The final quality of espresso coffee (EC) depends upon certain technical conditions, such as the extraction temperature used in preparing it. The aim of this work was to investigate the effects of water temperature (88, 92, 96 and 98°C) on the final quality of three types of EC (Arabica, Robusta Natural blend and Robusta Torrefacto blend) in order to select the optimal temperature. Volatile compound (analysed by Static headspace gas chromatography/mass spectrometry) and sensory flavour profiles were the most relevant parameters, whereas physicochemical, taste and mouthfeel parameters were not very useful for selecting the water temperature. For Arabica and Robusta Natural blend ECs, 92°C was the optimal water temperature. For Robusta Torrefacto blend EC the overall acceptability might lead to the selection of 88°C as the ideal water temperature, but the high percentages of key odorants related to roasty and earthy/musty flavours and the 'not hot enough' perception dictated the selection of 92°C in this case as well.

Three espresso coffee (EC) samples of different botanical varieties and types of roast were prepared in standard conditions using an experimental EC prototype: Arabica coffee, Robusta Natural blend, and Robusta Torrefacto blend (a special roast by adding sugar). The ECs were characterized with regard to the physical parameters, amount of total solids, total solids on filtrate, lipids, caffeine, trigonelline, and chlorogenic acids by HPLC, and sensory descriptive analysis related to foam appearance, taste, and mouthfeel. Principal component analysis (PCA) was applied to differentiate the EC samples. Arabica and Robusta samples were separated successfully by principal component 1 (55.3% of variance) including physicochemical and sensory parameters related to foam and taste of ECs. Torrefacto and Robusta Natural EC samples were separated by principal component 2 (20.7% of total variance) including mouthfeel and other attributes of color foam. Some interesting correlations among sensory and physicochemical variables were found. A very simple discriminate function was obtained by discriminate analysis allowing the classification of each EC sample into its respective group with a success rate of 100%.

No abstract

No abstract

No abstract

No abstract