The science of extraction - rewritten
Why a new equilibrium model is changing the way we think about brewing ratios, TDS and yield
Why a new equilibrium model is changing the way we think about brewing ratios, TDS and yield
Extraction is at the heart of every coffee preparation process. And yet it is often treated as a kind of black box: Water meets coffee grounds, after a certain amount of time an aromatic liquid emerges - sometimes stronger, sometimes weaker, sometimes too acidic, sometimes too bitter. Between theory and practice lies a zone of experiential knowledge characterised by trial and error and sensory intuition. But this is precisely where the next scientific turning point comes in: A new, elegant model for describing extraction in full immersion brewing now provides clarity - and makes extraction more controllable than ever before.
The approach: extraction can be described as a desorption process in equilibrium. Instead of dealing with individual molecules or complex transport models, the new model is based on an average equilibrium constant that represents the behaviour of all soluble components. The result is a precise predictability of brewing strength and yield as a function of the brewing ratio - and a radical simplification of previous extraction concepts.
TDS can be calculated - exactly
The central point of the model is the relationship between the brewing ratio (i.e. the ratio of water to coffee), the maximum possible extraction and the equilibrium constant K. This means that the TDS - i.e. the proportion of dissolved substances in the final liquid - can be derived almost exactly from the brewing ratio. And that's not all: for typical brewing ratios, there is a simple mathematical relationship in which K, E_max and R_brew are the decisive variables.
In practice, this means that if you know how much coffee and water you are using, you can predict the subsequent brewing strength with a high degree of precision. The brewing strength, which is otherwise often ‘felt out’ through experience or device tuning, thus becomes a predictable variable - regardless of the grind, roasting depth or exact temperature (at least in the range between 80-99 °C).
Yield is stable - and more difficult to influence
While the brewing strength is strongly influenced by the ratio of water to coffee, the yield (i.e. the proportion of soluble components that are actually extracted) is remarkably stable. In the model, it is directly related to the equilibrium constant and the maximum possible extraction potential of the coffee - but remains independent of the brewing ratio. This is a paradigm shift: many believe that the yield can be ‘optimised’ by using more or less water. The model shows: You only change the concentration, not the proportion.
From a quality control perspective, this opens up enormous advantages. Instead of laboriously finding the ‘sweet spot’ of extraction by varying the amount of water, you can rely on the fixed points of the model - and use the brewing ratio to control the strength without risking unintentional changes to the extraction yield.
Caution with oven drying: the hidden source of error
One particularly interesting aspect of the model concerns the widely used method for determining extraction: so-called ‘gravimetry’ using oven drying. This involves measuring the mass of the coffee residue before and after drying in order to calculate the extracted mass. However, this is precisely where a systematic error lurks: some of the extracted liquid remains bound in the wet coffee crumbs and is not completely removed - not even by pressing or decanting.
The model proves that this retained liquid contains dissolved substances that are incorrectly categorised as ‘not extracted’. As a result, the measured yield is systematically lower than the actual yield. The stronger the coffee, the greater this effect. This is a crucial realisation for research and product development: anyone working on the basis of oven drying data regularly underestimates the actual yield.
Brew ratio as the primary control lever
The essence of the model calculation can be summarised in a simple formula: In full immersion extraction, the brew ratio is the most important control lever for controlling the TDS - and at the same time the least important for influencing the yield. This sounds counterintuitive at first, but opens up new methodological clarity.
The brewing ratio thus becomes a targeted control variable: more water per gram of coffee results in a lighter cup with lower TDS, less water leads to a stronger concentration - without necessarily over- or under-extracting. The quality of the extraction remains stable, the character of the cup can be varied. So if you want to produce different beverage styles (cold brew concentrate, ready-to-drink, filter, cupping) from the same bean and the same degree of roast, you can do so by adjusting the ratio - with a high degree of predictability and reproducibility.
Technology implications for sensor technology and product development
The practical implications are far-reaching. In sensor technology, the model allows the targeted preparation of samples with precisely defined brewing strengths - regardless of the device, grinding degree or other variables. New scope is created in product development: a cold brew concentrate with 3.5 % TDS can be produced just as precisely as a light filter coffee with 1.2 %. Without lengthy testing, without guesswork.
For industrial applications, such as the production of RTD products, the model provides a mathematical basis for scaling: what works in the laboratory also works in a 100-litre batch - as long as the ratio is maintained. This transferability is rare in the coffee world, where many processes are highly sensitive to environmental influences, bean properties or machine parameters.
