Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
~Z786.'39
This invention relates to an improved vinyl chlo-
ride stripping process.
Polyvinyl chloride polymer (PVC), whether a homo-
polymer or a copolymer containing predominantly interpoly-
merlzed units of vinyl chloride monomer, is well known to the
art as a versatile plastic. PvC is made using many types of
processes, including emulsion (latex), solution, suspension
and bulk polymerization processes. No matter which process
is employed, however, total conversion of vinyl chloride
monomer to polymer is not obtained. This results in residual
vinyl chloride monomer (VCM) often dissolved in or entrapped
in the PVC polymer.
One method to remove the residual VCM from the
polymer is to heat up the PVC to about 180F. (82C.) to
volatilize the VCM and evaporate off the VCM. The process is
performed under reduced pressure (vacuum) to facilitate the
removal of VCM. As an example of the state of the art on
stripping of VCM, a typical stripping operation would be
conducted at about 170F. and 400 to 450 millimeters of
mercury absolute. Significantly higher temperatures are not
employed for fear of degrading the PVC. A patent recently
issued to Solvay and Company (Belgium Patent 793,505, issued
on June 29, 1973) discloses a technique of stripping VCM from
PVC consisting of condensing steam onto PVC polymer thereby
heating the PVC to above its glass transition temperature,
and then applying vacuum to evaporate off the water and VCM.
The evaporation of water and VCM cools the PVC to below its
glass transition temperature. Again, the stripping of the
VCM is done under vacuum.
~`'i7`~' :
F
~2786.39
In accordance with the invention an improved
process of reducing the residual monomeric matter, in parti-
cular VCM, content of PVC polymers comprises contacting a PVC
polymer containing unacceptably high levels of monomeric
matter, in particular VCM, with a hot gas, for example,
steam, at a temperature of from about 200F. ( 132C~ ) and at
atmospheric pressure or above. The hot gas and monomeric
matter, in particular, VCM, mixture is withdrawn and the
monomeric matter, in particular, VCM, recovered for reuse.
In particular, the process may be carried out to
remove residual vinyl chloride monomer present, after poly-
merization, in vinyl chloride polymers that are in the form
of an aqueous dispersion by contacting the dispersion with
the hot gas, in accordance with the invention, and removing
from the contact area a mixture comprising the hot gas and
vinyl chloride monomer.
In one embodiment the process is applied to an
aqueous dispersion of the polymer which dispersion is intro-
duced into the upper portion of a column containing a
plurality of sieve plates or perforated plates. In parti-
cular the openings of the plates are effective to allow the
aqueous dispersion to descend in the column. The dispersion
is contacted with the hot gas, for example, steam, in the
column for a period of 1 to 60 or more minutes, typically 1
to 15 minutes. In particular the total area of openings in
each sieve plate or perforated plate may be 1 to 10% of the
cross-sectional area of the plate.
Polyvinyl chloride homopolymers or copolymers
(hereinafter referred to as PVC~ can be prepared using
3G emulsion, suspension, solution, or bulk polymerization
lZ786;39
techniques known to the art. Unfortunately, no polymeri-
zation technique or process converts the total amount o~
vinyl chloride monomer (hereinafter referred to as VCM) to
polymer. Much of the unreacted VCM is dissolved in or
entrapped by the PVC polymer. The VCM, if not removed, is
later released upon further processing and/or use of the
polymer. Because of recent pollution and toxicity standards
proposed by the Environmental Protection Agency and set by
the Occupational Safety and Health Act Board, a level of
residual VCM content in PVC in the hundreds of parts per
million (ppm) is unacceptably high. The PVC must be post-
treated to remove the VCM to low levels (down to 10 ppm and,
preferredly, down to less than 1 ppm).
A known process to remove residual VCM from PVC
polymer is to heat the polymer up to about 180F. under
vacuum to release and vaporize off the VCM. It has unex-
pectedly been discovered that residual VCM can be efficiently
and effectively removed from PVC polymer by contacting the
polymer with a hot gas at a temperature of from about 200F.
