GB1580225A

GB1580225A - Removing monomer from vinyl chloride polymer slurry

Info

Publication number: GB1580225A
Application number: GB3129777A
Authority: GB
Original assignee: ICI Australia Ltd
Current assignee: Orica Ltd
Priority date: 1976-07-27
Filing date: 1977-07-26
Publication date: 1980-11-26
Also published as: SE438861B; JPS5334888A; NZ184661A; FR2359848B1; IT1143747B; SE7708564L; DE2733679A1; FR2359848A1; NL7708168A; ES461073A1

Description

(54) REMOVING MONOMER FROM VINYL CHLORIDE POLYMER SLURRY (tl) We, ICI AUSTRALIA LIMITED, of 1 Nicholson Street, Melbourne, Victoria, Australia, a Company organised and existing under the laws of the State of Victoria, Australia, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement:- This invention relates to a process for the stripping of vinyl chloride monomer (VCM) from slurries of particulate poly(vinyl chloride) (PVC) polymers in water (the term "poly(vinyl chloride) polymers" is intended to embrace homo- and Co- polymers of vinyl chloride).

Following the recognition of the health hazards associated with exposure to VCM vapour there has been a major effort throughout the world to find means of eliminating the release of VCM to the atmosphere during the production and processing of PVC polymer. Considerable advances have been made in reducing the release of VCM to the atmosphere during the production of PVC polymer, however, atmospheric contamination through release of unreacted VCM from the polymer both during and after processing is still a largely unsolved problem.

PVC polymer can be produced by a number of methods including emulsion, gasphase, bulk, solution, dispersion and suspension polymerization processes. However, in all processes it is usual practice to terminate the reaction cycle before complete conversion to the monomer, either because the reaction rate becomes uneconomical to maintain or a polymer of high porosity is required. Thus unreacted monomer is left adsorbed, dissolved or otherwise occluded in the polymer. In the suspension polymerization process conversion is normally between 83 and 90% and most of the unreacted monomer is recovered for recycling by vacuum stripping the reaction product slurry either in the reaction vessel or in a special stripping vessel. However, traces of VCM still remain in the polymer and this residual VCM is difficult to remove.

It is known that residual VCM can be removed from the polymer by heating, indeed one of the reasons why residual VCM needs to be removed from PVC polymer is the-health hazard associated with the release of the residual VCM during processing of the polymer. However, it is also well known that at temperatures above 1000C PVC begins to slowly decompose, the rate of decomposition increasing with increasing temperature. Any appreciable thermal degradation of the polymer can seriously affect its physical properties and make it unsuitable for processing. As a result, the processes developed to strip unreacted monomers from more heat stable polymers, such as for example, acrylonitrile from poly(acrylonitrile) and styrene from styrene-butadiene rubber, are not applicable to the stripping of residual VCM from the relatively heat sensitive PVC polymer. Thus it is of considerable importance that an economic process be found by which residual VCM may be efficiently removed from PVC polymer without affecting the physical properties of the polymer.

In the search for such a process PVC polymer producers have now turned to stripping the residual VCM from PVC polymer, either in the reactor vessel or in a special stripping vessel by applying heat, usually in the form of live steam, to the polymer. However, such processes have not proved successful in reducing the VCM content of the polymer to the required levels of below 10 parts per million (ppm) and preferably below 1 ppm without causing undesirable thermal degradation of the polymer.

Stripping VCM from a slurry of particulate PVC polymer in water by passing the slurry down a column of common sieve tray design with downcomers while passing steam up the column results in an improved stripping efficiency when compared to batch stripping in a vessel by the injection of live steam. However, in such a column the particles have a wide residence time distribution, some of the particles passing through the column quickly and not being effectively stripped while some of the particles remain on the column for a long time thereby being effectively stripped but also undergoing undesirable thermal degradation. Increasing the median residence time using such a column increases the percentage of effectively stripped particles, but because of the wide residence time distribution more particles are subject to undesirable thermal degradation. Thus a column of common sieve tray design with comers is not satisfactory for stripping residual VCM to an acceptable level.

An ideal continuous process for the removal of residual VCM from particulate PVC would appear to be a steam stripping process using a column in which all particles could be held on the column for the same length of time, thereby subjecting all particles to the same stripping conditions.

