The Swiss company hoping to capture 1% of global CO2 emissions by 2025

Print Friendly, PDF & Email

Carbon Brief

In the roof of a waste incinerator outside Zurich, the Swiss firm Climeworks has built the world’s first commercial plant to suck CO2 directly from the air.

Climeworks says that its direct air capture (DAC) process – a form of negative emissions often considered too expensive to be taken seriously – costs $600 per tonne of CO2 today. This is partly covered by selling the CO2 to a nearby fruit and vegetable grower for use in its greenhouse.

Climeworks hopes to get this down to $100/tCO2 by 2025 or 2030. It aims to be capturing 1% of global CO2 emissions each year by 2025.

Carbon Brief travelled to the opening of the plant and interviewed co-founder Christoph Gebald to find out more about Climeworks’ ambitions, how the technology works and how it might contribute to global climate goals.

Negative emissions

Under the 2015 Paris Agreement on climate change, nearly 200 countries agreed to limit warming to “well below” 2C above pre-industrial levels and to aim for no more than 1.5C.  The accord also calls for a “balance” between greenhouse gas emissions sources and sinks in the second half of the century, equivalent to reaching global net-zero emissions.

Even before Donald Trump said the US would pull out of the deal, national climate pledges, when viewed cumulatively, fell far short of what would be needed, with the carbon budget for 1.5C set to be used up within as little as four years.

This has sparked a growing realisation that so-called negative emissions might be necessary to meet the goals of Paris, where an overspend against the carbon budget is paid back by pulling CO2 from the air.

Some estimates suggest as much as five billion tonnes of CO2 (GtCO2) would have to be removed from the atmosphere, and then locked away underground, each year by 2050. (Last year, Carbon Brief produced a series of articles on the need for negative emissions, the options available and whether they are feasible – or merely a distraction that encourages complacency).

Direct air capture (DAC) is one of those options, with DAC machines often described as “sucking CO2 from the air” or “artificial trees”. It has a number of attractive features, including a limited land footprint, the ability to site units near to CO2 storage sites and a clarity around how much CO2 it sequesters, in contrast to negative emissions that use biomass.

Christophe Jospe, an independent consultant and former chief strategist at the Center for Negative Carbon Emissions at Arizona State University. points out that DAC avoids the complex question of biomass carbon accounting, which clouds the other leading negative emissions option, bioenergy with CCS (BECCS). “You have an unambiguous carbon sink, whereas trees can get cut down.”

miele greenhouse

miele greenhouse 2

Prohibitive costs?

Yet direct capture also has an achilles heel. There is a widely held perception that it is extremely costly, in both energy and financial terms. Academic estimates for the cost of CO2 capture, transport and storage, along with regeneration of chemicals used in the process, range from $400 to $1,000 per tonne of CO2.

These estimates are based on extrapolating what we know about carbon capture and storage (CCS) at power plants, where CO2 levels in flue gases are much higher than in ambient air. This is often thought to mean that the costs and energy needs for DAC will be many times larger than for CCS. (This view of high costs is disputed, however – see below).

Jospe tells Carbon Brief: “The big question in people’s minds is cost…That’s the potential nail…[in] the coffin [for] DAC.”

According to a 2016 Nature paper, DAC would require a theoretical minimum of 0.5 gigajoules (GJ) of energy to remove and store each tonne of CO2. Or, perhaps, as much as 12GJ/tCO2 once inefficiencies and other stages of the process are taken into account.

On this basis, the paper says that capturing 12 billion tonnes of CO2 equivalent (GtCO2e) per year (around a third of annual global emissions) would require 156 exajoules (EJ) of energy. This is more than a quarter of total annual global energy demand for all uses, of around 550EJ.

The paper says the costs and energy requirements would be “prohibitive” and that research and development is required to bring them down. Its calculations also do not include the greenhouse gas emissions that might be associated with providing the large energy needs of direct capture.

In a recent perspective for Science magazine, Prof Chris Field, a former co-chair of the Intergovernmental Panel on Climate Change’s (IPCC), and Dr Katharine Mach, director of the Stanford Environment Facility, wrote:

“Engineered, nonbiological approaches [to negative emissions], such as enhanced weathering and direct air capture…are energy-intensive and expensive [but] may eventually provide useful options for [CO2 removal] at scale. At this point, however, their technological immaturity means that estimates of future costs, performance, and scalability are speculative.”

