Note: Descriptions are shown in the official language in which they were submitted.
1089879
Background and Prior Art
This invention relates to the oxychlorination of
ethylene in a fixed bed catalytic process to produce
chlorinated hydrocar~ons, particularly 1,2-dichloroethane.
It is well known that hydrocarbons such as ethylene
may be chlorinated by reacting them with hydrogen chloride
and gases containing elemental oxygen, particularly air,
in the presence of a catalyst at elevated temperatures and
pressures. Such a process is generally termed an "oxy-
; 10 chlorination" or a Deacon process and usually employs a
catalyst comprising a chloride of a metal having at least
two valences, generally on a porous refractory support.
The most common catalyst for such processes comprises
cupric chloride on a particulate material such as activated
alumina, silica, alumina-silica, diatomaceous earth,
etc. Activated alumina, in one or another of its various
forms, isthe most common support utilized. In addition,
the catalyst may contain additives such as alkali metal
chlorides, rare earth metal chlorides, and other metallic
compounds which assist in promoting the desired reaction
and/or inhibiting the progress of side reactions. Particu-
larly, potassium chloride has been utilized as an additive
to such a catalyst when it is desired to produce 1,2-
dichloroethane from ethylene since potassium chloride is
known to s~ppress the formation of ethyl chloride. The
; amount of potassium chloride used, however, is kept low
as it also tends to decrease the activity of the catalyst
towards the primary reaction.
In the conduct of such a fixed bed oxychlorination
process one of the concerns is the control of the reaction
temperature. The oxychlorination reaction itself is highly
:
1089879
exothermic and, in addition, control of the reaction temper-
ature is hampered by the fact that the catalyst bed itself
has a low heat conductivity. As a result of these two factors,
there is a danger of the formation of undesirably extra-
ordinarily high localized temperature zones in the catalyst
bed. Numerous expedients have been proposed in the art
aimed at preventing or at least minimizing the existence of
such exceptionally high localized temperatures. For example,
it has been variously proposed to control temperature by
adjusting the ratio of the reactants; by diluting the feed ;
with an inert gas or an excess of one or more reactant
gases; by utilizing a tubular reactor with controlled
external cooling and/or tubes of varying diameters; by -
diluting the catalyst particles with inert particles; and
by varying the particle size of the catalyst and/or inert
particles. ~,,
Particularly when the reaction is conducted in
tubular reactors, it is known in the art, for example as
described in U.S. Patent 3,184,515, that the problem of
undesirably high localized temPeratures does not exist through-
out the entire reactor. This is a result of the nature of
the oxychlorination reaction itself, which becomes progressive-
ly less vigorous in the direction of flow of the reaction
mixture. At the inlet of the,catalyst bed, the reaction
proceeds rapidly and strongly and control of both the
temperature and location of the hot spot (point o~ highest
temperature) in the bed is important. ~owever, as the
reactants proceed through the bed, the reaction becomes
somewhat less vigorous as oxygen is consumed. This is
particularly the case when, as is known in the art, the
oxychlorination process is carried out in a series of two
1089879
or more catalytic reactors with the total air (or oxygen)
feed being split between the several reactors. Thus, as
the oxychlorination feed mixture reacts, oxygen is used up
toward the outlet of each reactor and the concern of a
runaway reaction or overly high localized temperature in
this zone is less than in the portion of the reactor closer
to the inlet.
One solution which has been proposed for accomplish-
ing acceptable conversions and selectivity to dichloro-
ethane as well as obtaining reasonable control of the
reaction temperatures and formation of high localized
temperatures is to dilute the catalyst with inert particles
intermingled with the catalyst particles. The inert
particles may consist of, for example, silica, alumina,
graphite, glass beads, etc. In some processes, the pro-
portions of catalyst to diluent in the oxychlorination
reactor have been varied. At the inlet side of the reactor,
it is desirable to have a less concentrated catalyst because
of the danger of runaway reactions or high localized temper-
atures. However, further from the inlet, as the reaction
proceeds and becomes less vigorous, these dangers are not
quite so prominent and in fact, it would be desirable to
have a more highly active catalyst to continue promoting
the reaction as oxygen is consumed. ~hus, it is common
when using diluted catalysts to divide the bed into two
or more zones, each zone containing a different ratio of
catalyst to diluent, with the ratio of catalyst to diluent
increasing toward the outlet end of the reactor. For example,
in U.S. Patent 3,184,515, a process is described (Example 1)
3Q using a diluted catalyst in which the reaction tube is divided
into four zones, the first zone containing 7 volume percent
~Q89879
catalyst anà 93 volume percent graphite diluent, the second
zone containing catalyst to diluent in a 15:85 volume percent
ratio, the third containing a 40:60 mixture, and the fourth
containing 100 percent catalyst. In another variation,
a fairly highly active catalyst may be placed at the very --
inlet of the reactor in order to initiate the reaction,
immediately followed by a much less active catalyst to prevent
the formation of hot spots in the adjacent reaction zone.
