A Comprehensive Analysis and Detection Methodology Using Near-Infrared (NIR) Spectroscopy to Unveil the Deceptive Practice of Milk Adulteration ^†

Abstract

The process of milk adulteration is undergone by adding various substances to milk with the intent of increasing the volume or improving the appearance of the product. The very general adulterants include water, urea, starch, detergent, and even animal fats. This practice is harmful to consumers, as it lowers the nutritional value of the milk and exposes them to potential health risks, such as bacterial infections, kidney damage, and gastrointestinal disorders. Milk adulteration is a widespread problem in many countries, particularly in developing nations, where regulations are often lax or poorly enforced. To combat this issue, various measures have been taken, such as implementing stricter regulations and penalties for violators, increasing public awareness about the dangers of contaminated milk, and encouraging farmers to use proper milking and storage practices. Overall, milk adulteration poses a serious threat to public health and safety, and it is essential that consumers remain vigilant and informed about the type of product being consumed. In the current study, it is observed that dairies and milk farms are using adulterants to such an extensive amount that it is leading to various health issues, as milk is used as a product by every age of the human race. The adulterant used in the current study was urea with a concentration of 10%. The NIR spectroscopy used in the study was used as a tool to identify the difference between an unadulterated and adulterated milk samples.

1. Introduction

Milk is a substance that is produced by female mammals, such as cows, goats, and sheep, to nourish their young. Milk is treated as a food, as it contains a range of required nutrients, including proteins, carbohydrates, fats, vitamins, and minerals, which are important for maintaining good health [1]. In many cultures, cow’s milk is the most consumed type of milk. It is a rich source of calcium, which is important for building strong bones and teeth, and also contains vitamin D, which helps the body absorb calcium. Milk is also a good source of protein, which is essential for muscle growth and repair. In addition to its nutritional benefits, milk is a versatile ingredient that can be used in a wide range of recipes, including smoothies, baked goods, and sauces. It is also used to make dairy products such as cheese, yogurt, and butter. While milk is a nutritious food, some people may be intolerant or allergic to its proteins or lactose, which is a sugar found in milk. Fortunately, there are many non-dairy milk alternatives available, such as soy, almond, and oat milk, which can provide similar nutritional benefits [2]. Overall, milk is an important food item that offers a range of health benefits and can be enjoyed in many ways.

Milk adulteration refers to the practice of adding various substances to milk with the intention of increasing its volume or improving its appearance. This practice is typically carried out to increase profits and reduce costs for unscrupulous milk producers and sellers. Adulterants commonly added to milk include water, starch, urea, sugar, and even animal fat. These substances can significantly dilute the nutritional value of milk and expose consumers to potential health risks such as bacterial infections, kidney damage, and gastrointestinal disorders. Milk adulteration is a common problem in many countries, particularly in developing nations, where regulations may be lax or poorly enforced. It can be difficult for consumers to detect adulterated milk, as it often looks and tastes like pure milk. Keeping track of milk quality is vital to maintain food safety measures [3]. India’s share in world milk production is 17% [4], as milk production is an important sector of India’s agricultural economy and provides employment and income opportunities for millions of rural households. Most of the milk in India is produced by small-scale farmers who own one or two cows or buffaloes and sell their milk to local cooperatives or private dairies. However, milk and its products help humans to absorb the nutrients recommended for healthy life and also help protect them from chronic diseases [5].

Table 1. Literature survey of papers in reference to the objective.

