Skip to main content

Magnetic Sorting and Ultrasound Sensor Technologies for Production of High Purity Secondary Polyolefins from Waste

Final Report Summary - W2PLASTICS (Magnetic Sorting and Ultrasound Sensor Technologies for Production of High Purity Secondary Polyolefins from Waste)

Executive Summary:
The European annual consumption of plastic materials has increased from 24.6 M tons in 1993 to about 50 M tons in 2008 (nearly half being polyolefins) and it is likely to keep growing, so that Europe is faced with the daunting challenge of managing millions of tons of waste plastics. At the same time, polymer recyclers and manufacturing industries have a problem in buying feed materials and secondary polymers of sufficient volume and consistent quality, due partly to the strong pull of China and India on raw material resources. In addition, considering that about two kg oil is needed to produce one kg plastic it is clear that the production and use of plastics has a range of environmental impacts. For the reason that only 2% of the economic value of the € 100 billion plastics market turnover in the EU comes from recycling, also relevant political-economic potential impacts are associated with the use of recycled plastics for the production of high-grade products (reduction of raw materials import and green jobs creation).
Today, Polyolefin recyclers focus mainly on relatively pure post-industrial or other mono-plastic type wastes, since these wastes can be easily turned into high-purity product materials by existing and cost-effective process technology. However post-industrial wastes are increasingly exported outside the EU, and so the Polyolefin recycling industry is forced to look for alternative resources. In principle, post-consumer wastes, such as, household waste, construction and demolition waste, WEEE and ASR provide such resource. They are a much larger potential reservoir of polyolefin’s than post-industrial wastes, but these wastes are also much more complex mixtures of materials and hence much more difficult and expensive to recycle in an effective way. For this reason, only a small fraction of polyolefins yearly sold in Europe is currently recycled into products where the recyclate replaces virgin resins. As a result, most of the polymer resources in complex wastes are largely unused. For example in 2011, out of the 18 million tons of polymer used for packaging, only about 5 million tons (7 billion euro virgin polymer value) were recycled into less than 2 billion euro of recycle polymer product. The main reason for this huge gap is the low efficiency of the current sorting and separation technologies for the secondary plastics, which recover little value from the polymer, and in consequence create a low incentive to recycle.
Technologies that are to address post-consumer wastes need to be extremely powerful, since they must be relatively simple to be cost-effective, but also accurate enough to create high-purity products and able to valorize a substantial fraction of the materials that are present in the waste into useful products of consistent quality in order to be economical. On the other hand, the potential market for such technologies is large.
Currently, in many EU countries post-consumer plastic packaging waste is sorted in sorting centers into mono plastic streams, such as HDPE, PP and PET, using NIR optical sorting technology. However, if the input material contains substantial amounts of non-bottle packaging (e.g. film and trays), such as in Germany and The Netherlands, only a minority of the input can be sorted into mono-plastic streams that is suited for high-grade applications. The larger part of the input is downcycled or even incinerated, which makes sorting according to the current practice a costly step. For this reason, only easy-to-sort packaging objects, such as bottles and jars, are collected in most European countries.
Therefore it is unlikely that the current plastics recycling rate will increase simply because the overall economics is not attractive for all stakeholders. The process will become economically attractive and self-sustained only when a flexible and robust sorting technology is available to exploit the large sources of complex plastics waste.
W2Plastics developed a Magnetic Density Separation (MDS) technology to recover high-purity polyolefin’s from complex wastes at low cost. The most important feature of this new concept (which includes new quality-control systems based on ultrasounds based and hyperspectral) derives from its ability to accurately separate many different materials in a single process step, using an environmentally friendly and cheap process fluid. Because of its intrinsic flexibility, this technology can easily be implemented to separate any kind of plastic materials in an economically sound way.
All of the following objectives were achieved within W2Plastics:
1. All valuable recyclable materials as well as all important contaminants relating to the economic value and ecologic impact of complex wastes have been identified. The quantification of limits and the setting of standards for waste generating facilities as well as their relation to the relevant properties of actual wastes to the acceptance strategies, logistics and process technology have been achieved;
2. Models necessary to develop the MDS technology and to foresee separation processes have been realized. The models which take into account the fluid dynamics, material properties and the chemical engineering are in good agreement with the experimental results;
3. A sensing technology based on hyperspectral imaging that is able to characterize feed and product streams in terms of the relevant composition parameters has been implemented.
4. The ultrasound principles have been studied and experimentally verified. The Ultrasound system has been tested in situ in semi-industrial setting. It has proven to be capable of providing vital process information for the quality of the MDS polyolefins.
5. A robust MDS pilot plant able to recover more than 90% of high-grade polymers from complex wastes has been built and tested in an industrial environment. The MDS facility has shown the capability of producing high-grade plastics streams to be used for prime applications. Because of its flexibility and competitive investment and operating costs, the MDS technology will contribute to reduce the environmental impact of plastics recycling industry, to minimize depletion of resources and to promote at the same time business opportunities and improved competitiveness of European industry and SMEs.
Project Context and Objectives:
The European consumption of plastics increased from 25Mtons in 1993 to about 50Mtons in 2008 and its growth rate exceeds that of the economy as a whole. At the same time, polymer recyclers and manufacturing industries have a problem buying feed materials and secondary polymers of sufficient volume and quality, partly as a result of the pull of China and India on all raw material resources. The alternative of using more primary plastics has a range of environmental impacts and needs more resources (about two kg oil for one kg plastic). The polymer resources in complex wastes, such as Construction and Demolition Waste (CDW), household waste, automotive shredder residues (ASR) and waste from electric and electronic equipments (WEEE) are largely unused, because of the problem to produce high-purity products from such sources at acceptable costs. W2Plastics aimed at developing a cost-effective and clean technology based on Magnetic Density Separation (MDS) and Ultrasound process control and Hyperspectral Imaging quality control, to recover high-purity polyolefin’s from complex wastes. The ambition of W2Plastics was to bring about a fundamental change: to transform materials that are largely land-filled or incinerated to resources of high-value secondary polyolefin’s (PO). A substantial effort is spent on making the new technologies fit in between the state-of-the-art technology of waste processors and the demands of the compounding and manufacturing industries by defining standards and best practices as well as effective quality-control tools (hyperspectral imaging). The integrated set of technologies and standards aimed at changing the status of complex wastes to a resource of high-purity polyolefin’s for a wide range of industries. The development of such technology is in line with the European legislation (COM/2001/0031, 99/31/EC, 2000/53/EC, 2002/96/EC, 2003/108/EC) aiming at fostering the development of environmental friendly technologies to reduce the environmental impact of human activities, to protect the environment, to minimize depletion of resources and to promote (at the same time) business opportunities and improved competitiveness of European industry and SMEs.
Project Results:
In order to reach the goals of the project, the scientific and technological objective of W2Plastics project pursued the following three main routes: Route 1. Market, Ecological & LCA analysis. Route 2. MDS technology development. Route 3. Sensors for process & quality control.
Concerning route 1 several activities towards the characterization of different sources of polyolefins wastes have been carried out (WP1). The study of packaging waste recycling infrastructure in Europe, the study of official data (sorting practice, logistics, etc) and the field study (interpreting official data) have been done.
Market analysis test and applications of secondary plastics including questionnaires have been done in collaboration with industrial partners (WP2).
Preliminary LCA and life cycle social analyses have been performed (WP3). Parallel to the above activities, investigations aiming at improving the chemical and physical quality of the recycled polyolefins have been implemented (WP5). Investigations on the most suitable technology to deal with process residues have been performed (WP4)
Herewith relevant outcomes achieved so far are mentioned:
(a) One outcome is the conclusion that WEEE as source of polyolefins is not as interesting as it is currently believed. On the contrary it has been found that an important and attractive source of polyolefins is CDW. The separation limit between PE and PP in CDW was well determined: the fractions above 0.92 g/cm3 contain mainly PP while PE appears in the range of 0.93-0.96 g/cm3.
(b) Another important result of route1 is the discovery that in order to reuse the secondary PP and PE as raw materials for high quality applications, it is not strictly necessary nor always sufficient to reach high grades (97% or higher). Investigations show that the compatibility of PP and PE mixtures strongly depends from their Melt Flow Index (MFI). Therefore it is important to distinguish not just PP, LDPE, HDPE, but also blow molded PE and injection molded PE. Rheological characterization on thermoplastic melts have been carried out and relevant information about polymer processing have been achieved. As expected, blow molded materials were found to be more viscous than injection molded ones. Tests were performed to study the rheological behavior of the materials applied for blow molded and injection molded polyolefins especially at the melting temperature. It has been found out that the difference of blow molded and injection molded in wall thickness allows some sorting by flake thickness, for instance using ballistic separation. Combining these discoveries with the current methodologies which can change the polymer viscosity, high quality secondary polyolefin products can be achieved.
Constant MFI of the waste fraction to be recycled was also one of the most important criteria named by the plastic industry companies in the market analysis on the use of secondary plastics carried out within the project. The melt flow index reflects the viscosity of a molten polymeric material. It is an important parameter in the manufacturing of plastic products, during which the ultimate shape of a given plastic product is obtained by moulding. Depending on the desired shape of the product, different moulding techniques can be used to obtain optimum results, such as injection and blow moulding or sheet extrusion. The required melt flow index for different moulding techniques can differ as much as two orders of magnitude. Although it has little to no influence on the physical properties of the finished plastic product, the right melt flow index is a main requirement for the moulding process, as consulting several converters using recycled compounds has revealed. Therefore, if polyolefins are to be recycled on a large scale, the issue of melt flow index should be considered already in the collection and mechanical processing phase of recycling.
