What actually makes lead-acid battery recycling clean

Andreas Manhart is a senior researcher at Oeko-Institut, where he has spent more than a decade studying the recycling of used lead-acid batteries (ULAB) in Africa and Asia.

He has walked through more low- and high-standard smelters than almost anyone, and his team’s field assessments and standard operating procedures have shaped how regulators in several countries think about the problem.

We talked about how he stumbled into the field through e-waste, why ULAB recycling pollutes the way it does, what separates a clean plant from a dirty one, and where he’d point limited enforcement capacity first.

From e-waste to batteries

How did you start working on lead-acid battery recycling?

Fifteen or twenty years ago, the big topic in Europe was e-waste. A lot of it was being shipped, used or as waste, to African countries. We were active in a couple of projects in West Africa, looking at where it was going, what the drivers were, what was done with it, and how it could be improved, all with African partners. A lot of it was being reused, which made it hard to ban, as under WTO rules you can’t ban a functioning device crossing a border.

At one point, a partner from the environmental authorities put us in his car and said, “I’ll show you something.” He took us to a lead smelter.

It was eye-opening. We’d been to all these informal e-waste sites, and suddenly we were somewhere where you thought, “if I would have to earn my living in either e-waste or ULAB recycling, I’d go back to the informal e-waste market,” because this was so obviously terrifying.

So we did a small fundraising campaign and reached out to African NGOs to ask whether this was a systematic pattern. We teamed up with five organizations, all of which are still active in the field today, and it turned out the answer was yes. It was very similar across many countries. That became the report The Deadly Business. It’s old now, but unfortunately many of the findings still hold.

From there, every e-waste project also took a side look at batteries, and then the solar projects came along with their own battery disposal problem. We developed the subject from project to project.

When was this?

We published around 2014 or 2015, so we started a year and a half earlier — 2012 or 2013.

Were many people working on ULAB recycling back then?

Not at all. There were the Basel Convention technical guidelines and a handful of people who knew about it, but the whole view that everyday products can cause severe harm in countries where they aren’t regulated came up first with pesticides, then with e-waste, and only then did ULABs land on the radar.

We tried to convince donors, but they kept saying, “This is the old battery — look at lithium-ion, that’s the modern subject.” It was a hard time finding project money, but we strung together a series of projects over the years.

ULABs get a lot more attention now than they did then. What changed?

Epidemiologists found that lead is this under-recognized problem in low- and middle-income countries, which almost nobody was aware of ten years ago. People thought about drinking water, air pollution, HIV — but not lead. I can’t trace exactly when it climbed up the rankings into the top DALY factors, but suddenly it was there.

In OECD countries we looked at lead back in the seventies and eighties, fought our own problems, and then assumed it was resolved. We didn’t really look at what was happening in LMICs.

Why ULAB recycling pollutes

Why is ULAB recycling such a problem in many LMICs?

From a value side, it’s high-value scrap. Roughly two-thirds of a battery is metal you can recover and sell, there’s a market for it, and it doesn’t take rocket science to extract — you break open the battery and take and process the lead fraction.

The value is quite consistent over time, and in environments where people are desperate for income, that’s an activity. We’ve seen investors jump onto it in the last decade or so, doing urban mining for lead. And as soon as you start taking things out of the battery box, you have a leakage and exposure risk.

Walk me through the different kinds of recycling and how they cause pollution.

In the early informal clusters we saw people breaking batteries open, taking the plates out, melting them in open pans, casting ingots, and shipping the ingots. We don’t see much of that in Africa anymore, for a technical reason: roughly half the lead in a battery isn’t elementary — it’s lead oxide and lead sulfate. To turn that into a sellable commodity you need more than an open pan. You need a smelter, with energy and a reducing agent, to convert it back to elementary lead.

That’s the opposite of e-waste. With e-waste, the informal operators who separate value from non-value always have an economic edge. With ULABs, if you bring equipment and a smelter to the ground, your efficiency is much higher. You’re still polluting — the potential is still high if you don’t do it properly — but what you can extract from one battery is so much greater.

From what I’ve seen in Nigeria, you used to have backyard, small-scale recyclers, and over time they got out-competed by bigger, low-standard firms with a furnace but few pollution controls — so a handful of bigger, very high-polluting firms becomes the equilibrium. Is that right?

Yes, that’s what we see in the mass market.

There are still niches for the small guys. A craftsman who needs solder paste finds the cheapest source is buying batteries and melting the plates. Someone making fishing-net weights or dumbbells needs something heavy to work with, and battery lead is the easiest source. So informal melting still occurs in those niches.

But you can’t consume all that volume in dumbbells and fishing weights. It just makes sense to sell it to someone who offers more than you could recover yourself. That’s what we observe — the backyards switched, over the years, to supplying [by selling scrap batteries to] the bigger operators.

How does low-standard ULAB recycling actually generate pollution, step by step?

It starts the moment you do anything with a battery. Most smelters don’t want the acid — it corrodes their equipment — so they ask suppliers to drain it off first. But when you drain the acid, you wash out lead particles. Batteries often fail because lead sludge builds up and shorts them, so pouring out the acid pours out some of that sludge. That’s the first point of pollution.

Then there’s transport — a shaky open truck with already-broken batteries loses material. In the plant, everything from removing the lead to all the mechanical handling produces dust, and the dust is what harms.

During smelting, a furnace above a thousand degrees evaporates lead, which condenses into microfine particles in the off-gas stream — so if you don’t control it well, you get massive emissions.

And finally there’s the slag. Every smelting process produces slag with some residual lead. Getting it down to three or four percent would be possible, but some have more, and the slag is of no use, so it’s dumped somewhere — and that’s where more lead goes.

Why does the slag contain lead?

The slag is the top, non-usable residue phase that comes off the furnace, capturing sulfur and other residues. In the rotary furnaces we see in Africa, it’s a soda-iron slag that decomposes and can’t be used safely — it’s hazardous waste.

There are ways to reduce how much lead ends up in it, or how much slag you generate at all, but the whole pattern follows “make as much money as possible while not investing too much.”

That’s the economic equilibrium, and it’s why they pollute. Optimizing for emission minimization is possible, but it’s the extra mile that costs a lot of money without giving you much more lead in return.

Do we have a sense of how much pollution each step creates relative to the others?

We’re doing some of that math now, so we have an idea, but it depends on the process chain — what kind of smelting, and crucially the filter fabric and whether it has holes burned in it by sparks, which is common. With a hole, stack emissions are at a completely different level.

By volume, slag is quite relevant. But some streams are small in volume and extremely risky — for example, take-home lead on workers’ clothes. Most plants we have seen in Africa don’t have in-house laundries, so workers wash clothes themselves or just go home dusty. It may not be much lead per day, but every day adds up.

The same goes for the plastic from the battery cases — it’s shredded and sold to the local plastics manufacturers, often without proper washing, with lead oxide still visible on it. Small volume, but you never know where it ends up — probably furniture or kitchenware.

Pollution controls

Explain how the controls around smelting work.

First you have to capture the off-gas: develop enough suction, then cool it down and slow it. You have lead vapor in the system that must condense back into filterable particles, so you need cooling towers, basically a series of pipes that bring the temperature down.

Once it’s cool enough, it passes through a filter medium, the baghouse. You normally run two baghouses in parallel, so if one has trouble you switch over. This needs good maintenance and the right filter fabric.

After that it goes to a wet scrubber — an alkaline solution that washes out the sulfur — and then to the stack. Ideally there’s a monitoring system tracking how the baghouse performs and how much reaches the stack.

And the captured dust goes back into the furnace?

The captured dust is essentially lead oxide, a powder that can be recharged into a proper furnace. It’s hazardous and has to be enclosed and managed well, but with a good modern rotary furnace you can run a closed-loop system. It’s done in many countries, including Germany and the US.

Even good plants are no rocket science — it’s a well-developed process. What it requires is well-maintained, well-monitored filter plants. We often see filter plants that aren’t monitored at all.

Furnace types

How much does the furnace type matter?

The simplest type, which goes back to Roman times, is the blast furnace. It’s sort of like a bakery oven with a stack. You load the charge, reducing agent, and fuel together and scoop the lead out the bottom. It’s almost impossible to control its emissions. The most common modern type is the rotary furnace, which gives you a lot of advantages, including recycling the furnace dust. But having a rotary furnace doesn’t mean things are fine. You can mess up the best furnace and create a lot of pollution.

If a recycler had no pollution controls at all, would you still rather they have a rotary furnace than a blast furnace?

No pollution control is just a no-go. It’s such a terrible emission point. Even a blast furnace can and should have a baghouse.

Right — but everything else equal, which is worse?

On one end of the scale, with a rotary you can do quite well. With a blast furnace it’s theoretically possible to reach a high standard, there may be a case here and there in the world. but it’s extremely difficult.

I’ve heard some Nigerian recyclers adopted rotary furnaces for economic reasons, not just pollution control. Is a rotary more profitable?

