DG
are
preferably used in a molar ratio of hydrogen:DG = 5 to 50, preferably
hydrogen:DG = 5
to 30.
CA 2798971 2018-02-15
CA 02798971 2012-11-08
= PF 70552
In a reaction in the gas phase, hydrogen (H2) and DG are preferably used in a
molar
ratio of hydrogen:DG = 40 to 220, particularly hydrogen:DG = 50 to 120.
The molar ratio (MR) of hydrogen:DG can be adjusted via the pressure and/or
dilution
5 with an inert gas, e.g. N2 or Ar.
Preference is given to effecting the reaction in the absence of a solvent.
In the process according to the invention, the catalysts are preferably used
in the form
of catalysts which consist only of catalytically active material and
optionally a shaping
aid (for example, graphite or stearic acid), if the catalyst is used as a
shaped body, i.e.
do not comprise any further catalytically active accompanying substances.
In this context, the oxidic aluminum oxide (A1203) support material is
considered to
belong to the catalytically active material.
The catalysts are used in such a way that the catalytically active material is
arranged in
the reactor after grinding, mixing with shaping aids, shaping and heat
treatment in the
form of shaped catalyst bodies ¨ i.e. in the form of tablets.
The concentration figures (in % by weight) of the components of the catalyst
are based
in each case ¨ unless stated otherwise ¨ on the catalytically active material
of the
finished catalyst after the last heat treatment thereof and before the
reduction thereof
with hydrogen.
The catalytically active material of the catalyst, after the last heat
treatment thereof and
before the reduction thereof with hydrogen, is defined as the sum of the
masses of the
catalytically active constituents and of the abovementioned catalyst support
material,
and comprises essentially the following constituents:
aluminum oxide (A1203) and oxygen compounds of copper, and preferably oxygen
compounds of sodium.
The sum of the abovementioned constituents of the catalytically active
material,
calculated as A1203, CuO and Na20, is typically 70 to 100% by weight,
preferably 80 to
100% by weight, more preferably 90 to 100% by weight, further preferably 98 to
100%
by weight, further preferably 99% by weight, most preferably 100% by weight.
The catalytically active material of the catalysts used in the process
according to the
invention may further comprise one or more elements (oxidation state 0) or the
inorganic or organic compounds thereof, selected from groups I A to VI A and I
B to
VII B and VIII of the Periodic Table.
Examples of such elements and compounds thereof are:
transition metals, such as Ni and NiO, Co and CoO, Re and rhenium oxides, Mn
and
Mn02, Mo and molybdenum oxides, W and tungsten oxides, Ta and tantalum oxides,
Nb and niobium oxides or niobium oxalate, V and vanadium oxides or vanadyl
CA 02798971 2012-11-08
PF 70552
6
pyrophosphate; lanthanides, such as Ce and Ce02 or Pr and Pr203; alkali metal
oxides,
such as K20; alkali metal carbonates, such as Na2CO3; alkaline earth metal
oxides,
such as CaO, Sr0; alkaline earth metal carbonates, such as MgCO3, CaCO3 and
BaCO3; boron oxide (B203).
The catalytically active material of the catalysts used in the process
according to the
invention comprises, after the last heat treatment thereof and before the
reduction
thereof with hydrogen,
20 to 75% by weight, preferably 25 to 65% by weight, more preferably 30 to 55%
by
weight, of aluminum oxide (A1203) and
to 75% by weight, preferably 30 to 70% by weight, more preferably 40 to 65% by
weight, most preferably 45 to 60% by weight, of oxygen compounds of copper,
calculated as CuO,
0 to 2% by weight, preferably 0.05 to 1% by weight, more preferably 0.1 to
0.5% by
15 weight, of oxygen compounds of sodium, calculated as Na20,
s 5% by weight, for example 0.1 to 4% by weight, preferably less than 1% by
weight,
for example 0 to 0.8% by weight, of oxygen compounds of nickel, calculated as
NiO.
The catalytically active material of the catalyst comprises, before the
reduction thereof
20 with hydrogen, more particularly less than 1% by weight, for example 0
to 0.5% by
weight, of oxygen compounds of cobalt, calculated as Co0.
The catalytically active material of the catalyst used in the process
according to the
invention most preferably does not comprise any nickel, any cobalt and/or any
ruthenium, in each case either in metallic (oxidation state 0) form or in an
ionic,
especially oxidized, form.
The oxygen compounds of copper are especially copper (I) oxide and copper(II)
oxide
preferably copper(II) oxide.
