An Introduction to Near Infrared Spectroscopy

Skip to Navigation

Education Article

  • Published: Jul 1, 2014
  • Channels: Raman / Infrared Spectroscopy
thumbnail image: An Introduction to Near Infrared Spectroscopy

Jerome (Jerry) Workman

Jerome (Jerry) Workman, Jr., an ASTM Fellow, is vice president of Research and Engineering for Argose Inc. in Waltham, Mass.. During his career, Workman has focused on molecular spectroscopy and chemometrics. He has published six books and more than 200 scientific papers on these subjects. He is the recipient of the ASTM International Award of Merit, and the Eastern Analytical Symposium 2002 Award for Achievements in NIR Spectroscopy.

Infrared energy is the electromagnetic energy of molecular vibration. The energy band is defined for convenience as the near infrared (0.78 to 2.50 microns); the infrared (or mid-infrared) 2.50 to 40.0 microns; and the far infrared (40.0 to 1000 microns). However, even though official standards, textbooks, and the scientific literature generally state that the NIR spectral region extends from 780-2500 nanometers (12821 - 4000 cm-1), a simple set of liquid phase hydrocarbon spectra demonstrates that the vibrational information characterized by the harmonic vibrations of the C-H stretch fundamental and their corresponding combination bands occurs from approximately 690 to 3000 nm. The predominant near-infrared spectral features include: the methyl C-H stretching vibrations, methylene C-H stretching vibrations, aromatic C-H stretching vibrations, and O-H stretching vibrations. Minor but still important spectral features include: methoxy C-H stretching, carbonyl associated C-H stretching; N-H from primary amides, secondary amides (both alkyl, and aryl group associations), N-H from primary, secondary, and tertiary amines, and N-H from amine salts.

The advantages touted for NIR measurements over other vibration techniques have proven themselves true throughout the 1980s up until today, they include: (1) C-H associated vibrational information is repeated 8 times from 690 nm to 3000 nm; (2) Simple harmonics may be selected or more information rich combination regions; (3) Low cost instruments with high signal-to-noise (SNR) are simple to make and typically exhibit signal-to-noise ratios (SNR) of 25000-100000:1; (4) High NIR throughput is possible, even when employing low cost fiber optics; (5) Variable pathlengths for industrial use are possible, typically 1 mm to 10 cm or more using different NIR spectral regions; and (6) NIR Light penetrates plant and animal tissue easily for biomedical applications (when using 900 nm and longer). Table 1 shows the relative intensities of C-H stretch bands for infrared and the various NIR overtone regions (first through fourth overtone). This repetitive information gives a great deal of flexibility for pathlength selection and information content. For example, the closer one approached the fundamental region the more detailed is the vibrational information.

Table 1. Relative intensities of C-H stretch bands for infrared and the various NIR overtone regions


Wavelength region

Relative Intensity


Optical Density

Fund. (n)

3380 - 3510 nm


0.01 mm


1st over. (2n)

1690 - 1755 nm


1.0 mm


2nd over.(3n)

1127 - 1170 nm


1.0 cm


3rd over. (4n)

845 - 878 nm


10.0 cm


4th over. (5n)

690 - 780 nm


10.0 cm


Qualitative and quantitative near infrared (NIR) spectroscopic methods typically require the application of multivariate calibration algorithms and statistical methods (i.e. chemometrics) to model NIR spectral response to chemical or physical properties of the samples used for calibration {1-2}. The NIR method relies on the spectra-structure correlations existing between a measured spectral response caused by harmonics of the fundamental vibrations occurring at infrared frequencies. These harmonic vibrations occur at unique frequencies depending upon the quantity of absorber (analyte), type of absorbing molecules present within the sample, and the sample thickness. Quantitative methods are possible where changes in the response of the near infrared spectrometer are proportional to changes in the concentration of chemical components, or in the physical characteristics (scattering/absorptive properties) of samples undergoing analysis. Recent refinements of the NIR measurement technique include the emergence of chemometrics and the diminishing distinction between near infrared, and infrared as measurement techniques. Rather the techniques are complementary, with each spectral region providing unique advantages for the analyst. This article introduces the use of near infrared spectroscopy as a technique for identification, quantitative analysis, and structure-correlation assessment of NIR functional group frequencies. For a wide range of NIR applications, particular attention is given to the appearance of methyl, methylene, methoxy, carbonyl, and aromatic C-H groups; hydroxy O-H; and N-H from amides, amines, and amine salts.

Near infrared spectroscopy is used where multicomponent molecular vibrational analysis is required in the presence of interfering substances. The near infrared spectra consist of overtones and combination bands of the fundamental molecular absorptions found in the mid infrared region. Near infrared spectra consist of generally overlapping vibrational bands that may appear non-specific and poorly resolved. The use of chemometric mathematical data processing and multiple harmonics can be used to calibrate for qualitative of quantitative analysis despite these apparent spectroscopic limitations. Traditional near infrared spectroscopy has been most often used for analysis of lignin polymers (2270 nm), paraffins and long alkane chain polymers (2310 nm), glucose based polymers such as cellulose (2336 nm), amino acid polymers as proteins (2180 nm), carbohydrates (2100 nm), and moisture (1440 and 1940 nm). When analyzing synthetic and natural materials NIR spectroscopy has shown unprecedented industrial success in multiple applications utilizing tens of thousands of instruments in grains, forages, baking products, flour, beverages, feeds, pharmaceuticals, dairy products, hydrocarbons and petrochemicals, fine chemicals, radioactive and hazardous materials, and medical imaging and diagnostics. The basic uses of near infrared spectroscopy have been for process control, for quality assessment, for identification of raw materials and process byproducts, and for chemical quantitative analysis of complex mixtures.