We have now found a process whereby VCM can be effectively stripped from PVC polymer by passing a slurry of polymer particles in water through a column fitted with perforated plates but no downcomers, the slurry in effect being held on the perforated plates by a flow of steam up the column through the holes in the platens, the steam flow being adjusted to allow the slurry to move down the column by "weeping" through the holes in the perforated plates. Surprisingly, the particle residence time distribution in the process of the invention is much narrower than that which can be obtained using a column of common sieve tray design with downcomers, and as a result the particles may be stripped to a very low VCM content without causing undesirable thermal degradation of the polymer.

Accordingly we provide a process for the removal of vinyl chloride monomer from a poly(vinyl chloride) polymer which process comprises passing a slurry of particles of said poly(vinyl chloride) polymer in water through a substantially vertical column provided with a series of horizontal perforated plates but with no liquid downcomers, the slurry being fed onto the column at or near the head of the column and a flow of steam being fed into the column at or near the foot of the column at a rate sufficient to allow the slurry to weep through the perforations against the flow of steam without flooding the column and wherein: the temperature of the slurry in the column is between 90"C and 150"C; the median residence time of the particles in the column is between 1 and 20 minutes; the semi interquartile range of the residence time of the particles in the column is less than 5 minutes; the stripped slurry is removed from the column at or near the foot of the column; and steam and vinyl chloride monomer are removed from the column at or near the head of the column.

The PVC polymer may be a homopolymer of vinyl chloride or a copolymer of vinyl chloride with one or more other copolymerizable vinyl monomers such as, for example, vinyl acetate vinylbenzene, vinylchloride, vinyl bromide, acrylic and methacrylic acids and their esters, ethylene and propylene. However, due attention must be paid to the softening point and thermal stability of the polymer when determining the operating temperature and median residence time. A polymer which softens appreciably or decomposes at an appreciable rate below 90"C may not be suitable for treatment by the process of the invention. The process of the invention finds its best application in stripping residual VCM from PVC homopolymer.

The process of the invention may be used to strip residual VCM from PVC polymer prepared by any of the known methods of PVC polymerization including, for example, emulsion, gas-phase, bulk, solution, dispersion and suspension polymerization processes. However, for treatment by the process of the invention the PVC polymer must be in the form of a slurry of PVC polymer particles in water, and therefore if the PVC polymer is not produced in particulate form it must be reduced to such form before treatment. In the suspension polymerization process PVC polymel is produced in the form of a slurry of particles in water and thus the process of the invention is particularly convenient for stripping residual VCM from PVC polymer produced by suspension polymerization.

The size of the PVC polymer particles is not critical in respect to stripping efficiency. For example, particles of a PVC homopolymer prepared by the suspension polymerization process are aggregates of primary particles, and in the process of the invention the rate of stripping of VCM is dependent on the rate of diffusion of VCM from the primary particle and is therefore not dependent on the overall size of the particle as measured by typical screen analysis. However, in the process of the invention the particles weep through the perforations in the plates and therefore it is important that the particles are screened before being introduced onto the column to ensure that the particles will move freely through the perforations.

The proportion of solid PVC polymer in the slurry is not critical and can range from a very low percentage to a high percentage limited by whether it can be pumped.

Typically the percentage of PVC polymer in the slurry varies from about 10% to about 60% by weight and in practice is preferably in the range of 20% to 45% by weight.

The size of the column is not narrowly critical, the column diameter and number of plates basically being determined by the size of the operation. Satisfactory results have been obtained from the process of the invention using columns containing from 5 to 60 perforated plates and columns from 16 cm to 61 cm in diameter, but these figures should in no way be regarded as limiting. For some operations where continuous processing is not important it may be preferable to multiple pass a batch of slurry through a column with a small number of plates rather than make a single pass through a multiplate column.

The stripping column used in the process of the invention is designed to obtain a desired throughput and reduction in VCM content of the polymer without thermal degradation of the polymer. The reduction in VCM content of a polymer is controlled by the temperature of operation and the period of time for which the particles are heated. Ideally all of the particles should be heated for the same length of time and in practice the residence time distribution of the particles around the median residence time should be as narrow as possible. In practical terms this means that the column should be constructed so as to avoid dead spaces, to ensure that the walls of the column and the undersides of the plates are washed and that the slurry held on the plates by the flow of steam is well agitated.

