Near-infrared (NIR) spectroscopy is a technique that measures the amount of heat absorbed or emitted by certain materials. It is used in a variety of applications, but in the agricultural world, it is often used to determine the quality and composition of mixed materials such as stock forage. It uses electromagnetic radiation in the 800 to 2500 nm range, which is just beyond the limits of our vision.
The patterns of heat absorption is determined by the functional groups present in the chemicals that make up the sample, and a degree of structural information can be obtained from the NIR spectrum of a sample. A better region to determine the structure of chemicals is in the mid-infrared region of 2500 nm to 25 000 nm. Organic chemists routinely use this region to identify the functional groups present in unknown chemicals, and there are a number of readily available charts (including Wikipedia’s effort, a 24 page practical guide and a cartoon version) that help users interpret this part of the infrared spectrum. As the NIR region is (quite rightly) not used for this degree of precision, very few charts exist. There are a couple of big, expensive books which will tell you everything you need to know (like the recently published Practical Guide and Spectral Atlas for Interpretive Near-Infrared Spectroscopy by Workman and Weyer), but in terms of getting a general idea of what functional groups absorb what parts of the spectrum, it’s a bit of a struggle to find anything.
This week I’ve been helping out a colleague do a bunch of plotting in R. It’s been good fun, and I’ve learned a number of tricks to modify the appearance of R plots. I was able to use this aspect of my week to rectify the problem identified above. I was able to track down a copy of Goddu and Delker’s 1960 near-infrared spectra-structure correlation chart, and use the information contained therein to create a dataset which I then plotted in R, to give the image shown above (a larger, better quality image is available here).
The chart shows the regions where peaks may be expected. The colours correspond to the molar absorptivity of the group at that wavelength. Darker colours correspond to a greater molar absorptivity.
The data and R commands used to create the image is available on GitHub. Caution is of course required when using this chart. The data is over 50 years old, and much research has been done since then. However, for those of us without a copy of Workman and Weyer 2012, it’s a guide to what might be going on in the region.
The International Code of Zoological Nomenclature
Edgardo Donato—Colecciòn 78 rpm
Kiwi scientists taking drastic action to fund research—primarily focused on health research, but symptomatic of NZ science funding in general.