Table of Contents Author Guidelines Submit a Manuscript
Journal of Spectroscopy
Volume 2013 (2013), Article ID 138728, 5 pages
Research Article

Application of Fourier Transform Infrared Spectroscopy for the Oxidation and Peroxide Value Evaluation in Virgin Walnut Oil

1Faculty of Life Science and Technology, Kunming University of Science and Technology, Kunming 650500, China
2Faculty of Chinese Traditional Medicine, Yunnan University of Chinese Traditional Medicine, Kunming 650500, China

Received 11 May 2013; Revised 27 August 2013; Accepted 9 September 2013

Academic Editor: Christoph Krafft

Copyright © 2013 Pengjuan Liang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Recent developments in Fourier transform infrared spectroscopy-partial least squares (FTIR-PLSs) extend the application of this strategy to the field of the edible oils and fats research. In this work, FT-IR spectroscopy was used as an effective analytical tool to determine the peroxide value of virgin walnut oil (VWO) samples undergone during heating. The spectra were recorded from a film of pure oil between two disks of KBr for each sample at frequency regions of 4000–650 cm−1. Changes in the values of the frequency of most of the bands of the spectra were observed and used to build the calibration model. PLS model correlates the actual and FT-IR estimated value of peroxide value with a correlation coefficient of 0.99, and the root mean square error of the calibration (RMSEC) value is 0.4838. The methodology has potential as a fast and accurate way for the quantification of peroxide value of the edible oils.

1. Introduction

The production of walnut constitutes a significant proportion of the income of farmers in many countries. In the recent years, the peroxide value of virgin walnut oil (VWO) has received great attention because of its sensory qualities and biological activities. Epidemiological studies show that VWO not only reduces serum cholesterol but also has nutritional cranial nerve cells which can adjust plant nerve function. Other experts discover that walnut oil does not only act as an officinal but also can be used as the “old man and the infant nutrition oil” as well as aerial work and flight personnel senior’s health care oil [1].

In this way, VWO production is a promising factor to the viability of ecological economy. Nevertheless, VWO can be a target of adulteration with cheaper oils or even metamorphic oils because of its higher prices in the global market. The adulteration comprises a great hazard not only for the economic development and prosperity of those communities but also for the health and safety of VWO consumers. Therefore, the development of rapid and accurate analytical techniques capable of detecting the quality of VWO is currently highly demanded.

In fact, the quality of oils depends on its chemical composition that changes qualitatively and quantitatively; one of the most important indicators of performance and shelf-life is the oxidative stability of oils, that is, their resistance to the oxidation process. Previous study showed that the unsaturated fatty acid content of oils is higher; that is, the unsaturated level is higher, and the oxidation is easier [2]. Because of the unsaturated fatty acid in VWO is 90%, and there is about 47.4% of linolic acid and 15.8% of linolenic acid [3], VWO is easier to be oxidized.

Generally, the sample of oil is oxidized when subjected to air or oxygen flow, heating, exposure to light, catalysers, and so forth. Although the mechanism of the oil degradation process was influenced by the oxidative conditions, it has normally been established as being a free radical mechanism yielding hydroperoxides, also called primary oxidation products [4]. Because hydroperoxides are the primary products formed as autoxidation commences and serve as precursors to the subsequent formation of secondary oxidation products: aldehydes, ketones, lactones, alcohols, acids, and so forth, they are expressed as peroxide value (PV) in the foodsafety [5]. The methods to quantify PV which is used to determine the degree of the oxidation process are mostly related to the measurement of the concentration of primary or secondary oxidation products or both, or to the amount of oxygen consumed during the process. Among those, PV, which measures hydroperoxide concentration, is one of the most popular [4].

In the past years, the traditional method of establishing the PV of edible oils is titration, and the method has been standardized by ISO (3960:1977) or GB/T 5538-1955, ISO (3960:2001) or GB/T 5538-2005. The determination process of titration is susceptible to a variety of external factors, and the measured value has poor reproducibility and low sensitivity, only can be applied to the large concentration of hydroperoxide determination.

The recent developments in FT-IR spectroscopy instrumentation and the application of this technique expand in food research, facilitating particularly the studies on edible oils and fats. FT-IR methods have demonstrated themselves to be rapid and nondestructive analytical tools with minimum sample preparation necessary. Due to the fact that the intensities of the bands in the spectrum are proportional to concentration (i.e., Beer’s law is obeyed) and coupled with the chemometric techniques, FT-IR spectroscopy becomes an excellent tool for quantitative analysis. Several applications have been performed using this analytical approach together with chemometric methods, to detect walnut oil adulteration [6]; evaluate olive oil freshness [7]; monitor fatty acid composition of virgin olive oil [8]; and assess oil oxidation [911].