The melt flow index of a given polymer can be altered during compounding in some cases. It is, for example, known that the melt flow index of PP can be increased using additives during compounding. This means that it is relatively easy to convert a blow mould quality PP into injection mould quality. Unfortunately, this method does not work the other way around. Moreover, opportunities to alter the melt flow index of (HD)PE are very limited compared to PP, according to converters. Therefore, incorporating the melt flow index into the set of quality criteria when designing a strategy for the mechanical processing of mixed polyolefin waste streams can greatly improve the marketability of the recycled compounds.
As mainly the desired shape and thickness of the plastic product dictate the conversion technique to be used, there is an opportunity to effectively sort polyolefins into groups of high and low melt flow index at some point in the mechanical processing chain of the polyolefin waste.
Because of the importance of the above results, further study to investigate the maximum acceptable levels of cross contamination in polyolefins, especially PP and HDPE and vice versa, has been carried out during the last reporting period (fourteen months). The goal of the study is to generate thorough knowledge on the blending properties of PP and HDPE, which allows an optimization of the sorting strategy using the MDS process by defining the required separation sharpness in density separation more precisely than currently possible.
Literature data indicate that the main problems in cross contamination of PP with HDPE and HDPE with PP are de-lamination and the deterioration of the strain at break. These observed effects indicate poor mixing of the PP and HDPE in the molten phase during extrusion.
First, from a study of the relationship between the spatial structure of the molecules of different types of polyolefins, their material densities on a macroscopic level and the degree of incompatibility in compounding, a hypothesis was formulated, which states that the degree of incompatibility between two different types of polyolefins reduces with diminishing differences between their densities. A comprehensive literature study carried in this reporting period to test this hypothesis has revealed no strong indications that would either verify or falsify the above hypothesis. In addition, no study has been found yet with an explicit focus on the differentiation between the degree of incompatibility of PP and different (HD)PE types in conjunction with their densities. However, evidence supporting our hypothesis was found in the fact that commercially available compatibilizers for blending PP with HDPE often contain certain LLDPE types having an intermediate density as a main ingredient. Further, compounders using recycled polyolefin regrind reported that a moderate amount of LDPE usually has a positive influence on the mechanical properties of PP-HDPE mixtures.
(c) Concerning the contaminants reduction efficiencies (route 1, WP1), the study of existing separation techniques revealed that the contaminant removal efficiency of mechanical separation devices may strongly vary when processing plastic waste from different sources and different compositions. This variation was found to lie in the single or combined effect of different parameters that for practical purposes can be divided into operational and input material related parameters. Examples of the first category are the actual feed rate of the applied separation device compared to its design capacity and the underlying separation principle utilized in the device together with the level of technological stage incorporated in the device. In general, increasing the feed rate of a given separation device beyond its design capacity progressively reduces its contaminant removal efficiency. Examples of parameters related to the input material are the property distributions (e.g. density, particle size and shape) of both contaminant and valuable component. This implies that differences in, e.g. particle size or shape of otherwise equal plastic materials containing the same type of contaminant may lead to significantly different separation results using the same separation device.
As the majority of reclaimers currently processing post-consumer plastic packaging waste to be converted into high-quality resins applies wet cleaning methods and due to the fact that post-consumer plastic packaging waste is usually relatively strongly contaminated with different types of organic and inorganic matter, water treatment issues become important from both a resin quality and a processing cost point of view. Our analysis of the current practice of cleaning post-consumer packaging waste indicates a technological vacuum in this field. Although wet processing is an efficient cleaning method, the contaminants are transferred to the water and therefore, they have to be continuously removed from the water prior to be re-circulated to the process. Unfortunately, the removal of contaminants originating from packaging waste from the process water is complex and requires several different processing steps, as there are many different types of contaminants that require different treatment steps to be removed efficiently.
On the other hand, our analysis revealed substantial advantages in applying dry-cleaning methods, which are currently offered by only a few suppliers, especially for post-consumer plastic packaging waste. The great advantage of dry-cleaning methods is that the different types of contaminants do not require distinct processing steps, but can be treated and disposed of together. As the contaminant removal efficiency of dry-cleaning methods is lower, they cannot usually replace wet-cleaning techniques, but should be used complementary. Although there are currently not sufficient supporting data, it appears that a proper combination of dry- and wet cleaning methods provides both cleaner plastic flakes and an overall cost saving in the cleaning of post-consumer plastic packaging waste.
(d) An important outcome of route 1 has been the calculation of the total amount of plastic packaging waste actually generated by households or companies on an annual basis for the Dutch case. This work has been carried out in order to verify official figures that are based on market-entry data (i.e. the amount of plastic packaging material put on the market by companies). For an independent verification, the calculation of the annual consumption of plastic packaging materials from end-of-life data has been made. It was found that market-entry-based data indicate 37 % lower amounts than our calculations, 454.000 ton/a versus 621.000 ton/a, respectively.
The study of the current sorting and processing infrastructure of post-consumer plastic packaging waste was done in detail in relation to the potential influence of a large-scale implementation of the MDS technology on this infrastructure and on the purity of the reclaimed PP and HDPE flakes. From analyses of the composition of reclaimed flakes derived from DKR-compliant HDPE and PP mono streams it was concluded that the actual cross contamination level of such reclaimed flakes lies in the range that can be achieved by flake sorting using the MDS method. This means that a large-scale implementation of the MDS technology at reclaimers would allow a substantial improvement of the current sorting process.
This work provided reliable first-hand information on market-wide accepted quality criteria regarding the cross contamination between PP and HDPE. These are important data for determining the required separation sharpness for the MDS process.
(e) Another results from route1 (WP2) comes from the market analysis. As general summary on the market analysis, it could be stated that usually the plastic supplier companies only reuse their own waste, only a few companies accept waste from external sources. Currently the ratio of the applied secondary plastics in the production of the primary products varies between 5 and 25%. The ratio of recycled plastics, which can be added to the product, depends from the technology, currently it can be up to 50%. Recycled plastic materials can be used as secondary raw material mainly in non-visible, non-coloured, non load-bearing parts, when no further processing is done, in non-prestige brands. Constant composition, hardness, elasticity, UV-stability, (long)-term heat resistance, recyclability were the most important parameters, which should be considered in order to increase the ratio of the secondary plastics.
(f) To define the application of secondary PO (route 1, WP2) market study was done about the products prepared from recycled polymers already available on market in Western Europe, Hungary and Romania. As for the products prepared from recycled plastics in Western Europe, all products are produced from Post Industrial Waste (PIW), the percentage recycled- vs. virgin-feedstock used in these products is always above 20%, some of the products are already produced of 100% recycled material. As for the products prepared from recycled plastics in Hungary, various products already available on the market could be found, mainly in the field of construction, traffic control signs, horticulture, agriculture, household, packaging. Both mixed plastic waste and separated polyolefins, PE, PP, PVC are reused, mainly by extrusion or injection moulding. Generally, mixed plastic waste is rather used for low-profile products, while products made from separated polymer waste have more added value. As for Romania, there are companies, which collect, transform plastic waste into granules and then they use themselves the granules in their final products or resale the granules and also companies, which only sell these granules. Generally, it can be concluded that in order to enter the market of higher value-added recycled products more efficient and economic solutions for the separation of polymeric waste will be needed, especially if post consumer wastes will be also used in addition to the hardly contaminated post industrial waste. Besides the need for more sufficient separation, the upgrading of plastic waste developed in WP5 is also of crucial importance to introduce recycled products in the industries with higher quality expectations and this way reuse the wastes in the same area, where they occurred, and not only in low-profile applications.
FTIR and Raman spectrometry proved to be a suitable tool for the characterization of the polyolefinic waste mixtures as FTIR and Raman spectra can give information on the nature of major and minor compounds and on their ratio in the mixture. Also they allowed determination of the polymer’s oxidizing degree expressed by Carbonyl Index (CI). DSC gave information not only about the purity of the fractions, but also about their probable thermal behaviour during processing by determining their melting and degradation temperatures. MFI gives information about the processing technologies, which can be applied in case of the characterized density fraction. Although it provides a simple and quick method to determine the flow properties, it has to be noted, that the purity of the waste or waste density fractions greatly influences the proper implementation of the MFI test.
As general conclusion on the results, it can be affirmed that the separation limits determined on the basis on the FTIR and Raman spectrometry (reported in deliverable 2.4) seemed to be in most cases correct. Comparing the different types of polymer wastes selected to density fractions, CDW (PC2) seemed to be the clearest and the most appropriate source for further applications. Clean fractions could be also separated from automotive waste (ALCUFER2) as well, according to the results, the <0.92g/cm3 density fraction polypropylene fraction can be processed under same conditions as virgin polypropylene at the shear rate of extrusion and injection moulding, which is promising for further applications.
According to the results of the static, dynamic and fatigue tests (route 1, WP2) it has been found that the mechanical properties greatly depend on the quality of the separation process. By choosing appropriate separation limits, in most cases significantly better mechanical properties can be achieved in comparison to the waste without separation. Besides testing the density fractions, mechanical properties of a product developed in the frame of WP2, a car door opener were also tested both by static and dynamic tests and its fatigue properties were determined as well. Most properties of car door openers made from recycled polymer upgraded with glass fibers were quite close to the ABS based products (this primary material is generally used for these purposes), in case of fatigue test even the recycled polymer had much better properties than ABS.

