VII. Testing Analysis Methods
The 2016/2017 Sorghum Harvest Quality Survey samples (each about 2500 grams) were sent directly from the local grain elevators to Amarillo Grain Exchange (AGE) in Amarillo, Texas. Upon arrival, the samples were dried, if needed, to a suitable moisture content to prevent any subsequent deterioration during the testing period. The samples were then split into two 1100- to 1250-gram subsamples using a Boerner divider. The divider splits the complete sample into two, while keeping the attributes of the grain sample evenly distributed between the two subsamples. One subsample was shipped to the Cereal Quality Lab (CQL) in the Department of Soil and Crop Sciences at Texas A&M University in College Station, Texas, to be analyzed for chemical composition and other physical factors, following either industry norms or well-established procedures in practice for many years. AGE graded and performed the mycotoxin testing for the other subsamples. AGE is an official grain inspection service provider in Texas as designated by U.S. Department of Agriculture (USDA) Federal Grain Inspection Service (FGIS). The grade testing procedures were in accordance with FGIS’s Grain Inspection Handbook, and the mycotoxin testing was performed using FGIS protocol and approved test kits.
A. Sorghum Grading Factors
1. Test Weight
Test weight is a measure of the quantity of grain required to fill a specific volume (Winchester bushel). Test weight is a part of the FGIS Official U.S. Standards for Sorghum grading criteria.
The test involves filling a test cup of known volume through a funnel held at a specific height above the test cup to the point where grain begins to pour over the sides of the test cup. A strike-off stick is used to level the grain in the test cup, and the grain remaining in the cup is weighed. The weight is then converted to and reported in the traditional U.S. unit, pounds per bushel (lb/bu).
2. Broken Kernels and Foreign Material (BNFM)/Foreign Material
Broken kernels and foreign material (BNFM) and foreign material are part of the FGIS Official U.S. Standards for Sorghum.
This test determines the amount of broken kernels and foreign material contained in the sample. Broken kernels is defined as all material which passes through a 5/64th-inch triangular-hole sieve and over a 2.5/64th-inch round-hole sieve. Foreign material is defined as all material, except sorghum, that remains on top of the 5/64th-inch triangular-hole sieve and all matter other than sorghum which passes over a No. 6 riddle. Foreign material is reported as a sum of the mechanically-separated foreign material as a percent of the dockage-free sample weight and the handpicked foreign material as a percent of the handpicked sample portion weight. BNFM is reported as the sum of broken kernels as a percent of the dockage-free sample weight and the foreign material.
3.Total Damage/Heat Damage
Total damage is part of the FGIS Official U.S. Standards for Sorghum grading criteria.
A representative working sample of 15 grams of BNFM-free sorghum is visually examined by a properly trained individual for content of damaged kernels. Types of damage include germ-damaged kernels, ground- and/or weather-damaged kernels, diseased kernels, frost-damaged kernels, heat-damaged kernels, insect-bored kernels, mold-damaged kernels (surface and/or internal), mold-like substance, purple-pigment-damaged kernels, and sprout-damaged kernels. Total damage is reported as the weight percentage of the working sample that is total damaged grain.
Heat damage is a subset of total damage and consists of kernels and pieces of sorghum kernels that are materially discolored and damaged by heat. Heat-damaged kernels are determined by a properly trained individual visually inspecting a 15-gram sample of BNFM-free sorghum. Heat damage, if found, is reported separately from total damage.
The moisture recorded by the elevators’ electronic moisture meters at the time of delivery is reported. Electronic moisture meters sense an electrical property of grains called the dielectric constant that varies with moisture. The dielectric constant rises as moisture content rises. Moisture is reported as a percent of total wet weight.
C. Chemical Composition
1. NIR Proximate Analysis
Proximates are the major components of the grain. For sorghum, the proximate analysis measures oil, protein, and starch concentration (or total starch) using Near-infrared transmission spectroscopy (NIR). The NIR uses unique interactions of specific wavelengths of light with each sample. It is calibrated to traditional chemistry methods, to predict the concentrations of oil, protein, and starch in the sample. This procedure is nondestructive to the sorghum.
Chemical composition tests for protein, oil, and starch were conducted using an approximately 50-gram sample in a Perten DA 7250 Near-Infrared Reflectance (NIR) instrument. The NIR was calibrated to chemical tests, and the standard error of predictions for protein, oil, and starch was about 0.3%, 0.4%, and 0.5%, respectively. Results are reported on a dry basis (percent of non-water material).