The use of a diluted catalyst such as described
in the preceding paragraphs, however, possesses several dis-
advantages. In the first place, it requires loading of the
catalyst in several different reaction zones and thus, the
formati~n of several catalyst-diluent mixtures of different
proportions. Of greater concern however, is the fact that
the mixing of catalyst and diluent can result in a non-uniform
mixture There is, therefore, a likelihood of formation of
undesirably high localized temperatures due to a concentration
of catalyst particles in a particular section of the reaction
zone if the mixing is not carried out to a sufficiently
thorough degree. Additionally, in many cases the diluent
particles are not of the same general size or shape as the
catalyst particles. It has been proposed, for instance,
to dilute cylindrical catalyst particles or spherical catalyst
particles with diluent particles of a different shape or size.
In such cases, the overall mixture does not provide a rea-
sonably uniform surface to the reactants and the pressures -
and/or pressure drops occurring during the reaction may be
other than advantageous. Even if the diluent particles
possess the same relative shape or size as the catalyst
3~ particles, there is still the li~elihood that the diluent
and catalyst will not be satisfactorily mixed and hot spots
lQ89879
can result, as well as the nuisance of having to mix up
different catalyst-diluent mixtures for the different reaction
zones.
Another solution which has been proposed has been
to provide a catalyst, either with or without diluent
particles, in which the particle size decreases from inlet
to outlet. Such a concept is described in U.S. Patent
3,699,178. However, such a practice, even if the catalyst
is utilized without a diluent, re~uires careful manufacture
to assure that the particle sizes are as desired and requires
the manufacture of at least two, and very likely three or
four different catalysts, in order to attain the objectives
of this concept. In the alternative embodiment of this
concept, in which the catalyst is intermingled with diluent
particles, the disadvantages of utilizing diluent particles
are added to those involving the preparation of different
sized catalyst particles.
It is an object of the present invention to pro-
vide an oxychlorination process for the production of 1,2-
dichloroethane from ethylene.
It is a further object of this invention to pro-
vide an oxychlorination process for the production of l,2-
dichloroethane from ethylene in which hot spot location and
temperature can be readily controlled.
~t is another obiect of the present invention
to provide a process for the production of 1,2-dichloroethane
by oxychlorination of ethylene at high hydrogen chloride
conversion rates.
Another object of the present invention is to
pro~ide such a process in which a substantially uniform
pressure drop can be maintained for a reasonably long period
-- 6 --
1089879
of time.
Yet another object of the present invention is to
provide such a process in which selectivity of conversion of
ethylene to 1,2-dichloroethane is acceptably high, and in
which excesses of reactants such as ethylene and/or air can
be kept to a minimum.
Still another object of the present invention is
to provide such a process which can be carried out at high
flow rates of reactants.
Another object of *he present invention is to
provide a system and process for the oxychlorination of
ethylene which can be run using either air or oxygen as the
oxygen-containing gas.
Yet another object of the present invention is
to provide a new catalyst for use in conducting the oxy-
chlorination of ethylene to 1,2-dichloroethane with the above
advantages.
Summary of the Invention -
In brief, the invention herein comprises the -
providing of a novel oxychlorination process utilizing a
novel catalyst comprising cupric chloride and potassium
chloride supported on a spherically shaped activated alumina
having a high surface area and a high structural integrity.
Ethylene is converted to 1,2-dichloroethane at high rates of
conversion and selectivity with good hot spot control and
very little increase in pressure drop over long periods of
time by utilizing such a catalyst in a series of three
reactors, in which the catalyst is utilized substantially
without the presence of an inert diluent, the content and
weight ratio of cupric and potassium chlorides in the
catalyst being varied within each reactor. Tn each of the
first two reactors, the catalyst bed is diveded into two zones,
10~9879
the catalyst in the zone nearest the inlet preferably having
a somewhat lower content of cupric chloride and a higher weight
ratio of potassium chloride to cupric chloride than the
catalyst in the zone nearest the outlet.
In yet another aspect, this invention comprises a
system for the carrying out of an oxychlorination process,
comprising the aforesaid catalyst situated in reactors as
herein described.
More particularly, in one aspect, the invention
herein comprises a process for the oxychlorination of ethylene
in a fixed bed comprising the steps of:
(a) reacting ethylene, hydrogen chloride and a
first portion of an oxygen-containing gas in a first reaction
zone in contact with a first catalyst bed substantially
undiluted by catalytically inert particles and comprising
cupric chloride and potassium chloride supported on spherical
particles of activated alumina, the first catalyst bed being
divided into two portions in the direction of flow of reactants
therethrough, the first portion comprising between about
45~ and about 7S% of the bed, the second portion comprising
between about 25% and about 55% of the bed, the first portion
comprising between about 4.5 and about 12.5 weight % cu~ric
chloride and between about l.S and about 7 weight % potassium
chloride, in a weight ratio of cupric chloride to potassium
chloride of between about l.S:l and about 4 1, the second
portion comprising between about 12 and about 25 weight %
cupric chloride and between about 0.5 and about 4 weight %
pota~ssium chloride, in a weight ratio of cupric chloride to
potassi~m chloride of between ahout 5:1 and about 15:1;
(b) reacting the products of step (a) and a
second portion of an oxygen-containing gas in a second
- 1089879 --
reaction zone in contact with a second catalyst bed sub-
stantially undiluted by catalytically inert particles and
comprising cupric chloride and potassium chloride supported
on spherical particles of activated alumina, the second
catalyst bed being divided into two portions in the
direction of flow of reactants therethrough, the first
portion comprising between about 45% and about 75% of the bed,
the second portion comprising between a~out 25% and about 55%
of the bed, the first portion of the bed comprising between
about 5.5 and about 15 weight % cupric chloride and between
about l and about 5 weight % potassium chloride, in a weight
ratio of cupric chloride to potassium chloride of between
about 2:1 and about 6:1, the second portion of the bed com-
prising between about 12 and about 25 weight % cupric chloride - --
and between about 0.5 and about 4 weight % potassium chloride,
in a weight ratio of cupric chloride to potassium chloride of
between 5:1 and about 15:1; and,
(c) reacting the products of step (b) and a
third portion of an oxyqen-containing gas in a third reaction
zone in contact with a third catalyst bed substantially
undiluted by catalytically inert particles and comprising
between about 12 and about 25 weight % cupric chloride and
about 0.5 and about 4 weight % potassium chloride, supported
on spherical particles of activated alumina, in a weight ratio
of cupric chloride to potassium chloride of between about 5:1
and about 15:1.
