verticity
I'm interested in things
Right, we are actually talking about a completely different technique from what is described in the paper. I think it might work. Using the modeling and fitting techniques I outlined above I believe it might be possible to measure individual kavalactones.The papers show how to closely duplicate HPLC results using NIRS. HPLC physically separates the kavalactones so that each can be measured individually by UV spectroscopy. This takes time (about 13 minutes with the method and column I use) and requires sample preparation. NIRS measures the absorptivity (~400-2500nm) of the dry powder (no prep necessary) and gives results in seconds. The advantage is that a large amount of samples can be tested very quickly. The disadvantage is that the NIRS apparatus capable of doing this is costs well in excess of $50,000, where a basic HPLC system can be assembled for around $3000. Also note these papers only describe measurement of kavalactones; to validate the same process for FK's would take much more research.
If I'm understanding you correctly, you are proposing the use of UV spectroscopy (~200-400nm) alone. While this would probably show some different characteristics of different cultivars it can't measure the individual quantities of specific compounds, because they don't react to UV the same as they do to NIR. To quote wiki, "Near-infrared spectroscopy is based on molecular overtone and combination vibrations." Since these vibrations don't occur at UV wavelengths, the approach you're proposing won't work.
Another big problem with NIR reflectance spectroscopy is that since it uses dry powdered samples, the fineness of the grind and texture of the powder could actually effect the spectra. If I had $50K+ to blow on this, I would get a GC-MS instead.
Last edited: