Coffee brewing is a complex process that involves extracting soluble compounds such as caffeine, sugars and acids from roasted coffee beans. One of the important factors for the extraction of these compounds and the exquisite taste of coffee is the effect of the brewing water composition.
Lockhart's famous 1957 pioneering work relates the quality of brewed coffee to two main metrics: "brew strength," the mass fraction of dissolved solids in the brewed liquid (usually characterized as total dissolved solids), and "extraction efficiency," the mass fraction of dissolved solids removed from the coffee beans (usually referred to as percent extraction). These quantities are linked through conservation of mass in relation to the water/coffee bean ratio and are combined by Lockhart in his classic "Coffee Brewing Control Chart" (below).
In the graph, the TDS labeled as "strong" or "weak" represents the opposite vertical boundaries, while the PE labeled as "bitter" / "improved" (mass fraction of soluble solids extracted from coffee grounds, percent extraction, more clearly the ratio of coffee extract extracted during a coffee brewing process to the total weight of coffee used) represents the opposite horizontal boundaries, and the graph is divided into 9 regions. The region in the center in terms of TDS values, with TDS values of 1.15-1.35% and PE values of 18-22%, is indicated as "ideal" and is defined as the "Golden Cup Standard" of the Specialty Coffee Association. Thus, let's summarize the "Golden Cup Standard" that everyone has been working on before. Although the Coffee Brewing Control Chart is widely used, it is problematic from the perspective of modern sensory methodology. This graph is not compatible with hedonic or quality descriptors ("ideal"), scale measures ("weak", "strong") and descriptive language (“bitter”). It is also completely unclear which consumers consider the coffees in the center of the chart to be “ideal.” Perhaps most importantly, this chart leaves out the wide variety of sensory attributes possible in coffee. The Coffee Flavor Wheel from World Coffee Research includes over 100 different flavor and aroma attributes that have been identified in roasted coffee. Many of these are desirable (e.g., caramel, fruity, dark chocolate, nutty) and undesirable (e.g., rubbery, medicinal, musty). Green and roasted coffees have been chemically characterized in many studies, and it is now common knowledge that coffee flavor and aroma are a complex mixture of carbohydrates, acids, lipids, proteins, antioxidants, and volatile aroma compounds. These components can be manipulated at every step, from the source, through the terroir of the coffee cherry (we might call this the soil stage), through post-harvest processing methods, to roasting, and finally to brewing methods. In addition, brewing temperature is also thought to affect coffee flavor and aroma. However, in the current literature, the number of studies investigating the effect of coffee brewing temperature on flavor characteristics is limited. Therefore, experts state that the temperature range of 92-96 °C determined for coffee preparation temperature does not provide data on why these temperatures are preferred. It has been reported that coffees served at higher temperatures are perceived as more roasted and bitter than coffees served at lower temperatures, while the perception of most other taste notes is significantly reduced at higher serving temperatures. In fact, since the studies are still insufficient, we can scientifically comment that high temperatures increase dissolution – as is the case for many substances, excluding exceptions. However, gustatory analyses are unfortunately quite unique, personal and variable analyses. Therefore, the concept of “ideal” that we have emphasized in previous articles is invalid.
In this study, roasted coffee was prepared according to Lockhart's table at nine different positions and three different water temperatures (87 °C, 90 °C and 93 °C) and a total of 27 sample types were obtained. In this study, the same roasted coffee was used, the same type of brewing equipment was used, the extraction time, flow rate, dosage (i.e. the amount of ground coffee) and grinding size were systematically changed at different brewing temperatures and controlled at the same TDS, PE and serving temperatures. In this study, where five-way ANOVA analysis (which is a statistical analysis method and measures the effect of five factors simultaneously. These factors are usually defined as variables of an experimental design) was performed, it was seen that 17 out of 31 taste notes tested were significantly different depending on TDS, PE, brewing temperature and/or TDS x PE interaction. The study aimed to show how the taste and aroma of coffee change depending on factors such as TDS (total dissolved solids), PE and brewing temperature.
The results that temperature has a negligible effect on perceptible sensory quality in the ranges measured in the study, when TDS and PE are kept constant, are quite striking for us coffee makers.
