John Sims
Golder Associates (UK) Ltd.

Radioactively contaminated plastic sheet material is notoriously difficult to compact, incinerate and encapsulate, even with a modest degree of pre-treatment such as shredding. The difficulties arise because, in the case of compaction, the plastic material has quite extraordinary recovery characteristics once the compaction force has been removed. Also, if incinerated the plastic feed will contain high volumes of air which make the control of the incinerator difficult. The encapsulation of shredded plastic tends to produce a matrix with a layered structure and hence low strength.

The author reports on the development work carried out on a system to hot extrude plastic wastes into a drum, excluding unwanted air, and the difficulties of extruding used pvc sheeting and in some cases, mixed plastic type wastes. Also discussed are the process modifications developed to overcome the initial problems, producing a specific waste form, of benefit for compaction, incineration or encapsulation processes.

Treatment of radioactively contaminated plastic type materials has always proved difficult. High Force Compaction of drums of plastic waste has resulted in large recovery factors because of the high hysterisis characteristics of the material. Recovery factors in the order of 50% have been witnessed after applying loads in excess of 1600 tonnes to drums filled with plastic. Although some of the recovery can be eliminated by mixing the soft plastic with hard solid waste, that mixing operation is never easy, particularly in a radioactive environment.

Incinerators fed with quantities of plastic type waste are notoriously difficult to control, resulting in poor off-gas performance, short fire brick life, and therefore, costly maintenance schedules. The prime reason for this problem is the volume of air trapped within the plastic feed stock and the control system’s inability to cater for large changes in the feed composition.

There are some nuclear sites that mix their plastic materials in a shredded form with a cement matrix to encapsulate them prior to long term storage or disposal. However, the mixing process employing a paddle in a drum tends to produce a matrix in which the plastic is in horizontal layers, which in turn, results in a low structural strength of the waste product for final disposal. An undesirable quantity of shredded material always adheres to the blade, producing a matrix which is not homogeneous and therefore, does not meet with some country’s regulatory requirements.

Recycling of plastics wastes collected at large shopping complexes is a well developed industry, employing hot extrusion process equipment. The author has investigated such processes and reviewed the systems to see if they can be applied to one of the most voluminous of nuclear wastes, radioactively contaminated sheet plastic, to produce an output in a minimum volume form ready for immediate disposal. The investigations included a development programme employing typical but non radioactive materials.

The aim was to extrude the air free product directly into a 200 litre drum, which when cooled, would be ready for disposal, avoiding the need for further treatment. As the development phase proceeded it became clear that the pvc content gave rise to operator exposure difficulties as the plastic degraded and, as a consequence, a revised strategy was researched. The result was the addition of equipment to pelletise the extruded plastic immediately it exited the equipment, avoiding the degradation problem. Although the pellet form of the end product was not quite in the air free form envisaged at the commencement of the development work, it was in such a condition to be

• In a reduced volume form
• More easily mixed with hard waste for compaction, thus reducing the recovery factors
• Provide a consistent feed stock for incinerators, reducing the control problems and
• More easily mixed with an encapsulant, providing a higher strength end product.

Hot extrusion of plastics is a well developed industrial process which relies on feeding pellets of the appropriate plastic into a heated chamber fitted with a screw shaft, the latter driving the melting product towards a tapered hole in the end of the chamber. The residence time of the plastic in the heated chamber is so arranged that by the time the product reaches the tapered hole it is in a semi-liquid state and is forced through the hole under the pressure induced by the screw shaft. The air free product which exits the extruder is allowed to fall into a cooled mould where the plastic will quickly solidify, providing a copy of the shape of the mould. There is a thriving industry recycling used plastics arising from shopping malls, producing such objects as gate and fence posts, which due to the nature of the material, are impervious to water and rot.

It was aim of this development programme to investigate the use of this extruder process to accept sheet material, representing the typical feed stock materials which arise at nuclear establishments rather than pelletised plastic employed in normal moulding applications. The other objective was to review the possibility of allowing the hot extruded product to fall directly into a 200 litre disposal drum, analysing if it maintained its molten form sufficiently to permit it to slump, filling the drum completely with a product very close to the plastic’s natural density. Such a product would have little, if any, trapped air, due to pressure extrusion, providing a good volume reduction ratio. A prime consideration behind the investigation into potential equipment was that the process should be readily acceptable to operations within a radioactive environment, that is, meet the nuclear site’s safety requirements, be convertable to a radiologically contained condition and still be maintainable.

Having determined the objectives of the research programme, a suitable extruder and shredder were obtained. At the same time some typical but uncontaminated plastic materials were obtained from a nuclear facility so that the development work would utilise exactly the same specification materials as used on nuclear sites. It was envisaged that the development work should investigate a range of mixed waste commonly found on nuclear sites, such as pvc, hyperlon, neoprene, rubber (leaded and unleaded), polyethelene, cellulosics, etc.

