| Author | Message | ||
     alunevans |
Hi I am new to this subject and have just been dropped in at the deep end in work. Part of my job is using NIR for gasoline blending in the petro refining industry and I have only the basic degree level understanding of the NIR technique. I have only a couple of months to look at the feasibility of using NIR to detect sulphur at concentrations of 20ppm and less. Any one else done this before? I have been trying for a couple of weeks now and it doesnt seem to work. Also any assistance in helping me look for literature on the background of NIR specifically used in the petro chemical industry would be appreciated. | ||
| Bruce H. Campbell (Campclan) |
You may be trying to do the impossible. Most detection limits are about 100 to 500 ppm (0.01 to 0.05%) and those are for hydrogen bonded to carbon, nitrogen, or oxygen (although for water the detection limit is much lower). Typicallly, sulfur (or sulphur) has even higher detection limits. There may be a route that would involve some chemistry combined with NIR. Many metal oxides have very intense, sharp bands. If the sample was reacted with a metal that would form bonds with the sulfur, there may be useful bands of the compounds so formed. Bruce | ||
| alun evans |
ok thanks bruce I will take that on board and see if we can devise a technique with using the metal oxide route. | ||
| Gabi Levin |
Hi Alunevans, Sulfur - the 64,000 dollar question - Here is why it will not work - 1. The sulfur carbon bonds don't provide usefull absorption 2. The sulfur -H bonds have mcuh lower absorption coefficients , thus need high concentration to be detected. Like Bruce, I suspect that even 500ppm will be questionable. 3. Even if this was not enough to prevent the measurement, there isn't just one sulfur compound. The reference method will give you total sulfur. However, same amount of sulfur may be associated with differing spectra because of variations in the mixture of the sulfur compounds. I hope the advice from Bruce will prove usefull, although I am not familiar with the behaviour of metal sulfides. Only one thought - if you do conventioal chemistry, that means you work off line. If you do, you might as well do sulfur by acceptable ASTM methods and then you don't have to worry about acceptability of the results by the petroleum community. Gabi Levin | ||
| alun evans |
Thanks for your reply Gabi. As you can probably tell I have very limited knowledge of this area. My background is mostly in hard testing in the refining industry but have recently been asked to take over responsibility for the NIR in our laboratory. Is there a definitive work for NIR in the petro chemical industry? We have next to no literature on the subject in our dept and I am looking for something I can refer to when developing and improving our models for gasoline components and charecteristics such as RON, MON, vapour pressure,density and distillation. | ||
| Bruce H. Campbell (Campclan) |
Alun, Send me an email to me at{[email protected]}. There are some aspects I believe I could help you determine without going through the list server. Bruce | ||
| Michel Coene (Michel) |
You can reliably detect over 20 parameters in gasoline blending with a single NIR spectrum, but as mentioned above, sulphur is not one of them. If you start looking around, you will find some companies who sell on-line equipment. Some of these have cooperated with large oil companies and have therefore built up a large database of samples, but if your incoming oil originates from a limited number of wells, then it is certainly feasible to build up a calibration yourself. For the sulphur, there are some other analysers which can reach these levels. X ray fluorescence is worth checking out, though I can't right now remember detection limits. Otherwise there are specialised analysers using ovens and other smelly things. |