Objective	Author	Dataset	Adulterant	Model	Results
Study milk adulteration and methods of detection of various chemical adulterants qualitatively	Riya et al. [4]	In total, twenty-five samples were collected from local vendors, dairies and farms from different areas, and brands that come under Delhi-NCR	Sugar, pulverized soap, skim milk powder, benzoic acid, salicylic acid, vanaspati, synthetic milk, and starch	Chemistry laboratory experiments	The most harmful adulterants found in milk are water, benzoic acid, and sugar
Chemometric model for rapid detection of urea and hydrogen peroxide in milk	Basak et al. [6]	A total of 35 preparations (calibration set) of standard solutions (Table 1) containing a mixture of milk, urea, and hydrogen peroxide were prepared for the development of a chemometric calibration model	Urea and hydrogen peroxide	Partial least square regression (PLSR), chemometric method, Multiplicative Scatter Correction (MSC), Savitzky–Golay derivative	From this study, we can conclude that FTIR spectroscopic data and PLS regressions can be used to determine urea (R2 = 99%) and hydrogen peroxide (R2 = 95%) as adulterants in milk when the spectroscopic data are pretreated with MSC and MSC+S-G filtering
Rapid detection of economic adulterants in fresh milk with liquid chromatography–tandem mass spectrometry	Grant Abernethy et al. [7]	Extractions were performed with 40 samples or blanks per 96-well extraction plate	An amino acid	Analysis was carried out using a Shimadzu ultra-high-performance liquid chromatograph and CTO20AC column oven at 40 °C.	This rapid and qualitative survey method may be adapted and deployed to quickly reduce the numbers of samples identified as outliers by high-throughput infrared instruments and to direct analysis to appropriate quantification methods for the detection of nitrogen-based adulteration in fresh milk
Near-infrared spectroscopy as an efficient tool for the qualitative and quantitative determination of sugar adulteration in milk	Uma Kamboj et al. [8]	A total of twenty-four samples were prepared using three different varieties of milk, out of which three samples were pure and the rest had sugar present in them	Sugar	Chemometric software (CAMO Unscrambler version X 10.3), Principal Component Analysis (PCA), partial least square regression model (PLSR)	The regression model revealed quite good results, with a coefficient of correlation higher than 0.9, and the root-mean- square error of validation (RMSEV) was 0.04

describes various studies conducted on the milk adulteration process.

According to FSSAI 2014, the addition of foreign substances has increased and the quality of food deteriorates due to the changes made. The food products no longer have the recommended parameters and come with a low-quality baseline [6]. There are various contents in milk, and the quality of the product is measured through the various components that are present, like protein, fat, lactose, etc., where protein is the significant indicator which helps to indicate the quality of milk [7]. To boost the economy and using illegal means, dairy product manufacturers and producers are constantly manipulating the components of milk by adding various adulterants, like water, urea, detergents, starch, boric acid, salicylic acid, hydrogen peroxide, sugar, sodium carbonate melamine, etc., which ultimately leads to health issues. To refrain from the practices followed by fraudulent companies, there are various major scientific instrumentations used [8]. This detailed study was conducted using various research papers, and the authors implemented various experimental techniques to visualize the absorbance spectrum and to conclude the results of adulteration and various absorbance peaks obtained during the experimental observations.

2. Milk Adulterants and Their Detection

Adulteration as it is understood is declining the quality of food and also leading to health issues in the human race; qualitative detection of these adulterants in milk is easily performed using various chemical reactions, while quantitative detections may be complex and diverse [9]. Researchers have used various methods and measures for solving the complexity of the process through spectroscopy and chromatography. These analytical methods also use sensors and capillary electrophoresis. Machine learning has helped researchers easily predict and detect adulterations in milk through the least square method, SVM, etc. [10]. The process of cheese-making is complex, which has been well described by authors Dadausis et al., occurring through a coagulation technique called milk coagulation properties [11].

2.1. Preservatives

Manufacturers are using certain preservatives as adulterants in milk and its products. Sodium carbonate (Na₂CO₃), sodium bicarbonate (NaHCO₃), and formalin (HCHO), which are easily available in chemical factories, are added to prevent curdling. To identify the mentioned preservatives, thermos-acoustic analysis can be used [12].

2.2. Cheese Whey

A byproduct form of the cheese-making process is cheese whey. It is the liquid that is left over after milk has been coagulated and curdled to form cheese. Cheese whey is a rich and prominent source of protein, lactose, vitamins, and minerals, and is generally used in food and beverage production, as well as in the production of animal feed and fertilizers. Cheese whey contains about 4–5% lactose, which can be used to produce other food products, such as whey protein concentrate and whey powder. Cheese whey protein concentrate is a high-quality protein supplement used in the food and beverage industry, while whey powder is often used in baking and cooking as a substitute for milk or as a thickening agent [13]. Proteomic analysis is an analytical technique that can identify and quantify specific proteins in a sample, including those present in cheese whey.