Although a lot of work has already been published on the determination of PV by FTIR spectroscopy, different articles have different analysis methods. A primary FTIR spectroscopic method for the determination of PV in edible oils, based on the stoichiometric reaction of triphenylphosphine (TPP) with the hydroperoxides present in edible oils to produce triphenylphosphine oxide (TPPO), accurate quantitation of the TPPO by the measurement of the absorption band at 542 cm−1, provides a simple means of determining PV [5, 12]. In an other study, the authors analyzed PV of oil by near-infrared spectroscopy and established three models including PLS-regression, multiple linear regression (MLR), and principal component regression (PCR). Comparative analysis indicated that the most reliable model was selected among the established models in predicting peroxide value of soybean oil [13]. In this study, MIR spectroscopy (4000–650 cm−1) applied to VWO sample gives information on the molecular bonds and therefore on the functional groups of molecule. Furthermore, samples can be directly detected without additional processing. Compared with other articles, the method in this study is more convenient.

In this paper, FT-IR spectroscopy is used for the oxidation and PV evaluation of virgin walnut oil. The oxidation experiments were carried out in all cases by heating the samples in a convection oven at 70°C at different time, the PV of each sample were determined by chemical and FT-IR spectroscopy method. PLS model correlates the actual and FT-IR estimated values of PV with a coefficient of determination () of 0.9918. This approach could represent a faster and more accurate recognition tool for the determination walnut oil peroxide value since it is characterized by the simplicity of sample preparation.

2. Materials and Methods

2.1. Sample Oxidation

Virgin walnut oil (VWO) was purchased from the local market. The samples were placed in a convection oven in uncovered polystyrene Petri dishes and heated at 70°C for 1, 2, 3, 4, 5, 6, 8, and 10 d, respectively.

2.2. Chemical Method

PV determination was performed according to the method proposed by ISO (3960:2001) or GB/T 5538-2005. 2.00–3.00 g of VWO sample was placed in iodine flask and dissolved in a blended solution of 50 mL isooctane-glacial acetic acid (2 : 3, v/v). A saturated solution of KI (0.5 mL) was added. The mixture was shaken by hand for 0.5 min and kept in the dark for another 3 min. After the addition of 30 mL distilled water, the mixture was titrated against sodium thiosulphate (0.01 M) until the yellow colour almost disappeared. Then, about 0.5 mL of starch indicator (0.05%) solution was added. Titration was sustained until the blue colour just disappeared. A blank was also determined under similar conditions. PV (meq/kg) was calculated as follows: where is the sodium thiosulphate concentration (M), and represent the volumes of sodium thiosulphate exhausted by the samples and the blank, respectively (mL), and is the mass of VWO (g).

2.3. FT-IR Spectra Acquisition

FT-IR spectra of samples were obtained using Shimadzu IRPrestige-21 equipped with DLATGS as detector and potassium bromide (KBr) as beam splitter and controlled with the IRsolution software. A small quantity (~20 μL) of the sample was deposited with the use of a Pasteur pipette between two well-polished KBr disks, creating a thin film in IR region of 4000–650 cm−1, by accumulating 40 scans with the resolution of 4 cm−1. These spectra were subtracted from reference spectrum of air, acquired by collecting a spectrum from the cleaned KBr disks blank before the measurement of each oil sample replication. At the end of every scan, the surface of KBr disk was cleaned with hexane twice, dried with special soft tissue, then cleaned with acetone, and finally dried with soft tissue following the collection of each spectrum. The samples of spectra were collected in triplicate and displayed as the average spectra.

2.4. Chemometric

The chemometric analyses were performed using the software SIMCA-P. Quantification of the PV of VWO was carried out using partial least square (PLS) [14, 15]. PLS algorithm is an effective chemometric tool. It takes the advantages of multiple linear regression (MLR) and principal component regression (PCR); it has strong predictive ability, relatively simple model, and so forth, which makes Fourier transform infrared (FT-IR) spectroscopy more powerful and useful [16, 17]. For quantitative analysis, 18 samples were selected for calibration and 10 samples for validation. The PV of each sample used in both calibration and validation is presented in Table 1. Each sample was subjected to FT-IR analysis.

Table 1: The peroxide value (PV) of each sample used in both calibration and validation.