The catalytically active material of the catalyst used in the process
according to the
invention most preferably does not comprise any zirconium dioxide (Zr02),
titanium
dioxide (Ti02) and/or silicon dioxide (Si02).
In a particularly preferred embodiment, the catalytically active material of
the catalysts
used in the process according to the invention does not comprise any further
catalytically active component, either in elemental or in ionic form.
In the particularly preferred embodiment, the catalytically active material is
not doped
with further metals or metal compounds.
Preferably, however, typical accompanying trace elements which originate from
the
metal extraction of Cu, and optionally Ni, are excluded therefrom.
CA 02798971 2012-11-08
PF 70552
7
For preparation of the catalysts used in the process according to the
invention, various
processes are possible. They are obtainable, for example by peptizing
pulverulent
mixtures of the hydroxides, carbonates, oxides and/or other salts of the
aluminum,
copper and optionally sodium components with water, and subsequently extruding
and
heat treating the material thus obtained.
The catalysts used in the process according to the invention can also be
prepared by
impregnating aluminum oxide (A1203), which is present, for example, in the
form of
powder or tablet moldings.
Aluminum oxide can be used here in different polymorphs, preference being
given to a-
(alpha), y- (gamma) or 0-A1203 (theta-A1203). Particular preference is given
to using
y-A1203.
Shaped bodies of aluminum oxide can be produced by the customary processes.
The aluminum oxide is likewise impregnated by the customary processes, as
described, for example in EP 599 180 A, EP 673 918 A or A. B. Stiles, Catalyst
Manufacture - Laboratory and Commercial Preparations, Marcel Dekker, New York
(1983), by applying an appropriate metal salt solution in each case in one or
more
impregnation stages, using, as the metal salts, for example corresponding
nitrates,
acetates or chlorides. After the impregnation, the material is dried and
optionally
calcined.
The impregnation can be effected by what is known as "incipient wetness"
method, in
which the inorganic oxide (e.g. aluminum oxide) is moistened with the
impregnating
solution up to a maximum of saturation according to its water absorption
capacity. The
impregnation can, however, also be effected in supernatant solution.
In multistage impregnation processes, it is appropriate to dry and optionally
to calcine
between individual impregnation steps. Multistage impregnation should be
employed
advantageously particularly when the inorganic oxide is to be contacted with a
relatively large amount of metal.
To apply a plurality of metal components to the inorganic oxide, the
impregnation can
be effected simultaneously with some or all metal salts, or in any desired
sequence of
the individual or plural metal salts.
Preference is given to preparing the catalyst used in the process according to
the
invention by employing precipitation methods. For example, they can be
obtained by a
coprecipitation of the components from an aqueous salt solution by means of
mineral
bases in the presence of a slurry of a sparingly soluble oxygen-containing
aluminum
compound, and then washing, drying and calcining the resulting precipitate.
The
CA 02798971 2012-11-08
= PF 70552
8
sparingly soluble oxygen-containing aluminum compound used may, for example,
be
aluminum oxide. The slurries of the sparingly soluble aluminum compound can be
prepared by suspending finely divided powders of this compound in water while
stirring
vigorously. These slurries are advantageously obtained by precipitating the
sparingly
soluble aluminum compound from aqueous aluminum salt solutions by means of
mineral bases.
Preference is given to preparing the catalysts used in the process according
to the
invention by means of a co-precipitation precipitation of all components
thereof. For
this purpose, an aqueous salt solution comprising the catalyst components is
appropriately admixed under hot conditions and while stirring with an aqueous
mineral
base, especially an alkali metal base - for example sodium carbonate, sodium
hydroxide, potassium carbonate, or potassium hydroxide - until the
precipitation is
complete. The type of salts used is generally not critical: since the
principle concern in
this procedure is the water solubility of the salts, one criterion is the good
water
solubility thereof, which is required to prepare these comparatively highly
concentrated
salt solutions. It is considered to be obvious that, in the selection of the
salts of the
individual components, the salts selected are of course only those with anions
which do
not lead to disruption, whether by causing undesired precipitation or by
complicating or
preventing precipitation as a result of complex formations.
The precipitates obtained in these precipitation reactions are generally
chemically
inhomogeneous and consist, inter alia, of mixtures of the oxides, oxide
hydrates,
hydroxides, carbonates, and insoluble and basic salts of the metal(s) used. It
may be
found to be favorable for the filterability of the precipitates when they are
aged, i.e.
when they are left alone for a certain time after precipitation, optionally
under hot
conditions, or while passing air through.