It’s partly culture — in South Asia rotary furnaces are well recognized. They’re useful because they’re batch processes; depending on your battery supply, you can stop them after a few batches. Other furnace designs force you to run for weeks on end, which is a management challenge if you do not have a stable ULAB supply.

But I wouldn’t lobby too hard for one furnace type. There are other types that can demonstrably be well controlled — the reverberatory furnaces used in China is apparently well controlled there, though I haven’t checked those myself. A rotary is the backbone of a scalable, versatile operation, but it’s not the only answer.

The economics

Walk me through the economics of a recycler in one of these countries, and how a new entrant thinks about what equipment to buy.

Think from the end. If you’re a recycler, not a battery producer, your product is refined lead or a lead alloy, priced off the London Metal Exchange. You can add a bit of value through alloying — antimony alloys for battery production, for example — but your product is traded internationally and you get essentially the same price as your competitors. So you model what you can produce against the LME price to know your revenue, and that has to cover everything upstream.

You don’t usually collect batteries yourself. You buy them from collectors, even under deposit-return systems. So you spend money buying batteries, you earn money selling lead, and the spread has to cover energy, maintenance, labor, and eventually the investment. In countries with strict regulation and emission control, high-standard plants are profitable — you don’t need a subsidy.

The danger is competitors who pay collectors more than you can. If you don’t have that competition, you can run very ambitious emission controls and still pay your suppliers a reasonable price.

What’s the cost difference between a low-standard plant that has reached some economies of scale and a genuinely high-standard one?

The low-standard investors bring in the equipment they need to drive up lead recovery — to get to ninety, ninety-five percent. The last few percent, plus a genuinely safe plant, is proportionally more expensive.

In the paper we just published, the average plant we currently see in African countries — with an output of ten to twenty thousand metric tons of lead per year — comes in around $3–8 million. The same plant operated responsibly is around $13.5–27.5 million. The good plants recover a bit more lead, but not enough more to close that gap.

How much of that is fixed capital cost versus ongoing operating cost? If you got a baghouse for free, how much cheaper would clean operation be?

Those figures are upfront investment, the hardware. But the same decisions recur in operation.

Maintaining a good standard means constantly choosing to do the maintenance, keep the workplace clean, run the dust-control steps. If a plant manager wants to prioritize throughput and save costs, they neglect the monitoring, the baghouse filter changes, all the small daily steps that hold emission control together.

So it’s both — and the result is a plant that gives a reasonable lead return while polluting enormously, and is still more profitable because it still recovers most of the lead.

What does the profit-margin difference look like?

What we studied in Nigeria is the input side. Polluting recyclers can pay collectors maybe 10 to 20% more for batteries, because they have more money left over. So instead of a thousand US dollars a ton, a battery collector might get $1,150. Collectors sell to whoever pays most, it’s a logical business decision.

Engitec built a smaller-scale plant design a couple of years ago. Are scaled-down clean plants relevant if they can’t operate profitably against informal competition?

Those attempts are extremely important. What counts as a medium-sized plant in Africa would be small-scale in Germany or the US — the markets are smaller, so the plants are smaller.

An equipment manufacturer designing tailor-made solutions for medium-sized plants is very useful. I know Engitec had one under construction recently; I don’t know its current status. Whether it breaks even depends on its competition and, ultimately, on regulation and enforcement in the country.

Walking into a plant

When you walk into a recycling plant, what are the first things that tell you whether it’s likely to be complying?

Before you even arrive, watch where you are — populated area, industrial site, or empty land? What’s the neighborhood?

Then your first impressions: is the facility orderly and clean? Then you wait for a safety induction. If the managers walk you in wearing sandals and no protection, with no warning signs, the alarm bells are already ringing. This is heavy industry — heat, pollutants. If you’re not told to wear a mask, stay with the guide, and what to do if there’s an alarm, you already know the safety culture isn’t well developed.

I always hope for a positive surprise. There have been a few. Most of the time there isn’t. Very often, just from the battery offloading and breaking area, you know it’s not the best plant.

Does one badly-done step predict the rest? If the breaking is all manual, are the odds worse that the baghouse is run correctly?

In about eighty percent of cases, yes. There’s a company culture that carries across. We’ve seen plants where some sections were done amazingly well and others so crudely it was a sharp contrast, but that’s the exception.

If you’re a regulator with very limited enforcement capacity and capital-constrained recyclers, which parts of the process matter most to upgrade first?

I wouldn’t name just one or two, but there are some points you may deprioritize.

Early on we had whole checklists, down to how many fire extinguishers there were — and some regulators started trading twenty more fire extinguishers against a new baghouse. That’s not the intention.