3. Results and Discussion

3.1. FT-IR Spectrum of VWO

As a descriptive example a virgin walnut oil FT-IR spectrum is presented from 4000 to 650 cm−1 in Figure 1. The assignment of functional groups responsible for IR absorption peak is as follows: 3009 (C–H stretching vibration of the cis-double bond); 2947–2954 (symmetric stretching vibration shoulder of CH3); 2926 and 2854 (symmetric and asymmetric stretching vibration of CH2, resp.,); 1745 (C=O stretching vibrations); 1465 and 1458 (bending vibrations of the CH2 and CH3); 1398 (bending in plane vibrations of CH cis-olefinic groups); 1377 (bending vibrations of CH2 groups); 1238, 1163, 1120, and 1099 (C–O stretching vibration) [1823].

Figure 1: IR spectra of virgin walnut oil (VWO) sample before oxidation.
3.2. Evaluation of Oxidation

Figure 2 shows the infrared spectra of the walnut oil sample under the oxidative conditions of heating at 0, 6, 8, and 10 days, respectively. The infrared spectra of the samples are similar in their original state; however, there are some differences which are related to the width and intensity of some bands, as well as the presence or absence of others [24, 25]. The frequency of the cis-double bond stretching vibration band near 3009 cm−1 remains almost unaltered or suffers a very slow shifting toward smaller wavelength values. The band at 2926 cm−1 increases its absorbance and width. The band at 2854 cm−1 and the shoulder at 2949 cm−1, increase their intensity due to surrounding chemical changes as a consequence of the oxidation process. There is another major absorbance near 1746 cm−1 and the change of this band can be associated with the appearance of saturated aldehyde functional groups [26, 27] or other secondary oxidation products that cause an absorbance at 1728 cm−1, which overlaps with the stretching vibration at 1746 cm−1 of the ester carbonyl functional group of the triglycerides. A very weak band of the spectrum is near 1399 cm−1 which has been proved to be closely related to the proportion of the oleic acyl groups in the sample [24], and some authors have associated it with bending in plane vibrations of CH cis-olefinic groups [28]. The frequency of this band remains practically unchangedor increases slightly. Two other bands present in all samples of oil are near 1238 and 1163 cm−1, and have been proved to be related to the proportion in the sample of saturated acyl groups [24]. The frequencies of both bands suffer similar changes during the oxidation process and increase their intensity, although the changes are sharper in the band near 1163 cm−1. The frequency of band near 1099 cm−1 does not generally suffer large variations except in oil samples rich in oleic acyl groups. As the oxidation process advances, the frequency of this band diminishes and reaches a minimum value.

Figure 2: FTIR spectra of the virgin walnut oil (VWO) sample after 0, 6, 8, and 10 days, respectively, under oxidative condition at region of 4000–1030 cm−1. (A) 0 day. (B) 6 days. (C) 8 days. (D) 10 days.
3.3. Calibration Model

The calibration model of PV was carried out by using PLS algorithm. The PV of each sample used in calibration is presented in Table 1. The spectral regions used for PLS calibration models are 3009, 2989, 2949, 2926, 2881, 2854, 2744, 1790, 1746, 1710, 1487, 1465, 1462, 1458, 1408, 1398, 1392, 1377, 1311, 1238, 1217, 1163, 1126, 1120, 1110, and 1099, 1041 cm−1 (Figure 1). The selection of these frequency regions was based on the optimisation processes in which they offered the highest values of and the lowest values of error, either in calibration or in prediction models. Although specific band directly related to PV was at approximately 3500 cm−1, samples with 4–90 PV at this band were very weak, and it was very difficult to differentiate from noise. So the band of 3500 cm−1 was totally removed.

Figure 3 shows the PLS calibration model which correlates the actual and estimated PV of VWO obtained from FT-IR spectra at the specified regions. The relationship between the actual and the observed value of PV showed a good correlation with coefficient of determination () values being 0.9918. The evaluation of the error in calibration model was obtained by calculating the root mean square error of calibration (RMSEC) [29]. RMSEC value is 0.4838; therefore, the high value of and low value of RMSEC indicate the success of calibration model.

