Abstract
The effect of degree of roast on resulting caffeine content in brewed coffee has been extensively researched, but conflicting methodologies and results have muddled development of a general conclusion. In this study, 30 unique combinations of green coffee variety, degree of roast, and brew time were investigated regarding extraction yield and caffeine content. An AeroPress brewer was used to prepare brew samples using a 15:1 mass ratio of brew water to ground coffee. Refractometry and HPLC were respectively used to measure extraction yield and caffeine content of brewed samples. Scanning electron microscopy was used to measure porosity of roasted seeds and showed increasing porosity with the degree of roast. Extraction yields generally decreased for roast batches with roasting mass losses greater than (\sim)12–14%, and caffeine concentrations in 10-min brews decreased for roast batches with drop temperatures greater than (\sim)400–420 (^{\circ })F. Under identical brewing conditions, caffeine concentrations in brewed samples were generally lower for dark roasts than light and medium roasts. However, at identical extraction yields, dark roasts generally exhibited higher caffeine concentrations than lighter roasted coffees. It is likely that the volatilization or decomposition of soluble compounds and increased porosity due to roasting act as competing mechanisms that determine compound concentrations in resulting brews.
I prefer to stick to lighter roasts because they taste better IMO. The resulting cup tends to contain more caffeine too, which is good or bad depending on your relationship with it.
BTW, what do you mean by “forcing extraction”?
I prefer lighter roasts for flavor too, brewed and held around 200F so it maintains the floral notes. As far as 'forcing extraction', I mean in something like an aeropress or a french press, rather than percolation.