In another aspect, this in~ention comprises a
catalyst comprising cupric chloride and potassium chloride
supported on a base comprising spherical particles of
activated alumina, said base having a BET surface area of
~Q89879
between about 225 and about 275 m2/g, an attrition hardr.~ss
of at least 90%, a total nitrogen pore volume of between
about 0.3 and about 0.6 cc/g, an average pore diameter of
between about 50 and about lO0 A, between about 20 and about
50% of the pore volume composed of pores having a diameter
of between about 80 and 600 A and further characterized by an
X-ray diffraction pattern as hereinafter defined and sub-
stantially no observable grain boundaries after etching with
HF.
In a third aspect, this invention comprises a
reaction system for oxychlorination of ethylene in the presence
of a catalyst comprising cupric and potassium chlorides
supported on spherical particles of activated alumina, the
reaction system comprising:
(a) a first reaction zone comprising a first
catalyst bed divided into two portions in the direction of
f low of reactants therethrough, the first portion comprising
between about 45% and about 75% of the bed, the second portion
comprising between about 25~ and about 55% of the bed, the
first portion comprising between about 4.5 and about 12.5
weight % cupric chloride and between about 1.5 and about
7 weight % potassium chloride in a weight ratio of cupric
chloride to potassium chloride of between about 1.5:1 and
about 4:1, the second portion comprising between a~out 12
and about 25 weight % cupric chloride and between about
0.5 and about 4 weight % potassium chloride, in a weight
ratio of cupric chloride to potassium chloride of between
about 5:1 and about 15:1;
(b) a second reaction zone comprising a second
catalyst bed divided into two portions in the direction
of flow of reactants therethrough, the first portion comprising
-- 10 --
lQ8g879
between about 45% and about 75% of the bed, the second
portion comprising between about 25% and about 55% of the bed,
the first portion comprising between about 5.5 and about 15
weight % cupric chloride and between about 1 and about 5
weight % potassium chloride, in a weight ratio of cupric
chloride to potassium chloride of between about 2:1 and about
6:1, the second portion of the bed comprising between
about 12 and about 25 weight % cupric chloride and between
about 0.5 and about 4 weight % potassium chloride, in a - -.
weight ratio of cupric chloride to potassium chloride of
between about S:l and about 15:1;
(c) a third reaction zone comprising a third
catalyst bed comprising between about 12 and about 25 weight
% cupric chloride and between about 0.5 and about 4 weight
potassium chloride, in a weight ratio of cupric chloride to
potassium chloride of between about S:l and about lS:l.
Detailed Description of the Invention
The invention will be described with reference
to the Figure, which depicts an illustrative flow sheet for
carrying out the process and utilizing the catalyst and system
of the present invention.
The catalyst is prepared by conventional impreg-
nation techni~ues utilizing a~ueous solutions of cupric
and potassium chlorides, as further described in the examples
which follow. The support which is impregnated as a spherical
particulate high surface area activated alumina, that is,
an alumina havlng a BET surface area of at least 100 m /g,
preferably between about 225 and about 275 m /g, and also
an attrition hardness of at least 90%, a total nitrogen pore
~olume of between about 0.3 and about 0.6 cc/g, and an
average pore diameter of between about 50 andabout 100 A,
1089879
wherein between about 20 and about 50% of the pore volume is
composed of pores ha~ing a diameter of between about 80 and
about 600 A. The alumina is further characterized by an
X-ray diffraction pattern having three main peaks: diffuse
peaks at 1.39 and 1.98 A and another peak at 2.34 A, the
ratio of intensity of the peak at 1.39 A to that at 1.98 A
being between about 1.5:1 to about 5:1, and by showing sub-
stantially no observable grain boundaries after etching with
HF. Supports having these properties have been found to have
good structural integrity even over lengthy periods of
operation. Examples of suitable supports of this type are
aluminas currently available under the designations HSC-ll ~,
from Houdry Process and Chemical Company and SCM-250~,
from Rhone-Progil.
