3.1. Validate analytical methods for detecting palm oil in artificially adulterated milk.
2.1. Physicochemical Analysis Of Fluid Milk Containing Milk Fat And Palm Oil At Different Levels
As observed in Table (1), the saponification number (SN) of the studied milk mixtures ranged from 198.6 in pure palm oil to 224.4 in pure milk fat (MF), meaning that the saponification number decreases as the percentage of palm oil (PO) in the milk fat mixture increases. These results for milk fat were in the normal range reported by (Özkanlı and Kaya 2007 ; Samet-Bali et al. 2009). Also, our findings agreed with Hamed et al. (2019), who suggested that the increase of added vegetable oils to buffalo butter, reduced the saponification number and concluded that the saponification number is outside the normal range of pure butter in mixtures containing more than 50% of vegetable oils. Furthermore, Hamed et al. (2019) concluded that if the added level of vegetable oils is lower than 25%, saponification numbers cannot be successfully used in the detection of adulteration in buffalo butter. The saponification number in cow milk fat should be in the range of 211.7–243.3 mg KOH/g oil (Egyptian standard 154-P5/2005), the SN of the sample (25% PO of its fat) was 216.3 mg KOH/g oil, this is in the normal range recommended by ES. Therefore, the SN could not be used to detect replacement milk fat by palm oil less than 25%.
According to Table (1), the iodine number (IN) ranged from 32.94 in pure milk fat to 50.90 in pure palm oil. Our results indicated that the IN increased by increasing the palm oil level in milk, and there is a direct relationship between the percentage of palm oil in the milk fat mixture and the iodine number. These findings agree with (Zaidul et al. 2007), who assumed that iodine number increased with the presence of a high amount of unsaturated fatty acids, especially oleic and linoleic acids, in the fat, therefore it helps decide the level of hardness as a high iodine number. The same results were recorded by (Abd El-Aziz et al. 2013 ; Singhal 1980). In addition, Kumar et al. ( 2010) suggest that the iodine number was out of the normal range by the addition of vegetable oils to butter in a ratio of 50% or more. The IN should be in the range of 26.4–43.1 of milk fat (ES 154-P5/2005). In this study, the IN of a sample (50% MF: 50% PO) was 40.07. This meant the IN could not be used to detect replacing milk fat with palm oil at a level of 50% or less.
The refractive index (RI) of the fat extracted from milk samples in this study ranged from 1.452 in 100% milk fat to 1.460 in 100% palm oil (Table 1). The RI should be in the range of 1.4522–1.4543 of milk fat (ES 154-p 5/2005). The RI of the sample (75:25 MF: PO) was 1.4549. This is out of the normal range of pure milk fat, but the results are not significant with pure milk fat (P ≥ 0.05). These findings agree with those of Dehanzadeh et al. (2018) who concluded that there is a direct relationship between the amount of palm oil in milk fat and the refractive index. Butyro refractometer reading of fat extracted from different samples of fluid milk had the highest reading in 100% milk fat (43.33) compared to the lowest value in 100% palm oil (41.07). By using butyric reading, it was not possible to find any significant difference between the fluid milk samples containing different levels of palm oil, except for 100% palm oil (p ≥ 0.05).
The refractometer index of the fluid milk samples had very close values that were nearly the same but somewhat higher in MF: PO 50:50% (1.4564). However, it was found that there was a non-significant difference between all the fluid milk samples regarding the refractometer index (p ≥ 0.05).
3.2. Fatty acid profile analysis by Gas Chromatographic (GC):
In GC separation, several types of fatty acids were detected in the samples. Each peak was detected and quantified by comparing the obtained chromatogram with the standard chromatogram of fatty acids, and the value of each peak was expressed in fatty acid percent of total fatty acids in the sample, (Table 2 and Figure 1). According to the fatty acid profile analysis by GC (Table 2), a sample of 100% MF had the highest values of capric acid (C10:0), lauric acid (C12:0), myristic acid (C14:0), and stearic acid (C18:0) (3.10, 3.79, 12.16, and 13.54, respectively of total fatty acids. All these corresponding values were the lowest in the 100% PO sample. In addition, the 100% MF sample exhibited the greatest value of saturated fatty acid (SFA) composition (68.31) compared to the lowest value in the 100% PO sample (49.88).