(g) LCA results (WP3) have shown that the recycling of polyolefins is almost always beneficial compared to other waste management options.
For the environmental LCA, considering the maturity (availability) of data, the study should only serve as a first LCA iteration of the functioning pilot MDS line and the results can only give an indication of the potential environmental impacts. Data of BAT waste plastic recycling system for five countries and three types of waste were gathered. Even though these data are not complete the LCA could be performed. Several assumptions had to be made while building the models. These assumptions have been justified and presented in the list of assumptions Nonetheless, it is concluded that for ecosystems and human health, MDS has a net environmental benefit since the avoided production of energy and virgin material counterbalance the burdens of the processing. On the contrary both incineration and landfill cause a net environmental burden. In terms of resources both systems have net environmental benefits: the benefits from incineration are due to the avoided materials used otherwise for electricity production of the grid. The benefits of MDS are also due to the increased recycling and thus the avoided production of virgin plastic material. The superiority of the MDS against its alternatives depends on the waste quantity processed by the MDS and the proportion of PE and PP in it. There is a trade-off between a) the environmental benefit due to material recovery and b) the environmental burden due to extra processing and transport: except for PE and PP, the rest of the plastic fractions (PP, PVC, and PS) cannot be separated by the MDS unit. In this respect it is important to ensure that the waste composition includes enough PP and PE so that the benefits from recycling will counterbalance the burdens from processing. As an overall conclusion, given the assumption of best available technologies, and the assumption that incineration is used in all European countries for the production of energy to the grid, MDS is evaluated as a beneficial technology if the economic value of recycled PE/PP through MDS is more than 50-60% of the value of virgin PE/PP, the MDS efficiency is >70%. These figures assume further that the MDS unit is placed as an additional processing step into the existing recycling chain without substantially changing the current recycling infrastructure, i.e. the way of collection, sorting and processing to clean flakes. However, applying the MDS technology offers a great opportunity to profoundly improve the efficiency of the whole recycling chain of, e.g. post-consumer plastic packaging waste. Such a change leverages the benefit of the MDS technology to a multitude when compared to the marginal benefit of placing the MDS as an additional processing step into an existing infrastructure. In that case the above indicated price and efficiency conditions become even milder.
Due to the sensitivity to the allocation principle, the study is inconclusive regarding the potential benefits in the area of protection ‘Resources’.
The methodology for life cycle social analysis went through all indicators currently proposed by e.g. a UNEP-SETAC working group on the subject. On this basis it is suggested that the social consequences considered in the assessment are limited to be the physical working environment for the workers, the creation of jobs and the educational profile of these, and the local community assessment of each system. Performing the social LCA with these indicators, the results are a bit more inconclusive. It shows that the assumption about whether mixed PO from existing waste treatment facilities can be assumed to substitute virgin PE and PP or whether it due to its lower quality only can substitute wood or coal has large implications for the results. With regards to the physical working environment, this production of virgin PE and PP resulted in significantly higher incidences of working accidents and diseases. Being more labour intensive this production of virgin PE and PP also resulted in creation of more jobs than the alternative including the MDS, most notably when it came to the high educational jobs. If, on the other hand, it was assumed that the mixed PO streams could substitute virgin PE and PP very limited differences between the alternatives could be identified. The inclusion of the MDS would in these alternatives tend to be slightly more labour intensive, and as a direct effect of this also result in slightly more accidents and diseases. Somewhat different was it when it came to the expected local community acceptance. Here the production of virgin PE and PP seemed to be of minor importance. Rather what was important was whether the inclusion of the MDS resulted in additional waste handling facilities being made and the nuisance level of these to the local community, and whether the wood or coal, which the mixed PO stream was assumed to substitute in some alternatives, were produced close to dwellings to whom it most likely would create significant nuisances. It was, however, estimated that in most cases, these nuisances will probably not realise. If this is the case, very limited differences could be found in relation to the local community acceptance of the assessed alternatives.
Concerning the economic feasibility of an industrial MDS plant the estimates on costs shows good revenues depending on placement of the sorting plant and on which type of polyolefin is sorted.
So whereas the social LCA has inconclusive results, showing insignificant changes whether the MDS is used or not, both the environmental and the economic evaluation shows overall beneficial effects of a future use of MDS.

