Separation by pervaporation of para and meta xylene in the presence of tetrabromide
by Randi Wright Wytcherley
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Chemical Engineering
Montana State University
© Copyright by Randi Wright Wytcherley (1987)
Abstract:
The separation of para and meta xylene was investigated by pervaporation in the presence of CBr
across a polypropylene membrane. Data was collected for experimental runs varying percent p-xylene
at 10%, 30%, 50%, 70%, and 90% in the para and meta xylene mixture, for temperatures of -20°C,
5°C, 22°C, 50°C, and 60°C, with zero, 10, and 24 mole % CBr4 added to the xylene mixture.
Evaluation of the data determined if the degree of separation was affected by any of the variables. The
results were compared using a calculated separation factor which is somewhat similar to the relative
volatility in distillation. Since p-xylene is the more volatile of the two isomers, and the membrane used
is selective for p-xylene, it was chosen as the basis when calculating the separation factor.
When CBr4 was added to a mixture of para and meta xylene, a solid complex was formed between the
CBr and the p-xylene under certain conditions. An association between the CBr4 and p-xylene was
present in the liquid phase under certain conditions. The complex or association formation was
dependent on the concentration of p-xylene in the xylene mixture, the amount of CBr4 added, and the
temperature. Both the solid complex and the association tied up the p-xylene in the feed and reduced
the amount of p-xylene available to permeate through the membrane. The amount of m-xylene which
was available to permeate through the membrane was unchanged. Therefore, more m-xylene permeated
through the membrane than p-xylene and so the m-xylene was concentrated in the product. The result
was an increased separability of the pervaporation process for m-xylene.
The greatest separation occurred at -20°C with 90% p-xylene in the para and meta xylene mixture, and
24 mole % CBr4 added to the xylene mixture. These conditions yielded a separation factor for p-xylene
of 0.05. The inverse of the p-xylene separation factor is the separation factor for m-xylene, so under
these conditions, the resulting m-xylene separation factor was 20.
In general, the separation of para and meta xylene by pervaporation can be significantly enhanced when
24 mole % CBr4 was added to the feed side of the membrane with high p-xylene content in the feed at
temperatures between 5°C and -20°C. SEPARATION BY PERVAPORATION OF PARA AND META XYLENE
IN THE PRESENCE OF CARBON TETRABROMIDE
by
Randi Wright Wytcherley
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Chemical Engineering
MONTANA STATE UNIVERSITY
Boz eman, Montana
May 1987
MAIN LIB.
V37?
l/9?Y
ii
APPROVAL
of a thesis submitted by
Randi Wright Wytcherley
This thesis has been read by each member of the thesis
committee and has been found to be satisfactory regarding
content, English usage, format, citation, bibliographic
style and consistency, and is ready for submission to the
College of Graduate Studies.
Date
Chairperson, Graduate Committee
Approved for the Major Department
Date
Approved for the College of Graduate Studies
Date
Graduate Dean
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis
requirements
for
a
in partial' fulfillment of the
master's
degree
at
Montana
State
University, I agree that the Library shall make it available
to borrowers under rules
from this thesis
are
of
the Library.
Brief quotations
allowable without special permission,
provided that accurate acknowledgment of source is. made. ,
Permission for extensive
of this thesis may be
granted
his absence, by the Dean
of
quotation from or reproduction
by my major professor, or in
Libraries when, in the opinion
of either, the proposed use of the material is for scholarly
purposes.
for
Any copying or use of the material in this thesis
financial
gain
permission.
Signature.
Date_____
>5H I an
shall
not
be
allowed
without
my
iv
ACKNOWLEDGMENTS
The author would like to
the
Chemical
Engineering
University, for their
and
encouragement
research by
appreciated.
my
thank the faculty and staff of
Department
guidance
given
advisor.
The author
P. Mundy and doctorate
Chemistry Department at
and
also
F.P.
the
course
State
The advice
of
this
McCandless, is greatly
wishes to thank Dr. Bradford
candidate
Montana
Montana
assistance.
throughout
Dr.
at
Mr.
Dave Barnekow of the
State University for their
assistance with the infrared spectroscopy.
V
TABLE OF CONTENTS
Page
APPROVAL.............. ......... . ...... .....
ii
STATEMENT OF PERMISSION TO U S E ..............
xii
ACKNOWLEDGMENTS...........
iv
TABLE OF CONTENTS........
v
ABSTRACT. . .................... ....... ‘........
INTRODUCTION..........................
Separating the Xylenes...... . . .
Pervaporation....................
Previous Related W o r k .... ......
Selective Complex Formation with
Aromatics................................
Separation Factor........................
viii
xiv
I
w
LIST OF FIGURES...............
vii
m
LIST OF TABLES................................
7
11
RESEARCH OBJECTIVES.... .....................
13
EXPERIMENTAL APPARATUS AND PROCEDURE..... . .
14
Pervaporation Test Ce l l......
Constant Temperature Bath and
Circulating System.......
Vacuum System and Cold Tra p s............
Product Analysis..............
Experimental Procedure........
RESULTS AND DISCUSSION............
Discussion of Results......
P-Xylene, m-Xylene and CBr^ System.....
14
17
18
18
18
21
38
40
vi
TABLE OF COMTENTS— Continued
Page
Formation of CBr • p-xylene Association
in Liquid Phase.........................
Carbon Tetrabromide in Product..........
Advantage of Using Pervaporation
Process............... .............* . . . .
Effect of Complexing Agent on Fl u x .....
54
55
CONCLUSIONS...................................
59
SUGGESTIONS FOR FUTURE RESEARCH. .............
61
LITERATURE CITED.........
62
APPENDIX.......
65
Cross Plots for Figures 13 through 17...
52
53
65
vii
LIST OF TABLES
Table
Page
1.
Physical Data for the C g Aromatics..........
2
2.
Various Mole % CBr Added to Xylene Feed
Mixture to Complex with the p-Xylene........
19
Initial Crystallization Temperatures for
Binary Mixtures of p-Xylene and m-Xylene___
39
3.
4.
Melting Points for the CBr • p-xylene complex
at High Concentrations of p-Xylene in the
Xylene Feed Mixture..........................
53
5.
CBr^ Present in the Product of the 30/70
Ratio (para to meta) Xylene Run with
10 mole % CBr^................................
53
Flux for the Runs at O r I O r and 24
Mole % CBr
57
6.
viii
LIST OF FIGURES
Figure
1. Initial crystallization temperature of
binary mixtures of carbon tetrachloride
and a xylene with the eutectic points
circled......................
Page
g
2. Initial crystallization temperature of
binary mixtures of carbon tetrabromide
and a xylene with the eutectic points
circled......................................
10
3. Pervaporation Equimpment Diagram...........
15
4. Pervaporation Test C e l l.....................
16
5. Separation factors vs temperature at
varying percent p-xylene in the para
and meta xylene mixture across a
polypropylene membrane with no
CBr4 added.......................
22
6 . Percent p-xylene in the product as a
function of the % p-xylene in the feed
at varying temperatures with no CBr
added.......... ..................... ; .......
23
7. Separation factor vs temperature for
10% p-xylene and varying % CBr
additions...................... ?
24
ix
LIST OF FIGURES— Continued
Figure
Page
8 . Separation factor vs temperature for
30% p-xylene and varying % CBr
additions...........................
25
9. Separation factor vs temperature for
50% p-xylene and varying % CBr
additions.... ................. t ............
26
10. Separation factor vs temperature for
70% p-xylene and varying % CBr
additions........
27
11. Separation factor vs temperature for
90% p-xylene and varying % CBr
additions...... ............... * ...... ......
28
12. Percent p-xylene in product as a function
of % p-xylene in feed with 10 and 24
mole % CBrif added at 6 0 °C..................
29
13. Percent p-xylene in product as a function
of % p-xylene in feed with 0 and 24
mole % CBrjf added at 50°C.........
30
14. Percent p-xylene in product as a function
of % p-xylene in feed with 0 and 24
mole % CBr added at 22 0C ..................
31
15. Percent p-xylene in product as a function
of % p-xylene in feed with 0 and 24
mole % CBrjf added at 5 0C ...................
32
4
X
LIST OF FIGURES— Continued
Figure
Page
16. Percent p-xylene in product as a function
of % p-xylene in feed with 0 and 24
mole % CBr^ added at -200C .................
33
17. Freezing point diagrams for mixtures of
p-xylene, m-xylene, and CCl , and p-xylene,
m-xylene, and CBr^.........7 ................
42
18. Mole % CBr present in liquid phase as a
function of p-xylene concentration in the
para and meta xylene mixture at 2 2 0C with
an initial concentration of CBr at
24 raole%........................ T ...........
46
19. Freezing point diagram for mixtures
of p-xylene, m-xylene, and CBr^............
48
20. Freezing point diagram for mixtures
of p-xylene, m-xylene, and CBr4 ........... .
51
21. Average flux for pervaporation process
with 0, 10, and 24 mole % CBr4 added......
56
22. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 10%
p-xylene in the xylene mixture with no CBr
23. Separation factors (a) produced as a
function of temperature from the pervapdrat
pervaporation
separation of a feed mixture containing 30%
p-xylene in the xylene mixture with no CBr
added.............. .........................?
67
xi
LIST OF FIGURES— Continued
Figure
Page
24. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 50%
p-xylene in the xylene mixture with no CBr
added..................................... . .? 68
25. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 70%
p-xylene in the xylene mixture with no CBr
added....................................... ?
69
26. Separation factors (ot) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 90%
p-xylene in the xylene mixture with no CBr
added....................................... ?