As seen in the Figure, the process is conducted
in three reactors in series, designated Rl, R2 and R3. Each
of the reactors contains a catalyst bed, 12, 14, and 16 re-
spectively. The reactors are preferably tubular reactors,
having tubes packed with the catalyst. In reactor Rl, the
catalyst bed is divided into two parts in the direction of
the flow of reactants therethrough. The first part, at the
inlet side of the reactor, comprises between about 45 and
about 75%, preferably between about 5~ and about 65~, of the
length of the catalyst bed and contains a catalyst having
between about 4.5 and about 12.5, preferably between about
~ and a~out 8, and most preferably between about 5.5 and
about 6.5 weight % cupric chloride, and between about 1.5
and about 7, preferably between about 2 and about 4, and
most preferably between about 2.7 and 3.3 weight % potassium
chloride, wherein the weight ratio of cupric chloride to
potassium chlorideis between about 1.5:1 and 4:1, preferably
1089879
between about 1.5:1 and 3:1, and most preferably about 2:1.
The second half of the bed, at the outlet end of the reactor,
comprises correspondingly between about 25 and about 55%,
preferably between about 35 and about 45%, of the bed length
and contains a catalyst having between about 12 and about
25 weight %, preferably between about 15 and about 20 weight
%, cupric chloride, between about 0.5 and about 4 weight %,
preferably between about 1.5 and about 3 weight ~, of potassium
chloride, the weight ratio of cupric to potassium chlorides
being between about 5:1 and about 15:1, preferably between
about 5:1 and about 12:1, most preferably about 10:1. Thus,
the catalyst bed 12 of reactor Rl contains two types of
catalysts: a first catalyst in the section of the bed
towards the inlet, having a relatively lower activity in order
to ensure that the reaction does not become uncontrollable
in its early stages, and a second catalyst having a higher
cupric chloride content and therefore, a higher activity,
in the portion of the reactor towards the outlet, to continue
the reaction at the point at which the reaction begins
to lose vigor due to the consumption of oxygen.
~eactor R2 contains catalyst bed 14 which is also
divided into two portions. The first portion of the catalyst
bed, at the inlet side of the reactor, comprises between about
45 and about 75%, preferably between about 55 and about 65%,
o~ the bed length and the second portion, toward the outlet
end, comprises between about 25 and about 55%, preferably
between about 35 and about 45%, correspondingly, of the bed
length. The catalyst in the first portion of the bed 14 in
reactor R2 can be somewhat stronger or more active than the
catalyst in the first portion of the bed 12 in reactor ~1 -
since the reaction has already proceeded partway towards
- 13 -
1089879
completion. In reactor R2, the catalyst in the first portion
of the bed contains between about 5.5 and about 15%, pre-
ferably between about 7.5 and about 12.5%, and most preferably
between about 9 and about 11 weight % cupric chloride,
and between about 1 and about 5, preferably between about
1.5 and about 3.5, and most preferably between about 2.5
and about 3.5 weight % potassium chloride, the weight ratio
of cupric chloride to potassium chloride being between about
2:1 and about 6:1, preferably between about 3:1 and about 4:1,
most preferably about 10:3 (3.3:1). Similarly to reactor Rl,
the second portion of the catalyst bed of reactor R2 contains
a stronger catalyst, which has between about 12 and about 25
weight %, preferably between about 15 and about 20 weight
%, cupric chloride and between about 0.5 and about 4 weight
~, preferably between about 1.5 and about 3 weight % potassium
chloride, the weight ratio of cupric to potassium chloride
being between about 5:1 and about 15:1, preferably between
about 5:1 and about 12:1, most preferably about 10:1. Similar-
ly to reactor Rl, since the reaction has been dropping in
intensity as the gases pass through the catalyst bed, due to
the consumption of oxygen, the danger of undesirably high
localized temperatures in the portion of the bed nearest
the outlet is much less than that in the portion near the
inlet and therefore a more active catalyst can, and in fact
should, be used for best conversions.
In both reactors Rl and R2, the weight ratio of
potassium chloride to cupric chloride is preferably higher in
the first portion of the catalyst bed than in the second
portion.
~0 Entering reactor R3, the reaction has proceeded
most of the way toward completion and ~here is much less
- 14 -
~()89879
danger of undesirably high localized temperatures in comPari-
son to reactors Rl and R2. Therefore, the entire bed of
reactor R3 can consist of the more active catalyst utilized
in the lower sections of beds 12 and 14, that is a catalyst
containing between about 12 and about 25 weight %, preferably
between about 15 and about 20 weight % cupric chloride, and
between about 0.5 and about 4 weight %, preferably between
about 1.5 and about 3 weight ~, potassium chloride, with
the weight ratio of cupric to potassium chlorides being
between about 5:1 and about 15:1, preferably between about
5:1 and about 12:1, most preferably about 10:1. It may also
be possible to utilize such a catalyst in reactor R3 and/or
the portions of the catalyst beds 12, 16 nearest the outlet in
reactors Rl and R2 with an even lower potassium chloride
content and higher weight ratio of cupric to potassium chloride.
Tn the operation of the process of the present
invention, hydrogen chloride qas is introduced through line 1
and preheated in preheater 22. Ethylene is introduced in line
2, preheated in preheater 21 and combined with the hydrogen
chloride feed in line 1. An oxygen-containing gas, which may
be air, molecular oxygen or oxygen-enriched air is
introduced through line 3 and divided into three portions in
lines 3a, 3b and 3c, respectively. The di~ision may be into
three e~ual or unequal portions. The portion in line 3a
is combined with the mixed hydrogen chloride ethylene feed
in line 1 and introduced into reactor Rl through line 4.
