Thermal behavior of palm oil samples drawn from the batch crystallizers that failed during crystallization and of a control oil that was drawn from a batch that produced good crystallization were analyzed by differential scanning calorimetry under constant heating and cooling conditions. Four po[ymorphs--~, o~, 13~, and ffl--were observed, and their temperatures were tabulated. A rapid and sudden surge of heat demand was observed for samples from failed crystallizers. Less supercooling values were obtained from the control oil compared to the higher values for samples from failed crystallizers. In crystallization thermograms, a sharp high-temperature exotherm (high-T peak) and a broad low-temperature exotherm (Iow-T peak) were observed. Low-T peaks were found almost invariably stationary at -5.1 to -5.6°C, and high-T peaks varied depending on the saturation level of the oil. A new peak, sandwiched between the high-T and low-T peaks, was observed for the control oil. JAOCS 72, 1529-1532 (1995).

Non-aqueous reversed-phase high-performance liquid chromatography (NARP-HPLC) with refractive index (RI) detection is described and used for palm olein and its fractions obtained at 12.5 °C for 12-24 h. The calculation formula for fatty acid methyl esters (FAMEs) and carbon number (CN) from the data obtained by NARP-HPLC is described and correction factors for all carbon numbers and fatty acids are tabulated. The results were compared with those obtained from FAMEs analysed on 10% SP 2330 Supelco packed column gas-liquid chromatography (GLC) and CN analysed on 3% OV-1 Supelco packed column GLC. The results were found to agree well for C48, C50, C52, C16:0, C18:0 and C18:1 (correction factor ≈ 1.0); however, a slight variation was observed for components C54, C14:0 and C18:2 (correction factor 1.0 ± 0.37).

The composition of purified palm olein crystals formed at room temperature (25°C) was identified in this study. Two peaks were obtained when the crystals were analyzed by reverse-phase high-performance liquid chromatography (RP-HPLC). The retention times of these peaks suggested that they were not triglycerides. Gas-liquid chromatography of fatty acid methyl ester analysis of the crystals showed the presence of C16, C18:0, and C18:1 fatty acids. Further analysis by gas-liquid chromatography of carbon number, after collecting the fractions from RP-HPLC, concluded that the major peak A was 1,3dipalmito-glycerol. The minor peak B was tentatively identified as 1-pa[mito-3-stearo-glycerol and/or 1-palmito-3-oleo-glycerol due to unavailability of respective standard glycerides. The differential scanning calorimetry thermogram of the crystals show that A and B were indeed the high melting glycerides, with melting and crystallization points of 70.4 and 53.8°C, respectively. JAOCS 72, 343-347 (1995).KEY WORDS: Crystals, 1,3-dipalmito-glycerol, HMGs, 1-palmito-3-oteo-glycerol, 1-palmito-3-stearo-glycerol, palm olein. L,I.I z z o o. z m z 9 I J MIN FIG. 3. Reverse-phase high-performance liquid chromatography chromatogram of purified palm olein crystals. See Figure 1 for abbreviation.

Clouding was obtained by storing palm olein at 12.5~ for up to 24 h and was separated by centrifugation. The fat crystals were collected after washing with cold acetone. The crystals were identified according to their fatty acid and triglyceride composition, carbon number and degree of unsaturation. Palmltic-oleic-palmitic (POP) and paimitic-oleic-stearic (POS) levels were high in the cloud material, especially in that recovered between 15-18 h of storage. The increase in POP and POS concentrations was concomitant with a decrease in the content of palmiticoleic-oleic (POO). The least amount of POO was also obtained in clouds collected between 15-18 h of storage, compared to the original oil sample. The increase of palmltic acid explained that triglycerides conjugated with palmitic acid molecules in their acyl chains could have been crystallized, and crystallization could have taken place in the order of the number of paimitic molecule in the acyl chain. The concentrations of the other triglycerides did not change much throughout the storage period. Further storage caused the composition of the cloud to become similar to that of the original oil sample.

Thermal behavior of crude palm oil (CPO) is important to determine the optimal fractionation process and product yield. In this study, the effects of repeated heating on thermal behavior of CPO were examined by differential scanning calorimetry. CPO was heated at 80°C for 5 min, and heating was repeated five times to simulate the common conditions experienced by an oil before reaching the refinery. The result revealed that the thermal behavior of CPO changed after heating. The change, however, occurred only in the behavior of the high-melting stearin peak but not in the low-melting olein peak. Overheating split the stearin peak at 17.30°C to two peaks at 18.88 and 17.30°C and formed a new peak at 11.28°C. Apparently, a new substance has been synthesized.Crude palm oil (CPO) has been heated several times before it reaches the refinery. Every time CPO has to be removed from a storage tank for transportation, it must be heated up above its melting point (about 37°C) to facilitate pumping of the liquid oil. The frequency of being heated will increase if only a part of the stored CPO is removed. It is recommended that heating does not exceed 55°C (1). However, in practice, it is quite frequently violated. The typical reason for such violations is to shorten the heating time and to avoid claims on demurrage. Storage in tanks that are not equipped with agitators could also induce localized heating. Another reason is to keep the transported CPO in liquid form without heating during transportation because most tankers are not equipped with heating system.One instrument that can be used to determine thermal properties of materials is the differential scanning calorimeter (DSC). The DSC method has been used in several works related to oils. Hampson and Rothbart (2) used DSC to determine the specific heat of triglycerides. Hagemann and Rothfus (3) used DSC to find out whether polymorph properties of saturated monoacid triglyceride were influenced by the length of odd and even chains. deMan et al. (4), D'Souza et al. (5), and Herrera and Anon (6) also used DSC to study melting and crystallization of some vegetable oils.DSC studies on palm oil and palm kernel oil have also been conducted by several workers (7-17). Busfield and Proschogo (7,8) studied heating thermograms of palm stearin and its product of hydrogenation. Their results showed that there was a relationship between thermogram profiles and crystal forms. Che Man and Swe (9) studied the possible cause of poor crystallization of palm oil during fractionation. The cooling thermogram of a palm oil from a failed batch differs from that produced after good crystallization. Yap et al.(10) studied polymorphism of palm oil and palm oil products with DSC and X-ray diffraction. The results showed that storage times of palm oil and its products affected their DSC heating curves.The success of palm oil fractionation determines both quality and yield of the products, namely stearin and olein. Technically, fractionation is done based on the thermal behavior of the oil fractions. R...