70
27. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 10%
p-xylene in the xylene mixture with 10 mole %
CBr^ added...................................
71
28. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 30%
p-xylene in the xylene mixture with 10 mole %
CBr added............. .....................
72
4
29. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 50%
p-xylene in the xylene mixture with 10 mole %
CBr^ added........................... .......
73
xii
LIST OF FIGURES— Continued
Figure
Page
30. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 70%
p-xylene in the xylene mixture with 10 mole %
CBr4 added...................................
74
31. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 90%
p-xylene in the xylene mixture with 10 mole %
75
CBr added...................................
32. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 10%
p-xylene in the xylene mixture with 24 mole %
CBr4 added...................................
76
33. Separation factors (ot) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 30%
p-xylene in the xylene mixture with 24 mole %
CBr added.................. ................
77
34. Separation factors (ot) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 50%
p-xylene in the xylene mixture with 24 mole %
7B
CBr added.................. ................
35. Separation factors (ot) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 70%
p-xylene in the xylene mixture with 24 mole %
CBr4 added...................................
79
xiii
LIST OF FIGURES— Continued
Figure .
Page
36. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 90%
p-xylene in the xylene mixture with 24 mole %
CBr added............. ............... .
80
xiv
ABSTRACT
The separation of para and. meta xylene was investigated
by pervaporation in
the
presence
of CBr
across a
polypropylene membrane. Data was collected for experimental
runs varying percent p-xylene at 10%, 30%, 50%, 70%, and 90%
in the para and meta xylene mixture, for temperatures of
-200Cj 5°C, 22°C, 50°C, and 60°C, with zero, 10, and 24
mole % CBr^ added to the xylene mixture. Evaluation of the
data determined if the degree of separation was affected by
any of the variables.
The results were compared using a
calculated separation factor which is somewhat similar to
the relative volatility in distillation.
Since p-xylene is
the more volatile of the two isomers, and the membrane used
is selective for p-xylene, it was
chosen as the basis when
calculating the separation factor.
When CBr^ was added to a mixture of para and meta
xylene, a solid complex was formed between the CBr and the
p-xylene under certain conditions.
An association between
the CBr4 and p-xylene was present in the liquid phase under
certain conditions.
The complex or association formation
was dependent on the concentration of p-xylene in the xylene
mixture, the amount of CBr
added, and the temperature.
Both the solid complex and the association tied up the px y Iene in the feed and reduced the amount of p-xylene
available to permeate through the membrane. The amount of
m-xylene which was available
to permeate through the
membrane was unchanged.
Therefore, more m-xylene permeated
through the membrane than p-xylene and so the m-xylene was
concentrated in the product.
The result was an increased
separability of the pervaporation process for m-xylene.
The greatest separation occurred at -20°C with 90% pxylene in the para and meta xylene mixture, and 24 mole %
CBr4 added to the xylene mixture. These conditions yielded
a separation factor for p-xylene of 0.05.
The inverse of
the p-xylene separation factor
is the separation factor for
m-xylene, so under these conditions, the resulting m-xylene
separation factor was 2 0 .
In general, the separation of para and meta xylene by
pervaporation can be significantly enhanced when 24 mole %
CBr4 was added to the feed side of the membrane with high pxylene content in the feed at temperatures between 5 0C and
-20°C.
I
INTRODUCTION
Xylenes are important
of plastics and dyes.
ingredients
But
in
for the manufacture
order for the xylenes to be
useful they must be high in purity.
Xylenes are produced in
petroleum refining as a by product from catalytic reforming.
They are produced as
meta (m ) , and para
a
(p ),
mixture
of the isomers, ortho (o ) ,
along with ethylbenzene in varying
compositions depending on
the
conditions
in the reformer.
The structure of these isomers are:
ethylbenzene
o-xylene
The boiling points and
along
with
the
combinations
emphasize
presents.
the
of
m-xylene
melting
relative
the
points for the isomers
volatilities
isomers
challenging
p-xylene
are
shown
separation
of
the
in
Table
problem
various
that
I
to
this
2
Table I. Physical Data for the C q Aromatics
-ing points and me.Lting points
ISOMER
B P
ethylbenzene
o-xylene
m-xylene
p-xylene
b.
136.2
144.4
139.1
138.4
Df
C
C
C
C
the C arcDmatics
M P3
-94.4
-25.2
-47.9
13.3
C
C
C
C
Relative volatility for binary combinations (I)
RELATIVE
BINARY MIXTURE
VOLATILITY
ethylbenzene
p-xylene +
m-xylene +
ethylbenzene
ethylbenzene
p-xylene +
+ o-xylene
o-xylene
o-xylene
+ m-xylene
+ p-xylene
m-xylene
1.34
1.28
1.22
1.08
1.06
1.01
Separating the Xylene Isomers
In order for the isomers to be effectively separated by
distillation the relative
1.0.
If the relative
separation
of
the
volatility
volatility
compounds
is
than
1 .0 ,
listed in Table lb is
concentrated
relative volatility is between zero
in the product.
lb
be greater than
equal to 1.0, then no
occur.
volatility is greater
component listed in Table
must
then
If
the
relative
the first component
in the product.
If the
and 1.0 then the second
would be the one concentrated
For economical separation by distillation a
relative volatility of about 1.25 is necessary (2).
The ethylbenzene and ortho xylene are usually separated
from the mixture
number
of
plates
of
isomers
(3).
The
by
distillation with a large
challenge
comes
with
the
3
separation of
volatility
the
of
para
their
and.
meta
binary
mixture
therefore separation by normal
as it would require an
xylenes.
is
The relative
equal
to
1 .0 1 ,
distillation is not feasible
extremely
large number of stages at
very high reflux ratio.
Several
processes
separation of para
based
on
have
and
meta
crystallization
presence.of
HF-BF3
(5).
This
rotary
allows
operation.
the
Some processes are
extraction
in
the
the most successful
process is based on adsorption by a
that
which
for
Parex process, which was developed
zeolite in a process
valve
solvent
Currently,
separation method is the
by U O P , Inc.
developed
xylene.
or
(4).
been
utilizes a complex and expensive
simulation
of
a
moving
bed
In principle, this adsorption system operates as
a chromatographic column.
The
xylene mixture is fed to the
unit as a pulse followed by
a
unit
alternatively
outlet
is
collected
pulse of the d e sorbent.
as
The
m-xylene and
desorbent or p-xylene and desorbent (6 ).
More
recently,
principle, has been
another
developed
type of zeolite with an
in order to
of
the
improve
by
Asahi
based
on
(7).
the same
A different
appropriate desorbent fluid is used
the displacement chromatography effect
separation.
Both
selective adsorption of a
on zeolite.
process,
processes
liquid
are
based
on
the
mixture of xylene isomers
In the above processes the separation of liquid
mixtures of the xylenes
has
been carried out by adsorption
4
on zeolite particles.
same
separation
continues
in
in
this
Recently
Carra
the
gaseous
area
since
et
al. proposed the
phase
the
(8 ).
Research
commercial
methods
currently used are very complicated and energy intensive.
Pervanoration
Pervaporation is a membrane
fractionate
liquid
mixtures.
processes,
membrane
processes
separate it into two
product
contact
In
with
Like
take
side
of
other
feed
separation
stream
and
In one the target
the other the target species
pervaporation
one
a
streams.
species is concentrated and in
is depleted.
separation process used to
a
a
liquid feed is placed in
nonporous
membrane.
The
components of the feed mixture pass through the membrane and
leave the downstream side as a vapor.
separation is the
difference
in chemical potential between
the liquid and the vapor phases.
permeation
may
be
The driving force for
The driving force for the
attributed
to
the
pressure
and
concentration differences across the membrane.
There are a
few
permeating molecules
theories
during
as
surface in
contact
I.
the
migration through the body of
the permeating material
et
Solution
with
at
the behavior of the
pervaporation.
theory was proposed by Binning
three step process:
to
al.
(9) which involves a
of
liquid into the film
liquid
the film;
the
One possible
charge
mixture;
2.
3 . vaporization of
downstream interface where
5
the permeate is immediately
swept
away.
the permeation rate and separation
predicted
by
the
components in
the
permeation
feed
of the mixture cannot be
rates
since
With this theory,
the
of
the
individual
membrane structure may
change due to swelling (1 0 ).
Another proposal as to
molecules is that in
that the
screen.
membrane
the
which
acted
behavior of the permeating
Michaels et al.
as
a
simple
(11) postulated
molecular sieve or
In this case the permeation rate of a mixture could
be predicted from
the
permeation
rates
of the individual
components.
The
pervaporation
process
uses
nonporous membranes.
The permselectivity of these membranes comes from properties
inherent
to
the
membrane
material.
defined as the rates of flux
of the two isomers under equal
partial pressure driving
forces
generally depends on the
diffusivity
the species being separated
(12)..
in
the
The permselectivity
and the solubility of
the membrane material.
permselectivity is best when there
either the diffusivity or
Permselectivity is
The
is a large difference in
solubility of the permeating
species and the rejected species.
Diffusivity is dependent on molecular size and shape as
well as the mechanical properties
of the polymer.
the chemistry of the
also very important.
polymer
chemical interactions affect
is
the
However,
The
solubility of the species
6
in the membrane.
In
principle, the more soluble a species,
the higher the permeability (13).
Pervaporation
can
sometimes
separation of certain mixtures
separate.
Liquid
mixtures
have very close vapor
chemical
membrane,
on
interactions
a
which
which
helpful
in
the
are otherwise hard to
form azeotropes or which
pressures are virtually impossible to
separate by conventional
separation depends
prove
distillation.
the
mechanical
between
successful
Since pervaporation
properties
the
molecules
separation
of
close
and the
and
the
boiling
components is sometimes possible.
