We demonstrated a quality evaluation of cheeses using terahertz spectroscopy. Strip preparations of seven kinds of commercially produced cheeses were measured and the characteristic spectra were obtained. Casein, milk lipid, and sodium chloride, as dominant components of cheese, were scanned using a terahertz spectrometer based on a Fourier transform infrared spectrometer, and we found that the absorption coefficient of casein became larger than that of sodium chloride below 75 cm-1. We analyzed the spectrum of Camembert cheese at 75.2 cm-1 to obtain a correlation with the crude protein concentration using the Kjeldahl method, and a significant correlation was shown. However, since the shape of the terahertz spectrum of casein was different from that of defatted and dehydrated cheese, this suggested the possibility that not only the crude protein concentration but also the micro structure consisting of casein micelles or fat particles in cheese was included in the terahertz spectrum.
We propose a new label-free allergen sensing technique using a metallic mesh and a nitrocellulose membrane, which is generally used for fixation of protein. Our sensing method relies on a change in the transmittance of terahertz (THz) radiation passed through the metallic mesh on which a sample was placed. Label-free detection using THz radiation is a new technique in biomedicine that enables us to make easier and faster food allergen inspections. By using this technique, we demonstrated detecting the interaction between a streptavidin - biotin and α-casein- anti α-casein antibody on each membrane. The streptavidin - biotin interaction could be detected with 1.6 - 5.0 ng/mm2 streptavidin. The possibility of detecting the α-casein - anti α-casein antibody interaction was shown with 9.0 ng/mm2 α-casein. These results indicate that our sensing system is sensitive enough to detect allergens in food.
The performance of a near-infrared(NIR) spectrometer that evaluates fatty acid composition in samples depends on the NIR spectroscopic features of the samples. When you develop a NIR instrument, you have to be aware how the NIR spectra of the samples behave. In some cases, you don't need a high SN ratio or high stability. In other cases, no matter how stable the system is, you can't get meaningful data. It is wasteful to develop a highly stable instrument in this case. You should research the NIR characteristics of the samples before starting designing to avoid waste and to get a proper design. We examined the pork fat spectra before starting designing and were able to get satisfactory result. The process and aspects of measuring fatty acid composition by NIR spectroscopy are described.
We developed a mid-infrared (MIR) spectroscopic evaluation method of brewed coffee, whose quality and taste depend highly on properties such as the geographical origin and the roasting, grinding, blending, and extraction conditions, using an FT-IR spectrometer equipped with an attenuated total reflectance (ATR) accessory. In addition, the simultaneous content determination of the main components of coffee drink was performed by applying the FT-IR/ATR method. We studied the effects of different varieties of coffee and the degree of roasting on the MIR spectral characteristics of brewed coffee and determined the caffeine and chlorogenic acid content using the FT-IR/ATR method. Arabica (from Colombia, Brazil, and Guatemala) and Robusta (from Indonesia) coffee roasted to city was prepared as the sample. Indonesian coffee roasted to full city was also used. The differences between the second derivatives of the ATR spectra of the brewed Arabica and Robusta coffee were observed around the several absorption peaks, and the brewed coffee from Brazil had different spectral features from those of the other varieties of Arabica coffee. Additionally, the roasting conditions of the Indonesian beans were reflected in the spectral features of the brewed coffee. Moreover, the caffeine and chlorogenic acid content in brewed coffee was determined by spectroscopy. For the coffee drink that was composed of instant coffee, sugar, and coffee whitener, the content of each component was excellently determined using the FT-IR/ATR method by analyzing the spectroscopic interactions between the components.
Multiple linear regression (MLR) analysis of spectra (500-1000 nm) of tomatoes (n = 82) gave a calibration equation using d2 log 1/R at 568, 602, 626, 692, 826, and 946 nm with a multiple correlation coefficient of 0.97. The MLR calibration was validated using other tomato sample lots, and the tomatoes were predicted well with a root mean square of 1.28 mg/100g (n = 42). Therefore, visible/near-infrared technology is a potentially effective way to non-destructively determine lycopene in tomatoes.
A low-pressure mercury (Hg) cold cathode lamp (CCL) was filled with neon at a few kilopascals and mercury (Hg). It was important to grasp the difference in electrode characteristics between normal lighting condition (Hg-vapor exists in the CCL) and end-of-life lighting condition (Hg-vapor is used up in the CCL) when we developed the CCL electrode. Then we investigated the variation of the electrode characteristics in some electrode materials with the CCL (Hg plus) filled with Hg and neon (Ne) and the CCL (Hg less) filled with only Ne. It is clear that when the CCL was filled with Hg, the cathode fall voltage (CFV) was higher and the negative glow of the CCL (Hg plus) spread larger (the current density on the electrode was lower). The former result is due to the fact that the value of γ (Hg2+) was lower than that of γ (Ne+), and the later result is due to the fact that the Hg ion's mobility was higher than the Ne ion's one. By observing the CFV waveform, it is clear that the CCL (Hg less) was in abnormal glow discharge, whereas the CCL (Hg plus) was in normal glow discharge. According to these results, the electrode consumption increased rapidly when the Hg was used up in the CCL.