| Summary: | Soybean quality is affected by temperature (T), moisture content (w) and split beans content (x_s), elevated levels of which can decrease safe storage time (t_s) because of accelerated grain deterioration. Depending on the final use of soybean, t_s can be based on dry matter loss (DML), germination, or mold growth. DML can be estimated by measuring the amount of respired carbon dioxide (CO2) during grain storage, and so has become the basis of previous recommendations for t_s of grains, oilseeds, and feedstocks. The main objective of this thesis research was to measure dry matter loss rates (v_DML) of soybeans in a series of respiration tests using two different grain respiration measurement systems (GRMS) and to understand the effects of GRMS itself, x_s, and w at elevated T. Several researchers have attempted to determine v_DML or time to reach a particular DML threshold (t_DML) in grain respiration studies. Two measurement approaches have been used to measure respired CO2: static and dynamic systems. In a static grain respiration measurement system (S-GRMS), grain is placed in a sealed chamber wherein a limited amount of oxygen gas is available for respiration. The respired CO2 accumulates in the sealed chamber and is measured over time. In a dynamic grain respiration measurement system (D-GRMS), air passes continuously through a bed of grain so the oxygen supply for respiration is maintained. The constant airflow carries the respired CO2 into a measurement system. The availability of oxygen for respiration or GRMS used in the study is expected to affect v_DML and t_DML estimates, but the effect by GRMS used has not been quantified before. The specific objectives of the research were to: (1) determine the effect of GRMS on v_DML estimates for 18% moisture content (m.c.) soybeans stored at 30°C; (2) develop a damage multiplier (M_D) for soybeans with an x_s range of 4 to 16% (w/w) from a baseline of 0% (w/w) stored with 18% m.c. and 35°C using S-GRMS; and (3) estimate v_DML of 14, 18, and 22% m.c. soybeans stored at 30°C using D-GRMS. The w and T parameters in these tests were chosen based on typical soybean harvest and initial storage conditions in Mato Grosso, Brazil, which has produced 32 million tons of soybeans per year, in recent years. Since soybeans are an important source of plant-based proteins and oils, a preliminary test to correlate DML to changes in chemical composition and lipid oxidation byproducts was also conducted. Secondary byproducts of lipid oxidation were measured via change in peroxide value (∆PV) and 2-Thiobarbituric Acid (∆TBA) value of soybean samples from before and after respiration tests. v_DML values from the third objective were also used to estimate time to reach 0.5% DML (t_0.5), the threshold used for maximum allowable storage time (MAST) guidelines for shelled corn by the ASABE Standard D535 and “approximate” MAST guidelines for soybeans in many university extension publications. Results showed that v_DML estimates for 18% m.c. soybeans at 30°C when measured using D-GRMS were 1.20 times higher than those from S-GRMS. While the difference between GRMS at this single set of storage conditions was found to be non-significant (p = 0.09), respiration and v_DML reported in the literature for grains stored in S- and D-GRMS vary greatly. Estimates with S-GRMS system tend to be lower than for D-GRMS. Thus, care should be taken when using v_DML rates from literature to estimate t_s. Damage by splits was expected to have a greater effect on v_DML. Soybeans are inherently prone to cracking, splitting, lipid oxidation, and protein degradation – all of which lead to accelerated dry matter and quality losses. Results showed that v_DML increased with increasing x_s and that average v_DML increased by 1.10 to 1.70 times greater than the base case (0% splits) when x_s increased from 4 to 16%. M_D was defined as the v_DML of 0% splits soybeans relative to v_DML with x_s % splits, and it decreased from 1.0 to 0.60 as x_s increased from 0 to 16% splits. M_D for soybean was found to be 1.25 times as sensitive to x_s when compared to the M_D for corn, which decreases from 2.08 to 1.42 as damaged kernels content increased from 0 to 16% (w/w). These results and the procedure developed for quantifying a soybean M_D are useful in the future as v_DML data for a wider range of w and storage conditions become available. v_DML was found to increase significantly with w at 30°C – approximately by a factor of four as w increased from 14 to 18% m.c. and by a factor of 14 when w increased from 14 to 22% m.c. Using v_DML of 14 to 22% m.c. clean soybeans at 30°C, estimates of t_0.5 were calculated and found to be four to five times longer than recommended MAST for corn and soybeans at the same T, w, and water activity (a_w). The discrepancy between estimated t_0.5 and recommended MAST values can be attributed to the fact that the soybeans used in this study were clean, intact soybeans, while the MAST values were for corn with a typical damage content of 30%, and for soybeans at the same a_w of corn at this damage level. Results from proximate analyses showed that DML had a -0.40 correlation coefficient with a decrease in carbohydrates change content (∆C, p = 0.18). Results from testing for lipid oxidation byproducts suggested that the first and second stages of oxidation occurred during the respiration tests, but a correlation coefficient of 0.35 was found between DML and ∆PV (p = 0.65) and of -0.38 between DML and ∆TBA (p = 0.62). However, to gain a better understanding of lipid oxidation byproducts and their rate of increase with v_DML, samples must be tested throughout a respiration test – as was done with DML measurements – for which the current D-GRMS was not designed. Nevertheless, these preliminary results support the idea that DML measurements may provide an indirect measure of lipid quality of soybeans during storage. The research reported in this thesis provide detailed development and testing of robust GRMS, test protocols, and data analyses for the development of MAST guidelines, which are sorely needed to mitigate postharvest losses of soybeans during storage, especially in expanding soybean production in sub-tropical regions of the world. The methodologies and analyses presented here are directly applicable to other crop systems, such as wheat, rice, and pulses – all of which are important sources of protein and nourishment to an ever-growing global population.
|
|---|