Previous Related Work
Selective complex
formation
used which forms a complex
component.
occurs
when
an agent is
between the agent.and a specific
The formation of this complex can increase or at
least change the selectivity of membrane processes.
Work has been done
past where the
complexes (14).
membrane
at
Montana State University in the
itself
was
modified using Werner
The result was an increased selectivity for
the target species.
Recently, work has
been
done
in coupled transport or
facilitated transport which somewhat overcomes the membranes
inability to make clean separations (15).
from the
membrane.
incorporation
This
carrier
of
a
forms
specific
a
Selectivity comes
carrier within the
complex
with the target
I
species on the feed side
diffuses through
the
of
the membrane; the complex then
membrane
and
the
target species is
released on the product side of the membrane.
Another approach is based
nature of the target
species
on
the idea of changing the
in
membrane will be either more or
the
feed solution so the
less permeable to it.
has been done with reverse osmosis (RO) where a
formed resulting in a
The RO membrane is
less
to
the
heavy
permeable
the
to the higher molecular
target
species.
ultrafiltration
heavy metal ions (17).
up
complex was
higher molecular weight species (16).
weight complex than to
has been applied
This
Complexing
metal
ions,
ultrafiltered, the metal
for
removal of toxic
agents were used to tie
the
complex
The same idea
solution
was
then
broken and the complexing
agent recovered for re-use.
Selective Complex Formation with Aromatics
Previous
work
indicated
the
complexes
between
discussed
formation
Cg,
by
of
Cg,
tetrahalogenated methanes.
Egan
solid
Cg ,
An
and
et
al.
(18)
molecular
C iq
temperature
binary mixtures of
single
eutectic
indicates no
of
CCl^
observed
complex
was
in
with
equimolal complex is formed
initial
and
addition
aromatics
between CCl^ and p-xylene which freezes at -3.90C.
shows the
has
the
the
formed.
Figure I
crystallization for the
xylene
case
isomers (19).
of
meta
A
xylene
Other tetrahalogenated
8
-
25.00
P-XYLENE
50.00
LU O
Q^O
-
CZ
-
100.00
-7 5 . 0 0
M-XYLENE
TEMPERATURE “C
25.00
o
o
0 .00
20 . 00
4 0 . 00
60.00
8 0 . OO
100. OO
MOLE % AROMATIC
Figure I. Initial crystallization temperature of binary
mixtures of carbon tetrachloride and a xylene,
with the eutectic points circled.
9
methanes were also
with p-xylene.
found
One of
to
form solid addition compounds
these
was
form an equimolal complex that
shows
the
temperatures
occurred for the
isomers (20).
freezes at 53.30C.
where
binary
Carbon
only p-xylene.
initial
mixtures
of
as
Figure 2
crystallization
CBr^
tetrabromide
with both of the xylene isomers
in forming complexes
CBr4 , which was found to
and the xylene
had complex formation
and so was not as selective
CCl4
which
formed a complex with
Complex formation is indicated for both para
and meta xylene with
CBr4
for each binary mixture.
since
two eutectics are present
The eutectic points are circled in
Figures I and 2.
This research project
involves
the
separation of the
para and meta xylene isomers in the presence of CBr4 .
Under
certain conditions a molecular complex is formed between the
CBr4 and
the
p-xylene.
complex of CBr4 and
the
amount
of
The
p-xylene
p-xylene
presence
in
which
through the polypropylene
xylene present
in
the
is
available
to permeate
This decrease in the
permeate results in less p-
product.
available to permeate through
the molecular
the feed solution reduces
membrane.
amount of p-xylene available to
of
The
amount of m-xylene
the membrane is not affected.
The result is an increase of separability of the process for
m-xylene, with more m-xylene
in the
feed.
CBr4
was
appearing
chosen
as
in the product than
the complexing agent
10
-50.00
0.00
P-XYLENE
-
100.00
M-XYLENE
TEMPERATURE 0C
5 0 . 00
O
O
d
0 . 00
20 . 00
40.00
60.00
8 0 . 00
1 0 0 . 00
MOLE % AROMATIC
Figure 2. Initial crystallization temperature of binary
mixtures of carbon tetrabromide and a xylene
with the eutectic points circled.
11
since it has a
higher
temperature of complex formation and.
it is less toxic than CCl4 .
Due to the limited availability of membrane material, a
Dupont polypropylene membrane
which
was
on hand was used.
\
Since the complexing agent
tied
membrane was selective to
the para to meta
xylene
up
the p-xylene while the
the p-xylene, the separability of
varied greatly, depending on which
effect was greater.
Separation Factor
The
separation
factor
(<x)
is
calculated
using the
following equation:
cc = y( 1-x)
X(l-y)
'
<x = separation factor with respect to p-xylene
y = fraction p-xylene in product
x = fraction p-xylene in feed
This equation gives a separation factor in terms of pxylene.
The separation
relative volatility
factor
is
greater
transfer of p-xylene
the membrane would be
separation factor is
factor
in
is
somewhat similar to the
distillation.
than
1 .0 ,
across
there
the
selective
between
0.0
If the separation
would
be
a greater
membrane than m-xylene or
to
the
and
p-xylene.
If the
1.0 the membrane is
selective for the m-xylene.
In the research, the interest
lay in decreasing the
of
feed,
which
could
amount
diffuse
p-xylene available in the
through
the
membrane,
by
12
completing it
amount
of
with
m-xylene
the
carbon
available
tetrabromide.
to
permeate
Since the
through
the
membrane was unchanged, the m-xylene was concentrated in the
product.
This would appear
as a lower separation factor or
an increased selectivity of the process for m-xylene.
13
RESEARCH OBJECTIVES
This research was conducted to determine if the addition
of carbon tetrabromide
as
a
completing agent would change
the pervaporation separation of
a liquid mixture containing
para
effects
and
meta
xylene.
tetrabromide to the
The
xylene
determine if the p-xylene
The effect
of
mixture
adding
carbon
were examined to
available to permeate through the
membrane decreased, thereby
product.
feed
of
increasing
concentration
the m-xylene in the
of
the CBr
and p4
xylene along with the
c
temperature was investigated for both
the complexing and the separability.
14
EXPERIMENTAL APPARATUS AKfD PROCEDURE
A pervaporation
experiments for
Figure
3
test
separation
illustrates
concentration
cell
of
of
the
liquid
temperature
the
feed
was
was
bath
to
conduct the
and meta xylene.
setup.
placed
The
in
known
(B)
constantly mixed (E).
and
controlled the temperature of
used
para
equipment
pervaporation cell where it
constant
was
circulation
the
system
the
The
(A)
liquid feed in the cell.
The vacuum system (D) pulled the vapor product into the cold
traps (C) where it was crystallized using liquid nitrogen as
the cooling medium.
When a sufficient amount of product had
been collected in Cl„ the vacuum pump was turned off and the
product was weighed
on
a
Mettler
P1200
balance to ±0.01
grams and analyzed using a gas chromatograph (GC).
Pervaporation Test Cell
The pervaporation
Figure 4.
The
test
test
cell
cell
was
is
shown
constructed
in
of
detail in
two 8 cm
diameter stainless steel flat face flanges made by modifying
a large pipe union.
stainless steel n u t .
These
were
held together by a large
Inside the bottom flange was space for
a 3.8 cm diameter perforated disc and filter paper which was
used
to
support
the
membrane.
This
flange
was
15
do— i
A. Controlled Temperature Bath
and Circulating System
B. Pervaporation Cell
C. Cold Traps
D. Vacuum Pump
E. Mixer
Figure 3. Pervaporation Equimpment Diagram.
16
BULK LIQUID
T==IT.!:.. MEMBRANE
FILTER PAPER
IPERFORATED PLASTIC DISC
RINGS
I_
I BULK VAPOR
^
r
f. ™
Figure 4. Pervaporation Test Cell.
17
connected to the vacuum
1/2
teflon tubing.
system with swedgelock fittings and
The top flange had threads to match the
large nut and an open cavity
feed
was
constantly
to hold the liquid feed.
mixed
to
avoid
any
This
concentration
gradients against the membrane.
The Dupont polypropylene,
Clysar® 350P-1A3, 20ymr
was
membrane
flanges and the large nut
leakage.
held between the two
tightened securely to prevent any
Two viton "0" rings
were offset on the flanges to
provide the protection against leakage.
The liquid feed was
where the mixer
placed
simulated
mixer consisted of a'
a
in the upper flange cavity
perfectly
converted
mixed system.
The
variable speed drill with a
variac power control.
Constant Temperature Bath and Circulating System
The constant
temperature
Circulator made by
the
capability
mixture held
Masterline
of
at
bath
a
Forma
circulating
specified
an
was
a
2095
Bath and
Scientific.
This had
ethylene
temperature.
glycol/water
The ethylene
glycol/water mixture was circulated through insulated tubing
into insulated casing which
flanges.
The temperature
surrounded both upper and lower
of
the liquid feed was monitored
by a 2108A digital thermometer (manufactured by Fluke).
feed could be
±1°C.
controlled
to
The temperatures ranged
on the temperature desired.
a
The
temperature usually within
from -230C to 6 1 0C depending
18
Vacuum System and Cold Traps
Two cold
finger
Pyrex
condensers
series to the vacuum line and
glass fittings and spring
the permeate
while
other
diffusion from the vacuum
condensers were placed
nitrogen.
A
two
stage
a
a
mercury
used
to
prevent hack
and mercury manometer.