So: there are core backbones that are non-negotiable, core management steps, and then things that matter but shouldn’t be traded against the backbones.

What are the backbones?

It’s not only hardware. For example, a factory hall with a smooth, orderly floor and a system to capture liquids, so you can do regular wet cleaning for dust control. Lead dust buildup is a key concern. Many plants have completely broken floors, everything stored everywhere — the dust settles and there’s no way to capture it. A good plant captures dust inside the facility and in the immediate outside area, with regular wet cleaning and damp floors.

So it’s not only the baghouse, the furnace, and the fume hoods — though those matter. Dust control is another big one.

Stacks and baghouses

A US EPA report I saw suggested a large share of lead loss happens during smelting, and people at a major equipment manufacturer told me that if they could mandate one thing, it would be the baghouse and smelting controls — partly because of stack height and how far the lead spreads. Fair?

It’s certainly a major hotspot, especially because of the particle size and the distances those particles travel for population exposure — most likely yes.

For occupational exposure there are more points. And something often ignored: we see different patterns from industrialized countries. Decentralized battery breaking and acid drainage, for instance. If you tell suppliers to deliver batteries drained, they break them in urban collection areas to cut ten to twelve percent of the weight.

And slag dumping — rotary-furnace slag breaks down into a reddish material that looks like tropical soil, so it gets dumped in a field, but it can be three to ten percent lead. Those patterns weren’t researched in Europe or the US, because we never had decentralized breaking or dumping at that scale. But yes — stack emissions are certainly a hotspot for the wider population, if you have them.

What does a baghouse cost?

Many are self-manufactured. In Germany every sawmill has a baghouse, it’s not lead-specific. So there are quite cheap options. But it must be properly designed, use the right filters, run the right temperature regime, and be maintained. You have to stop sparks burning through the bags, replace the bags, monitor for holes or blinded filters, and manage the captured dust properly. It’s not just the baghouse itself.

Should regulators worry about maximum stack height?

There are old established rules — in some former British colonies I think they kept the old requirements with a hundred-foot minimum stack height — from the days when you hoped a tall stack would spread emissions as far as possible. By today’s logic you’d do the opposite: make the plant footprint large, fence-control it so there’s nothing nearby, and keep the stack low so emissions don’t travel far.

But honestly, not much should be coming out of a well-run stack anyway — mostly hot air, CO₂, and moisture. So I wouldn’t treat stack height as a big lever. A low stack plus fence-line air monitors and a larger plot sound more appropriate today, but I wouldn’t prioritize it highly.

Relocation and industrial zones

How do you think about relocating recyclers where enforcing controls is very hard? Jenna mentioned the Dhaka government convincing a big recycler to move five kilometers outside the city.

A dirty plant in a central location is terrible for the whole city. But in many African countries, and I assume also Asia, smelters built ten or fifteen years ago outside the city are now inside it, because the city grew.

Many first-generation smelters in Africa are now inner-city. Newer ones tend to be built outside. So it’s not only relocating; you need controlled industrial zones with a clear, enforced non-encroachment policy. Africa has fast-growing cities and populations — you have to think about what the country looks like in ten or fifteen years.

There are two perspectives. Yes, you should move them out. But in Europe, with hundreds of years of industrial development, we often have smelters next to populated areas, monitored so strictly it’s possible, though still risky. I know one plant in Germany where they asked the managers to live next to it, with their families.

Monitoring

What does standard monitoring look like for a recycler in Nigeria or Ghana today — and what does best-in-class look like?

In every LMIC I’ve worked in, setting up an industrial plant requires an environmental impact assessment first: approval covering emission controls, plant layout, the stack. That’s one reason almost all plants have a baghouse: the authorities require it. A system with just a furnace and a stack, I’ve seen that in Myanmar years ago, but not in Africa.

Then the license has to be renewed periodically, with inspections from environmental, occupational-safety, sometimes health and fire authorities. They can extend or withdraw a license, often conditional on improvements. That’s the inspection-based monitoring we have in most African countries.

The problem is that inspectors cover every kind of industry — warehouses, power plants — so they’re trained for general assessment, not lead-specific assessment. Some write “regular maintenance is needed,” which isn’t specific enough to enforce. You need to say exactly what a plant must change. That’s where a lot of authorities need more know-how.

We shouldn’t replace inspections — we need more layers. In Germany and many European countries you have continuous stack monitoring, not just for lead recyclers but for any furnace industry, linked to both the operator and the authorities. If a substance peaks, lead, cadmium, whatever, it automatically rings a bell at the authority.