20 (93C.) to about 270F. (132C.) and at atmospheric pressure
or above. Residual VCM contents of the PVC as low as 0.5 ppm
have been obtained using the novel process. The PVC polymer
is not degraded in the process~
Any PVC polymer, whether homopolymer or copolymer
F
1~86~91
or higher enumerated polymer, can be used in the process.
Of course, the use of a PVC polymer having a low thermal
stability or a melting point or softening point well below
200F. (93~C.) is not favorable. The molecular weight of
the PVC polymer is not critical. Preferredly, the PVC
polymer used is porous, and excellent results have been
obtained using a PVC polymer having uniform porosity.
Polymers of interpolymerized units o~ vinyl
chloride monomer with copolymerizable vinylidene monomers
such as vinyl bromide, vinylidene chloride, ~-olefins such
as ethylene and propylene, acrylic and methacrylic acid,
acrylates and methacrylates such as ethyl acrylate and methyl
methacrylate, vinyl aromatics such as styrene and vinyl
toluene, and the like, and mixtures of these monomers are
known in the art or canbe prepared. Any or all of such PVC
polymers can contain unacceptably high levels of VCM. Hence,
the novel process of this invention can be used to remove
the residual VCM.
The PVC polymer can be prepared using any method
or technique known to the art. Emulsion, suspension,
solution, and bulk polymerization processes can be employed.
The process of the invention applies to PVC in particulate
form containing residual amounts of VCM, and the polymerization
method used to prepare the PVC is not critical. However, if
the PVC polymer is not prepared in a polymerization that
yields particles of PVC on completion, the polymer should be
isolated in a particulate form before using the VCM stripping
process. The actual particle size is not critical as
stripping of the VCM will occur in all cases. However,
stripping of the VCM is faster if the PVC particle size is
under 200 microns. Typically~ the PVC polymer used in the
stripping process has a residual VCM content of above 1000 ppm
1278~;.'3~
by weight of VCM in the polymer, and can contain up to
lO0,000 ppm of VCM and more.
The temperature range employed in the process is
from about 200~1. (93C.~ to abollt c70E. (1~2"C.~, arld more
preferredly from about 212"F. (100C.) to about ~4Q'F.
(118C.). The stripping is performed at atmospheric pressure
or above. A normal pressure range is from 0 psig to about
25 psig. The PVC polymer is contacted with a hot gas which
serves to both heat up the PVC and act as a carrier for the
VCM. The gas is preferredly an inert gas such as nitrogen
or helium, and is not a gas that promotes oxidation of the
polymer such as oxygen. Hot air, as it contains mostly
inert gases, is useful. A preferred gas to use is saturated
water vapor (saturate~d steam). Temperatures and pressures
of saturated steam are well known and can be found in
saturated steam tables (see Chemical Engineers' Handbook,
3rd Ed., McGraw-Hill Book Co., Inc. (1950), pages 277-278).
The use of saturated steam as the hot gas heats up the PVC
polymer, provides a positive pressure in the stripping area,
and acts as a carrier for the VCM.
The PVC polymer is used in the form of a slurry
of PVC polymer particles in a liquid carrier. The slurry
form facilitates pumping and agitation of the particles.
The liquid carrier can be any non-solvent for PVC polymer
having a relatively high boiling point (above 70C.).
Examples of such liquid carriers are ethanol, butanol,
cyclohexane, water, and the like. Water is the preferred
liquid carrier. The total solids content of the PVC slurry
can be from a very low percent by weight of solids to a
total solids content wherein the slurry can just barely be
pumped. As a practical matter, the total solids of the PVC
slurry ranges froln about 5~ by weight to about 80~ by weight
~27~ 9
of PVC polymex in the slurry. The PVC polymer slurry can ~e
contacted wit}- the hot gas in a variety of l~ays. The PVC
slurry and hot gas can be mixed together in a c]osed kettle,
the PVC slurry and hot gas can be mixed and expelled together
into a lower pressure area, or the PVC ~lurry and hot gas
can be contacted with each other in a counter-current flow
operation.
