The Plastic Recycling Process from Bale to Pellet

The plastic recycling process turns baled waste into clean flake or pellets through six mechanical stages: sorting, size reduction, washing, separation, drying and pelletizing. A 1,000 kg/h line draws around 250 kW of installed power, consumes 8–10 m³ of water an hour and needs 6–8 operators on shift.

Most descriptions of that sequence stop at the verbs. They tell you the material is sorted, then shredded, then washed — which is true and completely useless if you are about to specify a line. What decides whether a plant makes money is not the order of the stages but the numbers set inside each one: the screen size in the crusher, the temperature and alkali dose in the wash tank, the residual moisture leaving the dryer, the filter change interval on the extruder. This article walks the line stage by stage and gives the judgement behind each of those settings, along with the measured consumption figures for five line sizes. If reading the settings is not enough, you can also watch these stages running end to end on a working floor before you specify anything.

One thing worth stating before the first stage. This is mechanical recycling: the polymer stays a polymer throughout, and every stage is a physical operation. Chemical recycling breaks the chain back to monomer and is a different industry with different economics. Nothing below applies to it.

Stage One — Sorting, and When You Need a Manual Pick Station

The line starts when a bale is broken open. A bale breaker loosens the compressed block, a trommel screen tumbles out stones, sand and loose caps, and a metal detector sits ahead of everything with a rotating knife.

That last point is the whole argument for spending money here. A single steel fastener that reaches the cutting chamber can take out a set of knives, and the cost is not the knives — it is the shift you lose changing them. Sorting is not a cosmetic step at the front of the line; it decides whether the machines downstream are protecting value or grinding contamination into it.

The question buyers actually have to answer is whether to add a manual pick station on the belt. Three situations make it necessary:

• Mixed post-consumer bales. If the incoming material contains PVC, no downstream stage will remove it — PVC sinks alongside PET and survives every wash. It has to be taken off the belt by hand or by an optical sorter.
• Colour separation. Colour cannot be washed out. If any part of your output is aimed at a market that pays for clarity, the coloured items leave here or not at all.
• Unknown or variable feedstock. A plant that buys bales from several brokers is really buying several different materials. A pick station is the cheapest form of insurance until the supply stabilises.

Conversely, a plant running its own clean industrial offcuts — a single polymer, a known source, no labels — can usually skip the pick station entirely and put the money into the wash instead.

Where labels are the problem rather than the polymer, a dry friction label remover placed before crushing takes off 80–90% of labels and caps while the bottles are still whole. Sequence is everything: a label on an intact bottle is one large object that friction can strip, while the same label crushed alongside the bottle becomes thousands of fragments the same size as the flake, and separating those costs far more. This is standard on a PET bottle washing recycling line for exactly that reason.

Stage Two — Size Reduction, and the Difference Between Shredding, Crushing and Pelletizing

Three words get used interchangeably in this industry and they describe three different machines doing three different jobs. Getting them confused is the most common way a first line ends up mis-specified.

The practical distinction is torque versus screen control. A plastic shredder machine exists to make large, awkward objects small enough to handle — drums, crates, pipe, baled film. It works slowly and with enormous force, and it does not give you a controlled particle size because there is no screen deciding when a piece may leave.

A heavy-duty plastic crusher does the opposite. It runs fast, cuts against fixed blades, and nothing exits the chamber until it is small enough to pass the screen. That is what produces the 12–20 mm flake the washing stages are designed around, and it is why the screen is the single most important spec on the machine.

Most lines need both, in that order, and the ones that need only one are the exceptions: clean film going straight to a compactor, or small rigid offcuts that already fit the crusher throat. Choosing between the shredder types is a separate decision driven entirely by the shape of your material — that is covered in detail in how to choose a recycling shredder by material.

One detail that surprises people: on a PET line the crushing is done wet, with water already in the chamber. It suppresses dust, starts lifting the surface contamination, and stops the flake from taking on static that would make separation harder later.

