Infrared Spectroscopy for Detecting Adulterants in Food and Traditional Indonesian Herbal Medicine

Abstract

Adulteration in food has a detrimental effect on the product’s quality, which may result in nutritional deficiency. It can also be harmful, particularly for medicinal herbal products, as certain adulterants are very toxic to the body. It is thus critical to screen and identify the presence of any probable adulterants in food and herbal products in order to ensure the customers’ safety. Infrared Spectroscopy is a very viable technique for such purposes, as specific molecules absorb light at specific wavelengths, which correspond to the vibration frequency of the molecule’s bonds. Adulterants typically have their own unique molecular fingerprints, which exhibit their own vibrational spectra. On the basis of these principles, IR Spectroscopy is sensitive enough to detect the presence of potentially detrimental as well as harmful compounds added to food and medicinal products. This chapter describes how Infrared Spectroscopy can be beneficially used in detecting the presence of adulterants that are unintentionally or irresponsibly added to food or traditional herbal products. The last one is more of a result of a lack of knowledge and awareness of food adulterants and their deleterious impacts.

1. Introduction

Adulteration is the irresponsible act to downgrading the quality of food products delivered to the market, either by the replacement of some ingredients with other substitute components or by addition of certain component into its ingredients. The substitute components added into food products are then called as “adulterants” [1]. Responsible authorities must guarantee the safety of food products supplied to consumers in order to prevent any possible health hazards caused by foodborne illness or injury. Adulterations potentially occur at any stage in the food product’s supply chain, starting from the farm, in the production facility, through distribution system, and eventually all ways to the market where the products are purchased by customers [2, 3, 4].

There are numerous possible practices for such reckless adulteration, which can be broadly classified into intentional and unintentional [1, 5]. Adulteration practices can be driven by many motivations, either to get economic gains or to cause harm to society, i.e., public health disasters, economic reasons, or terror-related purposes. According to the United Nations’ Committee on World Food Security [6], specific goals related to the previously described issues are the food security, which means that all people, at all times, have physical, social, and economic access to sufficient, safe, and nutritious food that meets their food preferences and dietary needs for an active and healthy life. In a more visualized way, the classification of any efforts and motivations to undermine food security can be clearly depicted as in Figure 1.

Figure 1.

Food risk matrix (adapted from [7]).

Adulteration is also found in many medicinal or pharmaceutical products (drugs). The types of adulteration made in the drugs are varied, from replacing the main drug’s active ingredients with similar but cheaper substitutes. These substitutes may have similar or weaker strengths or even induce complementary effects that might affect other health problems in their consumers. In some drugs that are sold illegally, a.k.a. street drugs, psychoactive substances that provide mood-altering, stimulant, or sedative effects are added. In these kinds of drugs, psychoactive substances (such as caffeine, paracetamol, amphetamines, cocaine, and morphine) are added to modify their physicochemical or psychological activity [8, 9, 10, 11, 12].

Indonesian traditional herbal medicine (a.k.a. as “Jamu”) is predominantly made from extracted natural herbs, such as roots, bark, flowers, seeds, leaves, and fruits. Traditionally, Jamu is produced in liquid form, and it has been used by the Indonesian people for centuries to maintain health and alleviate illness. Even though Western (standard) medicine is becoming more prominently used in Indonesia, Jamu remains popular both in rural and urban areas, since people believe that it is regarded safer than any chemical synthetized drug. The production of Jamu is done traditionally with knowledge that is inherited from generation to generation. Unfortunately, many of these traditional producers have lack knowledge about how to process their Jamu in a hygienic manner. Often, the marketed products are found to be severely contaminated with bacteria, yeasts, and molds. Meanwhile, nowadays, Jamu is also processed in various forms, such as capsules, tablets, and powders, and is also labeled with many producers’ brands. Unfortunately, some irresponsible manufacturers mixed them with other chemical drugs as adulterants, which could be toxic and potentially cause health risks [13].

In this chapter, it will be described how Infrared Spectroscopy could beneficially contribute to screening and detecting the presence of adulterants, either in food or Indonesian traditional herbal medicine (Jamu), and how the recognition of adulterants’ molecules can be sensed using this measurement technique and the possible recent developments in customizable handheld systems for making the screening task easier.

2. Food fraud and modes of adulteration

The terminology Food fraud is described in the first quadrant is defined as any suspected intentional activity by businesses or individuals with intention of deceiving consumers and gaining an economic advantage. These categorized activities are also known as Economically Motivated Adulteration (EMA), which happens when an individual intentionally omits, removes, or replaces a vital element or component of a product. Additionally, EMA occurs when a substance was added to a product with the intention of improving the appearance or value. More specifically, as illustrated in Figure 2, there are several forms of food fraud.

