What Role is Spectroscopy Playing in Food Analysis?

Over the past few years, the rise of machine learning (ML) and artificial intelligence (AI) have helped advance spectroscopic techniques in evaluating the quality and purity of essential food products. Techniques such as near-infrared (NIR) spectroscopy and laser-induced breakdown spectroscopy (LIBS) have been used to both detect food adulteration and identify the elemental composition of different varieties of various food items. The recent literature shows that this technological shift is moving us toward ensuring global food safety and protecting public health through automated data analysis.

In this Q&A overview, we highlight how spectroscopic techniques are being used in the food and beverage industry, with a particular emphasis on how automated data analysis is unlocking the potential of these advanced techniques.

Why have spectroscopic techniques become essential in modern food and beverage analysis?

Spectroscopic methods are critical because they offer powerful tools for the detailed classification and quantification of essential parameters, nutrients, and contaminants.¹ Spectroscopic methods are often rapid, non-destructive, and require no sample preparation, whereas traditional chemical tests, which were previously used constantly in food and beverage analysis, are labor-intensive and destructive.^1,2 This is vital for sustaining the global food supply, as these techniques allow for the detection of quality, safety issues, and various forms of economic adulteration across diverse food matrices.¹

How is Raman spectroscopy specifically applied to ensure food safety?

Raman spectroscopy is a non-destructive, molecular technique. Because of its simplicity and speed, it has emerged as a prominent technique of choice in this industry.¹ Meanwhile, the success of Raman spectroscopy in this space has led to the rise of advanced variants like surface-enhanced Raman spectroscopy (SERS) being used for molecular imaging and detecting low concentrations of analytes, such as trace toxic substances.¹ When integrated with molecularly imprinted polymers (MIPs), SERS sensors can effectively recognize specific targets while mitigating interference from complex food matrices.¹ Other applications include measuring ethanol and toxic alcohols in beverages through a "through the container" approach and analyzing foodborne pathogens using microfluidic platforms.¹

In the context of honey, which is frequently targeted for fraud, how are FT-IR and NIR used for detection?

Honey is often adulterated with sucrose, maltose, or industrial syrups. Fourier-transform infrared (FT-IR) spectroscopy identifies chemical bonds and functional groups, allowing researchers to distinguish pure honey from samples containing 0–50% adulterants.^1,3 By analyzing the 1800–650 cm⁻¹ spectral region, cluster analysis can successfully separate authentic honey from deliberately adulterated market samples.¹

Specifically, for Kelulut honey, the researchers demonstrated how portable NIR spectroscopy (900–1700 nm) has been used to detect rice syrup adulteration.³ Because rice syrup is a C3 plant derivative, which is similar to honey, traditional carbon isotope ratio analysis is often ineffective, making NIR a superior alternative.³ In these studies, Principal Component Regression (PCR) models achieved a high prediction accuracy (R² = 0.914) for detecting the percentage of added syrup.³

What are the advantages of laser-induced breakdown spectroscopy (LIBS) for analyzing staple grains like rice?

LIBS is an "in-situ" technique that uses a powerful focused laser to create an intensely hot plasma on a sample's surface.² As the excited atoms emit light at characteristic wavelengths, the system generates a detailed elemental profile.² This is particularly useful for detecting heavy metal contamination (like Mn or K) in different rice varieties, which can stem from soil pollution or irrigation practices.² LIBS is favored because it is faster and more cost-effective than traditional elemental analysis methods like inductively coupled plasma–mass spectrometry (ICP-MS).²

How do researchers use hyperspectral imaging (HSI) to determine the freshness of animal-derived foods?

HSI combines the benefits of optical detection with spatial information, creating a "repository" of a sample’s spectral features across many bands. This has been used to assess chicken freshness, as an example. A recent peer-reviewed study demonstrated how HSI systems operating in the visible and near-infrared (VNIR) and short-wave infrared (SWIR) ranges are employed.⁴ These systems detect changes in protein and water content, often correlated with total volatile basic nitrogen (TVB-N) levels, which is a standard metric for spoilage.⁴ Researchers found that absorption peaks near 430 nm and 550 nm are linked to different myoglobin states, which shift as meat loses freshness.⁴

What role does machine learning and "data fusion" play in enhancing these analytical methods?

Spectroscopic data is often high-dimensional and complex, requiring chemometrics and ML for interpretation.² Techniques such as Principal Component Analysis (PCA) are used for dimensionality reduction, while supervised algorithms like support vector machines (SVM), random forest (RF), and artificial neural networks (ANN) perform classification.^2,4

Data fusion further enhances accuracy by integrating data from multiple sources (for example, combining VNIR and SWIR). For example, a Residual Network (ResNet), which is a deep learning architecture, achieved a 98.87% accuracy rate in classifying chicken freshness when combined with feature-wavelength selection algorithms like competitive adaptive reweighted sampling (CARS).⁴

Are there specific techniques for identifying the geographical origin of food products?

Yes. Atomic spectroscopy techniques, in particular, have been used to identify the geographical origin of food products. For example, inductively coupled plasma (ICP) techniques, including ICP-MS and ICP-OES, are used for high-sensitivity trace elemental analysis.¹ By analyzing sixty different elements in chicken meat, researchers used canonical discriminant analysis to achieve 100% accuracy in identifying the geographic origin of the samples.¹ Similarly, weakly selective fluorescence probes combined with pattern recognition can trace rice back to its specific region of origin.¹

What is the significance of portable and handheld instruments in this field?

Portable instruments are becoming more routine in analytical workflows, as more researchers demand smaller instruments that can be taken on-site.⁵ The benefit of using these instruments in food analysis is that they can have positive externalities on the production and supply chain processes.¹ For example, these devices are used for sorting and grading crops like apples based on soluble solids and firmness, and for performing on-site bulk monitoring of honey.^1,3 Portable Raman devices also enable non-invasive measurements of alcohol strength and purity directly through glass bottles, significantly increasing the efficiency of market inspections.¹

References

1. Workman, Jr., J. A Review of the Latest Spectroscopic Research in Food and Beverage Analysis. Spectroscopy. DOI: 10.56530/spectroscopy.ob9768p3 (accessed 2026-05-11).
2. Iroshan, A.; Feng, J.; Han, B. In-situ Detection of Rice Using Laser Induced Breakdown Spectroscopy and Machine Learning. Spectroscopy 2025, 40 (3), 14–19. DOI: 10.56530/spectroscopy.bn8372e6
3. Ying, L. L.; Saleena, L. A. K.; Solihin, M. I.; et al. Physicochemical Analysis and Detection of Rice Syrup Adulteration in Kelulut Honey Using Portable Near-Infrared Spectroscopy. Spectrosc. Suppl. 2024, 39 (s10), 29–37. Available at: https://www.spectroscopyonline.com/view/physicochemical-analysis-and-detection-of-rice-syrup-adulteration-in-kelulut-honey-using-portable-near-infrared-spectroscopy
4. Chen, S.; Tang, S.; Yu, Z.; et al. Detecting Chicken Freshness Utilizing VNIR, SWIR Spectroscopy, and Data Fusion. Spectroscopy 2024, ASAP. DOI: 10.56530/spectroscopy.sI7768t8
5. Crocombe, R. Spectrometers in Wonderland: Shrinking, Shrinking, Shrinking. Spectrosc. Suppl. 2022, 37 (s11), 6–11. DOI: 10.56530/spectroscopy.Iz8466z5