Note that a near infrared spectrum consists in the convolution of the measuring instrument function with the unique optical characteristics of the sample being measured (i.e. the sample is an active optical element of the spectrometer); the reference values are those chemical or physical parameters to be predicted using the NIR spectroscopic measurements. A spectrum may, or may not, contain information related to the sample chemistry measured using any specific reference method. Spectra-structure correlation provides a basis for the establishment of a known cause and effect relationship between instrument response and reference (analyte) data, in order to provide a more scientific basis for multivariate-based near infrared spectroscopy. When performing multivariate calibrations, analytically valid calibration models require a relationship between X (the instrument response data or spectrum), and Y (the reference data). The use of probability alone tells us only if X and Y 'appear' to be related. If no cause-effect relationship exists between spectra-structure correlation and reference values the model will have no true predictive importance. Thus, knowledge of cause and effect creates a basis for scientific decision-making.

Factors affecting the integrity of the teaching samples used to calibrate spectrophotometers for individual NIR applications include the variations in sample chemistry, the variations in the physical condition of samples, and the variation in measurement conditions. Teaching Sets must represent several sample 'spaces' to include: compositional space, instrument space, and measurement condition (sample handling and presentation) space. Interpretive spectroscopy is a key intellectual process in approaching NIR measurements if one is to achieve an analytical understanding of these measurements.


The discovery of the infrared region in 1800 is credited to William F. Hershel's famous work, "Experiments on the Refrangibility of the Invisible Rays of the Sun", read April 24, 1800 at the Royal Society, Phil. Transact. Roy. Soc. 90, 284-292. O.W. Wheeler {3} described the near infrared region as extending "from about 2 m into the visible at about 0.7 m." R.F. Goddu and D.A. Delker {4} demonstrated the spectra-structure correlations and average molar absorptivity for a number of functional groups for the NIR region (1.0 to 3.1 m), and the maximum recommended pathlengths for twelve solvents (useful for NIR spectroscopy) over the wavelength region 1.0 to 3.1 m.

Professor J.W. Ellis {5} reviewed work below 3 microns for absorption of organic liquids. W. Kaye {6} provided a summary review of the work in near infrared spectroscopy from the late 1920s until April 1954. R.F. Goddu {7} compiled an extensive review of near-infrared spectrophotometry prior to 1960. K.B. Whetsel {8} reviewed the significant work in near infrared spectrophotometry prior to 1968. Stark et al. {9} review work related to near-infrared analysis (NIRA) prior to 1986. Schrieve et al. {10} discuss applications for the short-wave near infrared (SW-NIR) region, referring to synonyms such as "the far-visible", the "near, near-infrared", or the "Herschel-infrared" to describe the range of approximately 700 to 1100 nm of the EMS (electromagnetic spectrum).

Since 1912, research investigations into the molecular structures of organic compounds using infrared spectroscopy has grown. This early work by W.W. Coblentz reported on the IR absorption of water. Coblentz used the spectral region from 1 to 8 microns. The new decade of the 1960s brought about a prolific series if papers related to 'direct determination' and the measurement of light transmittance and reflectance properties of intact biological materials.

In 1973, P. Williams reported the use of a commercial NIR grain analyzer for analyses of cereal products following the pioneer work of Norris and others. Later Williams and Karl Norris would edit a comprehensive text {11} on the subject of NIR analysis for commercially important biological materials. An early textbook written by B. Osborne and T. Fearn describes the uses of NIR in the food and beverage industries {12}. Early work, most of which used multiple linear regression to identify key calibration wavelengths, used both filter and dispersive scanning instruments to relating NIR spectral response to reference analytical data. P. Williams and co-workers described the work in flour milling using near-infrared spectroscopy for the determination of moisture in cereals and cereal grains. The authors describe the uses of NIR for protein and moisture analysis in hard and soft wheat flours.

Forage analysis using NIR measurement has been a major application of the technique largely due to the work of J.S. Shenk, M. Westerhaus, W. Barton, G. Marten, N. Martin, and a host of others who improved upon the technique and worked toward its widespread use and acceptance among scientists as a valid analytical technique. One could not mention NIR and Forage analysis without listing the primary reference source in the field since August 1985. The handbook edited by G.C. Marten, then of the University of Minnesota; J.S. Shenk of The Pennsylvania State University; and F.E. Barton of the Richard Russell Research Center, USDA; has become the most used handbook for NIR forage analysis {13}. A more recent comprehensive information source for NIR analysis of forages is a book edited by C. Roberts, J. Workman, and J. Reeves {14}.