(h) The first step in finding Green solutions for process residues (route1, WP4) was the analyzing of the output residues after MDS of inputs from: MSW from Romania (PP, PE in the light fraction; PP, PE, PS, PET, PVC, PA and PC in the heavy fraction), ASR from Austria (Cellulose, PU foam, STPe, epoxy adhesive, PP in the light fraction; PP, PE, PET, PS, PA, PVC, in the heavy fraction) and CDW from France (Polybuthene, LLDPE, ethylene/propylene, polyacryl amide in the light fraction; PPi, PE, Polybutadiene in the heavy fraction).
As far as the solutions for process residues is concerned, heat combustion measurements have been carried out using residues coming from MDS separation. All waste streams taken into account, shown high calorific values (MSW between 24213 J/g and 45444 J/g. ASR between 32721 J/g and 36591 J/g, CDW between 38010 J/g and 40526 J/g). Considering the complex composition of such residues, energy recovery seems a suitable solution to avoid landfilling.
There were performed also the pyrolysis researches on the MDS process residues. Having a high content of polyolefins, the MSW is more effective for oil production by pyrolysis by comparing to ASR wastes, the most effective being the light fraction (70.6%). Light fraction of ASR led to higher residue (22.1%). The same results were obtained for heavy fraction of MSW (25.4%). The ash can be used as construction filling material.
Composites made from contaminant fraction of PO-s (heavy fraction) and Chitin/cellulose antimicrobial fibers have been prepared. A good fibers inclusion has been noted. These types of composites could be used in building and construction as panels, carving, interior roof, plated ceiling, ornaments, outdoor garden recipients or pipes, due to their low density, capacity to absorb a low amount of water that will avoid water condense, good chemical resistance, due to each component resistance.

