Integrated Physicochemical, Spectroscopic, Granulometric, and Sensory Dataset of Single-Serve Coffee Capsules in Espresso and Lungo Preparations
Description
This dataset contains the experimental results obtained from the physicochemical, spectroscopic, granulometric, and sensory characterization of single-serve coffee capsules compatible with the Nespresso® system and intended for Espresso and Lungo preparations. The dataset includes eight Espresso references and ten Lungo references, providing an integrated analytical characterization of 18 commercial coffee products. Physicochemical measurements include headspace gas composition (O₂, CO₂, and N₂), initial and final headspace pressure, water activity (aw), sample temperature, and CIELAB color parameters (L*, a*, and b*). These measurements were performed on three independent capsules per commercial reference, resulting in 54 experimental units. Particle size distribution was determined by laser diffraction and includes D10, D50, D90, mean and modal particle size, span, coefficient of variation, surface area, and particle-size fractions, including the proportion of fine particles below 100 µm. Granulometric data are provided for the 18 commercial coffee references together with calibration measurements. The spectroscopic component comprises FT-NIR and ATR-FTIR data. The FT-NIR dataset contains 54 spectra acquired from 14,000 to 4,000 cm⁻¹ with 5,001 spectral points per spectrum. The ATR-FTIR dataset covers approximately 650–4,000 cm⁻¹ and contains 162 spectra corresponding to 18 coffee references, three independent capsules per reference, and three spectral measurements per capsule. Sensory data were obtained from three experienced coffee assessors. The evaluation includes descriptive intensity scores for fragrance, aroma, flavor, aftertaste, acidity, sweetness, and mouthfeel, together with affective quality scores for these attributes and overall impression. Descriptive intensity was evaluated on a 0–15 scale, while affective quality was assessed on a 1–9 scale. The dataset is intended to support research on the relationships among coffee particle size distribution, physicochemical properties, spectral fingerprints, packaging headspace conditions, and sensory quality in single-serve coffee systems. It can also be used for multivariate statistical analysis, chemometrics, principal component analysis, clustering, classification, and the development of predictive models based on FTIR and FT-NIR spectroscopy.
Files
Steps to reproduce
To reproduce the dataset, select 18 commercial single-serve coffee capsule references compatible with the Nespresso® system, comprising eight Espresso and ten Lungo products. Samples should be coded consistently by preparation type, commercial reference, and capsule replicate (e.g., E-S1-C1 or L-S1-C1). For physicochemical characterization, analyze three independent capsules per commercial reference, resulting in 54 experimental units. Measure the headspace gas composition (O₂, CO₂, and N₂), initial and final internal pressure, water activity (aw), sample temperature, and CIELAB color parameters (L*, a*, and b*). Maintain the measurements associated with each individual capsule to preserve replicate-level traceability. For particle-size characterization, analyze the roasted and ground coffee using laser diffraction with a Beckman Coulter LS 13 320 particle-size analyzer. Record the complete particle-size distribution and calculate D10, D50, D90, mean particle size, modal particle size, span, coefficient of variation, surface area, and particle-size fractions, particularly the fraction below 100 µm. The final dataset contains granulometric measurements for the 18 commercial coffee references, together with instrument calibration records. For FT-NIR analysis, acquire spectra for three independent capsules from each commercial reference. Spectra should be collected in diffuse reflectance mode over the range of 14,000–4,000 cm⁻¹, generating 5,001 spectral points per spectrum. This produces 54 FT-NIR spectra corresponding to 24 Espresso and 30 Lungo samples. For ATR-FTIR analysis, collect spectra from roasted ground coffee using an ATR-FTIR spectrometer over approximately 650–4,000 cm⁻¹. Analyze three independent capsules per commercial reference and acquire three spectral measurements per capsule. After excluding non-experimental test spectra and correcting sample identifiers, the final matrix should contain 162 spectra, corresponding to 18 commercial references × 3 capsules × 3 spectral replicates, with 3,596 wavenumber points per spectrum. For sensory analysis, prepare the beverages according to their intended extraction format: Espresso and Lungo. Evaluate each commercial reference using three experienced coffee assessors. Record descriptive intensity scores for fragrance, aroma, flavor, aftertaste, acidity, sweetness, and mouthfeel using a 0–15 scale. Record affective quality scores for the same attributes plus overall impression using a 1–9 scale. Keep assessor-level observations in long format to allow evaluation of sample and assessor effects. Finally, organize all analytical results using the same sample coding system so that physicochemical, granulometric, spectroscopic, and sensory data can be linked across commercial references.
Institutions
- Universidad SurcolombianaHuila Department, Neiva