The need to drive up the quality of recycled material in the UK has long been a concern for the sector. But what is the best way to go about it? Many have thought that the key to this would be government’s Material Recovery Facility (MRF) Code of Practice, which was finally published in draft form in January this year. At the time of writing, the proposals will make it mandatory for UK MRFs to test the quality of their inputs and outputs, though only those that process over 1,000 tonnes of dry recyclate per annum will be required to do this. Meanwhile, the document recommends that the sample sizes that operators test can be as low as 20 to 25 kilogrammes (kg). Some stakeholders have been far from impressed.

“They’re not really serious”, says Ray Georgeson, Chief Executive of the Resource Association, the industry body that represents reprocessors (which ultimately have to deal with MRF material). “The lack of frequency and the small sample sizes will make it very easy for rogue operators to circumvent. There is no provision in the regulations for the Environment Agency (as the regulator) to make unannounced inspections of MRFs and there is not any agreement yet on how sampling is done, the volume of weights, frequencies and the degree of sophistication that might be needed in relation to different material streams.

“I’m genuinely baffled why the waste management industry and Defra appear resistant to any movement on these technical aspects. Surely it’s in their interest to raise the bar, to set a high standard of environmental compliance and good practice?” We asked Defra for a response to these questions but did not receive a reply.

However, there are others who are broadly in favour, albeit with some reservations. “Generally, I’d say it’s a move in the right direction”, says Keith Freegard, Director of Sales and Marketing at Axion Recycling, which specialises in plastics. “I just think they could have taken a big rather than a tentative step forward. I would have liked it to have embodied some industry agreed standards. The chief stakeholders in each material stream and MRFs could have got their heads together and worked out a set of standards that would be applicable to that particular stream. You could have implemented some sort of gold/silver/bronze standard on bale quality.”

Sample size is obviously still a source of some debate, but how will the sampling itself be conducted and what sort of technology will be used? “I suppose ‘fit for purpose’ is the slick answer”, says Matthew Farrow, Director of Policy at the Environmental Services Association, the trade body that helped draft the code of practice, and which represents many of the UK’s MRFs. “Obviously the MRFs themselves use quite sophisticated high-tech optical sorters and magnets to pull apart the input material and make sure that it’s all separated out. But in my mind, the code of practice doesn’t require that and as far as I’m aware MRFs that use automation are no more accurate than those that rely on hand sampling.”

In the immediate future, it seems cutting-edge solutions are not on the agenda. “My take on it is that most sampling is going to be incredibly low-tech”, says Sam Reeve, Operations Director of the consultancy Resource Futures, who has conducted research into MRF sampling techniques. “It doesn’t need to be high-tech, and by making it high-tech you’re adding a cost that I don’t think they need to incur – that’s one of the arguments by many operators, that the cost of doing this
is prohibitive.”

Reeve suggests the technology used for testing will not extend to much more than a bin on the floor or a larger container for plastics. “If the samples are only going to be around 20-25 kilos, that is all you need. It will probably be a case of putting said container on the ground and taking a random grab of input material and dropping it over said container so you randomly get your amount of material you need.”

Yet as far as Reeve is concerned, there are still issues. “How it’s done is really important. So it’s things like if they’re managing the input pile, due to the nature of the material you will get all the dense stuff sinking to the bottom. When it comes to sampling, if your shovel load is only going into the top of the pile then you’re missing some of the denser stuff. So having guidance on how this low-tech sampling is done is important. It’s not rocket science, but it’s things like being aware that you need to take a section to ensure that it is representative.”

Over at Axion, Freegard largely concurs. “The only sampling technology you need is a basic understanding of O-Level statistics. Plus some scoops, a floor, bins and baskets, a sorting table and a set of scales that are accurate to a gramme maybe. And a little bit of software on a spreadsheet. The total investment in the special technology is probably less than £1,000.

