Procedia Engineering 47 ( 2012 ) 977 – 980
1877-7058 © 2012 Elsevier Ltd....Selection and/or peer-review under responsibility of the Symposium Cracoviense Sp. z.o.o.
Proc. Eurosensors XXVI, September 9-12, 2012, Kraków, Poland
Chemometric Discrimination of Philippine Civet Coffee
Using Electronic Nose and Gas Chromatography Mass
E. Ongoa,b, M. Falasconic, G. Sberveglieric, A. Antonellid, G. Montevecchid,
V. Sberveglierid, I. Concinac, F. Sevilla IIIb*
Industrial Technology Development Institute, DOST, Gen. Santos Ave., Bicutan, Taguig City, 1631 Philippines
University of Santo Tomas, Graduate School , Espana Blvd Blvd., Sampaloc, Manila, 1008 Philippines
CNR-IDASC SENSOR & Brescia University, Via Valotti 9, 25133 Brescia, Italy
Dipartimento di Scienze-Agrarie e degli Alimenti, Universita di Modena e Reggio Emilia, Via, G. Amendola 2, 42122 Reggio
This study presents a practical and promising approach to profile the headspace aroma attributes of Philippine
civet coffee using electronic nose (E-nose) and gas chromatography mass spectrometry (GCMS). Chemometric
pattern method was applied to enhance the discrimination of civet coffee against its control coffee beans (not eaten by
civet animal). E-nose analysis revealed that aroma characteristic is one of the most important quality indicators of
civet coffee. The result was supported by GCMS analysis. The chromatographic fingerprints indicated that civet
coffee differed with their control beans in terms of composition and concentration of individual volatile constituents.
Chemometric discrimination of E-nose and GCMS data demonstrated a clearly separated civet from their control
coffees indicating that cultivar and geographic origins dictate the aroma and volatiles variations in coffee.
© 2012 Published by Elsevier Ltd.
© 2012 Elsevier Ltd....Selection and/or peer-review under responsibility of the Symposium Cracoviense Sp. z.o.o.
Keywords: civet coffee: electronic nose; gas chromatography mass spectrometry; chemometric discrimination
Civet coffee is the top most expensive coffee in the world . It is made from coffee cherries, which
have been eaten and passed through the digestive tract of Asian palm civet (Paradoxurus
* Corresponding author. Telefax.: +632-7314031
E-mail address: email@example.com
Available online at www.sciencedirect.com
978 E. Ongo et al. / Procedia Engineering 47 ( 2012 ) 977 – 980
hermaphroditus). Civets naturally select and consume the ripest and sweetest coffee cherries whereby the
undigested inner beans are excreted.The unique aroma and flavor quality of civet coffee has been
associated to the breaking down of proteins during civet digestion . The increasing popularity and
appreciation of civet coffee and its limited annual supply could possibly lead to the occurrence of
adulterated and fraudulent civet coffee beans in the market. It is therefore important to discriminate civet
coffee from non-civet coffee. At present, there is no internationally accepted method of verifying whether
a bean is civet coffee. Traditionally, coffee aroma has been used to characterize coffee quality. The
volatile compounds in coffee contribute greatly to its overall aroma quality .
Sensory panel evaluation is commonly used to assess the aroma profile of coffee. However, this
technique has some drawbacks such as the difficulty to train the panel effectively in order to limit
subjectivity of human response to odors and the variability between individuals . The emergence of an
electronic nose (E-nose) for the discrimination of odors has become attractive and useful to the food
industry because of its ease of use, versatility, and broad range of applications . Similarly, the
application of GCMS in coffee analysis has proven to be a valuable technique to characterize the volatile
compounds responsible for the aroma quality of coffee . In this work, E-nose and GCMS analysis of
Philippine civet coffee were carried out for the first time to determine the potential of the instruments to
discriminate civet coffees with their control coffee beans (not eaten by civet). Chemometric analysis was
applied to visually display the similarities and distinction between civet and control coffee beans.
2.1 Coffee Samples and Instruments
Four different commercial brands of Philippine coffees were analyzed. Three coffees (civet and
control) were taken from Northern part of Luzon (Kalinga, Asipulo, and Cordillera) and one from
southern part of the Philippines (Matutum, South Cotabato). The coffee samples were Arabica (Cordillera
and Matutum) and Robusta (Asipulo and Kalinga) varieties.
The Electronic Olfactory System, EOS835 was manufactured by Sacmi Imola scarl, Italy . E-nose
analysis was carried out using six metal oxide (MOX) sensors whose characteristics and fabrication
parameters are shown in Table 1. Headspace generation was held at 40 °C for 10 min with 1 min shaking.
Then, 2 ml volume was extracted and injected into carrier line at 50 °C with injection speed of 4 ml/min.
Table 1. Metal oxide (MOX) sensors characteristics.
No. Sensing layer
Catalyst Operating T
a Tin Oxide-Rheotaxial Growth and Thermal Oxidation technique
b Tungsten Oxide
c Tin-Indium Oxide
CA, USA equipped with crossbond carbowax capillary column (Restek 11023). Headspace was generated
at 70 °C for 10 min and volatiles extraction at 70 °C for 20 min.
GCMS analysis was performed using Hewlet Packard GCMS model HP6890/5973 (Hewlett Packard,
E. Ongo et al. / Procedia Engineering 47 ( 2012 ) 977 – 980
3. Results and Discussion
The responses of the MOX sensors to the coffee samples varied greatly. A comprehensive view of the
PCA score plot of the E-nose data is illustrated in Figure 1 and the dendrogram graph is shown in Figure
2. Chemometric analysis through principal component analysis (PCA) and cluster analysis reveals
groupings that differentiate civet and non-civet (control) coffee. The discrimination indicates that the
headspace vapor, and therefore the aroma of each civet coffee is different from their corresponding
control coffee beans. The distinct data structure of the individual civet coffee shows that the passage of
the beans through the digestive tract of civet affects the aroma attributes of coffee .
Fig. 1. E-nose PCA plot of civet and control coffee beans. Fig. 2. Dendrogram of civet and control coffee beans.
Data obtained from gas chromatography-mass spectroscopy (GCMS) indicates the presence of at least
47 major components (Figure 3) in the headspace of civet and non-civet (control) coffees. It was observed
that the composition of volatile compounds in civet coffee is almost similar to their controls but of
different concentrations. PCA plot (Figure 4) of prominent coffee volatiles exhibits a good classification
between civet and control coffee beans. The separation in the GCMS-PCA plot between civet and control
coffees is complementary with the E-nose results. Chemometric discrimination of civet from non-civet
coffee reveals that the aroma characteristic is one of the most important quality indicators of civet coffee.
Fig. 3. GCMS chromatographic profile of civet and
control coffee beans
Fig. 4. GCMS PCA plot for civet and control coffee beans.
The integration of E-nose and GCMS data of the PCA plot shown in Figure 5 reveals a good
correlation between aroma quality and coffee volatiles of civet coffee considering that individual civet
980 E. Ongo et al. / Procedia Engineering 47 ( 2012 ) 977 – 980 Download full-text
coffee was successfully discriminated with its control beans. Cluster separation among civet coffees
suggested that the aroma quality of civet coffee is region-specific. Results provide proof that E-nose is a
promising and practical tool in evaluating the authenticity of civet coffee.
Fig. 5. PCA plot of E-nose and GCMS data for civet and control coffee beans.
E-nose (EOS835) has proven to be useful in discriminating the aroma quality of Philippine civet
coffee. The finding was supported by the differences between the relative GCMS traces of the different
coffees. Chemometric analysis through PCA and cluster analysis demonstrated groupings that
differentiate civet and non-civet coffee. Results revealed that aroma characteristic is one of the most
important quality indicators of civet coffee. The unique aroma quality of civet coffee can be attributed to
the intestinal digestive process of civet. The cluster separation among civet coffee samples indicated that
varietal and geographical origins dictate the aroma variations in coffee. Thus, it can be concluded that the
aroma characteristics of civet coffee is varietal and region-specific. This remarkable performance
provides proof that E-nose can be used in discriminating the authentic aroma quality of civet coffee.
This research work was funded by the Philippine Council for Industry, Energy and Emerging
Technology Research and Development?(PCIEERD), Department of Science and Technology (DOST),
Philippines, the University of Modena and Reggio Emilia and the University of Brescia, Italy. One of the
authors (E.O) gratefully acknowledges the sandwich thesis grant provided to her by DOST-PCIEERD.
Lee, H.Y. (July 2006) Wine and food feature: Most expensive coffee, Forbes Magazine, USA
Marcone, M.F. (2004) Composition and properties of Indonesian palm civet coffee (Kopi Luwak) and Ethiopian civet coffee,
Food Research International 37, 901–912.
Bernard, M.C. Roberts, D.D., and Kraehenbuehl, K. (2005) Interactions between Volatile and Nonvolatile Coffee
Components. 2. Mechanistic Study Focused on Volatile Thiols, J. Agric. Food Chem., 53, 11, 4426–4433
Shilbayeh, N.F. and Iskandarani, M.Z. (2004) Quality control of coffee using an electronic nose system. American Journal of
Applied Sciences 1, 2, 129-135.
Aparicio, R., Rocha, S.M., Delgadillo, I., & Morales, M. (2000) Detection of rancid effect in virgin olive oil by the electronic
nose, J. Agric. Food Chem., 48 853-860.
Franca, A. S., Oliveira, L. S., Oliveira, R.C.S., Agresti, P.C.M., Augusti, R. (2009) A preliminary evaluation of the effect of
processing temperature on coffee roasting degree assessment, Journal of Food Engineering 92, 345-352
 Falasconi, M, Pardoa, M., Sberveglieria, G., Ricco, I., Bresciani, A., (2005) The novel EOS835 electronic nose and data
analysis for evaluating coffee ripening, Sensors and Actuators B 110, 73–80