login
quiescence.eu
Skip to content

Semi-finished product · Electronic components

Printed circuit board assembly

One kilogram of assembled printed circuit board

This page shows what the product is made of and where each figure comes from. It does not show a death toll. The toll is calculated from this list of materials, and will be published only once it has a range and the model has passed its test on past years.

A printed circuit board by mass, as sixteen studies average it: 34.7% metals — copper 21.4%, iron 3.3%, tin 3.1%, aluminium 3.0%, lead 1.9%, nickel 1.0%, zinc 0.7%, silver 0.13%, palladium 0.05%, gold 0.04%, tantalum 0.01% — and 65.3% glass fibre, epoxy resin and other non-metals. Boards differ by product and age; the source's standard deviations are as large as some of the means.

Is something on this page wrong? Report it →

The human cost, per item

One item is one kilogram of assembled printed circuit board. Scenario Economic-embedded, using FIGARO 2023. Each figure is the middle result of 2,000 runs of the calculation, with the range that holds nine in ten of them beside it, and the grade that range earns.

Deaths

1.6 × 10⁻⁶ per item(0.60–3.4 × 10⁻⁶)LOWS-MAX · run 2259 · methodology v1.0 · data dv-20261007-44r

Deaths per item: the rangeThe range that holds nine in ten of the runs, on a scale where each step is ten times the last: 0.60–3.4 × 10⁻⁶. The dot is the middle result; a figure graded indicative has none.10⁻⁷10⁻⁶10⁻⁵10⁻⁴

The product’s list of materials traces 39% of its weight to raw materials the model knows. The rest carries no harm in this method, which is why this method gives a floor.

Injuries

1.3 × 10⁻⁵ per item(0.50–2.8 × 10⁻⁵)LOWS-MAX · run 2259 · methodology v1.0 · data dv-20261007-44r

Injuries per item: the rangeThe range that holds nine in ten of the runs, on a scale where each step is ten times the last: 0.50–2.8 × 10⁻⁵. The dot is the middle result; a figure graded indicative has none.10⁻⁶10⁻⁵10⁻⁴

The product’s list of materials traces 39% of its weight to raw materials the model knows. The rest carries no harm in this method, which is why this method gives a floor.

Deaths and injuries at work in these figures come from the 53 countries that reported work deaths and the 52 that reported work injuries for 2023.

Harm at work is counted as each country reports it to the UN’s labour statistics (ILOSTAT), and never estimated for a country that did not report: such a country is missing from that year, not counted as zero. So when two years’ figures differ, part of the difference can be a change in which countries reported. The model’s own year-to-year checks compare only countries that reported in both years.

Deaths: Argentina, Australia, Austria, Azerbaijan, Belarus, Belgium, Bulgaria, Canada, Colombia, Croatia, Cyprus, Czechia, Denmark, Ecuador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Rep., Latvia, Lithuania, Luxembourg, Malta, Mauritius, Moldova, Mongolia, Netherlands, Norway, Panama, Poland, Portugal, Romania, Russian Federation, Singapore, Slovak Republic, Slovenia, Spain, Sri Lanka, Sweden, Switzerland, Thailand, Turkiye, Turkmenistan, United States, Uzbekistan

Injuries: Argentina, Australia, Austria, Azerbaijan, Belarus, Belgium, Bulgaria, Canada, Colombia, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Latvia, Lithuania, Luxembourg, Malta, Mauritius, Moldova, Mongolia, Netherlands, Norway, Panama, Poland, Portugal, Romania, Seychelles, Singapore, Slovak Republic, Slovenia, Spain, Sri Lanka, St. Vincent and the Grenadines, Sweden, Switzerland, Thailand, Turkiye, Turkmenistan, United States, Uzbekistan

The money method — per €1,000 spent

Two methods, never added together. The materials method counts only what the product’s list of materials reaches, so it is a floor. The money method counts the whole supply chain of the industry for each euro spent, so in principle it misses nothing — but it cannot see what this particular product is made of.

In words

This page gives the model's estimate, for the reference year 2023, of deaths and injuries attributed to the supply chain, per kilogram of printed circuit board assembly, under the Economic-embedded scenario. The figure is a floor and counts the production phase only. On the material route, which follows only the raw materials in the bill of materials (39% of its mass is traced), deaths are 1.6 × 10⁻⁶ deaths per item(0.60–3.4 × 10⁻⁶)LOWS-MAX · run 2259 · methodology v1.0 · data dv-20261007-44r and injuries are 1.3 × 10⁻⁵ injuries per item(0.50–2.8 × 10⁻⁵)LOWS-MAX · run 2259 · methodology v1.0 · data dv-20261007-44r. On the monetary route, which covers the industry's whole supply chain, deaths are 1.4 × 10⁻⁶ deaths per €1,000 spent in its industry(0.94–2.2 × 10⁻⁶)MEDIUMS-MAX · run 2259 · methodology v1.0 · data dv-20261007-44r and injuries are 4.2 × 10⁻⁵ injuries per €1,000 spent in its industry(2.8–6.3 × 10⁻⁵)MEDIUMS-MAX · run 2259 · methodology v1.0 · data dv-20261007-44r. That corresponds to one death for every 290,000 – 1.7 million kilograms.