On the other hand, the 100% PO sample had the highest composition of palmitic acid (C16:0), oleic acid (C18:1 n-9), and linoleic acid (LA; C18:2 n-6), 45.40, 39.92, and 9.55, respectively of total fatty acids (Table 2). Unsaturated fatty acids (USFA) and USFA/SFA had the highest values in the 100% PO sample, 49.92 and 1.00, respectively (Table 2). The milk samples containing different levels of palm oil were easily detected by GC, as, for example, the reduction in myristic acid (C14:0) was at a level of 24% in a milk sample containing 25% palm oil of total fat.
Results of the current study agree with (ELAASER 2017), who found that milk fat samples from local Cairo markets had an apparent increment in the palmitic (C16:0), and a decrease or absence in some other fatty acids, as (C4:0), (C6:0), (C8:0) and (C10:0), or fatty acids that were found in low content, as (C12:0) and (C14:0), and concluded that this sample characterized shortening palm oil. Furthermore, Calvo et al. (2007) discovered that the fatty acid profile of the cheese fat was significantly changed by replacing milk fat with vegetable oils in many processed kinds of cheese on the Egyptian market.
3.3 Determination of the sterol content of the fat blenders in milk at the levels by RP-HPLC.
RP-HPLC results for of determination the cholesterol levels (Table 3), showed that there was a pronounced decrease in cholesterol content as the palm oil percentage increased in the milk (392.76 µg/ml in 100% MF and 56.97 µg/ml in 100% PO), while Beta-Sitosterol concentrations were much higher as the palm oil percentage increased in the milk (3.06 µg/ml in 100% PO and 0.10 µg/ml in 100% MF).
The presence of plant sterols in milk fat indicates that the product was adulterated. Determination of cholesterol in animal fats and oils is an important topic in the food industry since high amounts of cholesterol in foods are closely related to cardiovascular disease risks. Cholesterol has also been commonly used to detect mixtures of animal oils in vegetable oils.
The content of sterols and cholesterol is widely accepted as the most important indicators for the detection of the presence of vegetable oil in milk fat. Therefore, the sterol profile of oils can identify the origin of the oils much more than their fatty acid compositions (Khorsandmanesh et al. 2020). These findings were close enough to those of (ELAASER 2017) who studied three doubtful milk fat samples purchased from the local market. Also, Hamed et al. (2019) found that the cholesterol content decreased from 278.34 to reach 117.5 mg/100 g fat, while β-Sitosterol in buffalo butter increased from 0.10 mg/100 g fat in the buffalo butter to reach 39.65 mg/100 g fat when palm oil was replaced 75% of total milk fat.
Moreover, Contarini et al. ( 2002) mentioned that pure butter contains only cholesterol, except traces of isomer 7-cholesterol, which is usually less amount than 1% but could not contain other sterols. Also, Nurseitova et al. (2021), noticed that a large replacement of milk fat by vegetable oils changed the sterol profiles completely, as they observed that six sour cream samples contained more than 95% cholesterol (pure butter) while one contained only 2.1%, which could be regarded as adulterated. Furthermore, our results correspond well with the work of Khorsandmanesh et al. (2020), who studied samples containing 1, 2, 5, 10, 20, and 50% of palm oil of total fat and suggested that among all sterols, β-sitosterol at a level of 5% in all the samples could be a good indicator for the detection of milk fat adulteration.
On the other hand, other researchers found lower cholesterol levels in milk fat (204.3 - 382.4 mg/100 g-1 (Precht 2001), and butter (200 - 250 mg/100 g (Adányi and Váradi 2003). Additionally, (Molkentin 2006), reported that the cholesterol content in butter fat was 302.6 mg/100 g-1. While higher cholesterol content (an average of 921.17 mg/kg) in cream samples was detected by (Kolarič and Šimko 2022), with no detectable stigmasterol or β-sitosterol amount. Furthermore,(HAN et al. 2007), recorded that the average cholesterol content of the cream (36% fat) was 1.370 mg/kg, while it was 769 mg/kg (38% fat) as determined by (Piironen et al. 2002) using GC.
3.4. Ftir Instrumental Analysis Of Artificial Adulterated Milk With Palm Oil
Different levels of percentages at different wavelengths of absorbance for each functional group were recorded for the milk fat samples (Table 4 and Fig. 2). The absorbing and relative intensities of wavenumbers differ slightly. However, after careful observation, some variations could be reported regarding the stretching of > CH2 of acyl chains as asymmetric (63% in 100% PO and 72% in 100% MF mixtures), the stretching of > CH2 of acyl chains (76% in 100% MF and 82% in 100% PO mixtures), C–O–C stretching (68% in 100% MF and 82% in 100% PO mixtures) and PO stretching (symmetric) of > PO2 Polyphosphate phospholipid (80% in 100% MF and 87% in 100% PO mixtures).
