Recovered, unrecovered or unrecoverable? The fate of end-of-life steel

Is the steel in the Titanic scrap? Or available for scrap? Is it unrecovered, or unrecoverable?

We know roughly what end-of-life scrap is. It’s defined in ISO 20915: 2018 as ‘scrap from after the end of life of final products‘ (see ‘What is scrap?’). But that little word ‘after’ is doing some heavy lifting.

Does the steel in a product that has reached the end of its life in use become end-of-life scrap immediately – or is it just ‘available for scrap’ – or perhaps ‘potentially available for scrap’ until it has actually been recovered? Is the steel in the Titanic available for scrap? How about a redundant railway line? Is the steel embedded in concrete in the foundation of a building scheduled for demolition recoverable? If it is not recovered immediately, is there a point at which we should consider it unrecoverable, and no longer even potentially available for scrap?

The figures below illustrate two ways to think about recoverability.

End-of-life steel recovery – model (a)
End-of-life steel recovery – model (b)

In model (a) all end-of-life steel is considered to be at least potentially recoverable – and so potentially available to become scrap. Some is quickly recovered, whether for re-use or recycling. And some is not initially recovered – it joins an accumulating stockpile of unrecovered end-of-life steel. Steel from this stockpile might be recovered and returned to the recycling stream eventually (shown with a ‘*’), or rust away, or simply remain where it is. It is sometimes described as ‘hibernating’.

In model (b) some end-of-life steel is defined as being unrecoverable from the start. By definition, that amount is not even potentially recoverable, and hence never even potentially available to become scrap. Just as in model (a) the rest is potentially recoverable – although it may or may not be recovered in practice.

The models imply an important difference in the calculation of the recycling rate. In both models the recycling rate is the amount of end-of-life steel that is actually recovered and recycled, as a proportion of the potentially recoverable end-of-life steel. But in model (a) the amount of potentially recoverable end-of-life steel refers to all steel reaching the end of its life in use, whereas in model (b) any end-of-life steel deemed to have been unrecoverable has already been subtracted.

In either model we need to exclude any additional scrap that was recovered from the end-of-life steel stockpile – but we’ll come back to that.

Whether we are using model (a) or model (b) makes a big difference to the claimed end-of-life recycling rate. Suppose that in a given year there is one million tonnes of steel in products reaching the end of their lives in use, and that 800,000 tonnes of end-of-life scrap is recovered and recycled. The recycling rate on the basis of model (a) is 80%. But if 10% of the original end-of-life steel was pre-excluded from our estimate of potentially recoverable end-of-life steel on the grounds that it was unrecoverable, then we have an end-of-life steel recycling rate of 89%. – significantly higher.

Such differences are not just theoretical – according to the Dovetail report ‘Understanding Steel Recovery and Recycling Rates and Limitations to Recycling’ this partly explains wide discrepancies between estimated recycling rates for the USA. For example, a recycling rate of 89% reported by the Steel Recycling Institute, having pre-excluded unrecoverable scrap, compared to a recycling rate of 59% estimated by the US Geological Survey (59%) which didn’t.

Is one or other approach more appropriate? Well, yes. There are several reasons why model (a) is preferable. The first is that model (b) is misleading. If we say that the ‘end-of-life steel recycling rate’ is 89%, then that clearly implies that just 11% remains unrecovered. And if we also assume that at least some scrap is unrecoverable we would probably conclude that recycling rates are about as high as they ever can be – which could well be wrong.

That should be reason enough to choose model (a) already. But there are others.

The issue of recoverability is ultimately a social and economic question. All of the metal in any end-of-life product is potentially recoverable, if money and social constraints are no object. That is true of the steel in the Titanic, the steel in cans and domestic goods buried in landfill, and even steel contaminated with radioactive material. But whether it is recoverable in practice depends on factors like the quality of the scrap, its local market value, the cost of collecting it and, importantly, the price of iron ore as an alternative raw material. One consequence of this is that it is not possible to assign any fixed value to the proportion of end-of-life steel which is recoverable or unrecoverable: the proportion will change over time, and will be different in different places depending on the economic context. Another is that end-of-life steel that was considered to have been ‘unrecoverable’ one year may in fact be recovered in a later year if conditions change – in other words, it wasn’t unrecoverable at all, it was just ‘hibernating’.

So back to the Titanic. Unrecoverable? Almost certainly. But it’s an economic and social question, and we don’t need to make that call. We just need to know that it is unrecovered, and is for now part of the world’s end-of-life steel stockpile.

What are the conditions under which it might be recovered and re-enter the steel cycle in some year in the future? That’s a question of elasticity of the end-of-life scrap supply. And that’s for another post and another day…


Categories:

,

Tags: