Meat Science 88 (2011) 91–95

Contents lists available at ScienceDirect

Meat Science
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m e a t s c i

Analysis of pork adulteration in beef meatball using Fourier transform

infrared (FTIR) spectroscopy
A. Rohman a,b,c, Sismindari a,c, Y. Erwanto a,d, Yaakob B. Che Man b,e,⁎
a
Halal Research Group, Integrated Research and Testing Laboratory, Gadjah Mada University, Yogyakarta, 55281, Indonesia
b
Halal Products Research Institute, Universiti Putra Malaysia, 43400, Selangor, Malaysia
c
Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Gadjah Mada University, Yogyakarta, 55281, Indonesia
d
Faculty of Animal Sciences, Gadjah Mada University, Yogyakarta, 55281, Indonesia
e
Department of Food Technology, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400, Selangor, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: Meatball is one of the favorite foods in Indonesia. The adulteration of pork in beef meatball is frequently
Received 30 October 2009 occurring. This study was aimed to develop a fast and non destructive technique for the detection and
Received in revised form 1 July 2010 quantification of pork in beef meatball using Fourier transform infrared (FTIR) spectroscopy and partial least
Accepted 2 December 2010
square (PLS) calibration. The spectral bands associated with pork fat (PF), beef fat (BF), and their mixtures in
Available online 10 December 2010
meatball formulation were scanned, interpreted, and identified by relating them to those spectroscopically
Keywords:
representative to pure PF and BF. For quantitative analysis, PLS regression was used to develop a calibration
Beef meatball model at the selected fingerprint regions of 1200–1000 cm−1. The equation obtained for the relationship
Pork between actual PF value and FTIR predicted values in PLS calibration model was y = 0.999x + 0.004, with
Adulteration coefficient of determination (R2) and root mean square error of calibration are 0.999 and 0.442, respectively.
FTIR spectroscopy The PLS calibration model was subsequently used for the prediction of independent samples using laboratory
Partial least square (PLS) made meatball samples containing the mixtures of BF and PF. Using 4 principal components, root mean square
error of prediction is 0.742. The results showed that FTIR spectroscopy can be used for the detection and
quantification of pork in beef meatball formulation for Halal verification purposes.
© 2010 The American Meat Science Association. Published by Elsevier Ltd. All rights reserved.

1. Introduction Recently, there have been some issues associated with the
adulteration of pork in beef meatball products. The presence of
Meatball, or known as ‘bakso’ in Indonesia, is processed commi- porcine derivatives such as pork in food products is serious matters in
nuted meat which can be classified as restructured meat and is very view of religious concerns, because some religions like Islam, Judaism
popular among some countries within the Asian region and certain and Hinduism forbid their followers to consume any foods containing
European countries. It can be prepared using beef, chicken, pork, or porcine and its derivatives (Regenstein, Chaudry, & Regenstein, 2003).
fish meat and the one that is very popular and widely found in the Numerous analytical methods have been used for the analysis of
market is beef meatball (Purnomo & Rahardiyan, 2008). To gain pork and or lard obtained from extraction of pork such as differential
economical benefits, the substitution of beef in meatball with meat of scanning calorimetry (Coni, Pasquale, Cappolelli, & Bocca, 1994;
lower price such as pork takes place frequently. Kowalski, 1989), gas chromatography (Farag, Abo-raya, Ahmed,
The occurrence of food adulteration is a major issue in the food Hewedi, & Khalifa, 1983), high pressure liquid chromatography
industry, and is causing concerns among costumers and food (Marikkar, Ghazali, Che Man, Peiris, & Lai, 2005; Rashood, Shaaban,
manufacturers. One of the major authenticity issues of food product Moety, & Rauf, 1995; Saeed, Ali, Rahman, & Sawaya, 1989), electronic
involves the alteration of the true labeling of food ingredients nose (Che Man, Gan, NorAini, Nazimah, & Tan, 2005), and DNA-based
whereby high value raw materials are substituted by cheaper method (Aida, Che Man, Raha, & Son, 2007; Aida, Che Man, Wong,
materials (Al-Jowder, Kemsley, & Wilson, 1997). The detection and Raha, & Son, 2005). Some of these methods are too laborious and time
quantification of adulterants are very important for the protection of consuming, consequently, an analytical technique offering rapid and
wealth and health of consumers; therefore, some analytical methods reliable must be developed.
have been developed for food authentication studies. FTIR spectroscopy has received great attention to be used in the
quantitative analysis of edible fats and oils. Previous research has also
⁎ Corresponding author. Halal Products Research Institute, Universiti Putra Malaysia,
shown the potential of FTIR spectroscopy for analysis of lard in cake
43400, Selangor, Malaysia. Tel.: +60 3 89430405; fax: +60 3 89439745. formulation (Syahariza, Che Man, Selamat, & Bakar, 2005) and
E-mail address: [email protected] (Y.B. Che Man). chocolate products (Che Man, Syahariza, Mirghani, Jinap, & Bakar,

