Under pressure: poroelastic regulation of flow in espresso brewing

TL;DR

This work shows that espresso extraction is controlled by a poroelastic coupling between the dissolving coffee bed and the flowing liquid, producing a non-linear, pressure-dependent flow that saturates at high driving pressures. The authors develop a minimal quasi-static poroelastic model and a time-dependent extension that ties bed porosity to the amount dissolved, validating both with experiments on a café-grade machine and time-resolved solute measurements. They demonstrate that dissolution dynamics largely govern the temporal evolution of flow, linking early wetting and swelling to later extraction, and providing a framework to predict flow and solute transfer by coupling elasticity with mass transfer. The results imply that optimizing espresso, and similar porous-media flows, requires accounting for reactive dissolution and bed deformation rather than assuming constant permeability.

Abstract

The sensory richness of coffee is widely recognised and arises from the complex chemistry and immersion in cultural practices of coffee preparation. In contrast, the physical complexity of espresso has received less attention. The multiphase reactive flow through a dissolving, elastic porous medium remains challenging to describe. Using a controlled experimental setup based on a café-grade espresso machine, we demonstrate that the interplay between elasticity and porosity governs the long-time flow rate during espresso extraction and, consequently, the concentration of solubles in the final beverage. We introduce a minimal model that captures the resulting non-linear pressure-flow relationship and propose a methodology capable of reproducing the time-dependent behaviour of the espresso brewing process. Finally, we show that dissolution dynamics play a central role in determining the temporal evolution of flow during extraction.

Figures (10)

• Figure 1: The espresso brewing process. First the weighed grounds are placed in the brewing chamber (called portafilter), then clumps are removed with a needle tool (panel (A), necessary for good reproducibility). Then the grounds are tamped level with a tamper (panel (B), manual tamper pictured for demonstration only) to ensure consistent preparation pressure, the espresso can then be brewed (panel (C)). (D) Simplified diagram of the heat exchange espresso machine (typical for commercial cafés). Water flows from the pump via pressure regulator and flow meter, then through a heat exchanger (large amount of water in a boiler is kept at high temperature (typically 200 ℃) ensuring repeatable conditions) and into the brewing group. Inside the brewing group the water passes through a shower screen then the coffee grounds and filter basket sieve. Finally coffee ends up in the cup resting on digital scales. Data from digital pressure gauge, flow meter and scales were collected simultaneously.
• Figure 2: Brewing assembly pressure drop calibration.(A) Espresso machine with empty basket with second pressure sensor and control valve added below the portafilter. (B) Combined measurements from multiple series and fitted quadratic approximation. Two with a control valve sweep and two with pressure regulator sweep. Brewer pressure drop depends only on the flow rate through it. Brewer pressure drop, $p_1-p_2$, as a function of the flow rate can be then used to obtain basket pressure from driving pressure.
• Figure 3: Particle abundance as a function of size. The coffee grounds show a characteristic bimodal distribution with smaller component called "fines" of typical size of 50 $\mu$m and larger grounds of typical size below 200 $\mu$m.
• Figure 4: Theoretical equilibrium flow rate as a function of pressure Full theoretical expression \ref{['eqn:hat_solution_long']} for normalised flow rate $\hat{Q}$ as a function of normalised pressure $\hat{P}$ at several values of control parameter $\Phi$ together with limit $\Phi \to 0$. Within the range of possible values of $\Phi$ the dependence is negligible.
• Figure 5: Experimental data on espresso flow at controlled pressures.(A) Basket pressure as function of time. (B) Simultaneously measured mass in cup. (C) Derived mass flow rate as a function of time. At low pressures (dark blue) increase in pressure results in an increase in flow rate, this is contrasted with opposite trend for large pressures (light yellow). Each line is an average result of multiple experiments with shaded regions denoting standard deviation of the mean.