Fourier-Transform infrared (FT-IR) spectroscopy is a simple, fast and highly specific technology that can provide valuable insights into the complex chemical make-up of foods. Infrared provides tools, especially in the fingerprint region of the spectrum, to detect specific compounds in biological systems without the use of time-consuming methods or the use of hazardous organic solvents. Advances in FT-IR instrumentation and pattern recognition techniques have made it possible to extract information related to composition and conformation of food components from the spectra. We have evaluated the capability of infrared spectroscopy in classification and quantification of chemical compounds of interest for the dairy (butter) and tomato industries.
Authentication is a critical quality issue for organic products since consumers are willing to pay 10-40% price premiums. There is a need for rapid and reliable analytical tools for determination of authenticity since traditional methods often involve time-consuming and laborious processes. Our objective was to evaluate the application of infrared spectroscopy combined with pattern recognition techniques to discriminate among organically and conventionally-produced butter in relation to quality and authenticity. Spectra from butter purchased from a local market (Columbus, OH) were collected by using Attenuated total reflectance (ATR) spectroscopy and analyzed using soft independent modeling of class analogy (SIMCA), a multivariate classification technique. This simple protocol generated unique mid-infrared signature profiles that permitted the chemically-based classification of butter samples based on manufacturer and production practice (organic vs. conventional). By using the spectral region from 1400-800 cm-1, multivariate (SIMCA) modeling showed well-separated clusters that discriminated among butter samples according to manufacturer, due to -HC=CH- trans bending out of plane vibration modes, (966 cm-1) presumably attributed to conjugated fatty acids. Infrared spectroscopy combined with multivariate analysis provides a simple and efficient tool for monitoring butter authenticity with minimal sample preparation.
The objective of the second study was to develop a simple, accurate and cost effective protocol using ATR-IR spectroscopy and multivariate analysis to determine sugars in tomatoes. Tomatoes, the second most produced and consumed vegetable in the United States, are classified for use as fresh or processed tomato products based on their sugar and acid profile. Current methods to analyze sugar content of tomatoes are time and labor intensive making efficient assays for quantification desirable. Samples were obtained from genetically diverse tomato varieties that encompassed hybrids and elite parents used in the processing and fresh market industry. Samples were centrifuged, the supernatant vacuumed dried on a ZnSe crystal and infrared spectra collected. Enzymatic kits for glucose and fructose were used as reference methods. Multivariate models (PLSR) accurately predicted glucose and fructose using the supernatant with R-values > 0.98 and SECV <0.25g/100g, using the fingerprint infrared region of 1200-900cm-1 for sugars. Vacuum drying of the sample onto the ATR crystal caused spectral artifacts in some samples. ATR-IR combined with chemometrics could provide the tomato industry with a simple and high throughput method for determination of sugars in tomatoes that could lead to improved varieties with enhanced characteristics for industry and consumer demands.