I spoke to a Chinese recycler who gets a phone call within an hour of his emissions going over a threshold. For stack monitoring, measuring airborne lead directly is hard and expensive — do you use proxies like sulfur dioxide or particulates?

Yes, that’s what you do. SOx is useful, but you mainly want to monitor the lead emissions. But direct measurement of lead is hard, so you gauge the particle emissions. Once you know the average lead content of this dust, you can caluculate the lead emissions. Things are done a bit differently in different countries. In a lead plant, if there’s dust in the stack, it’s very likely high in lead. So you work with proxies.

How fine do the particle measurements need to be?

It depends, and you have to correlate it. But the other backbone of monitoring is regular blood-lead testing of workers, and also offering it to surrounding communities. You can’t force communities to test, but you can offer it, and it tells you what’s actually entering the human body. That’s a very good proxy. Regular occupational blood-lead testing is the absolute backbone of monitoring.

Right — if the workers are all lead-poisoned, the controls aren’t working.

You can get worker levels down to quite reasonable levels. In a really well-controlled plant you still see elevated levels compared to the general population, but not extremely so.

The Nigeria story and global supply chains

There was a big New York Times story about Nigeria. What broke down in the global system that let American companies buy from such low-standard recyclers?

Lead was out of focus. Many big companies have sourcing departments looking for volumes, price, and availability. If there is no clear policy to do otherwise purchasing decisions get made without much thought — not by individuals, not by departments, not by the company. That’s the commodity-trade tradition: you don’t ask much.

Have you seen that change in Nigeria since the story?

Industry players don’t want to be associated with those negative stories anymore, so they’re developing supply-chain due-diligence methods for risk mitigation and responsible sourcing. Trafigura, the commodity trader, has something on its website along those lines, checking suppliers, and they sell to customers who also want that risk mitigation.

Has that pushed recyclers to upgrade, or do they just sell to other buyers?

When your major customers run a checklist on what you’re doing, it’s supportive. It signals that it’s not only the government, NGOs, or researchers watching. The world is changing so that you don’t get away with everything anymore.

You might camouflage it here and there, and there will always be buyers who take the risk — there are stories of lead being crafted into pipes, exported, and remelted to dodge commodity tracking. So tricks will happen. It might not revolutionize the market, but more people asking for similar things helps shape it.

What’s next

What are you working on now?

We’ll support the governments of Nigeria and Cameroon with a policy development and systematic assessment of Cameroonian plants over the coming months. It’s important to keep building capacity in affected countries — regulatory, research, and NGO capacity.

We want to do more detailed work on the questions you raised: at what point in the chain do we see what emissions, how can each be prevented, and then test those measures. What does a pollution-control monitor change on the regulator’s side? Can satellite imagery help local regulators?

Africa is the main focus, but probably not the limit. Many Asian countries are heavily affected, and we’d like to compare recycling chains across regions and learn from each other.

Is there anything that would be particularly helpful to you right now? The audience here includes a few hundred people very interested in ULAB recycling.

We need buy-in from the proactive parts of the industry. We’ll have lead and lead-acid batteries around us for a long time — this isn’t anti-industry, but we need the proactive industry to be part of it.

We need the authorities — they are sometimes occupied with routine work and may appear to be slow, but many of them have a keen interest in positive change. And we need the groups in these countries who really care.

I’d also flag that there’s real room for hardcore research on exactly who is affected, how strongly, by ULAB recycling: how far the particles travel, what it means for IQ losses and blood-lead levels across cities and countries. I don’t have the final answer.

We see ourselves more on the “how do we reduce it, how do we reform the industry” side, and we’d encourage others to keep pushing on the exposure side.

Which country that you don’t currently work in are you most interested in working in next?

We use a double radar. You can ask where the problem is biggest, or where the preconditions for success are best. We often weight the second: authorities that are genuinely desperate for support because they recognize the problem, plus a strong partnership. That’s why we’re moving into Cameroon — the ministry really wants to move the market, and there are excellent people there, even though in absolute numbers it’s probably not comparable to Nigeria or India. You need to match both criteria.

In Africa, beyond where we already work, I’d rank Kenya high, and Côte d’Ivoire, because of the high awareness in Kenya and upcoming Engitec plant in Côte d’Ivoire, which is a chance to set a good environment from the start and challenge the other recyclers.

Then the Asian countries, because of the sheer size of the problem and because the recycling practices there sometimes differ. From a research perspective - but also from the need to develop and test sector upgrade strategies - we want to work with partners in these countries, cover the full bandwidth and design and test solutions with them.

Hugo Smith asked the questions and published this interview first on his blog ‘Lead Battery Notes’.