The temperature of the mixed gaseous feed is generally about
120-220Oc~ preferably about 135-180C. The mixed feed passes
through catalytic bed 12 which, as explained previously,
preferably consists of catalyst pac~ed in tubes, and is with-
drawn from the reactor in line 5.
10~9879
The products of reactor Rl in line S are com-
bined with the second portion of oxygen-containing gas in
line 3b, introduced into reactor R2 via line 6 and contacted
with catalyst in bed 14. Reaction products are removed via
line 7, contacted with a third portion of oxygen-containing
gas in line 3c and introduced via line 8 into catalytic bed
16 of reactor R3. The reaction products are removed in line 9,
preferably cooled in heat exchanger 23 and condensed in pro-
duct condenser 24. The reaction products are recovered in
line 10 and consist primarily of 1,2-dichloroethane, with
small amounts of ethyl chloride and other chlorinated
hydrocarbons.
In general, the process is carried out at an
overall system pressure of between about 30 and about 100
psig, preferably between about 40 and about 90 psig. In order
to maintain temperature control in reactors Rl, R2 and R3,
these reactors are preferably constructed as jac~eted reactors,
the jacket surrounding the tubes or catalystbedscontaining
a heat exchange fluid such as boiling water, steam or
Dowther ~ fluid. In general, the reaction is conducted at
temperatures of between about 180 and about 340C, preferably
between about 235 and about 300C. In reactor ~1' the hot
spot temperature is generally maintained below about 330C,
preferably below about 300C. In reactor R2, the hot spot
temperature is maintained generally below 330C and preferably
also below 300C and in reactor R3, the hot spot is maintained
at generally below 320C, preferably below 300C. Another
important factor is the control of the location of the hot
spot in the reactors In each reactor, the hot spot should be
located toward the inlet end of the catalyst bed. In
reactors Rl and R2, the hot spot should, in fact, be located
- 16 -
lQ89879
toward the inlet side in the first, or less active portion
of the bed. If the hot spot occurs too far toward the outlet
of the bed, it is an indication that the reaction is pro-
ceeding too slowly in the bed, that is, that the catalyst
is not being used efficiently. Additionally, if the hot
spot occurs too far toward the outlet of the bed, it may
produce a cumulative result with-the reaction-boosting effect
of the stronger catalyst in the second portion of the bed,
resulting in an undesirably high temperature level at that
point.
The division of the oxygen-containing gas among
lines 3a, 3b and 3c can be performed such that equal amounts
enter all three reactors or, as desired, the amount entering
each reactor can be varied as is known in the art. A variance
in the amount of this gas introduced into each reactor can
effect the temperature and the location of the hot spot in
the reactor.
For purposes of illustration, the Figure depicts
a system in which all three reactors are of the downflow type;
that is, reactants are introduced at the top of the reactors
and products withdrawn at the bottom. The invention may,
however, be practiced in either downflow or upflow reactors,
with suitable catalyst hold-down devices utilized for upf~ow
reactors. Construction of the reactors as alternately down-
flow and upflow may reduce piping costs. Thus, reactors R
and R3 could be upflow reactors and reactor R2 a downflow
one. Conversely, reactors R1 and R3 could be downflow and
R2 upflow.
The hydrogen chloride feed need not all be
introduced into the first reactor; a portion may be split
off and introduced into reactor R2, with the second portion
- 17 -
1089879
of the oxyge~-containing gas.
In general, the process is operated with an excess
of both oxygen-containing gas and ethylene with respect to
hydrogen chloride in order to ensure as complete conversion
of hydrogen chloride as possible. When using air, it is
desirable to maintain as low an excess of these reactants
as possible in order to avoid the handling of large amounts
of gas. In such case, the ethylene excess is maintained at
a maximum of about 35%, preferably between about 5 and about
20~, and the air excess is maintained at a maximum of about
25%, preferably between about 5 and about 20%.
The invention is described herein primarily in
the context of an oxychlorination process in which the
oxygen-containing gas is air. However, the catalyst, system
and process herein described are also suitable for use in an
operation in which molecular oxygen or oxygen-enriched air
is employed. In one such process, as described in U.S. Patent
3,892,816, a large excess of ethylene is utilized with
respect to the oxygen and hydrogen chloride to prevent over-
reaction and excessively high hot spots. The excess ethyleneis preferably recovered from the reaction product, and recycled
to the oxychlorination reactors; thus in such a process,
minimizationofethylene excess is not an objective.
If steel equipment is utilized in the hydro~en
chloride flow circuit, small amounts of ferric chloride
may be formed and introduced into the catalyst with the feed
stream. Corrosion of the oxygen-containing gas supply
system can produce iron oxides which can be converted to
ferric chloride by reaction with hydrogen chloride in the
reactor. The catalyst may contain minor amounts of iron as
impurities; however, it has been found that additional
- 18 -
1089879
ferric chloride, even in small amounts, tends to contaminate
and deactivate a cupric chloride oxychlorination catalyst.
