Principles for determining moisture content and mass of individual peanut kernels by microwave resonator measurements are discussed, and data are presented showing the application of these principles. By measuring the shift in the resonant frequency and the change in the cavity transmission characteristics when a peanut is inserted into the cavity, it is possible to obtain the moisture content of the kernel independent of its mass and apparently independent of peanut type, as far as Runner and Virginia market types are concerned. It is also possible to determine the kernel mass independent of kernel shape simultaneously with the moisture determination. Moisture contents in the range from 4 to 14%, wet basis, were determined with an uncertainty of 0.9% moisture, and the uncertainty in the mass determination was less than 0.05 g. Keywords, Moisture content. Microwaves. Moisture content is an important characteristic of peanuts, Arachis hypogaea L., during harvesting, storage, marketing and processing. In the Southeast, peanuts are allowed to dry on the vines after being dug at harvest to an average moisture content of 18 to 22%, wet basis, before being combined. After combining, the drying process must start immediately to reduce the moisture content to less than 10.5% for marketing. They can be stored safely at this moisture level if adequate ventilation is provided. Mass of individual kernels is also an important characteristic, because it is correlated with size, and larger kernels are usually more mature and of better quality. Conventional electronic instruments for moisture content determination in grain, seeds and nuts provide a moisture content reading which is an average for the sample. This average does not provide any information about the range of individual kernel or seed moisture contents that may be present in the sample. The variation of kernel moisture contents would be valuable information at many points in the storage, handling, and processing of peanuts, such as drying and roasting. Thus, an instrument that could determine the moisture content and mass of individual kernels within samples would be useful for research and application in the peanut industry. A moisture profile meter for peanut samples was developed which consisted of a commercial crushing-roller moisture meter equipped with a millivolt recorder and a peanut singulation device to feed individual peanuts to the crushing rollers (Hutchison and Holaday, 1978). Recorded signal peaks, which resulted from the DC conductance of the kernels as they passed between the rollers, were correlated with moisture content and provided moisture Article was submitted for publication in May 1992; reviewed and approved for publication by the Food and Process Engineering Inst, of ASAE in December 1992. The authors are Andrzej W. Kraszewski, Research Electronics Engineer, and Stuart O. Nelson, Agricultural Engineer, USDAAgricultural Research Service, Athens, GA. content estimates for the individual peanuts and information on the range of kernel moisture content in a 50-kernel sample. Recently, some success has been achieved in detecting single-kernel moisture content by impedance measurements on individual peanut kernels witih capacitive sensors at radio frequencies (Kandala and Nelson, 1990; Nelson et al., 1990). Measurements of capacitance and conductance of a small parallel-plate capacitor with the peanut kernel between and in contact with the plates, performed at two frequencies (1 MHz and 4.5 MHz), allowed prediction of the kernel moisture content within ±1% of the values determined by a standard oven drying method. Resonant cavity techniques are widely used in measuring the microwave properties of materials by measuring the shift in the resonant frequency and the change in the Q factor of the cavity when the object is inserted into the cavity (Altschuler, 1963). This technique has proven to be an interesting alternative to existing methods for kernel moisture determination in soybeans and com (Kraszewski et al., 1989; Kraszewski and Nelson, 1992), because it offers nondestructive and relatively fast measurements. The purpose of this research was to determine the effects of peanut kernels of various shapes and dimensions and different moisture levels on the parameters of a microwave resonant cavity and to evaluate the feasibility for determining the moisture content and mass of individual kernels by microwave measurements. MATERIALS AND METHODS PEANUT KERNEL PREPARATION Shelled peanuts of the Runner market type (Florunner cultivar) separated into two lots according to size (jumbo and medium), were obtained for this study from the USDA, ARS, National Peanut Research Laboratory, Dawson, Georgia, after the 1990 harvest season. Forty-six peanut kernels, including those of both medium and jumbo sizes, were randomly selected for the measurements. Some kernels were lightly sprayed with distilled water to VOL. 36(1): JANUARY-FEBRUARY 1993 127 increase their moisture contents to about 16% from the initial level of 10 to 12%. The weight of wet kernels ranged from 380 to 940 mg. These kernels were permitted to dry under ambient conditions for various time intervals between the microwave measurements, and after every drying period each kernel was individually sealed in a glass vial and kept at 4° C to equilibrate, usually for 2 to 3 days, before it was measured again. In this way each kernel was measured 3 to 5 times before it was finally dried to determine its dry mass. The dry mass was determined by placing individual kernels in copper moisture dishes (Nelson and Lawrence, 1989) and drying them in a forcedair oven for six hours at 130° C (ASAE, 1991). Kernels and moisture dishes were cooled in a desiccator over anhydrous CaS04 upon removal from the oven before being weighed. Subsequently, the kernel moisture content, wet basis, was calculated as a ratio of mass of water to mass of wet kernel as determined for each microwave measurement sequence. In 1991, four additional lots, two each of the Runner (Florunner cultivar) and Virginia market types in jumbo and medium sizes, were furnished for verification tests. Ninety-one kernels were randomly selected, with weight of wet kernels ranging from 420 to 1030 mg among Florunner peanuts and 460 to 1280 mg among the Virginia peanuts. Some of the kernels were lightly splayed with distilled water to increase their moisture over the original 6% level. After that, all kernels were kept individually in glass vials at 4** C for 72 h to equilibrate, and another 24 h at 23** C for temperature stabilization before microwave measurements. The previously described procedure was then followed in every detail. ELECTRICAL MEASUREMENTS The cavity consisted of a section of standard S-band WR-284 rectangular waveguide (inside dimensions: 72 x 34 mm) 305 mm long. It was coupled with external waveguides through two identical coupling holes 20.6 mm in diameter at each end of the cavity. A Plexiglas tube of 15.8-mm outside diameter and 12.4 mm inside diameter was installed in the center of the cavity as shown in figure 1. The resonant frequency of the empty cavity operating in the Hjos (TEjos) mode was 3175.9 MHz and its Q factor (Q )̂ was 865. The cavity was located between two waveguide-tocoaxial transitions, which allowed it to be connected to an automatic network analyzer calibrated in the transmission mode. The analyzer generated 801 discrete frequencies within a range of 16 MHz spanning the resonant frequency of the cavity. This allowed measurement of the transmission through the cavity in increments of 20 kHz by reading the coordinates of a marker on the test set CRT display. The resonance of the cavity appears as a peak in transmission through the cavity. To determine the resonant frequency, the frequency of a signal coupled to the cavity is varied until the maximum of the transmission is observed. The second parameter of the resonance curve, as shown in figure 2, is its shape. The coupling devices connected to the cavity, energy losses in the cavity walls and dielectric tubing, and losses in the sample introduced into the cavity all lower the apparent Q-factor. Thus, when a peanut kernel is introduced into the cavity, its resonant frequency will decrease and the Q-factor will be lowered, causing a broader, flatter resonance curve, shifted to the left. A "marker to maximum" command automatically accomplished the determination of the resonant frequency with an accuracy better than 10 kHz and the transmission coefficient (S21) through the cavity with an accuracy of 0.02 dB. In the following experiments, the shift of resonant frequency is denoted by AF = f̂ f̂ , where subscripts o and s refer to the empty cavity and cavity loaded with a sample (kernel) at the center of the cavity, respectively. Energy dissipated in the kernel is expressed as a change in the cavity Q factor: J 1 1 [̂ o i ) A T