The precipitates obtained after these precipitation processes are processed
further as
usual to give the catalysts used in accordance with the invention. After
washing, they
are preferably dried at 80 to 200 C, preferably at 100 to 150 C, and then
calcined. The
calcination is preferably performed at temperatures between 300 and 800 C,
preferably
400 to 600 C, especially 450 to 550 C.
After the calcination, the catalyst is appropriately conditioned, whether by
adjusting it to
a particular particle size by grinding and/or by mixing it, after the grinding
thereof, with
shaping aids such as graphite or stearic acid, pressing it by means of a press
to the
moldings, i.e. tablets, and heat treating. The heat treatment temperatures
preferably
correspond to the temperatures in the calcination.
The catalysts prepared in this way comprise the catalytically active metals in
the form
of a mixture of the oxygen compounds thereof, i.e. especially as the oxides
and mixed
oxides.
CA 02798971 2012-11-08
PF 70552
9
The catalysts prepared in this way are stored and may be traded as such.
Before the
use thereof as catalysts, they are typically pre-reduced. They can, however,
also be
used without pre-reduction, in which case they are reduced by the hydrogen
present in
the reactor under the conditions of the hydrogenating amination.
For pre-reduction, the catalysts are first exposed to a nitrogen-hydrogen
atmosphere at
preferably 150 to 200 C over a period of, for example, 12 to 20 hours, and
then treated
in a hydrogen atmosphere at preferably 200 to 400 C for another up to approx.
24
hours. This pre-reduction reduces a portion of the oxygen-containing metal
compound(s) present in the catalysts to the corresponding metal(s), such that
they are
present in the active form of the catalyst together with the different kinds
of oxygen
compounds.
The catalyst is preferably characterized by a micropore volume of < 0.5 cm3/g,
particularly < 0.4 cm3/g, for example < 0.3 cm3/g, (measured to DIN 66135-1).
(According to the 1984 IUPAC recommendation, micropores are defined as pores
with
pore sizes below 2 nm: K.S.W. Sing et al., Pure & Appl. Chem. 57 (1985) 4, 603-
619).
In addition, the catalyst is preferably characterized by the following pore
size
distribution: if normalized to pores with a pore size of > 0 to 5 20 nm
(measured to DIN
66134 (for the mesopores, pore size 2 to 5 20 nm) and DIN 66135-1 (for the
micropores)), 530% of the pores have a pore size up to 5 nm and more than 70%
of
the pores have a pore size of > 5 to 20 nm.
The reaction in the process according to the invention is preferably effected
in a tubular
reactor.
The reaction in the tubular reactor by the process according to the invention
is most
preferably effected in a cycle gas mode.
The cycle gas consists of predominantly hydrogen or a mixture of hydrogen and
an
inert gas (e.g. N2) and serves to evaporate the reactants and/or as a reactant
for the
amination reaction.
In cycle gas mode, the starting materials (DG, TBA) are preferably evaporated
in a
cycle gas stream and supplied to the reactor in gaseous form.
The reactant (DG, TBA) can also be evaporated as aqueous solutions and passed
on
to the catalyst bed with the cycle gas stream.
Examples of suitable reactors with a cycle gas stream can be found in
Ullmann's
Encyclopedia of Industrial Chemistry, 5th Ed., vol. B 4, pages 199-238, "Fixed-
Bed
Reactors".
CA 02798971 2012-11-08
PF 70552
The cycle gas rate is preferably in the range from 40 to 2500 m3 (at operating
pressure)/[m3 of catalyst (bed volume) - h], especially in the range from 100
to 2000 m3
(at operating pressure)/[m3 of catalyst (bed volume) = h].
5
The cycle gas into a gas/liquid mixed phase mode comprises preferably at least
10%,
particularly 50 to 100%, and very particularly 80 to 100% by volume of H2.
The cycle gas into a gas phase mode comprises preferably at least 10%, more
10 preferably 20 to 80%, and very particularly 30 to 60% by volume of H2.
The preferably isothermal reaction in the process according to the invention
is effected
preferably with a temperature deviation of not more than +/- 8 C, particularly
not more
than +/- 5 C, especially not more than +/- 4 C, very particularly not more
than +/- 3 C,
for example not more than +/- 0 to +/- 2 C or not more than +/- 0 to +/- 1 C.
