| Author | Message | ||
     Jan Ramaekers (johnson) New member Username: johnson Post Number: 2 Registered: 1-2009 |
Thank you Ralf for the interesting second point. To answer your questions: The matrix of our products is from corn or wheat origin. Parameters are rather classic: moisture, protein, ash, fat content, starch, ... Today we use classical NIR, which performs perfect for our products, but these are quite sensible to wear especially in production environment. Therefore we want to switch to more robust FT-NIR or Diode Array(because of the fixed set-up). But we're trying also to evaluate the benefits of both systems in a future online installation. Regards, Jan | ||
| Elvira Fernandez de Ahumada (elvira_fernandez) New member Username: elvira_fernandez Post Number: 1 Registered: 5-2006 |
Hi Jan, As part of my PhD, we have evaluated diode array instruments to be implemented on the manufacturing process of animal feeds. We measured crude protein, crude fiber and also the percentage of ingredients included. You may take a look of these papers and see if they are useful for you: - Feasibility of Diode-Array Instruments to Carry Near-Infrared Spectroscopy from Laboratory to Feed Process Control. E. Fern�ndez-Ahumada, A. Garrido-Varo and J.E. Guerrero-Ginel. Journal of Agricultural and Food Chemistry 56 (2008) 3185-3192. - Taking NIR Calibrations of Feed Compounds from the Laboratory to the Process: Calibration Transfer between Predispersive and Postdispersive Instruments. E. Ferna�ndez-Ahumada, A. Garrido-Varo, J.E. Guerrero, D.Pe�rez-Mari�n, T. Fearn. Journal of Agricultural and Food Chemistry 56 (2008) 10135-10141. If you have no access to the journal, write me to the e-mail: [email protected] Hope this helps. Kind regards, Elvira | ||
| Ralf Marbach (ralf) New member Username: ralf Post Number: 5 Registered: 9-2007 |
Hi Jan, Various semiconductor materials are used in diode arrays, and choice is a matter of various trade-offs. Your question is straightforward but can only be answered with more info. You mention the analyte you are after, but you don't mention "against" whcih other ingredients you need to measure it (specificity). You also don't mention integration time, focus spot area, etc(sensitivity, representative sampling). Everything is this business is application dependent ... With regard to your 2nd question, fiber vs. direct illumination, in an on-line situtation. Fibers absorb at longer than 2.3 micron. At shorter lambdas, they don't reduce the INTENSITY intensity of the light, they merely restrict the ETENDUE (throughput). A single fiber gives typically (0.6 mm * 0.22 steradian)^2 of optical thoughput. Direct illimunation can typically give, say, (3 mm * 0.4 sr)^2, so roughly 100x more illumination light. Because scattering breaks the law of conservation of etendue, you will see 100x more INTENSE light on your detector, so the advantage is direct. Generally speaking, the summary may be like this: Use fibers if you can, they are convenient and off the shelf. If the fiber probe can be in contact with the sample, then generally fine. In an ON-LINE situation, if the situation asks for anything special, e.g., high speed, very large sampling spots, shaped sampling spot, stand-off distance to the sample stream, passline (height) variations of the sample stream, ..., then a custom interface is usually worth it. Hope this helps. If you like to discuss further, I suggest we talk on the phone. Writing is too cumbersome for the details. Regards, Ralf [email protected] +358 40 770 8375 | ||
| Jan Ramaekers (johnson) New member Username: johnson Post Number: 1 Registered: 1-2009 |
Standard diode array detectors used in NIR equipment today don't detect all the spectral information of the first overtone and nothing of the combinations. Can this be an issue to analyse at quantitative level feed product, especially for the detection of protein content? Or is this loss completely compensated by the higher penetration of the NIR radiation into the sample itself in the case of diode array NIR? A second question, looking at online measurements two options seem to appear. 1) Fiber optics and 2) Diode array NIR Does somebody has experience with fiber optics? Is the low intensity of the light throughput causing problems? Can somebody give an opinion on the use of both technologies for online measurements? Thank you for your opinion and sharing your experience. |