Figure 2.

Numerous forms of food fraud.

Among the seven forms of food fraud, four of them can be considered food adulteration, i.e., dilution, substitution, concealment (masking/hiding), and unapproved enhancements. Meanwhile, the other three, namely counterfeiting, gray market trading, and mislabeling, are all forms of fraud that are not classified as adulteration.

Dilution is often accomplished by increasing the volumetric amount of liquid food or medicinal items by adding additional water or another solvent, thereby reducing the quantity of main ingredients. Sometimes, to conceal the dilution, additional substances such as sugar and artificial flavors are added. The principal reason is economic, since the price of the main ingredients in the diluted product will be lower for the same volume sold, whereas the consumers cannot realize the fraud attempts. One widely practiced example of this kind of fraud is the dilution of pure coconut water with ordinary water and the concealment of the dilution effect with additional sugar [14]. Another common example is the dilution of fruit juices, which are considered to be the most targeted food commodities for adulteration and fraud. It is common practice in the fruit juice industry to dilute fruit juice commodities with more water in order to reduce the required fresh fruit and conceal it with the addition of sugars, pulp wash, and other additives such as fruit flavor, as well as the undeclared addition of a lower-quality juice to a product (a.k.a. juice-to-juice adulteration) [15].

Substitution is another common adulteration practice, which is done by replacing the whole or part of the main ingredients with cheaper and harmless substitutes. Milk and dairy products adulteration are good examples of this kind of adulteration practiced in many places on earth. Common substitutes used in milk and dairy products adulteration include more watering, additional vegetable protein, milk from different animal species, whey protein, synthetic urine to boost the nitrogen content, hydrogen peroxide, and even urea and melamine [16]. The last was a well-known Chinese milk scandal that happened in the year 2008 [17].

Unapproved enhancement is the last category of food fraud, which is considered as adulteration. It involves the addition of unknown and unreported substances to food items in order to improve its quality characteristics. This kind of fraud is normally found in many supplemental medicines [18], among others are as dietary and weight-loss supplements [19, 20], cognitive enhancement supplements [21], and supplements for sexual enhancements [12].

3. Fraud and adulteration in medicinal products

Besides being practiced in food commodities, drug fraud and adulteration are also happening in medicinal and pharmaceutical products. Typical adulterants are introduced to raise the product’s weight, mimic or enhance its pharmacological action, or facilitate drug transport. Drug adulteration is considered as type of fraud in which legal or illicit drugs are cut or modified to decrease their quality below the level at which they are sold.

There are many facets of pharmaceutical products’ counterfeiting that are normally practiced, i.e.,

1. Perfect imitation of the original products, i.e., with the same packaging and the same Active Pharmaceutical Ingredients (APIs) at the right concentration.

2. The packaging is the same as the original pharmaceutical products, but it contains the APIs at different concentrations from those declared.

3. The fake drug looks exactly like the real thing, but it does not have any APIs in it at all.

4. The product contains ingredients different from those declared.

5. The product’s packaging has been falsified.

Hence, the points numbered 2, 3, 4 listed above can be considered as adulteration, since it includes modification of the product’s ingredients.

Drugs adulteration is harmful and may bring lethal consequence since these adulterated drugs may induce risk of overdose and toxicity that is dangerous to human tissues and organs. The adulterants can be either inert and/or pharmaco-logically active substances that have similar properties to the active ingredients in the original drug itself. These include:

1. diluent substances, i.e., inactive, inert, or structurally different compounds that are added to the drug in order to increase the size of the dosage form and share certain properties such as color, consistency, and taste. Commonly used diluents are anhydrous lactose, lactose monohydrate, talc, and sugar alcohols such as sorbitol, xylitol, and mannitol.

2. adulterants normally used in drugs, among others are:

   1. Caffeine: even often being used as common bulking agents (diluents), caffeine provides stimulant properties similar to cocaine and amphetamine.

   2. Paracetamol: often used as diluents, but also provides analgesic properties similar to heroin.

   3. Lidocaine: has similar anesthetic properties to cocaine.

   4. Procaine: in addition to providing anesthetic properties, also has been found to vaporize heroin at a lower temperature and therefore facilitate smoking

Contamination is another type introduction of unwanted elements into or onto a system during manufacturing, packing, or transport, and the unwanted elements are thus referred to as contaminants. These substances will have an adverse effect on the product or process. Contaminants that are normally found in drugs can be categorized as:

1. Physical contaminants: among others are metal-, glass-, mineral-, and insect fragments, machinery wear debris, greases and oils, dust, and corrosion, fibers, and hairs

2. Chemical contaminants: consist of heavy metals and elemental contaminants; cleaning agents (bleach and detergents); hydrocarbons (oils, gasoline, and diesel); preservatives and colors; and cross-contamination resulting from packaging, manufacturing, or storage.

3. Biological contaminants include viruses, bacteria, and fungi that may be introduced from circulated water and air inside the production system or may be transmitted via the workers that do not strictly follow the Standard Operating Procedures (SOPs).

4. Indonesian herbal medicine and possible adulterants

Indonesia is the world’s largest archipelagic country and consists of 18,110 islands with a diverse range of habitats and an extremely complicated geological history similar to neighboring Southeast Asian countries [22]. The climatic, geological, and biodiversity conditions in these regions stimulate the evolution of megadiverse plants with a lot of endemic and ecologically highly adapted species [23]. The biodiversity in Indonesia is considered to be the second largest in the world and has a lot of indigenous medicinal plants. It is thus understandable that many Indonesian communities, particularly in rural areas, still rely on the use of traditional herbal medicines to maintain and improve their health conditions, as well as to prevent and treat diseases [24].

In Indonesia, people generally named these traditional herbal medicines with the word Jamu. This word etymologically came from ancient Javanese language, as a contraction of two words, i.e., jampi and usodo, which means prayer and health, respectively [25]. The historical records on Jamu as ancient traditional medication were engraved as wall relief sculptures at several Hindu–Buddhist temples such as Borobudur, Prambanan, and Penataran temples, as can be seen in Figure 3 [25, 26].

Figure 3.

Historical record on ancient used of Jamu engraved as wall relief sculptures at the Borobudur temple.

Jamu as herbal medications are widely practiced in many communities living in the islands of the Indonesian archipelago, spreading from the main islands of Sumatra [27, 28, 29], Java and Madura [30, 31, 32, 33], Kalimantan [34, 35, 36], Sulawesi [37, 38, 39], Papua [40, 41], as well as their smaller neighboring islands. Traditionally Jamu is prepared from raw herbal ingredients in two methods, i.e.:

1. The fresh Jamu (a.k.a. Jamu segar): by grating and pounding the raw herbal ingredients using a grater tool or a mortar pounder, respectively, then mixing with fresh drinking water, and squeezing using muslin cloth to extract the herbal juice. The extracted juice is then ready to be consumed.

2. The boiling Jamu (a.k.a. Jamu godhog): by boiling mixed herbal ingredients with water in a stewing pot, usually up to the boiling water reduced by half. Upon cooling and filtering, the decoction of herbal extract is ready to be consumed.

These preparation methods are normally adopted as personal consumption in many households or by traditional Jamu’s handlers, which can be found in many cities and urban areas in Indonesia. People called these Jamu’s handlers according to how they carry their sold products, i.e., as bakul Jamu gendong, bakul Jamu dorong, bakul Jamu sepeda, respectively, as can be depicted in Figure 4 (i.e., the left, middle, and right).

Figure 4.

Typical Jamu handlers in many cities and rural areas in Indonesia.

Meanwhile, in modern processing methods, the Jamu product preparation is done following complex, standardized, and strict methods under controlled standard operating procedures (SOPs). These SOPs are implemented in every processing stage, starting from collecting raw herbal ingredients from farmers, where the originality, purity, water content, and content of active ingredients of each raw herbal component are always checked, since they will contribute to the quality of the final products. Next step is the preprocessing stages including cleaning and washing, chopping into smaller chips of herbal ingredients, drying, and storage for further processing. These preprocessed ingredients are the being weighted and mixed in composition according the prepared prescription, milled, and sieved to get a smooth powdered preparation. This final preparation is then packaged as solid products in capsules, pills, or powdered forms, or being bottled in liquid form upon dilution in prescribed solvents. List of main Jamu factories in Indonesia can be seen in [24].

Either in the traditional or modern processing modes mentioned above, there are open opportunities for adulteration of the prepared products.

Most traditional Jamu handlers are typically honest and humble people who strictly hold the values of responsible business. So, they do not have any intention of adulterating their processed and sold products with ingredients other than those taught by their ancestors. Most problems in the products sold by these traditional handlers are solely due to contamination, either chemically or biologically. This contamination is unfortunately caused by a lack of knowledge on how to maintain the hygiene of their raw herbal ingredients, processing tools, and product storage bottles. They do not even understand how to maintain and control the quality of the Jamu they produce. Adulteration in traditional Jamu is only found to be done by a small number of naughty producers and handlers that are hoping to make more profit from their business. They usually use artificial sweeteners such as cyclamate and saccharin [42], which can have negative health effects if consumed for an extended period of time.

Adulteration is also not possible with the Jamu that are produced by main Jamu factories, since their production activities are under the strict supervision of the National Agency of Drug and Food Control (BPOM). Quality screening is regularly done to assure that all licensed factories fulfill the standard of how to make good traditional medicine (abbreviated as “CPOTB” in Indonesian). Factories that are found not to conform to the standard can be further investigated, with the possibility of suspending their operation or even having their production license revoked. Adulteration in modern processed Jamu is often done by unregistered and illegal small home industries, whose adulterated products are usually distributed by street drug handlers, which can be found on many road sides in some areas in big cities, as depicted in Figure 5.

Figure 5.

Typical street drug handlers that usually sell adulterated and illegal modern processed Jamu.

5. Infrared spectroscopy: principle of molecule’s spectra quantification

Infrared (IR) photon spectra are optical electromagnetic radiation components with their wavelengths ranging from 700 nm to 30 $\mathrm{\mu m}$. This spectral range is located between visible light and microwave radiation, and the photon’s energy of these electromagnetic radiation can be expressed mathematically using a well-known Planck-Einstein equation of E = hν = h(c/λ). So, the photon energy range of infrared electromagnetic radiation (800 nm–30 μm) spread between$1.78x{10}^{-1}$ and $4.14x{10}^{-3}\mathrm{eV}$. One electron volt (eV) is defined as the unit of energy required to accelerate an electron through a potential difference of 1 V.

In Infrared Spectroscopy, IR photons are used to probe the spectral absorption behavior of the investigated molecules. When photons of infrared radiation interact with molecules, one possible mode of light-matter interaction is absorption. Different molecules possess specific spectral absorption behaviors that can be used as a determining signature for recognition of the molecule, a.k.a. molecule’s spectral fingerprint.

The molecule’s spectral absorption profile represents the number of absorbed photons by molecules at different wavelengths of the incoming photons. This spectral absorption behavior is usually represented as percent transmittance (% T) of the infrared radiation that is received by the detector as a function of a photon’s wavelength (in nm). But most spectroscopists prefer to represent the absorbance behavior as a function of wavenumber (in ${\mathrm{cm}}^{-1}$). The units for wavelength and wavenumber are inversely proportional to each other, so spectral range of 700 nm–30 μm correlated to wavenumber of 12,500–333.3 ${\mathrm{cm}}^{-1}$.

5.1 Molecular vibrations

Atoms in a molecule are connected by chemical bonds having bonding properties determined by their electronegativity (EN), which determines an atom’s ability to attract a bonding pair of electrons. An atom’s electronegativity is determined by atomic number as well as the configuration of most outer electrons. The difference in electronegativity between two atoms can be used to predict the bonding types between two atoms in a molecule. Based on these differences, chemical bonds between two atoms can be classified into four different classes, i.e.,

• nonpolar covalent bond $\left(0<\mathrm{EN}<0.39\right)$

• moderately polar covalent bond $\left(0.4<\mathrm{EN}<0.89\right)$

• very polar covalent bond $\left(0.9<\mathrm{EN}<1.69\right)$, and

• ionic bond $\left(\mathrm{EN}>1.7\right)$

The higher the difference in electronegativity, the stronger the bond between two atoms will be, which means that the higher the required photon energy to break their bond.

The behavior of molecular vibration can be analogically modeled using a harmonic spring-mass system as usually used in explaining well-known Hook’s law. The law states that the force $F$ required to extend or compress a spring by some distance $∆x$ is linearly proportional to the distance, with the proportionality constant $k$, which representing the spring’s stiffness. The diatomic model can be seen in Figure 6.

Figure 6.

Analogical harmonic spring-mass system for diatomic molecular vibration, displaced by $\mathit{∆}\mathit{x}$, either compressed (black color) or extended (red color).

When a photon of energy $∆E$ comes and is absorbed by a molecule, this photon energy will stimulate atoms in a molecule to vibrate, i.e., with the vibration frequency $\mathit{\nu }$ according to the following relation, i.e., $∆E=\mathrm{h\nu }$ where $h$ is the Planck’s constant. This photon absorption will stimulate fundamental vibration if the molecule is initially in its ground state. Meanwhile, absorption of multiple photons will stimulate the first or even higher overtones.