These temperature deviations relate to the particular temperatures in the
particular
catalyst bed, specifically on entry of the reactants into the catalyst bed and
on exit of
the reaction mixture out of the catalyst bed.
It is possible for a plurality of catalyst beds to be connected in parallel or
in series.
When a plurality of catalyst beds are connected in series, the temperature
deviations
mentioned in the isothermal method preferred in accordance with the invention
relate to
the particular temperature in the catalyst bed, specifically on entry of the
reactants into
the first catalyst bed and on exit of the reaction mixture out of the last
catalyst bed.
In a preferred embodiment, the temperature of the reactor tube is controlled
externally
with a heat carrier stream, in which case the heat carrier may, for example,
be an oil, a
salt melt or another heat-transferring liquid.
Advantages of the inventive reaction regime over a synthesis in the liquid-
only phase
and in particular over a non-isothermal synthesis include those of better
yields and
greater safety with regard to runaway reactions.
As a result of the preferably isothermal gas phase mode or gas/liquid mixed
phase
mode, preferably gas phase mode, the potential of a runaway reaction during
the
synthesis is greatly reduced. The material present in the reactor which would
be
available for a runaway reaction is only a fraction of the material in a
liquid phase-only
process.
The process according to the invention is preferably performed continuously,
in which
case the catalyst is preferably arranged as a fixed bed in the reactor. In
this case, flow
toward the fixed catalyst bed either from above or from below is possible.
TBA and DG are preferably used in a molar ratio of TBA:DG = 1 to 4,
particularly in a
molar ratio of TBA:DG = 1 to 3, more particularly in a molar ratio of TBA:DG =
1 to 2.
CA 02798971 2012-11-08
= PF 70552
11
The process according to the invention is preferably performed at an absolute
pressure in
the range from 1 to 200 bar, preferably 2 to 100 bar, more preferably 3 to 50
bar.
The process according to the invention is preferably performed at a
temperature in the
range from 165 to 205 C, more preferably 170 to 200 C, further preferably 175
to
195 C.
The catalyst hourly space velocity is preferably in the range from 0.1 to 2.0
kg, preferably
0.1 to 1.0 kg, and more preferably 0.2 to 0.7 kg of DG per liter of catalyst
(bed volume)
and hour. The use of higher catalyst hourly space velocities is possible.
The pressure in the reactor, which arises from the sum of the partial
pressures of the
TBA, DG and of the reaction products formed at the given temperatures is
appropriately increased by injecting hydrogen to the desired reaction
pressure.
The water of reaction formed in the course of the reaction generally does not
have a
disruptive effect on the conversion, the reaction rate, the selectivity and
the catalyst
service life, and is therefore appropriately removed therefrom only on workup
of the
reaction product, for example by distillation.
The excess hydrogen and any excess aminating agent present are removed from
the
reaction discharge, after it has appropriately been decompressed, and the
crude
reaction product obtained is purified, for example by a fractional
rectification. Suitable
workup processes are described, for example, in EP 1 312 600 A and EP 1 312
599 A
(both BASF AG).
Unconverted reactants (DG and/or TBA) and also any suitable by-products
obtained
are more preferably recycled back into the TBADG synthesis. Unconverted TBA
can,
for example, in batchwise or continuous mode, after condensation of the
products in
the separator, be passed over the catalyst bed again in the cycle gas stream.
Unconverted reactants can also be recycled into the synthesis after one or
more
continuous or batchwise workup step(s), for example, distillation(s), in pure
form or else
optionally as a mixture with a suitable secondary component.
All pressure figures relate to the absolute pressure.
Examples
A series of experiments on the preparation of 2-(2-tert-
butylaminoethoxy)ethanol
(TBADG) from tert-butylamine (TBA) and diethylene glycol (DG) in the presence
of
hydrogen was conducted in batchwise and continuous modes. Conditions were
CA 02798971 2012-11-08
PF 70552
12
selected in order to study the conversions and selectivities in the liquid
phase and in
the gas phase. The catalyst was in each case first activated and then used.
The
analysis was performed by means of gas chromatography (GC) on an Rtx-5-Amine
column (with length 30 m, internal diameter 0.32 mm, coating 1.5 pm) and with
a
temperature program of 60 C to 280 C at 4 C/rnin. The quantitative analysis
was
effected by determining factors for DG, TBADG, N-tert-butylmorpholine (TBM)
and
2,2'-di(tert-butylamino)diethyl ether (DAE) with diethylene glycol dimethyl
ether
(DGDME) as the standard. For technical reasons, it was not possible to analyze
the
TBA quantitatively (decompression losses). The conversion was calculated based
only
on the DG.