By raw material, Gold has the largest share, 94%. By country of harm, Sudan has the largest share, 45%. By conflict, Sudan: Government has the largest share, 38%. On the monetary route, Mining and quarrying has the largest share, 59%. These are separate views of the same total and do not link one to another.

Written by an AI language model (claude-sonnet-5-5 (Claude Code; reviewed by claude-opus-5-5)) on 2026-10-14, from the figures on this page. It wrote no numbers itself: each number was put in by the page from the calculation, and any text that wrote a number of its own, or mentioned anything its evidence does not contain, was thrown away unseen.

In human terms

One death for every 290,000 – 1.7 million items.

By the money method, about €450 million – €1.1 billion of spending in its industry for each death.

Rounded so that it stays within the range, so the sentence never claims more than the range does.

Under each scenario

Five answers to how much of the harm counted in a place is linked to its economy. Same product, same year, same trade table: only the rule for linking harm to the economy changes.

Deaths per item under each scenarioEach line is one scenario’s range, on a scale where each step is ten times the last; the dot is its middle result.10⁻⁷10⁻⁶10⁻⁵S-MAXEconomic-embeddedS-EVIDEvidence-weightedS-RESResource-onlyS-ARMSArms and resourcesS-CONSConservative

Where it comes from

By raw material — the materials method

By raw material — the materials methodEach bar is one contributor’s share of the deaths in this figure; the rest is grouped last.Gold — 94%Gold94%Copper — 2.7%Copper2.7%Silver — 1.6%Silver1.6%Tin — 1.1%Tin1.1%Tantalum — 0.11%Tantalum0.11%Nickel — <0.1%Nickel<0.1%Zinc — <0.1%Zinc<0.1%Iron ore — <0.1%Iron ore<0.1%Everything else — <0.1%Everything else<0.1%

By industry — the money method

By industry — the money methodEach bar is one contributor’s share of the deaths in this figure; the rest is grouped last.Mining and quarrying — 59%Mining and quarrying59%Manufacture of basic metals — 14%Manufacture of basic metals14%Water transport — 5.2%Water transport5.2%Land transport and transport via pipelines — 4.0%Land transport and transport …4.0%Air transport — 3.2%Air transport3.2%Manufacture of computer, electronic and optical products — 3.2%Manufacture of computer, elec…3.2%Crop and animal production, hunting and related service activities — 3.0%Crop and animal production, h…3.0%Warehousing and support activities for transportation — 2.2%Warehousing and support activ…2.2%Everything else — 5.9%Everything else5.9%

Where the harm was counted

Where the harm was countedEach dot is a country, placed at its capital, sized by its share of the deaths in this figure.Sudan 45%Burkina Faso 30%Mali 11%Mexico 8.2%Congo, Dem. Rep. — 2.7%Colombia — 0.64%Myanmar — 0.58%United States — 0.29%Canada — 0.29%Uzbekistan — 0.25%Russian Federation — 0.24%Brazil — 0.22%Papua New Guinea — 0.17%Australia — <0.1%Argentina — <0.1%Nigeria — <0.1%Ghana — <0.1%Indonesia — <0.1%Poland — <0.1%Peru — <0.1%Philippines — <0.1%Ukraine — <0.1%Ethiopia — <0.1%South Africa — <0.1%Sweden — <0.1%India — <0.1%Mozambique — <0.1%Bolivia — <0.1%Turkiye — <0.1%Guatemala — <0.1%Uganda — <0.1%Portugal — <0.1%Thailand — <0.1%Burundi — <0.1%Finland — <0.1%China — <0.1%Ireland — <0.1%Greece — <0.1%Germany — <0.1%Mongolia — <0.1%Bulgaria — <0.1%Czechia — <0.1%Pakistan — <0.1%Mauritania — <0.1%Venezuela, RB — <0.1%Romania — <0.1%Niger — <0.1%Iran, Islamic Rep. — <0.1%Afghanistan — <0.1%Slovenia — <0.1%Korea, Rep. — <0.1%Hungary — <0.1%Saudi Arabia — <0.1%Japan — <0.1%Slovak Republic — <0.1%Georgia — <0.1%Eswatini — <0.1%Bangladesh — <0.1%Norway — <0.1%Armenia — <0.1%
Each circle is a country, at its capital, sized by its share of the deaths in this figure. The largest are named; point at or tap any other circle to see its name.
Every material, in full
By raw material — the materials method
MaterialShareDeaths per item
Gold 94%1.5 × 10⁻⁶
Copper 2.7%4.2 × 10⁻⁸
Silver 1.6%2.6 × 10⁻⁸
Tin 1.1%1.7 × 10⁻⁸
Tantalum 0.11%1.7 × 10⁻⁹
Nickel <0.1%3.9 × 10⁻¹⁰
Zinc <0.1%1.3 × 10⁻¹⁰
Iron ore <0.1%7.0 × 10⁻¹¹
Coal and lignite <0.1%3.4 × 10⁻¹¹
Bauxite <0.1%1.5 × 10⁻¹¹