The vibration frequency in wavenumber ($\stackrel{\sim }{\mathit{\nu }}$) can be expressed as:

$$\stackrel{\sim }{\mathit{\nu }}\left[{\mathit{cm}}^{-1}\right]=\frac{1}{2\mathit{\pi c}}\sqrt{\mathit{k}\frac{\left({\mathit{m}}_1+{\mathit{m}}_2\right)}{{\mathit{m}}_1{\mathit{m}}_2}}\left(x0.01\right)$$E1

Factor 0.01 in the parenthesis is used to convert the unit ${\mathit{m}}^{\mathit{-}1}$ into ${\mathit{cm}}^{\mathit{-}1}$. The vibrational frequency thus depends strongly on the bond’s strength, molecular geometry, and the masses of connected atoms. Meanwhile, the bond’s strength itself is affected by chemical environment factors and other influencing factors, i.e., hydrogen bonding, coupled vibrations, electronic effects, and Fermi resonance.

5.2 Vibrational modes

As explained previously, the possible vibrational motions in diatomic molecules are compression and extension. Meanwhile, in triatomic or higher polyatomic molecules, the possible motions are more complex. These possible vibrating motions of connected atoms are called vibrational modes, and IR Spectroscopy deals with spectrally quantifying the behavior of these molecular vibrations.

Basically, there are six fundamental (normal) vibrational modes of polyatomic molecules, which can be classified into two classes, i.e., stretching (either symmetrical or asymmetrical) and bending (wagging, twisting, scissoring, and rocking). These bending vibrations can be further divided into two types, i.e., the in-plane (happened on the same plane, i.e., the scissoring and rocking) and out-of-plane (happened not in the same plane, i.e., the wagging and twisting) vibrational modes. Besides these six normal vibrational modes, there are also local modes, which happen as the excitation energies get higher and stimulate the overtones [43, 44].

As an example, let us see the vibrational modes of water molecules, as water is the most common adulterant/diluter that is normally found in food and medicinal products. Water molecule $\left(H_{2}O\right)$ contains three atoms, so it has $3N-6=3\left(3\right)-6=3$ vibrations, i.e.:

1. Symmetric stretch $\left({\stackrel{\sim }{\nu }}_1\right)$ at $3280{\mathrm{cm}}^{-1}$ (3048.8 nm)

2. Symmetric bending $\left({\stackrel{\sim }{\nu }}_2\right)$ at $1654{\mathrm{cm}}^{-1}$ (6045.9 nm)

3. Asymmetric stretch $\left({\stackrel{\sim }{\nu }}_3\right)$ at $3490{\mathrm{cm}}^{-1}$ (2865.3 nm)

These normal modes and related spectrum in the Vis and IR regions can be depicted in Figure 7.

Figure 7.

Normal vibrational modes of water molecules.

5.3 Infrared spectroscopy measurement techniques

There are basically two types of spectrometers used in IR Spectroscopy, i.e., the Dispersive IR (DIR) and the Fourier Transform IR (FTIR) spectrometers. The first typically uses radiation from a broadband source that passes through the sample, which is then dispersed by a monochromator into component frequencies that are further directed onto a detector for recording the spectrum. This recorded spectrum is then compared with the reference beam, as most of the Dispersive IR spectrometers have a double-beam configuration.

Meanwhile, the second type adopts the interferometric measuring principle, where the splitted reference and measuring beams are combined to make an interference pattern (interferogram). This recorded interferogram is then processed by using a Fourier Transform algorithm to extract the spectrum. The FTIR is preferable to the DIR due to superior speed and sensitivity, since all frequencies are examined simultaneously in a shorter time. Detailed comparison of these two measurement techniques can be found in [45].

5.4 Processing and interpretation of IR spectra

In order to obtain accurately identified spectra of examined samples, it is critical to ensure that a number of steps, beginning with sample preparation and measurements, preprocessing, reading, and interpreting the measured spectra, are carried out correctly and precisely. As a rule of thumb, there are a number of key considerations, as concisely given in [46], and these are:

1. Clearly understand how the spectra are measured, i.e., knowing the instrumental resolution, sampling method used, and any spectral preprocessing, i.e., subtraction, smoothing, baseline correction, and other processing algorithms applied like spectral derivation. All of these factors will affect the appearance of the spectra.