The results of the experiments are reported in the tables as diethylene glycol
conversion (DG conversion) in mol /0 of the DG used, as the TBADG selectivity
(TBADG selectivity) in mol% of the DG converted, as the molar ratio of TBADG
to the
TOM by-product (TBADG/TBM molar) and as the TBADG yield (calculated from the
DG
conversion and the TBADG selectivity).
Preparation of the nickel-free catalyst A (Al and A2)
The nickel-free copper catalyst A possessed the composition of 50% by weight
of CuO
and 50% by weight of gamma-A1203 (after the last heat treatment thereof and
before
the reduction thereof with hydrogen). Catalyst A was prepared by
coprecipitation of
copper oxides and aluminum oxides from the nitrate solution thereof (according
to
DE 30 27 890 Al, page 14 if., examples 1 and 2), catalyst Al, or by
impregnating
gamma-A1203 powder with an aqueous copper nitrate solution, catalyst A2. The
tableting was effected in each case by the customary method. Before
commencement
of the reaction, the catalyst was reduced in a hydrogen stream (see below).
Micropore volume of the catalysts Al and A2 thus obtained: 0.03 cm3/g.
Pore size distribution in the catalysts Al and A2 thus obtained: normalized to
pores
with a pore size of > 0 to 20 nm (measured to DIN 66134 (for the mesopores,
pore
size 2 to 20 nm) and DIN 66135-1 (for the micropores)), about 18% of the pores
had a pore size of less than 5 nm.
Comparative example 2 (CE-2) (batchwise process in the mixed phase)
DG (36.3 g, 0.34 mol), TBA (100.0 g, 1.36 mol), and catalyst (5 g, activated
comparative catalyst CC1, 28% by weight of Ni as NiO, 13% by weight of Cu as
CuO
on zirconium dioxide) were initially charged in an autoclave (300 ml). The
autoclave
was inertized with nitrogen and hydrogen was injected to 50 bar. After heating
to
195 C, the pressure was adjusted to 100 bar with hydrogen and, when the
pressure
decreased during the reaction, hydrogen was injected again to 100 bar. Samples
were
taken and analyzed by means of GC. The result is entered in table 1, CE-2.
Example 2 (E-2) (batchwise process in the mixed phase)
DG (32.6 g, 0.31 mol), and TBA (90.0 g, 1.23 mol) were initially charged in an
autoclave (300 ml) provided with a catalyst basket. The activated catalyst Al
(5 g) was
CA 02798971 2012-11-08
PF 70552
13
introduced into the catalyst basket and placed into the autoclave. The
autoclave was
inertized with nitrogen and hydrogen was injected to 50 bar. After heating to
205 C the
pressure was adjusted to 100 bar with hydrogen and, when the pressure
decreased
during the reaction, hydrogen was injected again to 100 bar. Samples were
taken and
analyzed by means of GC. The result is entered in table 1, E-2.
Comparative example 3 (CE-3) (continuous process in the gas phase)
For the continuous preparation of TBADG, TBA and DG were used to prepare a
feed
mixture in a molar ratio of 2:1.
The amination was performed in an oil-heated jacketed glass reactor (D = 40
mm,
L = 900 mm) with capacity approx. 1000 ml. The reactor was safeguarded to a
pressure of 0.2 bar gauge with a glass relief valve, and was operated at
standard
pressure. The feed was metered in at the top of the upright reactor together
with the
heated hydrogen by means of a feed pump. As a result of the temperature and
the
sufficient amount of hydrogen, the feed mixture was evaporated at the top of
the
reactor and conducted over the catalyst in gaseous form within the reactor. At
the
reactor outlet at the bottom was a receiver with a jacketed coil condenser,
where the
starting materials and the products were condensed and collected. The reactor
was
filled with approx. 200 ml of catalyst and, above that 800 ml of V2A metal
rings. The
catalyst comprised 46% by weight of Cu as CuO and 11% by weight of Ni as NiO
on
aluminum oxide, and was reduced before the start of the experiment at 180-200
C first
with a hydrogen/nitrogen mixture and later with pure hydrogen (activation).
Then
150 ml/h of the feed mixture (by calculation, 54.2 g/h of DG) and 160 I
(STP)/h of
hydrogen were fed to the reactor from the top downward at 190 C. The result of
the
experiment is entered in table 2, CE-3.
