← Curriculum
Module 04 · 16 min

Contamination and Quality

the make-or-break variable

Before we start

Pick up a black plastic ready-meal tray. The plastic it's made from is often perfectly recyclable. And yet that tray will almost certainly end up in a landfill or an incinerator — not because of what it is, but because of its colour. The sorting machine literally cannot see it.

That single example contains the whole lesson of this module. Recycling does not fail only because of bad materials or bad intentions. It fails because of quality, contamination, and the hard physical limits of the machines that do the sorting. Quality is the variable that decides whether collected material becomes a product you can sell or a cost you have to pay to dispose of.

By the end of this module you should be able to:

  1. Define contamination and name its main types.
  2. Explain, with numbers, how contamination destroys both yield and value at once.
  3. Explain why black plastic is the field's clearest example of 'recyclable in theory, not in practice.'
  4. Make a basic economic case for keeping a stream clean at the source.

Part A — The concept

What contamination actually is

Contamination is anything in a recycling stream that shouldn't be there: the wrong material, or the right material fouled with food or liquid. Three types to know.

  • Wrong-material contamination — items that don't belong (bags, textiles, hoses, electronics, batteries). Some, like film and 'tanglers,' physically jam the machines.
  • Residue contamination — food and liquid left in or on otherwise-recyclable items, which spoils paper and gets loads downgraded.
  • Wishcycling — the well-meaning version, doubtful items added in hope. It feels good and it makes things worse.

Contamination destroys yield and value at the same time

Two things happen when a stream is dirty, and you must hold both in mind.

First, yield falls — which you proved to yourself in Module 2: pushing contamination from 15% to 30% sharply cut the final recyclate. At global scale, the OECD estimated that about 40% of the plastic collected for recycling is disposed of as residue rather than turned into new material (OECD, 2022). That is the yield cost of impurity.

Second, value falls. Recyclables are commodities priced on purity. A clean bale sells; a contaminated one sells at a discount or not at all, and can flip from something you sell into something you pay to remove. When China tightened its import standards in 2018 it demanded a contamination level of just 0.5% — a level so strict that most Western material could not meet it, which is a large part of why the export route collapsed (Earth.Org, 2021; see Module 2).

So contamination is not one problem. It is a yield problem and a value problem hitting at once, which is why operators obsess over it.

The part they don't teach in class: a recyclable plastic, defeated by its colour

Figure · Why an NIR sorter cannot see black plastic
NIR sensor ▾ (emits + reads)→ air jets sort
PET
↑ reflects
HDPE
↑ reflects
PP
↑ reflects
BLACK
absorbed
PS
↑ reflects
Sorted chutes
PET · HDPE · PP · PS → recycled
Residue chute
BLACK passes through → landfill / incineration
Carbon-black pigment absorbs near-infrared across the range, so the item returns no signature to the sensor.

Most sorting plants identify plastics using near-infrared (NIR) light: each polymer reflects a distinct signature, and the machine reads it and fires an air jet to sort the item. But black plastics are coloured with carbon black, a pigment that absorbs near-infrared light across the whole range. So a black item reflects almost nothing back to the sensor. To the machine it is invisible, and it passes straight through to residue — headed for landfill or incineration — even though the underlying polymer (often PP or PET) is recyclable (SC Johnson; Beyond Plastics; Plastics Engineering, 2026).

The honest update, because this field moves: the problem is being worked on. Newer mid-infrared and AI-assisted sorters, and NIR-detectable black pigments that replace carbon black, can now identify some black plastics (Packaging Dive, 2024–25). But across most existing plants today, black plastic still falls through. Watch for this distinction in any 'we can now sort black plastic' claim: is it in real plants at scale, or in a demonstration?

"When in doubt, throw it out"

This sounds wrong to environmentally minded students, so let me be blunt. In recycling, a confident wrong 'yes' is usually worse than an honest 'no.' One contaminated item can pull down a batch; one careful exclusion costs only that item. Good systems make it easy to get this right.

Part B — Company & Tech Spotlight

Dossier № 04
Dossier · Norway

TOMRA — sensor-based sorting

If contamination is the enemy, sensor-based sorting is the main weapon — and the clearest example is TOMRA.

What it is

TOMRA is a Norwegian company founded in 1972, headquartered in Asker. It is known for two things: reverse vending machines for deposit-return systems (Module 5) and sensor-based sorting for recycling plants, where it reports thousands of installed systems across more than 100 markets (TOMRA company materials).

How it fights contamination

On a fast belt, sensors — commonly near-infrared, plus cameras and other detectors — identify each item by material, and a precisely timed air jet ejects the target into the right chute. This is how a plant separates PET from HDPE, or pulls stray non-target items out of a stream, at high speed and high purity. Raising the purity of an output fraction is the same thing as removing contamination from it.

Where the limits are, honestly

The black-plastic problem above is exactly a limit of standard NIR sorting, and TOMRA and its competitors are now adding lasers, AI, and other sensing to close it. Two more honest points worth teaching: heavily mixed or fused materials still defeat automated sorting, and a sorter that hits 95% recovery in a clean demonstration can drop sharply during a real shift as dust builds on its lens. Technology raises quality; it does not excuse a filthy input. The cheapest contamination to remove is the contamination that never entered the bin.

Who hires for this kind of work

Sensor-sorting companies hire automation and controls engineers, optical and software engineers for the recognition systems, process engineers for plant layout, and field-service engineers to commission and maintain installations. If you like machines that make thousands of correct decisions a second, this is a clear target.

Part C — Do the work

Task 1 — Quantify the cost of contamination

Figure · The contamination flip — quality is money
Below threshold (≤15%)
+€60/t
Processor pays the city for clean recyclate.
Revenue ↑
Above threshold (>15%)
−€40/t
City pays a disposal charge for the same load.
Cost ↑
Threshold
15%
One percentage point on the wrong side of the threshold can flip a load from revenue to cost.

