Generating value from fermentation CO2

Abstract

Around 8% of the total CO₂ emissions for the Australian wine sector are from fermentation of grapes (Emissions Reduction Roadmap, 2023). Although regarded as biogenic CO₂ and therefore not required to be included in carbon accounting, fermentation emissions offer a unique opportunity for the wine sector to capture CO₂, and some customised systems are being used in wineries in Spain and France. This project worked collaboratively with several large wineries in Australia to evaluate a range of options for capture and reuse or storage of CO₂ from wine fermentations. Included was a comprehensive assessment of their economics, carbon footprint outcomes, feasibility for Australian wineries and impact on supply chain resilience.

Summary

A review was undertaken of options for recovering carbon dioxide (CO₂) from winery ferments. This included analysis of data from collaborating wineries on quantities of CO₂ generated and consumed, review of systems in use at wineries and breweries in Australia and overseas, discussions with equipment and gas suppliers, and analysis of pricing and carbon impacts.

Across three vintages at five large wineries, CO₂ generated each vintage exceeded each wineries annual CO₂ consumption by 1.7-39x. Sites that packaged wines on site had lower ratios (i.e. consumed relatively more CO₂), while the site with the greatest CO₂ generation relative to consumption was a predominantly red winery with no packaging on site. In contrast to ferment CO₂ generation, CO₂ consumption showed little repeatable seasonality.

All wineries were currently experiencing inflated CO₂ purchase prices, typically double or even triple the price they had paid in previous years. Most sites ran out of CO₂ for short periods before the gas suppliers secured alternative sourcing at higher prices. The largest cause of the shortages in South Australia was the shutdown of Air Liquide’s CO₂ recovery plant on the gas-fired Torrens Island Power Station, as the power station shifted to operating in peaking mode before a planned 2026 closure. Air Liquide and BOC are currently constructing CO₂ recovery plants attached to the natural gas conditioning plant at Longford in Victoria. These will come on-line in 2024 and they believe that these will alleviate the CO₂ shortages, albeit with 20-30% higher prices for South Australian customers. Air Liquide estimates a 15-year life for this CO₂ source, and it is plausible that CO2 recovery on a natural gas conditioning plant is at lower-risk of closing than a natural gas or other fossil fuel fired power station. However, it is still closely linked to fossil fuel production or combustion, as are all their other current and planned production sites. This presents a significant supply risk in the longer-term that it would not be sensible to ignore, given the shortages and price shock that is being experienced right now because of their poor planning.

Fermentations present an alternative CO₂ supply. Before any processing, the raw ferment gas is already far purer than the natural gas or flue gas being used as CO₂ sources currently. It presents a major opportunity for the wine sector for internal use and sale to other sectors in the longer-term as the world decarbonises. There are technical and economic challenges to overcome, but is important to overcome the technical challenges, so that it become a viable supply option when the economics allow.

The accepted beverage CO₂ industry purity standard is published by the International Society of Beverage Technologists (ISBT). This specifies a minimum CO₂ purity of 99.9% and limits for multiple key contaminants. This purity level may or may not be required for internal winery use. It is expected that purity will be more important when the CO₂ is used for carbonation or sparging compared with when it is used for headspace management due to the smaller gas-liquid surface area in the latter. While raw wine ferment gas is believed to be fairly pure (often >99% CO₂), there is little published data proving this or evaluating what the impurities are relevant to CO₂ re-use in wineries and other industries. When obtaining designs and prices for CO₂ recovery plants, this lack of data was a problem, and it is recommended that in the 2024 vintage, purity measurements be performed on raw winery ferment gas. This will be foundational information relevant to multiple potential future uses of fermentation CO₂.

Discussions and inspections of fermenters at wineries identified that for white fermenters it should be relatively easy to pipe off CO₂ for collection (except for those tanks with lids sitting in water traps), but that many red fermenters present major challenges. Collection from open-top fermenters and rotary fermenters will be very difficult but these only represent a relatively small fraction of red fermentation volumes. However, most other red fermenters in wineries have irrigators sitting in the necks that make it difficult to close the lid without a redesign and others use Pulsair and rack and returns. Pulsair creates pressure pulses that need to be managed and will dilute the headspace CO₂, and rack and returns may also dilute the headspace CO₂ concentration. It is important to find a viable means of collecting relatively pure CO₂ from these tanks because being able to collect CO₂ from both white and red fermenters will allow the length of CO₂ capture (and associated use of an expensive recovery plant) to be doubled from around 40 to 80 days per year. Therefore, it is recommended that work be performed in the 2024 vintage to resolve how to get CO₂ from different fermenter designs to a central collection point. Again, this is foundational knowledge that will be useful for a range of potential uses of fermentation CO₂.

Economic analysis of brewery-style CO₂ recovery plants identified that CO₂ storage costs are likely a larger expense than the recovery plant itself because of the highly seasonal nature of wine production and therefore large storage volumes. This makes investment in brewery-style plants uneconomic without subsidies - although a few are installed in German wineries, and it is likely that there will be Australian grants at some point that would improve the payback period for installations here.

To try and address the high cost of liquefied CO₂ storage, it is recommended that CO₂ storage based on the conversion of sodium carbonate to bicarbonate with later reversal of the reaction and generation of CO₂ using heat, also be experimented with. However, this presents its own challenges with regards to the low solubility and therefore large volumes of water needed, and the unknown feasibility of regenerating pure CO₂. Liquefied CO₂ storage is the accepted CO₂ industry standard, and despite its current economics, most likely it will be the technique used by wineries in the future (in some format) – rarely if you have something pure, does it make sense to make it less pure, when you later need the pure substance again in the future. Therefore, it is recommended, that all work relating to capturing ferment CO₂ be ultimately compatible with this storage mechanism.

The reaction of carbonates with CO₂ can also be used to precipitate and sell bicarbonates and this is being performed by a small number of wineries in France. However, managing bicarbonate production represents another operation that will mainly be performed during vintage when resources are already stretched. To meet customer demands for purity and moisture content, the bicarbonates may also require further processing including energy intensive drying. It is also moving further away from the wineries core-business of making wine. Similar arguments can also be made for other products that might use ferment CO₂ as an ingredient – eFuels, methane, algae, etc.

During the review, several wineries reported that significant volumes of purchased CO₂ are being wasted from pressure tanks, during Charmat wine production. Preliminary economic analysis suggests that there could be a benefit in recovering and reusing this. It is recommended that as part of the current project that this opportunity is quantified in some detail and other potential solutions considered. For example, some of the current losses may potentially be mitigated by changes to how headspaces of tanks are connected, utilising adjacent empty pressure tanks for CO₂ storage, using non-caustic cleaners so that tanks can be cleaned under pressure, and possibly even substituting CO₂ with nitrogen in some operations. The practicality of these alternatives versus installing a separate pressure storage tank and compressor will vary between sites. It is a worthwhile part of the project, but it should not become the prime focus because while the payback appears more favourable than fermentation CO₂ capture, the magnitude of savings is relatively small and there is not the same step-change in opportunity for the industry.