The
dewer flask containing liquid
duo-seal
capacity provided a vacuum of
monitored by
One condenser collected
was
pump
in
connected in
to the test cell using ground
clamps.
the
were
vacuum
0.5
U-tuhe
pump running at
pm Hg.
monometer
The vacuum was
and
at times a
McCloud guage.
Product Analysis
A quantitative analysis of the permeate was found using
a
Varian
Aerograph
Series
1400
Sargent-Welch recorder, model
with Bentone 34
para and
meta
modified
xylene
gas
SRG.
with
chromatograph
The column was packed
diisodecylphthalate for the
analyses,
but
for' the quantitative
determination of CBr^,
SR-30
packing
was required.
packing is also
as
boiling
point
known
analyses usually required
a
I
and
hour
to
I
packing.
SR-30
The
1/2 hours and the
peaks quantified using a disc integrator.
Experimental Procedure
At the beginning of a
between the flanges.
The
run
a fresh membrane was placed
upper
flange
was greased with
19
vacuum grease and placed on
a sheet of Dupont polypropylene
membrane to serve as
the
cutting pattern. The membrane was
cut to
the
flange.
The
to
wrinkling the membrane
size
around
carefully placed together
avoid
flanges were then
and securely fastened together with the large nut.
Twenty to forty grams
in the
feed
cavity.
of
The
the feed mixture were placed
amount
of
feed
required was
dependent on the number of runs
to be made with a membrane.
A
for
fresh
membrane
composition.
Runs
was
used
were
made
each
varying
different
feed
the following three
parameters:
1. concentration of the complexing agent
2 . ratio of the para to meta xylene
3. temperature
The combinations
studied
are
presented
in
Table 2.
Each set of runs made using a single membrane consisted of a
specific mixture of CBr^, p-xylene, and m-xylene, separated
Table 2. Various Mole % CBr4 Added to Complex with p-Xylene
in the Para and Meta Xylene Feed Mixture.
!TEMPERATURE 0C
RATIO OF D / m
10/90
30/70
50/50
70/30
90/10
-20
0,10,24
0,10,24
0,10,24
0,10,24
0,10,24
5
0,10,24
0,10,24
0,10,24
0,10,24
0,10,24
22
0,10,24
0,10,24
0,10,24
0,10,24
0,10,24
50
0,10,24
0,10,24
0,10,24
0,10,24
0,10,24
60
10,24
10,24
10,24
10,24
10,24
20
at the
various
temperatures
220C r 5 0 0C, and 60°C.
to significantly
the
Therefore
where there was some
approximately -20°C, S 0C r
The warmer temperatures did not seem
improve
CBr^ was present.
of
CBr^
separation
the
obtained when no
B O 0C runs were only made
present.
Under these conditions
the CBr^ was completely soluble in the feed mixtures.
Prior to each run, the cold finger pyrex condenser used
to collect the product was
cleaned, dried and weighed.
dry weight was recorded to
use in determining the amount of
product collected.
After
the
liquid feed was allowed to
system
pressure.
product
with
The run
the
time
collected.
examination.
was assembled, the
reach a steady state temperature
and the vacuum pump was started.
recorded along
The
At this point the time was
feed
temperature, and barometric
varied
This
depending on the amount of
was
determined
by
visual
The maximum run time was during the -200C runs
and was about 48 hours.
At
B O 0C, the run time was usually
only about I hour.
After a sufficient amount
the
cold
finger
disconnected.
then warmed to
Pyrex
The
room
Pyrex
of
product was collected in
condenser,
condenser
temperature
and
the
with
vacuum
line
was
the product was
weighed.
temperature, and product weight were recorded.
The time,
21
RESULTS AMD DISCUSSION
Figure 5
illustrates
temperature obtained,
varying percent
from
p-xylene
with no CBr^ added.
in terms of an x
y
separation
the
pervaporation separation at
across
Figure
vs
the
a
factors
(a) vs
polypropylene membrane
6 presents the same results but
diagram,
or
the % p-xylene in the
product vs the % p-xylene in the feed with no CBr4 added.
The separation
factors
agent was added can be
obtained
when
the complexing
seen plotted against temperature
Figures 7 through 11.
Figures
effect of adding CBr^ in terms
12
in
through 16 present the
of % p-xylene in the product
vs % p-xylene in the feed.
Figure 5 illustrates how the selectivity of the process
changed depending on temperature
para and meta xylene
feed
xylene in
the
feed
mixture,
selective
for
p-xylene
and
mixture.
at
the
all
separation factors were greater
concentrated in the
for p-xylene.
product
For the
the separation process
the warmer temperatures.
the % p-xylene in the
With
10% and 30% p-
separation
temperatures.
than
and
process was
When the
1 .0 , the p-xylene was
the process was selective
50%, 70%, and 90% p-xylene mixtures,
was
selective
With
50%
for p-xylene only at
and 70% p-xylene in the
feed mixture, when the temperature was lowered to -200C, the
I, OO
0.80
0.60
SEPARATION FACTOR
0.00
0.20
0.40
((X)
1.20
1.40
22
TEMPERATURE
(bC)
Figure 5. Separation factors vs temperature at varying
percent p-xylene in the para and meta xylene
mixture across a polypropylene membrane with
no CBr added.
4
23
o
% P-XYLENE IN PRODUC
O
22
0. 40
0. 60
C
% P-XYLENE IN FEED
Figure 6 . Percent p-xylene in product as a function of
% p-xylene in feed at varying temperatures with
no CBr4 added.
24
1.40
o
CO
m-xylene
selective
I. OO
0. 60
0.80
p-xylene
selective
24% CSr4
.40
(CX)
S E P A R A T I O N FACTOR
1.20
0% CBr4
- 40.00
-
20.00
0 . 00
TE M P E R A T U R E
20 . 00
4 0 . 00
60 . 00
(eC)
Figure 7. Separation factor vs temperature for
10% p-xylene and varying % CBr^ additions.
25
o
tJ-
I
p-xylene
Oa I
z
«4- w
m-xylene
selective
10% C B r 4
0.60
0.80
1.00
■^
24% CBr4
.20
0.40
S E P A R A T I O N FACTOR
(CX)
0% CBr4
-40.00
0 . 00
2 0 . 00
40. 00
TEMPERATURE
Figure 8. Separation factor vs temperature for 30% p-xylene
and varying % CBr4 additions.
26
o
0% CBr4
0.80
1.00
p-xylene
selective
m-xylene
selective
10% C B r 4
((X)
0.60
SEP A R A T I O N FACTOR
1.20
CO
.40
24% CBr4
-
20.00
0 . 00
20 . 00
TEMPERATURE
40. 00
60. 00
80. 00
(0C)
Figure 9. Separation factor vs temperature for 50% p-xylene
and varying % CBr^ additions.
27
o
10% C B r 4
0.80
m-xylene
selective
0.60
CBr 4
(CX)
.20
0.40
S E P A R A T I O N FACTOR
1.00
CBr 4
p-xylene
selective
-40.00
-
20.00
0.00
20.00
40. 00
60. 00
TE M P E R A T U R E
Figure 10. Separation factor vs temperature for 70% p-xylene
and varying % CBr^ additions.
28
o
CM
p-xylene
selective
1 0 % CB r 4
0.40
0.60
(CX)
m-xylene
selective
CBr 4
.00
0.20
S E P A R A T I O N FACTOR
0.80
CBr 4
-40.00
-
20.00
0. 00
2JQ. 00
40. 00
60. 00
TEMPER A T U R E
Figure 11. Separation factor vs temperature for 90% p-xylene
and varying % CBr4 additions.
29
O
C
CS r 4
0 . 80
P-XYLENE IN FEED
Figure 12. Percent p-xylene in product as a function of
% p-xylene in feed with 10 and 24 mole % CBr^
added at 60°C.
30
O
O
0% CSr4
CS r 4
P-XYLENE IN FEED
Figure 13. Percent p-xylene in product as a function of
% p-xylene in feed with 0 and 24 mole % CBr
added at 500C.
4
31
% P-XYLENE IN PRODU
O
O
0 % CS r 4
24% CSr4
% P-XYLENE IN FEED
Figure 14. Percent p-xylene in product as a function of
% p-xylene in feed with 0 and 24 mole % CBr
added at 220C.
4
32
O
O
O o'"
0 % C B r4
2 4 % CB r4
'o. oo
o. 40
X
0. 60
C
I. oo
% P-XYLENE IN FEED
Figure 15. Percent p-xylene in product as a function of
% p-xylene in feed with 0 and 24 mole % CBr
added at 5 0C.
4
33
% P-XYLEME IN PRODUC
C
O
0% C B r 4
24% CSr4
0.40
0. SO
C
P-XYLENE IN FEED
Figure 16. Percent p-xylene in product as a function of
% P-xylene in feed with 0 and 24 mole % CBr
added at -20°C.
4
34
selectivity of the
xylene was
separation
concentrated
factors are
between
in
zero
selective for m-xylene.
process
the
and
xylene, the selectivity of
product.
1.0
For
changed
a
when
feed
so that xn-
The separation
the
process was
mixture with 90% p-
the
process was for m-xylene at
Figure 6 presents the same
results as in Figure 5, but
both 5 0C and -20°C.
on an x vs
y
diagram
which
product as a function of the
shows
%
Points
% p-xylene in the
p-xylene in the feed.
diagram conveniently illustrates
of the process.
the
the
This
degree of selectivity
on the diagonal represent the cases
of no separation, where there is the same amount of p-xylene
in the feed as in the product.
As the degree of separation,
or the selectivity of the process, increases, the difference
in the concentration of
p-xylene
the product becomes greater.
in
the feed from that in
This is shown by the points on
the x vs y diagram which are farther away from the diagonal.