Fourier Transform Infrared (FTIR) spectroscopy has wide applications in food analysis, including milk, butter, cheese, fat, and oil ( (Du et al. 2019 ; Saputra et al. 2018). FTIR spectroscopy offers food manufacturers rapid quantitative quality control tools, and the means of verifying the identity, authenticity, and purity of the raw materials and ingredients that are used in the food industry (Rutkowska et al. 2015).
Table (4) illustrates the FTIR spectra of milk containing different palm oil levels (0%, 25%, 50%, 75%, and 100% of total fat in the sample). The FTIR spectra of milk samples in the current study were in the range of 3012–725 cm− 1. The spectra of milk fat were dominated by typical peaks assigned to functional groups. These findings agree with Hamed et al. (2019), who revealed that FTIR spectral bands of butter gradually increased with increasing the addition level of palm oil, as FTIR spectral bands of buffalo butter were lower in absorption in the regions of 3,000–2,800 nm than those of vegetable oils. The maximum absorption of bands for buffalo butter was at 2,925 and 2,855, which shifted to 2,928 and 2,857 nm for vegetable oils, respectively. Also, these results correspond with Cuibus et al. (2015), who studied the ATR-FTIR spectra of seven different butter samples spiked with 1–44% palm oil and found that the FTIR spectral was in the range of 3873 − 690 cm− 1.
The large variety of functional groups makes the overall butter spectra very complex, and not easy to understand, and it might be difficult to identify small variations within the spectra due to adulterant traces. The absorption peaks correspond to different and specific wavenumber ranges, which were close enough to those reported by other authors (Gori et al. 2012 ; Subramanian et al. 2011; Rodriguez-Saona et al. 2006; Karoui et al. 2005), as they found 3873–3000 cm− 1 for O–H stretching modes of water absorbing, the CH stretching vibrations in fatty acids (3000–2800 cm− 1), the stretching of C = O bonds in acids and esters
(1750–1650 cm− 1) and C = O and CCC stretching of acids (1200–800 cm− 1). So, the FTIR spectra could not observe significant differences among all samples, and this instrument could not be validated as a method to detect the adulteration of milk fat.
3.5. Sensory Evaluation Of Artificial Adulterated Milk With Palm Oil
Sensory evaluations of artificially adulterated milk with palm oil samples were carried out by a group of 15 students and postgraduate students at the Faculty of Agriculture, Alexandria University, according to the scorecard suggested by Bodyfelt et al. (1988), who gave flavor 10 points, texture 5 points, and appearance 5 points. There was no significant difference regarding sensory traits (Table 5), such as the appearance and flavor of artificially adulterated milk with different levels of palm oil and the control of normal milk (p = 0.103 and 0.224, respectively), while a significant difference was recorded regarding the texture (p = 0.048).
The color and appearance of artificial adulterated milk with palm oil and normal milk ranged from 4.70 (in 100% MF) to 3.67 (in 100% PO). The study of (Abd- ElGhany et al. 2020), who investigated the color analysis of fresh cream substitution of red palm oil, revealed that lightness exhibited a decreasing trend with the increase of the percentage of red palm oil replacement in milk.
In the current study, the flavor scores of artificially adulterated milk with palm oil decreased gradually but not significantly as the palm oil percentage increased (from 9.00 in 100% MF to 7.40 in 100% PO). These findings agree with Abd- ElGhany et al. (2020), who declared that the incorporation of red palm oil affected the flavor of the resultant ice milk. They revealed low scores for both mouth feel and flavor for iced milk with red palm oil at different levels, which could be attributed to the slight flavor of red palm oil, which was not commonly accepted by some panelists. In addition, Corradini et al. (2014) reported that a decrease in fat content resulted in a lower flavor release in ice milk. Regarding the texture of milk samples, milk containing 100% MF had the highest score (4.57) among other samples (Table 5), but there were no significant differences between normal milk sample (100% MF) and all other samples containing 25, 50, and 75% PO of total milk fat. The only significant difference was between natural milk (100% MF) and a sample containing 100% PO. This discovery should pique the interest of government officials in detecting palm oil in milk.