Catalytic Production of Cyclohexane-

oxime via Cyclohexanone Ammoximation

Final Spring 2018
Date submitted: May 11, 2018
By: Katelynn Ryan and Kevin Thomas
Theory and Background
Ammoximation is the direct synthesis of oximes from ammonia, oxygen and ketones or
aldehydes in the presence of a catalyst. An oxime is a chemical compound belonging to the
imines with the general formula R RI C=NOH , where the carbon is double bonded to the
nitrogen. Ris an organic hydrocarbon chain while R I may be a hydrogen, forming an aldoxime,
or another it could be another organic hydrocarbon group, forming a ketoxime. This project was
based on producing 99.999% of cyclohexanone-oxime from the ketone, cyclohexanone. The
reaction was between cyclohexanone, ammonia and hydrogen peroxide in the liquid phase using
titanium silicalite as the catalyst. Due to the presence of polar compounds such as water, the
Non-random-two-liquid (NRTL) model was used to calculate the activity coefficient of the liquid
phase. It is recommended for non-ideal chemical systems, and can be used for VLE and LLE
applications for separation of a liquid mixture. Reactions are the heart of the chemical processes
in which raw materials are converted into favorable products. The important factors from
chemical reactions are determining what are the reactants and products needed, to what extent
the reaction will proceed, and how fast it will proceed. Reactions can be classified as reversible
and irreversible reactions.In reversible reactions, the reactants are converted to products at a
certain rate while the products are converted to reactants at a different rate. At equilibrium, the
two rates become equal. In irreversible reactions, the rate of conversion of products to reactants
is zero. Another important classification is based on the phase or phases involved. A
homogeneous reaction is defined as a reaction which occurs in one phase.A heterogeneous
reaction requires the presence of two or more phases for the reaction to take place.A CSTR is a
continuous stirred-tank reactor with an agitator and flow coming in and out of the reactor. It is
assumed perfect or ideal mixing, constant density for liquids, isothermal conditions, and steady
state. Therefore, the mass balance equation is:
accumualtion=¿−out+ generation
dN i
=Fio −F i +V r i
dt
Where:
F io=the molar flow rate inlet of species i, F i=the molar flow rate outlet , r i =reactionrate , N i=moles

The reaction rate is generally dependent on the reactant concentration, where the molar flow rate
of species A is equal to the concentration of species A multiplied by the constant volumetric flow
rate, and the rate constant, k. The rate constant can be determined by using the reaction rate that
is adjusted for temperature using the Arrhenius equation. Generally, as temperature increases so
does the rate at which the reaction occurs. The rate expression for catalytic reactions can be
modeled using the Langmuir Isotherm which consists of three important steps: the adsorption of
the reactants to the catalyst surface; the surface reaction for reactants to form products; the
desorption of the products from the catalyst surface to the surrounding environment. From this
information the following expression can be developed:
 (driving force expression)
rate=(kinetic factor )
( adsorption term )
The kinetic factor represents the surface reaction, and is represented by:

Kinetic factor=k T n e−E / RT

When T o is not specified
Where:T =Temperature, k =rate constant,n=0,E=activation energy, R=gas constant
The driving force expression represents the desorption of product from the catalyst surface, and
the adsorption term represents the adsorption of reactant to the catalyst surface. Both the driving
force expression and adsorption term are defined by equilibrium constants. These equilibrium
constants are of the form:
B
ln ( K ) =A + +C ln ( T )+ DT
T
Detailed Modeling Description
Figure [1]: Flowsheet

In the catalytic production of cyclohexanone-oxime (CXOXM) via cyclohexanone

ammoximation, we set up a unit process which included a CSTR, two distillation towers, and a
splitter. A continuous stirred tank reactor (CSTR) is a reactor that is run at steady state with a
continuous flow of reactant and products. The feed stream for the CSTR is at a uniform
composition throughout the reactor and the exit stream has the same composition as it was inside
the CSTR. The advantages of using a CSTR is that it is good for running at a set temperature, it
can easily adapt to two phase runs, and it also has a low cost to operate. To start the process,
there were two streams that were fed to the CSTR. The first stream consisted of pure hydrogen
peroxide (H2O2), and the second stream consisted of cyclohexanone (HEX) and ammonia
(NH3). The H2O2 stream was pure and was fed into the CSTR with a flowrate of .0311 kmol/hr.,
a pressure of 1 atm, and a temperature of 353 K. The second stream, (HEX-NH3) consisted of
30% ammonia, 30% cyclohexanone, and 40% of tert-butyl-alcohol (TERT--01), and it was fed
into the CSTR at a flowrate of .0988142 kmol/hr., a pressure of 1 atm, and a temperature of 353
K. The CSTR was set at a volume of 10 m3 with 10 kg of catalyst, a catalyst density of 3093
kg /m3, a temperature of 353 K, and a pressure of 1 atm. The reaction that is taking place in the
CSTR is shown below.

HEX + N H 3 + H 2 O2 →CXOXM +2 H 2 O

The rate law for this reaction is as follows:

 kmol
−93200
[ N H 3 ][ H 2 O2 ]
−r [
kgcat sec ]
=1.8333E10 e RT
0.05+ [ H 2 O 2 ]

Where: k = 1.8333E10, E = 93200, Driving force expression = [ N H 3 ][ H 2 O2 ], Adsorption term =

0.05+ [ H 2 O2 ], and Adsorption coefficient = 1

The flow coming out of the CSTR (CSTROUT) had a flowrate of 1.03283 kmol/hr., and
consisted of HEX, NH3, H2O2, CXOXM, H2O, and TERT--01. This stream was then fed into a
distillation tower.

