Discrimination of Green Arabica and Robusta Coffee Beans by Raman Spectroscopy

Abstract

This paper presents an approach that may be applied as an accurate and rapid tool for classifying coffee beans on the basis of the specific kahweol content. Using Fourier-transform Raman spectroscopy with 1064 nm excitation it is possible to monitor the characteristic Raman bands of kahweol in green coffee beans without chemical and physical processing of the beans. The procedure was optimized on the basis of 83 and 125 measurements of whole and ground beans, respectively, using coffee samples of two different species, Coffea arabica L. and Coffea canephora L. (var. Robusta), and different origins (Asia, Africa, and South America). The relative contribution of the kahweol in individual beans can be determined quantitatively by means of a component analysis of the spectra, yielding a spectral kahweol index (σka) that is proportional to the relative content of kahweol in a coffee bean. The reproducibility of the spectroscopic measurement and analysis was found to be 3.5%. Individual beans of the same type and origin reveal a scattering of the σka values. Nevertheless, an unambiguous distinction between Arabica and Robusta samples is possible on the basis of single-bean measurements as the σka values are greater than and less than 10 for Arabica and Robusta coffees, respectively. Measurements of whole and ground beans afforded very similar results, despite the heterogeneous distribution of kahweol within a bean. Unlike conventional analytical techniques, the single-bean sensitivity of the present approach may also allow for a rapid detection of unwanted admixtures of low-value Robusta coffee to high-quality and more expensive Arabica coffee.

Looking into individual coffee beans during the roasting process: direct micro-probe sampling on-line photo-ionisation mass spectrometric analysis of coffee roasting gases

“…The diterpenes kahweol (m/z 314), cafestol (m/z 316), and 16-O-methylcafestol (m/z 330) enable a discrimination of Arabica and Robusta cultivars: While Arabica contains kahweol and cafestol, Robusta solely contains cafestol and 16-O-methylcafestol [11,42,43]. However, these compounds are of rather low volatility and stability and thus are difficult to be analysed on-line in a μ-probe measurement.…”

“…Higher Lower Furan derivatives (4.20-9.65 g/100 g) (1.66-3.53 g/100 g) (98, 126m/z) → higher abundant in Arabica Lipids [47] Higher Lower Fatty acids (12.8-16.2 g/100 g) (6.70-8.70 g/100 g) (256, 280, 284, 312m/z) → higher abundant in Arabica Sulphur compounds [44] Lower Higher Methanethiol (CH 3 SH 48m/z) → higher abundant in Robusta α-Dicarbonyls [44] Higher Lower Carboxylic compounds (42, 70m/z) → higher abundant in Arabica Cafestol [11,42,43] Occurs Occurs Anhydrous cafestol [43] (298m/z) → higher abundant in Arabica Kahweol [11,42,43] Occurs Does not occur Anhydrous kahweol [43] (296m/z) → present in Arabica solely different roasting times, which were observed during cracking of the beans. Phenol represents a relatively small molecule with high volatility such as acetaldehyde, acetone, pyridine or furfuryl alcohol, which show a similar behaviour.…”

“…Several studies used on-line spectroscopic methods such as Infrared [9,10] and Raman spectroscopy [11,12]. Gas sensors or sensor arrays ("electronic noses") have been applied as well [11,13,14]. "Electronic noses" consist of an array of several, more or less selective and sensitive, gas sensors combined with a "pattern recognition type"-multivariate statistical analysis of the data.…”

“…On-line real-time analytical methods inherently are better suited to follow the fast dynamics of the roasting process. Several studies used on-line spectroscopic methods such as Infrared [9,10] and Raman spectroscopy [11,12]. Gas sensors or sensor arrays ("electronic noses") have been applied as well [11,13,14].…”

Green Synthesis of Metal Nanoparticles via Natural Extracts: The Biogenic Nanoparticle Corona and Its Effects on Reactivity

“…The spectra in Figure 5 display signatures characteristic of esters (1760-1730 cm -1 ), polyphenols and chlorogenic acids (1660-1600 cm -1 ), coffeespecific diterpenes (1567 and 1480 cm -1 ) and aliphatic groups (1450, 1375 and 1263 cm -1 ). [46][47][48] Peaks were found to disappear within the time needed for recording ~2 spectra (2 s); this could be due to beam damage of coffee components during Raman, as previously reported, 46 or to heating effects of laser illumination. To confirm that the observed peaks result from surface-enhanced Raman spectroscopy (SERS) of organics adsorbed onto Ag, we carried out control experiments using carbon particles incubated in coffee and subsequently rinsed, which did not yield spectra as the ones in Figure 5.…”