[Standard liters (I (STP)) = volume converted to standard conditions (20 C, 1
bar)].
Example 5 (E-5) (continuous process in the gas phase)
150 ml/h of a mixture of TBA and DG in a molar ratio of 2:1 and 1601 (STP)/h
of
hydrogen were fed continuously at standard pressure (1 bar) and 190 C to 200
ml of
catalyst Al in the same system as in CE-3. Before the start of the experiment,
the
catalyst was reduced at 180-200 C first with a hydrogen/nitrogen mixture and
later with
pure hydrogen (activation). The result is entered in table 2, B-5.
Example 6 (E-6) (continuous process in the gas phase)
150 ml/h of a mixture of TBA and DG in a molar ratio of 3:1 and 160 I (STP)/h
of
hydrogen were fed continuously at standard pressure (1 bar) and 190 C to 200
ml of
catalyst Al in the same system as in CE-3. Before the start of the experiment,
the
catalyst was reduced at 180-200 C first with a hydrogen/nitrogen mixture and
later with
pure hydrogen (activation). The result is entered in table 2, B-6.
CA 02798971 2012-11-08
= PF 70552
14
Examples 7,8 and 10, 11 (E-7, E-8 and E-10, E-11)
The following series of experiments (E-7 to E-11) was carried out in an oil-
heated
jacketed V2A reactor (D = 6 mm, L = 12.5 m) of capacity approx. 350 ml. The
feed and
the hydrogen were metered in continuously at the top of the upright reactor.
Example 7 (E-7) (continuous process in the mixed phase)
260 ml/h of a mixture of TBA and DG in a molar ratio of 4:1 and 1500 I (STP)/h
of
hydrogen were conducted continuously at 200 bar and 180 C to 200 ml of
catalyst Al.
Before the start of the experiment, the catalyst was reduced at 180-200 C
first with a
hydrogen/nitrogen mixture and later with pure hydrogen (activation). The
result is
entered in table 2, E-7.
Example 8 (E-8) (continuous process in the mixed phase)
260 ml/h of a mixture of TBA and DG in a molar ratio of 4:1 and 15001 (STP)/h
of
hydrogen were conducted continuously at 50 bar and 180 C to 200 ml of catalyst
Al.
Before the start of the experiment, the catalyst was reduced at 180-200 C
first with a
hydrogen/nitrogen mixture and later with pure hydrogen (activation). The
result is
entered in table 2, E-8.
Example 10 (E-10) (continuous process in the mixed phase)
700 ml/h of a mixture of TBA and DG in a molar ratio of 2:1 and 1500 I (STP)/h
of
hydrogen were conducted continuously at 25 bar and 185 C to 200 ml of catalyst
Al.
Before the start of the experiment, the catalyst was reduced at 180-200 C
first with a
hydrogen/nitrogen mixture and later with pure hydrogen (activation). The
result is
entered in table 2, E-10.
Example 11 (E-11) (continuous process in the mixed phase)
700 ml/h of a mixture of TBA and DG in a molar ratio of 2:1 and 1500 I (STP)/h
of
hydrogen were conducted continuously at 5 bar and 185 C to 200 ml of catalyst
Al.
Before the start of the experiment, the catalyst was reduced at 180-200 C
first with a
hydrogen/nitrogen mixture and later with pure hydrogen (activation). The
result is
entered in table 2, E-11.
Examples 12 to 16 (E-12 to E-16)
In the following series of experiments (E-12 to E-16), the influence of
nitrogen together
with hydrogen as a carrier gas for the reaction of DG with TBA in the gas
phase was
studied. The amination was performed in an oil-heated jacketed V2A reactor
(D = 41.1 mm, L = 3500 mm) of capacity approx. 5 I. The feed was metered in at
the
top of the upright reactor together with the heated hydrogen by means of a
feed pump.
As a result of the temperature established and the sufficient amount of
hydrogen, the
feed mixture was evaporated at the top of the reactor and conducted over the
catalyst
in gaseous form within the reactor. At the reactor outlet was a high-pressure
separator
at an operating temperature of 40 C, in which the hydrogen was removed from
the
CA 02798971 2012-11-08
PF 70552
liquid discharge and circulated by means of a pump back into the oven. Two
additional
pumps were used to return a particular amount of fresh hydrogen gas and a
particular
amount of nitrogen to the cycle gas. A pressure-regulating valve regulated the
offgas
rate; this kept the pressure in the system constant. The reactor was filled
with 1 liter of
5 catalyst Al. Before the start of the experiment, the catalyst was reduced
at 180-200 C
first with a hydrogen/nitrogen mixture and later with pure hydrogen
(activation).