Limits and potential
Like every model, this one also has its limits. It describes the equilibrium, not the dynamics. It can therefore only provide approximations for short brewing times, interrupted extraction or complex fermentation profiles. Microbiologically influenced extracts, for example from controlled fermentations, also sometimes deviate from the predicted behaviour.
Doch gerade in der Welt des klassischen Brewings – ob für Cuppings, Cold Brew, French Press oder automatisierte Batch-Brewer – liefert es eine neue Form von Kontrolle. Eine Kontrolle, die nicht auf Erfahrung beruht, sondern auf Physik. Und die aus einer jahrzehntelang geführten Erfahrungsdiskussion eine mathematisch lösbare Aufgabe macht.
Appendix– Literature
The essence of the model calculation can be summarised in a simple formula: In full immersion extraction, the brew ratio is the most important control lever for controlling the TDS - and at the same time the least important for influencing the yield. This sounds counterintuitive at first, but opens up new methodological clarity.
The brewing ratio thus becomes a targeted control variable: more water per gram of coffee results in a lighter cup with lower TDS, less water leads to a stronger concentration - without necessarily over- or under-extracting. The quality of the extraction remains stable, the character of the cup can be varied. So if you want to produce different beverage styles (cold brew concentrate, ready-to-drink, filter, cupping) from the same bean and the same degree of roast, you can do so by adjusting the ratio - with a high degree of predictability and reproducibility.
Technology implications for sensor technology and product development
The practical implications are far-reaching. In sensor technology, the model allows the targeted preparation of samples with precisely defined brewing strengths - regardless of the device, grinding degree or other variables. New scope is created in product development: a cold brew concentrate with 3.5 % TDS can be produced just as precisely as a light filter coffee with 1.2 %. Without lengthy testing, without guesswork.
For industrial applications, such as the production of RTD products, the model provides a mathematical basis for scaling: what works in the laboratory also works in a 100-litre batch - as long as the ratio is maintained. This transferability is rare in the coffee world, where many processes are highly sensitive to environmental influences, bean properties or machine parameters.
Limits and potential
Like every model, this one also has its limits. It describes the equilibrium, not the dynamics. It can therefore only provide approximations for short brewing times, interrupted extraction or complex fermentation profiles. Microbiologically influenced extracts, for example from controlled fermentations, also sometimes deviate from the predicted behaviour.
Doch gerade in der Welt des klassischen Brewings – ob für Cuppings, Cold Brew, French Press oder automatisierte Batch-Brewer – liefert es eine neue Form von Kontrolle. Eine Kontrolle, die nicht auf Erfahrung beruht, sondern auf Physik. Und die aus einer jahrzehntelang geführten Erfahrungsdiskussion eine mathematisch lösbare Aufgabe macht.
Appendix– Literature
- Liang, J., Chan, K. C., & Ristenpart, W. D. (2023). An Equilibrium Desorption Model for the Strength and Extraction Yield of Full Immersion Brewed Coffee. Research Square. https://www.researchsquare.com/article/rs-127037/v1
- Liang, J., Batali, M. E., Routt, C., Ristenpart, W. D., & Guinard, J.-X. (2024). Sensory analysis of the flavor profile of full immersion hot, room temperature, and cold brewed coffee over time. Scientific Reports. https://doi.org/10.1038/s41598-024-69867-6
- Guinard, J.-X., Frost, S., Batali, M., Cotter, A., Lim, L. X., & Ristenpart, W. D. (2023). A new Coffee Brewing Control Chart relating sensory properties and consumer liking to brew strength, extraction yield, and brew ratio. Journal of Food Science, 88, 2168–2177. https://doi.org/10.1111/1750-3841.16531
Author:
Dr. Steffen Schwarz
Dr Steffen Schwarz is an internationally recognised expert in the field of coffee, an entrepreneur and a lecturer.
As the founder of the Coffee Consulate in Mannheim, he has for many years been dedicated to the training and further education of professionals across the entire coffee value chain. His work combines scientific insights with practical experience and covers topics such as coffee quality, sensory analysis, sustainable cultivation, and the processing and preparation of coffee.
Drawing on his many years of expertise, he advises companies worldwide, develops training programmes for the coffee industry and is committed to the responsible and quality-oriented handling of coffee. As an author, Dr Steffen Schwarz conveys complex specialist knowledge in an accessible and practical way, offering his readers new perspectives on one of the world’s most fascinating luxury foods.