2. When using wavenumber (preferable for most spectroscopist), the spectra are being plotted with highest wavenumber at the most left of the graph and followed by smaller wavenumber running to the right.

3. The peaks of the spectra should be on scale, i.e., between absorbance units of 0 and 2 (or transmittance units of 10–90%); otherwise, it is considered to be off scale

4. Measured spectra should be in a “good quality,” i.e., low noise, little or no baseline offset, a flat baseline, peaks on scale, and no spectral artifacts (i.e., the unwanted spectra that are usually contributed during the sample measurement). The use of background correction, sealing and desiccating the measuring instrument, as well as purging measurement area with dry N₂ gases are typical efforts used to minimize occurrence of spectral artifacts.

5. Identify any possible spectral artifacts before assigning other peaks.

6. Use other supporting knowledge about the examined sample, i.e., how the samples being prepared, its physical properties and appearances, or when possible, analysis results using other measurement techniques such as Raman, UV–Vis, or nuclear magnetic resonance (NMR) in order to make further accurate interpretation tasks easier.

7. Identify the positions of common functional groups or chemical bonds that might be picturized in the acquired spectra. Spectral absorption ranges from common functional groups given in Table 1.

8. Assign the most intense bands first, followed by ones with lesser intensities. One does not need to identify all peaks in the spectra, but most importantly, it is the spectra that are most important for the goal of analysis.

9. In the case that the changes in the intended observed effect are not clearly seen in an overlapped spectral band, deconvoluting the spectra into its constructing peaks would be helpful to discern the observed effect.

10. In several cases, the pattern changes in observed spectral bands are not clearly “readable” from the spectra itself. Thus, an additional analysis tool is then required to make the interpretation of the pattern of spectral changes easier to extract. Chemometrics analysis can be beneficial to discern these spectral patterns.

Position of peaks		Functional groups or Chemical Bond
$\left[{\mathbf{cm}}^{\mathbf{-}\mathbf{1}}\right]$	[nm]
3300–3500	2857.1–3030.3	$\mathbf{N}\mathbf{-}\mathbf{H}$ stretch
3200–3400	2941.2–3125.0	$\mathbf{O}\mathbf{-}\mathbf{H}$ stretch
2850–3100	3225.8–3508.8	$\mathbf{C}\mathbf{-}\mathbf{H}$ stretch
2100–2260	4424.8–4761.9	$\mathbf{C}\mathbf{=}\mathbf{N}$ strech; $\mathbf{C}\mathbf{\equiv }\mathbf{C}$ strecth
1650–1800	5555.6–6060.6	$\mathbf{C}\mathbf{=}\mathbf{O}$stretch
500–1000	10,000–20,000	$\mathbf{C}\mathbf{=}\mathbf{C}$ stretch; $\mathbf{C}\mathbf{-}\mathbf{H}$ wagging aromatic ring fingerprint region

Table 1.

Positions of some common functional groups or chemical bonds.

5.5 Chemometric analysis

Infrared Spectroscopy provides spectral data that contains extremely large amounts of information, as a fingerprinting tool to extract several important properties of the measured sample [47]. Interpretation of infrared spectra is a complex process, since the spectra show peaks and vibrations of molecular bonds in the measured sample, as well as their combinations and overtones. Chemometrics is thus required to interpret this data [48].

Chemometrics, according to the International Chemometrics Society, is the science of relating measurements made on a chemical system or process to the state of the system using mathematical or statistical methods. Since its first introduction in the 1960s, chemometrics has become dispersed in many areas [49]. It is also widely used as an essential tool in modern analytical Vibrational-Spectroscopy-based instruments to efficiently extract the maximum useful information from spectra in a relatively short time.

The most commonly used chemometric methods in IR spectral analysis can be classified into the following groups [50]: the processing techniques (e.g., normalizations, baseline corrections, centering, derivatives, multiplicative corrections, and required combinations of them) to enhance the information contained in the spectra; the classification methods (either supervised or unsupervised), which link any a priori knowledge about the samples to be classified; and the regression methods, which link the spectra to quantifiable properties of the samples, mainly those based on multivariate regression through principal component analysis (PCA) and partial least squares regression (PLSR) and Discriminant Analysis (DA) approaches. In PCA, the correlated response variables in spectral data are transformed into non-correlated variables known as principal components (PC’s) [51]. Meanwhile, the PLSR is used to reduce the predicting variables to a smaller set of predictors, which are then used to perform a regression [52].

