Master of Science / Food Science Institute / David L. Wetzel / Grain processing and handling requires quality determinations to ensure wholesome products that meet or surpass legal standards and specifications required by the end consumer. Near infrared spectroscopy has proven to be a useful and versatile tool to enable grain processers to make adjustments as needed. Near infrared chemical imaging also provides spatial information within the image and relative composition of chemically distinct components within the product. The potential use of chemical imaging to determine extraneous material in bread baking quality flour was addressed. A specimen preparation technique was developed. Insect fragment spiked specimens were imaged to determine their imaging effectiveness for application near the allowable limit of insect fragment concentration. Imaging was achieved using indium antimonide array detection of diffusely reflected radiation. The detector array of 81,920 pixels collected radiation from an area of 30.72 mm by 38.4 mm with a pixel size of 120 µm². Spectra were collected simultaneously from each pixel without moving parts by scanning with a liquid crystal tunable filter. Partial least squares analysis of each pixel within the sample allowed a summation of the insect quantity. The chemical structural distinction of chitin in the high protein matrix of the insect residue was in contrast to the non-digested carbohydrate residue in the lesser protein matrix of the flour. The method developed provided a linear response for a concentration range from approximately half the allowable limit to twice the limit for two insects that commonly contaminate flour. For the two insects studied the slopes are comparable with a slight off-set over a practical working range, so that insect concentration can be determined independently of species recognition.
| Identifer | oai:union.ndltd.org:KSU/oai:krex.k-state.edu:2097/38803 |
| Date | January 1900 |
| Creators | Nickoley, Tyler R. |
| Source Sets | K-State Research Exchange |
| Language | en_US |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0015 seconds