Points above the diagonal
because there is a
higher
product than in the feed.
indicate selectivity for p-xylene
concentration of p-xylene in the
Points located below the diagonal
represent process selectivity for m-xylene, or less p-xylene
in the product than in
greatest change in
-200C and where the
the
feed.
selectivity
feed
Figure 6 shows that the
occurs
mixtures
at a temperature of
have a concentration of
35
p-xylene of 50%
and
higher.
Under
these conditions the
process is selective for m-xylene.
The separation factors obtained at various temperatures
with varying percent additions of CBr4 are shown on separate
graphs for
each
mixture.
different
The
experimental
concentration of p-xylene
Under the
percent
conditions
results
was
with
p-xylene
10%
no
are
CBr4
in
when
the feed
the
feed
shown on Figure 7.
and
10
mole
% CBr4
additions, the selectivity of
the
at all temperatures.
When 24
mole
feed mixture and the
temperature was -200C, the selectivity
of the process was for
process was for p-xylene
% CBr4 was added to the
m-xylene.
warmer temperatures resulted in
Experimental runs at the
the process being selective
for p-xylene.
Figure 8 illustrates the results
xylene in the feed
selectivity of
the
mixture.
When
process
addition of 10 mole % CBr4 ,
when there was 30% p-
was
the
no CBr4 was added, the
for
With the
selectivity of the process
was for p-xylene at temperatures of
-200C the selectivity changed
p-xylene.
so
5°C and warmer.
But at
the process was selective
for m-xylene.
When
24
mole
%
CBr4
was
selective for p-xylene at 22°C
at both
5 0C
and
-200C
was
evaluate the reproducibility
which had 30% p-xylene in
the
added r
the
and warmer.
for
of
m-xylene.
the
feed
process
was
The selectivity
In order to
data, the experiments
mixture with 24 mole %
36
CBr^ added
were
duplicated.
The
results
are
shown on
Figure 8 .
Figure 9
presents
the
experimental
xylene feed composition
was
50%
separation process when no CBr^
m-xylene at -200C.
At
was still selective
for
p-xylene.
the warmer temperatures the process
p-xylene
to the feed mixture, the
10 mole % CBr4
xylene.
when
there
was no CBr4
With 10 mole % CBr4 added
process was selective for m-xylene
For the remaining experimental results at
addition,
A 24 mole
temperatures
This time the
was added was selective for
added to the xylene feed mixture.
at 5 0C and -200C.
results when the
of
%
the
CBr4
-200C,
process was selective for p-
addition to the feed mixture at
5 0C,
separation process being
and
selective
60 0C
for
resulted
m-xylene.
in
the
At 22°C
and SO0C the process was selective for p-xylene.
Figure 10 depicts
the feed
mixture
the
experimental results found when
contained
70%
p-xylene.
separation process was selective
the amount of CBr
mole %
CBr4
added.
added,
When 24 mole %
selectivity
of
GBr4
the
process
was
the
for m-xylene regardless of
For the cases with no CBri and 10
where
warmer, the separation
At -200C the
temperatures
was selective for p-xylene.
added
process
were 5 0C and
to the xylene mixture, the
was
for
m-xylene
at
all
temperatures.
Figure 11 illustrates the experimental results obtained
with 90%
p-xylene
in
the
feed
mixture.
The separation
37
process with this feed
mixture
amount of CBr^ added, was
CBr4 had been added, the
xylene at 5°C, but
at
selective
xylene at all
When no
the warmer temperatures, the process
With
the feed mixture, the process
5 0 °C.
xylene at 60°C.
for m-xylene.
process was still selective for m-
was selective for p-xylene.
5 0C, 22°C, and
at -20°C, regardless of the
10
mole % CBr^ added to
was selective for p-xylene at
But
the
selectivity changed to m-
The separation process was selective for mtemperatures
when
24
mole
% CBr^ had been
added to the feed mixture.
Figure 12 shows the
temperature of 600C in
experimental results produced at a
terms
of
results when 10 mole % and 24
feed mixture of
varying
The
for m-xylene when
mole
mixture containing
xylene.
a
x
mole % CBr
vs y diagram.
4
concentrations
xylene are compared.
24
an
were added to the
of p-xylene and m-
separation process was selective
%
CBr^
concentration
In these cases.
The
Figure
was
of
added to the feed
50%
or greater p-
12 shows the % p-xylene in
the product was less than the % p-xylene in the feed.
Figures 13 through 16
when 24 mole % CBr^
when no CBr^ was
when the
was
compare the experimental results
added
present.
separation
to
the feed, with the runs
Figure 13 presents the results
process
was
operated
selectivity was for m-xylene
at
or greater
experimental
p-xylene.
Figure 14 were collected
The
at
a
at
50°C.
The
feed concentrations of 70%
temperature
results shown in
of
22°C.
The
38
selectivity of the process was
p-xylene
in
the
feed
for
mixture. .
experimental results
obtained
when operated
temperature
at
a
process was selective for
Figure
from
15
shows
the
the separation process
of
5 0C.
The separation
m-xylene at all concentrations of
p-xylene in the feed except 10%.
temperature of -20°C,
m-xylene at 70% and 90%
Figure 16 shows that at a
all
experimental runs were selective
for m-xylene regardless of
the concentration of p-xylene in
the feed mixture.
Discussion of Results
Background separability data
for
para and meta xylene
across a polypropylene membrane were produced at the desired
temperatures.
This background
which the effect of
could
be
separation factors
But at the higher
As
These
the
p-xylene with the CBr^
are
the
Figure
greater
is
concentrations
are
the
5
shows,
than
membrane
temperatures the pervaporation
xylene.
served as a basis from
complexing
compared.
surprising since
data
1.0.
of
This
the
is not
selective for p-xylene.
of p-xylene and the lower
process
points
most
below
is selective for mthe 1.0 separation
factor line.
The pervaporation process is
selective for m-xylene at
high concentrations of p-xylene and low temperatures because
of the crystallization that occurs
membrane.
Pure
p-xylene
freezes
on
the feed side of the
at
13.30C,
m-xylene is added, the freezing point is depressed.
but
when
Table 3
39
presents the
initial
crystallization
temperatures for the
binary mixtures of p-xylene and m-xylene (2 1 ).
Table 3. Initial Crystallization Temperatures for Binary
Mixtures of p-Xylene and m-Xylene.
% P-Xvlene
Temp 0C
100
90
70
50
30
13
10
0
13.3
10
- I
-13
-29
-52.8
-52
-47.8
For any concentration of
temperature is
temperature,
lowered
pure
para
and meta xylene, as the
the
initial crystallization
below
p-xylene
crystallizes
minimum temperature of -52.B 0C
the binary eutectic of
the
is
mixture
xylene and m-xylene crystallize
the solid phase at the
reached.
eutectic
out
until
the
At that point
is reached and both p-
out.
The concentration of
is 13% p-xylene and 87% m-
xylene.
Crystallization
affected
the
separation
of
the
pervaporation process at feed concentrations of 50% p-xylene
and higher when the temperature
was
xylene, the p-xylene
does
begin
-290C.
experiments
Since
the
temperature of -230C,
the
were not affected by the
not
30%
below
were
5 0C.
At 30% p-
to crystallize until
run
at
a
minimum
and lower p-xylene mixtures
crystallization of p-xylene.
This
40
can be seen on
Figure
6,
as
temperature the pervaporation
even
process
xylene at 30% and lower p-xylene
and 70%
p-xylene
feed
and
was selective for p-
mixtures.
mixtures ,
crystallize at about -13°C
for runs at the lowest
the
- I 0C
In both the 50%
p-xylene
begins to
respectively.
-200C runs had crystallization, but not
So the
the S 0C run.
For a
90% p-xylene mixture, p-xylene will begin to crystallize out
at
IO0C,
therefore
both
the
5 0C
and
-200C
runs
had
crystallization.
As can be seen, in
every run where crystallization was
present, the selectivity of the
xylene.
Since
the
p-xylene
cannot permeate through the
So although
the
that
is
in
crystalline form, it
membrane, but the m-xylene can.
membrane
conditions are such
membrane process was for m-
is
selective
crystallization
for p-xylene when
is present,
the
membrane process is selective for m-xylene.
P-Xylene, m-Xvlene,
and CBr 4-System
----- --------C---The complexing agent, CB r ^ , was
mole % and 24 mole
mixtures.
It was
%
to
the
various para and meta xylene
expected that an equimolal C B r ^ • p-xylene
complex would form as shown by Egan
mole % CBr^ runs, the
mixture
xylene to utilize all the
at the 10/90 ratio
added in amounts of 10
other run at 10 mole %
(22).
In the 10
in which there was enough p-
CBr^
p-xylene
et al.
as an equimolal complex was
to m-xylene.
Therefore, every
CBr^ had a p-xylene concentration in
41
excess of that required to
the 24 mole % CBr^
runs r
complex
For
an equimolal ratio of p-xylene to
CBr^ existed where there was
the feed mixture.
with all the CBr^.
There
a
30/70 para to meta ratio in
was
excess p-xylene present for
para to meta ratios greater than this.
The results
was
selective
which
for
m-xylene
through 16, and can be
m-xylene, and CBr^
indicate
the pervaporation process
are
presented
interpreted
freezing
in
Figures 7
in terms of a p-xylene,
point diagram, and selective
complex formation between the CBr4 and p-xylene.
The behavior of para and meta xylene in the presence of
CCl4 was
studied
conveniently
by
Egan
represented
by
diagram (shown in Figure
divided into four areas
et
al.
the
17a).
to
(23).