Microwave techniques, methods and instrumentation can be utilized in agriculture to improve the efficiency of crop production, handling and processing, and improve the quality of products. Pertinent microwave theory relative to such applications is discussed, and a review of microwave research on agricultural problems is presented. A wide range of applications is treated briefly, including examples of successful applications of microwaves in agriculture, and some future prospects are discussed.

Use of broadband horn/lens antennas with a collimated beam in a free-space measurement set-up with a network analyzer operating between 3 and 13 GHz and applying time-domain gating improved the accuracy of permittivity (dielectric properties) measurements on grain. Measurements were taken in the transmission mode on samples from five different classes of wheat over ranges of moisture content, temperature, and bulk density. Dielectric properties were divided by bulk density of the grain for comparison of different wheat lots. No differences in permittivity data were noted among the different wheat classes, so data were combined for presentation. The new data agree well with those reported previously. For further research, to provide dielectric properties over a very broad frequency range for a given grain sample, use of time-domain spectroscopy is suggested.

The importance of cereal grain moisture content in determining time of harvest and in preserving grain quality is described. Techniques for moisture measurement, including electronic moisture meters, are also discussed briefly, and the need for better moisture sensing techniques for modern agricultural on-line moisture monitoring is described. Principles of microwave free-space measurements involving attenuation and phase determination are explained, and density-independent functions of the grain permittivity are presented that permit reliable moisture sensing applicable to moving grain in which bulk, density variation occurs. A new moisture calibration function, based on complex-plane plots of the density-normalized dielectric constant and loss factor, is described, which promises a universal calibration for different types of grain and seed. This important advantage should encourage commercial development for practical use.

