Author
 |
Reeves Iii, James |
|
Submitted to: Spectroscopy Conference Proceedings
Publication Type: Abstract Only Publication Acceptance Date: 8/31/1997 Publication Date: N/A Citation: N/A Interpretive Summary: Technical Abstract: The objective of this research was to determine if mid-infrared diffuse reflectance spectroscopy could be used to determine the composition of dried, ground forages. The results obtained from a variety of sample sets including tall fescue hays collected over a three year period and treated forages and by-products will be discussed. Ground samples were scanned using diffuse reflectance on Fourier transform spectrometer equipped with a diffuse reflectance attachment. Samples were scanned under a variety of conditions with variations in the number of co-added scans per sample, scanning resolution, with or without KBr dilution, and using a static sample holder or a custom made sample transport device. All data sets were analyzed using PLS with data pretreatments consisting of a first or second derivative and "Multiplicative Scatter Correction" or "Mean and Variance Scaling" using a one-out cross validation analysis. Results have shown the following: 1. Mid-infrared spectra can be used to determine the composition of dried forages with an accuracy equal to or greater than that obtained using near-infrared spectra. 2. Sample dilution with KBr is not needed. 3. Increasing the amount of sample scanned by the use of a sample transport device significantly increased calibration accuracy. 4. Increased spectral resolution often resulted in increased calibration accuracy. 5.Calibration robustness may not be as high with mid-infrared spectra as with near-infrared spectra. And, 6. Spectral interpretation is much easier in the mid-infrared than in the near-infrared. These results demonstrate that mid-infrared spectra of neat samples of forages and by-products can be used to accurately determine sample composition. |