The residence time distribution of the particles in the column according to the process of the invention is also influenced by the liquid level on the plates. For a given column throughput a low liquid level on a large number of plates results in a narrower particle residence time distribution than a higher liquid level on a correspondingly smaller number of plates The capacity of the column and the median residence time of the particles on the column are to a large extent determined by the size of the column, the number of plates in the column, the percentage free area of the plates and the flow of steam up the column. Design features which affect the operation of a column include the size of the perforations in the plates, the percentage free area of the plates, the nature of the perforations, the thickness of the plates and the spacing between the plates.

The size of the perforations in the plates should be large enough to allow the slurry to weep through easily, but for economic operation should not be so large that the velocity of steam required to hold the slurry on the plates is extremely large. The percentage free area or perforation area of the plates is to a large extent determined by a balance between the required median residence time (how well the polymer is to be stripped) and steam usage (economics). In practice if the size of the perforations is too small tliey are prone to blockage, and if too large fewer perforations are required for the same percentage free area, resulting in a decrease in slurry agitation and a corresponding increase in particle residence time distribution. In order to obtain good agitation of the slurry the perforations are preferably substantially evenly spaced over the surface of the plate.

By substantially evenly spaced we mean that the perforations should not be all grouped in one section of the plate, but should be spaced over the plate surface.

A perforation size equivalent to a hole having a diameter in the range of 2 mm to 20 mm (i.e. area of 3 mm2 to 300 mm2) is satisfactory with a plate-free area ranging from 5% to 15%. Preferably perforations of hole diameter in the range of 5 mm to 15 mm (i.e. area of 20 mm2 to 180 mm2), more preferably about 10 mm (i.e. area of about 80 mm2), and a free area of 5% to 10%, more preferably about 7%, is used together with a steam velocity in the holes of 3 to 20 m/sec. However, the teaching of this specification, those skilled in the art will be able to determine other combinations by simple experimentation.

In order to avoid the build up of PVC polymer particles on the surface of the underside of each plate it is preferable that some of the slurry which weeps through the perforations is made to run along the surface of the underside of the plate so that the underside of the plate is continuously washed. Thus, although the perforations may be of any desired shape, we have found that cylindrical, unbevelled, sharp edged holes without any projecting pieces of metal, drilled vertically through the plates and spaced substantially evenly over the surface of each plate, are particularly satisfactory in causing the underside of the plates to be washed. Those skilled in the art will be able to employ other expedients to cause the under surface of the plates and column dead spaces to be washed.

The thickness of the perforated plates used in the process of the invention is not critical. However, as the slurry weeps through the perforations in the plate against the flow of steam through the perforations, for a given perforation size and percentage free area and steam rate, the weeping rate will be slower for a thick plate than for a thin plate. In practice we have found that 3 mm to 10 mm thick plates are satisfactory; however, this thickness should not be construed as limiting, as plate thickness will to some extent be determined by the column size and plate material.

The spacing of the plates in the column is not narrowly critical. However, the plates should be spaced far enough apart to prevent premature flooding of the column, but close enough together to allow the splashes from a plate to wash the walls of the column up to the next plate in order to avoid PVC polymer build up on the walls of the column. In practice we have found a plate separation in the range of 10 cm to 40 cm to be satisfactory and 1S cm to 20 cm preferable.

By flooding we mean the situation wherein the space between the adjacent plates of the column is filled with so much slurry that there is no free space between the top of the slurry and the under surface of the top plate. In a flooded column the space between adjacent plates is completely filled with a foam of slurry and stripping fluid.

It should be noted, that when the PVC polymer is introduced onto the column above the uppermost plate of the column, it may be necessary to provide a means of washing the walls of the column above the uppermost plate to ensure that there is no polymer build up on the walls of the column above the uppermost plate. Suitable means include directing a flow of water, preferably hot water, onto the walls of the column above the uppermost plate to ensure that polymer particles splashed onto the walls are washed back onto the perforated plate. Other means will be evident to those skilled in the art.