As an embodiment o~ the novel proce~s, the PVC
slurry can be stripped in a kettle. The PVC polymer slurry
is placed in a closed vessel, which can be a polymerization
vessel or a holding tank, and hot gases introduced intv the
vessel. To insure good contact between the PVC and the hot
gases, the hot gas is normally introduced into the bottom
of the vessel. Agitation of the PVC can aid the contact.
Pressure in the vessel is atmospheric or above and normally
ranges from 0 psig to about 20 psig. Temperatures in the
vessel and of the PVC polymer range from about 200F. (93C.)
to about 250~F. (121C.), and more preferredly, from about
200F. (93C.) to about 220F. (104C.). Contact times vary
as to the capacity of the vessel, and range from about 5
minutes to 60 minutes or more. Stability of PVC at elevated
temperatures is a time-temperature phenomenon. Therefore,
shorter contact times should be employed as the temperature
is raised. The hot gas and VCM are withdrawn from the vapor
space in the vessel and the VCM recovered. The PVC is
pumped from the vessel into a hold tank or directly into a
drier. Residual VCM contents of down to 4 ppm and lower can
be obtained. It was unexpectedly discovered that not only
was residual VCM effectively and efficiently removed from
the PVC polymer, butthat the PVC polymer product was not
signlficantly degraded in the process. Prior to this
discovery it was widely believed that any process operation
-- 6 --
lZ786~'39
that would heat PVC to above 180F. (820C.) would have a
severe detrimental effect on the PVC and its subsequent
stability. Under the preferred operating conditions, little
or no PVC degradation was observed.
Another embodiment of the novel process is to
admix the PVC polymer slurry and hot gas at the temperature
and pressure range described and inject the mixture into an
area (such as a kettle) of lower pressure (not a vacuum).
The process can employ a single flash stripping, recycle
flash stripping wherein the hot gas and VCM are withdrawn
from the vapor space in the kettle after the injection into
the kettle and the PVC is then pumped back into an area for
re-mixing with hot gas~ and multi-stage flash stripping
wherein the PVC slurry is mixed with the hot gas and the
mixture injected into a first kettle and then the process
repeated in subsequent kettles. Temperatures of up to and
over 250F. (121C.) were investigated using the flash
stripping operation. The hot gas and PVC slurry mix was
flashed to a kettle at atmospheric pressure. The PVC
slurry was used at about 35~ total solids by weight.
A preferred embodiment of the novel process is to
contact the PVC polymer slurry and the hot gas in a stripping
column. The hot gas used is saturated water vapor (steam).
As opposed to a batch process conducted in a kettle, herein
the PVC polymer, in the form of a slurry, is pumped into a
stripping column at or near the top of the column. The feed
rate can vary depending upon the capacity of the column,
the level of residual VCM in the PVC, the particle size and
porosity of the PVC polymer, and with operating temperatures
and pressures. Feed rates employed in production facilities
could vary from about 100 pounds of resin per hour to 20,000
pounds of resin per hour and higher. The steam is introduced
-- 7 --
127B6.'39
into the column at or near the bottom of the column. ~ence,
the PVC and steam will run counter-current to each other
Pressure and temperature within the column can be controlled
using any known technique including external jacketing,
internal heating coils, and the use of compressed gases.
However, it is both practical and convenient to use saturated
steam in a colurnn that can withstand pressure, to both heat
and pressurize the PVC and the column. For example, saturated
steam at a pressure of 20 psi absolute has a temperature of
2280F. (109C.), and at 25 psi absolute has a temperature
of 2400F. (116C.). The amount of steam introduced into the
column varies according to the feed rate of the PVC slurry
and column design.
The steam and PVC polymer make contact in the
stripping column. It is preferred that the stripping column
be of a tray column design to help with control of flow and
aid in more uniform contact. The height and width of the
column and the number of trays and their spacing and design
are all design variables readily calculable having knowledge
of flow rates and properties of the PVC slurry.
In a preferred method of operating the stripping
column process, the PVC polymer is pumped from a PVC slurry
feed tank into a tray-design stripping column near the top.
