Applications for portable NIR spectroscopy in food analysis

Near-infrared spectroscopy (NIR) refers to the study of the interaction between matter and light within the electromagnetic spectrum's near-infrared region which ranges from 750 to 2500 nm ^[1]. When infrared light interacts with a sample's molecules, the amount of electromagnetic energy that is reflected, transmitted, and absorbed for each wavelength depends on the bond types present in the sample ^[1]. The C-H, N-H, and OH vibration bonds are the most prevalent in the NIR region, determining the spectra shape of a given substance.

NIR spectroscopy is commonly used to measure and quantify a sample's proximate composition, such as protein, moisture, dry matter, fat, and starch. Additionally, the NIR spectrum reflects its physical properties or characteristics ^[1]. As a result, when applied to food, the NIR spectra of samples can provide information not only about the food's chemical composition but also about its functionality via a non-destructive, rapid, and clean approach that does not require the use of reagents ^[2].

Only recently has NIR technology evolved towards miniaturized devices, making it possible to bring NIR analysis from lab to field. Portable near-infrared (NIR) spectroscopy is an excellent tool for monitoring crop quality and determining optimal cultivation conditions and harvesting time. The importance of controlling food quality cannot be overstated given the high vulnerability of foods to content variation, the need to maintain freshness to prevent quality loss, and the possibility of illegal adulteration. Moreover, the complex nature of the food production, delivery chain and the need to reduce analysis time to a minimum has made portable spectrometers a revolutionary step forward in this field ^{[5] [6]}.

Appling computational techniques to the absorption spectra obtained with a NIR analyzer, Parastar et al. were able to discriminate fresh from thawed meat as well as correctly classify chicken fillets according to the growth conditions of the chickens with good accuracy ^[3]. Using similar tools, Kucha and Ngadi were able to assess the freshness of minced pork meat ^[4]. These computational methods, usually referred to as ‘chemometrics’ use a variety of algorithms and statistical techniques such as multiple linear regression, partial least squares regression, and principal component analysis to analyze the data coming from the spectrometer. These methods translate the spectral information into chemical and functional properties associated with the sample ^[2].

Portable NIR analyzers have been utilized for on-farm monitoring of feed and forages to assess their quality. In this process, a sample of the feed is placed in front of a scanner that analyzes it, providing results to the farmer or nutritionist. This allows them to make prompt management decisions regarding the feed, significantly reducing the time required for obtaining results from a few days to a few seconds. For instance, the dry matter content of corn silage in cattle feed can vary greatly from day to day, up to 41 % over six months. By making adjustments on the spot, cows receive a more consistent ration leading to an improvement in the general health of the herd. This is observed through changes in blood parameters and a reduction in mastitis, resulting in increased milk production. Moreover, this technology can potentially decrease feed wastage, thereby reducing costs and increasing revenue^[7].

Another valuable application field for portable NIR spectrometry is the on-field evaluation of crops in all phases of produce growth. Tardaguila et al. have studied the absorption wavelength of 160 individual grapevine leaves from eight different varieties grown under various environmental conditions. They specifically targeted water content assessment to identify an optimization strategy for irrigation in the wine industry ^[8]. During harvest season, NIR spectroscopy has been used to assess the on-tree ripeness of olive fruits ^[9], grapes ^[10], and tomatoes ^[11] enabling harvest time optimization and even automated fruit picking using agricultural robots. Following harvest, spectroscopy techniques in the NIR range could be helpful to farmers, consumers, and quality control officers to conduct a rapid nondestructive examination of produce quality. This technique also allowed for the detection of pineapple fraud due to mislabeling of conventionally produced fruits as organic ones ^[12].

There are two main methods for analyzing the absorption spectrum of organic materials in the NIR spectrum. The first method is diode-array-based spectroscopy. This technique uses dispersive grating to separate the light reflected or transmitted from the sample into its wavelength components. Each component is then focused on a different pixel of a linear detector array. This method is considerably fast and can be used for real-time measurements. However, the light throughput of the diode-array spectrometer is inversely proportional to its spectral resolution, which limits its effectiveness. Additionally, the high cost of linear arrays sensitive in the near-infrared region may limit their adoption for certain applications, especially in agriculture and food. 

The second method for obtaining an absorption spectrum is Fourier transform interferometry. In this method, the incoming light is split into two paths, with one directed toward a fixed mirror and the other toward a movable mirror. When these paths are recombined, an interferogram is obtained.

By performing a Fourier transform of this interferogram, the spectrum of the incoming light can be obtained, and with proper calibration, the absorption spectrum of the sample can be determined. Using this technique, all wavelengths are measured simultaneously, providing better throughput and higher sensitivity without compromising spectral resolution (usually referred to as ‘Fellgett’s advantage’). In this technique, only a single NIR photodetector is used instead of an array, keeping the cost low.

Additionally, absorption peaks at 2300 nm were used to estimate the alcohol concentration in each of the beverages. This measurement showed values in accordance with the actual presence of alcohol in the liquids, confirming how accurate estimation is possible using this compact device and method.

While the portion of light irradiated onto a sample is regularly reflected by its superficial particles, the rest penetrates the sample. Here, the light is repeatedly diffused through refractive transmission, light scattering, and surface reflection until it makes it out of the sample to be measured. The diffuse reflection spectrum obtained through this measurement is similar to the absorption spectrum of the sample.

The obtained spectra were characterized by peaks at 1600 nm-1800 nm, caused by the first overtone of CH stretching from lipids and protein. The differences among the various spectra were more evident when observing the second derivative of the spectra. Classification of the different kinds of nuts was possible through the principal component analysis method.