They add: “Direct air capture could become a major industry if the technology matures and prices drop dramatically…Direct air capture might require much less land [than other negative emissions techniques], but entail much higher costs and consumption of a large fraction of global energy production. The required land would operate as an immense array of industrial facilities.”

Commercial Climeworks

It’s into this somewhat sceptical arena that Swiss firm Climeworks recently launched the world’s first “commercial” direct CO2 capture plant at Hinwil, a small town just outside Zurich. (It’s worth noting up front that while the plant is selling its CO2, it is not covering its full costs.)

The firm is a spin-off from ETH Zurich, founded in 2009 as the brainchild of two students who met at the university: Christoph Gebald and Jan Wurzbacher. In the past two years, Climeworks has grown rapidly, reaching 45 employees today. Its $20m in financing includes $5m in Swiss government grants and $15m from private equity.

Describing the inspiration for the company, Gebald tells Carbon Brief:

“The original idea to capture CO2 from the atmosphere comes from Prof Aldo Steinfeld. He is a professor at ETH Zurich, working on solar fuels, and he needed atmospheric CO2 in order to produce renewable fuels. In his research programme, Jan and I were students, so that was the original spark of starting what we did, eight years ago.”

In fact, this spark remains, as the company is partnering with German carmaker Audi to develop renewable fuels. Climeworks’ Hinwil plant cost $3-4m to build and sits in a favourable location, perched on the roof of a municipal waste incinerator, which supplies the low-cost heat that it needs.

With a backdrop of green farmland and solar-clad barns, the plant is also a stone’s throw from the Gebrüder Maier fruit and vegetable company, which uses the captured CO2 to boost the growth of cucumbers, tomatoes and aubergines in its large greenhouses.

This location, a lengthy tanker-trip away from the usual industrial sources of CO2, such as refineries, presumably adds to the price that Climeworks can command for its product. The market price in Switzerland, for small amounts of CO2, is $200-250/t, Wurzbacher says in a press call.

Driving the Climeworks process uses 2.5 megawatt hours (MWh) of heat, at around 100C, for each tonne of CO2, along with 0.5MWh of power. This energy requirement is roughly equivalent to the 12GJ/tCO2 estimates set out above, though the firm hopes to shave 40% off this figure, bringing it down to around 7GJ/tCO2. Gebald says an increase in energy resources – he points to wind and solar – would be needed to scale up direct capture.

Capture process

The Climeworks machine consists of a series of three, stacked, shipping container-sized units, each of which contains six CO2 filters. A large hot water storage tank sits alongside, along with two further containers housing control equipment.

It works by adsorptiondesorption, with fans blowing ambient air over the capture material, which traps CO2 and water, until they are released with heat. Water is a by-product of the process. The machinery is capable of capturing 900tCO2 per year.

Gebald tells Carbon Brief:

“With this plant, we can show costs of roughly $600 per tonne [of CO2], which is, of course, if we compare it to a market price, very high. But, if we compare it to studies which have been done previously, projecting the costs of direct air capture, it’s a sensation.

“Studies so far assume that what we do will cost $600 per tonne – this is a study by the American Physical Society, and one study by MIT, Stanford and Berkeley even assumed $1,000 per tonne – for plants which are a) industrialised and b) have a much bigger scale, so not one thousand tonnes per year, but a factor one thousand more, so a million tonnes per year.

“So we are very confident that, once we build version two, three and four of this plant, we can bring down costs. We see a factor [of] three cost reduction in the next 3-5 years, so a final cost of $200 per tonne. The long-term target price for what we do is clearly $100 per tonne of CO2.”

For consultant Christoph Jospe, the modular nature of direct capture machinery offers the potential for cost savings that he likens to the solar industry, where prices plummeted as manufacturing scale, and experience, increased. Gebald explains the ways that Climeworks hopes to cut costs:

“In order to achieve this factor-three cost reduction, it’s a combination of facts. It will be procurement, so, purchasing larger volumes, it will be professionalising our production infrastructure. The plant which we are starting today is more or less handmade in Switzerland, which is maybe not the definition of the cheapest way of producing things. So we are starting to automate production steps: rather than people, drilling or screwing stuff, that’s like, robots, etcetera, can do this. By going step by step, these means that I just mentioned, we can achieve these cost reductions of a factor of three.

