Nitrogen+Syngas 367 Sept-Oct 2020
30 September 2020
Carbon dioxide as a feedstock
CO2 TO CHEMICALS
Carbon dioxide as a feedstock
Falling costs for production of hydrogen by electrolysis are encouraging more serious consideration of using recovered carbon dioxide as a feedstock for chemicals and even fuels production.
The cost of generating hydrogen by electrolysis of water and the cost of renewable wind or solar energy to drive that electrolysis has fallen dramatically over the past decade. The Hydrogen Council, an alliance of industry leading companies in hydrogen production, noted in its report earlier this year1 : “A Path to Hydrogen Competitiveness” that: “electrolysis fed with renewable electricity – the most common production method to produce ‘renewable hydrogen’ – has become 60% more affordable as low-carbon and renewable electricity prices have dropped and electrolysis capex has fallen. The cost of solar and wind power, the largest driver of renewable hydrogen production costs, has seen an 80% decrease over the past decade. Recent subsidy-free offshore wind auctions in Europe and bids close to or below $20/MWh for solar photovoltaics (PV) and onshore wind plants have been seen. This downward cost trajectory for renewables should continue, with 14 times more solar capacity projected to become available in 2030 than was previously estimated. At the same time, electrolysis capacity has also started to accelerate, with at least 55 times more capacity expected by 2025 versus 2015, which will result in a similar cost drop in electrolysis capex.”
Two thirds of all new power generation capacity installed last year globally was represented by wind and solar power. As we have noted in articles this year, this potentially makes the production of ammonia and downstream nitrogen products from renewables a cost-competitive proposition, depending upon tax and other incentives. But while generating hydrogen from electrolysis avoids the emission of carbon dioxide from partial oxidation of natural gas, coal or other feedstocks to make syngas, it has also led to interest in whether carbon dioxide itself could become a feed for chemical production, generating ‘carbon negative’ or at least ‘carbon neutral’ chemicals.
Carbon recycling
Although research is ongoing into using polymer membranes or cryogenic separation methods, at the moment, on an industrial scale recovery of carbon dioxide is usually accomplished using solvents, including monoethanolamine or methyldiethylamine (MEA/MDEA), methanol, such as in Linde and Air Liquide’s Rectisol process, methyl pyrolidone, and potassium carbonate (the Benfield process). These processes have all been demonstrated at large scales.
Once the CO2 is captured, the question arises as to what to do with it. At present, there are essentially three options: pumping it into the ground in an existing oil and gas reservoir, either to achieve enhanced oil recovery (EOR), or to store it permanently (carbon capture and storage – CCS), with the final option being converting it into a more useful chemical form. To date, most large applications have been based around EOR, as – absent government incentives – there is no monetary return for CCS, and it would only be preferred where EOR is not possible or where a process can only meet environmental legislation by using CCS – a number of coal gasification project proposals tend to include a CCS component, for example, as it is unlikely that a coal gasification project in e.g., the US or Europe would be permitted without it. But if the chemical industry is to achieve its full ‘green’ potential, the best use would be to use the carbon dioxide as a feedstock for generating chemicals rather than oil, gas or coal, in effect ‘recycling’ the CO2 into chemicals production.
A number of potential routes for this are available. On a large scale basis, chemicals such as formaldehyde and dimethyl carbonate can be produced from CO2 . But perhaps the two largest potential consumers of CO2 are urea and methanol. Indeed, using carbon capture to boost urea production is already an established technology, for example as deployed by Mitsubishi Heavy Industries at the Ruwais urea plant in the UAE in 2007, which recovers 400 t/d of CO2 to boost urea production2 .
Methanol
While the global market for urea is over 180 million t/a, and that of methanol slightly smaller at around 100 million t/a, attention has focused particularly on methanol as an intermediate way of using CO2 , both because production of methanol opens the way to the manufacture of many other compounds, from formaldehyde, acetic acid and methyl methacrylate, to dimethyl ether, methyl t-butyl ether, olefins such as propylene and ethylene, and even gasoline, and because methanol can be used as a fuel directly, either on its own or in blends with more conventional fuels.
Methanol is normally conventionally produced from synthesis gas; a mixture of carbon monoxide and hydrogen, as follows:
CO+2H2 →CH3 OH (1)
However, using carbon dioxide as a feedstock requires the less exothermic CO2 hydrogenation route to methanol:
CO2 +3H2 →CH3 OH+H2 O (2)
This must also compete with the endothermic reverse water-gas shift reaction:
CO2 +H2 →CO+H2 O (3)
which can then supply CO for a conventional conversion for methanol via (1). Both (2) and (3) however generate water which can reduce the lifetime of conventional copper-zinc-alumina (CZA) methanol synthesis catalysts, and which must be separated from the methanol during a downstream purification step, increasing the cost3 . This has led to experimentation with other catalyst systems to minimise water formation.