For a given column, the parameters which control the operation of the column include column temperature, slurry feed rate and steam rate.

In stripping VCM from PVC polymer according to the process of the invention the rate of stripping is determined by the rate of diffusion of the VCM from the polymer. The rate of diffusion is a function of the nature of the polymer and the temperature of the stripping operation. Thus the amount of VCM stripped from a polymer will depend on the temperature at which it is stripped and the time for which it is held at that temperature. The rate of stripping increases with increasing temperature, however, PVC polymer undergoes slow decomposition at temperatures above 1000C and therefore it is essential that the polymer not be heated above 1000C for an extended period of time.

In a column stripping process it is possible to control the median residence time of the particles on the column (the median residence time being that time taken for half the number of particles in a given charge of slurry to pass through the column). However, if the residence time distribution of the particles is wide, some of the particles will pass rapidly through the column without effective stripping, while others will remain on the column for a period of time considerably longer than the median residence time. A convenient measure of the residence time distribution is the semi-interquartile range which is defined as half of the difference in time between the first of the second quartile and the last of the third quartile to pass through the column. It is therefore evident that in order to effectively strip VCM from PVC polymer without thermal degradation of the polymer it is essential that the column temperature and the median residence time of the particles on the column be carefully controlled and that the residence time distribution of the particles on the column be as narrow as possible.

In practice we have found column temperatures ranging from 900C to 1SO"C, median residence times ranging from 1 to 20 minutes and semi-interquartile residence time of less than 5 minutes satisfactory. However, it should be recognised that the temperature of operation, median residence time and the semi-interquartile range are inter-related. At the higher end of the temperature range the rate of thermal degradation of the polymer will be faster. Thus in order to effectively strip VCM from PVC polymer without thermal degradation of the polymer, when a temperature in the higher part of the temperature range is used the median residence time and semi-interquartile range should preferably be in the lower part of the range. Conversely when a temperature in the lower part of the temperature range is used, the median residence time may if necessary be in the higher part of the range. Precise limitation of the temperature, median residence time and semi-interquartile range is not possible as these parameters will depend on the column, the grade of the PVC polymer being treated and upon the stabilizing agents and other additives present in the polymer. However, we have found column temperatures in the range of 100"C to 1200C preferable combined with median residence times necessary to give the required degree of stripping without thermal degradation of the polymer. For highly temperature sensitive grades of PVC polymer we prefer to use temperatures in the lower part of the range and operating conditions such that the semi-interquartile residence time is less than 2 minutes. Such conditions may be readily achieved using the process of the invention.

In the case of a readily stripped polymer it may be desirable to reduce the residence time by introducing the slurry feed lower down the column rather than run the column with a very low steam rate.

The pressure in the column is normally substantially atmospheric; however, the pressure may be above or below atmospheric pressure in order to obtain the desired operating temperature. Moderately super-atmospheric pressures such as for example between 1 and 1.5 atmospheres are preferred. Preferably the pressure drop from the foot to the head of the column is kept low so that the temperature gradient is small.

In the perforated plate column without downcomers according to the process of the invention, for a given column the median residence time of the particles on the column is determined by both the feed rate of the slurry onto the column and the steam rate. This is in direct contrast to a normal sieve plate column with downcomers in which, for a given column, the median residence time is determined only by the feed rate.

In controlling the median residence time of the particles in the column the feed rate and steam rate control the column liquid and vapour loading. We have found that stable column operating conditions according to the process of the invention may be obtained when the slurry feed rate and steam rate are adjusted to give column vapour and liquid loading between 50% and 90% of the column flood point provided that the vapour loading is always greater than the liquid loading.

Under certain conditions of column operation and slurry feed temperature the amount of steam required to maintain stable operating conditions is insufficient to raise the temperature of the slurry to the desired operating temperature. Under these conditions it is advantageous to heat the inlet slurry before feeding it into the column. This may be simply done by injecting live steam into the slurry feed line immediately before the slurry is fed onto the column.