In one embodiment of the present invention, therefore, ferric
chloride contamination of the catalyst is avoided, as shown
in the copending U.S. Patent 4,000,205 to Ramsey G. Campbell,
entitled "Purification of Gas Streams Containing Ferric
Chloride'' by passing either the hydrogen chloride stream or
the mixed gaseous feed stream through a bed of activated
alumina impregnated with between about 5 and about 25,
preferably between about 10 and about 20 weight % sodium
chloride or potassium chloride. ~he bed may be placed in the
hydrogen chloride line 1, either before or after preheater 22,
or in the combined feed line 4. Alternatively, it may be
located inside Rl at the space 30 adjacent the inlet. In a
preferred embodiment, in which tubular reactors are utilized,
the bed material may be loaded into the portions of the tubes
of reactor Rl adjacent the inlet, ahead of the cupric chloride
catalyst in the direction of flow of reactants, and may be
separated from it by a screen or the like to prevent migration
of removed ferric chloride into the cupric chloride catalyst.
In a preferred embodiment, the alumina utilized is the same
as that utilized as the catalyst support.
~ t should be noted that the operation of each of
the reactors Rl and R2 is only partially affected by the
operation of the other. Thus, for instance, should the
location or temperature of the hot spot in either of reactors
; Rl and R2 be temporarily outside the desired range, the
operation of the other reactor would not necessarily be
adversely affected or, if it were, it could be possible to
separately control operation by adjusting the coolant temper-
ature, di~ision of air between the reactors, etc. Similarly,
- 19 --
1~89879
it is possible, though much less preferred, to utilize
the catalyst disclosed herein in the disclosed proportions
and bed divisions, in either of reactors Rl or R2 and another
suitable, though probably less effective catalyst, in the
other. Such an arrangement would not be expected to possess
all the advantages of the present invention in terms of
throughput, pressure drop, conversion, selectivity, etc.,
but may be necessary in the event of an emergency or temporary
unavailability of catalyst.
To further illustrate the invention in its various
embodiments and modifications, the following examples are
presented. These in no way are intended to limit the in-
vention, but merely to serve as illustrations thereof.
Catalyst Preparation
The catalysts utilized in the following examples
were prepared as follows:
A quantity of dry unimpregnated spherically shaped
alumina particles was placed in a beaker and weighed. The
alumina utilized was HSC-114~ alumina obtained from Houdry
20 Process and Chemical Company, and had the following properties:
Surface Area, BET 250 + 25 m2/g
Bulk Density 37-44 lb/cu. ft.
Loss of Ignition (300C) 5 wt. %, max.
Attrition ~ardness 90%, min.
Pore Volume, N 0.44 cc/g
2 O
Average Pore Diameter, BET 64-70 A
Pore Volume,
80-600 A Pores -- 20-37% of total pore volume
Screen Analysis (Tyler Screen)
+ 3 mesh 1.0 max. wt. ~ -
- 20 -
1089879
- 3, + 4 mesh 25-70 wt.
- 4, + 5 mesh 25-70 wt. %
- 5, + 6 mesh 10 max. wt. %
- 6 mesh 3 max. wt. %
X-ray diffraction data showed three main peaks:
at 1.39, 1.98 and 2.34 A. The peaks at 1.39 and 1.98 A
were diffuse. The X-ray diffraction pattern of gamma-alumina,
as given in ASTM file 10-425, has main peaks at the same
values. Xowever, ASTM File 10-425 shows the intensities of
the peaks at 1.39 and 1.98 A to be about equal, whereas the
diffraction data for the catalysts used in these examples
shows the intensity of the peak at 1. 39 A to be about twice
that of the 1.9 8 A peak.
Results of this type indicate that the support
is comprised of (crystalline) gamma-alumina plus an amorphous
form of alumina. The amorphous form is possibly rho-alumina,
the X-ray diffraction pattern of which has a single broad
band at 1.40 A.
Purther investigation of the catalyst's micro-
structure was performed by HF etching. Samples of new andused catalyst were mounted on a special szmple holder and
imbedded in an epoxide resin. The samples were cut and
polished and chemically etched on a cross-section for 2-5
minutes at room temperature with a 20~ HF solution. Ob-
servations were made after 20 sec. and 4 minutes at 21X,
and after S minutes at 85X. ~o evidence of grain boundaries
was observed even after 12 minutes of etching.
Water was added to the bea~er in small amounts,
the contents being stirred with a glass rod after each
addition, until moisture was observed in the bea~er. At
that point, the alumina was saturated with water. From the
- 21 -
1(~89879
final weight of the beaker, the percentage by weight of the
absorbed water can be derived. Thus, the amount of water
by weight absorbed by the chosen weight of base is determined.
The ~uantity of dry alumina base to be impregnated
was weighed. The quantities of cupric and potassium chlorides
which constituted the desired weight percent with reference
to the dry base were weighed and dissolved in water. Additional
water was added to the resulting solution to bring the volume
up to the absorptive capacity of the base as previously
determined. It should be noted that it may be necessary
to reduce this volume by about 3% to prevent there being a
large excess of impregnating solution which is not absorbed
by the base. The base and impregnating solution were placed
in a drum which was covered, sealed, and rotated for a period
of 15 minutes to achieve impregnation.
After 15 minutes, the drum was removed, immediately
opened and the saturated alumina particles placed in either
glass or ceramic trays to a depth of approximately three
inches. The trays were placed in a forced draft oven where
the temperature was maintained at 140C for 16 to 24 hours.
The catalyst was then removed and allowed to cool prior to
being stored in air-tight containers.
The weight percents of the salts in the catalyst
thus prepared are within ~.5% of the desired values. The
accuracy can be enhanced by thorough drying of the
alumina base prior to weighing. ~rying should immediately
follow impregnation; a delay can be detrimental to the
physical structure o~ the resulting catalyst.
