Abstract Five amylose preparations of different origins (cassava, potato, smooth seeded pea, wheat and maize), obtained from native starch granules by thymol complexation, were ultracentrifuged as the final decontamination step to remove a high molecular weight population contaminating the amylose solutions. The efficiency of this ultracentrifugation procedure was assessed by dynamic light scattering (DLS): decreases in apparent hydrodynamic radii, R? H , from 46·1–72·6 nm before ultracentrifugation to 16·1–29·3 nm after were observed. Amylose solutions were then characterised by size exclusion chromatography coupled on-line to multi-angle laser light scattering (SEC–MALLS). Under these conditions M? w of cassava, potato, smooth seeded pea, wheat and maize amyloses were, respectively, 1·05×10 6 , 7×10 5 , 6·2×10 5 , 5·1×10 5 and 3·4×10 5 g/mol. Using a specific optimisation algorithm, experimental molecular weight distributions (MWD) were fitted by two mathematical models of ‘Most Probable’ distribution and ‘log-normal’ distribution. The best fit was obtained for the second model, but fitted M? w were higher than experimental M? w . When a ‘Most Probable’ model was used, the fitted M? w were consistent with experimental M? w but with a lower quality of fit. Exponents c in the power law R G = KM c were between 0·6 and 0·7, indicating an extended linear random coil in the range of MW analysed (3×10 5 –9×10 6 g/mol). This procedure was also applied to the characterisation of different commercial amylose products.
Read full abstract