2.3. Urea

Urea is a chemical compound which is a common fertilizer. It can also be used as an adulterant in milk and milk products. Urea is added to milk to increase its protein content, which is measured by the amount of nitrogen present. As is known, urea contains nitrogen, which can artificially inflate the protein content of milk to an extent, making it appear as if the milk has a higher nutritional value. Consuming milk adulterated with urea can be harmful to human health. Urea is not a natural component of milk, and it can cause digestive problems, kidney failure, and even death in many cases. Additionally, urea can react with other compounds in milk to form harmful chemicals such as formaldehyde, which is a known carcinogen [6]. Infrared spectroscopy (IR) and nuclear magnetic resonance (NMR) spectroscopy are powerful analytical techniques that can detect urea in milk based on its unique molecular properties. These methods can provide rapid and accurate results.

2.4. Hydrogen Peroxide

Hydrogen peroxide (H₂O₂) is a strong oxidizing agent and can be used as an adulterant in milk to improve its appearance and prolong its shelf life. H₂O₂ is added to milk in small amounts to inhibit the growth of bacteria and to whiten or brighten the milk’s color. The usage of hydrogen peroxide in milk as an adulterant is not legal in most countries, as it can cause serious health problems. Additionally, consuming milk adulterated with hydrogen peroxide can cause digestive problems, ulcers, and other health issues. The purity of milk can be tested using various methods, such as the lactometer test, which measures the specific gravity of the milk, or the adulteration test, which detects the presence of common milk adulterants, like hydrogen peroxide. It is important and required to be aware of the potential health risks associated with consuming adulterated milk and to take appropriate measures to ensure that the milk you consume is safe and pure [6].

2.5. Sugar

As in other cases, sugar is also added to milk to make it taste sweeter. But high additives can result in calorie intake, increased risk of tooth decay, increased blood sugar levels, and nutrient displacement. Hydrometers, which measure the specific gravity of a liquid, or electronic methods such as refractometry may also be applied to detect the presence of added sugar. Chemical analysis techniques and methods such as high-performance liquid chromatography (HPLC) or gas chromatography (GC) can be used to detect the presence of added sugars in milk.

2.6. Sodium Chloride (NaCl)

The cryoscopic effect is a phenomenon in which the freezing point of a liquid is lowered when a solute is dissolved in it. The presence of dissolved solids in milk, such as lactose, proteins, and minerals, causes a reduction in the freezing point of the product as compared to water in general. This effect is commonly referred to as the cryoscopic effect in milk. The cryoscopic effect in milk is used to measure the milk’s purity and quality. The measurement of the cryoscopic effect is commonly used in the dairy industry to determine the authenticity of milk and to detect the presence of adulterants such as water, urea, and salt [14,15,16,17,18,19].

2.7. Water

Water is easily available; hence many manufacturers use water as an adulterant in milk. But using the cryoscopic phenomena, as in water is added as an adulterant or not, can be detected easily by determining the freezing temperature of the sample [20]. To measure the cryoscopic effect in milk, a device called a cryoscope is used. By measuring the degree of freezing point depression, the cryoscope can provide an accurate estimate of the milk’s total solids content, which is a significant indicator of its purity, quality, and nutritional value.

2.8. Starch

Starch is a complex carbohydrate commonly found in grains, vegetables, and fruits, and it is not typically used as a milk adulterant. However, in rare cases, unscrupulous milk vendors may add starch to milk to increase its volume and thickness or to mask the taste of the degraded quality of the milk. The addition of starch as an adulterant to milk can affect its texture and taste, making it thick and starchy to consume. Additionally, consuming milk adulterated with starch can have harmful effects on health, particularly for individuals who have digestive problems or are sensitive to carbohydrates [21].

3. Methodology

The present review of milk adulterants is based upon the various research studies and implementations performed to date, which have already been discussed in the previous sections. Various statistical and experimental analyses and methods used in the detection of adulterants through experimental methods or automation are given in a comparative analysis for various methods.

