WO 2005/021149 PCT/KR2004/002193
METHOD OF PRODUCING UNSATURATED ALDEHYDE AND UNSATURATED
ACID IN FIXED-BED CATALYTIC PARTIAL OXIDATION REACTOR WITH
ENHANCED HEAT CONTROL SYSTEM
Technical Field
The present invention relates to a process of producing
unsaturated aldehydes and unsaturated acids from olefing by
catalytic vapor phase oxidation, particularly a process of
producing acrolein and acrylic acid from propylens, as well
as a heat exchanger-type reactor for use in such a process,,
particularly a fixed bed shell-and-tube heat exchanger-type
reactor, The process of producing unsaturated aldehydes and
unsaturated acids corresponds to typical catalytic vapor
phase oxidation.
Background Art
Generally, catalytic vapor phase oxidation is carried
out by charging one or more kinds of granular catalysts into
a reactor tube (catalytic tube}, supplying feed gas into a
reactor through a pipe, and contacting the feed gas with the
catalyst in the reactor tube. Reaction heat generated during
the reaction is removed by heat exchange with a heat transfer
medium whose temperature is maintained at a predetermined
temperature, , The heat transfer medium for heat exchange is
provided on the outer surface of the catalytic tube so as to
perform heat transfer. The reaction mixture containing the
desired product is collected and recovered through a pipe,
and sent to a purification step.
Since the catalytic vapor phase oxidation is a highly
exothermic reaction, it is very important to control reaction

WO 2005/021149 PCT/KR2004/002193
temperature in a certain range and to reduce the size of" the
temperature peak at a hot spot occurring in a reaction zone.
It is impossible to satisfactorily control reaction
heat from the catalytic vapor phase oxidation only by uniform
circulation, of the heat transfer medium in the reactor, and
the serious temperature peak at the hot spot mentioned above
often occurs, causing excessive oxidation at local sites in
the reactor. This results ±n an increase in an undesirable
combustion reaction, thus reducing the yield of the desired
product, in addition, the catalyst is always locally exposed
to high temperature caused by the presence of the hot spot,
thug reducing the life cycle of the catalyst.
The partial oxidation of olefin uses a mixture of
molybdenum, and bismuth or vanadium oxide or mixed oxide
thereof, as a catalyst. Typical examples thereof include a
process for the production of acrolein or acrylic acid by the
oxidation of propylene, a process for the production of
phthalic anhydride by the partial oxidation of naphthalene or
orthoxylene, and a process for the production of maleic
anhydride by the partial oxidation of benzene, butylene or
butadiene.
Generally, accylic acid, a final product, is produced
from propylene by a two-step process of vapor phase catalytic
partial oxidation, In the first step, propylene is oxidized
by oxygen, dilution inert gas, steam and a certain amount of
a catalyst, so as to mainly produce acrolein, and in the
second step, the produced acrolein is oxidized by oxygen,
inert dilution gas, steam and a certain amount of a catalyst,
so as to produce acrylic acid, The catalyst used in the first
step is a Mo-Bi-based. oxidation catalyst which oxidises
2

WO 2005/021149 PCT/KR2004/002193
propylene to mainly produce acrolein. Also,, some acrolein is
continuously oxidized on such a catalyst to produce acrylic
acid. The catalyst used in the second step is a Mo-V-based
oxidation catalyst which oxidizes acrolein, mainly producing
acrylic acid,
A reactor for performing such a process is provided
either in such a manner that both the two-steps can be
performed in one catalytic tube or in such a manner that the
two steps can be perfonmed in different catalytic tubes, US
patent No. 4,256,783 discloses such a reactor,
Meanwhile, acrylic acid manufacturers now conduct
diversified efforts either to improve the structure of such a
reactor so as to increase the production of acrylic acid by
the reactor, or to propose the most suitable catalyst to
induce oxidation, or to improve process operations.
In part of such prior efforts, propylene which is
supplied into the reaction is used, at high apace velocity or
high concentration. In this case, there are problems in that-
rapid oxidation in the reactor occurs, making it difficult to
control the resulting reaction temperature, and also a high
temparature at hot spot in the catalyst layer of the reactor
is produced so as to increase reaction temperature, resulting
in an increase In the production of byproducts, such as
carbon monoxide carbon dioxide and acetic acid, thus
reducing the yield of acrylic acid,
Furthermore, in the case of producing acrylic acid
using high space velocity and high concentration of
propylene, as the reaction, temparature abnormally behaves in.
the reactor, various problems, such as the loss of active
ingredient a from the catalyst layer, a reduction in the
3

WO 2005/021149 PCT/KR2004/002193
number of active sites caused by the sintering of metal
components, are caused, thus deteriorating the function of
the catalyst layer.
Accordingly, in the production of acrylic acid, the
control of the heat of reaction in the relevant reactor is
important of all things- Particularly, not only the formation
of hot sports in the catalytic layer but also the accumulation
of heat around the, hot spots must be inhibited, end the
reactor must be effectively controlled such that the hot
spots do not lead to reactor runaway (a state where the
reactor is not controlled or explodes by a highly exothermic
reaction)-
Thus, it is very important to inhibit hot spots and
heat accumulation around the hot spots go as to extend the
life cycle of a catalyst and inhibit side reactions, thus
increasing the yield of product such as acrylic acid. To
achieve this inhibition, various attempts have been steadily
made.
A fundamental method is to form several catalyst layers
having activities that vary according to the moving direction
of reactants. Namely, at a reactor inlet side where hot spots
generate, a catalytic layer with low activity is formed, and
catalyst layers whose activities increase slowly toward a
reactor outlet side are formed. Typical methods for
controlling catalytic activity include: a method of making
catalytic particles by mixing a catalytic meterial with
inactive materials (e.g., US patent No, 3,301,634, Japanese
patent Ho, 53-306988, and Japanese patent No, 63-38831}; a
method of controlling activity and selectivity by either
changing the kind of alkali metals and controlling the amount
4

WO 2005/021149 PCT/KR2004/002193
thereof (e.g., US patent No. 4,837,360); a method of
controlling activity by adjusting the occupied volume of
catalytic particles (e.g., US patent Nos, 5,190,581 and
5,719,318); and a method for controlling activity by
controlling sintering temperature in the preparation of a
catalyst (e.g., US patent No. 6,023,220), However, such
methods have scene effects but still need to the improved.
The method of mixing the catalytic material with the
inactive materials is the simplest method to control
activity. However, in this method, the filling length of a
catalytic layer must be significantly lengthened due to the
use of the inactive materials and uniform mixing of the
catalytic particles and the inactive particles acts as
excessive load since it requires a significant effort and
time-
The method of using alkali metals is a significantly
good method since it allows not only the control of activity
out also an increase in selectivity. However, since alkali
metals in a catalytic composition cause a change in activity
and selectivity even at a very low amount, thare is a risk in
that a small error in the preparation of a catalyst will lead
to a great reduction in yield.-
Even when significant parts of the above mentioned
problems are solved, in order that the above technologies are
more effectively used, a reactor system needs to be designed
such that it is suitable for oxidation with excessive heat
generation. Particularly, in order to inhibit the
inactivation of a catalyst caused by excessive heat
generation, it is necessary to establish an efficient heat
control system capable of controlling a hot spot and runaway.
5