6. Detecting adulterants in food and traditional Indonesian herbal medicine using infrared spectroscopy

In this section, we will devote our discussion on how to recognize the adulteration based on the IR Spectra acquired. The use of necessary preprocessing techniques will be also addressed to make the changes in spectral pattern more easily readable. Cases of adulterants discussed in this section are as previously described in Sections 2-4.

6.1 Over-dilution with water or other solvents

Dilution is the process of adding additional water or other solvents to a product in order to increase total volume while decreasing the concentration of active ingredients. This procedure is often used to adulterate milk, juices, and other liquid food or medicinal products. In order to recognize the dilution, one should observe and analyze peaks of water molecule’s spectral signature embedded in the spectrum of the investigated samples. Absorption spectrum of water molecule is given in Figure 8.

Figure 8.

Water molecule absorption spectrum (reproduced from [53]).

Breast milk is regarded as the optimum exclusive source of nutrition for the first 6 months of baby’s life and may remain part of the healthy infant diet for the first 2 years of life and beyond. The WHO and UNICEF recommend that a baby initiate breastfeeding within the first hour of birth and be exclusively breastfed for the first 6 months of life [54]. Unfortunately, not all postpartum mothers can provide sufficient breastfeeding for their babies; therefore, in some countries there are noble initiatives to establish a human milk banking [55]. These milk banking systems collect breast milk from donor mothers, who have more than their babies need, process and redistribute the donated milk to the needed babies, i.e., the premature and sick babies whose mothers do not have enough milk for them.

Sadly, these noble initiatives were disgraced by irresponsible acts to adulterate the milk. In this context, efforts to develop a screening system to differentiate between real BM and its adulterated ones are necessary to guarantee that the needed babies receive the required nutrition available in BM. Infrared Spectroscopy proved to be beneficial as a screening technique for this problem.

One interesting case to discuss here is the detection of adulterated milk as described in [56]. This paper analyzed the difference among Human Breastmilk (BM) and related adulterated with water (W) and cow milk (CM), semi-skimmed cow milk (SSCM), and skimmed cow milk (SCM). The recorded FTIR spectra of the comparison among original BM and adulterated variants are given in Figure 9. From the spectra we can observe the prominent absorption peaks of water, as in Figure 8, i.e., in regions 3500–3000 cm⁻¹ and 1730–1600 cm⁻¹ due to H₂O stretching and H-O-H bending vibration. These peaks seem to be overlapped with other absorption bands of other molecules in milk, make them difficult to clearly discern visually. The other absorption bands observed in the ranges 1630–1680 cm⁻¹ and 1510–1570 cm⁻¹ are due to C=O stretching or N-H and C-H bending vibration of the milk proteins, and the bands of 2920, 2850, and 1743 cm⁻¹ may be due to the antisymmetric and symmetric stretching of CH₂ and C〓O groups from the fatty milk components, respectively [57].

Figure 9.

FTIR spectra comparison of pure BM, CM, SCM, and SSCM.

Meanwhile the absorption bands in the ranges 3200–3700, 1030–1200, and 900–450 cm⁻¹ have been associated with carbohydrate structures. That is why other processing technique such as derivative, PCA, and PLSR are needed to get better insight into the separation between real and adulterated milk, as can be seen in Figure 10.

Figure 10.

(a) 3D score plot by PCA and (b) by PLSR-DA [56].

6.2 Boosting the nitrogen content in milk

In the year 2008, the world was shocked by the deadly milk scandal in China, where around 300,000 children were poisoned by melamine, a chemical that is usually used to make plastic that was added to powdered milk to increase nitrogen content in milk. Melamine consumption may result in reproductive damage, bladder or kidney stones, and bladder cancer. Infrared Spectroscopy was beneficial as a rapid and accurate technique in screening the availability of melamine in milk. There is a distinct absorption band of melamine, i.e., characterized by the out-of-plane bending of the 1,3,5-s-triazine ring of melamine at 814 cm⁻¹, which is absent in pure milk. This spectral difference can be depicted in Figure 11. Using region of 851.62–798.39 cm⁻¹ as calibration, Jawaid et al. [58] obtained a good PLS model for very low detection range up to 1%, i.e., with a good correlation coefficient of R² = 0.999 and accuracy and high precision (i.e., 0.12–1.10% error and relative standard deviation of 1.38–2.07%, respectively), and with limit of detection (LOD) can achieve 1 ppm.

Figure 11.

FTIR transmittance spectra of milk and melamine [58].