Saturated steam at about 235F. (113C.) is introduced into
the column near the bottom. Pressure in the column is about
22 psi absolute. Total contact time in the column varies
as to feed rate and column capacity. With a 30 inch O.D.
column having 17 trays and a feed rate of abo-ut 5000 pounds
of PVC per hour, total residence time in the column is from
about 3 minutes to about 15 minutes. The PVC polymer exits
the colurnn at the bottom of the column and is pumped to a
hold tank or into a drier. The steam with the VCM monomer
1~86~g
exits at the top of the column and goes into a condensate
receiver where the steam is then condensed to water and the
VCM recovered. PVC polymer entering the column has an
average of about 20,000 ppm of residual VCM. The PVC polymer
exiting the column has an average of about 10 ppm of residual
VCM in the polymer. Residual VCM content in the PVC polymer
of lower than 1 ppm can be obtained.
Residual VCM contents in the PVC polymer were
determined by Gas Chromatograph analysis of the particulate
PVC, using a preset calibration. Stability of the PVC resin
both before and after the stripping operation can be determined
by using comparative oven aging tests run against a control
or by using a Capillary Viscometer Heat Stability Test
wherein the resin is admixed with a set level of stabilizer
(if desired), placed into the barrel of the capillary
viscometer, heated to 210C. and slowly extruded. Darkening
of the resin upon extruding indicates degradation of the
PVC polymer. Comparative tests between PVC polymer not
subjected to the novel stripping process and PVC stripped
by the process shows little if any change in the time to
develop color (darkening of the resin).
The following examples further illustrate the
invention.
EXAMPLES
Experiments wherein the PVC polymer slurry and
hot gas (steam) were mixed using a kettle-type apparatus
showed the feasibility of using such a method to strip
residual VCM from PVC polymer at temperatures above 200~F.
and pressures at atmospheric and above, without degrading
the polymer. Residual VC~ contents in the PVC of less than
10 ppm can be obtained. It was believed, though, that more
efficient VCM stripping could be obtained with a counter-
~27~36~9
current flow of PVC slurry and hot gas where maximum
diffusion of VCM would be promoted. Furthermore, less
volume capacity and greater productivity would result from
using a continuous stripping operation as opposed to the
batch operation in a ke-ttle-type apparatus. Hence, experiments
focused upon the use of a stripping colwnll apparatus to conduct
the novel ~-rocess. Tllese experimellts were porlorlned il) two
stages; i.e. at a bench-scale level and at a semi-works level.
At the bench-scale level, the apparatus used was a
6-inch diameter, 8-tray stripping column. The column
accommodated different tray designs, and six different tray
designs were evaluated (one bubble cap design and five
variations of sieve plate designs having open areas ranging
from 1 to about 10 percent determined by hole size and
number). Liquid level on all the trays, regardless of dcsign,
W.IS about .~.~) i.llCIleS for all experiment~l rU~lS. 'L'llo l'VC
slurry was used at 25~o by weight total solids in water as
the liquid carrier. The PVC slurry was pumped into the
column above the first tray at a feed rate of about 2 to
about 45 pounds of resin per hour. The feed rate controlled
residence time. ~s column capacity was about 1.4 gallons,
residence time in the column, based upon the feed rates given
above, ranged from about 2 minutes to about 45 minutes.
Initial residual VCM in the PVC polymer was about 1000 to
2000 ppm. Levels of residual VCM in the stripped PVC as
low as 10 ppm were obtained. VCM content was determined
using gas chromatograph analysis.
Experiment A
A series of runs were made in the strippingcolumn
using a common sieve-tray design throughout the series. The
runs were performed at about 215F. (102C.) and about 16 psi
absolute. The PVC used was a suspension PVC homopolymer
- 10 -
l~q86~9
5 having an averag~ particle si~e of about 130 microns ~nd a
porosity of ahout 0.14 cc./gm. At ~eed rates of the PV~
slurry ~rom about 12 to about 44 pounds of PVC per hour,
residence time varied from 2 minutes -to about 8 minutes.