0309-1740/$ – see front matter © 2010 The American Meat Science Association. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.meatsci.2010.12.007
92 A. Rohman et al. / Meat Science 88 (2011) 91–95

2005). Guillen and Cabo (1997) also used FTIR spectroscopy to the next sample. Cleanliness was verified by collecting a background
characterize lard and other edible oils. Che Man and Mirghani (2001) spectrum and compare to the previous one. These spectra were
as well as Jaswir, Mirghani, Hassan, and Mohd Said (2003) have recorded as absorbance values at each data point in triplicate.
developed FTIR spectroscopic method for detecting lard in the
mixture with other animal fats. Currently, Rohman and Che Man 2.4. Statistical analysis and validation
(2010) have also used FTIR spectroscopy in combination with partial
least square (PLS) and discriminant analysis for quantification of lard The software TQ Analyst™ version 6 (Thermo electron Corpora-
in binary mixtures with other animal fats of beef, chicken and mutton. tion, Madison, WI) was used for chemometrics analysis of partial least
Combined with multivariate calibration of PLS, FTIR spectroscopy squares (PLS). The “leave-one-out” cross-validation procedure was
is often used for quantitative analysis of complex mixtures. This used to verify the calibration model. The values of root mean square
combination enables the rapid and simultaneous determination of error of calibration (RMSEC) and coefficient of determination (R2)
each component in the complex mixtures without time consuming were used as the validity criteria for the calibration. The validation
and with minimal sample preparation. PLS was based on the data was further investigated using the mean difference (MD) and
compression and inverse calibration (Paradkar, Sivakesava, & Iru- standard deviation of difference (SDD) for accuracy and
dayaraj, 2002). The objective of this study is to investigate the use of reproducibility.
FTIR spectroscopy combined with PLS to detect and to quantify the
presence of pork in beef meatball formulation through analysis of fats 3. Results and discussions
obtained from pork.
3.1. FTIR spectral analysis
2. Materials and methods
Fats obtained from the extraction of pork or beef are analyzed
Beef and pork were randomly obtained from five different using FTIR spectroscopy in the mid infrared region (4000–650 cm−1).
slaughter houses in Jogjakarta, Indonesia by taking into account the FTIR spectroscopy can be an ideal technique for analysis of fats and
different feeding of corresponding animals. The materials used for oils, because they are essentially single component systems of
making meatball formulation were purchased from local market. All triglycerides (TGs) and can be applied directly in their neat form to
solvents used for analysis were of pro analytical grade. ATR crystal or passed through a flow cell. The application of FTIR
spectroscopy for analysis of fats and oils has been widely proposed,
2.1. Meatball preparation and extraction of fats because once the instrument has been calibrated; it can be used for
routine analyses. The importance of IR spectroscopy for the qualitative
Meatball was prepared by emulsifying 90% of fine ground meat analysis originates from the much information contents obtained and
(beef and or pork) with 10% of starch and mixing it with salt and the possibility to assign certain absorption bands related to functional
certain spices, and finally shaping it into balls. It is then cooked in groups. In fats and oils, most of the peaks and shoulders of the
boiling water for 10 to 20 min. Meatball was further cut into small spectrum are attributable to specific functional groups (Bendini et al.,
pieces using commercial blender and subjected to fat extraction 2007).
according to traditional Soxhlet method as described by Association of The representative spectra of fats obtained from the extraction of
Official Analytical Chemists, AOAC (1995) using hexane as an pork and beef meatballs are demonstrated in Fig. 1. These spectra look
extraction solvent. Fats yielded were further used for FTIR analysis. very similar and show a typical characteristic of absorption bands for
common TGs. The assignments of prominent peaks were compiled in
2.2. Calibration and validation standard Table 1.
An observation using the naked eyes, the whole range looks very
The calibration sets was prepared by spiking pork to meatball in similar because the main component composed of these fats are TGs.
concentration range of 1.0; 3.0; 5.0; 10.0; 25.0. Meatball containing of However, due to the fingerprint technique, meaning that there is no two
100% beef and 100% pork was also made to observe the spectral compounds or samples having the same spectra in terms of amount and
differentiation. For validation or prediction, another series of meatball intensity of peaks, FTIR spectroscopy can be used to extract a difference
containing the mixture of pork and beef were prepared. The meatball among these fats. Upon a closer scrutiny, the minor differences (peak
was further subjected to extraction. The fats obtained were analyzed heights) can be attainable at 1117 and 1097 cm−1 (l and m)
using FTIR spectroscopy. The spectral regions where the variations corresponding to C–H bending vibration and C–H deformation vibrations
were observed were chosen for developing PLS model. of fatty acids, respectively. Hence, these frequencies, which FTIR spectra
variations were observed, are used as a basis for choosing the spectral
2.3. FTIR spectroscopy analysis regions in the quantification of pork fat in meatball samples.

