Near infrared spectroscopy has been used as an analytical technique for almost a century. But improved understanding of agricultural and biological materials, chemical processes and reactions is making the method even more important to today's manufacturing and control applications.
Spectroscopic methods for monitoring chemical, biological and industrial processes are becoming attractive because of their potential for remote, online, multi-point and real-time analysis. Especially vibrational spectroscopy, which reflects detailed changes in chemical and physical states of molecular species or reaction and properties related to the final product.
Spectroscopy primarily deals with the interaction of light and matter. Different forms of spectroscopy are possible depending on the absorption, transmission, emission and scattering characteristics of light. Some of the techniques used for quantitative and qualitative characterization of chemical, agricultural, and biological systems are UV-visible, near-infrared, mid-infrared, fluorescence, raman and nuclear magnetic resonance.
Most spectroscopy users have limited formal training and theoretical background.
Know the basics
Infrared radiation is commonly defined as electromagnetic radiation with frequencies between 14,300 and 20 [cm.sup.-1] (0.7 [micro]m and 500 [micro]m). When a normal molecular motion such as vibration, rotation, rotation/vibration or lattice mode, combination, difference or overtone results in a change in the molecule's dipole moment -- a molecule absorbs infrared radiation in this region of the electromagnetic spectrum. The corresponding frequencies and intensities of these infrared bands, representing the functional or chemical groups of a molecular structure, constitute a spectrum.
The theoretical basis of infrared absorption can be explained using the "ball and spring" concept from classical mechanics. When an oscillating electric field of radiation interacts with the oscillating molecular dipole, an energy transfer occurs. The transfer happens when the frequency of radiation matches the resonant frequency of the assembly of balls and springs. If the atoms constitute point masses and the bonds represent weightless springs, the resonant frequency can be given by v = (1/2[pi])[square root](k/[micro]) where v is the frequency, [micro] is the reduced mass ([micro] = [m.sub.1] [m.sub.2] / [m.sub.1] + [m.sub.2]), and k is the force constant.
As an example, consider the carbon hydrogen (CH) bond for which k is 463 N/in and [micro] is 1.53 X [10.sup.-27]. The resonant frequency from equation 1 is 8.75 X [10.sup.-13] Hz or 2,919 [cm.sup.-1]. This is characteristic of the asymmetric stretch of the [CH.sub.2] functional group that can be observed around 2,925 [cm.sup.-1].
Other examples are the nitrogen-hydrogen (N-H) stretch of the amide group at 1,650 [cm.sup.-1] for amide I and 1,600 [cm.sup.-1] for amide II -- for protein, the carbonyl group represented by carbon oxygen (C = O) at 1,740 [cm.sup.-1] for ester, which is fat related. Similar assignments can be made for other chemical groups that constitute different food system components.
The widely used near-infrared (NIR) spectroscopy is a routine procedure used for determining the amount of protein, fat and moisture in food. Other applications include:
* evaluating meat and fruit quality using a combination of spectral data and statistics,
* monitoring cell density and lactic acid during fermentation,
* routine examination of conformity of pharmaceutical process,
* starch hydrolysis,
* livestock feed evaluation,
* plant tissue analysis and
* real-time grain evaluation during harvest.
Online NIR systems for quantitative food component measurement and quality evaluation of industrial and agricultural products are on the rise. Key NIR advantages are sampling ease and procedure simplicity. Disadvantages are a lack of qualitative information because NIR bands are broad and mainly due to overtones rather than fundamental molecule vibrations.
Another commonly exploited region is the mid-infrared (between 670 and 4,000 [cm.sup.-1] or 2.5 to 14.9 [micro]m), also known as Fourier transform infrared (FTIR) spectroscopy. FTIR spectroscopy is based on the Michelson interferometer configuration and is a standard analytical technique in chemistry and related fields.
