In the wake of my illness and enforced period of rest I found the time to dive into three very different books on the history of the iron industry and put together my own thoughts on the possible future for this central substance in our evolving civilisation. I will briefly cover the most interesting points from each book, then round things off with my own speculations.
“Iron Making in the Olden Times” (Nicholls, 1966) is a reprint of two books written in the mid 1800s that detail many aspects of the deeper history of the forest of Dean near the British border of Wales. This region was exploited for its timber, stone and iron ore since Roman times, leaving enormous hand dug caverns throughout the territory. The combination of critical resources made it a major centre of iron production for the last two millennia. As medieval workers developed more efficient smelting techniques the residue of Roman activities was resmelted to extract more metal. Coal was first used in 1200 as sea-coal (washed ashore from under water formations), later dug by hand in the 1300s. By the 1600s this energy source was beginning to substitute for wood derived charcoal for the smelting of iron, though needed to be charred into coke first to drive off unwanted substances. The shift to this new energy source removed one major resource limitation to iron production. Steam engines began to substitute for waterwheels to run the blast furnaces in the mid 1700s, removing the final environmental limit on production. From this point on iron output increased dramatically, leading to local deforestation since wood was still used extensively for structural purposes in the mines. Eventually water pumping costs at ever greater depths made mining unprofitable and the last iron ore mines closed in 1946, then the last coal mines in 1965.
“Iron Men” (Waller, 2016) explores the enormous impact of the Maudslay factory in catalysing the industrial revolution. In 1800 the British navy used 100 000 pulleys, each hand crafted and prone to failure due to irregularities (which could cripple an entire sail ship). By 1808 the first modern assembly line factory was constructed by a young Henry Maudslay, replacing 100 skilled craftsmen with half a dozen unskilled men and boys to operate the high precision machines. The last of these machines only stopped operating in 1968. A rope factory built around the same time continues to use their original equipment to this day.
Henry Maudslay was an obsessive and creative engineering genius who was responsible for perfecting and standardising the design and creation of a wide range of basic interchangeable parts like nuts, bolts and screws that made the mechanisation of the industrial revolution possible. The first machines featured handmade fasteners, such that every single bolt and screw needed to be individually labelled in some kind of bespoke Ikea instruction manual from hell. Maudslay’s bench micrometer was capable of measuring components within 0.00079 of an inch in a world accustomed to measuring to 1/32th of an inch at best.
It is interesting to note that the physics of many of Maudlay’s machines was not fully understood until the 20th century, much like the advances in metallurgy that happened around the same time through empirical experimentation. All of this explosion in creativity was made possible by the collapsing prices of metals and energy that began in the mid 1700s, which opened the door to prolonged periods of tinkering and wastage with materials which had once been in short supply.
The final book I read was “Still the Iron Age” (Vaclav Smil, 2016), a tour de force journey through the iron and steel industry overflowing with data and analysis that mere mortals rarely get to see. He starts by exploring the transition from the relatively easy to smelt metals of the Bronze age, through a prolonged transition to the more technically demanding smelting of iron, a process which often took a millennium to achieve. Iron production increased massively after 1800 by harnessing coal instead of wood, followed by an increase in annual per capita steel consumption from 20 kg in 1900 to 230 kg in 2010 due to improvements in processes for transforming iron. The falling cost of iron and steel, coupled to the production of a range of specialty alloys, supported the mechanisation of virtually every major industry.
Medieval iron production placed a huge burden on local forests, consuming the output of 10-50% of wood production in different countries. Well forested countries like Russia and Sweden dominated iron exports before industrialisation for this reason. Only Brazil produces iron from charcoal today, in numerous small smelters which rely on illegal deforestation of the Amazon. Once the dependence on coal was realised the centre of iron production shifted from Britain, to the USA, to Japan and finally today to China. Most of China’s stupendous overproduction of iron has gone into reinforcing concrete buildings to support their utterly insane housing bubble. Global iron and steel companies have experienced a persistent economic slump due to overproduction and overcapacity in recent years, with a rash of mergers doing little to improve economic performance. This consolidation into a smaller number of megacompanies mirrors the earlier trend of building ever larger smelters to approach the thermodynamic limits for efficient metal production.
As a result the world is shifting to rely on the recycling of steel and iron for ever more of its needs, up to 57% at present. The magnetic nature of the metal makes recovery from mixed waste streams relatively simple. Despite being 5% of the earth’s crust, remaining economically useful iron deposits are heavily concentrated in Australia and Brazil. The more conveniently located resources in Eurasia and the USA have been mostly exhausted by a century of production. Remaining iron resources are projected to last another 27 years at current consumption rates, though the broader category of reserves should last 160 years. The economic accessibility of a geological formation depends on the current economic environment, which translates into availability of useful surplus energy. Just as the rate of production of medieval iron was limited by the amount of charcoal and water power available, iron reserves today are limited by the amount of coal and oil on hand. Current operations rely on access to refined diesel to operate the giant machines which dig and transport iron ore and coal. Geopolitical realities mean that 69% of the world’s iron ore is transported vast distances to be smelted, something that was impossible in the pre-industrial era. Declining ore quality means ever more energy is used in various purification techniques before transport and smelting.