CRI
One company that has pioneered producing methanol from carbon dioxide and sustainable hydrogen is Carbon Recycling International (CRI). CRI was founded in Iceland in 2006 to explore the idea of recycling carbon dioxide emissions by using it to make useful products. Since 2012, the company has been using geothermal energy in Iceland from natural volcanic processes as a source of energy to heat steam in a turbine and generate electrical energy. This in turn is used to split water into hydrogen and oxygen, and the hydrogen is then combined with CO2 in a process which CRI calls ‘Emissions to Liquids’ (ETL) to produce methanol. The CO2 is captured from the Svartsengi geothermal power station which also generates the power for electrolysis. Around 5,000 t/a of CO2 is used to generate about 4,000 t/a of methanol, which is then blended into gasoline as a fuel additive for use locally. While burning the methanol releases the CO2 back to atmosphere again, it is at least a carbon neutral fuel.
Horizon 2020
As part of the EU Horizon 2020 research programme4 , CRI has also been awarded a e1.8 million grant to help scale its technology to larger scale production plants, a concept referred to as ‘CirclEnergy’, as the technology is designed to support and enable the transition to a ‘circular economy’. The technology has been used at a coal-fired power plant operated by RWE in Niederaussem, Germany, where carbon capture is used to gather CO2 emissions from combustion of coal and combine it with hydrogen from electrolysis in the MefCO2 project. MefCO2 , commissioned in 2019, produces only 1 t/d of methanol, but was designed as a test bed for the system’s capacity to react automatically to variations in hydrogen availability. The automated ETL process system was able to adjust rapidly to fluctuating electricity generation and the methanol synthesis system followed the changes in the availability of hydrogen. Results showed that conversion efficiency in the ETL reactor was not affected by these fluctuations and exceeded design criteria, presenting the opportunity for methanol production to be used in so-called ‘grid balancing’, storing excess energy generated from wind and solar at times when it exceeds demand for electricity, when marginal cost of electricity generation is effectively zero.
Now that this project is over, the process module is being transferred to another strand of Horizon 2020, where CRI is also working with Sweden’s Swerea MEFOS research centre facility in Luleå, Sweden. The module will be configured to convert residual blast furnace gases into methanol, using both hydrogen-rich off-gas and CO2 from direct iron oxidation to make methanol for use as a shipping fuel for Swedish ferry operator Stena, which operates the world’s first methanol fuelled passenger ferry, the Stena Germanica.
Meanwhile, on a larger scale, CRI is working with Chinese chemicals corporation Henan Shuncheng Group to produce low carbon intensity methanol in China. The $90 million project will involve building a plant at Anyang city in Henan province to recycle about 150,000 t/a of CO2 , along with other waste gases to produce 180,000 t/a of methanol and LNG. Commissioning is expected by the end of 20215 .
Other projects
In Germany, bseEngineering and the Institute for Renewable Energy Systems at Stralsund University of Applied Sciences (IRES) have demonstrated the conversion of wind power into renewable methanol using CO2 captured from flue gas. The technology is now being tested under ‘dynamic conditions’ of fluctuating power input over the course of a year-long programme. Basic production of methanol is 20 t/d (16,400 t/a) using one of bse’s modular reactors and a BASF catalyst, but the system has the flexibility to run at between 10 and 120% of design capacity according to bse6 .
Clariant is providing its expertise in catalysis and syngas production in partnership with Air Liquide (who are licensing their Lurgi methanol process) and gas to liquids (GTL) technology company Ineratec – a spin-off from the Karlsruhe Institute of Technology – to develop a power to fuels and chemicals programme. Ineratec have their own modular containerised process for conversion of CO2 to liquid fuels in a microchambered reactor. The technology relies on Clariant’s HyProGenR-70 ® catalyst to generate renewable syngas using recovered CO2 and hydrogen from electrolysis via a reverse water-gas-shift reaction. Clariant’s MegaMax catalyst also powers the methanol synthesis reaction. Clariant said at a recent webinar organised by the Methanol Institute that it is looking at a three-stage methanol synthesis process with inter-condensation of water and a recycle loop that leads to higher single pass conversion and less water flow on the catalyst. It also claimed that the process can achieve power to liquids economics competitive with current advanced biofuels7 .
Haldor Topsoe meanwhile is part of a consortium called Liquid Wind with Siemens, Nel Hydrogen and Carbon Clean Solutions to develop a methanol plant based on hydrogen from wind energy. The company is aiming to build its first ‘eMethanol’ facility in northern Sweden, with initial operation slated for 2023. The company plans to develop five further facilities in Scandinavia before 2030, at which point the model will be replicated and licenced internationally. Topsoe is providing its new MK-317 Sustain methanol synthesis catalyst as part of the project8 .
Direct air capture
Most CO2 capture technology has concentrated on removal of carbon dioxide from industrial process streams or large scale combustion, where the CO2 is already concentrated and in gaseous form. However, Carbon Engineering, a company based in Squamish, British Columbia which was founded in 2009 by Harvard professor David Keith, generated some headlines last year when they declared that they would be scaling up their technology for direct processing of carbon dioxide from air. Carbon Engineering calls this Direct Air Capture (DAC)9 .
The process begins with an air contactor; a large structure modelled off industrial cooling towers. A giant fan pulls air into the structure, where it passes over thin plastic surfaces which have potassium hydroxide solution flowing over them. This solution binds with the CO2 molecules, removing them from the air and trapping them in the liquid solution as a carbonate salt.