As stated hereinbefore, in stripping VCM from PVC polymer according to the process of the invention the rate of stripping is determined by the rate of diffusion of VCM from the polymer which is in turn dependent on the stripping temperature and the nature of the polymer. One often used measure of the ease with which a polymer can be stripped is the porosity of the polymer. We have found that the process of the invention may be used to satisfactorily strip VCM from PVC having a wide range of porosity. The invention is particularly suited to stripping VCM from all grades of PVC produced by the suspension polymerization process, without thermal degradation of the polymer.

In practice we have stripped PVC polymer of porosity as low as 0.7 cc/g at 1 100C with a steam usage of 0.25 to 0.35 units per unit of polymer to give a polymer with a residual VCM content of less than 1 ppm without undesirable thermal degradation of the polymer. More porous PVC may be stripped more readily with a corresponding lower steam usage.

As stated hereinbefore, PVC polymer is subject to thermal degradation when heated for extended periods above 1000C and it is essential in a column stripping process than the residence time distribution be narrow. The accompanying drawing illustrates the narrow residence time distribution obtained using a column according to the process of the invention in comparison to the residence time distribution obtained using a sieve plate column with downcomers. Curve 1 shows the residence time distribution for a 57 plate column according to the process of the invention with a mean residence time of 18 minutes and a semi-interquartile range of 3.9 minutes. Curve 2 shows the residence time distribution for a 45-plate sieve plate column with downcomers not of this invention with a mean residence time of 26 minutes. The figure clearly illustrates the narrow residence time distribution obtained using a perforated plate column according to the process of the invention in comparison to the broad residence time distribution of the sieve plate column with downcomers. It is clear that in the sieve plate column with downcomers some particles pass through the column well before the mean residence time and, more importantly, a considerable proportion of the particles slowly "tail" off the column well after the mean residence time.

It is completely unexpected that the process of the invention should give such a narrow residence time distribution. In the past columns similar to that used in the process of the invention, and known as dual flow columns, have been used for the distillation of liquids where the liquids are of such a nature that there is a risk that blocking of downcomers would occur as the distillate is removed due to the precipitatlon of undesired solids or tars. In such a situation narrow residence time distribution of the solids is not required and no adventitious observation of this property has ever been made. Moreover, there is no problem of downcomer blockage when stripping a slurry of PVC polymer particles in a sieve tray column with downcomers and thus there is no obvious advantage to be gained in using a column without downcomers.

Furthermore, normal plates used in dual flow columns for the distillation of liquids are provided with punched holes usually with sharp downward protrusions to cause the liquid passing through the plates to form drops on the sharp protrusions which drop down onto the next plate. In such a system the under side of the plates are not washed by liquid passing down through the holes, and hence, if used in the process of the invention, would not achieve the desired narrow range of residence time of slurry particles. In addition, in conventional dual flow columns used for the distillation of liquids, the plates are usually spaced much further apart than the spacing preferred according to the process of the invention. Hence, if used in the process of the invention such spacing would not enable splashes from a plate to wash the walls of the column up to the next plate in order to avoid polymer build-up on the walls of the column and this would not achieve the desired narrow range of residence time of slurry particles. In conventional dual flow column operations any build-up of solids on the inner surface of the column is not detrimental to the process, as such solids are only undesired products of the process.

As stated hereinbefore, PVC polymer may be stripped according to the process of the invention to very low final VCM content. The efficiency of the process of the invention is illustrated in Table I below for the stripping of PVC homopolymer of porosity about 0.07 cc/g prepared by suspension polymerization.

TABLE I

Stripping Median Residence VCM Concentration (ppm) Temperature Time OC Minutes Inlet Outlet 100-105 15.0 c. 40,000 < 1 110 12.0 14,000 < 0.1 113 4.9 560 (0.2 115 6.0 670 < 0.2 117 4.8 745 -(0.5 Samples of PVC stripped by the process of the invention were carefully examined for signs of undesirable thermal degradation. Tests included polymer colour (Yellowness Index), dynamic heat stability (Haake Dynamic Heat Stability Test) and various static heat stability tests. The test results showed that the process of the invention had little, if any, effect on the physical properties of the PVC, the stripped PVC having an acceptable colour and heat stability for all commercial applications. Thus the process of the invention offers a novel method for stripping PVC to a final VCM content of below 1 ppm without effecting the physical properties of the polymer.