The
ternary
data was
freezing
point
The triangular diagram is
indicate the composition of the
solid phase that crystallizes out first when a solution of a
given composition is cooled.
In the same
manner,
has been constructed
system.
a
for
ternary freezing point diagram
the
This diagram is shown
the triangle represents
a
p-xylene, m-xylene, and CBr4
in Figure 17b.
binary
mixture.
Each side of
Information on
the initial crystallization
temperatures
of the binary mixtures were
taken from plots located in the
article by Egan et al.
An equimolal complex of
and the eutectics
(also shown in Figures I and 2) (24).
p-xylene
and
CBr4 is shown on the
42
P-XYLENE
F.P .13.3 0C
MOLE BASIS
dashed lines are isotherms
solid lines are eutectics
p-xylene
\\F.P
CCl •p-xylene
m-xylene
M-XYLENE
F.P.
F.P. -22.S 0C
P-XYLENE
F.P. 13.3 0C
12.B 0C
p-xylene
m-xylene
M-XYLENE
F.P.
'Br -m-xylene
Figure 17. Freezing point diagrams for mixtures of p-xylene,
m-xylene, and CCl , and p-xylene, m-xylene,
and CBr .
4
43
right side of
the
triangle
and
has
a
freezing point of
53.3°C.
CBri^ also forms a
conditions.
Therefore
complex
the
for p-xylene,
m-xylene,
labled areas.
Each
ternary
and
area
with m-xylene under certain
freezing point diagram
CBr4 ,
is
indicates
divided
into five
the composition of the
solid phase that crystallizes out first when a solution of a
given
composition
is
separating the areas
assumption
in
cooled.
are
Figure
quaternary (point A)
The
the
17b
binary
concerns
between
the
five
solid
eutectics.
the
lines
The main
location
of the
CBr4 , the complexes r and
the xylenes.
The eutectic between para and meta xylene is shown as a
horizontal dashed line following
the -52,S 0C isotherm.
The
quaternary will be located somewhere along this p-xylene and
m-xylene eutectic, at the point of intersection of the other
eutectics.
The
separation
factor
resulting
from
the
experimental run having a feed mixture of 30/70 para to meta
xylene
with
10
mole
temperature of -200C
solid
formation
%
CBr4
was
addition
below
occurring
which
1.0.
(point
B)
at
a
This indicated some
effectively
changed the
separation factor making the process selective for m-xylene.
If
the
solid
forming
CBr4 • p-xylene complex,
about "-34.4°C (note
below point B ) .
the
were
the
temperature
-34.40C
Therefore
p-xylene
the
instead
would
of
the
have to be.
isotherm located slightly
solid
forming must be the
44
CBr^' p-xylene complex
-20°C.
So,
the
since
the
temperature
is
only at
p-xylene — CBr4 • p-xylene eutectic must be
located to the left of point
B.
Determination of the exact
location for the
is
beyond
research.
The
quaternary
phase
diagram
is
the
scope of this
nevertheless helpful in
interpreting the experimental results.
The quaternary formed is located at the intersection of
the lines
representing
p-xylene — CBr4 • p-xylene eutectic,
CBr4 ' p-xylene — CBr4 - m-xylene
m-xylene
eutectic,
(point A).
the
and
the
eutectic,
CBr4 - m-xylene —
m-xylene — p-xylene
eutectic
A ternary is also shown at point C, intersecting
CBr4 - p-xylene - CBr4
eutectic,
CBr4 - CBr4 - m-xylene
eutectic, and CBr4 - m-xylene — CBr4 - p-xylene eutectic.
The isotherms in
lines.
Figure
The isotherms in
in Figure
17a.
This
17
Figure
is
17b are more vertical than
due
to
freezing points of CBr4 and CCl4 .
in Figure 17b go from
triangle since there
right side
of
the
the
are
left
no
difference in the
the bottom side of the
negative
temperatures on the
Although
xylene, and CBr4 ternary
freezing
approximation, it proves
helpful
on
the
The negative temperatures
to
triangle.
effects of crystallization
are represented by dashed
the p-xylene, m-
point diagram is only an
in
the discussion of the
the pervaporation separation
of para and meta xylene in the presence of CBr4 .
The results of the
10/90,
to meta xylene mixtures at
30/70, and 50/50 ratio para
22 0C and above were all similar.
45
The change in separation factors,
a separation factor less
than
of CBr^ added except at 50%
6 0 0C.
At concentrations
feed,
no
solid
if any, did not result in
1.0 regardless of the amount
p-xylene with 24 mole % CBr4 at
of
50%
and
CBr4 • p-xylene
less p-xylene in the
complex
formed
at
room
temperature of 220C.
It was desired to know what composition of p-xylene was
required in the feed mixture
complex at
room
temperature
conducted to find
formed.
to form a solid CBr4 • p-xylene
at
what
of
22 0C.
ratio
of
p-xylene the complex
The analyses were conducted with 24 mole % CBr4 for
varying ratios of para to meta xylene.
%
CBr4
in
liquid
as
a
function
concentration at 22°C and
decrease
GC analyses were
in
CBr4
in
liquid
concentration of CBr4 was 24 mole
xylene mixture complexes
with
CBr4 that is detected by
the
CBr4 detected
in
the
of
the
feed p-xylene
illustrates graphically where the
the
solid complex formation is
Figure 18 shows the
occurs.
%.
the
initial
When the CBr4 in the
p-xylene, the amount of
GC is reduced.
indicated
liquid
The
by
phase.
Therefore the
a decrease in the
The CBr4 - p-xylene
complex first forms when there is 24 mole % CBr4 present and
when the ratio of para
to
meta
xylene is 60/40.
remains in solution at
concentrations of p-xylene less than
50% of the para and meta xylene mixture.
in the liquid decreases as
to the formation of
the
the
The CBr4
The amount of CBr4
% p-xylene is increased due
CBr4 - p-xylene
complex.
There is
20.00
10.00
.00
MOLE
% CBr4
30.00
46
° 0 . 00
20. 00
40. 00
60. 00
80. 00
% P-XYLENE IN XYLENE MIXTURE
ioo. oo
Figure 18. Mole % CBr present in liquid phase as a
function or p-xylene concentration in the
para and meta xylene mixture at 22 0C with
an initial concentration of CBr^ at 24 mole %.
47
also a corresponding decrease
in the p-xylene concentration
in the remaining liquid.
This explains how even though the
membrane is selective f or
p-xylene, when the CBr4 * p-xylene
complex is formed,
for m-xylene.
the
pervaporation
Since some of
complex, it cannot permeate
m-xylene continues - to
greater concentration
results in a very good
the p-xylene is tied up in the
through
permeate
of
process is selective
the membrane.
and
m-xylene
so
in
the
the
But the
result
is a
product.
This
separation of m-xylene. from p-xylene
by the pervaporation process.
For the 10/90 runs, a
enough to cause any
temperature
xylene
of 5 0C was not cold
crystallization.
Since the p-
xylene concentration was low, no complex was formed.
-20°C there
was
a
significant
factor with the addition of 24
the separation factor can
phase diagram in Figure
temperature was low
phase that would
xylene.
be
CBr4 -P-Xylene complex.'
the separation
mole % CBr4 .
This change in
At
the
crystallize
10
in
explained by referring to the
19.
enough,
Therefore at
change
At 24
mole
% CBr4 , if the
composition
out
mole
10
%
mole
of the solid
first
would
be pure in-
CBr4
there would be no
% CBr4 , the solid phase
which would form first is the CBr4 -p-xylene complex.
-200C
the
CBr4 -P-Xylene
effectively
pervaporation
changes
process.
the
But at
complex
forms
separation
The
in
the
factor
pervaporation
So, at
feed and
for
process
the
is
48
100% P-XYLENE
F.P 13.3 0C
p-xylene
MOLE BASIS
dashed lines are isotherms
solid lines are eutectics
CBr •p-xylene
/,10/90 ratio xylene
• /rni v h i iy a Tzri
I OSc P-R
100% M-XYLENE C CBr "m-xylene
F.P. -47.S0C
Figure 19. Freezing point diagram for mixtures
of p-xylene, m-xylene, and CBr ^ .
100% CBr
F.P. 86=
n*
m-xylene
49
now selective for m-xylene
as
the m-xylene is concentrated
in the product.
For the experimental runs
meta xylene feed
mixture,
with
solid
a
complex
feed did occur at -200C for
both
additions.
containing
Only
the
feed
solid complex formation at 5°C.
process became more
amounts of CB r ^ .
the 10 mole
%
complex was formed.
of complex was greater
more
CBr4
was added, more
the amount of p-xylene available
membrane
xylene available to permeate
m-xylene permeation
mole % CBr^ had
24 mole % CBr4 runs, the amount
so
to permeate through the
24
was an excess of p-xylene for
as
In the
the 10 and 24 mole % CBr
for m-xylene with increasing
Since there
case,
formation in the
At -20°C, the pervaporation
selective
CBr4
30/70 ratio para to
rate
was
through
less.
With less p-
the membrane, and the
unchanged,
the
m-xylene ends up
being concentrated in the product.
For the 50/50 ratio
solid
complex
para
formation
mixtures with both 10 and
to meta xylene mixture runs r
occurred
at
5 0C
24
%
CBr .
mole
and
The more CBr
4
present, the more complex
was
leads to the m-xylene being
formed,
the feed.
The p-xylene
and as before, this
from p-xylene.
At -20°C,
p-xylene would crystallize out in
in crystalline form cannot permeate
through the membrane so this
process to m-xylene.