Example 12 (E-12) (continuous process in the gas phase)
3.11/h of a mixture of TBA and DG in a molar ratio of 4:1 and 8.3 m3 (STP)/h
of
10 hydrogen were fed in continuously at 5 bar and 180 C with no nitrogen.
The hydrogen
to diethylene glycol ratio was 56:1, and the residence time approx. 2.5
seconds. The
result is entered in table 3, E-12.
[Standard cubic meters (m3(STP)) = volume converted to standard conditions (20
C,
1 bar)].
Example 13 (E-13) (continuous process in the gas phase)
3.1 l/h of a mixture of TBA and DG in a molar ratio of 4:1 and 15.3 m3 (STP)/h
of
hydrogen were fed in continuously at 10 bar and 180 C with no nitrogen. The
hydrogen
to diethylene glycol ratio was 112:1, and the residence time approx. 2.6
seconds. The
result is entered in table 3, E-13.
Example 14 (E-14) (continuous process in the gas phase)
2.1 l/h of a mixture of TBA and DG in a molar ratio of 4:1 and 19.3 m3 (STP)/h
of
hydrogen were fed in continuously at 20 bar and 180 C with no nitrogen. The
hydrogen
to diethylene glycol ratio was 217:1, and the residence time approx. 3.8
seconds. The
result is entered in table 3, E-14
Example 15 (E-15) (continuous process in the gas phase)
2.1 l/h of a mixture of TBA and DG in a molar ratio of 4:1 and 19.3 m3 (STP)/h
of a
hydrogen/nitrogen mixture of 1:1 were fed in continuously at 20 bar and 180 C.
The
hydrogen to diethylene glycol ratio was 107:1, and the residence time approx.
3.9 seconds. The result is entered in table 3, E-15.
Example 16 (E-16) (continuous process in the gas phase)
2.1 l/h of a mixture of TBA and DG in a molar ratio of 4:1 and 19.3 m3 (STP)/h
of a
hydrogen/nitrogen mixture of 1:1 were fed in continuously at 20 bar and 180 C.
The
hydrogen to diethylene glycol ratio was 71:1, and the residence time approx.
3.9 seconds. The result is entered in table 3, E-16.
4 PF 70552 CA 02798971 2012-11-08
16
Discussion of results:
In the presence of a nickel-containing copper catalyst, CE-2 (table 1), the
best yield of
42% of TBADG was attained after 8 h, with the best selectivity of 64% at a
conversion
of 65%. The selectivity at the start (4 h) is good and declines at the end (12
h) the
higher the conversion becomes.
In the presence of a nickel-free copper catalyst, example E-2, a yield of 53%
was
achieved, with a DG conversion of 71% and a selectivity of 74% after 12 h. The
selectivity of 74% was not achieved on the nickel-containing catalyst in CE-2.
One advantage of low-nickel, especially nickel-free, copper catalysts is the
increase in
the yield by approx. 30% by the improvement in the TBADG selectivity by
increasing
the TBADG/TBM ratio and by reducing the decomposition of DG.
In order to demonstrate the advantage of a nickel-free copper catalyst over a
nickel-
containing copper catalyst, experiments were also carried out in the gas phase
at
1 bar. In comparative example CE-3 (table 2) the reaction was performed on a
catalyst
comprising 46% by weight of Cu as CuO and 11% by weight of Ni as NiO on
aluminum
oxide, and in example E-5 the reaction was carried out under the same
conditions on
the nickel-free copper catalyst. The Ni-containing catalyst produces much more
of the
undesired TBM than the Ni-free catalyst. Even as a result of reduction in the
temperature by 10 C to 180 C (CE-4), the TBM remains as the main product. On
the
nickel-free catalyst in E-5, the desired TBADG product is the main product and
is
produced with a molar ratio to the TBM of approx. 13.3. This is a crucial
advantage with
regard to material costs.
In the further experiments E-7 to E-11 in table 2, the influence of the
pressure on the
reaction in a continuous process was studied. Under all conditions in these
experiments, a liquid phase and a gaseous phase are present in the reactor.
When the
pressure is reduced from 200 bar to 50 bar, the yield can be improved from
approx.