Near infrared has been used for analysis of gasoline, fine chemicals, polymers and pharmaceuticals, both with dispersive and Fourier-Transform (FT-NIR) based instruments {15}. Pharmaceutical applications include: identification of raw materials and product quality, moisture and solvent content in drying or solvent removal processes, residual drug carryover in manufacturing facilities, mixing quality evaluation, and imaging of tablets and packaging systems. Additional pharmaceutical applications include real-time monitoring of fermentation processes, and products. More recently, medical applications for near-infrared have proliferated into areas of blood analyte monitoring and imaging of materials including tissue {16}.

When measuring hydrocarbon mixtures, such as fuels or solvent mixtures, near infrared spectroscopy does not directly measure hydrocarbon classes such as olefins or naphthenes, as such, but rather it measures functional group absorptions such as methyl, methylene, methine, and aromatic stretching and deformation vibrations. The ratios of these various absorptions, when correlated (using a variety of well described multivariate calibration techniques to known physical or compositional parameters, for a learning or teaching set, provides a correlation estimate of the hydrocarbon class composition from various unknown complex mixtures.

In summary, the NIR spectral region is information rich. For example, C-H stretch combination bands occur four times, the C-H overtones occur four times, the O-H combination bands (three times), the O-H overtones (three times), and the N-H overtones (three times). The important molecules for NIR measurements have most often been water (as O-H stretch) and associated solutes, proteins, carbohydrates, fats and oils, and hydrocarbon classes. Comprehensive descriptions and references on spectra-structure correlations and additional technical details of applications for near-infrared spectra are given in references {17-18}. NIR continues to provide a valuable measurement technique, applicable to both natural and synthetic materials, for use in continuous or real-time process monitoring.


1. H. Martens and T. Naes, Multivariate Calibration, John Wiley and Sons, Chichester, 1992.

2. H. Mark and J. Workman, Statistics in Spectroscopy 2nd Edition, Elsevier, Amsterdam, 2003.

3. O.H. Wheeler, Near Infrared Spectra. A neglected field of spectral study, J. Chem. Education, 1960, 37, 234-236.

4. R.F. Goddu, and D.A. Delker, "Spectra-Structure Correlations for the Near-Infrared Region, Anal. Chem. 1960, 32, 140-141.

5. J.W. Ellis, Molecular Absorption Spectra of Liquids Below 3 m, Trans. Faraday Soc. 1928 {sic}, 25, 888-898.

6. W. Kaye, Near-infrared spectroscopy; A review. I. Spectral identification and analytical applications, Spectrochimica Acta 1954, 6, 257-287.

7. R.F. Goddu, Near-Infrared Spectrophotometry, Advan. Anal. Chem. Instr. 1960, 1, 347-424.

8. K.B. Whetsel, "Near-Infrared Spectrophotometry," Appl. Spectrosc. Reviews 1968, 2(1), 1-67.

9. E. Stark, K. Luchter, and M. Margoshes, "Near-Infrared Analysis (NIRA): A Technology for Quantitative and Qualitative Analysis," Appl. Spectrosc. Reviews 1986, 22(4), 335-399.

10. G.D. Schrieve, G.G. Melish, and A.H. Ullman, "The Herschel-Infrared--A Useful Part of the Spectrum," Appl. Spectrosc. 1991, 45, 711-714.

11. P. Williams and K. Norris, (Eds.), Near-Infrared Technology in the Agricultural and Food Industries 2nd Edition, American Association of Cereal Chemists, St. Paul, Minnesota, USA, 2001.

12. B. Osborne and T. Fearn, Practical NIR Spectroscopy with Applications in Food and Beverage Analysis 2nd Edition, John Wiley & Sons, Inc., New York, 1993.

13. G.C. Marten, J.S. Shenk, and F.E. Barton II, Near Infrared Reflectance Spectroscopy (NIRS): Analysis of Forage Quality, United States Department of Agriculture - Agricultural Research Service, Agriculture Handbook No. 643, Issued August 1985.

14. C. Roberts, J. Workman, and J. Reeves (eds.), Near-infrared Spectroscopy in Agriculture, ASA-CSSA-SSSA, Madison, WI, 2004.

15. D. Burns, and E. Ciurczak, Handbook of Near-Infrared Analysis 2nd Edition, Marcel-Dekker, Inc. New York, 2001.

16. E. Ciurczak and J. Drennen, Near-Infrared Spectroscopy in Pharmaceutical and Medical Applications, Marcel-Dekker, Inc. New York, 2002.

17. J. Workman, Handbook of Organic Compounds: NIR, IR, Raman, and UV-Vis Spectra Featuring Polymers and Surfactants, Volume 1, Academic Press, 2001.

18. L. Weyer and S.-C. Lo, "Spectra-Structure Correlations in the Near-infrared," In Handbook of Vibrational Spectroscopy, Volume 3, Wiley, U.K., 2002, pp. 1817-1837.

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Most Viewed

Copyright Information

Interested in separation science? Visit our sister site

Copyright © 2018 John Wiley & Sons, Inc. All Rights Reserved