The shredder was required to reduce the nuclear sites typical sheet material into a condition acceptable to the extruder inlet. It was designed to yield a postage stamp sized product, after passing through a screen with 15 mm perforations and which would be transferred into the extruder via a crammer/feeder, the latter employed to prevent the shred material bridging at the extruder inlet and thus avoiding an intermittent flow condition. The extruder was provided with 5 heating stages and 1 cooling stage positioned progressively along the horizontal length of the unit, all controlled by a semi-programmable control unit. This extruder had the capacity to produce temperatures up to 200° Celsius. The movement of the melting product through the extruder is achieved employing a screw shaft, the rotation speed of the shaft being variable to suit the residence time required to melt the material being extruded. At the end of the extruder is a plate which, when the extruder was delivered, had numerous, small diameter holes to produce a multi strand extrusion. For the purposes of the initial development work, the multi hole plate was changed to a single hole system, the single hole being some 15 mm diameter. Catch pots of a variety of sizes and volumes could be positioned below the extruder hole to accept the falling, molten rope of plastic material.

At a later stage in the development programme, the multi hole extrusion system was placed back on to the extruder and a pelletiser added. This was basically a spinning cutter blade which chopped the multi strand extruded product into small capsules.

The arrangement of the development equipment was as indicated in Figure 1.

It was decided to heat and extrude a range of ‘plastic’ type materials, first testing individual materials, followed by a range of mixed products, the mixture ratios having been determined by discussion with a typical nuclear site. The main objective of the test programme was to determine the temperature of melting of each combination of material and the time constraints, if any, at that temperature.

The first and probably the most important test was with pvc (polyvinylchloride), mainly due to the chemical make up of the material and the high proportion of this material on a typical nuclear site. The chemical formulation of pvc is such that there can be a risk of the material degrading if the time / temperature limits are exceeded, resulting in the emission of hydrochloric acid gases, which are carcinogenic and highly corrosive. If the melting temperatures and the safe residence time in the extruder could be established for pvc, it was thought perhaps easier to establish similar factors for pvc mixed with other products, because of the dilution factor and a diminution of the quantity of pvc and the possible off-gassing effect.

It is important to state here that the dangers of extruding pvc as a semi-molten product can be overstated. Pvc, when heated, is time/temperature dependent and those factors are well known for this material, when new and are indicated in Figure 1. However, as it was intended to subject the materials under test to what amounted to a recycling process, and the exact composition of the mixture was unknown, it was decided to employ a ventilation system over the extruder outlet in case any degradation did occur.

The initial tests involving extruding new pvc material into a small (½ litre) catchpot were highly successful with the equipment operating within the residence time and temperature recommended for new pvc. The extruded material poured into the catcthpots and over a short period of time, the material slumped to fill the complete pot. The sectioned ½ litre samples showed a remarkably solid material with no blemishes indicating how the extrusion of new pvc produced the high density product aimed for.

Similarly, catchpots of 1 litre were apparently successfully filled without difficulty. However, Figure 2 shows views of the exterior and a cross section of the 1 litre sample, indicating a small degree of degradation in the form of small gas bubbles at the centre of the sample. On closer inspection, these 1 litre samples appeared to have little evidence of degradation near the external surface of the catchpot, where cooling was most rapid.

The next phase of the development programme was to investigate the possibility of extruding the molten mixed product containing primarily pvc but with small proportions of hyperlon, neoprene, rubber (leaded and unleaded), polyethelene, cellulosics, etc. Each of these tests proved that small volumes (half litre) of the mixed products could be successfully extruded, although in some cases the added materials (e.g. rubber) had not changed their form and were, in effect being encapsulated by the molten pvc. However, when attempting to fill 1 litre catchpots with the extruded, mixed product, an increase in the degradation of the pvc took place, presumably because the other materials in the mix were preventing the sample cooling sufficiently quickly, this being the only difference between these and the original all pvc tests. Figure 3 shows how the mixture of materials has provided a product which is of rougher texture on the exterior surface, while the cross section indicates a degree of degradation at the centre of the block. The metal scrap material was placed in the catchpot to act as a conductor in an attempt to improve the heat transfer characteristics of the material.

A further phase of the development work was to undertake a half scale investigation into the possibility of filling a 200 litre drum with molten pvc, followed by the mixtures already tested in smaller volumes. The result justified the provisioning of the ventilation systems as the pvc degraded very rapidly in the drum, producing copious quantities of hydrochloric acid gas. As in the case of the 1 litre samples, the mixed plastics samples also degraded, but in this case, even more rapidly than the new and pure pvc sample.