A processor pays a city €60 per tonne for clean recyclate, but if contamination in a load exceeds 15%, the deal flips and the city must pay a €40 per tonne disposal charge for that load.

The city sends 200 tonnes this month at 12% contamination and another 200 tonnes at 18% contamination. Calculate the net financial result for the month, then write one sentence on what a single percentage point of contamination is worth here. (Simplified model for intuition; real contracts use sliding scales, but the lesson holds: quality is money.)

Task 2 — The black-plastic hunt

Spend five minutes in your kitchen and bathroom. List every black plastic item you find (trays, lids, bottles, containers). For two of them, state the likely polymer and explain, in one line, why a standard NIR sorter would miss it. Then suggest one design change a brand could make so the item becomes sortable.

Task 3 — Design a source-level fix

Pick one common contaminant (plastic bags, food residue, batteries, tanglers). Propose one intervention at the source (not at the plant) to reduce it, and name one way you would measure whether it worked.

Deliverable. Your monthly financial result with the value of one contamination point, your black-plastic list with two explained items and a design fix, and your source-level fix with a measurement.

Watch

One is enough here. Good sorting-plant videos exist, but most are company-owned and not openly licensed — so I've kept this deliberately to a single Creative Commons piece that reinforces the module's theme.

Watch
Shannon Odell · TED-Ed
· ~5 min
Confused about recycling? It's not your fault

A short animation on why so few plastics are actually recycled and what the symbols really mean. Notes that only about 9% of plastic is recycled — reinforcing the quality-and-confusion theme of this module.

Open on YouTube ↗

From the field

Sources

  1. OECD (2022), Global Plastics Outlook: ~40% of plastic collected for recycling disposed of as residue. https://www.oecd.org/environment/plastics/
  2. Black plastic and near-infrared sorting — carbon black absorbs NIR, making black plastic undetectable to standard NIR sorters; ~15% of plastic recyclables; mid-infrared/AI sorters and NIR-detectable pigments emerging. SC Johnson (2023); Beyond Plastics fact sheet; Plastics Engineering (2026); Packaging Dive (2024–25). Detection technology is improving — re-verify current state before publishing.
  3. China 2018 import restrictions, 0.5% contamination limit: Earth.Org (2021). https://earth.org/chinas-national-sword-policy/ (see Module 2 sources).
  4. TOMRA company website / history — founded 1972, Asker; sensor-based sorting (AUTOSORT) and reverse vending; 100+ markets. https://www.tomra.com/
  5. Shannon Odell, 'Confused about recycling? It's not your fault,' TED-Ed. https://ed.ted.com/lessons/confused-about-recycling-it-s-not-your-fault-shannon-odell (Creative Commons).

All URLs accessed June 2026. Where figures are paraphrased, the linked primary documents are the authoritative reference — quote directly only with attribution.

Do the work · 3 tasks

Submit your work for feedback

Each task is graded immediately by an AI tutor using the rubric below. An instructor can override the AI score and leave a final comment.

Task · max 10 pts

Quantify the cost of contamination

A processor pays €60/t for recyclate at ≤15% contamination but charges €40/t disposal above that. Two loads of 200 t arrive — one at 12%, one at 18%. Fill the financial breakdown and one insight.

Grading rubric

Full credit for: (1) correct net result — 200 t × +€60 = +€12,000 for the clean load; 200 t × −€40 = −€8,000 for the dirty load; net +€4,000. (2) A sensible 'value of one point' statement (e.g. the 3-point gap from 15% to 18% flipped €20,000 of value, so roughly €6–7 k per point in this scenario; any defensible derivation OK). (3) Visible working, not just a final number. Score out of 10.

Item€ value (+ income / − cost)How you got there
Clean load (200 t at 12%)
Dirty load (200 t at 18%)
Net result for the month
Sign in to attach a file.
Task · max 10 pts

Black-plastic hunt + design fix

Walk through your kitchen and bathroom; list the black plastic items you find, identify the polymer for two of them, then propose one brand-side design change so they become NIR-sortable.

Grading rubric

Credit for: (1) a real list of at least 3 items; (2) two items with a plausible polymer guess and a correct NIR explanation (carbon-black absorbs NIR → no signature); (3) one concrete, realistic design fix (e.g. NIR-detectable pigment, switch to natural/white colour, remove sleeve). Score out of 10.

ItemItem nameLikely polymer (PP/PET/PS…)Why standard NIR misses it
Item A
Item B
Sign in to attach a file.
Task · max 10 pts

Source-level fix with a measurement

Pick one common contaminant. Propose an intervention at the source (household / retail / product design — not at the plant) and one concrete way you would measure whether it worked.

Grading rubric

Credit for: (1) intervention is genuinely at the source — household, retail, or product design — not 'add another sorter'; (2) named contaminant matches the intervention; (3) measurement is concrete and tied to the intervention (bin-audit % before/after, fire incidents per month, reject-load tonnage, etc.). Score out of 10.

Sign in to attach a file.
Module quiz · 6 questions · pass at 80%

Check your reading

  1. 1.

    Which of these is NOT one of the three types of contamination introduced in this module?

  2. 2.

    According to OECD (2022), roughly what share of plastic collected for recycling is disposed of as residue?

  3. 3.

    Why does a standard NIR sorter miss black plastic?

  4. 4.

    Roughly what share of plastic recyclables is estimated to be black plastic?

  5. 5.

    What contamination limit did China's 2018 import standards demand?

  6. 6.

    Founded in 1972 in Asker, Norway, this company is best known for sensor-based sorters and reverse vending machines:

0/6 answered
Module checklist

Apply what you've learned