Skip to main content
ARS Home » Southeast Area » New Orleans, Louisiana » Southern Regional Research Center » Commodity Utilization Research » Research » Publications at this Location » Publication #76152

Title: MEASUREMENT OF EDIBLE OIL COLOR: AUTOMATED VERSUS VISUAL METHODS

Author
item Wan, Peter

Submitted to: Oilseed Processing and Utilization World Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 9/8/1997
Publication Date: N/A
Citation: N/A

Interpretive Summary: Color of edible oil is an important quality factor for the processors as well as the end users. The color determination of oil is traditionally done by visual measurement which subjects operator to operator variations. Both the results on cottonseed oils refined and bleached in the lab and in international collaborative study on oil color of eighteen samples from seven different types of oils demonstrated that good correlation between readings from a microcomputer based automated colorimeter with the conventional visual color measurements can be established.

Technical Abstract: Color as a fundamental quality of edible oils has been determined primarily by visual comparison methods for many decades. These methods require experienced operators and are subject to operator variability. Recently, several versions of automatic colorimeters were introduced. One study using an automatic colorimeter, Colourscan, to measure the color of refined and refined-bleached cottonseed oils showed good agreement (r to the 2nd = 0.99) with visual color measurement using Lovibond-AOCS Color Scale. Effect of temperature and path lengths of color measurements were also studied. Another international collaborative study established a broad scale correlation between the automated colorimeter and visual color measurement. In this effort, factory processed refined, and refined-bleached-deodorized (RBD) canola, corn, cottonseed, peanut, sunflower and soybean, as well as refined palm olein, RBD palm, washed-dried-filtered and deodorized tallow were used. Fourteen labs from the United States and Canada, and sixteen labs from twelve countries outside of North America, participated. The correlations between automated and visual color measurements were good (r to the 2nd = 0.94-0.99) for red of both refined and RBD oils and yellow of RBD oils. The exception was yellow color for refined oils which exceeded the maximum yellow scale.