Spider web plots visualizing differences in feature intensities by all variables significant for at least one of the three factors for (A) TDS, (B) PE, and (C) brew temperature. Asterisks indicate statistical significance (p < 0.05). The inner ring corresponds to a feature intensity of 0, while the outer ring corresponds to a feature intensity of 50; intermediate rings are equally spaced in increments of 5 intensity points.
To summarize the above table, it was revealed that 17 different attributes differed significantly for at least one factor. These attributes can also be visualized using spider web plots to show the average differences for overall TDS, PE and temperature (Figure 4). It should be noted that the shape of the spider web plots changed significantly for different TDS values (Figure 4a). The most significant changes were for Bitter , Sour and Astringent tastes, while smaller but statistically significant changes were observed for the other 13 attributes. The shape of the spider web plots changed less for different PE values (Figure 4b). The largest differences were observed for Sour , Black Tea and Citrus tastes. Finally, almost no differences were observed in the shape of the spider web plots for different brewing temperatures (Figure 4c). Only the Hazelnut taste note showed a slight but statistically significant change with temperature.
TDS is the factor with the greatest impact on sensory attributes, especially when visualized (table above), while PE has a lesser but noticeable sensory impact. In contrast to the lack of an effect of brewing temperature in the study, the results confirm that both TDS and PE have significant effects on the sensory attributes of the brew.
The figure below shows that out of the 31 tested sensory attributes, only one, at 90 °C water temperature, showed a very slight increase compared to the other temperature values, namely the perception of hazelnut taste. However, this difference is very small compared to the significant differences recorded in the other attributes, with only one point difference on a scale of 100. Although the difference in hazelnut taste was found to be statistically significant (p < 0.05), further testing is required to confirm that this result is not simply a type I error.
Perhaps the most important part of the study can be described as the “Ideal” box in the classic Coffee Brewing Control Chart. The point to be emphasized is that the study is entirely quantitative (i.e. hedonic preference rather than sensory descriptive analysis). However, previous studies have identified some characteristics that determine the reasons for consumers’ likes or dislikes. According to a previous study, it was shown that most consumers do not like bitterness and harshness in coffee products, while a floral aroma has a positive effect. This situation shows that the lower third of the coffee brewing control chart (the “Weak” part) may be more “ideal” for many consumers. However, the same study revealed that intense “roasted” aroma is also one of the reasons for liking, while other studies have indicated that there are important consumer groups that are interested in bitter and strong coffee.
Although there is relatively little data on consumer preferences for specific flavors in coffee, these observations suggest that the “ideal” box is not ideal for everyone and thus may not be a useful descriptor for coffee brewing.
The results also suggest that the current industry guidelines’ focus on brewing temperature may be misguided. For example, home brewers currently fail certification if they can’t maintain their temperatures in the 92-96°C range, but the data shows that when grind size, brew ratio, and flow rate are adjusted to hold TDS and PE constant, coffee brewed at 87°C is sensorially indistinguishable from coffee brewed at 93°C.
One of the areas that is particularly emphasized in this study is that lower brewing temperatures also have significant consequences in terms of environmental sustainability, which we constantly emphasize as a company. Since a cup of coffee is approximately 99% water, the energy required to heat the water represents a large portion of the overall energy use throughout the entire coffee supply chain. It has been shown that energy use in a café accounts for 45% of the overall carbon dioxide emissions of brewed coffee production, even when all other steps are taken into account, including farming, post-harvest processing, and roasting on the farm. Reducing the average brewing temperature can have a disproportionate impact on overall sustainability efforts in coffee production. The results presented here demonstrate that improvements in consumer health and the global environment can be achieved without sacrificing the sensory quality of brewed coffee. Overall, the study yielded two main results. First, in the tested range, brewing water temperature was found to play a minimal role in coffee sensory attributes as measured by a trained descriptive analysis panel. Secondly, by reporting how certain sensory properties change with brewing parameters, it was noticed that the differences in total dissolved solids in brewing were more pronounced than the differences in percentage removal (TDS-PE difference).
The results revealed in the study not only enable coffee professionals to better reach the flavors they are looking for, but also contribute significantly to the sustainability effort towards minimizing energy during brewing.
DUYGU KURTULUŞ
Co-Founder / Chemist / Nanotechnology Engineer / Hazardous Chemical Consultant / Chemical Evaluation Specialist