The chain, step by step

The largest step at each stage of the model. Each links to the full account, where every stage adds up to the headline figure.

  1. Printed circuit board assembly
  2. Gold 94% of the figure
  3. Sudan 45% of the figure
  4. Conflict and political violence 98% of the figure
  5. Sudan: Government 38% of the figure

The full breakdown →

Sources behind this figure

Every data file that went into this figure in this calculation, with the credit its licence asks for.

What this figure is not

Production phase only. Harm from using the product — driving the car, charging the phone — is not in this figure.

How to read these numbers →

Beside the most similar products

The two products whose deaths per kilogram are closest to this one’s, by the materials method.

Cite this figure

HumanToll 2026, run 2259, methodology v1.0. https://quiescence.eu/humantoll/product/circuit-board/

What it is made of

Each line is one part of one kilogram of assembled printed circuit board: how much of it goes in, how much had to be produced to get that much (some is wasted along the way), and where the figure comes from.

What it is made of
Component Quantity Source
Copper
ingredient 		214 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Cu 21.44 wt% (sd 9.14)
Tin
ingredient 		31.4 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Sn 3.14 wt%; solder
Steel
ingredient 		32.8 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · low
Fe 3.28 wt%, carried as steel
Aluminium
ingredient 		30.2 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Al 3.02 wt%
Nickel
ingredient 		10.3 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Ni 1.03 wt%
Zinc
ingredient 		7.30 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Zn 0.73 wt%
Silver
ingredient 		1.30 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Ag 0.13 wt%
Gold
ingredient 		0.400 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Au 0.04 wt%
Tantalum
ingredient 		0.100 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Ta 0.01 wt%
Lead
ingredient 		18.6 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Pb 1.86 wt%
Palladium
ingredient 		0.500 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
Pd 0.05 wt%
Non-metallic fraction — glass fibre, epoxy resin and other
ingredient 		653 g Mori de Oliveira, Bellopede, Tori et al. (2022), Gravity and Electrostatic Separation for… · medium
100% − 34.73% metals

Down to raw materials

The whole chain worked out — everything one kilogram of assembled printed circuit board needs, through every product in between, with all the waste along the way included.

Down to raw materials
Input Per item
Bauxite 121 g
Coal and lignite 19.4 g
Copper 214 g
Gold 0.400 g
Iron ore 37.2 g
Nickel 10.3 g
Silver 1.30 g
Tantalum 0.100 g
Tin 31.4 g
Zinc 7.30 g
Electricity, electric arc furnace route (not in the register) 0.0226 MJ
Lead (not in the register) 18.6 g
Limestone (not in the register) 7.07 g
Non-metallic fraction — glass fibre, epoxy resin and other (not in the register) 653 g
Palladium (not in the register) 0.500 g
Recycled steel (scrap) (not in the register) 9.86 g

An input marked “not in the register” is counted by weight but is not yet linked to any supply chain. Services such as transport appear as the euros, or tonnes carried a kilometre, that the list of materials gives them.

How much of it is known in detail

100% of this product’s weight is listed in detail in its own list of materials.

Its whole chain uses 1.13 kg of materials, waste included. 39% of that is raw materials the model knows — the part it can trace harm to — 61% is inputs not yet on the list, and 0% is an openly stated remainder.

In the table of what industries buy from each other it belongs to “Manufacture of computer, electronic and optical products” (FIGARO industry C26).

In the table of what industries buy from each other it belongs to “Manufacture of computer, electronic and optical products” (FIGARO industry C26).

What uses it

Products whose own list of materials includes this one.

Sources

Products →

About

Business-intelligence and B2B platforms, plus free privacy-first browser tools. Built and hosted in Europe.

Links

Privacy Cookies Terms Legal Notice
© 2026 QUIESCENCE.EU