Efficiency of VCM stripping was evaluated, at the different
flow rates, by measuring VCM content both before entry into
the column and a~ter exit. Percent removal of residual VCM
varied from about 60% to about 98% by weight (based upon the
original VCM content in the PVC polymer). Higher stripping
efficiency occurred at longer residence times (6 to 8 minutes)
indicating that a balance between feed rate and operating
conditions exists for optimum VCM removal in the stripping
operation.
Experiment B
A series of runs was made in the stripping column
using a common sieve-tray design and a co~non feed rate for
all the runs. The PVC polymer and slurry used in Experiment
A was used for these tests also. The purpose of this series
of runs was to explore temperature and pressure effects upon
stripping of VCM from PVC polymer. Saturated water vapor
(steam) was used as the hot gas. Again, residual VCM in
the PVC was measured at entry and upon exit in the column.
Results of the runs are given in the following table.
Temperature PressurePercent Removal
(C.) (mm. of Hg)of Residual VCM
Residence time of 4 minutes
(7 300 20
91~ 630 55
100 760 75
109 1100 97.5
110 1150 98-5
Residence time of 12 minutes
77 300 5
94 630 85
100 760 93
109 1100 over 99
110 1150 over 99
~zq86~9
The data shows the unexpected result that much
more effective and efficient removal of residual VCM is
obtained at conditions of both higher temperature and
pressure. Prior to this discovery, it was generally
believeA str-ipping of ~CM under reduced pressure (vacuum)
was requir~d f`or et'fectivc renloval of' residllal morlomer i~
the polymer. Operating conditions of from about 200~.
(93C.) to about 270~, (132C,) at pressures from atmospheric
to about 40 psi absolute were evaluated, Within this range,
little or no degradation of the PVC polymer was observed,
Semi-Work Stage
Based upon favorable results obtained in the
bench-scale stripping column experiments, tests were
scheduled to evaluate the feasibility of using the method
on a larger scale. Again, a stripping column apparatus was
chosen in which to conduct and evaluate the novel process,
The column used was a 30 inch diameter stripping column
having 18 trays (sieve type), Capacity of the column was
limited by its flood point which was about 50 gal,/minute
of liquid feed, The PVC slurry was used at about 30% total
solids by weight in water. PVC polymer feed rates were
from about 1700 pounds per hour to about 660o pounds per
hour, Steam feed rate ranged from about 1000 to about
4500 pounds~hour, Residence times in the column varied
from about 1 minute to about 10 minutes (based on a column
holdup of about 98 gallons), Extensive tests were conducted
at various PVC polymer slurry and/or steam feed rates and
different operating temperatures and pressures,
EXAMPLE I
A PVC homopolymer used as a 30% by weight slurry
in water was fed to the stripping column which was operating
at 212F. (lOO~C.) and atmospheric pressure, The polymer
- 12 -
lZ786.'39
has an average particle size OI about 100 microns and a
porosity of about 0.12 cc./gm. Resul-ts of the tests show
that 80% to 90% of the residual VCM was removed :from the
PVC p o lym~ :r .
1~786~9
L~ O O
.,~
~ F~l
o
v n,
O O O
v ~ N
F~
.
U~
CU
~ a~
~n
a) r
E~
~ ~D
a
~\
F4 ~ O g O
N ~) N ~1
td a) N ~ ~ :~
a~ ~
~ ~ ~ .
q~
o
_
L~
~1 l~i N (Y~ tr
v a
F~ F~
~¦ ~I N
P~
- 14 -
~27~36'39
EXAMPLE II
The PVC homopolymer slurry used in Example I was
fed into the column at a feed rate of 15 gpm (2500 l'~/hr. of
PVC). Steam feed rate was 1500 ~-/hour, and residence time
was about 6.5 minutes. Operating conditions were 215F.
(102C.) and about 15 psi absolute. Initial VCM content
of the PVC polymer was 2460 ppm, and final residual VCM
content after stripping was 330 ppm, indicating a removal
of 86% by weight of VCM.