WO 2005/021149 PCT/KR2004/002193
For the establishment of the efficient heat control system,
studies have been performed on the introduction of a
perforated shield plate (e.g., US patent No, 4,256,763,
European patent No. 293224A, and Japanese patent No, 52-
839361, the establishment of circulation pathway of molten
salts by the placement of various baffles (e.g., US patent
No, 3,871,445), the design of an oxidation reactor integrated
with a cooling heat exchanger (e.g., US patent No.
3,147,004), a multi-stage heat control structure using an
improved heat exchanger system (e.g,, Korean patent
application Ho. 10-2002-40043, and PCT/KR02/02074), etc.
As described above, since the catalytic vapor phase
oxidation is an exothermic reaction, which, not only progresses
at high temperature but also has excessive heat generation,
it can cause a reduction in selectivity due to the generation
of a hot spot with very high temperature around a reactor
Inlet or the accumulation of heat around the heat spot, and
also can result in a significant reduction in the performance
of a catalyst in a long-term viewpoint, Particularly, if the
temperature of the hot spot is higher than the calcination
temperature of the catalyst , the life cycle of the catalyst
will be reduced to shorten the replacement time of the entire
catalyst layers, resulting in an economical loss. In
addition, if the activity of the filled catalyst layers is
reduced and the temperature of a heat transfer medium (molten
salt) is elevated in order to compensate for the seduction in
activity, the temparature of a hot spot and the accumulation
of heat in the hot spot will also be increased, and as a
result, a solution to solve this problem is required.
Brief Description of the Drawings
6

WO 2005/021149 PCT/KR2004/002193
FIG. i is a schematic diagram showing the structure of
a catalyst layer in a pilot reactor in which first-step
reaction and second-step reaction are successively performed
in one catalytic tube.
FIG. 2 is a schematic diagram showing the structure of.
a pilot reactor consisting of two catalytic tubes, and the
stricture of a catalyst layer in each of the catalytic tubes.
First-step reaction and second-step reaction is conducted in
the two catalytic tubes, respectively,, and a partition is
placed at a boundary between the catalyst layers in first-
step reactor.
FIG. 3 is a schematic diagram showing the structure of
a pilot reactor consisting of two catalytic tubes, and the
structure of a catalyst layer in each of the catalytic tubes -
First-step reaction and second-step reaction is conducted in
the two catalytic tubes, respectively, and a partition. is not
placed at a boundary between the catalyst layers in first
step reactor.
FIG. 4 is a schematic diagram showing the structure of
a pilot reactor consisting of two catalytic tubas , and the
structure of a catalyst layer in each of the catalytic tubes.
First-step reaction. and second-step reaction is conducted in
the two catalytic tubes, respectively, and a partition, is not
disposed in the pilot reactor,
Disclosure of the Invention
Technical Problem
In view of the above -mentioned problems occurring the
prior art, the present inventors have made improvements in a
fixed-bad shell-and-tube , heat exchanger-type reactor of

WO 2005/021149 PCT/KR2004/002193
producing unsaturated aldehydes and unsaturated acids from
olefins either using a single reactor where two-step reaction
is conducted or using two reactors which are connected in
series with each other. In the inprovements, the first-step
reaction zone of the reactor was divided into two or more
holies in an axial direction by a partition, and the
temperature of a heat transfer medium filled in each of the
divided shell spaces of the first-step reaction zone was
independently set to a temperature suitable for the activity
of a catalyst and the degree of reaction. As a result of such
improvements, the present inventors have found that a hot
spot and heat accumulation around the hot spot could be
inhibited. The present invention was perfected based on this
finding.
An object of the present invention is to provide a
production process in which the temparature difference
between the peak temparature of a catalyst layer in each of
the divided reaction zone and the temperature of a heat
transfer medium (molten salt} filled, in the shell space
corresponding to that zone is controlled, so that the process
is not only thermally stable but also can be operated without
a reduction in yield, even in the presence of a catalyst with
very high activity, as well as an improved heat exchange-type
reactor for use in this process,
Another object of the present invention is to provide a
production process in which a reaction inhibition layer is
inserted in order to facilitate the removal of heat
generation at a location where the partition is placed, as
well as a shell-and-tube heat exchanger-type reactor for use
in this process.
8

WO 2005/021149 PCT/KR2004/002193
Technical solution
In one aspect, the present invention provides a process
of producing unsaturated aldehydes and unsaturated acids from
olefins, particularly a process of producing acrolein and
acrylic acid from propylene, by fixed-bed catalytic partial
oxidation in a shell-and-tube heat exchanger-type reactor,
the reactor comprising one or more catalytic tubes each
including a first—step reaction zone of mainly producing the
unsaturated aldenydes., a second-step reaction zone of mainly
producing the unsaturated acids, or both the two zones, the
impressment wherein: the first-step reaction zone is divided
±nto two or more zones by a partition, each of the divided
shell spaces being filled with a heat transfer medium, the
heat transfer medium being maintained at isothermal
temperature or a temperature difference of 0-5°C, in which
the temperatures of the heat transfer medium in each of the
divided shell spaces are set to increase ,in the moving
direction of reactants, and/or a difference between the
temperature of the heat transfer medium and the temperature
of a hot spot is limited, and/or a reaction inhibition layer
is inserted into a location where the partition is placed.
In another aspect,the present invention provides a
shell -and-tube heat exchanger-type reactor which can he used
in a process of producing unsaturated aldehydes and
unsaturated. acids from olefins by fixed-bed catalytic partial
oxidation, the reactor comprising one or more catalytic tubes
each including a first-step reaction zone of mainly producing
the unsaturated aldehydes , a second-step reaction zone of
mainly producing the unsaturated acids, or both the two
zones, the improvement wherein the first-step reaction sons
9

WO 2005/021149 PCT/KR2004/002193
is divided into more than two zones by a partition, each of
the divided shell spaces being filled with a heat transfer
medium, the heat transfer medium baing maintained at
isothermal temperature or at a temperature difference of 0-5
OC, in which the temperatures of the heat transfer medium in
each of the divided shell spaces are set to increase in the
moving direction of reactants, and/or a difference between
the temperature of the heat: transfer medium and the
temperature of a hot spot is limited, and/or a reaction
inhibition layer is inserted into a legation where the
partition is disposed.
As used herein, the term "each of the shell spaces"
indicates an internal space surrounded by a catalytic tube, a
shell, a partition, a tube sheet, etc.
As described below, the present invention makes
improvements in the first-step reaction region of mainly-
producing unsaturated aldehydes and unsaturated acids, and
aims to use an improved heat control system to inhibit a hot
spot and heat accumulation at the hot spot and to increase
the yield in so intermediate, step, and at the same time, to
make stable the reaction in the second-step reaction zone
while increasing final yield.
In the inventive production process and heat exchanger
type reactor, the temperature of the heat transfer medium in
each of the divided shell spaces is set as nearly as possible
to isothermal conditions. according to the amount of heat
generation and the capacity of the heat transfer medium, the
temperature difference between portions of the heat transfer
medium, which correspond to both the ends of a catalyst layer
in each of the divided shell spaces, is preferably 0-5°C, and
10

WO 2005/021149 PCT/KR2004/002193
more preferably 0-3 OC.
Examples of the heat transfer medium include very
highly viscous media, for example a molten salt which
consists mainly of a mixture of potassium nitrate and sodium
nitrite. Other examples of the heat transfer medium include
phenyl ether media (e.g., "Dowtherm") , polyphenyl media
(e.g., "Therm S"}, hot oil, naphthalene derivatives (S.K.
oil) and mercury.
By controlling the flow rate of the heat transfer
medium, the reaction throughout the tube corresponding to
each of the shell spaces in the reactor can be carried out at
substantially the same temperature of the heat transfer
medium.
The temperature of the heat transfer media in the
adjacent shell spaces in each of the reaction zones is
preferably set to increase in an axial direction by a
temperature difference of. 0-50 OC, and more preferably 5-15
OC.
As the temperature of the heat transfer medium filled
in each of the shell spaced increase in the moving direction
(hereafter, referred to as "axial direction") of reactants,
the reactivity of the catalyst layer increases in the axial
direction.
If the temperature of the heat transfer medium in each
of the shel1 spaces divided by the partition increases in the
axial direction, a hot spot and heat accumulation around the
hot spot can be inhibited.
Since a reactor front portion with high react ant
concentration and high reaction pressure has the highest
reactivity, a hot spot with significantly high temperature is
11

WO 2005/021149 PCT/KR2004/002193
formed in the reactor front portion. If the temperature of
the heat transfer medium in a portion of the shell space,
which corresponds to a location where this hot spot is
formed, la decreased to the lowest possible active
temperature; the size of the hot spot can be reduced and heat
accumulation around the hot spot can be prevented while
causing no significant reduction in reactivity.
In the case of inducing a continuous increase in
temperature without the partition, excessive efforts are
required in ordar to remove reaction heat by catalytic
oxidation or to inhibit heat accumulation, and it is very
difficult to exactly set the desired temperature profile.
The location of the partition is preferably established
based on the exact prediction of a position where a hot spot
is formed.
A hot spot is produced by the generation of reaction
best resulting from catalytic vapor phase oxidation, and
determined by the composition of reactants, the flow rats
(L/min) of reactants, the temperature of a heat transfer
medium, etc., and has a certain position and size in a
certain process condition. However, the activity of a
catalyst can change with time.
In portions where heat control is problematic, a hot
spat can be generated in the front portion of a first-step
oxidation catalyst layer, in which olefin (propylene) , a main
reactant,and molecular oxygen , are present at high
concentrations. Also, if two or more catalyst layers are used
in the first-step oxidation zone, a hot spot can be generated
around the boundary between the adjacent catalyst layers-
The partition is preferably located at either a
15

WO 2005/021149 PCT/KR2004/002193
position where a hot spot or heat accumulation resulting from
the hot spot sure problemaqtic, or a position allowing the
largest possible removal of heat generation in each zone.
A hot spot is the site of the maximum temperature peak,
and each of catalyst layers generally has at least one hot
spot,
Also, it is preferred that the temperature of the heat
transfer medium (molten salt or heat transfer salt) in each
of the divided shell spaces is set in such a manner that a
catalyst has optical activity.
Particularly in the present invention, in order to
inhibit a hot spot and heat accumulation resulting from the
hot spot either in a catalytic tube for each reaction step or
in each reaction zone in one catalytic tube under high olexin
concentration or high oleffin space velocity, the temparature
of the heat transfer medium is changed ±n the axial direction
so as to reduce catalyst damage caused by a highly exothermic
reaction and to inhibit a reduction in yield caused by side
reactions, resulting in an increase in yield.
The peasant invention provides a production process and
raactor, wherein, when the shell spaces divided by the
partition in the first-step reaction zone are named, such as
zone 1, zone 2, zone 3, ,„ in the axial direction, Tb1-Tsatl is
< 150 °C, and preferably Tb1-Tsalt is < 110 °C, and Tb1-Tsatl is <
120 OC, and more preferably Tb1-Tsalt is <, 100 0C where N is an
integer of 2 or more
Here, Th1 is the peak temperature of a reaction mixture
in a catalyst layer corresponding to the first shell apace
(the peak temperature of "the catalyst layer; , and ThN is the
peak temperature of a reaction mixture in a catalyst layer
13,

WO 2005/021149 PCT/KR2004/002193
corresponding to the Nth shell space. And Tsalt is the
temperature of a beat transfer medium (molten salt) filled in
the first shell space, and Tsalt is the temperature of a heat
transfer medium filled in the Nth shell space.
In the first shell space,. the concentration and
pressure of reactants are high, so that the temperature
difference between, the peak temperature of the catalyst layer
and the temparature of the molten salt is higher than that in
the next shell space. For this reason, the temperature
difference range in the first shell space will be surely
wider than those in the next shell spaces, However, the
present invention provides a method by which the magnitude of
peak temperature in the first shell space- is minimised while
a temperature difference in the next shell space is also
limited to a certain range, so as to prevent local excessive
heat; generation, thus making smooth the shape of temparature.
profile, in which the limited temperature range is the result
of various experiments conducted over several years by the
present inventors. If operations are done without such a
limited range, the sintering of a catalyst in a hot spot, the
loss of important metal components, an increase in
byproducts, etc., can be caused, and particularly accidents
can also occur clue to a sudden exothermic reaction when
introducing raw materials into a, reactor. For thsse reasons,
the inventivs method is technologically necessary for safe
start-ups, stable operations, and safe shut-downs,
According to the present invention, the temperature
difference between the peak temperature of a catalyst layer
in each reaction zone and the temperature of a heat transfer
medium (molten salt) is controlled in the above specified

WO 2005/021149 PCT/KR2004/002193
range, so that the catalyst can show uniform activity in the
axial direction, and the degree of reaction can be suitably
controlled, so as to inhibit heat accumulation in a hot spot
and suppress side reactions, thus preventing a reduction in
yield.
Accordingly, the present invention can be stably
operated at high olefin concentration or high space velocity
without the control of activity of a catalyst filled in the
first-step reactor. Meanwhile, in the production of
unsaturated aldehydes and unsaturated acids from olefins, the
olefins can be introduced into a reactor inlet at a space
velocity of 50-130 hr-1.
The method of filling a plurality of catalyst layers
with varying ac±±vities in the first-step reaction zone can
achieve better performance by reducing the magnitude of peak
temperature at a hot spot and inhibiting heat accumulation.
However, in such a method, there is a shortcoming in that
different kinds of catalysts must be prepared and filled
separately, thus causing the problem of an increase in
catalyst costs. Furthermore, since it is very difficult to
control not only the size and shape of a catalyst
corresponding to each of the catalyst layers but also the
content of alkali metals and alkaline earth metals, it is a
great load to prepare various kinds of catalysts in such a
manner that the catalysts have uniform performances every
time. On the other hand, in the present invention , a multi-
step heat control system is applied even in the catalyat
layers with the same activity, so that a process and reactor
of producing acrolein and acrylic acid in an efficient and
stable manner can be provided.
15

WO 2005/021149 PCT/KR2004/002193
according to one embodiment of the present Invention,.
improvements are made in a process of producing acrolein and
acrylic acid by reacting propylene with molecular oxygen,
dilute inert gas and steam under a catalyst, optionally with
recycled off-gas which had not been absorbed into an
absorbing column. in the improvements the first—step
reaction step of producing acrolein and acrylic acid is
divided into two or more separate shell spaces in an axial
direction, and the temperature of a heat tansfer medium
filled in each of the shell spaces is set to increase in the
axial direction, so that the reactivity of the catalyst
layers increases in the axial direction.
In the first-step reaction zone where acrolein is
mainly produced from propylene, the temperature of the
catalyst layer with activity is set to about 290-420 OC and
the temperature of the heat transfer medium is get to about
2 90—350°C. In this reaction zone, acrylic acid, carbon
monoxide, carbon dioxide, steam, acetic acid and small
amounts of byproduct a are produced in addition to acrolein.
Since the oxidation in this step is a reaction which
progresses at high temperature and is highly exothermic, heat.
of 8l kcal per g-mol of propylene occurs and the temperature
in a hot spot reaches 370-400°C- In the second-step reaction
zone where actylic acid is mainly produced from acrolein, the
temperature of the catalyst layer with activity is set to
about 260-360OC and the temperature of the heat transfer
medium is set to about 230-330°C. In this second reaction
zone, unreacted acrolein, carbon monoxide, carbon dioxide,
steam, acetic acid, unreacted propylene and small amounts of
byproducts are produced in addition to acrylic acid Since
16