Stage Three — Washing, and How the Temperature and Alkali Dose Are Set

Roughly 80% of the work in a recycling line happens here. The public image of recycling is melting; the reality is cleaning, and everything after this stage is just handling material that has already been made clean or has not.

A full plastic recycling washing line runs three mechanisms in series, and each removes a different class of contamination.

Friction washing — mechanical, not chemical

A high-speed friction washer throws flake against itself and against the drum wall under a water spray. It removes what is physically stuck on: dirt, fibres, paper pulp, sand. No chemistry involved, and it is the cheapest cleaning per tonne in the line, which is why an under-sized friction stage is a false economy that shows up as chemical cost later.

Sink–float separation — density does the sorting

A sink-float separation washing tank exploits the fact that PET has a density above water while PE and PP have densities below it. PET flake sinks, cap and label fragments float, and the two streams leave by different routes. It is elegant, cheap to run and completely passive.

Its limitation is worth knowing before you rely on it. Density separation only works when there is a density gap. It removes PE and PP from PET reliably, but it cannot separate PP from PE — both float — and it cannot separate PVC from PET, because both sink. Any line whose contamination problem is PP/PE mixing needs a different answer.

Hot alkali washing — the expensive stage, and the one to justify

Hot washing runs at 85–95 °C with 1–3% alkali or detergent. That range is not arbitrary. Below roughly 85 °C the hot-melt adhesives holding labels on do not soften enough to release, and oils stay emulsified. Above 95 °C you are paying substantially more in heating energy for very little additional removal, and on thin film you start risking deformation.

The alkali dose is set by what is on the material, not by the polymer. Adhesive and food residue need the upper end; general surface dirt does not need alkali at all and a strong friction stage plus cold water will do it. This matters because hot washing brings heating energy, chemical dosing and a wastewater stream that has to be treated, and those costs recur every hour the line runs. The decision of when it is genuinely required is worked through case by case in hot wash versus cold wash and when each is enough.

Material type changes the whole balance. A rigid plastics washing line handles material that is thick, holds little water and tolerates aggressive mechanical action. A washing line for soft plastics deals with film that carries far more surface area per kilogram, traps sand in folds and cannot be treated roughly. Same six stages, different machines in almost every one of them.

Stage Four — Drying, and Why Moisture Has to Go Below 2%

Drying looks like a housekeeping step and is in fact a quality gate.

A dewatering centrifugal dryer spinning at around 1,500 rpm throws off free water and brings flake down to 0.5–2% residual moisture. A hot-air dryer then takes it below 1% where the downstream process demands it.

The reason the target is that tight is what water does in an extruder. Moisture flashes to steam in the melt and leaves voids in the strand, which show up as bubbles in the pellet and breaks at the die. Worse, in PET and other condensation polymers water at melt temperature drives hydrolysis, cutting the polymer chains and dropping the intrinsic viscosity of the entire batch. That loss is permanent. A batch that goes into the extruder wet cannot be recovered by drying the pellets afterwards.

Film is a different problem and needs a different machine. Soft plastic holds water in every fold, and a centrifuge alone will not get it out at any reasonable throughput. A plastic squeezer and densifier mechanically compresses the film, driving out water and simultaneously densifying it so the extruder can feed it consistently. Trying to run washed film into a pelletizer without this step is the classic way a film line ends up at half its rated output.

Stage Five — Pelletizing, and Where Single-Stage Stops Being Enough

Pelletizing melts clean, dry material, filters the melt and cuts it into uniform pellets. Extrusion, screen changer, die, cutter, cooling — the sequence is the same everywhere. The real specification decision is single-stage or two-stage.

Single-stage is one extruder doing everything. It is cheaper, simpler, uses less power and takes less floor space. It works when the input is genuinely clean and dry and consistent: in-house offcuts, well-washed rigid flake, single-polymer feed.