The universal character of a permittivity-based calibration function for moisture determination in granular materials such as cereal grain and seed is tested through measurement at microwave frequencies on materials of different structure and composition using two free-space measurement arrangements. Results obtained from measurements on wheat, corn and soybeans provide moisture calibration equations with temperature compensation that are very similar for both systems leading to a single moisture calibration equation and the potential of transferability of the calibration routine across instruments of different designs.

A feasibility study of a universal calibration method for nondestructive real-time sensing of moisture content in granular materials is described. The principle is based on measurement of the dielectric properties of granular materials at microwave frequencies and the use of a calibration function that is expressed in terms of these properties to predict moisture content from a single moisture calibration equation. Effectiveness of this method is shown for three granular materials exhibiting significant structural and compositional differences: wheat, oats, and soybeans. The resulting single moisture calibration equation, based on measurements of the dielectric properties at 9.46 GHz and 24 degreesC, provides moisture content in all three materials with a standard error of calibration of 0.46% moisture content, wet basis.

A new calibration technique was developed for implementation with microwave moisture sensors. The calibration permittivity function used for this purpose allows computation of moisture content in granular materials with significant differences in shape, dimensions, and composition, independent of bulk density and with temperature compensation. A 3D representation is used to plot the calibration permittivity function as it depends on temperature and moisture content in wheat and corn. For each material, data points form a plane surface. These planes have nearly the same coefficients, which can be utilized for the development of a "universal" calibration method for moisture sensing in natural and manufactured granular materials. Foundations of the method are discussed based on results obtained for wheat and corn over a wide temperature range and at moisture contents of practical interest.

A dielectric method for determining bulk density of granular materials from measurement of their dielectric properties at microwave frequencies is presented. A complex-plane representation of the dielectric properties, normalized to bulk density, is used to generate calibration equations at 7 GHz for cereal grain and seed over wide ranges of moisture content and temperature.

A universal calibration method for microwave moisture sensors for granular materials is presented. The method is based on measurement of the dielectric properties and the use of a complex-plane representation of the normalized complex permittivity. Effectiveness of this method in moisture content prediction is tested for granular materials presenting significant structural and compositional differences. A single moisture calibration equation is given at 9.46 GHz and 24/spl deg/C for wheat, oats and soybeans.

A nondestructive dielectric method for sensing bulk density of granular materials is presented. The bulk density is determined from measurement of the dielectric properties of these materials at a single microwave frequency without knowledge of their moisture content and temperature. Bulk density calibration equations are generated from a complex-plane representation of the dielectric properties normalized with respect to bulk density. The effectiveness of the method is shown through measurement of the dielectric properties at 7 GHz for materials with significant compositional and structural differences, i.e. wheat. oats, corn and soybeans. over wide ranges of moisture content and temperature. The standard error of calibration and the relative error calculated for each material indicate that the method is as accurate as or better than commonly used methods for on-line density determination. Because the density is expressed in terms of the relative complex permittivity, the method can be applied regardless of the measurement technique (using transmission lines. a resonant cavity. admittance or impedance).

Sensors UpdateVolume 7, Issue 1 p. 393-414 Sensor Market Recent Bibliography on Moisture Sensing: 1990—1998 (Sensors, Methods, Applications) A. Kraszewski, US Department of Agriculture, Athens, GA, USASearch for more papers by this author A. Kraszewski, US Department of Agriculture, Athens, GA, USASearch for more papers by this author First published: 11 April 2001 https://doi.org/10.1002/1616-8984(200001)7:1<393::AID-SEUP393>3.0.CO;2-QCitations: 2AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onEmailFacebookTwitterLinked InRedditWechat Citing Literature Volume7, Issue1January 2000Pages 393-414 RelatedInformation

A brief history of cereal grain moisture measurement by sensing the electrical properties of grain is presented. The basic principles are also described for using radio-frequency (RF) and microwave dielectric properties, or permittivity, of grain for sensing moisture through their correlation with moisture content. The development of density-independent functions of the permittivity is explained. The findings of recent research are summarized, which indicate that reliable density-independent moisture content determinations can be realized by measurements on grain at RF and microwave frequencies. Development of these techniques will provide useful instruments for on-line monitoring of moisture content in flowing grain to manage moisture in grain, prevent spoilage in storage and transport, improve processing, and provide information important for yield determinations in precision agriculture applications.

Phase measurements of the transmission coefficient are important when used for the dielectric characterization of materials. They are required for industrial material monitoring applications, where the phase is correlated with parameters such as moisture content and density, which need to be continuously determined, However, when the thickness of the material under test is greater than the wavelength in the material, a phase problem is encountered. Two methods are proposed to solve this problem. The first is based on the selection of the appropriate material thickness; the second requires the use of measurements at two frequencies. Advantages and limitations of both methods are discussed, and numerical validations are given for particulate materials.

