Statistical studies were carried out on the quantitative analysis in gaschromatography. The two gaschromatograph instruments attached with the heat conductivity detector cell were used. The factors which chosen are the flow rate of carrier gas, the column temperature, the kind of stationary phase, the detector oven temperature, and the amount of samples. The results of the variance analysis and S/N ratio analysis are as follows : About the other factors, the significant differences were not found. (1) The factors which affects to the peak areas are (a) the instruments (b) the flow rate of carrier gas (c) the amount of the samples (2) Instrument No. 1 showed the better accuracy which detects the infinitesimal change of high boiling components in the solvent mixtures than the other. Contrarily, instrument No.2 showed the better accuracy for the low boiling components. (3) The following conditions were recommended for the quantitative analysis of solve nt mixtures. Instrument 1 Instrument 2 Column Temperature 150s°C 120°C Detector Oven Temperature 250°C 200°C (4) From the relations between peak areas and compositions of the samples, the scatterings were minimum when the compositions were calculated as the volume ratios of the components with instrument 1, and with the instrument 2, the scatterings of the weight ratio calculated from peak areas and 3-constants were of minimum. Here β=Wi/Σw· Σa/ai Wi : weight fraction of the i-component at : peak area fraction of the i-component (5) The scattering of the results obtained with instruments No. 2 was nine times larger than that of instrument No. 1.
The curing reactions of acrylic copolymers (polymethylmethacrylic acid-n-butylmethacrylate) with the curing agent (hexakis methoxymethyl melamine) has been studied by the determination of crosslinking density of sample coatings, cured at various temperatures. Crosslinking densities of the sample coatings have been determined from the stress-strain measurement for swollen samples by using the equation from the theory of rubber elasticity. The equation we used is τ=nRT (α-1/α2) Vp, 1/3 Where τ is the force per unit cross-sectional area of the swollen unstrained film, n is the number of network chains per cc, T is the absolute temperature, α is the relative extension and vp is the polymer fraction of swollen sample. The experimental results from this study are as follows ; 1. Condensation reactions which occures in these series of samples are classified into ture types from the crosslinking structure. One is the effective reaction for crosslinking, and the other is the ineffective reaction. These reactions are recognized from the experimental results, and can be explained by the crosslinking model which is proposed by J. N. Koral, et al. As the experimental result, crosslinks increase with the increasing amount of crosslinking agent in the first region in composition (where lesser amount of crosslinking agent is present). While, the crosslinks decrease with increasing amount of crosslinking agent in the second region (where excess amount of crosslinking agent is present). 2. We have applied the superposition rule of time and temperature to the curing reactions, i. e., crosslinking densities are plotted vs. logarithmic value of curing time at each curing temperatures, 140°C, 160°C, 180°C, 200°C, and resulting each curves are shifted along the time axis to obtain the master curve in the temperature range from 140°C to 180°C. The sample films cured at 200°C, are not followed to the superposition rule of time and temperature. 3. The Arrhenius type plots of logarithmic values of shift factors vs. reciprocals of absolute temperatures the show the linear relations. The apparent activation energies (_??_H) for the crosslinking reaction have been calculated by using the following equation, _??_H= Rd (ln αT) / d (1/T) The apparent activation energies for the curing reaction of these sample coatings are in the range from 22 Kcal. /mol to 31 Kcal. /mol. The activation energies monotonously decrease with increase of melamine content.