Two-stage adds a second extruder after the first, with an additional filtration point and better degassing between them. You pay for it in capital and in energy, and you buy tolerance — for residual moisture, for volatiles from printing ink, for contamination levels that would blind a single screen pack within an hour. Post-consumer film with printing on it is the standard case where two-stage stops being optional.

The machine also follows the material. A SJ pelletizing machine for rigid plastics feeds flake directly into the screw, because rigid flake has bulk density and flows. An ML pelletizing machine with compactor for soft plastics puts a compactor in front of the extruder, because film is too light and too bulky to feed a screw on its own. For clean PE film in smaller volumes there is also a low-temperature air-cooling granulator that skips the water bath entirely.

Not every plant should pelletize at all. Washed and dried flake is a saleable product in its own right, and stopping there removes the most capital-intensive and most energy-intensive part of the line. Whether the extra step pays depends entirely on who is buying — the difference in how the two products are priced and judged is set out in plastic regrind and how buyers judge its value.

What One Tonne Actually Consumes

This is the part almost nobody publishes. The figures below are the specification of a PET bottle washing line across five capacities, as built.

Converted to consumption per tonne of throughput, which is the number that actually goes into a business case:

Read across those rows and the conclusion is hard to miss. Going from 500 to 3,000 kg/h cuts installed power per tonne by 44% and labour per tonne by roughly 55%. The largest single cost lever in plastic recycling is line size, not machine brand.

Two qualifications, both important. First, installed power is the connected load, not the running load — a line in normal operation typically draws 60–70% of it, and anyone quoting installed power as an electricity bill is either confused or selling something. Second, a bigger line only delivers those numbers if you can keep it fed. A 3,000 kg/h line running at half capacity has worse per-tonne economics than a 1,000 kg/h line running full, so feedstock supply, not machine capacity, is what should be settled first.

The same throughput, three different bills

The table above is one material. Put the three classes side by side at a common throughput and the spread is larger than anything inside a single table — which means the choice of stream affects your operating cost more than the choice of capacity does.

Four conclusions come out of that comparison, and none of them appears on an equipment quotation.

A film line draws two to three times the power of a rigid line at the same tonnage. 128 against 253–400 kWh per tonne. The reason is in the stages described above: film has to be squeezed, then hot-air dried, then air-conveyed, where rigid flake is thrown dry in a centrifuge and moved on a belt. Buyers comparing the two on machine price never see this layer.

A film line uses roughly three times the water. 5.7–7.1 against 13.3–20.0 m³ per tonne. Where water supply or effluent capacity is constrained, this single row can end a project whatever the equipment costs.

Scale works hard on energy and barely at all on water. A rigid line moving from 250 to 1,900 kg/h drops from 400 to 131 kWh per tonne, a fall of about two thirds. Over the same range water only moves from 16–20 down to 5.3–7.9 m³, and most of that is the smallest tier having been over-specified rather than the large one being efficient. All three classes flatten out on water by mid-range, so planning to dilute a water bill by building bigger does not work.

PET is the most labour-dense of the three. At the same tonnage it needs 5.3–6.7 people per tonne per hour against 2.1–3.6 on rigid, and the whole difference stands at the sorting belt described in stage one. That is why PET projects behave so differently in high-wage and low-wage economies even with an identical machine list.

At the small end the gap is narrower but the same shape: at 200–300 kg/h a rigid line runs about 100 kW over 300 m² with 2–3 operators, while a soft line at the same throughput needs 110–160 kW and 350 m².

These figures cover the equipment itself. Freight, duty, foundations, commissioning, wastewater treatment and electrical supply upgrades all sit outside them, and they are frequently larger than buyers expect — what adds up in plastic recycling plant cost works through the layers. For the machine list on its own, what machines are needed for a plastic recycling line separates the mandatory from the optional.

For context on the quality standards the output is measured against, the APR Design Guide sets out the recyclability and quality criteria used across the industry, and the US EPA publishes material-specific data on plastics in the waste stream that most post-consumer feedstock comes from.

Frequently Asked Questions