The energy needed to produce a ton of iron has dropped from 300 Gigajoules in 1750 to 20 GJ today (close to the theoretical limit of 10 GJ). Australian sourced iron is 40 GJ/t due to the extra transport costs. This suggests energy efficiency of iron production has passed its peak as ore needs to be sourced from ever more remote locations. By comparison aluminium is an order of magnitude more energy intensive at 200 GJ/t, plastics need around 100 GJ/t, paper 30 GJ/t, glass 10 GJ/t, lumber and cement under 5 GJ/t. Recycled metal consumes only 1 GJ/t by comparison, so it isn’t surprising this is coming to dominate total supply.
Of the 51 Gigatons of steel produced between 1850 and 2015 Smil estimates 10% was oxidised or destroyed, 25% recycled, 15% embedded in structures, leaving a remainder in accessible steel stock of 15 Gt, or 2.5 t/capita. Per capita steel consumption has been falling in recent decades as lighter metals, plastic and carbon fibre replace various uses.
Reflections and Predictions
I’ve been wanting to delve the history of the iron industry for so long, so thank you Coxiella burnetii (Q fever) for the enforced holiday and opportunity to do extra reading. The first thought was the observation that both copper working and then iron working often relied on people first finding rare examples of native metals. Copper sometimes forms metallic nodes on the surface (such as the formations that supported a short-lived copper working culture among the indigenous North Americans) or in the case of iron in the rare ferrous meteors that found their way into ancient Egyptian artefacts. Working metal is a lot easier than smelting it from ore. In some way this fits my mental model of accidental domestications of species needing very little modification coming before the much harder work of shaping a biological system more extensively to bring it into symbiosis. In this regard I see humans still at the “banging a metal meteorite into an amulet” stage of mastering biology.
Smelted metal can in many ways be viewed as the secretions of a very refined form of the original domesticated life form, that being fire. The use of fire to transform matter progressed beyond cooking food, to the hardening of wooden tools, then ceramics, then finally to ever more complex metals. Aluminium was the last major metal added to our repertoire, used to clad the top of the Empire State building during a time when it was vastly more expensive than pure gold.
The dependence of mechanisation on standardisation of parts was an interesting reflection of the reductionist mindset of the era. Even mechanical problems were broken down into a discrete series of steps, with the basic solutions like interchangeable screws forming the foundation for the next level of problem solving. Despite this mindset, the key advances of this incredible period in human history were drawn from imagination, inspiration, persistent tinkering and plain old serendipity, with theoretical science often taking more than a century to catch up with widely applied technologies. I believe a similar pattern will play out in the coming era of biotechnological advances, except this time the people nominally driving the process are attempting to apply a theory first approach, meaning they are likely to fall behind the tinkerers and the dreamers.
Getting back on the topic of iron and steel, it is interesting to speculate how current trends and capacities will respond to a new era of economic contraction. The current smelting capacity is excessive, but much of the smelting infrastructure is reaching the end of its projected lifespan. Investment in new smelters is virtually zero, though recycling of existing metal stocks is rapidly growing to dominate the market. The quality and accessibility of remaining iron and coal reserves continues to drop, and presumably one or the other will eventually cross invisible thresholds for viability, just as the miners in the forest of Dean found that pumping water from beyond a certain depth was never profitable.
This means the most likely near-term future is a world of declining population size and per capita iron and steel consumption. If this dynamic is combined with the enormous reserves of metal and the stunning energetic efficiency of recycling it over smelting new metal, then it is reasonable to project a complete cessation of iron smelting for multiple human generations. The same dynamic is likely to play out with other major metals like aluminium to a greater extent due to the enormous energy consumption during smelting.
The continuation of this dynamic over multiple generations during the downslope of the industrial era may set up a rare opportunity. Damascus steel is a particularly beautiful form of the metal, produced in India many hundreds of years ago. All the worlds’ scholars, using the most powerful tools of industrial science, took generations to rediscover the forgotten methods used to make this product. A similar pattern played out with the loss of the recipe for various forms of Roman concrete, which display superior properties over their industrial counterparts. All of this is to say that large, complex and highly technological civilisations have a long history of forgetting basic technological processes.
After a century of getting by on scavenged and reforged iron and steel, our descendants may well find that they have forgotten how to smelt fresh ore by the time their inheritance of pre-smelted runs out. They may also find that the surface iron ore, coal or wood, and water power resources no longer coexist to even support a medieval level of iron production.
We may well be living through the end of the iron age.
The implications of this suggestion are profound, given virtually everything we think of as “technology” relies in some way on access to iron. Further than that, virtually everything we think of as “agriculture” (including pre-industrial agriculture) also relies on access to a range of simple but indispensable metal tools.
With proper preparation this transition could represent the greatest opportunity for the improvement of our species and the planet in the last fifty thousand years. The opportunities and alternatives to an iron dependent society are something I will be exploring in a future post, so this will have to do for today. In the meantime, I will encourage you to unleash your imagination to explore what a world without iron might look like.
Just think. Before reading this essay, the deepest I'd thought about iron was "pumping iron" and curing my cast iron pan. I'm always impressed by how you can transform something seemingly mundane into an entertaining journey. Great work!
Q fever, nasty. And a bit tricky to diagnose.
And then there’s this: https://www.pbs.org/wnet/religionandethics/2003/10/24/october-24-2003-operation-whitecoat/15055/