WO 2005/021149 PCT/KR2004/002193
the second-step reaction is also a reaction which progresses
at high temperature and is highly exothermic, heat of 60 kcal
per g-mol of acrolein occurs and the temperature in a hot
spot reaches 310-3500C,
In the structure of the reactor according to the
present invention, the first-step reaction zone is divided
into two or more shell spaces by a partition such that the
temperature of a heat transfer medium filled in each of the
divided shell spaces can be controlled independently. The
temperature of the heat transfer medium filled, in. each shell
space is set to increase in an axial direction. For example,
if the shell space corresponding to the first-step reaction
region is divided, into three separate spaces, two partitions
will be vertically disposed to an axis of catalytic tube
inside the shell space to provide a structure with, three
shell spaces in which the temperature of the heat transfer
medium is controlled independently. The heat transfer mediums
filled in the spaces are set to increasing temperatures of,
for example, 300°C, 310°C and 315 0c, respectively, in an
axial direction from the reactor inlet to the outlet.
Meanwhile, the catalyst layer in the fitst-step
reaction zone may consist of one layer with axially uniform
activity, or if necessary, two or more stacked layers with
increasing activity toward the outlet. The catalyst layer in
the second-step reaction zone may consist of one layer with
axially uniform activity, or if necessary,, two or more
stacked layers with increasing- activity- toward the outlet.
The number of catalytic tubes in a commercial shell-
and-tube reactor of producing acrylic acid reaches several
thousands to several tens of thousands, and a partition
17

WO 2005/021149 PCT/KR2004/002193
disposed in the reactor has a very large "thickness of 50-100
mm. Thus, if the shall space in each of the two-step reaction
zones is divided into two ox more layers, the removal of heat
generation due to reaction in a position where a partition is
disposed is not easy, thus causing a problem in heat
transfer. To eliminate such a problem, the present invention
is also characterised by providing a layer made of an
inactive material alone or a mixture of an inactive material
and a catalytic material, i.e., a reaction inhibition layer,
within a portion of the catalytic tube, which corresponds to
a position where the partition is disposed.
Such a reaction inhibition layer is a layer with
different characters from an inactive layer which is filled
between the first step (propylene to acrolein) and second
reaction, step (acrolein to acrylic acid) to a thickness of
about 400-1,000 urn so as to induce cooling to a reaction
temperature suitable for the second-step reaction. This
reaction inhibition layer is a filling layer for minimizing
neat generation in a position where heat transfer is
problematic. The volume ratio of an inactive material to a
catalytic material -in this reaction inhibitian layer is 20-
100%, and preferably 80-100%. The filling height of the
reaction inhibition layer is 20-500%, and preferably 120-150%
of the thickness of the partition, indicating that the
reaction inhibition layer completely overlaps the thickness
of the partition. However, if the height of the reaction
inhibition layer must be made smaller than the thickness of
the partition, it is preferably filled in such a manner that
the largest possible area overlaps.
The inactive material -used in the reaction inhibition
18

WO 2005/021149 PCT/KR2004/002193
layer is designated as a material which is inactive to a
reaction of producing unsaturated aldehydes and unsaturated
acids from olefins, for example, a catalytic oxidation of
producing acrolein and acrylic acid from propylene. It can be
used in a sphere, cylinder, ring, rod, plate or wire mesh
shape, or a mass shape with suitable size, or a suitable
combination thereof. Widely known examples of the inactive
arterial include alumina, silica, stainless steel, iron-,
steatite, porcelain, various ceramics, and mixtures thereof.
The catalytic tube in the reactor may comprise one or
more catalyst particle layers for each oxidation Step,
Advanced Effect
As described above, the present invention provides the
improved heat control system for use in the two—step process
of producing unsaturated aldehydes and unsaturated acids by
the oxidation of olefins. This heat control system allows the
design of a reactor system to which the advantages of the
prior art can be sufficistit.ly applied even, under reaction.
conditions with high load. If this heat control system is
used, the formation of a hot spot or heat accumulation in the
hot spot can be inhibited, and as a result , unsaturated
aldehydes and unsaturated acid3 can be produced at high
productivity and also the life cycle of a catalyst can be
extended.
Mode for Invention
Hereinafter the present invention will be described in
detail with reference to the accompanying drawings. FIGS. 1
19

WO 2005/021149 PCT/KR2004/002193
to 4 show a pilot structure designed to transfer the idea of
the present invention, and the scope of the present invention
is not limited only to details shown in the drawings.
It is known that reactor behavior characteristics, such
as temperature, yield,, etc, in an actual shell-and-tube heat
exchanger-type reactor, can be represented by a reactor with
one catalytic tube. Thus, the effects of the present
invention will be proven by a pilot experiment with one
catalytic tube for each reaction step.
For the description below, 3 catalyst layer for each
step is named as follows, and the following sequence
coincides With a, reaction pathway:
The first catalyst layer in the first-step reaction: a
first step-layer A;
The second catalyst layer in the first-step reaction: a
first step-layer B;
The third catalyst layer in the first-step reaction: a
first step-layer C;
The first catalyst layer in the second-step reaction: a
second step-layer A;
The second catalyst layer in the second-step reaction:
a second step-layer B;
The third catalyst layer in the second-step reaction: a
second step-layer C;
If necessary, the catalyst layers may be disposed in
such a manner that their catalytic activity gradually
increases toward the layers A, B, C,
FIG, 1 shows the structure of a pilot reactor in which
20

WO 2005/021149 PCT/KR2004/002193
two-step reaction occurs in one catalytic tube, As shown in
FIG. l, a first-step reaction zone 10 end a second-step
reaction zone 20 are connected in series vith each other,
such that reactants fed into a reactor inlet are subjected to
first-step reaction and then to second-step reaction, thus
producing acrylic acid. If necessary, two or more catalyst
layers with different activities {except for an inactive
material layer and a reaction inhibition layer) can be
included in each of the reaction zones.
Hereinafter, a detailed description on a reaction
system based on two catalyst layers with different activities
for each reaction step will be made by way of an example.
Reference numerals 11, 12 and 21 in Fig. 1 denote shell
spaces (jackets) into which heat transfer media with
different temperatures are filled. FIG. 1 shows the structure
of the catalyst layers in the catalytic tube, and the
following layers are filled in the catalytic tube in an order
from the lower level to the upper level;
Inactive particle-layer A 16
First-step reaction zone:
First step-layer A 14
First step-reaction inhibition layer 17
First step-layer B 15
Inactive particle-layer B 31
Second-step reaction zone:
Second step-layer A 24
Second step-layer B 25
The first step-layer A and the first step-layer 5 can
be catalyst layers with the same or different activities, The
second step-layer A and the second step-layer B can be
21