Leucoanthocyanidins (catechins) and proanthocyanidins (tannins) are a class of flavonoids known as flavanols that react with vanillin in the presence of mineral acids to produce a red color. Vanillin reacts with the flavanols, but other flavonoid compounds can give specific color development. Values near or below 4.0 mg catechin equivalents (CE) per gram (g) sample by this method generally imply absence of condensed tannins. Type III tannin sorghums usually have values greater than 8.0 mg CE/g. The test involves grinding approximately 50 g of sound seed using a UDY grinder with 1-mm sieve, and accurately weighing 0.30 g of this sample for analysis. Extraction and analysis is performed using the vanillin-HCl test with blank subtraction to remove interference by sorghum pigments. Developed color is measured using a UV-Vis spectrophotometer at 500 nm. Standard curve is run using pure catechin. Tests are run in triplicates and the average value is reported as mg CE/g sample on a dry basis.
D. Physical Factors
1. 1000-Kernel Weight, Kernel Volume and Kernel True Density
The 1000-kernel weight (TKW) is determined from the average weight of 300 individual kernel replicates using the Perten Single Kernel Characterization System (SKCS 4100). The instrument weighs each seed to the nearest 0.01 mg and automatically calculates the TKW based on the average weight of the 300 individual seeds. The averaged TKW is reported in grams.
The kernel volume for an accurately weighed 80.00 ± 0.05 g kernel sample is calculated using a helium pycnometer and is expressed in mm3/kernel. The individual kernel volume is obtained by dividing the TKW (g) by the total seed weight (g) used in the pycnometer, and multiplying the recorded pycnometer volume (cm3) by this factor. The value obtained, cm3/1000-kernels, is equivalent to mm3/kernel. Kernel volumes usually range from 12 to 28 mm3 per kernel for small and large kernels, respectively.
True density of kernel samples is calculated by dividing the mass (or weight) of the 80.00 ± 0.05 g externally sound kernels by the pycnometer volume (displacement) of the same kernels, and is reported in grams per cubic centimeter (g/cm3). True densities typically range from 1.24 to 1.39 g/cm3 at “as is” moistures of about 12 to 15%.
2. Kernel Hardness Index
Grain hardness is measured using the SKCS 4100. The SKCS 4100 automatically selects individual kernels, weighs them, and then crushes them between a toothed rotor and a progressively narrowing crescent gap. As a kernel is crushed, the force between the rotor and crescent is measured. About 50 g of clean, externally intact seed is introduced into the instrument hopper. The instrument then automatically characterizes 300 individual seeds. The data are reported as average kernel hardness index, based on the 300 individual seeds. Samples are also classified as hard, mixed, or soft, depending on average hardness index value and hardness distribution among the 300 seeds. Kernel hardness index values can range from 20 to 120.
3. Kernel Diameter
Kernel diameter is measured using the SKCS 4100. The instrument records the individual diameter of 300 seeds, and calculates the average seed diameter in mm.
E. Mycotoxin Testing
Detection of mycotoxins in sorghum is complex. The fungi producing the mycotoxins often do not grow uniformly in a field or across a geographic area. As a result, the detection of any mycotoxin in sorghum, if present, is highly dependent upon the concentration and distribution of the mycotoxin among kernels in a lot of sorghum, whether a truck load, a storage bin, or a railcar.
The objective of the testing for the 2016/2017 Sorghum Harvest Quality Survey is only to report the frequency of occurrences of the mycotoxin in the current crop, and not to report specific levels of the mycotoxin in sorghum exports. To report the frequency of occurrences of aflatoxins and DON for the harvest samples, AGE performed the mycotoxin testing using FGIS protocol and approved test kits. FGIS’s protocol requires a minimum of a 908-gram (2-pound) sample from trucks to grind for aflatoxin testing and approximately a 200-gram sample to grind for DON testing. For this study, a 1000-gram laboratory sample was subdivided from the 2.5-kg survey sample for the mycotoxin analysis. The 1-kg survey sample was ground in a GIPSA-FGIS-approved Romer Model 2A mill so that 60-75% would pass a 20-mesh screen. From this well-mixed ground sample, a 50-gram test portion was removed for each mycotoxin tested. ROSA WET-S5 and ROSA DONQ2 quantitative test kits were used for the aflatoxin and DON analysis, respectively. The aflatoxins and DON were both extracted with water (5:1), with an added extraction powder used for aflatoxin. The extracts were tested using the ROSA lateral flow strips, and the mycotoxins were quantified by the Charm EZ-M system.
The ROSA WET-S5 quantitative test kits report specific concentration levels of the mycotoxin if the concentration level exceeds a specific level called a “Limit of Detection” (LOD). The LOD is defined as the lowest concentration level that can be measured with an analytical method that is statistically different from measuring an analytical blank (absence of a mycotoxin). The LOD will vary among different analytical methods developed for different types of mycotoxins and commodity combinations. The LODs for the ROSA WET-S5 and ROSA DONQ2 are 2.0 parts per billion (ppb) aflatoxins for diluted extract, and 0.1 parts per million (ppm) DON for diluted extract.
A letter of performance has been issued by FGIS for the quantification of aflatoxins and DON using the ROSA WET-S5 and ROSA DONQ2 kits, respectively.