The CO2 contained in the carbonate solution is concentrated, via separating the salt out from the solution into small pellets in a structure called a pellet reactor. These pellets are then heated in the third step, a calciner, in order to release the CO2 in pure gaseous form. This step also leaves behind processed pellets that are hydrated in a slaker and recycled back within the system to reproduce the original capture chemical.
Carbon Engineering announced in September 2019 that it was expanding the design for its first commercial DAC plant from s protected capacity of 500,000 t/a of CO2 removal to 1.0 million t/a of CO2 . The planned facility, in the Permian Basin, is being built in partnership with Oxy Low Carbon Ventures, a clean technology subsidiary of oil major Occidental, and private equity firm Rusheen Capital Management. Front end engineering and design will begin in 1Q 2021 and construction for the plant is expected to begin in 2022, with the plant becoming operational within approximately two years.
Of course, the major issue with DAC is the relatively low concentration of CO2 in atmospheric air. Even at the enhanced CO2 levels that industrialisation has led to, CO2 concentrations remain at only 400 parts per million (ppm), or 0.4%. This means that huge volumes of air must be processed relatively efficiently in order to recover large volumes of CO2 . Nevertheless, by operating on such a scale, Carbon Engineering claim to be able to recover CO2 at a cost of $100/tonne. The company also argues that as the concentration of CO2 is relatively constant globally, the technology is independent of location, and so could be constructed directly next to reservoirs suitable for carbon capture and storage (CCS), as in the Permian Basin plant, or plants capable of using the CO2 for industrial processes.
And as far as the latter goes, Chevron is also investing in Carbon Engineering’s DAC technology, though its interest is for synthetic fuel production – Chevron was involved in Fischer-Tropsch gas to liquids plants with Sasol in the 1990s, and developed a zeolite catalyst for F-T production, as well as providing isocracking and other technologies for product clean-up. In California, where Chevron and Carbon Engineering are developing CO2 to fuel technologies using hydrogen from renewables, the state’s Low Carbon Fuel Standard assesses a $200/ton penalty on excess CO2 in the fuel, so avoiding that penalty can make their synthetic fuel a the lower-cost option than conventional gasoline. That plant should be operational between 2022 and 2023.
Carbon Engineering argues that falling solar prices should enable them to bring “air-to-fuels” to market for about $1/litre (around $4/gallon) in the mid-2020s. This is equivalent to a methanol price of over $1,300/tonne, although the company says that the price will continue to fall from there.
Practicalities
The issue that bedevils many green technologies is their cost compared to conventional production routes. The dominant cost factor has generally been that of electrolysis of hydrogen – the hydrogen oxygen bond in water is strong, and breaking it is thermodynamically costly. However, as noted above, the cost of generating the electricity to power the electrolysis continues to fall. Topsoe has pointed out that each doubling of capacity in wind or solar photovoltaic power generation reduces the cost per unit of energy by about 20%. A Nature Energy paper last year 10 found that if market trends continue, green hydrogen could be economically competitive with natural gas on an industrial scale within a decade. Similarly, the International Energy Agency projects that the cost of clean hydrogen will fall 30% by 2030. Green hydrogen may already be nearly affordable in some places where periods of excess renewable generation drive down the costs of electricity to nearly zero. In a recent research note, Morgan Stanley analysts wrote that locating green hydrogen facilities next to major wind farms in the US Midwest and Texas could make the fuel cost competitive within two years.
However, a report last year in the Proceedings of the US National Academy of Sciences (PNAS) 11 has poured some cold water on the idea of a large scale move by the chemical industry to renewable feedstocks. The research team modelled the global chemical industry “to analyse the potential disruptive changes through large-scale CO 2 utilisation and resulting emission reductions.” It found that, if rolled out across the petrochemical industry, while these technologies could reduce greenhouse gas levels in the atmosphere by 3.5 gigatonnes, this would require 18.1 petawatt hours of renewable electricity per year, far more than the total amount of renewable electricity available anywhere in 2030 even on the most optimistic projections. The thermodynamics of breaking the H-O bond are, as we noted earlier, fundamentally against you.
But this is to assume that a move to lower carbon chemical production is necessarily an ‘all or nothing’ affair. In fact it is possible to lower the CO 2 intensity of methanol production without moving completely to all-renewable operation. Andrew Fenwick of Johnson Matthey described a wide range of options for doing so in this magazine earlier in the year 12 . And at the Methanol Institute webinar in August this year, Yawar Abbas Naqvi presented what he called a ‘hybrid plant’ concept 8 , as described in Figure 1. Table 1 shows the results of Topsoe’s analysis, assuming an electricity cost of e29/MWh, and a realistic price for natural gas in Europe. It shows that even at today’s prices, with the addition of the renewable hydrogen feed, a net negative carbon balance can still be achieved for a 1,000 t/d methanol plant which is largely working from natural gas, at a methanol price only a little way above current prevailing ones, and with the addition of environmental credits bringing it into the realms of commercial possibility.
References