FTIR is fast, sensitive, has high signal-to-noise ratio, predicts multiple analytes in a mixture and provides qualitative information of component structural groups and quantitative determinations. One general constraint is lack of an appropriate sampling procedure. However, due to a wide range of sampling accessories -- especially with the advent of "smart" sampling -- FTIR is becoming a major force in solid, liquid, gas and thin film characterization.
Most commonly used FTIR techniques include:
* Fourier transform infrared attenuated total reflectance (FTIR-ATR) -- for liquid, paste and thin film analysis,
* diffuse reflectance infrared fourier transform spectroscopy (DRIFTS) -- for powers and
* Fourier transform infrared photoacoustic spectroscopy (FTIR-PAS) -- for nondestructive characterization and depth profile analysis of powders, solids and thin films.
The FTIR region responds to water presence due to its strong absorbance in the 3,000 [cm.sup.-1] and 1,600 [cm.sup.-1] region overlapping or obscuring absorbance due to the chemical group in this region. However, by applying background spectra and powerful statistical data processing techniques, useful information can be extracted from the spectra.
"Chemometrics" which deals with measuring and analyzing chemical data gives spectroscopists different ways to solve the calibration problem in quantitative determinations. Understanding chemometrics does not have to involve understanding the method's mathematics. The key is in knowing which model to use for a given problem.
Other tools are spectral data enhancement techniques, such as the first and second derivative spectra, plus spectral libraries and chemical data bases for several materials.
An example of quantitative FTIR for fat measurement might depend on absorbance corresponding to the ester carbonyl groups of fat molecules at 1,745 [cm.sup.-1]. Protein measurement could involve examining absorbance at 1,538 [cm.sup.-1] by peptide bonds of protein molecules and lactose measurement based on absorbance at 1,042 [cm.sup.-1] by hydroxyl groups of lactose molecules.
Examples of qualitative characterization include monitoring the oxidation of edible oil, potato chips, frying oil quality, determining secondary structure of proteins and monitoring proteolysis and lipolysis during cheese ripening. A change in peak intensity corresponding to assigned wave number for a functional group indicates a change in molecular structure.
Wavelength assignments for typical functional groups in fatty acids are absorption bands around 1,745 [cm.sup.-1] for ester carbonyl groups, 2,930 and 2,853 [cm.sup.-1] for C-H stretch in methylene groups and 1,160 [cm.sup.-1] for carbon-oxygen (C-O) bonds of lipid. The wavelengths of protein-related functional groups are asymmetric and symmetric N-H stretch and hydrogen-bonded primary amide around 3,350 and 3,170 [cm.sup.-1]; the C = O stretch in amide band that overlaps that of N-H bend at 1,640 [cm.sup.-1] and the C-N stretch assigned to a wave number of 1,425 [cm.sup.-1]. Similar evaluations can be made by examining the entire spectrum for structural groups of interest.
Other applications include:
* environment monitoring for volatiles,
* examining biological cells, tissues and fluids for carbon-dioxide,
* detecting grain contamination,
* monitoring fermentation processes,
* characterizing and classifying bacteria,
* analyzing hazardous waste and environmental extracts,
* authenticating food ingredients,
* food adulteration studies,
* polymer and biomaterial characterization and
* biomedical analysis.
Another novel application for FTIR-PAS is with the generalized two-dimensional (G2D) correlation analysis for non-destructive depth profile study of multi-layer materials. An example of classifying honey with different sugar adulterants using discriminant FTIR data analysis (Figure 1) offers potential for detecting the adulterant type.
Potential topics for future research include standardizing the calibration procedure and online and real-time analysis for process optimization and product quality control in all sectors.
ASAE member Joseph Irudayaraj is a professor in the Agricultural and Biological Engineering Department, Pennsylvania State University.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||analytical technique for monitoring chemical, biological and industrial processes|
|Publication:||Resource: Engineering & Technology for a Sustainable World|
|Date:||Sep 1, 2000|
|Previous Article:||Looking at the Big Picture.|
|Next Article:||Calling Dr. Internet.|