Member Control Panel
Increasing The Use Of Recyclate in End Products Without Reduction in Quality
| By Kerry Bodin, Triton Technology Ltd Recycling environments can mean several things to different organisations; If we first consider a production line where articles are being extruded or injection moulded there will always be a certain amount of waste material. If the original material used is a virgin grade of known quality, the assumption is that this waste will be the same and can be used with little or no risk in the same process or bagged off and sold on as a known polymer. This may or may not be true. Contamination can occur if more than one type of polymer is present in the factory and polymers are also thermally labile and may or may not have degradation that could affect the final properties of the articles being produced. Consider also a second situation in a recycling company - many tons of mixed polymers will be present. Mechanical sorting using a variety of techniques can sort out a considerable portion of the material. Waste can be from domestic or industrial sources e.g. bulk waste from polymer converters which should have some known providence. Unfortunately, the quality in reality can be highly suspect unless substantial comprehensive testing is undertaken. Even the most sophisticated separation technologies will always fail to completely isolate all the different polymers. There are also many grades of each polymer type and these certainly are not easily sorted.  Generally before using and after sorting, all these materials are subjected to regrind and sold on with an indication of type and quality. The most common indication used, though not necessarily the only one is the melt flow index (MFI) value. This may or may not be adequate. In essence, a single point in the rheological profile of the polymer in question is used to indicate its quality. This is not a characterisation and will not profile the overall properties of the regrind material. Further more, this cannot provide a safe comparison against a virgin material. As mentioned earlier, a more comprehensive testing regime is required to characterise these materials sensibly and provide an assessment for use. Traditionally, these tests would in most cases be uneconomic and unsustainable for any operation in the recycling chain. The common approach for providing a more comprehensive characterisation of polymers uses thermal analysis techniques. Two in particular are very good at providing comprehensive thermo mechanical profiles of polymers. These are Dynamic Mechanical Analysis (DMA) and Differential Scanning Calorimetry (DSC). Unfortunately, these are expensive, time consuming and require careful evaluation of the results by a skilled scientist. Additionally, no statistical assessment of product quality is achieved using these instruments. A new product, the identiPol QA, produced by Triton Technology Limited, strives to solve many of these issues outlined above. In essence, a device has been produced that can profile the thermo mechanical properties as found on a DMA and DSC in one small and at a substantially lower package cost. Typically the unit is set up to take small samples of the material. A test sample is prepared in the unit and run rapidly, especially in comparison to the traditional equipment. This only takes 10 to 15 minutes in total, including the cool cycle, preparing the unit ready to run another test.  The first stage of evaluating the quality of an unknown material is to teach the unit what the thermo mechanical profiles typically look like. The user does not normally see these or indeed need to, but they can be accessed if required via a management programme. The user will typically need to ‘feed’ the unit with at least 10 to 20 samples of the material being assessed before a reasonable Quality Index Score (QIS) can be produced. The user should understand that the all polymeric material varies to some extent therefore the more material run the better the evaluation. After the learning set of a minimum number of samples have been run, the device starts to output QIS scores. The learning set produces an envelope of acceptable values for subsequent tests. If these fall in the envelope, they would normally be added until a substantial data set is reached. In order to reduce the risk of missing a low level drift in the material, a maximum number of samples in the training set is fixed as well as a minimum. This is typically 50. If an unknown material falls below the minimum envelope threshold, further samples must always be run to check if the material is definitely different or not. If low results continue to be produced, the thermo mechanical profiles can be examined by the manager or the data can be forwarded back to Triton for comment. 
|