(i) Concerning the functionalization of the polymers, chemical modification of waste polypropylene fraction with reactive agents was performed in batch procedure. It could be concluded, that the efficiency of intumescent fire retardants could be enhanced by interlayers that deliver the active components to the surface. By applying the concept of reactive modification the amount of necessary flame retardants could be successfully lowered. The impact of stabilizers was tested in different concentrations on reference polypropylene, polyethylene and also on plastic wastes containing mainly polypropylene and polyethylene. According to the results, it could be stated that the recycled samples need significantly more stabilizers than the reference virgin materials (instead of 0.1-0.2 mass% the amount of the required stabilizers is in the range of 0.8-1.0 mass% to reach the same level of stability). This may be explained not only by the higher oxidation degree of the waste polymers, but also on high initial filler content of the waste sample, which usually acts as catalyst of degradation. The experiments have proved that the chosen, relatively high amounts of stabilizer has successfully protected the material against degradation, the mechanical properties of the stabilized samples were unambiguously higher than of those without stabilization.
Chopped glass fiber was used as reinforcing component in recycled polypropylene waste separated from car shredder (ASR), with density below 0.9 g/cm3, and intumescent flame retardant system served for the reduction of flammability. Layered sandwich structure was used in order to eliminate both the “candlewick effect” caused by glass fibres and deterioration of mechanical properties caused by flame retardant additives. The application of the recycled shredder automotive waste instead of virgin materials did not cause considerable deterioration in terms of the fire retardancy of the flame retarded shell and the sandwich composite. The mechanical properties of the flame retarded recycled layered composite reached or exceeded the properties of the pristine polypropylene matrix, therefore the composites containing recycled materials are still proper for certain engineering applications.
The mechanical properties of automotive shredder plastic waste have been successfully improved by applying reinforcing polypropylene fabrics, five times higher tensile strength, two times higher flexural strength, while reduced flammability has been obtained by adding 18% flame retardant additive to the recycled matrix layers of the self-reinforced composites.
In case of glass fabric reinforced composites advantageous compatibilizing effect of the applied FR additive was observed, which effectively improved the adhesion between the glass fibers and the recycled matrix layers, resulting in significantly higher tensile strength but relatively lower perforation energy as a consequence. However, from environmental aspects the prepared self-reinforced composites are given preferences, because considering that, in opposition to glass fiber reinforced composites, they are homocomposites, they are easy to recycle with simple physical methods at the end of their life-cycle.