4
concentrated in the product and
a greater separation of m-xylene
with no CBr^ present, the
-20°C for
changes the selectivity of the
50
For both, the 70/30
and
90/10 mixtures, complexing was
apparent in the runs with 24
addition of 10
change the
added.
mole
mole
%
CBr^
to
selectivity
from
the
% CBr^ as expected.
the
The
feed mixture did not
case
where
no CBr4 was
This indicates the temperature was not low enough to
cause any solid complex formation.
factors for 0, 1 0 , and 24
this case, there
both 10 and 24
is
mole
solid
mole%
CBr
At -20oC, the separation
% CBr4 were very close.
complex
In
formation occurring at
additions,
and when no CBr
4
is
4
added, pure p-xylene crystallizes out.
The results for the
70/30 ratio para to meta run can be seen by referring to the
phase diagram in Figure 20.
24 mole % CBr4 ,
with
a
mixture for the other 76
Beginning at point A, which is
70/30
ratio (para to meta) xylene
mole
%, cooling the mixture would
crystallize out CBr4 - p-xylene
crystallize out until about
remaining crystallization
xylene — p-xylene
would
occur
(down
same
would
The complex would
(B),
to
approximately
remaining xylene mixture
This is roughly the
-12°C
eutectic
final temperature of
complex.
at which point the
along
point
-200C
the CBr4 - pC).
Mhen the
was obtained the
contain about 42% p-xylene.
product concentration of p-xylene
which was obtained experimentally.
At 10 mole % CBr4 following
E, pure
p-xylene
would
crystallize
temperature of -200C about 42
in the liquid.
As
for
the cooling line from D to
the
out.
At
the final
% p-xylene would again remain
case with no CBr4 added, using
51
P-XYLENE
F.P. 13.3°C
MOLE BASIS
dashed lines are isotherms
solid lines are eutectics
10% CBr
p-xylene
24% CBr
40% p-xylene
-23.3 °C
38% p-xylene
CBr »p- xylene
m-xylene
M-XYLENE
F.P.
Figure 20. Freezing point diagram for mixtures
of p-xylene, m-xylene, and CBr4 .
52
th.e
phase
diagram,
a
final
approximately 36 % would be
xylene
concentrations
p-xylene
expected.
are
concentration
of
Since these final p-
similar
to
concentrations found experimentally, it
the
product
is evident that the
solid complex formation or pure p-xylene crystallization are
the predominant separating
concentrations
and
low
mechanisms
at the high p-xylene
temperatures.
Both
situations
decrease the amount of p-xylene in the feed which is free to
permeate through the
membrane.
So, although the membrane
itself is selective for p-xylene, under the conditions where
the solid complex formation
present,
xylene.
the
of p-xylene crystallization are
pervaporation
The separation
of
process
is
selective
for m-
the xylenes is greatly enhanced
over the separation obtained by just pervaporation.
Formation of CBr^ * p-xvlene Association in Liquid Phase
For the experimental runs
50% and higher
selective to
change
in
p-xylene,
m-xylene
selectivity
formation between the
the
at
the
cannot
CBr^
and
melting points are as follows:
with a feed concentration of
membrane process became more
higher
be
due
temperatures.
to
solid
This
complex
p-xylene since the complex
53
Table 4. Melting Points for the CBr • p-Xylene Complex at
High Concentrations of p-Xylene in the Xylene Feed
Mixture.
% p-Xvlene in Feed
mole % CBr 4
50
70
70
90
90
Meltincr Point 0C
24
10
24
10
24
<22 °C
<22°C
37 °C
<22°C
45 0C
At temperatures of 50°C and 60°C, the complex is in the
liquid phase for all feed compositions examined.
in selectivity must
CBr^ and p-xylene
be
due
which
The p-xylene in the
to
an association between the
effectively
association
through the membrane and
The change
ties up the p-xylene.
is unavailable to permeate
therefore the process is selective
for m-xylene.
Carbon Tetrabromide in the Product
The products from the
10
mole
% CBr^ and 30/70 ratio
para to meta xylene mixture
were analyzed for CBr^ using an
infrared spectrometer (IR).
The
in all products
tested
with
presence of CBr^ was found
IR.
The products were then
analyzed with a GC and the results are shown in Table 4.
Table 5. CBr
Present
in the product of the 30/70 Ratio
(para to meta) Xylene Run with 10 mole % CBr^.
[Temperature I
pole % CBr^ |
-20°C
0.00
|
7°C
3.02
|
22°C
3.11
|
50°C
3.02
|
59°C
2.93
|
54
As can be seen the CBr^ did
product of the
-200C
run
indicated by the IR, it
not
show up on the GC when the
was
was
analyzed.
a
Although it was
very weak peak indicating a
very low concentration.
Advantacre of Using Pervaporation Process
The separations occurring
solid complex formation
to be
as
good
pervaporation.
as
If
in
the
feed resulting from
and p-xylene crystallization appear
these
separations
anything,
without pervaporation across
they
the
would
combined with
be
membrane
formation and crystallization both
the liquid phase of the
when
a bit better
since the solid
decrease the p-xylene in
feed, and the pervaporation process
would take what p-xylene is available in the liquid phase of
the feed and concentrate it in the product.
the m-xylene is being concentrated
on
the feed side of the
membrane, but diluted on the product side.
greater percentage of m-xylene
in
the
In other wor d s ,
There would be a
liquid phase of the
feed than in the product.
One advantage of
separation by means of
xylene
crystallization
present in the product.
using
the pervaporation process over
just
solid complex formation and p-
exists
in
the
reduction
of CBr^
Without pervaporation, the amount
of CBr^ present in the liquid
was still about 8 mole % with
an excess of p-xylene and no m-xylene (see Figure 18).
m-xylene was introduced into the
When
system, the amount of CBr^
55
remaining in the liquid increased.
The CBr4 present in the
product of pervaporation separation regardless of the amount
of para or meta xylene present was at most 3.11 mole %.
end uses of the separated
as to the effect of CBr4
The
xylenes would have to be examined
as
an impurity.
There would also
be some cost involved from the loss of the CBr4 .
Effect of Complexing- Aaent on Flux
The average flux, as a function of temperature for each
mole % CBr4 is
shown
Figure 21 that the
in
flux
Figure
21.
decreases
amount of the CBr4 -p-xylene
It is evident from
with an increase in the
complex.
This is probably due
to the fact the membrane is selective for p-xylene, and thus
p-xylene permeates faster than
is complexed with
the
CBr4 ,
m-xylene.
it
xylene present, which results in
reduces
When the p-xylene
the amount of p-
an overall decrease in the
total flux because of the lower permeation rate of m-xylene.
Therefore the more
lower the flux.
10, and 24 mole %
CBr4
The
added,
flux
CBr4
the
more
complex and the
for
the individual runs with O ,
added,
along with the average flux
for the runs at each temperature are located in Table 6 .
56
o
o
(kg/m2hr)
0% CSr4
1 0 % CS r 4
FLUX
24% CSr4
-
20 . 00
0.00
20 . 00
TEMPERATURE
40. 00
60. 00
(bC)
Figure 2 1 . Average flux for pervaporation process
with 0, 10, and 24 mole % CBr added.
4
80. 00
57
Table 6. Flux For the Runs at 0, 10, and 24 Mole % CBr^
a.
Flux in kg/m nr for varying temperatures and
concentrations with 0% CBr .
4
CONC.
ratio p/m 10/90
30/70
50/50
70/30
90/10
TEMP °C
-20
5
22
50
0.023
0.111
0.121
1.623
0.034
0.048
0.310
1.921
0.021
0.158
0.097
1.302
0.014
0.146
0.169
1.536
0.022
0.078
0.281
2.280
Flux in kg/m2hr for varying temperatures and
concentrations with 10 mole % CBr .
OTVTr*
4
CONC.
ratio p/m 10/90
30/70
50/50
70/30
90/10
TEMP 0C
AVE
0.023
0.110
0.200
1.730
b.
-20
5
22
50
60
0.010
0.055
0.076
1.106
1.911
0.014
0.080
0.215
1.129
2.019
0.007
0.064
0.103
1.358
1.922
0.032
0.088
0.110
1.256
2.260
0.018
0.100
0.131
1.106
1.896
Flux in kg/m2hr for varying temperatures and
concentrations with 24 mole % CBr .
CONC.
ratio p/m 10/90
50/50
70/30
90/10
30/70
TEMP 0C
AVE
0.020
0.077
0.130
1.190
2.000
c.
-20
5
22
50
60
0.002
0.040
0.041
0.650
1.313
The flux
did
0.008
0.025
0.025
1.060
1.527
0.015
0.018
0.080
0.743
1.580
not
appear
during separations where
flux was below 0.2
to
0.040
0.072
0.108
1.627
2.020
be
50°C.
for
most
At
50°C
0.014
0.030
0.080
1.110
1.760
appreciably reduced
crystallization
kg/m2hr
temperature was below
0.003
0.019
0.136
1.493
2.384
AVE
was present.
The
of the runs when the
the flux increased
58
significantly, ranging
from
1.3
JcgZm2Iar
to
2.3 JcgZm2Iir
This is about a 10 fold increase over the 2 0 0C flux.
59
CONCLUSIONS
1.
CBr^ forms a solid molecular complex with, the
p-xylene and therefore decreases the amount of
P-xylene in the pervaporation product at high
concentrations of p-xylene and low temperatures.
2.
In general, the addition of 24 mole % CBr^ changed
the separability of the pervaporation process so it
was selective for m-xylene at low temperatures for
all concentrations of p-xylene, but the 10 mole
% CBr^ addition only affected the separability
at low temperatures for higher concentrations of
p-xylene.