Using a Pasteur pipette, fats obtained from the extraction 3.2. Calibration and validation
procedure were placed in direct contact with attenuated total
reflectance (ATR) crystal on a multibounce plate at controlled Partial least square (PLS) was used for constructing calibration
ambient temperature (25 °C). A Nicolet 6700 FTIR spectrometer model. Frequencies at selected fingerprint (1200–1000 cm−1) was
(Thermo Nicolet Corp., Madison, WI) equipped with a detector of exploited for quantitative analysis. The ratio of peak height (R) at 1117
deuterated triglycine sulphate (DTGS), a beam splitter of KBr/ and 1097 cm−1 in beef fat (BF) was much higher than that of pork fat
Germanium, and connected to software of the OMNIC operating (PF). Consequently, increasing PF concentrations in meatball formula-
system (Version 7.0 Thermo Nicolet), were used during FTIR data tion will result in reducing the ratio value, approaching the height ratio
collection. To minimize water vapor interference, the instrument was of PF at 1117 and 1097 cm−1 (Fig. 2). The R average values for PF and BF
maintained with dehumidifier of silica gel. FTIR spectra were recorded from different origins are (1.054 ± 0.005) and (1.107 ± 0.009),
from 32 scans at a resolution of 4 cm−1 at 4000–650 cm−1. These respectively.
spectra were substracted against background air spectrum. After In PLS model, the samples of meatballs were divided into sets of
every scan, a new reference air background spectrum was taken. The calibration and prediction. Beef meatball adulterated with 1.0–100.0%
ATR plate was carefully cleaned in situ by scrubbing with hexane twice of pork was used for quantification purposes. The relationship
followed by acetone and dried with soft tissue before filling in with between actual value (x-axis) and FTIR predicted value of PF in PLS
 A. Rohman et al. / Meat Science 88 (2011) 91–95 93

Fig. 1. FTIR spectra of pork fat and beef fat in the mid infrared region (4000–650 cm−1).

calibration model was shown in Fig. 3. A good linear regression of shown in Fig. 3, with R2 and root mean square error of calibration
y = 0.999x + 0.004; was obtained with R2 and RMSEC values are 0.999 (RMSEP) obtained are 0.996 and 0.712, respectively.
and 0.128, respectively. R2 value tells how close the relationship The confirmation and validation of the analysis region used for
between actual and FTIR predicted values of analyte of interest (PF). developing the PLS model were performed by computing the
The closer R2 value to unity, the better the relationship. According to predicted residual error sum of squares (PRESS) values for different
International conference on Harmonization, R2 value higher than 0.99 factors or principal components (PCs). The PRESS value is a direct
is acceptable for such relationship. Meanwhile, RMSEC refers to the measure on how well a calibration predict the concentrations left out
calibration uncertainty. The smaller the RMSEC value, the better the during a cross validation (Smith, 2002). PRESS informed that the
calibration model. optimal factor number is 4 as revealed in Fig. 4, that express RMSEC
The calibration model was further cross validated using “leave- obtains a stable value, minimally after four-factor. This confirms that
one-out” technique by removing one standard at a time. In cross the spectral region used for developing the PLS model for quantifi-
validation, one of the calibration samples is taken out from PLS cation of PF exhibits significant correlation with its concentration.
calibration model and the residual samples are used to build PLS The statistical results were also calculated from the prediction
model. Subsequently, the removed sample is calculated using the new samples as mean difference (MD) and standard deviation of difference
PLS regression. This manner was recurred, leaving each sample out in (SDD) for determination of accuracy (a) and precision (r) of the
turn (Miller & Miller, 2005). The R2 and root mean square error of developed method. The term accuracy refers to the closeness of
cross validation (RMSECV) values of 0.998 and 0.442 were obtained. agreement between actual data and the predicted data of FTIR results.
PLS calibration model was also used to calculate the PF contents in In addition, the precision is a measure of the random error in a data
prediction samples. The scatter plot for the relationship between set. The low values of MDa (2.024) and SDDa (1.076) obtained
actual and FTIR predicted values of PF in prediction samples was indicated that FTIR spectroscopy combined with PLS is an appropriate

