For the purpose of detecting foreign fats in lard, differential thermal analysis of crystallized glycerides from acetone by the Bömer method was carried out. 1) The DTA curve of the crystallized glyceride itself by the dilution method showed each one fusion peak at 6365°C for lard, at 5860°C for beef tallow, at 60°C for mutton tallow, and at 59°C for hardened lard. It is impossible to judge the mixing ratio from this point alone. 2) The crystallized glyceride was melted, cooled rapidly, and then heat-treated at 18°C. By doing this treatment, lard showed three endothermic peaks at 48, 55, and 66°C, and one exothermic peak at 58°C. Beef tallow showed a large endothermic peak at 61°C and a small exothermic peak at 47°C, exhibiting a DTA curve entirely different from that of lard. Mutton tallow showed a curve quite similar to that of beef tallow, and hardened lard showed two endothermic peaks, a gentle slope peak at 44°C and a sharpe one at 56°C, and one exothermic peak at 47°C. There could be seen no any change of these peaks according to the duration of heattreatment. 3) Presence of more than 10% of beef tallow in lard can be detected from changes in DTA curve and, further, estimation of the mixing ratio is believed possible from the characteristics of the curve in the rough mixing range of 1060%. 4) Presence of 1080% of hardened lard in the lard makes difficult the judgement, whether the foreign fat is beef tallow or hardened lard, from only result of Bömer number. Whereas it is believed possible to estimate the rough mixing ratio from DTA curve.
In the previous papers, the authors reported the aromatic ο-substitution products which were formed by the disproportionation of the cyclohexadiene compounds produced in heated linseed oil in the presence of alkali or in heated ethyl β-eleostearate. In these experiments, linseed oil was heated at 275°C for 12 hrs in the presence of carbondioxide without use of catalyst and solvent. Heated linseed oil fatty acid methyl esters were prepared by the routine method. Cyclic monomers were obtained from the vacuum distillate (187°C/0.21.5 mmHg) of the esters by urea adduct separation. In the IR spectrum of the cyclic monomer fraction, the absorptions of aromatic ο-substituents and cyclohexadiene compounds were not observed, and in the GLC, there was a very small peak at position corresponding to cyclohexadiene compound and there was no any aromatic compound peak. Main parts of the cyclic monomers were cyclohexene compounds. The cyclic monomer fraction was processed by the following procedures, (1) dehydrogenation, (2) aromatization with N-bromosuccinimide and (3) further heating (275°C, 6 and 12 hrs). The results of these procedures showed that ; (1) aromatic ο-substitution products appeared slightly, (2) cyclohexadiene and aromatic ο-substituted compounds were produced in both and (3) the cyclic monomer fraction did not change further except polymerization. As above mentioned, in the case of heating in the absence of alkali, both cyclohexadiene compounds and aromatic ο-substitution products were not produced, because the disproportionation reaction did not occur on the cyclohexene rings.
Forty-five brands of household margarine were submitted ta measurexnent of hardness (f), plastic viscosity (ηpl) and apparent viscosity (η*) at 20°C. Fats were then separated and submitted to the measurement of S.F.I. and micropenetration, and their correlation was examined. 1) The values of hardness were 1020 × 104 dyne/cm2 (x 15.3×104), Plastic visosity 1020×106 poise (x 13×106), and apparent viscosity 110×105 poise (x 6.7×105) inmajority of the samples. Thesevalues are considered as standard range of Japanese household margarines. 2) Values of S.F.I. of the fats were mostly between 15 and 20. Micropenetration measulementindicated a hyperbolic relation and this relationship can be expressed by M/S=-1.303-0.765 M. 3) In order to find the correlation between the content of liquid fraction index (LF.I.) and amountof water, multiple regression analysis was carried out with F-test (0.01). It was thereby found that the values of both ηpl and η decreased with the increase of LFI, but were not affected by theamount of water. 4) There was a linear relation between S.F.I. and values of ηpl and η, and a fairly satisfactory result was obtained for the confidence limit, 95% of coeffcient of regression against the regxesslonline. 5) There was no special correlation between chemical and ffatty acid compositions because of thecomplex nature of the fats used.
Palm kernel oil was purified and hardened, and the samples at each stage of the process including crude oil were stored at -20, 5, 15 and 30°C for 6 months, in the same way as done with the coconut oil, of which A.V., P.O.V. and Co.V. were measured in every month. Correlation of these values with crystal growth was examined by electron microscope. 1) Palm kernel oil showed the same tendency as the coconut oil ; increase in A.V. was in the order of those at 15 and 5°C, and the increase was more marked in the hardened oil. No change was recognized in those stored at-20 and 30°C. 2) Crystal surface of the hardened palm kernel oil under electron microscope showed the presence of “whiskers” in the oil stored at 5°C and that stored at 15°C showed a very rough crystal surface, both being larger than those of hardened coconut oil. Single crystal was small and few in the oil stored at 20°C and the surface was smooth in the stored at 30°C. 3) These experimental results indicated that the changes of A.V. were greater in those with rough crystal surface.
An oxirane group was treated with 100200% excess of HCl, and remaining HCl was determined by applying a reaction in which I2, is liberated by the action of HCl to a mixture of KBrO3 and KI. Thus, 10 ml of 1/10 N-HCl, dioxane-CCl4 (1 : 1) solution, was added into a solution in which 0.10.2 g of epoxystearic acid was dissolved in 10 ml of CCl4, and swirled occasionally for 10 min. Then 10 ml of an aqueous solution containing 0.5 g of KBrO3 and 0.5 g of KI was poured into the solution, and the mixture was shaken for 1 min followed by allowing to stand for 4 min. Iodine liberated was determined with 1/10 N-Na2S2O3 solution after adding 30 ml of water. Similar procedure was accomplished without epoxy compound. No compensation for neutralization value was necessary. The value of oxirane oxygen content obtained accorded with that by the A.O.C.S. method. Anisaldehyde, benzoyl peroxide, cetyl and oleyl alcohols, and caprylic, stearic and oleic acids had no effect on the determination. Acetic acid, however, gave lower value of oxirane oxygen content. The present method was also suitable for the determination of low oxirane oxygen (<1%).
Authors attempted to presume the sort of lubricating oils by using eluogram that shows relation. between the eluted quantity of oil and retention time, prepared from the chromatogram by programmed temperature gas chromatography. By the use of this eluogram, the groups of spindle oils, machine oils, and engine oils were able to discriminate. The eluograms of electric insulating oils and liquid paraffin belonged to spindle oil group, and those of cylinder oils to engine oil group, and they could not differentiate from another, although the shapes of eluograms were different each other.