(l) In order to develop competitive products from recycled plastics which can compete with the primary ones on the market, along with the determination of exact density separation limits and the throughout characterization of the density fractions is necessary to improve their chemical and physical properties.
The work reported in the frame of task 5.4 aimed at upgrading automotive shredder plastic waste, by development of flame retarded self-reinforced composites from density separated secondary polyolefin fraction. The mechanical- and flammability properties of the prepared self-reinforced composites were compared to conventional glass fabric reinforced composites and to the recycled polymers.
The mechanical properties of the polyolefin waste have been successfully improved by applying reinforcing polypropylene fabrics: 5 times higher tensile strength, 2 times higher flexural strength and 4 times higher perforation energy were achieved. Although the tensile and flexural strength of the recycled matrix could be increased more effectively using glass fabric reinforcement, the prepared self-reinforced composites are given preferences from environmental aspects, considering that, their density is much lower and in opposition to glass fibre reinforced composites, they are homocomposites, so they are easy to recycle by simple reprocessing at the end of their life-cycle.
Significantly reduced flammability were obtained by applying intumescent flame retardant additives and the prominent mechanical properties of the recycled multilayer composites have not been remarkably influenced by the FR content of their matrix layers. Special beneficial effect on the effectiveness of the applied phosphorus-based intumescent flame retardant additive was found in the self-reinforced system resulting in flame-extinguishing behavior during horizontal and vertical burning (UL-94) test, LOI of 30 v/v%, furthermore in time shifted pkHRR reduced by 75% during combustion. This novel phenomenon promoting the possibility of cost-effective fire retardancy is under further comprehensive investigation in our laboratory.
Based on the described results it can be concluded that products of high technical value can be obtained from secondary raw materials with the preparation of self-reinforced composites upgraded with flame retardancy. It is assumed that the safe self-reinforced polyolefin composites made of low-cost recycled materials could become especially attractive for the industry.

In regard to route 2, the following outcomes have been achieved in the simulation and set up of the MDS facility.
The main challenge of the fluid simulation was to understand the mechanisms responsible for the turbulence observed during the experiment. To achieve this, the chosen methodology consisted in considering three sets of simulations (WP6):

1. Test models. This step consisted in simulating individual parts of the prototype in order to set up the right parameters of the simulation, which is mainly: numerical method; mesh size. The individual parts considered were the laminator and the screens.
2. Full models. Once the tests models were conducted, the full prototype was considered by assembling the individual parts. There, all the important pieces of the prototype were considered. These simulations were useful to depict the regions of interest and to have a global idea of the flow patterns in the prototype.
3. Reduced models. From the results of the test and full models, we understood which regions should considered in more details. They include the injection channel, laminator and separator.
Fine simulations of the reduced models have enabled to identify a zone of generation of flow instabilities. This zone is the intersection of the injection cylinder with the feeding channel. The incoming flow strongly interacts with the recirculating flow inside the cylinder, forming a shear layer. The characteristics of the instability (amplitude and period) correspond to the ones responsible for the particle dispersion at the outflow. Different optimized prototypes were therefore proposed, namely MDS2.1 and MDS2.2 to reduce the turbulence inside the separation channel.

• Simulations of MDS2.1 and MDS2.2 showed a drastically reduced turbulence inside the channel.
• Flow measurements carried out at TU Delft on the optimized MDS2.2 confirmed the results of the simulations.

The combined analysis of experiments and simulations enabled the design of an optimized MDS that meets with the initial requirements. The MDS is able to sort polyolefins with a separation accuracy of 6 kg/m3 that is better than the expectation (10 kg/m3).

