Near-Infrared peaks are not easily assignable as is the case for the mid-infrared. The Near-IR region ranges approximately from 12,800 cm-1 to 4000 cm-1, or 780-2500 nm and contains overtones and combinations of fundamental vibrations that lie in the mid-IR range (4000 cm-1 to 500 cm-1). Overtones are essentially higher harmonics of fundamental transitions, while combination bands involve vibrational transition of two different bonds/chromophores in the same molecule. The intensity of the overtone bands generally decreases by one to two orders of magnitude upon increasing the number of the harmonic (i.e., n --> n+1 harmonic). The bandwidth, on the other hand, increases with each consecutive overtone as a result of the increased anharmonicity. Thus, only vibrations that involve bonds that have strong absorptions in the fundamental mid-IR range, such as C-H, N-H, and O-H, can be readily seen in the Near-IR region. Bending, scissoring, and rotating motions are only seen as third or fourth overtones and are mostly lost in the baseline. Because of the increased density and bandwidth of transitions in the Near-IR region few pure peaks exist and little structural detail can be determined visually. However, these limitations are only of consequence in trace analysis and extremely small sample volumes. The ultimate advantage is that, unlike mid-IR spectroscopy, there is no sample preparation necessary because of the decreased intensities. In addition, information about other chemical parameters that may be lost during sample preparation (polymorphism, particle size, etc.) can often be extracted from the Near-IR spectrum.

 

The advent of relatively fast inexpensive microcomputers allowed the potential of the Near-IR to be tapped. The computers allow thousands to hundreds of thousand calculations to be performed in a reasonable amount of time. This is crucial to the success of Near-IR spectroscopy as many times the correlations are not apparent to the naked eye, nor are they attributable directly to first order principals. Successful Near-IR analysis many times require the testing of thousands of equations to determine which has the best fit.

Return