Table 1
Functional groups and modes of vibration of pork and beef fats.a

Assignment Frequency (cm−1) Functional group vibration

(a) 3005 cis C=CH stretching

(b) 2954 Asymmetric stretching vibration of
methyl (–CH3) group
(c) and (d) 2924 and 2852 Asymmetric or symmetric stretching vibration of
methylene (–CH2) band
(e) 1743 Carbonyl (C=O) functional group from the ester
linkage of triacylglycerol
(f) 1654 cis C=C
(g) 1465 Bending vibrations of the CH2 and CH3 aliphatic groups
(h) 1417 Rocking vibrations of CH bonds of cis-disubstituted alkenes
(i) 1377 Symmetric bending vibrations of CH3 groups
(j) and (k) 1228 and 1155 Vibrations of stretching mode from the C–O group in esters
(l) and (m) 1111 and 1097 –CH bending and –CH deformation vibrations of fatty acids
(n) 1033 C–O stretching
(o) 962 Bending vibration of CH functional groups of isolated trans-olefin
(p) 721 Overlapping of the methylene (–CH2) rocking vibration and to
the out of plane vibration of cis-disubstituted olefins
a
Source: Guillen and Cabo, 1997; Vlachos et al., 2006.
94 A. Rohman et al. / Meat Science 88 (2011) 91–95

Fig. 2. FTIR spectra of beef and pork fats as well its mixture obtained from extraction of meatball formulations, showing changes in the height ratio (R) at 1117 and 1097 cm−1 due to
the increase of pork fat percentages at 4000–650 cm−1.

analytical technique for quantitative analysis of PF in beef meatball.

Meanwhile, the low MDr (2.256) and SDDr (2.546) indicated that the
FTIR method is precise enough.

4. Conclusion

The adulteration of beef meatball with pork can be monitored by

means of FTIR spectroscopy on attenuated total reflectance (ATR)
technique. Partial least squares can be successfully used to quantify
the level of adulterant of pork at selected fingerprint region (1200–
1000 cm−1).

Acknowledgements

This research was supported by project grant from the LPPT-UGM

(Integrated Research and Testing Gadjah Mada University) through
Excellent Research for National Strategic (Riset Unggulan Strategis
National, RUSNAS) with contract no. LPPM-UGM/1309/2009. The
authors acknowledge Halal products research institute UPM for
providing the FTIR instrument.

References
Aida, A. A., Che Man, Y. B., Raha, A. R., & Son, R. (2007). Detection of pig derivatives in
food products for halal authentication by polymerase chain reaction–restriction
fragment length polymorphism. Journal of the Science of Food and Agriculture, 87,
569−572.
Aida, A. A., Che Man, Y. B., Wong, C. M. V. L., Raha, A. R., & Son, R. (2005). Analysis of raw
Fig. 3. Relationship between actual value of PF and FTIR predicted value (% w/w) at
meats and fats of pigs using polymerase chain reaction for Halal authentication.
selected figerprint region (1200–1000 cm−1). A = calibration; B = prediction. Meat Science, 69, 47−52.
Al-Jowder, O., Kemsley, E. K., & Wilson, R. H. (1997). Mid-infra red spectroscopy and
authenticity problems in selected meats: A feasibility study. Food Chemistry, 59,
195−201.
AOAC (Association of Official Analytical Chemists) (1995). Official methods of analysis
(16th edition). Gaithersburg, MD: AOAC International.
Bendini, A., Cerretani, L., Di Virgilio, F., Belloni, P., Bonoli-Carbognin, M., & Lercker, G.
(2007). Preliminary evaluation of the application of the FTIR spectroscopy to
control the geographic origin and quality of virgin olive oils. Journal of Food Quality,
30, 424−437.
Che Man, Y. B., Gan, H. L., NorAini, I., Nazimah, S. A. H., & Tan, C. P. (2005). Detection of
lard adulteration in RBD palm olein using an electronic nose. Food Chemistry, 90,
829−835.
Che Man, Y. B., & Mirghani, M. E. S. (2001). Detection of lard mixed with body fats of
chicken, lamb, and cow by Fourier transform infrared spectroscopy. Journal of
American Oil Chemists' Society, 78, 753−761.
Che Man, Y. B., Syahariza, Z. A., Mirghani, M. E. S., Jinap, S., & Bakar, J. (2005). Analysis of
potential lard adulteration in chocolate and chocolate products using Fourier
transform infrared spectroscopy. Food Chemistry, 90, 815−819.
Fig. 4. Root mean square error of calibration (RMSEC) of cross validation versus Coni, E., Pasquale, M. D., Cappolelli, P., & Bocca, A. (1994). Detection of animal fats in
principal component (PC) factor. butter by SC: A pilot study. Journal of American Oil Chemists' Society, 71, 807−810.
 A. Rohman et al. / Meat Science 88 (2011) 91–95 95