Analysis and experimental methods used for milk composition measurement are time-consuming, costly, and require manpower and are not automated [22]. NIR spectroscopy is a rapid and less complex method that can be used for real-time analysis of milk samples in a laboratory or in-line production environment. It is a cost-effective and efficient technique that can provide accurate and reliable results for milk analysis. In milk analysis, NIR spectroscopy can be used to describe and determine the various components present, such as protein, fat, and lactose content of the milk. This information is essential and important for ensuring the quality and safety of dairy products and can help detect milk adulteration.

NIR uses near-infrared radiation (12,500–4000 cm⁻¹). It provides information about the overtones and combinations of molecular vibrations. Additionally, it often analyzes samples in a non-contact manner within seconds (about 400 samples per hour). Milk samples were collected from a dairy, and then they were stored in a refrigerator (3 °C) and left at room temperature before analysis. In the current study, no prior treatments were administered to the samples, although some researchers have suggested that better results can be obtained in multivariate calibration models after pretreatment [23]. Every sample of unaltered milk was gently stirred, and the sample was analyzed in triplicate. Milk samples were analyzed on different days and in random order. The spectrum obtained from NIR analysis was monitored to find out the mean and standard deviation of each component (e.g., fat content, protein content, etc.) of the milk sample used. The quantity of urea was best visualized in the spectral ranges of 1649–1621 and 1611–1580 cm⁻¹, whereas partial least square regression was also implemented for predicting the spectral range [24]. Partial least square regression (PLS) was applied to the calibration set to build a calibration model that related the NIR spectra to the reference values. PLS is a commonly used chemometric method in NIR spectroscopy analysis that allows the relationship between the spectra and reference values to be modeled, even in the presence of collinear and noisy data.

Near-infrared (NIR) spectroscopy can be suitable for detecting specific adulterants, but it has limitations when it comes to identifying other potential adulterants.

Its suitability for specific adulterant detection is as follows:

• Rapid and non-destructive: This is defined as creating and maintaining a comprehensive spectral database of the known and familiar adulterants. It becomes possible to compare new samples against the reference spectra to identify adulteration quickly.
• Specific absorption band: Various compounds have distinct molecular structures, leading to specific absorption bands in the NIR region. This property enables the identification of specific adulterants based on their characteristic spectra and wavelengths.
• Quantitative analysis: NIR spectroscopy can be used for quantitative analysis, allowing the determination of an adulterant’s concentration in a sample. The above issue and information are valuable for accessing the severity of the adulterants used in the products.
• In-field and on-site applications: Portable NIR spectrometers are available, allowing for on-site and in-field analysis. This provides advantages for real-time screening, especially in remote locations or at various stages of the supply chain.

In summary, NIR spectroscopy can be suitable for specific adulterant detection when applied to well-characterized samples with known adulterants. However, its effectiveness depends on the complexity of the sample matrix and the availability of a comprehensive spectral database. To address the limitations, complementary techniques and methods may be used in conjunction with NIR spectroscopy for a more comprehensive adulterant-detection approach. The standard range of urea, i.e., 18–40 mg/dL in milk, indicates the presence of a sufficient amount of protein in the cow’s diet. The addition of synthetic urea to the milk falsely shows higher protein content. It also increases the shelf life and whiteness of the milk. Urea is a compound with low toxicity for human health, but its presence over the permitted limit, i.e., 70 mg/dL in milk (as per the relevant food safety standard authorities), may have necrotic and degenerative effects on the kidneys and liver, even with short-term exposure. Urea ingestion may also lead to gastrointestinal problems such as ulcers and indigestion. Urea consumption may also be carcinogenic.

The detection limits of NIR spectroscopy for reliable adulterant detection can vary depending on the following factors:

• Trace levels: in some cases, NIR spectroscopy can detect adulterants at trace levels, typically in the range of 0.1% to 1% or lower.
• Specific adulterants: The detection limits can vary depending on the specific adulterant being targeted. Some adulterants might exhibit strong NIR spectral features, making them detectable at lower concentrations, while others with weaker absorption bands might require higher concentrations for reliable detection.
• Calibration model: The quality of the calibration model used in the analysis is crucial. A well-developed and properly validated calibration model can enhance the sensitivity of the method and improve the accuracy of the results.
• Instrumentation: The sensitivity of the NIR spectrometer itself plays a role in determining the detection limits. Higher-end instruments with better signal-to-noise ratios can achieve lower detection limits.
• Pre-processing and data analysis: proper data preprocessing and analysis techniques can also improve the sensitivity of NIR spectroscopy for adulterant detection.
• Adulterant types: Different types of adulterants may have varying detection limits. For example, some inorganic substances might be easier to detect at lower levels compared to certain organic compounds.