“In order to again halve the cost, once we reach $200/t, we need R&D to happen. For example, you see a lot of steel behind you, actually stainless steel, which is also not the cheapest [way] to build things. So maybe in the future we can use cheaper materials than steel…I think we cannot reach $100/t simply by scale, or by procurement, but it’s not fundamental research that needs to be done, it’s simply optimisation work, which is well known.”

Several other firms hoping to commercialise direct capture of CO2 say they can already comfortably beat these costs. Graciela Chichilnisky is chief executive and co-founder of Global Thermostat, which is developing four projects to capture CO2 for use in carbonated drinks.

Global Thermostat says it can capture CO2 for just $50 per tonne, using hardly any electricity and waste heat at 85C. Chichilnisky tells Carbon Brief: “Our technology is unusually inexpensive. No other technology is even close to ours in terms of removing CO2 from air.”

This cost is viewed with scepticism, however. “Show me and I’ll believe it,” says one researcher in the industry. That’s “very, very ambitious,” says Climeworks’ Gebald. “As of today, we consider a price of $50/t as being highly challenging to unrealistic.”

Climeworks in numbers

  • 300 tonnes of CO2: Annual removal capacity of one shipping container-sized unit fitted with six Climeworks filters
  • 225 million tonnes of CO2: Climeworks goal for CO2 capture in 2025, equal to nearly 1% of global emissions
  • 100 units: Annual capacity of Climeworks’ current production line
  • 750,000 units: Number of units that would be needed to capture 225MtCO2 in 2025
  • 250 cars: Equivalent annual CO2 emissions that would be captured by a single Climeworks unit
  • $700 per car: Annual cost to remove car CO2 emissions from the air, at current costs
  • 675 terawatt hours: Energy needed to capture the 225MtCO2 target for 2025, at current energy use rates
  • 650 terawatt hours: Electricity used by Germany each year


Paris target

The context for the Climeworks opening is not lost on its founders, who make frequent reference to the need for technologies, such as DAC, in order to meet the net-zero emissions goal of the Paris Agreement. Gebald tells Carbon Brief:

“Paris says we have to go to zero gigatonnes, be CO2 neutral in 20 to 30 years, or by 2050, actually, which requires severe emissions cuts and a combination of all technologies which are available. Yes, it took us eight years to get to where we are today, but we started as college graduates, we started without any experience…I’m very confident that we can continue this strong growth, which we had especially in the last two or three years, in order to meet those targets.”

He says: “The vision of our company is to capture [one] percent of global emissions by 2025, which is super ambitious, but which is something that is feasible.” To reach this scale, the company would need to install hundreds of thousands of units, an impossible task on a commercial basis alone.

Gebald says: “Reaching 1% of global emissions by 2025 is currently not possible without political will, without a price on carbon, for example. So it’s not possible by commercial means only.”

Wurzbacher suggests pioneering companies could help kickstart this by aiming to become not only carbon neutral, but carbon negative.

For now, Climeworks is developing niche applications, including CO2 for carbonated drinks and renewable fuels. It plans to open seven facilities over the next two years, including a negative emissions plant in Iceland that will both capture, and then store CO2 underground. (This is a partnership including Reykjavik Energy, CarbFix and the University of Iceland).

The costs of this scheme are likely to differ, because the figures cited by Climeworks above do not include the cost of transport or storage of CO2. These steps may not add significantly to the total, however. Antonios Papaspiropoulos, global lead for advocacy and communications at the Global CCS Institute, tells Carbon Brief:

“In terms of the costs of transporting and storing CO2, these are obviously dependent on different variables – transportation distances, storage properties, etc. That said, as a rule of thumb, we would suggest US$10-$20 per tonne [of CO2] as a reasonable range.”