The mixture of steam and stripped vinyl chloride monomer passing out of the head of the column is conveniently passed through a condenser, the condensed water being returned to the column and ultimately being removed with the stripped slurry at the bottom of the column. The residual vinyl chloride after removal of steam may be fed into a conventional vinyl chloride recovery unit. Alternatively, the mixture of steam and vinyl chloride may be passed to a vinyl chloride recovery unit without prior condensation of the steam.

The invention is now illustrated by, but by no means limited to, the following examples. All VCM analyses were carried out by gas chromatography using standard methods.

Examples 1 to 4 illustrate the narrow particle residence time distribution obtained using the process of the invention to strip VCM from particulate PVC homopolymer.

Examples 5 to 16 illustrate the high efficiency of the process of the invention in stripping VCM from particulate PVC homopolymer using a small column.

Examples 17 and 18 illustrate the high stripping efficiency of the process of the invention in stripping VCM from particulate PVC homopolymer using a pilot scale column.

Example .19 illustrates the negligible effect that the stripping process of the invention has on the phvsical properties of PVC homopolymer.

In all Examples the PVC homonolymer was prepared by suspension polymerization and had a particle size between 100 and 200 microns.

Example 1.

A 0.16 m diameter glass column was set up containing seven stainless steel plates (5 mm thick) set 15 cms apart. The plates were provided at uniform spacing with holes 1 cm in diameter. The holes being drilled vertically through the plate. The holes occupied 5.5% of the area of each pla

TABLE

Time elapsed since coloured particles were introduced to Normalised exit slurry feed point (min) concentrations 0 0 0 1 0.0057 11/2 0.0758 2 0.1804 21/2 0.2180 3 0.1886 3/2 0.1351 4 0.0862 4 0.0509 5 0.0285 5'/2 0.0154 6 0.0081 61/2 0.0041 7 0.0021 7 0.0010 Median residence time is 21/2 minutes Mean residence time is 3 minutes Semi interquartile range is 3/4 minute Example 2.

A column was prepared as in Example 1 except that 19 plates were used at 15 cm spacing with a free area of 7.5% instead of 5.5%. All other dimensions were the same.

The column was operated in the same manner as in Example 1 except that in place of the coloured PVC particles a pulsed charge of a solution of radio active sodium ion was added. The residence time was monitored by measuring the radio activity of the product from the column. The following results were obtained.

Steam consumption was 0.3 Ib of steam per 1 lb of PVC processed. Slurry feed rate to the column was 1.6 l/m. VCM inlet concentration was 1250 ppm.

VCM exit concentration was 40 ppm of VCM in solid PVC.

The residence time distribution for the PVC particles is shown in Table III.

TABLE Ill

Time elapsed since radio isotopes were introduced to slurry feed Normalised exit point (min) concentrations 0 0 0 1 0 1 0 2 0 21/2 0 3 0.045 3 0.112 4 0.157 4 0.213 5 0.191 5 0.146 6 0.079 61/2 0.034 7 0.011 7 0.011 8 0 Median residence time is 4 minutes Mean residence time is 4.7 minutes Semi interquartile range is 3/4 minute Example 3.

The same column as used in Example 2 was operated under the following conditions. Steam consumption was 1 lb steam/lb PVC. Slurry feed rate was 0.8 l/m.

VCM stripping effectiveness was such that the VCM concentration in the slurry feed to the column was 550 ppm and no VCM could be detected in the slurry exit stream (meaning that the VCM concentration was below 1 ppm).

The residence time distribution is shown in Table IV. TABLE IV

Time elapsed since radio isotopes were introduced to slurry feed Normalised exit point (min) concentrations 0 0 0 1 0 1 0 2 0 2 0 3 0 3 0 4 0 4 0 5 0.005 5 0.014 6 0.035 6 0.062 7 0.081 71/2 0.119 8 0.141 8 0.136 9 0.122 9 0.103 10 0.076 10% 0.049 11 0.030 11 0.016 12 0.005 12l/2 0.003 13 0.002 13l/2 0 Footnotes for Table IV.

Median residence time is 8 minutes Mean residence time is 8.5 minutes Semi interquartile range is 1 minute Example 4.

