At present, the motherboards in our computers contain around fifty different metals. But when our computers are no longer usable, what can we actually recover?
23 July 2026
Les cartes-mères de nos ordinateurs contiennent une cinquantaine de métaux, dont cinq au mieux sont récupérés.

Les cartes-mères de nos ordinateurs contiennent une cinquantaine de métaux, dont cinq au mieux sont récupérés.

© Unkas Photo - Shutterstock

The motherboards in our computers contain around 50 different metals, of which, at best, five are retrieved.

It all starts in our homes. During a spring cleaning, the broken-down laptop that’s been lying around is finally taken to the recycling centre and placed in the waste electrical and electronic equipment (WEEE) bin. If it cannot be reused or put to another use, it is sent for recycling. 

What proportion of this waste can actually be recycled?

The first step is done by hand: the electrical and electronic waste is sorted, and the computer goes into a container designated for computers. Next, again by hand, the battery is removed and sent to a specialist recycling facility — this is a regulatory requirement.

At this stage, there are several possible routes: we will assume here that the technician who removes the battery also separates the motherboard; and that the remaining components – the screen and the casing – are sent to automated sorting processes which will recycle those metals present in large quantities (iron, copper, aluminium), and sometimes the plastics.

Now, let’s take a look at the motherboard. It is regarded as a high-value circuit board because it contains significant quantities of copper and precious metals: more than 20 per cent copper, more than 150 milligrams of gold per kilogram of circuit board, and more than 600 milligrams of silver per kilogram. 

As such, scrap motherboards sell for between €3 and €8 per kilogram. The more advanced the board is (with more RAM or more powerful processors, for example), the more likely it is to contain significant quantities of precious metals.

The boards are sorted into different categories, from the highest-value to the lowest-value — once again by hand, after purely visual inspection. This is the grouping stage: the aim is to group together boards of the same category into a single batch. This will make transport, storage and resale easier.

When the consignment reaches a certain weight (around twenty tonnes), it is sold to copper smelters. These are metallurgy industrialists who produce and sell pure copper; they are known as copper “smelters”. There are currently fewer than ten smelters in the world that recover copper alongside precious metals. There are none in France, but the majority are in Europe (Belgium, Sweden, Austria and Germany). Alongside these “metallurgy giants” (such as Umicore, Aurubis and Boliden), there are also a number of SMEs in the recycling sector (including, in France, WeeeCycling).

Most smelter plants were originally designed to process copper ore (rock extracted from a copper mine). They have gradually incorporated waste into their processes, but this still accounts for only a small proportion of their overall intake. The metals recovered there are copper, gold, silver, palladium and platinum, if present (that is, just five metals, whereas the board contains nearly fifty).

Boards that are not “rich” enough are sent to “smelters”, who recover only the copper.

Why not recycle more?

On the economic front, it is clear that the sector needs to be profitable: the costs of processing this secondary raw material (motherboards) must therefore be lower than those of processing the primary raw material (copper ore).

Our products (such as digital devices) are evolving very rapidly, and some of the metals used today may no longer be in use tomorrow. Should we set up temporary processing lines? Just look at, for example, the recycling chain for energy-saving light bulbs, which have been replaced by LEDs over the past decade?

From a technological perspective, the technologies for recovering these fifty or so metals already exist in the mining sector, and many can be adapted for use in the recycling sector.

Is it possible to recycle very small amounts?

This range of recycling technologies cannot reasonably be applied to all fifty or so metals used.

In both ore and waste, metal production is organised into major metal groups: the steel industry for steel, the copper, nickel and aluminium metallurgical industries, and the rare-earth group. For example, in nature, gallium is found in bauxite (aluminium) ores, so it is produced by the aluminium smelting industry.

This production organisation, however, is of little use for electronic boards. A computer motherboard is processed using only copper metallurgy, and none of these facilities recovers gallium. The 3 milligrams of gallium contained in a board are therefore lost.

Beyond these issues relating to supply chains, there is also the question of how to recover value from a metal that is present only in very small quantities. 

What is the energy cost of extracting 0.075 milligrams of europium from a motherboard?

Determining the elementary materials

Finally, a third reason for the partial recycling of boards is the lack of data on their make-up. In fact, computers and their motherboards are not currently supplied with a “material passport” detailing their exact composition, even though this varies depending on both the manufacturer and the time of production.

If we are to establish new recycling sectors, we need to understand this resource base in order to confirm the economic and technical feasibility of recycling: how much of a particular metal is available, and at what concentration does it occur? Accessing this data is difficult and requires extensive research.

As part of the ReviWEEE project, Solène Touze and her colleagues at BRGM are developing a methodology for sampling component materials and assessing uncertainties.