Quantum leap

Thomas Stocker is professor of climate and environmental physics at the University of Bern and was co-chair of working group one for the Intergovernmental Panel on Climate Change (IPCC) fifth assessment report. He tells Carbon Brief at the Climeworks launch:

“What we’ve seen here today is really a quantum step in implementation of technology that is able to capture carbon from the atmosphere and put it into use or capture it for good and store it. This very first element that we see here has shown the proof of concept and it is now in such a state that it is convincing people to buy the product. That is one of the key and crucial steps to achieve what Paris has laid out, and that is, implementation of technology that sits in the labs, that has not yet been scaled up.”

Stocker notes the challenge of the Paris goals and the ambition of the Climeworks targets, which would entail the creation of a significant new industry. However, he tells Carbon Brief:

“Look, if you have to climb a huge mountain, you start with the first step. You will not reach the summit if you have not done the first step…It’s in that perspective that I see this development: a very important first step has been taken. A step that proves to the outside world, outside of the laboratory, that this can work. Whether or not it’s scaleable at the scale of global emissions…that’s another question that cannot be answered here…It’s also not a silver bullet that can solve the problem, to keep temperature rise below 2C.”

Stocker says that negative emissions technologies must supplement, not substitute for aggressive investment in energy efficiency and renewables to “replace fossil energy at the fastest rate possible”.

climworks tech.

One interesting possibility raised by Gebald is the idea that direct capture could provide a backstop solution that effectively caps the cost of cutting emissions. He says:

“At the end of the day, I think that the price of air capture will determine the price of carbon. In the long run, it doesn’t make sense, if we can capture if for $100, like, in 20 years, but there is a price on carbon for $400.”

This can be compared to the marginal cost of cutting emissions in some models of the future. For example, a recent International Energy Agency report has a power-sector abatement cost reaching $600/tCO2 in 2060.


Climeworks’ opening of the world’s first commercial direct capture plant holds the potential to be a major milestone in the fight against climate change. Speaking at the launch event for the plant, Lawrence Livermore National Laboratory’s Dr Julio Freedman, an Obama appointee to the US Department of Energy, told the audience: “I truly believe [this] is a historic event.”

Yet questions remain over the cost and energy needs of the process. Dr Niall Mac Dowell, head of the clean fossil and bioenergy research group at Imperial College London, tells Carbon Brief:

“The cost of DAC continues to be an an area of some controversy, with a very wide range of estimates in the academic literature. There is, therefore, an urgent need for transparent and verifiable costs to be presented to the community in order to build confidence in what could be an incredibly valuable technology in fighting climate change…It’s particularly interesting to see [Climeworks’] cost-cutting objectives of a three to four-fold reduction [to around $200/tCO2] by 2020, which is just around the corner. That’s really exciting, but it’s important they demonstrate that in a transparent and understandable way.”

Climeworks is well aware of the pressure to open up its costings. Gebald tells Carbon Brief: “It’s unclear when we will do that, but clearly, it’s on our list, and we want to share this information.”

Source: Carbon Brief. Reproduced with permission.  

  • Bill V

    Would be better to capture methane?

    • Alastair Leith

      Or clamp down on fugitive emissions like Obama finally did at the end of his presidency. And Trump erased.

      In Ag sector we have a major methane emissions and land clearing problem contributing to the 56% of national emissions in Australia coming from the Land Use sector. Get the livestock of extended ranges of repeatedly cleared land and there’d be potential for some dietary intervention, but still it’s not a solution so much as a minor mitigation exercise.

  • Jo

    “… 675 terawatt hours: Energy needed to capture the 225MtCO2 target for 2025, at current energy use rates
    650 terawatt hours: Electricity used by Germany each year

    With Germany emitting 2.16% CO2 of the world the above data become a circular argument: you need all of Germany’s electricity to reduce half of it’s CO2 emission?????
    I am aware that I mess up CO2 emission and energy production because I did not have time to go into details. But just this crude analysis tells me the this idea cannot work. And it is dangerous because the project could be used by polluters to tell that cleaning up later is simple.

    • Ian

      It would be perfect if we only we could send over our ‘clean coal’ to power it 🙂 !