A distillation tower is used in order to separate a liquid or vapor mixture of two or more
substances into its desired purity component fractions. This is done by heating the mixture to a
temperature that is in between their respective boiling points. The tower is made up of a series
plates that are stacked on top of each other. When the feed enters the tower, it is separated into
the heavy key product, which goes out from the bottom of the tower, and into the light key
product, which goes out from the top of the tower. The liquid feed flows down over each plate
and the vapor bubbles go up through the liquid via holes that are in the plates. This process is
done by heating the mixture to a temperature that is in between their respective boiling points.
The tower can handle high liquid flow rates and it can also achieve better distribution. The liquid
that exits the tower is at the bubble point and the vapor that exits the tower is at the dew point.

In order to set up the RadFrac distillation tower, we were missing data. In order to obtain
the missing data, we set up a DSTWU distillation towers in place of the RadFrac towers. Both
the towers were set were set at a pressure of 1 atm, the light key was set as H2O, the heavy key
was set as CXOXM, and the first tower had 10 stages and the second had 20 stages. The first
tower produced two streams, the top stream (DIST1VAP), and the bottom stream (DIST1LIQ).
The top stream was fed into a splitter and the bottom stream was fed into the second distillation
tower. The second tower also produced a top stream (DIST2VAP) and the bottom stream
(PRODUCT). The top stream is then fed into the splitter.

The splitter divides the streams that are fed into it into two or more outputs that have the
same compositions. The splitter produced two streams. A purge stream (PURGE) and another
stream (SPLITREC) which is recycled back into the CSTR. By running this simulation, we were
able to get the data that we were missing for the RadFrac distillation towers from the DSTWU
towers. We took the data for the reflux ratio and the distillate to feed ratio to plug it into the
RadFrac towers. This information was just enough so that we can run the simulation using the
RadFrac towers. In order to achieve the desired molar flowrate and mole % of CXOXM in the
PRODUCT stream, we increased the reflux ratio and also set a design specification. For the
second RadFrac tower we increased the reflux ratio to 7 so that the DIST2VAP produced would
have better conditions in order to go into the splitter. The design specification was set so that we
can achieve the desired molar flowrate of .02959 kmol/hr. We set the spec. to run so that it
would change the flowrate of HEX-NH3 that was going into the CSTR to get the right flowrate
of CXOXM coming out of the PRODUCT stream. To get the 99.999% molar fraction of
CXOXM in the product stream, we increased the number of stages in the first RadFrac to 8
stages and the second RadFrac to 15 stages. By running this final simulation, we the desired
molar flowrate of CXOXM of .02959 kmol/hr. and molar fraction of .99999 in the PRODUCT
stream.

Results and Discussion

For our process, we were able to achieve 0.02959 Kmol/hr of cyclohexanone-oxime with a molar
fraction of 99.999%. The stream results are shown below in Table [1]:
Table [1]: Stream Results

The heat duties from the condenser and reboiler the first distillation tower (RadFrac1) was
determined to be -51318.4 kJ/hr and 52311.9 kJ/hr, respectively. The heat duties for the second
distillation tower (RadFrac2) are -1251.26kJ for the condenser and 1338.03kJ/hr for the reboiler.
Table [2]: RadFrac1 Heat Duties Table [3]: RadFrac2 Heat Duties

In order for a good separation to achieve the desired product, the yield should be 97% or higher
while the purge should be 1.07% or lower. In order to calculate yield and purge, the following
equations were applied:
F CXOXM − product
Yield ( mole basis )= × 100
F HEX− feed

Where:
F CXOXM − product =the molar flow rate of cyclohexanone−oxime∈the product stream , F HEX −feed =the molar flow rate o

FCXOXM −purge
Purge ( mole basis )= ×100
F HEX −feed

Where:
F CXOXM − purge=the molar flow rate of cyclohexanone−oxime∈the purge stream , F HEX −feed =the molar flow rate of cy

For our process, we were able to produce a yield of 99.82% and a purge of 0.122%, which is a
very good separation of the desired product.
0.02959 kmol/hr
Yield ( mole basis )= × 100=99.82%
0.0296442 kmol/hr

(3.61904 × 10−5 )
Purge ( mole basis )= ×100=0.122 %
0.0296442
Conclusions
In conclusion, we were able to successfully to run the system and achieve 99.999% conversion
of cyclohexanone-oxime by the process of ammoximation. We were also able to produce
0.02959 kmol/hr of cyclohexanone-oxime, and a yield of 99.82% and a purge of 0.122%.
Overall, the simulation ran the ammoximation reaction smoothly.

Nomenclature
F: Feed Stream
HEX: Cyclohexanone
CXOXM: Cyclohexanone-oxime
TERT--01: Tert-butyl-alcohol
References
1. Satmon, J.; L., S. B.; Tong, Z.; Zhu, M. Simulation and optimization of cyclohexanone
ammoximation process over TS-1 catalyst: equilibrium and kinetic reactor
https://www.irjet.net/archives/V4/i11/IRJET-V4I11310.pdf (accessed May 4, 2018).
2. Hussain, Y. A. Reactors http://www.just.edu.jo/~yahussain/files/Reactors.pdf (accessed
May 4, 2018).