WO 2005/021149 PCT/KR2004/002193
catalyst layers with the same or different activities.
Between the first-step reaction zone and the second-step
reaction zone, inactive particles (inactive particle-layer B)
are suitably filled such that the temperature of a reaction
mixture entering the second step is in the range of
activation temperature of second step-layer A. The shell
space in the first-step reactor is divided into two heat
control spaces which are heat-controlled independently. As
shown in FIG. l, a partition 13 in the first-step reaction
zone is located at a boundary between the catalyst layers.
Reference numeral 30 in FIG. 1 denotes a partition of
providing a division between the first-step and second-step
reaction zones, and the inactive material layer 31 is a
filling layer of inducing reactants to be cooled to a
temparature suitable for the catalyst layer 24 in the second-
step reaction zone.
Reference numeral 1 in FIG. 1 denotes the flow of
reactants consisting of propylene, molecular oxygen, dilute
gas, and steam. Reference numeral 3 denotes the flow of
Products consisting of main product acrylic acid, dilute gas,
molecular oxygen, unreacted propylene, unreacted acrolein,
and small amounts of byproducts, etc.
Reference numerals 11 and 12 in FIG- 1 denote two
divided shell spaces (jackets) in the first-step reaction
zone, and reference numeral 21 in FIG. 1 denotes a shell
space (jacket) in the second-step reaction zone.
FIG. 2 shows the structure of a pilot reactor in which
a first-step reaction zone and a second-step reaction zone
are divided from each other. In FIG, 2, the fundamental
structure of the reactor, and the structure of the catalyst
22

WO 2005/021149 PCT/KR2004/002193
layer, are the same as those in. FIG- 1.
FIG, 3 shows a structure where two or more catalyst
layers with different activities are filled in each of the
reaction zones. In this structare, a partition is placed away
from a boundary between the catalyst layers. Also, this
structure can be uaed when first heat control zone is defined
by a section, ranging from a reactor inlet to the peak
temperature zone of the first step-layer B, and the second
heat control zone is defined by the remaining section. Also,
this structure can be applied when two or more adjacent
temperature peaks are present in the catalyst layer so that
such temperature peaks need to be controlled at the same time
using one shell space (jacket).
For example, peak temperatures occurring in the first
step-layer A and the first step-layer B can be controlled
below a predetermined peak -temperature of catalyst layers to
be managed, by controlling the temperature of a heat transfer
medium filled in the first heat control zone. In FIG. 3,. a
method of positioning the partition, and a method of filling
the catalyst and the inactive material, can also be applied
to the reactor structure with one catalytic tube as shown in
FIG. 1, in the same principle,
In the location of the partition in the catalyst layer,
the partition can be disposed between the preceding catalyst
layer and the relevant catalyst layer in the filling order,
and a second partition can be disposed following the peak
position of the relevant catalyst layer.If the relevant
catalyst layer is a first catalyst layer either at a reactor
inlet or at the initiation point of each of the reaction
sones, one partition will be disposed following the peak
23

WO 2005/021149 PCT/KR2004/002193
point of the relevant catalyst layer. The shell spaces
divided by the partition may cover the position of one or
more peak temperatures occurring in a plurality of the
catalyst layers
The inventive haat control system can be applied in the
oxidation of olefins, and also in a reaction system where the
kind of faction, varies in an axial direction 30 that it
progresses for each step, as wall as a system where reaction
temperature must be changed according to reaction zones so
that it is controlled at the optimal tempersture,
Example 1 (improved heat control system) : changes in
yield and the magnitudes of temperature peaks at hot spot a
with change In setting temperature af molten salt
Aa shown in FIG, 3, a pilot reactor was provided in
which each of" first-step reaction and second-step reactions
ia conducted in one catalytic tube (included in the zone 10
or 20 of FIG. 3) . Each of the catalytic tubes is 26 mm in
inner diameter, and the first-step catalytic tube was filled
with catalytic layers with a height of about 1200 mm, and the
second-step catalytic tube was filled with catalytic layers
with a height of about 1100 mm. Reference numerals 11 and 12
in FIG. 3 denote the divided shell spaces of the first-step
reaction zone. The temperatures of molten salts filled in the
shell spaces are 300 OC and 305 OC, respectively. Reference
numeral 21 ±n FIG. 3 ±g a shall space into which a molten
salt is filled and set to 265 °C. The two catalyst layers
filled in the first-step reaction zone 10 are made of a
catalyst based on molybdenum (Mo) and bismuth (Bi), the
preparation of which is described in Korean patent No.
0349602 (Application No. 10-1997-0045132) - The two catalyst
24

WO 2005/021149 PCT/KR2004/002193
layers filled in the second-step reaction zone 20 are made of
a catalyst, based on molybdenum and vanadium (V) , the
preparation of which is described in Korean patent No.
0204728 or Korean patent No, 0204729,
The Catalytic tube for each reaction step was filled
with two catalyst layers whose activity gradually increases
from an inlet to an outlet, according to a method for
controlling catalytic activity as disclosed in US patent Nos,
3801634 and 4337360.
The catalytic tube in the first-step reaction zone was
filled with two catalyst layers having 320mm and 880 mm
respectively, in an axial direction, and the catalytic tube
in the second-step reaction, zone were filled with two
catalyst layers having 290 mm and 320 mm, respectively, in an
axial direction, A partition was disposed at the 600 mm
position of the first-step reaction zone, such, that it
covered both temperature peaks occurring in the two catalyst
layers. In a portion inside the catalytic tube corresponding
to the position of the partition, an inactive material layer
was filled to a thickness corresponding 120% of the thickness
of the partition, A pipe inducing a. flow represented by
reference mineral 2 in FIG. 3 serves to connect the two
catalytic tubes and is surrounded by an insulation material.
Starting material comprising propylene, steam, oxygen and
inert gas enter the reactor through a line 1, passes through
the reaction steps , and then flows out from the reactor-
through a line 3. The starting materials consist of
propylene, oxygen, steam and nitrogen gas, in which the
amount of propylane is 7% and the ratio of oxygen and
propyleps is about 1.80. Space velocity is 1400 hr-2 (STP) in
25

WO 2005/021149 PCT/KR2004/002193
the first-step reaction zone, and space velocity is 1530 hr-1
(STP) in the second-step reaction zone. Also, the space
velocity of olefin entering the first—step reaction zone is
98 hr-1 {STP) ,
In the zone corresponding to the first shell space ±n
the first-step reaction zone, a hot spot with a temperature
of 381.5 °C was farmed. The yields of acrolein and acrylic
acid were 8l.17% and 8.34%, respectively. In the second-step
reaction zone which is operated at isothermal conditions, the
temperature of a hot spot was 327 oC, and the yields of
acrolein and acrylic acid were 0.553% and 84.01%,
respectively.
Since a reaction In a reaction inhibition layer
(inactive material layer) did not occur, an abnormal increase
in temperature was not observed by a reduction in heat
transfer efficiency.
Example 2 (improved heat control system) : changes _in
yield and the magnitudes of temperature peaks at hot spots
with change in setting temperature of molten salt
This example was performed in the same manner as in
Example 1 except for the setting temperatures of a molten
salt in the first-step reaction zone (first-step reactor) .
The temperatures of the molten salt in the first-step
reaction zone were set to 300 °C and 310 °C, respectively, in
an axial direction.
In the zone corresponding to the first shell space in
the first-step reaction zone, a hot spot with a temperature
of 381-5 0C was formed. The yields of acrolein and acrylic
acid were 81.13% and 9.30%, respectively. In the second-step
reaction zone which is operated at isothermal conditions, the
26