As far as the development of the Laboratory Inverse Magnetic Density Separator facility (IMDS) is concerned (route2, WP7), considerable achievements have been made. The first Laboratory IMDS Facility developed at TUDelft did clearly proof the principle that a PP-PE mix can be separated in a continuous process using MDS. The separation accuracy of the first sorting line was 15 kg/m3. Simulation showed that the scale of the turbulence was about 1 cm. In order to achieve the accuracy of 10 kg/m3 the scale of the flow turbulence had to be reduced. Although the proof of principle has been successfully demonstrated with the first IMDS facility, an improved version has been designed based on the experience learned from the former version and has been tested. It is worth to underline that the developed separation technology intends to be a breakthrough in the separation technology field. Current sink-float technologies need about 3 minutes to separate materials from each other while the IMDS will take about 3-5 seconds.
The design of the second IMDS setup aimed to obtain a flow containing less turbulence. One of the improvement of this setup, compared with the first one, is that the injection tube has been narrowed and inclined. By narrowing the injection tube, the flow instability in the injection tube has been reduced and less turbulence is developed in the separation channel (as the the simulations carried out within WP6 showed). In order to produce high quality secondary polyolefins, the wetting technology, the magnetic fluid quality control unit, the ultrasound technology and the Hyper Spectral Imagining process (HSI), were all integrated with the second IMDS facility.
Herein more results obtained within the development of the MDS technology are listed.
(m) It has been discovered that good wettability of polymers is essential for high-accuracy sink-float separation in water-based media such as MDS, because the grade and recovery of the products are critically influenced by even a small percentage of air bubbles. Another important result concerns the cut-density of sink-float separations in magnetic liquids. It has been determined that since it depends on the magnetization of the process liquid, the magnetization needs to be controlled. Based on a magneto-gravimetric principle, a magnetization measurement tool was designed and shown to have sufficient sensitivity to detect and control fluctuations of the magnetization of the process liquid. Research on improving on wettability by immerging polyolefins in boiling water shows the contact angle between flake surface and water became smaller than without boiling. Boiled PP and PE were submerged in magnetic fluid, and the misplacement of the flakes after boiling 1min in water is less than 2 mm in magfluid.

(n) The experiments show that segregation of the magnetic fluid due to incomplete mixing can be avoided by utilizing a static mixer which is able to reduce the size of droplets of concentrated magnetic fluid. By combining a nano-filtration membrane and a static mixer, process liquid is properly controlled online as well. Based on a magneto-gravimetric principle, a magnetization measurement tool was designed and shown to have sufficient sensitivity to detect and control fluctuations of the magnetization of the process liquid in an industrial environment.

(o) The performance of the second IMDS setup on polyolefin recycling has been evaluated by experiments. Two types of polyolefin wastes were used in this study: household packaging waste (from Romania and the Netherlands) and dismantled automotive polyolefin components (from Hungary). The experiments demonstrated that the separation accuracy of the second IMDS can reach 10 kg/m3. For polyolefin flakes, with a thickness more than 0.5 mm, the IMDS process reached a separation accuracy of 5-6 kg/m3. The separation accuracy of the IMDS process is good enough to obtain PP and PE with both high grade (95%) and high recovery (95%) in one single step. The small residual fraction of each test indicates that almost all the input materials can be recovered by the IMDS process, and only a small amount of residues is generated.

(p) A model of IMDS which takes into account the characterization of particles, the quality of process liquid, the uneven magnetic field and the flow turbulences was developed and validated with experiments. The model contains an friendly-use interface, and the users can change the material property and other settings in the MDS according to their need. The comparison between the experiment and simulation show an excellent match. The model can be used to predict or optimize the separation process for polyolefin recycling.

Pertaining to route 3, the achievements in ultrasounds and hyperspectral sensors technologies are described below:

(q) Commercial medical ultrasound technology (imaging machine and sensor array probes) has been adapted to allow real-time image forming (WP8). It is concluded that the ultrasound method has great potential for embedded monitoring applications, such as trouble shooting and controlling settings of the MDS that directly determine the quality of the polyolefin products. Identification by ultrasound of different types of polyolefin plastics proved physically possible and technically feasible as shown by the acoustic materials database. Real-time ultrasound quantitative measurements such as identification proved possible if high quality images are available. For that purpose, real-time video streams have been produced by the medical imager and stored intermittently for image processing to allow for robust time-averaged measurement of the particles. The medical imager proved satisfactory, but is not the optimum machine for future industrial MDS processing. Therefore, research was performed into options for alternative imaging techniques (software) and real-time data acquisition and processing (hardware) to investigate the dedicated MDS ultrasound machine. It was shown that alternative imaging techniques may deliver improved details and more robust image forming. Moreover, it has been shown that some imaging techniques may also allow correct imaging of the back walls of polyolefin particles by which the particles may be sized more accurately, which in turn allows for improved real-time intermittent MDS throughput estimation in terms of polyolefin volume or mass. The ultrasound system have been incorporated into the new prototype MDS for testing its performance in polyolefin waste separation on pilot plant scale.

(r) An innovative hardware and software platform performing HyperSpectral Imaging (HSI) based analysis to characterize and quantitatively assess presence and typology of PO and contaminants inside plastic waste streams, has been developed (WP9). The HSI is operating in the visible (VIS) as well as in the near infrared (NIR) range. The system includes a certified innovative HSI spectral library of different PO waste and contaminants.
The main significant results achieved during the third period are synthetically reported in the following:
• Development, set-up, tuning and testing of the procedures based on HSI for quality control of plastic flake products at laboratory scale.
• Development, construction, integration and installation at Redox plant (The Netherlands) of the HW&SW HSI prototype for quality control of plastic flake products.
• Development, set-up, tuning and testing of the procedures based on HSI for quality control of plastic flake products at industrial scale, through the utilization of the HW&SW HIS prototype installed at Redox plant (The Netherlands).
Potential Impact:
There are environmental, economic and social benefits to be gained by recycling waste materials. However, recycling can only offer a sustainable, long-term solution if the economics of the recycling process is sound and if there is a genuine market demand for secondary materials.
W2Plastics brought about a fundamental change of the status of complex plastics wastes such as the ones contained in CDW, household waste, ASR and WEEE. Currently those materials are poorly recycled because the economics of the entire recycling process (including logistics) is not attractive. The recycling of polymer packaging waste in the Netherlands may serve as an example: there packaging producers use ca 1 billion euro of polymer to make about 600,000 tons of polymer packaging and pay a fee of between 75 and 100 million euro yearly for the separate collection and recycling of ca 100,000 tons of EOL packaging into ca 20 million euro worth of secondary polymer products. This poor performance is why most of the plastics wastes contained in the above main waste streams are largely landfilled or incinerated at the moment. Because of its robustness and separation efficiency, W2Plastics technology has the potential to change the status quo for packaging waste (the largest stream of polymer waste) by providing, for the first time, a very small residue stream and almost complete recovery of the input waste polymers into high-value products, resulting in a positive difference between revenues and processing costs in Eastern European markets. In Western Europe, external fees will still be necessary to cover the substantial cost of separate collection, but they will be significantly less than today.
So far the difference between selling price and cost has been negative so that the governmental contributions have been necessary to reach the (low) plastics recycling rates observed in Europe. This is a pity because 1 M €/yr of primary polyolefins relates to an import of 0.8 M €/yr of oil, and only 0.4 jobs, whereas 1 €/yr of recycle polymer is produced from locally available raw materials and creates 5 jobs. In order to effectively increase the plastics recycling rates, it is of outmost importance to reduce the processing costs of the whole recycling process and to provide high-value secondary raw plastics materials to the market. This will happen when the recycling technology will be able to tap into the large sources of plastics waste in an economically sound way while providing materials of consistent and high quality. W2Plastics provided a feasible innovative technology which has proven to have a positive gross margin and therefore is likely to provide positive effects (at environmental, economic and social levels) once it will be fully exploited on the market.
W2Plastics will trigger the market demand for secondary Polyolefin’s by separating complex plastics waste (including polyolefin mixtures) to such sorting precision that it will be possible to use them for high quality products.
W2Plastics will achieve the above advantages and jobs by providing the recycling industry with a “beyond the state of the art” sorting equipment allowing the production of high quality secondary polyolefin’s at competitive costs.
In addition to the direct environmental and social impact, the technology of W2Plastics is useful in other fields of raw materials, and has spin-offs particularly in heavy plastics, metals and minerals purification.

Concerning the dissemination activities, several actions have been carried out.
The activities carried out aimed at strengthening the collaboration between universities and industries and within universities and industries themselves. Therefore university staff members and experts from industry met with each other in order to make the knowledge flow smooth and productive. Five students from different partner universities have been exchanged between partners. They have been involved in research study concerning several WPs.
To contribute to transfer technologies and information needed for the production of high-purity polyolefins from complex wastes (one of the main objective of the project), two successful workshops have been organized by the consortium. The first workshop titled “Magnetic Sorting and Ultrasound Sensor Technologies for Production of High Purity Secondary Polyolefins from Waste; the W2Plastics Project” has been organized within the International Conference on Materials Science & Engineering BRAMAT 2009 conference hold at Transilvania University of Brasov on February 26th – 28th 2009.
During the workshop, seven papers have been presented by W2Plastics partner members. Later on all papers have been published in the Environmental Engineering and Management Journal or in Metalurgia International journals.
The second W2Plastics workshop has been held in Philadelphia within the Conference on Solid Waste Technology and Management on March the 27th-30th 2011. Seven papers have been presented by project members. Because of the above initiatives, dissemination in a European (and beyond) contest has been achieved.
Moreover nine articles concerning W2Plastics topics have been published on a special section dedicated to the W2Plastics project in the Open Waste Management Journal (Betham Science Publishers) and a dissemination leaflet have been published on the magazine The Parliament.
A training course has been organized during the exhibition Romenvirotec 2012 held in Bucharest in February 2012. Basic theoretical and practical information have been given to the industry technicians and management staff members who attended the training course. Besides background useful for the proper usage of the equipment, the most important knowledge concerned to the project (LCA, CFD, market analysis, quality assessment, mechanical properties, etc.) have been presented to the attendants.
The training course has been provided by W2Plastics participants both in English and in Romanian languages.
A TV programme dedicated to science and technology broadcasted an interview given by the project coordinator and other Romanian project partners. The importance positive effect of plastics recycling on the environment and the profitability of the W2Plastics technology as recycling tool have been emphasized. Two articles describing the W2Plastics project and the MDS technology have been published on Romanian magazine and web site. The following pictures show some of the above dissemination activities.
A workshop named: W2Plastics: A Recycling Example has been organized within the Romenvirotec 2012 held in Bucharest between February 28 and March 2, 2012.
During the workshop, W2Plastics partner members from industry and university introduced the attendants with the technologies developed and used within the project.
Because of the above activity, the dissemination in the Eastern Europe contest has been achieved.
Furthermore during the project implementation a total of 85 articles have been published in scientific journals, conference proceedings or magazines. Moreover two PhD thesis have been finalized and one more is going to be defended in June 2014.
All publications are the results of scientific research carried out by the consortium within the W2Plastics project.
List of Websites:
www.W2Plastics.eu

Contact details:

Dr. Francesco Di Maio
Delft University of Technology
Faculty of Civil Engineering and Geosciences
Stevinweg 1, 2628 CN Delft, The Netherlands
+31 15 278 81 48 (office)
+31 15 278 81 62 (fax)
+31 6 186 859 65 (mobile)