6.3 Artificial sweeteners

Sugar (sucrose) additions in food and drinks serve several functions; among other is changing the flavor profile to increase appetite and likelihood of consumption. It is also used to provide a sweet taste to lessen the bitter and sour taste of some ingredients in some medicinal or traditional herbal drinks (Jamu). Historically, the use of artificial sweeteners has been used a long time ago [59] as an enhancer of the sweet taste in food/drink products or as a substitute for sugar for those with a problem of sugar consumption. Unfortunately, it has been found in recent decades that consumption of these kinds of sweeteners may induce metabolic disorders [60, 61].

IR spectra from sucrose and other common artificial sweeteners can be seen in Figure 12.

Figure 12.

Absorbance spectra of several common artificial sweeteners, reproduced from [62].

There are many overlapped absorption peaks observed, even though there are some specific peaks observed that might characterize each of these sweeteners, as well as differ from sucrose, which can be used as a rapid non-destructive tool for sweetener identification. As in the case of milk, the use of chemometric analysis can accurately classify the distinct signatures of sweeteners.

Interesting work is reported by Wang et al. [63] that builds and tests PLS model to identify differentiate single and double mixtures of artificial sweeteners. Mixture of sweeteners usually done to reduce bitter off-taste following its consumption [64]. The developed model may have prediction performance with R² ranging from 0.9981 to 0.9996 for a single type of sweetener, and from 0.9397 to 0.9998 for a blend of two types of sweeteners. Blends of more sweetener types may produce lower fitting degree of prediction and measured value, i.e., 0.7648–0.9997, 0.7292–0.9994, 0.6617–0.9968, for three, four, and five sweetener mixtures, respectively.

6.4 Adulterated Jamu with active pharmaceutical ingredients (APIs)

Some active pharmaceutical substances found to be added in the illegally marketed herbal medicine in Indonesia. Paracetamol, phenylbutazone, and dexamethasone are among chemical substances that are usually added to Jamu, in order to provide pain relieving effects and comfortable sensations to the consumers. Unfortunately, many consumers do not realize the harmful side effects that can be induced by consuming these adulterated Jamu [65].

In order to identify the addition of APIs in Jamu, one should know how the IR spectrum of each API differs specifically. Different absorbance IR spectra from paracetamol, phenylbutazone, and dexamethasone are depicted in Figure 13. It is observed that some distinct unique absorption peaks can be used for recognition.

Figure 13.

Absorbance spectra of paracetamol, phenylbutazone, and dexamethasone, reproduced from [62].

There were also evidence that in some arthritic and rheumatic herbal medicine (jamu pegal linu) was adulterated with the Piroxicam, i.e., a type of nonsteroidal anti-inflammatory drug (NSAID) used to relieve the symptoms of painful inflammatory conditions such as arthritis and other similar chemical such as prednisone and naproxen [13, 65]. Figure 14 provides IR transmission spectra of piroxicam, and its typical absorption bands are given in Table 2.

Figure 14.

IR transmission spectra of piroxicam.

Frequency (cm^−1)	Band Assignments
3854	OH stretching
3338	NH stretching
2365	CH stretching
1629	Amide $\mathbf{C}〓\mathbf{O}$ stretching
1527	Aromatic $\mathbf{C}〓\mathbf{C}$ stretching
1435	$\mathbf{N}▬\mathbf{H}$ bending
1181	$\mathbf{C}▬\mathbf{S}$ stretching
938, 830	$\mathbf{C}▬\mathbf{H}$ bending

Table 2.

Infrared absorption bands of piroxicam.

Readers who are interested in deepening their understanding on how IR Spectroscopy can be beneficially used to authenticate herbal products can refer to these interesting review papers [66, 67, 68].

7. Summary and future outlooks

In this chapter, evidence of food fraud and types of adulteration in food and medicinal products has been described. Indonesian traditional herbal products, as traditional medication ways that are widely practiced among societies residing in the Indonesian archipelagos, are also prone to irresponsible adulteration practices.

Infrared Spectroscopy has proved to be a sensitive and rapid identification tool for recognizing these adulterants. Fundamental principles for generating, measuring, and analyzing the specific spectrum of adulterant molecules have also been described. Infrared spectra contain extremely large amounts of information as a fingerprinting tool to extract several important properties of the measured sample, and the elaboration of chemometric analysis makes the analysis of the investigated samples more clear and easier to handle.

Grasping better insight into the responsible spectral bands that are strongly correlated with the observed variables would be beneficial in developing a customized portable measurement system [69]. Such a system might be helpful in terms of technology to keep people safe from adulterated food and medicinal products.