WO 2005/021149 PCT/KR2004/002193
temparature of a hot spot was 320.0 °C, and the yields of
acrolein and acrylic acid were 1.18% and 84.35%,.
respectively,
Example 3 (improved beat control system) : changes in
yield and the magnitudes of temperature peaks at hot spots
with change in setting temperature of molten salt
This example was performed in the same manner as in
Example 1 except for the setting temperatures of a molten
salt in the first-step reaction zone (first-step reactor).
The temperatures of the molten salt in the first-step
reaction zone were set to 300 OC and 315 °c, respectively, in
an axial direction.
In. the zone corresponding to the first shell apace in
the first-step reaction zone, a hot spot with a temparature
of 381-2 OC was formed. The yields of acrolein and acrylic
acid were 79.02% and 11-46%, respectively. In the second-step
reaction zone which is operated at isothermal conditions, the
temperature of a hot spot was 327.5 °C, and the yields of
acrolein and acrylic acid were 0.607% and 84.95%,
respectively
Example 4 (improved heat control system) : changes ±n
yield and magnitudes of temparature peaks at hot, spots with
change in setting temperature of molten salt
This example was performed in the same manner as in
Example 1 except for the setting temperatures of a molten
salt in the first-step reaction zone [first-step reactor)
The temperatures of the molten salt in the first-step
reaction zone were set to 300 °C and 320 0C, respectively, in
an axial direction.
In a zone corresponding to the first shell apace in the
27

WO 2005/021149 PCT/KR2004/002193
first-step reaction zone, a hot spot with a temperature of
381.2 0C was formed. The yields of acrolein and acrylic acid
ware 78.37% and. 11.45%, respectively. In the second-step
reaction zone which is operated at isothermal conditions, the
temperature of a hot spot was 327.0 °C, and the yields of
acrolein and acrylic acid were 0.607% and 84.88%r
respectively.
Example 5 (improved heat control system) : changes in
yield and magnitudes of temperature peaks at hot spots with
change in setting temperature of molten salt
This example was performed in the same manner as in
Example 1 except for the setting temperatures of a molten
salt in the first-step reaction zone (first-step reactor) .
The temperatures of the molten salt in the first-step
reaction zone were set to 308 OC and 3l5 OC, respectively, in
an axial direction.
In a zone corresponding to the first shell space in the
first-step reaction zone, a hot spot with a temperature of
392.5°c was formed. The yields of actolein and acrylic acid
were 80-33% and ll.37%,, respectively. In the second-step
reaction zone which is operated at isothermal conditions, the
tempera.ture o£ a hot spot was 320.5OC and the yields of
acxolein and acrylic acid were 0.631% and 36.834,
respectively.
Example 6 (a case where each of reaction zones is
filled with catalysts having the same activity and a multi-
step heat control system is applied}.
Of two catalysts used in the first-step reaction zone
in Example 1, a catalyst with higher activity was filled in
the fitst-step reaction zone with a height of 1200 ran. Also,
28

WO 2005/021149 PCT/KR2004/002193
of two catalysts used In the second-step reaction zone in
Example l, a catalyst with higher activity was filled in the
second-step reaction zone with a height of 1100 mm. Example
6 was performed in the same manner as in Exanple 1 except for
the structure of the catalyst layer filled in each reaction
Zone, and the setting temperature of a molten salt- The
temperatures of molten salts in two shell spaces
corresponding "to the first-step reaction zone were set to 295
°C (a shell space carresponding to the first zone) and 305 oC
(a shell space corresponding to the second zone). The
temperature of molten salts in the second-step reaction zone
was set to an isothermal temperature of 265 °C.
In the first-step reaction zone which is operated at
isothermal conditions , the temperature of a hot spot in the
first zona was 392,3 °C, and the temperature of a hot spot in
the second zone was 363.6 0C. The yields of acrolein and
acrylic acid were 79.23% and 11.08%, respectively. In the
second-step reaction zone which is operated at isothermal
conditions, the yields of acrolein and acrylic acid were
0.704% and 85.54%, respectively.
Th1-Tsalt was 97.3 OC, and Th1-Tsalt was 53.6 OC.
Comparative Exanple 1 (the case of operations under
isothermal conditions without the application, of a multi-step
heat control system) : Changes in hot spot temperature and
yield with change in setting temperature of molten salt
Comparative Example 1 was performed in the same manner
as in Example 1 except for the setting temperature of molten
salts in the first-step reaction zone (first-step reactor]
(see FIG. 4). The temperature of the molten salts in the
first-step reaction zone was set to an isothermal temperaturs
29

WO 2005/021149 PCT/KR2004/002193
of 300 aC, and no partition was disposed. The temperature of
molten salts in the second-step reaction zone was set to an
isothermal temperature of 265 OC.
In the first-step reaction zone which is operated at
isothermal conditions, the temparature of a hot spot was
383.8 oC, and the yields of acrolein and acrylic acid were
81.3% and 8.18%, respectively. In the second-step reaction
zone which is operated at isothermal conditions, the
temperature of a hot spot was 320.1 °C, and the yields of
acrolein and acrylic acid ,were 1.583% and 83.11%,
respectively.
Comparative Example 2 (the case of operation under
isothermal conditions without the application of a multi-step
heat control system) : Changes in hot spot temperature and
yield with change in setting temperature of molten salt
comparative Example 2 waa performed In the same manner
as In Comparative Example 1 except for the setting
temperature of molten salts in the first-stop reaction zone
(first-step reactor), The temperature of the molten salts in
the first-step reaction zone was get to an isothermal
temperature of 305 °C
In the first-step reaction zone which is operated as
isothermal conditions, the temperature of a hot spot was
394.6 °C, and the yields of acrolein and acrylic acid were
81.91% and 8.35%, respectively. In the second-step reaction
zone which is operated at isothermal conditions, the
temparature of a hot spot was 320.3 °C, and the yields of
acxolein and acrylic acid were 1.424% and 84.07%,
respectively.
Comparative Example 3 (the case of_operations under
30

WO 2005/021149 PCT/KR2004/002193
isothermal conditions without the application of a multi-step
heat control system) : changes in hot spot temperature and
yield _with change in setting temperature of molten salt
Comparative Example 3 was performed in the same manner
as in Comparative Example 1 except for the setting
temperature of molten salts in the first-step reaction zone
(first-step reactor) . The temperature of the molten salts in
the first-step reaction zone was set to an isothermal
temperature of 310 0C,
In the first-step reaction zone which is operated at
isothermal conditions, the temperature of a hot spot was
405.7 °C, and the yields of acrolein and acrylic acid were
80.43% and 10.11%, respectively. In the second-step reaction
zone which is operated at isothermal conditions, the
temperature of a hot spot was 316.0 °C, and the yields of
acrolein and acrylic acid were 1.257% and 84.66%,
respectively.
Comparative Example 4 (a case_ where each of reaction
zones is filled with catalysts. having the same activity) :
changes in hot spot temparatura and yield with change in
setting temperature of molten salt
The first-step reaction zone was filled with the
catalyst used in Example 6 to a height of 1200 mm. Also, the
second-step reaction zone was filled with the catalyst used
in Example 6 to a height of 1100 mm, Comparative Example 4
was performed in. the same manner as in Example 6 except for
the setting temperature of molten salts in each of the
reaction zones, The temparature of molten salts in the first-
step reaction zone was set to 300 °C, and the temperature of
molten salts in the second-step reaction zone was set to 265
31