  • John Saint-Smith

    “One of the advantages is the small footprint”. That assumes that no space is occupied by the manufacture of these machines, or for the storage of the CO2. Algae based bio-diesel ‘factories’ could be similarly compact and require no costly energy input apart from sunlight.
    Surely we should first consider all the natural advantages of enhanced biological CCS, where the CO2 is sequestered as wood or some other synthetic carbohydrate material that could be used for building, road making, insulation?

  • Brunel

    Waste incinerator! Fine people for throwing plastic and glass bottles in rubbish bins.

  • Mark Roest

    Major carbon capture could also happen if we divide up the remaining work after 80% of jobs are lost to automation, and most of us help grow food forests (check on Permaculture practices), and each of us use maybe 450 kg of wood from selective thinning for pyrolysis (see All Power Labs, Berkeley, California’s pyrolysis refinery on a pallet). Use the liquid that results as a chemical feedstock to replace petrochemicals, and bury the biochar in gardens and wild-lands, wherever resilience-promoting storage of water and nutrients is needed. Use intermittent renewable energy and batteries to provide all transportation (except eg ships with sails to provide motion most of the time) and electricity.

    • Alastair Leith

      Container transport ships with sails, i saw them once on the “Beyond 2000” TV show. biochar no silver bullet but better than a fools promise of removing CO2 from the atmosphere in a scale that makes a difference in a time frame that stops the 5-8º C of warming Finkel’s modelling projected we aim for.

  • Carl Raymond S

    Where does the CO2 end up? The graphics says “collected”. Then what?

    • Mike Shackleton

      In the example here – the CO2 is sent to greenhouses to improve their productivity. The greenhouses pay the plant for the CO2. I thought that was pretty clear in the article.

      • Carl Raymond S

        So it returns to the atmosphere and won’t affect the Keeling curve.

        • Alastair Leith

          Exactly. Unless the CO2 it is locked away to never see the light of day it’s an expensive exercise in CO2 production for greenhouses not a Climate solution. Way better investment would be to reduce ag sector emissions 90% of which in Australia are associated with ruminant livestock.

          • Steve

            Not quite so quick to dispel the technology though.

            In a carbon neutral world CO2 can be combined with water and cheap energy to produce methane – to drive turbines when power can’t be generated by renewables. The benefit is that LNG remains in the ground.

            That methane can be used to generate REALLY clean avgas (as in, no sulphur or anything) to support net carbon neutral airline travel.

            And one of the difficulties with sequestering is that the areas where CO2 can be sequestered (say underwater basalt fields) are not necessarily where CO2 is generated. These may be positioned next to natural means of sequestering carbon.

          • Alastair Leith

            Yes, Steve, if this can play a role in power2fuel then great! Just don’t fancy the economics ATM. If CO2 input dependant industries can use it and drive the price down to something we could pay for when excess energy is being spilled every day on the grid from solar and wind when all the storage is already fully charged then I could see this having a future. Things like mineralisation of olivine/serpentine will compete with it I suppose and whoever gets cheapest will win the market for non-bio sequestration.

            Still think getting livestock of extended zone grazing allowing millions of acres of regrowth/woodlands/plantations would be more effective.

  • Gnällgubben

    Who is going to pay for capturing CO2 only to bury it in the ground?

    • Alastair Leith

      when excess PV and wind is being curtailed most days (or dispatching at negative prices due to the certificates they earn) and there is a price on carbon only then can this begin to add up, and costs would have to fall through the floor for the technology.

  • Alastair Leith

    “What we’ve seen here today is really a quantum step in implementation of technology that is able to capture carbon from the atmosphere and put it into use or capture it for good and store it.”
    If you feed it to plants that is not what you are doing, no different to putting it straight back into the atmosphere.

  • Alastair Leith

    “At the end of the day, I think that the price of air capture will determine the price of carbon. In the long run, it doesn’t make sense, if we can capture if for $100, like, in 20 years, but there is a price on carbon for $400.”

    Well yes, but $400/tC is a little ways off thanks to reactionary conservative politicians (and just straight out corrupt ones) and those who ultimately pull their strings. And $100/tC sequestered is fiction at this point.

  • Radbug

    I like carbon fibre from atmospheric CO2, using Li2O, Li2CO3, CO2 cell. Energy cost: 25% that of Hall Heroult.