Here, we compare NIR spectroscopy with some other common methods used for milk adulteration detection:

• High-performance liquid chromatography (HPLC):
  • Advantage: HPLC provides excellent specificity and sensitivity for detecting specific adulterants, such as melamine, antibiotics, and some chemical additives. It allows for precise quantification of adulterants at low levels.
  • Limitations: HPLC requires extensive sample preparation, which may be time-consuming and costly. It is typically limited to the detection of specific target analytes and may not be suitable for screening unknown or emerging adulterants in milk.
• Polymerase Chain Reaction (PCR) and real-time PCR,
  • Advantage: PCR-based methods can detect DNA-based adulterants, such as the presence of animal-derived material in milk. Real-time PCR offers high sensitivity and specificity for targeted DNA sequences.
  • Limitations: PCR methods are very specific to DNA-based adulterants and may not detect other types of adulterants that may be present. They can be impacted by the quality and integrity of DNA in the sample.

Comparative analysis:

• Speed and ease of analysis through NIR spectroscopy offers rapid and non-destructive analysis, requiring minimal sample preparation compared to HPLC, PCR, and ELISA.
• Adulterant diversity in NIR spectroscopy can detect a wide range of adulterants, including water, whey, and milk powder, making it versatile in milk adulteration detection. In contrast, HPLC, PCR, and ELISA are more specialized and limited to specific adulterants.
• NIR spectroscopy and HPLC are capable of quantitative analysis, which is crucial for assessing the severity of adulteration and determining adulterant concentrations.

4. Implementation

Out of all the adulterants that can be used, urea was used as an adulterant, as it is an available, cheap, and nitrogen-containing compound that can be used as an adulterant to increase the protein content of milk. Urea is a colorless, odorless, and tasteless crystalline compound that dissolves readily in water, making it easy to add to milk without detection. In milk, the protein content is an important indicator of quality, and it is directly related to the milk’s value.

When urea is added to milk, it increases the nitrogen content, leading to a false indication of higher protein content in the sample milk. Usually, it is Im < Iw (and T < 1) due to the presence of other milk components that affect the transmittance radiance, but it may also be Im > Iw (and T > 1) [25]. In Figure 1a, when 10% urea was added to milk, the protein peak on the graph became more pronounced and shifted to higher wavelengths as the concentration of protein increased. A shift in the lactose peak to a lower wavelength was observed due to the formation of a compound called ureidoisobutyric acid. There was also an increase in the fat content of the milk, resulting in a more pronounced fat peak on the graph, and in Figure 1b, the milk was not adulterated, and the peaks showed a varied difference in absorbance values.

In Figure 1b, the unadulterated milk had a characteristic NIR spectrum, with peaks at 970 nm, 1020 nm, 1050 nm, 1130 nm, and 1170 nm. These peaks were due to the presence of water, lactose, protein, and fat in the milk, whereas in Figure 1a, adulterated milk with urea showed a peak at 1040 nm in addition to the peaks mentioned above. This peak was due to the presence of urea in the milk. The absorbance of tainted milk samples usually decreases at a wavelength of approximately 1700 nm, which is the wavelength at which urea absorbs light. This occurs as a result of urea and milk water competing with one another to absorb light. The amount of urea in the milk sample correlates with the absorbance drop. The presence of urea in milk also changes the intensity of the other peaks.

5. Conclusions

This study highlights the methods used to detect milk adulterants and impurities in milk and milk products with various spectroscopic methods, like FTIR and NIR, by identification of the spectrum wavelengths corresponding to the ingredients or constituents present in the milk. This helps quality assurance and food safety teams to have control over the usage of adulterants in food products. This study also explores machine learning models that can be implemented to identify, detect, and predict the number of adulterants used in milk. In the present study, NIR spectroscopy was used to observe the absorbance values of various components in milk with adulteration and without adulteration. The study can be extended to implement machine learning models for predicting the adulterants used in the milk.