Sensors UpdateVolume 7, Issue 1 p. 51-64 Dielectric Properties and Sensor Technology Density-independent and Temperature-compensated Moisture Content Determination in Shelled Corn by Microwave Sensing A.W. Kraszewski, A.W. Kraszewski US Department of Agriculture, Agricultural Research Service, Athens, GA, USASearch for more papers by this authorS. Trabelsi, S. Trabelsi US Department of Agriculture, Agricultural Research Service, Athens, GA, USASearch for more papers by this authorS.O. Nelson, S.O. Nelson US Department of Agriculture, Agricultural Research Service, Athens, GA, USASearch for more papers by this author A.W. Kraszewski, A.W. Kraszewski US Department of Agriculture, Agricultural Research Service, Athens, GA, USASearch for more papers by this authorS. Trabelsi, S. Trabelsi US Department of Agriculture, Agricultural Research Service, Athens, GA, USASearch for more papers by this authorS.O. Nelson, S.O. Nelson US Department of Agriculture, Agricultural Research Service, Athens, GA, USASearch for more papers by this author First published: 11 April 2001 https://doi.org/10.1002/1616-8984(200001)7:1<51::AID-SEUP51>3.0.CO;2-9Citations: 5AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Citing Literature Volume7, Issue1January 2000Pages 51-64 RelatedInformation

Moisture content of shelled maize,Zea maysL., was correlated with the attenuation and phase shift of electromagnetic waves travelling through a layer of grain. Several calibration equations are presented in the paper that are based on measurements taken at 15·2 GHz for various grain densities, moisture contents and temperatures. Validation of the calibration equations indicated that moisture content can be predicted with an uncertainty less than ±0·72% moisture content w.b. at the 95% confidence level. Moisture predictions are largely independent of bulk density variations at temperatures from 4 to 45°C and moisture contents from 9 to 19% w.b. By following one approach described, the grain bulk density can be determined from the same measurements with an uncertainty of less than 25 kg/m3. With another approach, the grain bulk density cannot be determined, but moisture content is determined independent of the material bulk density and compensated for temperature. No differences among three maize hybrids were observed in the measured data.

A brief history of cereal grain moisture measurement by sensing the electrical properties of grain is presented. The basic principles are also described for using radiofrequency (RF) and microwave dielectric properties or permittivity, of grain for sensing moisture through their correlation with moisture content. The development of density-independent functions of the permittivity is explained. The findings of recent research are summarized, which indicate that reliable density-independent moisture content determinations can be realized by measurements on grain at RF and microwave frequencies. Development of these techniques will provide useful instruments for on-line monitoring of moisture content in flowing grain to manage moisture in grain, prevent spoilage in storage and transport, improve processing, and provide information important for yield determinations in precision agriculture applications.

The moisture content of granular materials can be determined indirectly and nondestructively from measurements of their dielectric properties at a given frequency and a given temperature. Several sensors have been developed based on this principle and successfully used in different industries including pharmaceutical, construction, and food and agricultural products processing. However, the effects of both the bulk density and structure are troublesome and cause significant errors if they are not taken into account or eliminated by defining appropriate calibration functions. It is possible to determine bulk density by an additional device, such as a /spl gamma/-detector, and compensate for its effect in the reading output of the meter. The structure, i.e., size and geometry in this instance, has a more subtle effect and although it is known to affect the scattering and absorption properties of the material, it is not possible to quantify this effect by measurement or theoretical means, and hence it cannot be compensated for in the final meter output. The only method applied to this time requires individual calibration for each particular material and instrument. In this paper, a new method is presented for moisture determination in granular material independent of bulk density and structure. This unified calibration method is based on the combination of two concepts, i.e., density independence and structure independence, embodied in a function fully defined in terms of the dielectric properties.

A method of moisture content determination in grain is discussed based on simultaneous measurements of two parameters of the electromagnetic wave transmitted through the layer of material. These parameters, attenuation, A, (decrease of the wave amplitude), and phase shift, F, (the phase delay in the material), are directly related to the amount of water in the material. Both parameters, however, are also dependent upon the material temperature and its density at the time of measurements. It has been observed empirically for many moist materials and agricultural and food products, including grain, that by measuring attenuation and phase shift, the effect of density variation can be minimized, thus leading to a density independent calibration equation. Temperature of the material is usually easy to measure, and it can be introduced as another variable into the calibration equation, making the measuring system both temperature- and density independent. This technique, however, is not limited to the frequency reported in this paper. It can be applied for much lower frequencies, as long as the ionic conductivity of the material is not a disturbing factor.