Example 3 was repeated using a 40% w/w PVC slurry, stripping temperature of 114"C, slurry feed rate of 1.7 1/min and column steam rate of 31 Ib/hr. Steam was injected into the slurry feed line at a rate of 32 Ib/hr to preheat the slurry before it entered the column. Overall steam use was 0.3 lb per lb of PVC. Inlet VCM concentration in the PVC was 83g ppm and outlet VCM concentration in the PVC was 6 ppm.

The residence time distribution is shown in Table V.

TABLE V

Time elapsed since radio isotopes were introduced to slurry feed Normalised exit point (min) concentrations 0.0 0.000 0.5 0.000 1.0 0.000 1.5 0.004 2.0 0.025 2.5 0.097 3.0 0.140 3.5 0.165 4.0 0.174 4.5 0.157 5.0 0.110 5.5 0.058 6.0 0.035 6.5 0.021 7.0 0.010 7.5 0.004 8.0 0.000 Median residence time is 4.2 minutes Mean residence time is 4 minutes Semi interquartile residence time is ?4 minute Examples 5-16.

A vertical 0.16 m diameter glass column was set up containing 19 stainless steel plates of 5 mm thickness set 15 cm apart. The plates were provided at a uniform spacing with 10 mm diameter holes, the holes having been drilled vertically through the plates. The total area of the holes on each plate was 7%. The plates were held in position by a centrally located rod. The head of the column was fitted with a water cooled condenser and the effluent from the top of the condenser was led into a VCM recovery system. The stripped slurry was removed from the foot of the column.

Steam was introduced into the column below the lowest plate and a slurry of PVC particles in water (40% w/w PVC) was added continuously to the top plate.

The column was operated under a range of conditions and the operating and stripping efficiency is shown in Table VI below.

TABLE VI

Example No Operating Parameters 5 6 7 8 9 10 11 12 13 14 15 16 PVC Porosity (cc/g) 0.08 0.07 0.25 0.27 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 Slurry Feed Rate (1/min) 0.8 1.2 1.5 1.5 1.5 1.9 1.4 1.7 1.7 1.6 1.5 2.0 Preheat Steam Rato (lb/br) 0 13 18 16 13 17 17 27 32 25 23 36 Column Steam Rate (lb/hr) 48 27 47 38 47 48 28 28 31 28 31 23 Steam Usage (lb/lb PVC) 0.98 0.37 0.52 0.30 0.51 0.43 0.33 0.27 0.3 0.29 0.34 0.19 Column Temperature ( C) 100 100 111 111 111 113 113 113 114 115 117 115 Mean Residence Time (min) 8.5 5.8 5.3 4.7 5.9 4.9 5.4 4.4 4.0 5.1 4.8 6.0 VCM Content Inlet 550 1180 365 271 636 560 1017 898 839 1081 745 670 of PVC (ppm) Outlet < 1 5 0.2 0.2 0.4 < 0.2 8 10 6 9.1 < 0.6 < 0.2 Examples 17 and 18.

A vertical 0.25 m diameter stainless steel column was set up containing 30 stainless steel plates set 20 cm apart. Each plate was provided at a uniform spacing with 54 10 mm diameter holes. The total hole area of each plate was 7%.

The column was operated using the same general procedure described for Examples 5 to 16. A slurry of PVC particles in water (40 /O w/w PVC) obtained directly from a pilot scale reactor was passed twice through the column representing stripping with a 60 plate column. The column operating conditions and results are shown in Table VII below.

TABLE VII

Example No Operating Parameters 17 18 Slurry Feed Rate (Ib/hr) 1030 1300 Total Steam Rate (Ib/hr) 130-140 160-175 Column Temperature (OC) 110 110 VCM Content Inlet 10400 14000 of PVC (ppm) Outlet 5 0.08 Product Colour1 5.5 7.3 Mean Residence Time (min) - 16 Yellowness Index determined using a Hunterlab E23D Colour and Colour Difference Meter.

Example 19.

Samples of polymer from Examples 7, 11 and 13, taken both before and after stripping, were formulated and tested for dynamic heat stability using a Haake Rheometer. The results are shown in Table VIII below.

TABLE VIII

Test Sample Appearance at Time (min) Type of Temperature Sample Formulation ( C) 3 6 9 12 15 Example 7 Stripped Pipe dry blend 210 Opaque Opaque Opaque Opaque white off white pale yellow brown+ Example 7 Un-Stripped Pipe dry blend 210 Opaque Opaque Opaque Opaque white off white pale ywllow brown+ Example 11 Stripped Bottle dry blend 180 Clear Clear Clear Clear Clear colourless colourless colourless pale yellow light brown+ Example 11 Unstripped Bottle dry blend 180 Clear Clear Clear Clear Clear colourless colourless colourless pale yellow light brown+ Example 13 Stripped Bottle dry blend 180 Clear Clear Clear Clear Clear colourles colourles colourless pale yellow light brown+ Example 13 Unstripped Bottle dry blend 180 Clear Clear Clear Clear Clear colourless colourless colourless pale yellow light brown+ + Significant discolouration.

Claims

WHAT WE CLAIM IS:1. A. process for the removal of vinyl chloride monomer form a poly(vinyl chloride) polymer which process comprises passing a slurry of particles of said poly(vinyl chloride) polymer in water through a substantially vertical column provided with a series of horizontal perforated plates but with no liquid downcomers, the slurry being fed onto the column at or near the head of the column and a flow of steam being fed into the column at or near the foot of the column at a rate sufficient to allow the slurry to weep through the perforations against the flow of steam without flooding the column and wherein: the temperature of the slurry in the column is between 90 C and 150 C; the median residence time of the particles in the column is between 1 and 20 minutes; the semi-interquartile range of the residence time of the particles in the column is less than 5 minutes; the stripped slurry is removed from the column at or near the foot of the column; and steam and vinyl chloride monomer are removed from the column at or near the head of the column.
2. A process according to claim 1 wherein the poly(vinyl chloride) polymer is a homopolymer of vinyl chloride.
3. A process according to claim ] or claim 2 wherein the particulate poly(vinyl chloride) polymer is prepared by suspension polymerization.
4. A process according to any one of claims l to 3 inclusive wherein the percentage of solid PVC polymer in the slurry is from 10 percent to 60 percent by weight.
5. A process according to claim 4- wherein the percentage is from 20 percent to 45 percent by weight.
6. A process according to any one of claims 1 to 5 inclusive wherein a perforation ranges in size from 3 square millimetres te 300 square millimetres.
7. A process according to claim 6 wherein the size range is 20 square millimetres to 180 square millimetres.
8. A process according to any one of claims 1 to 7 inclusive wherein the percentage free area of the perforated plates is from 5 percent to 15 percent.
9. A process according to claim 8 wherein the percentage is from 5 percent to 10 percent.
10. A process according to any one of claims 1 to 7 inclusive wherein the size of a perforation is about 80 square millimetres and the percentage free area of the perforated plates is about 7 percent
11. A process according to any one of claims 1 to 10 inclusive wherein the perforations are cylindrical unbevelled sharp-edged holes vertical to the plane of the plate and spaced substantially evenly over the surface of the plate.
12. A process according to any one of claims 1 to 11 inclusive wherein the spacing between the perforated plates is between 10 cm and 40 cm.
13. A process according to claim t2 wherein the spacing is between 15 cm and 20 cm.
14. A process according to any one of claims 1 to 13 inclusive wherein the temperature in the column is between 100"C and 1200 C.
15. A process according to any one of claims 1 to 4 inclusive wherein the semiinterquartile residence time is less than 2 minutes.
16. A process according to any one of claims 1 to 15 inclusive wherein the pressure in the column is between 1 atmosphere and 1.5 atmospheres absolute.
17. A process according to any one of claims 1 to 16 inclusive wherein the slurry feed rate and the steam rate are adjusted so that the column vapour loading and liquid loading are between 50 percent and 90 percent of the column flood point and the vapour loading is greater than the liquid loading.
18. A process according to any one of claims 1 to 17 inclusive wherein the porosity of the PVC polymer is greater than 0.07 cc/g.
19. A process for the removal of vinyl chloride monomer from a poly(vinyl chloride) polymer as in claim 1 and substantially as herein described with reference to the Examples.
20. Particulate poly(vinyl chloride) polymer slurry when stripped using a process according to any one of claims 1 to 18 inclusive.