Farag, R. S., Abo-raya, S. H., Ahmed, F. A., Hewedi, F. M., & Khalifa, H. H. (1983). Purnomo, H., & Rahardiyan, D. (2008). Review article: Indonesian traditional meatball.
Fractional crystallization and gas chromatographic analysis of fatty acids as a International Food Research Journal, 15, 101−108.
means of detecting butterfat adulteration. Journal of American Oil Chemists' Society, Rashood, K. A., Shaaban, R. R. A., Moety, E. M. A., & Rauf, A. (1995). Triacylglycerols-
60, 1665−1669. profiling by high performance liquid chromatography: A tool for detection of pork
Guillen, M. D., & Cabo, N. (1997). Characterization of edible oils and lard by Fourier fat in processed foods. Journal of Liquid Chromatography, 18, 26−31.
transform infrared spectroscopy. Relationships between composition and frequen- Regenstein, J. M., Chaudry, M. M., & Regenstein, C. E. (2003). The kosher and halal food
cy of concrete bands in the fingerprint region. Journal of American Oil Chemists' laws. Comprehensive Reviews in Food Science and Food Safety, 2, 111−127.
Society, 74, 1281−1286. Rohman, A., & Che Man, Y. B. (2010). FTIR spectroscopy combined with chemometrics
Jaswir, I., Mirghani, M. E. S., Hassan, T. H., & Mohd Said, M. Z. (2003). Determination of for analysis of lard in the mixtures with body fats of lamb, cow, and chicken.
lard in mixtures of body fats of mutton and cow by Fourier transform-Infra red (FT- International Food Research Journal, 17, 519−527.
IR) spectroscopy. Journal of Oleo Science, 52, 633−638. Saeed, T., Ali, S. G., Rahman, H. A., & Sawaya, W. N. (1989). Detection of pork and lard as
Kowalski, B. (1989). Sub-ambient differential scanning calorimetry of lard and lard adulterants in processed meat: Liquid chromatographic analysis of derivatized
contaminated by tallow. International Journal of Food Science & Technology, 24, triglycerides. Journal — Association of Official Analytical Chemists, 72, 921−925.
415−420. Smith, B. C. (2002). Quantitative spectroscopy: Theory and practice. Amsterdam:
Marikkar, J. M. N., Ghazali, H. M., Che Man, Y. B., Peiris, T. S. G., & Lai, O. M. (2005). Academic Press.
Distinguishing lard from other animal fats in admixtures of some vegetable oils Syahariza, Z. A., Che Man, Y. B., Selamat, J., & Bakar, J. (2005). Detection of lard
using liquid chromatographic data coupled with multivariate data analysis. Food adulteration in cake formulation by Fourier transform infrared (FTIR) spectroscopy.
Chemistry, 91, 5−14. Food Chemistry, 92, 365−371.
Miller, J. N., & Miller, J. C. (2005). Statistics and chemometrics for analytical chemistry. Vlachos, N., Skopelitis, Y., Psaroudaki, M., Konstantinidou, V., Chatzilazarou, A., & Tegou,
Pearson Education Limited (pp. 213−239). (5th ed.). England: Edinburgh Gate Harlow. E. (2006). Applications of Fourier transform-infrared spectroscopy to edible oils.
Paradkar, M. M., Sivakesava, S., & Irudayaraj, J. (2002). Discrimination and classification Analytica Chimica Acta, 573–574, 459−465.
of adulterants in maple syrup with the use of infrared spectroscopic techniques.
Journal of the Science of Food and Technology, 82, 497−504.