Figure 3: Iodometric and estimated values of peroxide value (PV) for calibration and validation sets (▲: calibration set; ■: validation set).
3.4. Validation Model

In order to estimate the prediction ability of the developed model, PLS calibration model was used to predict the values of PV of 10 VWO samples as validation data sets (Table 1). Figure 3 shows the PLS validation model which correlates the actual and estimated PV of VWO obtained from FT-IR spectra at the specified regions as well as calibration model. The values of and RMSEP [29] are 0.9903 and 0.3545, respectively. Based on this result, it can be stated that PLS appears to have a reasonable ability to estimate the PV of VWO samples. Table 2 compiled PLS performance in terms of , standard deviation (Std. dev.), RMSEC, RMSEP, and the number of principal components for quantification of the PV of VWO.

Table 2: PLS performance for analysis of calibration and validation model.

4. Conclusions

From the results here obtained it can be concluded that FT-IR spectroscopy is an adequate technique for detecting the changes produced in virgin walnut oil sample during an oxidation process. The changes observed in the frequency values of most of the bands of the spectra throughout the oxidation experiment provide accurate information about the different stages of the process and for this reason, FTIR-PLS could be useful to evaluate the oxidative state of VWO and to determine the PV, one of the most important indicators of performance and shelf-life, in a simple, accurate, and fast way.


The authors gratefully acknowledge the financial support of State Science and Technology Support Program (2011BAD46B00) and Yunnan Science and Technology Program (2011AB006).


