Global Cement speaks with Leah Ellis, co-founder of Sublime Systems, which is scaling up the room-temperature production of a low-CO2 ‘drop in’ alternative to conventionally-made cement.
GC: Please could you explain the background to Sublime?
Leah Ellis (LE): Following my PhD in lithium ion research, I was keen to continue to develop chemistry that would have a beneficial impact on the world. I was fortunate to receive a two-year grant from the Canadian government, which I used to work alongside Dr Yet-Ming Chiang at MIT. He is a prolific inventor who had enjoyed previous success spinning out companies. Like me, he is only keen to do research in areas that are at the forefront technologically and make sense on the commercial side.
It was clear to us that decarbonisation of electrical power was well underway, so we turned our attention to how electrochemistry might help us to reduce the other ~70% of global CO2 emissions.
Cement production was an obvious target. After some initial studies and deep dives into the literature, we began work on a ‘drop-in’ replacement for ordinary Portland cement (OPC) using electrochemical means - and at ambient temperatures. We founded Sublime in March 2020 to commercialise our approach.
GC: So what is the approach?
LE: Our approach started by calcining limestone in an electrolyser similar to those used to split water. However, instead of water, we used a different neutral electrolyte that converts into an acid around one electrode and into an alkali at the other. The strong bond between the calcium and the carbonate is broken easily by the acidic conditions at the anode (+ve). This releases CO2 a bit like the more dramatic ‘Mentos and Coke’ demonstration.
The Ca2+ ion has now been liberated into solution and heads to the cathode (-ve), where it encounters the alkali solution. Calcium is insoluble in alkalis, so it precipitates out as calcium hydroxide (Ca(OH)2). This is a solid that can be removed, with the liquid reused in the process.
In the early days, we thought that this could be used as a replacement for a calciner and we wrote a significant journal paper on how existing kilns could be adapted to use calcium hydroxide.1 However, we soon realised that this approach offered much more, the ability to produce cement at ambient temperature by blending our carbon-neutral lime with pozzolanic silicates. The pozzolanic silicates could be sourced naturally from clays and pumices, or they could be produced as a by-product of our calcium extraction process. The dried lime and pozzolanic silicates are simply blended together at ambient temperature.
The dry powder is now a cement as your readers will recognise it. When water is added, it undergoes a hydraulic reaction to give calcium silicate hydrates. It is a like-for-like replacement that large construction firms, contractors and individuals can use without needing to adapt their existing practices. Sublime’s product meets all of the relevant ASTM standards and is particularly good at late strength development. We are developing its performance in other areas and are in discussion with diverse stakeholders regarding the needs of the market.
GC: What kinds of calcium sources can be used?
LE: We envisage a few different variations on the main process, depending on the feed material. Most commonly, at least in the first instance, we expect that the process will use the same calcium sources as existing cement plants - limestone, chalk and marl. Beyond this, there are options to use natural calcium silicates, for example basaltic minerals, using a slightly different configuration of the process. There is also the possibility to use ‘unnatural calcium silicates,’ better known as waste concrete, although this would have to contain the right types of aggregates to be worthwhile.
GC: What are the benefits of Sublime’s approach compared to conventional means?
LE: Sublime’s approach avoids both the use of thermal fuels and the conventional decomposition of limestone. This allows it to make an analagous product to OPC with embodied CO2 emissions far below anything that can reasonably be envisaged for thermal approaches. There is even a path to net-zero CO2 emissions by using renewable energy. On top of this, when the CO2 bubbles out of solution, it is pure, clean and cold. It can easily be removed and condensed, used or stored as necessary, reducing the complexity and cost of the crucial ‘CCUS’ step by up to 90%.
There are also benefits for the downstream process. For example, the cement is already a fine powder, so there is no need to grind clinker, saving further energy. As well as this, there are negligible impurities in the precipitate, as these either don’t dissolve at the same pH as the calcium or don’t precipitate at the same pH. This could allow the Sublime process to offer more consistent quality cement. With further development, it may even be possible to valorise impurities in the limestone, for example magnesium, iron or aluminium.
Additionally, there is no stack, so there are no emissions of dust, NOx, SOx, HCl, CO, cyclic organic compounds, heavy metals or other impurities. Producers will be able to do away with all manner of current emissions control equipment and have fewer dealings with the environmental authorities!
GC: What scale has the process reached so far?
LE: We currently operate a 100t/yr pilot plant at our research facility in Somerville, Massachusetts. It has proven the concept and provides enough material to make blocks, perform testing and otherwise provide investors with insights into our process.
GC: What are the next steps for scale-up?
LE: In January 2023, we secured US$40m in Series A funding from venture capital company Lowercarbon Capital, The Engine, Energy Impact Partners and others. Siam Cement Group is also onboard as a strategic investor. This funding will contribute to a 100t/day demonstration scale plant, currently in the design and engineering stage. It will be built in 2024 and will be operational by the close of 2025.
GC: What are the cost implications of using the Sublime process?
LE: We have calculated that the economies of scale of the Sublime process are similar to those of conventional cement plants. A plant of 1Mt/yr should be able to compete with existing technology. It will, of course, be a long road. Cement is capital-intensive, regardless of the technology used. Even a mini cement plant is huge. There are US$6bn set aside by the US Office of Clean Energy Demonstrations. I would say that the work Sublime has done so far positions it well to hopefully receive some of that funding.
GC: How has Sublime set itself apart in what is now a very crowded low-CO2 cement field?
LE: Many of the low-CO2 cement innovations out there, while impressive, are ‘work-arounds.’ To use an analogy with a water leak, they constantly mop up the water, rather than turn off the tap. The battle with climate change means that we need ‘mops’ for CO2, no doubt, but we need to look beyond this solution as well.
Sublime shows that we can turn off the tap on CO2 in cement production - and that’s hugely exciting. We can stop emitting CO2 from the fuel step using renewable power. When we use calcium silicates as the feedstock, Sublime also does away with CO2 generation from the chemistry itself. As far as I am aware, ours is the only process that does both of these things, while meeting ASTM standards.
Of course, there are also many synergies to work alongside some of these ‘competitors’ too. We could generate CO2-neutral lime for manufacturers of low-CO2 blocks, for example, perhaps even pushing their products into CO2-negative territory.
GC: How would you like to see Sublime grow over the next 10 years?
LE: In the long term, we want to scale up this process globally, but we need to strike a balance between expanding swiftly and at scale. If we do this well, I see no reason why Sublime will not be the main method used to manufacture cement in the coming decades.
GC: Leah Ellis, thank you for your time today.
LE: Thank you too Peter.
Reference
1. Ellis, L. D.; Badel, A. F.; Chiang; M. L.; & Chiang, Y-M. Proc. Natl. Acad. Sci. U.S.A; 117 (23), pp. 12584-12591, 2019. www.pnas.org/doi/10.1073/pnas.1821673116.