The potential applications of NIR infrared spectroscopy in the food industry have been widely documented by numerous scientific publications for several years. The advent of portable instruments is now moving the analysis from lab to field reducing dramatically the time for results from days to seconds. Most notably this hardware miniaturization driven by Hamamatsu’s MEMS technology happens without compromising on sensitivity or resolution.

New computational techniques are continuously being developed to analyze and compare the absorption spectra and estimate the content of specific chemical compounds in food. These methods are making the technology more and more accessible to non-technical users across the industry.

Portable FTIR analyzers can be a valuable tool to address many vital challenges in the food industry. For example, they can help improve crop yield thus providing an alternative to deforestation when facing an increased food demand. Integrating these technologies into farming can limit water waste when optimizing irrigation and limiting food waste across the supply chain. Finally, FTIR analyzers can contribute to the improvement of our food quality making it safer and healthier for us and for all the animals depending on us.

[1] K. B. Beć, J. Grabska, and C. W. Huck, “Near-Infrared Spectroscopy in Bio-Applications”, Molecules, vol. 25, no. 12, p. 2948, Jun. 2020, doi: 10.3390/molecules25122948.

[2] D. Cozzolino, “The Ability of Near Infrared (NIR) Spectroscopy to Predict Functional Properties in Foods: Challenges and Opportunities”, Molecules, vol. 26, no. 22, p. 6981, Nov. 2021, doi: 10.3390/molecules26226981.

[3] H. Parastar, G. van Kollenburg, Y. Weesepoel, A. van den Doel, L. Buydens, and J. Jansen, "Integration of handheld NIR and machine learning to 'Measure & Monitor' chicken meat authenticity" in Food Control, vol. 112, pp. 107149, 2020. doi: 10.1016/j. foodcont.2020.107149.

[4] Kucha, C.T., Ngadi, M.O. “Rapid assessment of pork freshness using miniaturized NIR spectroscopy”. Food Measure 14, 1105–1115 (2020). https://doi.org/10.1007/s11694-019-00360-9

[5] J.-H. Qu, D. Liu, J.-H. Cheng, D.-W. Sun, J. Ma, H. Pu, and X.-A. Zeng, "Applications of Near-infrared Spectroscopy in Food Safety Evaluation and Control: A Review of Recent Research Advances" Critical Reviews in Food Science and Nutrition, vol. 55, no. 13, pp. 1939-1954, 2015. doi: 10.1080/10408398.2013.871693.

[6] K. B. Beć, J. Grabska, and C. W. Huck, “Miniaturized NIR Spectroscopy in Food Analysis and Quality Control: Promises, Challenges, and Perspectives,” Foods, vol. 11, no. 10, p. 1465, May 2022, doi: 10.3390/foods11101465.

[7] "Can On-Farm NIR Analysis Improve Feed Management?", Penn State Extension. [Online]. Available: https://extension.psu. edu/can-on-farm-nir-analysis-improve-feed-management.

[8] J. Tardaguila, J. Fernández-Novales, S. Gutiérrez, and M.P. Diago, "Non-destructive assessment of grapevine water status in the field using a portable NIR spectrophotometer", J. Sci. Food Agric., vol. 97, pp. 3772-3780, 2017. doi: 10.1002/jsfa.8241.

[9] A. J. Fernández-Espinosa, "Combining PLS regression with portable NIR spectroscopy to on-line monitor quality parameters in intact olives for determining optimal harvesting time", Talanta, vol. 148, pp. 216-228, 2016. doi: 10.1016/j.talanta.2015.10.084.

[10] G. Ferrara, V. Marcotuli, A. Didonna, A. M. Stellacci, M. Palasciano, and A. Mazzeo, “Ripeness Prediction in Table Grape Cultivars by Using a Portable NIR Device”, Horticulturae, vol. 8, no. 7, p. 613, Jul. 2022, doi: 10.3390/horticulturae8070613.

[11] H. Yang, B. Kuang, and A.M. Mouazen, "In situ Determination of Growing Stages and Harvest Time of Tomato (Lycopersicon Esculentum) Fruits Using Fiber-Optic Visible—Near-Infrared (Vis-NIR) Spectroscopy", Applied Spectroscopy, vol. 65, no. 8, pp. 931-938, 2011. doi: 10.1366/11-06270.

[12] C. L. Y. Amuah, E. Teye, F. P. Lamptey, K. Nyandey, J. Opoku-Ansah, and P. O. Adueming, "Feasibility Study of the Use of Handheld NIR Spectrometer for Simultaneous Authentication and Quantification of Quality Parameters in Intact Pineapple Fruits", Journal of Spectroscopy, vol. 2019, Article ID 5975461, 9 pages, 2019. doi: 10.1155/2019/5975461.

[13] Z. Husain and R.A. Schwartz, "Food allergy update: more than a peanut of a problem", International Journal of Dermatology, vol. 52, pp. 286-294, 2013. doi: 10.1111/j.1365-4632.2012.05603.x.

[14] S. H. Sicherer and H. A. Sampson, "Food allergy: Epidemiology, pathogenesis, diagnosis, and treatment", The Journal of Allergy and Clinical Immunology, vol. 133, no. 2, pp. 291-307.E5, Feb. 2014. doi: https://doi.org/10.1016/j.jaci.2013.11.020

[15] A. Luparelli, I. Losito, E. De Angelis, R. Pilolli, F. Lambertini, and L. Monaci, “Tree Nuts and Peanuts as a Source of Beneficial Compounds and a Threat for Allergic Consumers: Overview on Methods for Their Detection in Complex Food Products”, Foods, vol. 11, no. 5, p. 728, Mar. 2022, doi: 10.3390/foods11050728.