Characterisation of "glassiness" in commercially processed french fried potatoes
The relationship between the “glassiness” defect in frozen French fries and the moisture, starch and reducing sugar content of the affected potato tuber was investigated. The effect of soil water quality, cultivar, soil depth, storage duration, specific gravity and blanching conditions during French fry production on the occurrence of “glassiness” was determined. Fourier transform near infrared (FT-NIR) spectroscopy was used to identify possible classifications of defected tubers. No significant difference occurred between the moisture (p=0.10, trial 1 and p=0.15, trial 2), starch (p=0.76, trial 1 and p=0.70, trial 2) or reducing sugar (p=0.05, trial 1 and p=0.51, trial 2) content of potato sample with and without the “glassiness” defect. Samples of the cultivar Herta (Her) showed the lowest occurrence of the defect (23%, trial 1 and 0%, trial 2), while the cultivar Columbus (Col) showed the highest occurrence (70%, trial 1 and 84%, trial 2). The soil water quality prevailing in the area of cultivation contributed to the amount of “glassiness” occurring in the samples of the cultivar Col. Col obtained from the Parys area (electrical capacity (EC) = 145 mS.m-1) showed a 21% occurrence of “glassiness”. Col obtained from the Uitvlug (EC = 57 mS.m-1) and Zandrug (EC = 25 mS.m-1) areas showed a 91% occurrence of the defect. All samples cultivated in the Parys area during trial 1 showed a significantly lower occurrence of “glassiness” (p=0.01) than samples obtained from the areas Uitvlug and Zandrug. During trial 2 all samples obtained from the Thaaibos area (EC = 82 mS.m-1) showed a lower occurrence of the defect than samples obtained from the area Witklip (EC = 178 mS.m-1) although this difference was not statistically significant (p=0.06). Soil depth, specific gravity and storage duration did not contribute to a significant difference in the occurrence of “glassiness” between samples. Modified blanching conditions of 62ºC for 25 min instead of 80ºC for 20 min during frozen French fry processing had a reducing effect on the occurrence of the defect in the cultivars Fianna (Fia) (p=0.06), Pentland Dell (Pen) (p=0.05) and Col (p<0.01). The modified blanching conditions improved the texture uniformity in the French fry strip, reducing oil absorption during frying and prevented fry strips from breaking during subsequent processing steps. FT-NIR calibration models could not be successfully developed for the prediction of the moisture, starch and reducing sugar content in a potato sample. Principal component analysis (PCA) indicated no classification between potato samples affected by the “glassiness” defect and samples without the defect. The calibration models for moisture, starch and reducing sugar content yielded a standard error of prediction (SEP) of 1.62%, 2.28% and 0.07%, respectively. The respective correlation coefficients of these calibration models were 0.46, 0.42 and 0.41. The “glassiness” defect was most prominent in the cultivar Col. The occurrence of the defect was reduced and French fry quality improved by adjusting blanching parameters to 25 min at 62ºC. FT-NIR spectroscopy is not recommended for screening of potato quality prior to processing.