Location: Quality & Safety Assessment Research2012 Annual Report
1a. Objectives (from AD-416):
1. Develop cost-effective systems and sensors for rapid and nondestructive measurement of moisture content and density of cereal grains, oilseeds and nuts that can be used in static and dynamic situations (on line and in process applications). 2. Develop a portable system for moisture measurements of shelled and unshelled peanuts that can be used in the field and at peanut grading stations. 3. Improve peanut grading processes by developing a rapid system for nondestructive determination of peanut kernel moisture from measurements on pre-cleaned samples and samples consisting of a mixture of pods and foreign materials. 4. Develop methods for monitoring water migration in almonds and other nuts and its effect on their quality by dielectric and spectroscopic methods.
1b. Approach (from AD-416):
Developing a cost-effective microwave system for rapid moisture measurement will require analysis of existing dielectric properties data to determine the optimum frequency and measurement parameters. From these results, microwave components will be specified, assembled into a low-cost rapid prototype, compared to a network analyzer, then calibrated and validated for moisture content of wheat, corn, soybeans, almonds, oats, sorghum, and barley. For peanuts, dielectric properties from 2 to 18 GHz on Runner, Spanish, Valencia, and Virginia will first be collected on shelled and unshelled peanuts. As with the other grains above, these data will then be analyzed to determine the optimum dielectric properties parameters and associated algorithms for a density- and variety-independent measure of peanut moisture content, and a prototype low-cost rapid peanut moisture measurement system will be developed. From these measurements and system development for peanut moisture content of shelled and unshelled peanuts, further development of a system to measure the moisture content of peanut kernels without shelling will be developed. Additional dielectric properties of both the unshelled and shelled kernels will be collected as needed, along with the moisture content of both the shells and the kernels. From these data, moisture algorithms will be developed. Additional dielectric properties measurements on unshelled peanuts with and without foreign material will also be collected for development of moisture models that are also independent of foreign material (trash). Models will be developed for both pod and kernel moisture contents. Besides kernel moisture content, meat content and percentage of foreign material are also important grading parameters. Attempts to correlate the dielectric properties data with these parameters will also be made. For almonds, since no dielectric properties data exists, fundamental dielectric properties measurements of almonds of varying varieties, growing locations in California, and moisture contents will be collected with a network analyzer. Correlations between moisture content and dielectric properties data will be developed. To investigate the dynamics of water migration in the almond kernels, known amounts of water will be sprayed on almond kernels of known moisture content, mixed, and then sealed in a Styrofoam box. The sealed box will then be placed between two horn-lens antennas for free-space measurement of the dielectric properties between 2 and 18 GHz at room temperature. Changes in the dielectric properties will be recorded over time as water moves from the almond surface to water in equilibrium inside the kernels. The next stage will be the use of these dielectric spectroscopic methods to monitor water migration inside the almonds under controlled conditions of humidity and temperature. In this instance, the water will permeate the almonds from the atmosphere for varying relative humidities from 20 to 80% in a controlled environmental chamber. Additionally, measurements will be repeated over varying temperatures from 0 to 50 oC in 5 oC increments.
3. Progress Report:
Dielectric properties data for grains, seeds, and nuts obtained with a free-space system and a vector network analyzer between 2 GHz and 18 GHz for samples of different moisture content, temperature and bulk density were compiled and investigated both graphically and statistically. In particular, different algorithms for simultaneous and independent determination of bulk density and moisture content from measured microwave parameters including attenuation, phase shift, the dielectric constant, and the dielectric loss factor were tested and compared. For bulk density, performance of the different algorithms were equivalent. However, for moisture content, density-independent calibration functions performed better than the algorithms involving attenuation, phase shift, and dielectric properties. For peanuts, dielectric properties measurements were performed on cleaned and uncleaned samples. Different density-independent calibration algorithms were used for peanut kernel moisture determination from measurements on cleaned and uncleaned pods. The results from uncleaned pods look promising and are as good as those obtained from cleaned pods. If confirmed through more measurements on samples including different foreign materials, this will allow peanut growers and buying points inspectors to determine peanut kernel moisture content without having to either shell or clean the samples. This will result in significant cost and time reduction associated with the drying and grading of peanuts. Dielectric properties of minced onions were performed with an open-ended coaxial line between 200 MHz and 20 GHz for moisture ranging from 8.1% to 90.2% and temperature ranging from 0 deg. Celsius to 40 deg. Celsius. A density-independent calibration function was used for moisture prediction from measured dielectric properties. The best results were obtained for a frequency of 13.35 GHz with a standard error of performance of less than 1% moisture content. Dielectric measurements on whole muscle and minced chicken breast meat were carried out at radio frequencies between 200 MHz and 20 GHz and temperature ranging from -10 deg. Celsius to + 25 deg. Celsius. No significant difference in the measured dielectric properties was observed between the whole muscle and minced meat. However, influence of water binding status inside the meat on the dielectric properties was obvious at zero degrees Celsius similar to that which occurs in phase transition from ice to liquid water at microwave frequencies. Also, the effect of ionic conduction was most significant at lower frequencies between 200 MHz and 1.5 GHz.
Paz, A., Trabelsi, S., Nelson, S.O., Thorin, E. 2011. Measurement of the dielectric properties of sawdust between 0.5 and 15 GHz. IEEE Transactions on Instrumentation and Measurement. 60(10):3384:3390.