WO 2005/021149 PCT/KR2004/002193
°C. In Comparative Example A, no partition was disposed and a
reaction Inhibition layer was used.
In the first-step reaction zone which is operated at
isothermal conditions, the temperature of a hot spot was
407,9 °C, and the yields of acrolein and acrylic acid were
80.92% and 9.09%, respectively. In the second-step reaction
zone which is operated at isothermal conditions, the yields
of acrolein and acrylic acid were 0.807% and 84.21%,
respectively,
Th1-Tsalt was 107,9 0C
Comparative Example 5 (a cage where each of reaction
zones is filled with catalysts having the same activity)
changes in hot spot temperature and yield with change in
setting temperature of molten salt
Comparative Example 6 was performed in the same manner
as in Comparative Example 4 except for the setting
temperature of molten salts in each of the re-action zones.
Tne temperature of molten salts salts in. each of the reaction zone
was get to 305 °C, and the temperature of molten salts in the
second-3tep reaction zone was set to 265 o C.
In the first-step reaction zone which is operatad at
ieothermal conditions, the temperature of a hot spot was
410.3 oC, and the yields of acrolein and acrylic acid were
80.77 and 9,15%, respectively .In the second-step reaction
zone which is operated at isothermal conditions r the yields
of acrolein and acrylic acid were 0.934% and 84.30%,
respectively.
Th1-Tsalt was 113.3 oC-
Comparative Example_ 6 (use of a reaction inhibition
layer having a thickness corresponding to only 10% of the
32

WO 2005/021149 PCT/KR2004/002193
thickness of a pactition)
The procedure of Example 6 was repeated except that a
reaction inhibition layer hairing a thickness corresponding to
only 10% of the thickness of a partition was inserted into a
portion of the first-step catalyst layer, where the partition
had been placed.
Comparative Example 7 (no use_ of a reaction Inhibition
layer)
The procedure of Example 6 was repeated except that a
reaction inhibition layer was not inserted into a portion of
the first-step catalyst layer, where the partition had been
placed.
Table 1


WO 2005/021149 PCT/KR2004/002193

partition.
As apparent from the results for Comparative Examples 1
to 3, it can be found that if the first-step reaction zone is
set to isothermal conditions in order to increase yield, the
temperature of the molten salt must be increased, resulting
in a significant increase in the temperature of the hot spot
in the catalyst layer.
34

WO 2005/021149 PCT/KR2004/002193
In Example 3, the yield of acrylic acid was higher than
those of Comparative Examples 2 and 3, and also the hot spot
temperature was much lower due to the application of multi-
step heat control, thus making stable operations possible. As
evident from the comparison between Example 2, and Comparative
Example 2 having the same, average temparature of molten salts
in the first-step reaction (305 °C) and from the comparison
between Example 4 and Comparative Example 3 (310 0C) , it can
be found that if the first-step reaction is conducted in the
two-divided heat control zones as in examples, the final
yield of acrylic acid will be higher than that of the case
where the finest-step reaction is performed at isothermal
condition. Also, the temperature peaks of the hot spots in
examples will be much lower than those of the isothermal
condition case, thus making stable operations possible.
In Examples 1 to 4, as the temperature of the second
shell "space increased from 305 oC to 320 oC, the conversion of
acrolein produced in the first-step reaction zone to acrylic
acid was increased so that the yield of acrolein was
relatively reduced. Thus, the load of the second-step
reaction zone for converting acrolgin to acrylic acid was
slightly reduced, thus increasing the yield of acrylic acid.
In examples using multi-step heat control, a load, for
conversion, of acrolein in the second-step reaction zone was
lower than that of all comparative examples. Among them,
Example 3 showed the lowest load for conversion of acrolein
in the second-step reaction zone, indicating the highest
final yield of acrylic acid. From the results for Examples 1-
4, the sum of "the yield of the Intermediate products
(acrolein + acrylic acid} produced in the first-step reaction
35

WO 2005/021149 PCT/KR2004/002193
zone was the highest in Example 3, indicating that Example 3
had the highest selectivity to the final product (acrylic
acid} .
In Example 5, the temperatures of molten salts in the
shell spaces of the first-step reaction zone were set to 3O8
°C and 315 oC in an axial direction, such that the temperature
of the front portion of the first-step reaction zone was
higher than that of Examples 1-4, This resulted in 3 higher
yield than those of Examples 1-4, The temperature of the hot
spot was increased to 392.5 °C due to the increases in the
molten salt temperature. However it could be found that the
sums of the yields of acrolein and acrylic acid in the first-
step reaction zone was 91.7% for Example 5 and 90.47% for
Example 3, indicating that the conversion and selectivity to
the intermediate and final products were higher in Example 5.
As a result, the yield of acrylic acid, in the second-step
reaction zone was 86.93% which is the highest of all the
experimental results.
In addition, by Introducing the inactive rasterial layer
into a position where heat transfer is problematic due to the
installation of the partition, an abnormal increase in
temparature in the catalyst layer did not appear,
Meanwhile, as evident from the results given in Table
2, it can be found that even when the first-step reaction
zone is filled with catalysts with the same activity, the
effect of multi-step heat control is realised, In Example 6,
multi-step heat control was applied, and in Comparative
Examples 4 and 5 the first-step reaction zone was operated
under isothermal condition. As apparent from Table 2, in
Example 6, AT (Th-Tsaltl) was controlled in the specified range
36

WO 2005/021149 PCT/KR2004/002193
by multi-step heat control while yield and selectivity were
better than those of Comparative Examples 4 and 5. In
Comparative Examples 4 and Sr the temperature in the hot spot
was a very high temperature close to the sintering
temperature of the catalyst, resulting in the deterioration
of catalyst and also In side reactions due to the high
temperature abound the hot spot, thus leading to a reduction
on yield upon long-term operation- Particularly, in
Comparative Examples 4 and 5 in which AT {Th-Tsalt1) was closed
to or exceeded 110 oC which is a preferred value specified in
claims 1,3 and 15 o± the present invention, the yield of
acrylic acid, a final product, was at least 1% lower than
that of Example 6, due to an increase in byproducts, such ag
carbon dioxide, acetic acid, etc.
Table 3 shows the results of an experiment to solve a
problem in that heat transfer does not sufficiently occur due
to the insertion of the partition- In Example 6, the reaction
inhibition layer was inserted into a position where the
partition has been placed, and in this state, an experiment
was performed- Conparative Example 6 was the same as in
Example 6 except that the thickness of the reaction
inhibition layer was 10% of the thickness of the partition.
Comparative Example 7 was the same as in Example 6 except
that the reaction inhibition layer was not inserted into the
location of the partition. As evident from Table 3, sines
temperature control in the partition location where heat
transfer is problematic was not easy, Comparative Example 6
showed an increase in temperature of 61.4 °C as compared to
the temperature of molten salts, and Comparative Example 7
showed an increase in temperature of 100.7 °C. Such increases
37

WO 2005/021149 PCT/KR2004/002193
In temperature are significantly different from that of
Example 6. Also, it could be found that Tintf (the temperature
of catalyst layer at the partition location) was close to the
peak temparature of the catalyst layer. It is believed that
this Is mainly attributed, to heat accumulation by an
exothermic reaction which occurs since heat transfer around
the partition is not easy. If the reaction inhibition layer
is not used, heat accumulation will not be the only problem.
As apparent from Comparative Examples 6 and 7 where the
reaction inhibition layer was not suficiently ensured, the
yield of acrylic acid, a final product, was also reduced as
compared to Example 6. This indicates that reaction heat at
ths partition location did not easily flow out, so as to
cause an abnormal increase in temperature and finally a
reduction in selectivity, thus leading to an increase in the
amount of byproducts.
38

