Extraction Yield (EY)

Coffee grounds contain approximately 30% soluble material (the other 70% is insoluble cellulose and structural compounds). However, only 18% to 22% of the total mass produces balanced, desirable flavors. EY is calculated using TDS and brew ratio: EY% = (TDS% x beverage mass) / dose mass. For example, if you brew 15g of coffee and produce 250g of beverage with 1.35% TDS: EY = (1.35 x 250) / 15 = 22.5%. This would be slightly over-extracted by SCA standards. Extraction is not uniform: different compounds extract at different rates. The order of extraction is: acids and salts first (sour), then sugars (sweet), then bitter compounds. This is why under-extracted coffee tastes sour (acids extracted, sugars not yet) and over-extracted coffee tastes bitter (bitter compounds over-extracted).

The compounds extracted from coffee, in rough order of extraction rate, include: 1. Acids (citric, malic, acetic, chlorogenic): extracted first, contributing sourness and brightness. 2. Salts and minerals (potassium, magnesium): extracted early, contributing to overall flavor perception. 3. Sugars (sucrose, glucose, fructose): extracted in the middle phase, contributing sweetness. 4. Lipids (triglycerides, diterpenes): extracted throughout, contributing body and mouthfeel. 5. Melanoidins: extracted later, contributing body, color, and bitterness. 6. Caffeine: extracted throughout, contributing bitterness. 7. Bitter polyphenols and tannins: extracted last, contributing astringency and bitterness. This extraction order explains why short shots (ristretto) can taste sweet and balanced while long shots (lungo) can taste bitter.

The physics of Extraction Yield (EY) is characterized as a mass transfer process involving the migration of solutes from a solid matrix into a liquid solvent. This occurs through two primary mechanisms: advection and diffusion. Extraction typically follows a two-phase kinetic model. The initial phase, 'surface wash' or erosion, involves the rapid dissolution of compounds from cells mechanically ruptured during the grinding process. The second, slower phase is 'inter-particle diffusion,' where water must penetrate the complex, porous cellulose structure of the coffee grounds to reach internal solutes. This diffusion rate is described by Fick’s First Law, which posits that the flux of solutes is proportional to the concentration gradient between the coffee particle and the surrounding water. Grind size acts as a primary physical lever; a finer grind increases the total surface area-to-volume ratio and reduces the diffusion path length, thereby accelerating the rate at which EY is achieved. Additionally, temperature influences the process by increasing the kinetic energy and the diffusion coefficient of the water molecules. In percolation methods, fluid dynamics—specifically the 'boundary layer' around each coffee particle—play a role; moving water (advection) helps maintain a high concentration gradient by sweeping extracted compounds away from the particle surface, preventing local saturation and allowing extraction to continue efficiently.

EY represents the efficiency and completeness of extraction. The 18% to 22% range was established by the Coffee Brewing Center through consumer preference testing in the 1950s and 1960s. The curve of extraction vs preference is a bell curve: preference increases from 16% to 20%, peaks around 20%, and decreases from 20% to 24%. EY is affected by grind size (finer = more surface area = higher EY), water temperature (hotter = faster extraction = higher EY), brew time (longer = higher EY), agitation (more = higher EY), and water chemistry (higher mineral content can increase extraction). Channeling (water finding preferential paths through the coffee bed) causes uneven extraction, with some areas over-extracted and others under-extracted, resulting in a combined EY that can be within range while the cup tastes unbalanced.

EY% = (TDS% x beverage mass) / dose mass. Target range: 18 to 22% (SCA drip standard). Ristretto: 18 to 21%. Normale espresso: 18 to 22%. Lungo: 18 to 22% (but with different TDS). French press: 18 to 22%. EY is a single number that does not capture extraction uniformity. Two coffees with identical EY can taste different if one had uniform extraction and the other had channeling. EY should be used as a diagnostic tool alongside sensory evaluation, not as a replacement for tasting. Calculation example: 18g dose, 36g espresso, 10% TDS. EY = (10 x 36) / 18 = 20.0%.

If your coffee tastes sour or weak, your EY is likely below 18%. Fix this by grinding finer, increasing water temperature, or brewing longer. If your coffee tastes bitter or astringent, your EY is likely above 22%. Fix this by grinding coarser, lowering water temperature, or brewing shorter. Measure EY with a refractometer: record TDS, then calculate EY using the formula. Use EY alongside tasting to dial in recipes.

• •Higher extraction yield is always better. In reality, the ideal range is 18% to 22%. Beyond 22%, bitter and astringent compounds dominate.
• •EY alone determines coffee quality. In reality, EY measures extraction efficiency but not uniformity. Channeling can produce a balanced EY with an unbalanced cup.
• •You cannot have high TDS and low EY. In reality, ristrettos often have high TDS (10%+) but EY can be as low as 15% because less water passes through the coffee.

• •Lockhart (1955) established the 18% to 22% ideal extraction range through consumer preference panels, showing a bell curve of preference peaking near 20%.
• •Research by the Coffee Science Foundation has shown that different compounds extract at different rates, explaining the sour-to-bitter progression of extraction.
• •Studies on espresso extraction (Fasano et al., 2020) demonstrated that channeling causes simultaneous over- and under-extraction within the same puck.
• •Moroney et al. (2019) developed mathematical models of coffee extraction that predict EY based on grind size, water temperature, and flow rate.

• •Chapter 5: Extraction Science
• •Chapter 6: Brewing Methods

• •Lockhart, E.E. (1955). 'The Soluble Solids Content of Coffee Beverages.' Coffee Brewing Institute.
• •Moroney, P.F. et al. (2019). 'Mathematical Modelling of Coffee Extraction.' Journal of Mathematical Industry.
• •Fasano, A. et al. (2020). 'The Mathematics of Espresso Extraction.' SIAM Journal on Applied Mathematics.
• •SCA (2019). 'Coffee Brewing Standards.' Specialty Coffee Association.

• •Specialty Coffee Association
• •Coffee Science Foundation
• •Coffee Brewing Center
• •James Hoffmann