Submitted to: Applied Engineering in Agriculture
Publication Type: Peer reviewed journal
Publication Acceptance Date: 7/1/2007
Publication Date: 11/1/2007
Publication URL: naldc.nal.usda.gov/download/12871/PDF
Citation: Armstrong, P.R., Lingenfelser, J., Mckinney, L. 2007. The Effect of Moisture Content on Determining Corn Hardness from Grinding Time and Grinding Energy, and Hardness Prediction Using Near-Infrared Spectroscopy. Applied Engineering in Agriculture. Vol. 23(6):793-799. Interpretive Summary: Corn hardness is important in both the food and industrial processing of corn. Food grade corn generally requires kernels with a large portion of hard endosperm while many industrial processes can desire a softer corn. The Stenvert hardness tester is one method of measuring corn hardness and is used by measuring the grinding time of a 20 gram sample. It is agreed that samples should be equilibrated to a common moisture level to obtain good results. This is not always practical and usually takes a few days to do. This research developed methods to correct for different sample moisture levels and also compared grind-times with the energy to grind the sample. Generally, grinding energy was the preferable method as correction for moisture content was better. Near-infrared reflectance spectroscopy was also used to predict grinding energy and time but results were poor. Based on this work, better measurements of corn hardness can be made using the correction methods developed.
Technical Abstract: The Stenvert hardness test was used to determine the energy-to-grind (ETG) and time-to-grind (TTG) of 107 food-grade corn hybrids at different moisture content (MC) levels. ETG and TTG were significantly affected by moisture content. Across hybrids, ETG displayed the most consistent response to MC between 10% and 13% wet basis MC (MCwb). An equation was developed to adjust ETG and TTG to a common MC level in order to minimize moisture effects on corn hardness determination. ETG was considered to be the preferable method to measure corn hardness considering MC adjustments were more accurate using the developed equation. Results also support grinding at MC levels between 10% and 13% MCwb to obtain the most accurate results as opposed to higher MC levels. Grinder speed effects were also found to be significant but controllable and the repeatability of ETG and TTG were about the same. Near-infrared reflectance (NIR) spectroscopy was concurrently evaluated as a method to measure corn hardness in terms of ETG and TTG on whole kernel and the ground material from the grinder. Predictive models, using PLS regression, were poor using spectra (500-1700nm) of whole kernel and ground samples. The moisture adjustment methods developed in this work allow samples of corn to be tested over a broader range of MC. This provides more convenience and greater confidence in grinding parameters as a measurement of corn hardness.