  1. S. L. Zhao, C. Y. Chen, and F. Ge, “Progressive study on functions and components,” Journal of Yunnan University of Traditional Chinese Medicine, vol. 33, pp. 71–74, 2010. View at Google Scholar
  2. S. G. Li, W. T. Xue, and H. Zhang, “Research progress in analysis methods of peroxide value in edible oils,” Cereals and Oils, vol. 7, pp. 35–38, 2007. View at Google Scholar
  3. B. Y. Wan, H. Z. Dong, and H. Li, “Research on the properties and nutrition of walnut oil,” China Western Cereals and Oils Technology, vol. 26, pp. 18–19, 2001. View at Google Scholar
  4. N. Vlachos, Y. Skopelitis, M. Psaroudaki, V. Konstantinidou, A. Chatzilazarou, and E. Tegou, “Applications of Fourier transform-infrared spectroscopy to edible oils,” Analytica Chimica Acta, vol. 573-574, pp. 459–465, 2006. View at Publisher · View at Google Scholar · View at Scopus
  5. K. Ma, F. R. van de Voort, J. Sedman, and A. A. Ismail, “Stoichiometric determination of hydroperoxides in fats and oils by fourier transform infrared spectroscopy,” Journal of the American Oil Chemists' Society, vol. 74, no. 8, pp. 897–906, 1997. View at Google Scholar · View at Scopus
  6. P. J. Liang, H. Wang, C. Y. Chen et al., “The use of Fourier transform infrared (FT-IR) spectroscopy for quantification of adulteration in virgin walnut oil,” Journal of Spectroscopy, vol. 2013, Article ID 305604, 6 pages, 2013. View at Publisher · View at Google Scholar
  7. N. Sinelli, M. S. Cosio, C. Gigliotti, and E. Casiraghi, “Preliminary study on application of mid infrared spectroscopy for the evaluation of the virgin olive oil ‘freshness’,” Analytica Chimica Acta, vol. 598, no. 1, pp. 128–134, 2007. View at Publisher · View at Google Scholar · View at Scopus
  8. R. M. Maggio, T. S. Kaufman, M. D. Carlo et al., “Monitoring of fatty acid composition in virgin olive oil by Fourier transformed infrared spectroscopy coupled with partial least squares,” Food Chemistry, vol. 114, no. 4, pp. 1549–1554, 2009. View at Publisher · View at Google Scholar · View at Scopus
  9. M. D. Guillén and N. Cabo, “Fourier transform infrared spectra data versus peroxide and anisidine values to determine oxidative stability of edible oils,” Food Chemistry, vol. 77, no. 4, pp. 503–510, 2002. View at Publisher · View at Google Scholar · View at Scopus
  10. A. Bendini, L. Cerretani, F. Di Virgilio, P. Belloni, M. Bonoli-Carbognin, and G. Lercker, “Preliminary evaluation of the application of the ftir spectroscopy to control the geographic origin and quality of virgin olive oils,” Journal of Food Quality, vol. 30, no. 4, pp. 424–437, 2007. View at Publisher · View at Google Scholar · View at Scopus
  11. B. Muik, B. Lendl, A. Molina-Diaz, M. Valcarcel, and M. J. Ayora-Cañada, “Two-dimensional correlation spectroscopy and multivariate curve resolution for the study of lipid oxidation in edible oils monitored by FTIR and FT-Raman spectroscopy,” Analytica Chimica Acta, vol. 593, no. 1, pp. 54–67, 2007. View at Publisher · View at Google Scholar · View at Scopus
  12. X. Z. Yu, S. K. Du, Z. X. Li, and J. Y. Zhang, “Study on detection of peroxide value in edible oils using FTIR spectral reconstitution technique,” Journal of Chinese Institute of Food Science and Technology, vol. 11, pp. 169–175, 2011. View at Google Scholar
  13. L. Q. Wang, L. Y. Zhang, and X. C. Zhu, “Near-infrared spectroscopic analysis of peroxide value of soybean oil,” Food Science, vol. 31, pp. 205–207, 2010. View at Google Scholar
  14. H. Martens and T. Naes, Multivariate Calibration, John Wiley & Sons, Chichester, UK, 1989.
  15. E. V. Thomas, “A primer on multivariate calibration,” Analytical Chemistry, vol. 66, no. 15, pp. 795–804, 1994. View at Google Scholar
  16. E. Y. Zhu and P. Y. Yang, Chemometrics and Its Application, Science Press, Beijing, China, 2003.
  17. L. Zhang, L. M. Zhang, Y. Li, B. P. Liu, X. F. Wang, and J. D. Wang, “Application and improvement of partial-least-squares in Fourier transform infrared spectroscopy,” Spectroscopy and Spectral Analysis, vol. 25, no. 10, pp. 1610–1613, 2005. View at Google Scholar · View at Scopus
  18. N. B. Colthup, L. H. Daly, and S. E. Wiberly, Introduction to Infrared and Raman Spectroscopy, Accademic Press, New York, NY, USA, 3rd edition, 1990.
  19. M. K. Ahmed, J. K. Daun, and R. Przybylski, “FT-IR based methodology for quantitation of total tocopherols, tocotrienols and plastochromanol-8 in vegetable oils,” Journal of Food Composition and Analysis, vol. 18, no. 5, pp. 359–364, 2005. View at Publisher · View at Google Scholar · View at Scopus
  20. H. Yang, J. Irudayaraj, and M. M. Paradkar, “Discriminant analysis of edible oils and fats by FTIR, FT-NIR and FT-Raman spectroscopy,” Food Chemistry, vol. 93, no. 1, pp. 25–32, 2005. View at Publisher · View at Google Scholar · View at Scopus
  21. M. D. Guillen, N. Cabo, and J. Agric, “Methods for monitoring composition and flavor quality of cheese using a rapid spectroscopic method,” Food Chemistry, vol. 46, no. 5, pp. 1788–1793, 1998. View at Publisher · View at Google Scholar
  22. J. Coates and R. A. Meyers, Encyclopaedia of Analytical Chemistry, John Wiley & Sons, Chichester, UK, 2000.
  23. J. Workman, Handbook of Organic Compounds, Academic Press, London, UK, 2001.
  24. M. D. Guillén and N. Cabo, “Characterization of edible oils and lard by fourier transform infrared spectroscopy. Relationships between composition and frequency of concrete bands in the fingerprint region,” Journal of the American Oil Chemists' Society, vol. 74, no. 10, pp. 1281–1286, 1997. View at Google Scholar · View at Scopus
  25. M. D. Guillén and N. Cabo, “Relationships between the composition of edible oils and lard and the ratio of the absorbance of specific bands of their fourier transform infrared spectra. Role of some bands of the fingerprint region,” Journal of Agricultural and Food Chemistry, vol. 46, no. 5, pp. 1788–1793, 1998. View at Publisher · View at Google Scholar
  26. E. N. Frankel, Lipid Oxidation, Oily Press, 1980.
  27. R. J. Hamilton, “The chemistry of rancidity in foods,” in Rancidity in Foods, J. C. Allen and R. J. Hamilton, Eds., Elsevier Applied Science Publishers, London, UK, 1989. View at Google Scholar
  28. D. B. Dahlberg, S. M. Lee, S. J. Wenger, and J. A. Vargo, “Classification of vegetable oils by FT-IR,” Applied Spectroscopy, vol. 51, no. 8, pp. 1118–1124, 1997. View at Google Scholar · View at Scopus
  29. M. M. Paradkar and J. Irudayaraj, “A rapid FTIR spectroscopic method for estimation of caffeine in soft drinks and total methylxanthines in tea and coffee,” Journal of Food Science, vol. 67, no. 7, pp. 2507–2511, 2002. View at Google Scholar · View at Scopus