3.
The degree of separation varies with temperature,
the greatest separation in the presence of CBr^
occurring at -20°C where solid complex formation
was present.
4.
At 220C, 24 mole % CBr^ complexes with p-xylene
when the xylene mixture has at least 60 wt %
p-xylene (and 40 wt % m-xylene).
60
CONCLUSIONS— Continued
5.
At temperatures of 500C and S O 0C r where the solid
complex has melted and there is above 50% p-xylene
in the xylene feed mixture, an apparent association
is present between the CBr4 and the p-xylene which
ties up the p-xylene in the feed and changes the
process selectivity to m-xylene.
61
SUGGESTIONS FOR FUTTTRFl PRREARCH
I*
Different coitiplexing agents should, be experimentally
evaluated, to find one which complexes with m-xylene.
This would be advantageous since the complexing
agent would then be working with the membrane's
selective nature instead of against it, as with the
CBr .
2.
Determine the behavior of the molecules in terms
of a phase diagram so the conditions where the
CB r ^ -p-xylene complex forms can be more accurately
predicted for specific compositions and
temperatures.
3.
Continue the current experimental research at
temperatures above 6 0 0C to see if the trend
towards greater m-xylene selectivity continues.
Examine the association formed between the CBr
4
and the p-xylene at these higher temperatures.
62
LITERATURE CITED
1.
McCandless, F. P . , and D o w n s , W . B . , "Separation
of C Aromatic Isomers by Pervaporation Through
Commercial Polymer Films," Journal of Membrane
Science, V o l . 30, pp. 111-116 (1987).
2.
Y e h , A., "A Study of the Reversing of Relative
Volatilities by Extractive Distillation," Thesis,
Montana State University, Bozeman, MT (1980).
3.
Downs, W.B., "Temperature Effects on the
Separation of Isomeric Xylenes Using the .
Pervaporation Process," Thesis, Montana State
University, Bozeman, MT (1985).
4.
Morbidelli, M., Santacesaria, E., Giuseppe, S., and
Carra, S., "Separation of Xylenes on Y Zeolites in
the Vapor Phase. 2. Breakthrough and Pulse Curves
and Their Interpretation," Ind. Eng. C h e m . Process
D e s . D e v . , V o l . 24, pg 83 (1985).
5.
Morbidelli, M., Giuseppe, S., and Carra, S.,
"Comparison of Adsorption Separation Processes in
the Liquid and Vapor Pha s e . Application, to the
Xylene Isomer Mixture," Ind. Eng. Chem. Fundam.,
V o l . 25, pp. 89-95 (1986).
6.
Morbidelli, M., Santacesaria, E., Giuseppe, S., and
Carra, S., "Separation of Xylenes on Y Zeolites in
the Vapor Phase. 2. Breakthrough and Pulse Curves
and Their Interpretation," pp. 83-88.
7.
Morbidelli, M., Giuseppe, S., and Carra, S.,
"Comparison of Adsorption Separation Processes in
the Liquid and Vapor Phase. Application ot the
Xylene Isomer Mixture," I nd. Eng. C h e m . Fundam.,
V o l . 25,pp. 89-95 (1986) .
8.
Morbidelli, M., Santacesaria, E., Giuseppe, S., and
Dar r a , S., "Separation of Xylenes on Y Seolites in
the Vapor Phase.
2. Breakthrough and Pulse Curves
and Their Interpretation," pp. 83-88.
I
LITERATURE CITED— Continued
Binning, R.C., Lee, R.J., Jennings, J.F., and
Martin, E.C., "Separation of Liquid Mixtures by
Permeation," Industrial Engineering and
Chemistry, Vol 53, No I, pg 45 (1961).
Sikona, J.G., and McCandless, F.P., "Separation
of Isomeric Xylenes by Permeation Through Modified
Plastic Films," Journal of Membrane Science,
V o l . 4, p p . 229-241 (1978).
Michaels, A.S., Baddour, R.F., Bixler, H.J., and
Choo, C. Y . , "Conditioned Polyethylene as
Permselective Membrane. Separation of Isomeric
Xylenes.," Industrial and Engineering Chemistry
Process Design and Development, Vol I, No I,
pg 14 (1962).
McCandless, F.P., and Downs, W . B . , pg 112.
Torrey, S., "Membrane and Ultrafiltration Tech­
nology Developments Since 1981," Chemical Tech­
nology Review No. 226, Noyes Data Corporation,
pp. 425-435 (1984).
Sikona, J.G., and McCandless, F.P., "Separation
of Isomeric Xylenes by Permeation Through Modified
Plastic Films," Journal of Membrane Science,
V o l . 4, pp. 229-241 (1978).
Torrey, S., "Membrane and Ultrafiltration
Technology Developments Since 1981." p 438
Lonsdale, H. K . , MiTstead, C.E., Cross, B.P., and
Graver, F.M., "Study of Rejection of Various
Solutes by Reverse Osmosis Membranes," Technical
Report No. 447, Prepared by Gulf General Atomic,
Inc., for the Office of Saline Water, Available
Through NTIS (PB 203828), Springfield Virginia,
March (1969).
Strathmann, H . , and K o c k , K . , "Selective Removal
of Heavy Metal Ions from Aqueous Solutions by
Diafiltration of Macromolecular Complexes,"
in N.N. Li (Ed), Recent Developments in
Separation Science, V o l . IV, CRC Press, Boca Raton,
Florida, pp. 29-38, (1978).
64
H
CO
LITERATURE CITED— Continued
E g a n r C.J . r and Luthyr R . V . r "Separation of
Xylenesr" Industrial and Engineering Chemistry,
V o l . 47, No. 2, p p . 250-253 (1955)
19.
Ibid,
p.
252.
20.
Ibid,
p.
252.
21.
Ibid,
p.
251.
22.
Ibid,
p.
252.
23.
Ibid,
p.
251.
24.
E g a n , C .J ., and Luthy, R . V . , "Separation of
Xylenes," p 250.
v
65
APPENDIX
CROSS PLOTS FOR FIGURES 13 THROUGH 17
66
ALPHA
o
-
20.00
0.00
20 . 00
40. 00
60. 00
TEMPERATURE
Figure 22. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 10%
p-xylene in the xylene mixture with no CBr
added.
4
I . 16
, .08
I . 10
I . 12
1.14
AL P H A
1.18
1.20
67
" - 4 0 . 00
-20.00
0.00
20.00
TEMPERATURE
40.00
60.00
C
Figure 23. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 30%
p-xylene in the xylene mixture with no CBr
added.
4
68
CO
ALPHA
CM
-
20.00
20.00
40.00
TE M P E R A T U R E
60. 00
80. 00
C
xi
Figure 24. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 50%
p-xylene in the xylene mixture with no CBr^
add e d .
69
ALPH
O
0.00
20.00
TEMPERATURE
40. 00
60. 00
C
Figure 25. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 70%
p-xylene in the xylene mixture with no CBr
a d ded.
4
70
ALPHA
O
<N
-
20.00
0.00
20.00
40. 00
60. 00
TEMPERATURE
Figure 26. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 90%
p-xylene in the xylene mixture with no CBr
added.
4
1.20
71
+
0.80
+
0.00
0.20
0. 40
0.60
ALPHA
1.00
+
-
20.00
0 . 00
20.00
40.00
TEMPERATURE
60.00
80.00
C
Figure 27. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 10%
p-xylene in the xylene mixture with 10 mole %
CBr added.
4
72
O
ALPHA
CM
C
CO
Ol
o~
CO
CO
C
C
CO
G~
CM
- 4 0 . OO
-
20.00
0.00
20.00
TEMPERATURE
40. 00
60. 00
C
Figure 28. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 30%
p-xylene in the xylene mixture with 10 mole %
CBr 4 added.
73
o
AL PH A
CM
-
20.00
0.00
20.00
40.00
TEMPERATURE C
60. 00
80. 00
Figure 29. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 50%
p-xylene in the xylene mixture with 10 mole %
CBr added.
4
74
ALPHA
C
TEMPERATURE
Figure 30. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 70%
p-xylene in the xylene mixture with 10 mole %
CBr added.
4
75
O
0.80
0.20
0.40
0.60
ALPHA
1.00
CM
O
O
° - 2 0 . 00
0.00
20.00
40.00
60.00
80.00
T E M P E RATURE C
Figure 31. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 90%
p-xylene in the xylene mixture with 10 mole %
CBr added.
4
76
o
ALPHA
n
- 40.00
0.00
20.00
J
60 . 00
TEMPERATURE C
Figure 32. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 10%
p-xylene in the xylene mixture with 24 mole %
CBri
added.
4
77
ALPH
C
TEMPERATURE C
Figure 33. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 30%
p-xylene in the xylene mixture with 24 mole %
CBr added.
4
78
O
ALPHA
CM
-
20.00
20.00
40.00
TEMPERATURE C
Figure 34. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 50%
p-xylene in the xylene mixture with 24 mole %
CBr 4 added.
0.64
.32
0.40
0.48
0.56
ALPHA
0.72
0.80
0.88
79
_________ I
° - 4 0 . 00
-20.00
__________ __________ __________
0.00
20.00
TEMPERATURE
40.00
60.00
C
Figure 35. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 70%
p-xylene in the xylene mixture with 24 mole %
CBr added.
80
S
ALPHA
d
-20.CO
0.00
20.00
40. OO
temperature C
60. oo
Figure 36. Separation factors (a) produced as a
function of temperature from the pervaporation
separation of a feed mixture containing 90%
p-xylene in the xylene mixture with 24 mole %
CBr added.
4
ITnDlHIES
1762 luu I