Physico-elemental analysis of roasted organic coffee beans from Ethiopia, Colombia, Honduras, and Mexico using X-ray micro-computed tomography and external beam particle induced X-ray emission

Abstract

The physico-elemental profiles of commercially attained and roasted organic coffee beans from Ethiopia, Colombia, Honduras, and Mexico were compared using light microscopy, X-ray micro-computed tomography, and external beam particle induced X-ray emission. External beam PIXE analysis detected P, S, Cl, K, Ca, Ti, Mn, Fe, Cu, Zn, Br, Rb, and Sr in samples. Linear discriminant analysis showed that there was no strong association between elemental data and production region, whilst a heatmap combined with hierarchical clustering showed that soil-plant physico-chemical properties may influence regional elemental signatures. Physical trait data showed that Mexican coffee beans weighed significantly more than beans from other regions, whilst Honduras beans had the highest width. X-ray micro-computed tomography qualitative data showed heterogeneous microstructural features within and between beans representing different regions. In conclusion, such multi-dimensional analysis may present a promising tool in assessing the nutritional content and qualitative characteristics of food products such as coffee.

Keywords: Colombia; Ethiopia; External beam PIXE; Honduras; LDA, linear discriminant analysis; Mexico; Organic; PIXE, particle induced X-ray emission; Roasted coffee bean; X-ray micro-computed tomography; micro-CT, X-ray micro-computed tomography.

Graphical abstract

Fig. 1

Elemental distribution maps of K, Ca, and Fe of a hand-sectioned roasted organic coffee bean (top image). The intensity scale in maps was selected to emphasize certain elemental distribution features.

Fig. 2

A: Heatmap generated using complete linkage hierarchical clustering based on Euclidean distances showing the median elemental abundances (quantified as moderated z-scores) of clusters for mineral compositions of roasted organic beans from different regions. The relationship is presented by a dendrogram in which rows represent region and columns minerals, with a specific colour representing the magnitude of abundance (please refer to the text for a full explanation of the heatmap results). B: Linear discriminant analysis (LDA) to separate geographical lines by macro-elemental content (left) and micro-elemental content (right). LDA was performed in which linear discriminants were computed to represent the axes that maximize separation between multiple classes and show patterns encoding the highest variance in the data set together with the variables linked to such patterns.

Fig. 3

X-ray micro-computed tomography image of a coffee bean showing its microstructure in terms of porosity.

Fig. 4

Two-dimensional images of sagittal and transverse plane sections of roasted organic beans from Mexico (top row), Colombia (middle row), and Honduras (last row). Images in colour clearly indicate areas of high (red) or low (blue) pore interconnectivity, whilst grey-scale images reveal pore features. The smaller images show corresponding three-dimensional reconstructed images and the plane selected for 2D imaging. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

X-ray micro-computed tomography images of roasted organic beans from Ethiopia. The sagittal plane 3-D image (top image) shows pore interconnectivity (in colour) and pore features. The sagittal and transverse plane 2D colour images (row A) clearly indicate areas of high (red) or low (blue) pore interconnectivity, whilst grey-scale images (row B) distinguish pore features. The smaller images (in row B) show corresponding three-dimensional reconstructed images and the plane selected for 2D imaging. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)