Title: NIR spectroscopy as a tool for optimizing sorting of white corn kernels contaminated with mycotoxins Authors
Submitted to: Corn Dry Milling Conference Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: May 31, 2007
Publication Date: May 31, 2007
Citation: Wicklow, D.T., Pearson, T.C., Brabec, D.L. 2007. NIR spectroscopy as a tool for optimizing sorting of white corn kernels contaminated with mycotoxins [abstract]. Corn Dry Milling Conference Proceedings. Technical Abstract: Near infrared reflectance spectra (500-1700 nm) were analyzed to determine if they could be used to identify single whole white corn kernels contaminated with aflatoxin and fumonisin. Kernels used for the study were obtained from grain lots harvested in 2006 from commercial fields in southern Texas. Kernels were visually examined and grouped into four symptom categories: asymptomatic, showing 25% to 50% discoloration, showing over 75% discoloration, and discolored BGYF kernels. Friable kernels and fragments were not included in this study as they are usually removed by existing cleaning equipment at grain elevators. Spectra were acquired on both the germ side and endosperm side of each kernel. After spectra acquisition, kernels were weighed individually, then placed in groups of five according to their classification based upon symptoms of fungal infection and numerical sequence within each pill box. Aflatoxin B1 or total fumonisins (B1, B2, and B3) were measured with a fluorometer after extracts were purified with immunoaffinity columns (Aflatest or Fumonitest, Vicam, Watertown, MA). The aflatoxin or fumonisin level of each five-kernel group then was assigned to each individual kernel from that group. Kernels were analyzed in groups instead of individually to reduce cost and analysis time. For high speed sorting operations, whole spectra cannot be acquired at throughput rates that are economically feasible. Most commercial sorting machines are able to only measure one spectral band of light, while some machines can measure two bands. Discriminate analysis was used to select the optimal pair of wavelengths to identify kernels containing mycotoxins. It was found that using the wavelength pair of 500nm and 1200nm, approximately 87% and 93% of kernels having high levels of aflatoxin (>100ppb) and high levels of fumonisin (>40ppm), respectively, were correctly classified. Additionally, approximately 96% of the kernels having low levels of aflatoxin (<10ppb) and fumonisn (<2 ppm) were correctly classified as uncontaminated. Kernels having minor symptoms (25% to 50% discolorations) had lower classification accuracies (80%), than those with discolorations covering more than 75% of the kernel (88%), or discolored BGYF kernels (91%). A commercial sorting machine (Satake DE) was set up to sort corn using light at 500 and 1200nm and reject 4% to 9% of the incoming corn. Ten kg samples of corn were used in all sorting experiments. Incoming mycotoxin levels in three lots of corn used in these sorting experiments ranged from 23 to 150 ppb aflatoxin, and 0.4 to 0.6 ppm fumonisin. In one pass through the sorter, aflatoxin was reduced 19-75% (average 45%), and fumonisin was reduced an average of 60%. Two passes through the sorter reduced aflatoxin by 88%, and below 20 ppb in each of the three lots of corn. A limited number of samples run through a monochromatic sorter that has higher spatial resolution using the 515 nm filter reduced aflatoxin by 61% in one pass and 82% in two passes. It was found that some kernels of white corn with minor symptoms of discoloration could have very high contamination levels of aflatoxin without exhibiting external BGYF. Thus, two passes through the sorter were required to reduce aflatoxin below the regulatory limit of 20 ppb. In an earlier sorting experiment with commercially harvested yellow corn that was contaminated with equivalent levels of aflatoxin (average 75ppb), we were able to reduce aflatoxin below 20 ppb with a single pass through the same Satake DE machine (Pearson et al., 2004). Differences in pericarp thickness, kernel vitreosity (hardness), and the presence or absence of carotenoids, could influence both kernel symptom expression and the ability of NIR light to detect fungal damaged endosperm.