“When we do sampling, we arrive on site with a load of buckets, bins and a sorting table. We have some sorting people, a young guy who is confident on the laptop. It takes about 15 minutes to train the guys – saying ‘This is a PET bowl, this is a PE bowl. Sort it all out into various bins. One for PET, one for PE, weigh the samples, put it in a computer.’ Bang.”

He makes it sound simple. Certainly, more complex sampling technology still seems some way off. Freegard talks about the possibility of a conveyor chute system, where every so often a lever is switched to divert two or three seconds’ worth of material into a sampling bin. Reeve, meanwhile, sees infrared technology as playing some role in the future: “You could have a sensor that would be able to read the composition for you, whether that’s input or product. There could be some sort of scanner above your in-feed belt that could tell the difference between a plastic film and a dense plastic bottle, and could then spit it out onto your computer screen that your input feed is running at, say, 22 tonnes an hour and comprising of 40 per cent fibre, 16 per cent polymers, etc. I have to say, though, that I’m not aware of anything like this ever getting to the testing stage.”

In 2009, when WRAP conducted its MRF Quality Assessment report, the technology was nothing more elaborate than hand sorting 19 material streams over a 45-milimetre screen. Interestingly, the sample sizes used in the study were between 75 and 100 kg. The report concluded: ‘There is a requirement for future sampling to become more robust within MRFs, but a need to recognise the immense resource commitment in order to undertake the appropriate level of sampling/testing.’

Certainly it isn’t surprising that it is often the larger operators with more resources that have a reputation for producing higher-quality material from their MRFs. One of these is paper mill operator UPM.

“We have three processes”, explains Commercial Director Simon Walker. “Every load is visually inspected when it comes in – a lorry driver discharges it and there’s a guy there watching the material looking for major non-conformances, say if it’s very wet.
Then we would take a 220-kilo sample out of a trailer. We feel that that figure gives us a much more representative, realistic sample of a whole load. The difficulty you get if you take a smaller sample, [is that] you automatically exclude the bigger pieces of material, because you literally can’t fit them into your sampling bucket.

“That 220 kilos then goes into a hand-sorting mini-MRF where the material is separated into its constituent parts. Then we use a series of mechanical spotting techniques, and then optical sorting. From then on it’s all mechanically controlled – near infrared, eddy current separation, magnetic separation and gravity separation. Then finally it will pass in front of a quality control person who will just negative pick, i.e. pick the stuff that we don’t want off the system.”

UPM also has a dedicated testing facility. “That is really to give us much more accuracy for the local authority. It is a 100 per cent check and we do that on an ‘as per contract’ basis. We do between one in six and one in 10 vehicles, which is six to eight times the amount required in the MRF Code of Practice.”

It is this thoroughness, combined with the sampling and sorting tech, that is necessary to consistently produce quality recyclates. Walker explains: “Well, if you start with the principle of ‘If we know what’s going in, then we have every opportunity to work out what to get out’, then you can start tweaking the system, like working the system at different speeds – running it faster if you notice there is no contamination. We have an infinitely variable configuration of the MRF. It’s not one thing, it’s a series of adjustable features that make us produce the quality we produce.”

UPM has the resources to invest heavily in its facilities, but Walker still feels that the company’s model is one that the whole industry would do well to follow: “Quality is the future, and the modern, larger MRFs are all heading in this direction. I think the days of the dirty horrible MRF sat in the corner putting out fairly grim material are numbered. More and more paper mills, bottle manufacturers and glass manufacturers all want a higher standard, just to keep the quality of their outputs at the right level. Our view when we first set out was ‘Well, if we want higher quality, why wouldn’t the rest of our customers want the same?’”

Whether the MRF Code of Practice will be able to deliver higher quality across the board is, of course, another matter. Whilst its implementation is undeniably a step in the right direction, it may be that it will take a quantum leap forward in terms of investment across the board to bring about a rise in the overall quality of UK recycling.