13% to approx. 32%. The reduction of the pressure further to 25 bar and 5 bar
enables
a comparable yield of approx. 30% at double the space velocity. The space-time
yield
can thus be more than doubled by reducing the pressure from 50 bar to 5 bar.
The reduction of the pressure has two effects:
1) the concentration of hydrogen in the system is reduced,
2) more DG and TBA are present in the gas phase.
Another series of experiments (E-12 to E-16) was carried out under particular
conditions in order to maintain a single gas phase (and no liquid phase) in
the reactor.
The cycle gas rate and optionally the space velocity were adjusted in order to
prevent
the formation of a liquid phase at relatively high pressure.
The first three experiments show how the conversion of DG, the selectivity and
yield of
TBADG fall, when the pressure is increased from 5 to 10 and 20 bar. This
indicates
CA 02798971 2012-11-08
PF 70552
17
that not only the second effect mentioned above has an influence on the
reaction, but
also the concentration of hydrogen. More specifically, the concentration of
hydrogen in
the system can be described as the molar ratio of hydrogen to DG. The molar
ratio
(MR) of H2 to DG has to be increased in the specific examples at relatively
high
pressure from approx. 56 to approx. 217, in order to maintain a gas phase
system.
Experiments E-15 and E-16 were conducted under the same conditions, except
that
hydrogen has been replaced stepwise with nitrogen. However, the total cycle
gas rate
was kept constant in order to prevent the formation of a liquid phase. The MR
H2:DG
was reduced as a result from 217 to 107 and 71. In agreement with the
observations so
far, the lowering of the MR H2:DG improved the conversion of DG and the
selectivity
and the yield of TBADG.
..
-11
-n
--4
a
Table 1
al
al
MR Sample TBADG/TBM DG
TBADG TBADG IV
Experiment Cat. Temp. Pressure
TBA:DG x h (molar)
conversion selectivity yield
CE-2 CC1 195 C 100 bar 4:1 4 h 4.0
49% 63% 31%
8 h 2.9 65%
64% 42%
12 h 1.3 85%
46% 39%
, E-2 Cu/A1203 195 C 100 bar 4:1 12h 3.3
71% 74% 53%
Table 2
C)
0
n)
Space
l0
Experiment Pressure Temp Feed velocity H2 MR DG
TBADG TBADG/TBM TBADG " CD
VD
CO
...3
No. bar C g/h ml/h kg/I=h I (STP)/1-
h TBA:DG:H2 conversion selectivity ratio yield H
1.)
DG Total DG mol%
mol% (molar) mol% 0
H
I.)
1
CE-3 1 190 54.2 150 0.27 800 2 : 1 : 14 83.2
13.3 0.2 11.1 H
H
1
CE-4 1 180 54.2 150 0.27 800 2 : 1 : 14 64.6
36.1 0.8 23.3 0
CD
E-5 1 190 54.2 150 0.27 800 2: 1 : 14 67.8
70.0 13.3 47.4
E-6 1 190 54.2 200 0.27 800 3: 1 : 14 71.7
84.7 13.6 60.7
E-7 200 180 50 260 0.27 7500 4: 1: 13 15.2
89.0 11.9 13.5
E-8 50 180 50 260 0.26 7500 4 : 1 : 14
40.4 78.2 4.5 31.6
E-10 25 185 130 700 0.62 7500 4: 1: 6 30.9
84.3 6.4 26.0
E-11 5 185 130 700 0.62 7500 4 : 1 ; 6
38.6 78.8 5.4 30.4
-0
Table 3
0
cn
Exp.. Pressure Temp. Feed Sp. vet.
Cycle gas Fresh gas MR MR MR DG TBADG TBADG/TBM TBADG
No. (bar) [ C]
kg/h l/h (kg/I=h) [m3 (STP)/h] [I (STP)/h] TBA : H2: N2:
conversion selectivity ratio yield
DG tot. DG DG DG H2
mol% mol% molar mol%
E-12 5 180 636 3.1 0.64 8 300 4 56 0 30.0
64.8 1.9 19.4
0
E-13 10 180 636 3.1 0.64 15 300 4 112 0 27.1
51.0 1.1 13.8
OD
E-14 20 180 424 2.1 0.42 19 300 4 217 0 19.7
46.0 0.9 9.0
0
E-15 20 180 424 2.1 0.42 19 300 4 107 1 35.4
78.3 4.2 27.7 co
0
OD
E-16 20 180 424 2.1 0.42 19 300 4 71 2 30.3
83.4 6.2 25.2
