Hillside Aluminium could get green $0.03c/kWh lifeline by 2030 – Mallinson

18th January 2022 By: Martin Creamer - Creamer Media Editor

Hillside Aluminium could get green $0.03c/kWh lifeline by 2030 – Mallinson

Independent energy consultant Clyde Mallinson

JOHANNESBURG (miningweekly.com) – It’s all about South Africa focusing attention on the generation and storage of its wonderful African sunshine and prime African wind.

It’s all about going hammer and tongs for these three letters, SWS, which stand for solar, wind and storage ­– and eight hours of storage is all we need, which is an exceptionally low amount compared with most other countries.

This will give us the clean, green electricity that we must have – with the excess there for drawing on when needed.

By generating excess green electricity, we can send the South African economy into a new orbit that will cut unemployment, reduce inequality and lower poverty.

You won’t need to wait for nightfall to recharge your electric car. The middle of the day will be just as good – and there will be green hydrogen for trucks, trains, buses, ships and big logistics, as well as for hard-to-abate industries and probably things we haven't even thought of yet.

Having surplus will bring with it near-zero marginal cost that will open the way for industries to spring up.

The plan may knock on the door for 1.5% of our gross domestic product but it will bring with it much more globally accepted electricity and much-needed growth.

This is what the detailed modelling of independent energy consultant Clyde Mallinson clearly shows. Mallinson spoke to Engineering News & Mining Weekly in a Zoom interview. (Also watch attached Creamer Media video.)

South Africa's current Integrated Resource Plan (IRP) has a mix of energies but if this is changed to comprise only SWS, that would deliver not only a load-shedding-free electricity supply industry but also create the platform for new industries to be built on near-zero-marginal-cost excess green electricity, Mallinson calculates.

His alternative SWS IRP would involve the deployment, between 2021 and 2040, of 40 GW of new wind, 230 GW of solar photovoltaic (PV) and 35 GW/290 GWh of storage, comprising mainly of battery energy storage and mechanical gravitational potential energy storage.

It would deliver electricity – including to the energy-intensive Hillside Aluminium which must go green – at a cost of $0.03c/kWh (R0.45c/kWh in rand at prevailing exchange rates), well below Eskom’s current wholesale tariff of about $0.07c/kWh, which is poised to rise to $0.15c/kWh by 2030.

If diversified mining company South32, which is working on options to secure green electricity at Hillside Aluminium smelter in KwaZulu-Natal province, is unable to source clean electricity, this high-carbon operation risks becoming uncompetitive in the international market over time, given the emergence of carbon border tariffs and end-user demand for green aluminium.

Owing to the aluminium smelter’s significant economic and social importance, South32 has commenced 'just transition' planning to support options for the smelter if its energy transition is not commercially viable and the smelter becomes uncompetitive.

Around 30% of the sales of Hillside – which has played a key role in KwaZulu-Natal’s economic development as one of the largest industrial employers in the region – are to South African customers, thereby supporting local downstream industries and employment.

Mallinson is of the firm belief that with SWS, Hillside Aluminium could get green factory gate energy at about $0.03c/kWh by 2030.

“I need to specify at the factory gate but I don’t mean at Hillside itself. I mean the resources supplying Hillside, wheeling on the Eskom grid, could do it at round about an average cost of $0.03c/kWh by 2030," he adds.

While that might sound impossible, he points to Bid Window 5 renewable energy prices and notes how batteries have decreased in costs by 90% over the last decade and that they are due to decrease by another 80% over the next decade.


“We have, undeniably, the best wind and solar resources in the world. Whether it’s wind or solar, it might not be the very best, but when you put it all together, it certainly is. In other words, the inter-seasonal differences between wind and solar, and the times of day when wind and sun complement each other, there’s actually no one who can beat us," says Mallinson.

“We’ve got exceptional resources and we need to grab those opportunities presented by those resources and move into a future that will be dominated by wind and solar and storage, and embrace this disruption that’s taking place in the energy sector,” he adds.

The SWS IRP deviates materially from the current IRP2019, which envisages a mix that is also dominated by wind and solar PV, but still includes new coal, gas and imported hydro.

Mallinson’s alternative IRP allocates far more to solar PV and storage than is the case in the IRP2019, with the wind and the storage designed to “see us through the night” and the capacity of the storage calibrated to the country’s current peak demand of about 35 GW.

The actual ratios of wind to solar PV could vary depending on future cost reductions, with the SWS IRP assuming a steeper decline in solar PV costs relative to wind over the period, reflected in the preponderance of solar PV in the IRP.

Such a system would not only adequately supply at least 231 TWh of conventional demand yearly but would also deliver a further 461 TWh of “superpower”, a term coined by independent US think tank RethinkX, or near-zero-marginal-cost clean electricity.