The experimental data in the tables which follow
was obtained using catalysts prepared according to the fore-
going procedure in an apparatus arranged in a flow se~uence
1(!898~79
as in the Figure. The three reactors, Rl, R2 and R3, each
consisted of one schedule 40 nickel pipe, 12.5 feet long and
1 inch in diameter, jacketed for the entire length with a
2-1/2 inch schedule 40 steel pipe, and arranged for downward
flow of reactants. The heat of reaction was removed by re-
fluxing Dowtherm~ E in the annular space between the two
pipes. The hot spot temper~ture and location inside the
catalytic bed of each reactor was measured by means of a
mo~able thermocouple inside a 12-foot long thermowell which
was introduced at the bottom of each reactor.
As shown in the Figure, hydrogen chloride was
introduced into the system in line 1 through a preheater 22.
Ethylene was introduced into the system in line 2 through a
preheater 21. Air was introduced into line 3 and split into
subconduits 3a, 3b and 3c, the air in line 3a being introduced
- into reactor Rl, that in 3b into reactor R2, and that in 3c
into reactor ~3. The mixed feed to reactor Rl af~er preheating
was at a temperature of about 140C. A 100% throughput basis
was established as being equivalent to a hydrogen chloride
mass velocity in the reactor of 117.6 g.-mole/h-in.2. The
air and ethylene feed rates are expressed as a percentage
excess based on hydrogen chloride, assuming a stoichiometric
reaction to produce 1,2-dichloroethane.
The gas exiting from reactor ~3, after pressure
reduction, was cooled in a glass water-cooled condenser which
condensed all the unreacted hydrogen chloride as an aqueous
Fhase and the majority of the 1,2-dichloroethane produced as
a relatively pure (about 98.5%) organic phase. Hydrogen
chloride conversion was determined by titrating the aqueous
3Q phase with sodium hydroxide to obtain its hydrogen chloride
concentration in weight percent.
The condensed gas from the water-cooled condenser
1089879
was analyzed by gas chromatography. The analyses were used
to calculate the amount of ethylene oxidation and formation of
ethyl chloride, both expressed as a percentage of the ethylene
feed.
Table I illustrates conditions of one run of the
process according to the present invention, in a preferred
embodiment. The run, which continued for a total of 320 hours,
was divided into two sections~ In the first section, occupying
the first 298 hours, the equipment was operated at a 125%
throughput of hydrogen chloride and other reactants. During ~-
the last 22 hours, the equipment was operated at a 100
throughput rate.
The catalyst to be used was prepared in three
formulations. These are shown in Table I as catalysts A, -
B and C. The desired weight percents of cupric and potassium
chlorides were:
Catalyst A: 6.0 + 0.5% CuC12
3.0 + 0.3~ KCl
(weight ratio 2:1)
Catalyst B: 10.0 + 0.7% CuC12
3.0 + 0.3% KCl
(weight ratio 3.3:1)
Catalyst C: 18.0 + 1.8% CuC12
1.8 + ~.25% KCl
(weight ratio 10:1~
(Table ~ shows the amounts found to be present after analysis
of the catalysts. The amounts of cupric and potassium chloride
in catalyst C varied somewhat because catalysts from several
different preparations were used). In all three reactors,
the catalyst bed length was 135 inches; the first (upper)
portion, being 81 inches (60~ of bed length) and the second
- 24 -
10~9879
(lower) portion being 54 inches in length, or 40~ of the total
bed length. Catalyst A was utilized in the first portion of
the reactor Rl (bed 12); catalyst B was utilized in the
first portion of reactor R2 (bed 14): catalyst C was utilized
in reactor R3 and the lower portions of reactors Rl and R2.
TABLE I
Catalyst: A B C
Cupric Chloride - wt.% 6.1 9.9 18.4-19.6 -~
Potassium Chloride - wt.% 3.08 3.08 1.84-1.92
Iron as Fe - ppm ~800 C800 ~800
Sulfate as SO4 - wt.~ C0.5 ~0.5 <0.5
Silica as SiO2 - wt.% ~0.25 <0.25 C0.25
Loss on Ignition at 300C - wt.~ ~6.0 ~6.0 ~6.0
Hardness - wt.% 93 min. 94 95
Screen Analysis - wt.%
+3 3 max. 3 max. 3 max.
+4 40 min. 40 min. 40 min.
~5 45 max. 45 max. 45 max.
+6 5 max. 5 max. 5 max.
BET Surface Area - m /g 204 196.5 149-165
Total N2 Pore Volume - cc/g ~0.25 >0.25 >0.25
Pore Volume 80-600A
Pores - % of total pore volume ~30 ~30 ~30
Average Pore Diameter
(BET) - A ~60 >60 ~60
During the first 60 hours of the run, various
sources which produced unstable data were corrected and
stabilized. Similarly, hours 265 through 298, which occurred
between the two portions of the overall run, were occupied
by restabilizing the pilot plant following a one-month shut-
down between portions of the run.
During operation, as can be seen from the follow-
ing table, variations were made in the total system pressure,
- 25 -
1089879
air and ethylene excesses, and division of air between the
three reactors, in order to determine the effect of these
conditions on hydrogen chloride conversion, selectivity of
the reaction to 1,2-dichloroethane, and hot spot temperature
and location. The run data and results are given in the
following Table II.
. - 26 -
1(~89879
.
,~ ~ ~ ~ ~ ~ I ~ ~ N ~ ~ N ~
~ 3 ~ ~ ~ N ~
O o l ~D a~ co ~ co ~ ~ I~ c~ 1~ ~ o ~ co ~ o~
--~ ~ ~ ~ ~ J ~ N ~ ~`J ~ ~ ~ ~ ~I N ~I
O O O O
~_ ~o ~ o u~ o u~
H ~ ~ dP 1~ r` CO ~ CO C~ CO C~ ~ I C~ O I
. H m~- ~ ~ ~
Y1 ~ 7~ ~
o ~ ~ .~ ~ U
U~
~ ~ j; N ~ O
O ;~ ~ ~ ~ ~ N
U~ ~
dP O O O O o o O O O
I. ~q .
~^ ~ ~
~
4~
.
~g
o o o o o o o o o o o o ~ D ~ O ,~
~_1IN O O O O O ~ O O O O O O
~i 1~ ~ r r 1
_ o r- ~ o ~ U~ o~ u~ o o u o
o C~l ~ ~ t~ ct~ o ~ ~o ~ o ~1
~ ~
~ 8 r I ~1 0 0 1 1 0 ~ o ~ o u~ o
~d-- I 1 0 ~ ~ ~ N ~ ~I ~ ~
,
:
89879
Similarly to run number 1 described hereinabove,
additional runs designated in Table III below as runs 2 through ~ -
9, were conducted utilizing various combinations of catalyst
compositions and reaction conditions within the scope of this
invention. In all cases, as was the case in run number 1, the
catalyst bed was 135 inches long. In both Rl and R2, the first
zone, at the upper, or inlet, end of the reactor occupied 81
inches, or 60% of the bed, while the second zone occupied 54
inches, or 40% of the total bed length. The catalyst in reactor
R3 was similar to that in run number 1. Coolant temperatures
were maintained similarly to run number 1.
- 28 -
10~9879
CO ~ CO U~
00 CO ~ ~ IS~ N 1`
u) ~ I O
~D 1` CO N ~ O a~ 1 ~ N N ~ N _~
a-~r ~ O N t~
~` ~0 00 ~ 't O ~ U~ N U~
CC~ I` 1~ C'`l NCOO CO CO _i N ~) ~ ~ N e1~ _I
a~ ~ ~
N
~ N O CO
I~ CO CO ~ D N O
i ~ O t~l I I O
-- _I 1~ NN CO N CO C~ O U~ N
o """ ~OD
co co a~
I r~l ~ ~ ~ U') N O
I ~ ~ N O U~ ~ O I
~O Il~ I O ~ N ~1 ~ ~ ~r N N D O
CO ~ ~
U~ ~ - N - N ~ O ` r/ Ul O
~ I` O N - ~ - CO C~ 5~1 N O ~ t~
1~ ~ I N _I
co co ~ ~ r7 ~D
I N ~ ~ ~ N~ _~
H ~r 1~ I N - CO - COOl~ 9 0 _I N 1~1 1~ _I
~ U~l _I Cl~ _I ~ ~1CO_I Cl~ N _I ~ ~ tq ~r N ~
1~ C~ o~
N N ~ N--i I ~ ~ N
~) ~ I N In ~ N :~J ~ N Ul _I
~ 2
a~ 1` 1
CD a~ o u~ o ~ u~
N _i N
I U~ ~ - - - ~1 ~r O N u~ N O ~
~ r O N - ~ NCOC~ ~1 a~ C~l O ~D _I N
N ~ I ~ N _I ~ N ~ r~
dP ~ 3
~P C ~ ..... ~ ~ ~
` . : - N N NN~--1 a ~ c
s ~q ~ 3 3 ~~ ~ rN E Y
~ u~ a) o ~ u~
8 ~ ~ ~ ~ _1 N _I N ~ 0 0 8
# ~ ~ ~ ~ ~ 3 ~ ~
-- 29 --
10~9~79
In all the runs 1 through 9 above, it can be seen
that the use of the catalyst described herein, in the staging
patterns described, resulted in maintenance of hot spot
temperatures generally below about 340C at throughputs as high
as 125~ of the design throughput. The hot spots were in con-
trollable locations and, in fact, as can be seen, on many
occasions they fell far below 300~C. Of e~ual importance, the
pressure drop remained substantially constant throughout the
reaction system, even in run 1, in which the run length totaled
over 300 hours. Thus, a pressure reduction was not required
to maintain the hot spot control and the pressure drop did not
increase over extended run times due to deterioration of the
catalyst. Furthermore, even when operating at high throughput,
the ethylene excess could be kept within reasonable limits with-
out substantially affecting hydrogen chloride conversion. As
seen from the data in Tables II and III, it was quite possible
to operate at ethylene excesses of about 10% while still
obtaining as high as 99+% hydrogen chloride conversion.
Additionally, the catalyst in the staging pattern,
herein described, has also been found suitable for operation
at less than design throughput. Thus the catalyst and system
herein described possesses versatility and can be utilized in
a plant operation in which the throughput can be variable,
depending on conditions such as feedstock supply and mar~et
demand, over a range from substantially less than design
capacity to substantially greater than design capacity, without
xequiring replacement or modification.
- 30 -