At this phase in the test programme a change of pvc was necessary, due to the fact that the original supply was replaced with another site’s specified material. Immediately it was noted that every sample produced, even at the previously successful half litre size, resulted in severe degradation behaviour, made even worse when the pvc was mixed with other materials. A number of repeat experiments were undertaken to investigate the possibility of extracting the heat from the samples, thus effectively reducing the residence time at the melt temperature. Graphite powder and lead scrap were employed as heat conductors, with some limited success. (See Figure 3) Finally, the temperature of the extruder was reduced to eliminate the degradation, but the extruded product did not possess the flexibility to slump and successfully fill a catchpot with a minimum of trapped air.

Although some of the small tests were acceptable, the ultimate goal of filling a 200 litre drum full of extruded plastic had failed, due to the resultant degradation of the pvc, a factor which was worse with some types of pvc and when some of the other materials were in the mix. At this stage in the programme a change of direction for the research was decided. A decision was made to employ the multi-hole die that was supplied with the extruder and determine if better heat transfer was obtained with a spaghetti type product. It was quickly established that the slumped product in the drum was still subject to degradation and for identical reasons. It was decided that further investigation into multi-strand extrusion was required with the pvc manufacturers. It was soon discovered that process engineers have overcome the degradation problem by chopping the multi-strand output from the extruder into short pellets, which, due to the reduced volume of each pellet, cool and harden while they are falling into the catchpot. The pellets produced also maintain a regular shape. This pelletising operation was investigated and proved to overcome the degradation problem of pvc and other mixed products, producing pellets as shown in Figure 4.

There is no doubt that the fact that both pure pvc and pvc mixed with other ‘plastic’ type materials degraded rapidly when the size of the catchpot exceeded ½ litre, mainly due to the mass and the poor thermal conductivity of the product. Although it was thought that the other materials in the mix might improve the situation, they appeared to hamper the thermal transfer characteristics rather than helping. Another factor encouraging degradation is the use of pvc which has already been recycled and as a result, has lost some of the important stabilising chemicals. The objective of trying to fill a 200 litre drum with a solid product, therefore, proved disappointing and totally impractical. It is also impractical to try and insist that site operators only purchase new pvc as that could result in a more costly consumable product. In any event, the development work showed that using new pvc would only allow the production of 1 litre samples.

There was some benefit on the thermal conductivity by adding both graphite powder, lead scrap and aluminium tubes, but it could be foreseen that these would prove difficult to add within a nuclear environment, and in any case added to the waste volume, reducing the minimisation effect being sought. Although serious consideration was given to limiting the product to ½ litre size and filling drums with them, the extra handling required was a disadvantage, as was the large volumes of interstitial air trapped between the blocks. The fact that all the operations were confined to a radiological containment would also mean that the extra handling would result in additional radiation dose uptake.

The adopted solution of pelletising a multi strand extrusion product works exceedingly well, avoiding the degradation of the previous tests, as is clearly indicated in Figure 4, although more detailed work will be required to determine the exact size of the holes in the extruder die plate to accommodate the variety of mixed products likely to be found at other sites. Pelletising does result in a volume reduction over the untreated plastic although the degree of reduction is not as significant as was hoped. An indication of the reductions in volumes obtained is shown in Table 1 below.

Although the reported development work did not achieve its original goal of filling a 200 litre drum with a solid, air free plastic material, it did yield a pelletised product which was in a significantly reduced volume form. Not only could the product be disposed of directly in its pelletised form, showing reductions in disposal costs, but could also be used as:

• A high quality feed stock for incinerators with a constant air to plastic ratio allowing easier control of the burning process with resultant lower maintenance of the system. The mechanical feeding of the pelletised product into the incinerator should be easier than current practice
• Better packing factors when used with high force compactors making it easier to mix the pelletised product with hard, solid waste, avoiding the natural springback of plastic sheeting.
• Pellets will be easier to mix with an encapsulant, like cement, providing a higher strength matrix and no hold up of sheet plastic on the mixing paddle

In addition to the above benefits, the development programme always considered that if this process was ever to be adopted, the equipment employed should be able to be safely operated and maintained. Reference to Figure 1 will indicate that all of the process could be radiologically contained, from the point where the plastic sheeting would be fed into the shredder top door in a bagged form, through to the extruded, pelletised product being allowed to fall under gravity into a sealed drum.

The author wishes to acknowledge the support provided for this project by the Department of the Environment, London and Golder Associates (UK) Ltd.

1. J.A. BRYDON, Plastics Materials, published by Newnes & Butterworth
2. Degradation of Polymers - Comprehensive Chemical Kinetics - Vol 14
3. C.J. LLOYD, Portsmouth Polytechnic The Management of Polymer Contaminated Wastes Arising from Nuclear Fuel Cycle
4. A.J. PERKINS, Investigation of the Density of Extruded Plastic Waste
5. B. HARPER, Private communication
6. UK Patent Application 2 122 803 A (Sims and Hitchcock)
7. PAUL STUBBS, EVC Compounds Ltd., Chester, England. Private communication

BACK