This "superpower", Mallinson argues, could be used to bolster the competitiveness of existing economic activities such as mining and smelting, while opening prospects for new activities such as electric mobility, the low-cost desalination of water, or the production of green hydrogen for use in hard-to-abate sectors.

The large-scale output of the SWS fleet would be three times that of the current coal-dominated system, based on Mallinson’s reinterpretation for South Africa of the ‘Clean Energy U-Curve’ used by RethinkX to model a 100% clean-energy system for America.

That U-curve shows that the lowest-cost system, or “sweet spot” for South Africa, would involve building an SWS fleet that produces 1.67 times the annual output of South Africa’s current fleet, at a cumulative capital cost of $71-billion. Such a system would deliver 155 TWh of so-called "superpower", while meeting yearly conventional demand of 231 TWh.

However, Mallinson’s IRP is premised on a system that by 2040 will produce three times that of the current system, as it would deliver 461 TWh/y of near-zero-marginal-cost surplus green electricity for an investment only $25-billion larger than the least-cost solution.

“For only 38% more investment above the least-cost sweet spot, you get a disproportionate amount of superpower,” he explains, describing it as the 2040 target for which South Africa should be aiming.

“We really need to be building wind and solar resources as if our lives and livelihoods depended on it,” says Mallinson, who advocates creating new SWS headroom ahead of retiring the coal fleet on the basis that “you can’t scrap your old car before you buy your new one if you are a travelling salesperson”.

To fully exploit the "superpower" that will arise, however, will require a shift from a demand-driven electricity system to one where surplus generation is embraced.

“The generation profile will dictate what the demand profile morphs into, with new demand adapting itself to those periods when the near-zero-marginal-cost excess green electricity is available,” he says.

Engineering News & Mining Weekly: But how should South Africa go about attaining competitive power tariffs?

Mallinson: The incumbents that have been in charge of this for 50 to 100 years, and that includes Eskom, are very reluctant to ever have surplus electricity. It’s seen as a major impediment. Before system modellers model, they say let’s limit excess and curtail excess to a minimum amount. That’s valid when operating in an environment when the excess is not stored for use when there is a shortage. At the moment, we have a demand system that is dominating our thinking and we are desperately trying to plug the demand with resources. We put in some open-cycle gas turbines and we’re being dictated to by demand. In the new system, the generation will set the pace and we will then have new demand entities like electric vehicles and green hydrogen, which will feed off the generation at times when there are surpluses. It’s a complete flip around of the way that we work at the moment. One of the most interesting things is that we can cover our current electricity needs, with no load-shedding, with a combination of only wind storage and batteries. But here’s the thing. We have to get through the worst periods in winter. There will be a stretch in winter of three or four days when there is perhaps not that much sunshine and, of course, winter has fewer sunshine hours anyway. So we design a system that can get us through that worst period and then anyone can see logically, in every other period of the year, every other day, more or less, we’re going to de facto have a surplus. We can choose to build bigger batteries, and often with hydrogen storage and other energy carriers, people are looking at inter-seasonal changes. As I said originally, we have a very muted inter-seasonal change in South Africa. We are renowned for our sunny winter skies and so although we have fewer hours of sunshine in winter, we tend to not have rain in winter. We can either build bigger batteries with less wind and solar, or we can build more wind and solar with smaller batteries, and we put those three into the mix and we see what is the cheapest combination of wind, solar and batteries that meets our needs for security of supply, as well as our annual energy. When you do that, you find that there’s a sweet spot, where that whole system costs the least, and at that sweet spot, we de facto have a very, very large amount of surplus electricity, because of our design to get us through that worst patch, which means we have surplus on the other days. But is it really a surplus, because we know that to combat climate change, we need to decarbonise transportation, heavy industry, industrial heating and cooling, all sorts of things, and so it’s that surplus that we’ll then tap into and make use of. In other words, in very simple terms, you will charge your electric car in the future in the day time, not at night. You will charge your energy storage systems in the daytime and discharge them at night, whereas at the moment Eskom has a surplus at night and it charges its pumped storage at night to release in the day. That’s going to turn completely. We’ll be charging the storage in the day to release at night, a complete switch. That’s where we’re at and in this new system surpluses are an unbelievable benefit, not something to be curtailed through modelling or any other means. Of course, when you’re not looking at lots of storage, then those surpluses appear to be a problem because we know that you need to use electricity pretty sharply. If you don’t store it, it’s gone. You use it or lose it. You can't inject more into the system than the system’s taking out at any point in time, otherwise you trip the system. That’s where we’re heading and I think it’s the most exciting thing that’s happened in the energy space in the last century.

What sort of storage is needed?

The interesting thing is that an organisation in America did a study for California, Texas and New England. They were looking just simply at battery energy storage and they found that they needed 39 hours of storage at a capacity that more or less matched their peak load. So the idea is that if we peak at 35 GW, we need 35 GW of storage and, if we were in California, we would need 35 GW for 39 hours. The good news is that in South Africa – because of our exceptional resources and the fact that they mesh so well between the seasons – we only need eight hours of storage. We still need 35 GW because you can imagine that if the wind is not blowing and the sun’s not shining and the system needs 35 GW, then your storage needs to be able to deliver 35 GW, but that’s very seldom because that’s our peak usage and it’s in the evening. Although the sun is generally not shining then, the wind’s generally blowing in the evening. The eight hours of storage we need is an exceptionally low amount compared with most countries. Besides battery energy storage, there are also other fantastic possibilities. We can use underground pumped hydropower, for example. So, instead of using new pumped storage hydro dams, we can actually utilise old gold mines, many of them pumping out water as we speak, including those that have closed because they have to keep pumping that water out. They are so deep that you can flood one of the cavities above where you’re pumping out at the bottom and essentially build a pumped water storge system underground in an old gold mine, and then you don’t have any of the civil construction costs of building a new dam. You’ve basically got your dams underground. The other example is taking old tailings or ash from coal plants and compositing those into large bricks and then lifting and lowering those bricks, which are much bigger than ordinary bricks, each weighing maybe 35 t. But you simply lift and lower them and when you’ve got surplus electricity you lift them, and when you want to get electricity back, you lower them. The nice thing about that is it helps you clean up old tailings dams, ash dams and settling ponds. At Rand Water, for example, where they deposit thousands of tons of silt every day, cleaning up the water that we drink, silt dams are filling up and they have to build new ones and we could take that silt and turn it into batteries.

Couldn’t the delayed Tubatse pumped storage scheme also throw a green energy lifeline to the carbon-intensive Hillside Aluminium, which is poised to face closure over time if it fails to gain access to competitively priced green energy?

I’ve had a look at Hillside Aluminium because when one studies the energy system, it attracts one’s attention because it uses 1 200 MW on a 24-hour, seven days a week basis. One must remember that Hillside Aluminium is a very useful demand response device for Eskom. Eskom is often asking Hillside to switch off at peak times and they can do that because there still an inertia in the molten aluminium, and as long as they don’t switch off for too long, and the aluminium freezes, they can actually switch off. They’ve obviously got to catch up that thermal loss that does take place. The other point is that I actually believe that Hillside Aluminium could get green energy by 2030 at about $0.03c/kWh, with storage at the factory gate. I need to specify at the factory gate but I don’t mean at Hillside itself. I mean the resources supplying Hillside, wheeling on the Eskom grid, could do it at round about an average cost of $0.03c/kWh by 2030. Now that might sound impossible. But what were the prices in Bid Window 5? At today’s exchange rates, $0.03c is about exempting the average of Bid Window 5 and I’m talking by 2030, and batteries are decreased in costs by 90% over the last decade and they are due to decrease by another 80% over the next decade. So, it’s actually an easy solution and it could start immediately by leveraging under-utilised Eskom pumped storage. Eskom doesn’t use its existing pumped storage to its full potential for the simple reason that it doesn’t have enough energy to fill it. If you’ve got to get through a week with a water bottle, you can’t drink it all on the first day. You’ve got to ration it over the week and top it up over the weekend. But if you know that you can fill your water bottle every day when the sun shines, you can then saturate your thirst every day and at the end of the day, if it’s a hot and long day, you can then pour that water over your head, if you want to, because you know you can fill up the next day. But if you’ve got to see yourself through a week of so-called rationing because you don’t have surplus, you can’t do that. For something like Hillside Aluminium, the storage alone is not enough. I often say and it’s a bit of a cliché now, that you can only store when you have a surplus. I’ve never seen anyone store anything when they haven’t got a surplus and I can speak from personal experience because I haven’t stored any money because I don’t have a surplus. It’s almost Biblical. You build storage when you have surpluses. We need to think of a surplus electricity system that allows us to store. At the moment we have a shortage and we’re having to plug the gaps because we haven’t been able to store power.

So what should we get going on?

The right thing is that we need to look at all the problems in the electricity sector together. We need to look at the non-payment from Soweto, we need to look at the communities in the pollution prone shadow of the coal mines, we need to look at them all simultaneously. My solution is that the population of South Africa should invest in a new fleet through the issuance of green bonds that can even have a bonus element. Instead of buying lotto tickets, you buy green bonds and the bonus element could be a monthly draw where you could win something as long as you brought a green bond. It would mean then that ten million South Africans would earn the new fleet by virtue of having financed it, and Eskom, municipalities and independent power producers (IPPs) would run and manage that fleet on behalf of the owners of the fleet, the South African population. If you play lotto, if you win R200-million it probably makes you a marked person, when the relatives pitch up, but if you win R2-million, it will change your life and you can maybe stay under the radar, and you can invest that R2-million in more green bonds so that you are building up your own pension. It’s actually possible to over a period of 15 years to move ten million South Africans from Sassa grants to renewable energy dividends.