WO 2005/021149 PCT/KR2004/002193
Claims
1. In a process of producing unsaturated aldehydes and
unsaturated acids from olefins by fixed-bed catalytic partial
oxidation in a shell-and-tube heat exchanger-type reactor,
the reactor comprising one or more catalytic tubes each
including a first-step reaction zone of mainly producing the
unsaturated aldehydes, a second-step reaction zone of mainly
producing the unsaturated acids, or both the two zones, the
improvement wherein:
the first-step reaction zone is divided into two ox
more zones by a partition,
each of the divided shell spaces being filled with a
heat transfer medium,- the heat transfer medium being
maintained at isothermal temperature or a temperature
difference of 0-5 oC, in which the temperatures of the
heat transfer medium in each of the divided shell spaces are
get to increase in the moving direction of reactants and
when the shell spaces divided by the partition in the
first-step reaction zone are named, such as zone 1, zone 2,
zone 3, ..-, Tn-Tsalts is < 150 oC and Tn-Tsalts is < 120 °C,
wherein N is an integer of 2 or more, Th1 is the peak
temperature of a reaction mixture in a catalyst layer
corresponding to the first shell space (the peak temperature
of the catalyst layer) , Th1 is the peak temparature of a
reaction mixture in a catalyst layer corresponding to the Nth
shell space (the peak temperature of the catalyst layer),
Tsalts is the temperature of a heat transfer medium filled in
the first shell space, and Tsalts is the temperature of a heat
transfer medium filled in the Nth shell space.
39

WO 2005/021149 PCT/KR2004/002193
2. In a process of producing unsaturated aldehydes and
unsaturated acids from olefins by fixed-bed catalytic partial
oxidation in a Shell-and-tube heat exchangers-type reactor,
the reactor comprising one or more catalytic tubes each
including a first-step reaction zone of mainly producing the
unsaturated aldehydes, a second-step reaction zone of mainly
producing the unsaturated acids, ox both the two zones, the
improvement wherein:
the first-step reaction zone is divided into two or
more zones by a partition,
each of the divided shell spaces being filled with a
heat transfer medium, the heat transfer medium being
maintained at isothermal temperature or a temperature
difference of 0-5 oC,
in which the temperatures of the heat transfer media in
each of the divided shell. spaces are set to increase in the
moving direction of reactants, and
a reaction inhibition layer made of an inactive
material alone or a mixture of the inactive material and a
catalyst is placed in a position within the catalytic tube,
which corresponds to a position where the partition is
disposed.
3. The process of Claim 1 or 2, which is a process of
producing acrolein and acrylic acid from propylene.-
4. The process of Claim 1 or 2, wherein the temperature
difference between the heat transfer media filled in the
adjacent shell spaces is in a range of 0 oC-50 oC
40

WO 2005/021149 PCT/KR2004/002193
5- The process of Claim 1 or 2, where in the partition.
is disposed in such a planner that it covers at least one
temperature peak,
6. The process of Claim 5, wherein the temperature peak
occurs at the inlet of the reactor, the front portion of each
of the reaction zones, or a boundary between the adjacent
catalyst layers with different activities.
7. The process of Claim l, wherein a reaction
inhibition layer made of an inactive material alone or a
mixture of the inactive material, and. a catalyst is placed in
a position within the catalytic tube, which corresponds to a
position where the partition is disposed.
8. The process of Claim 2 or 7, wherein the volume
ratio of the inactive material to the catalyst material in
the reaction inhibition layer is 20-100%.
9. The process of Claims 2 or l, wherein the height of
the reaction inhibition layer is 20-500% of the thickness of
the partition.
10. The process of Claim 1 or 2, wherein each of the
reaction zones except for the inactive material layer or the
reaction inhibition layer is filled with either one catalyst
layer with the same activity or two or more catalyst layers
with different activities, the two or more catalyst layers
are being filled such that their activities increase in an
axial direction.
41

WO 2005/021149 PCT/KR2004/002193
11. The process of Claim 1 or 2, wherein the
temperature of the heat transfer medium filled in each of the
shell spaces can be controlled independently.
12. The process of Claim 1 or 2, wherein the space
velocity of the olefing introduced into the reactor inlet is
in 3 range of 50-130 hr-1.
13. The process of Claim 1 or 7, wherein Th1-Tslal is <
110 oC, and Th1-Tslal is < 100 0C
14- In a shell-and-tube heat exchanger-type reactor
which can be used in 3 process of producing unsaturated
aldehydes and unsatucabed acids from olefins by fixed-bed
catalytic partial oxidation, the reactor comprising one or
more catalytic tubes each including a first-step reaction
zone of mainly producing the unsaturated aldehydes,, a second-
Step reaction zone Of mainly producing the unsaturated acids,
or both the two zones, the improvement wherein: the first-
step reaction zone is divided into two or more zones by a
partition, each of the divided shell spaces being filled with
a heat transfer medium, the heat transfer medium in each of
the shell spaces being maintained at isothermal temperature
or a temperature difference of 0-5 °C, in which the
temperature of the heat. tranfer media in each of the
divided shell spaces are set to increase in the moving-
direction of reactants, and when the shell spaces divided by
the partition in the first-step reaction zone are named, such
as zone 1, zone 2, zone 3, .,., Th1-Tslal is < 150°C and Th1-Tslal
42

WO 2005/021149 PCT/KR2004/002193
is < 120 °c, wherein N Is an integer of 2 or more, Thl is the
peak temperature of a reaction mixture in a catalyst layer
corresponding to the first shell space (the peak temparature
of the catalyst layer), Thi is the peak temparature of a
reaction mixture in a catalyet layer corresponding to the Kth
shall space (the peak, temperature of the catalyst layer) ,
Tsalts is the temperature of a heat transfer medium filled in
the first shell space, and Tsalts is the temperature of a heat
transfer medium filled in the Nth shell space,
15. The shell -and-tube heat exchanger-type reactor of
Claim 14, wherein Th1-Tslal is < 110 °c, and Th1-Tslal is < 100
16. In a shell-and-tube heat exchanger-type reactor
which can be used in a process of producing unsaturated
aldehydes and unsaturated acids from olefins by fixed-bed
catalytic partial oxidation, the reactor comprising one or
more catalytic tubes each including a first-step reaction
zone of mainly producing the unsaturated aldehydes, a second-
step reaction zone of mainly producing the unsaturated acids,
or both the two zones, the improvement wherein the first-
step reaction zone is divided into two or more shell spaces
by a partition, each of the divided shell spaces being filled
with, a heat transfer medium, the heat transfer media being
maintained at isothermal temparature or a temperature
difference of 0-5 °C, in which the temperatures of the heat
transfer media in each of the divided shell spaces are set to
increase in the moving direction of reactants; and a reaction
inhibition layer made of an inactive material alone or a
42

WO 2005/021149 PCT/KR2004/002193
mixture of the inactive material and a catalyst is placed
within a portion of the catalytic tube, which corresponds to
a location where the partition is disposed.

The present invention provides a process
of producing unsaturated aldehydes and unsaturated acids
from olefins by fixed-bed catalytic portion oxidation in
a shell and tube heat exchanger type reactor,In this
process,the reactor comprises a first step reaction .In this
process,the reactor comprices a first step reaction done of
mainly producing the unsaturated aldehydes,a second step
reaction zone of mainly producing the unsaturated acids
or both the two zones.The first step reaction zone is
divided into two or more zones by a partitions .Each
of the divided shall spacer is filled with a best transfer
medium,and the heat transfer medium in each shell space
is maintained at isothermal temparature ora temparature
dufference of 0-5o C.The temparature of the heat transfer
media in each of the divided shell spacer the set so
increase in the moving dire3ction of reaction.In order
to facilitate the removal of heat generation or a location
where the portion in placed ,a reaction inhibition layer
is disposeds in the first step reaction zone.Also,in other
to protect the catalyst layer from a highly exothermic
reaction the process is performed at a limited temparature
difference between the temparature in a hot spot are the
temparature of a molten salt.If the improved heat control
system according to the present invention is used,the heat
stability of the catalyst layer will be secured and the yields
of intermidiate and final producer can be increased