EXAMPLE III
The PVC homopolymer slurry used in the previous
examples was used in a series of runs at varying temperature,
pressure, PVC slurry feed rate, and steam feed rate. Results
of the runs are given in the following table. In all cases,
until run No. 5 where the PVC slurry feed rate was approach-
ing 40 gpm, over 90% of the VCM was removed.
- 15 -
~LZ786''3~
~ cd
a~ ~
U o Lr\ , ~ ~o ~ ~
a~
a) ~
~;
~1ooooo
a~ ~ ~ ~ cu o
o ~,
O
V {~,
O O o o O
~' ~ ~ l O L~ L~
v H co N ~D 15
N N CU ~1
a) a~
L~
-rl~0 ~0 N
.
o o o o o
a) a) +~ ~ o o o o L~
O t~ \ L~ ~ LO~ O N
; ~~1 ~1 ~I N (~
~a)
~I P~
Cl~ ~L~ ~ O L~
~ t~O ~1 ~I r-l N~) S~
v a~ ~,
O
5~ O
td
~rl ~ O C`J ~ ~ O
O ~,
P~ ,Q
h
O
~d ~J LS~ O ~
N N ~ ~r)trl 1~4
o ~4 N N N N N
~o
a) r1
E~
~ ¦ ~I N ~) ~L~
cr;
- 16
1~786~
EXAMPLE IV
PvC ho~opolymer stripped at various temperatures
and pressures was checked for its stability using the
! Capillary Viscometer Heat Stability Test. A sample of the
same type o~ PVC homopolymer, which was not stripped using
the novel process, was run for comparative purposes. The
data ahows that the stripping process had little or no
effect on the heat stability o~ the PVC polymer.
Stripping Conditions Min. to Develop Color
Temperature, F Pressure, psia Control Stripped Sample
215 15 18 18
232 21 19 18
237 23 19 19
EXAMPLE V
A series of runs were made using a PVC homopolymer
having a porosity of about 0.15 cc./gm. and an average
particle size of about 130 microns. All the runs were
made at 217F. (103~C.) and 15 to 16 psi absolute. The PVC
polymer slurry feed rates and steam feed rates were varied
from 10 gpm (1650 pounds per hour of PVC) to 40 gpm (6600
pounds per hour of PVC)~ and steæm feed rates were set at
from 1500 to 4400 pounds per hour. Residence time varied
inversely as the PVC slurry feed rate and ranged from 2.5
minutes to 10 minutes. Percent removal of residual VCM
monomer varied from 86~ to 99.8~ by weight. At a condition
f 25 gpm PVC slurry feed rate (~100 pounds per hour o~
PVC) and 2000 pounds per hour of steam feed rate, a
residual VCM content of 3 ppm was obtained from a PVC
polymer having an initial VCM content of 1500 ppm.
EXAMPLE VI
Using the same PVC homopolymer as in Example V,
and operating at 235F. (112C.) and 23 psi absolute using
a slurry feed rate of 25 gpm and a steam feed rate of
2000 #/hour, the VCM content of the polymer was reduced from
- 17 -
3020 pprr, to 25 ppm, a 99.2 percent VCM rer~oval.
kXA~L~ VII
rr'he e~)erimental series of runs made in ~ample V
were essen-tlal~y repeated but for the use of a PVC homo-
polymer rtc-;in ilaving a porosity of about 0.25 cc./gm. and
an avera~e par-tic:Le size of about 130 microns. The PVC iS
more porous than those used in Example I (0.12 cc./gm ) and
Example V (0.15 cc./gm.). Particle sizes of the three ~ypes
of PVC hornopolymers are roughly equal, being about 100 to
130 ~nicrons on the average. In a range of temperature from
215~F. (102C.) i,o 236F. (113C.), pressures from atmospheric
to 23 psi absolute, PVC slurry feed rates of from 15 ~pm to
35 gpm (2500 to 5800 pounds per hour of PVC)g and steam feed
rates Of from 1500 to 2500 pounds per hour, residual VCM
content in the FVC polymer, after stripping was below 1 ppm
in all runs bUt for one out Of fourteen runs. In that one
run, the percent removal of VCM was still 99 percen-t by weight.
- 18 -
