Tobacco moisture measurement with Near IR technology is long established throughout the world as accurate and reliable. Tobacco nicotine measurement on-line is of questionable value for a number of reasons.
GC, GC-MS, HPLC and HPLC-MS can all be used for tobacco nicotine determination in the laboratory. Near IR full spectrum analyzers utilizing PLS or other suitable chemometric treatments can also be employed in the laboratory on ground samples. The simplest and least expensive technique is GC. The levels of detection in a laboratory are in the ppm range rather than the % range that may be obtained using filter photometer NIR equipment on-line.
The difficulty with on-line nicotine measurement is that nicotine is not distributed evenly throughout the plant. The stem and the leaf and even within a single leaf contain significant variation in nicotine content. So, if you have more stem or less stem in the sample that alone shifts the nicotine measurement. Moisture equilibrates throughout the tobacco product, nicotine does not.
To indicate the accuracy of an on-line trend measurement of nicotine we need to determine the sample collection reproducibility and the sample analysis accuracy. The best way to do this is with 3 sets (or more) of double blind samples analyzed by a certified laboratory for nicotine content.
This means that two separate grab samples from the measurement point need to be collected at three different times. Hopefully the nicotine level is at a different level for each collection time. Otherwise it will be necessary to collect additional samples to create a range of nicotine levels.
When the samples are sent to the lab, the lab does not know which samples were collected at the same time, they are “blind” to this information to avoid bias in the analysis. The resulting data will determine how reliable the sample collection method is and the homogeneity of the samples collected.
Nicotine and related alkaloids/metabolites can be measured in the Near IR with chemometric treatments. The wavelength selection is the not the limiting factor in the measurement, it is the nicotine stratification throughout the plant. We believe that sample variation due to nicotine stratification makes a claim of absolute measurement on-line with photometers questionable at best.
Two papers of interest on nicotine variation within the tobacco plant can be viewed via hyperlinks below.