During high-temperature refining of vegetable oils, 3-monochloropropanediol (3-MCPD) esters, possible carcinogens, are formed from acylglycerol in the presence of a chlorine source. To investigate organochlorine compounds in vegetable oils as possible precursors for 3-MCPD esters, we tested crude palm, soybean, rapeseed, sunflower, corn, coconut, and olive oils for the presence of organochlorine compounds. Having found them in all vegetable oils tested, we focused subsequent study on oil palm products. Analysis of the chlorine isotope mass pattern exhibited in high-resolution mass spectrometry enabled organochlorine compound identification in crude palm oils as constituents of wax esters, fatty acid, diacylglycerols, and sphingolipids, which are produced endogenously in oil palm mesocarp throughout ripening. Analysis of thermal decomposition and changes during refining suggested that these naturally present organochlorine compounds in palm oils and perhaps in other vegetable oils are precursors of 3-MCPD esters. Enrichment and dose-response showed a linear relationship to 3-MCPD ester formation and indicated that the sphingolipid-based organochlorine compounds are the most active precursors of 3-MCPD esters.
During the milling process of palm oil, the degree of palm fruit ripeness is a critical factor that affects the quality and quantity of the oil. As the palm fruit matures, its chlorophyll level decreases, and since chlorophyll in oil has undesirable effects on hydrogenation, bleachability, and oxidative degradation, it's important to monitor the chlorophyll content in palm oil during the milling process. This study investigated the use of light-induced chlorophyll fluorescence (LICF) for non-invasive and real-time monitoring of chlorophyll content in diluted crude palm oil (DCO) located at the dilution and oil classification point in palm oil mill. An LICF probe was installed at the secondary pipe connected to main DCO pipeline, and the system communicates with a computer located in a separate control room via a Wi-Fi connection. Continuous measurements were recorded with an integration time of 500 ms, averaging of 10, and a time interval of 1 min between each recording during the oil mill's operation. All data were stored on the computer and in the cloud. We collected 60 DCO samples and sent them to the laboratory for American Oil Chemists' Society (AOCS) measurement to compare with the LICF signal. The LICF method achieved a correlation coefficient of 0.88 with the AOCS measurements, and it also provided a direct, quantitative, and unbiased assessment of the fruit ripeness in the mill. By incorporating Internet of Things (IoT) sensors and cloud storage, this LICF system enables remote and real-time access to data for chemometrics analysis.