[Development of an asymmetrical flow field-flow fractionation system for the size characterization of starch granules].

Starch occurs naturally in the form of semicrystalline granules, and is composed of two types of carbohydrate molecules, amylose (AM) and amylopectin (AP). Starch granules and starch molecules have sizes in the range of 1-100 μm and 20-250 nm, respectively; these size ranges are among the key factors affecting the functional properties of starch. Asymmetrical flow field-flow fractionation (AF4) is a size-based separation technique. The major difference between AF4 and dynamic light scattering or microscopy techniques is that AF4 enables the separation of particles based on their size; consequently, the elution profile can be converted to the size distribution of the samples. In the last two decades, AF4 systems, when coupled online with multiangle light scattering (MALS) and differential refractive index (dRI) detectors (AF4-MALS-dRI), have demonstrated to be applicable for the size characterization of starch at the molecular level. Unlike size exclusion chromatography (SEC), AF4 systems use an open channel that does not require a stationary phase or packing materials. Thus, the shear scission of AP molecules during AF4 separation is minimized. The size detection range of a commercial AF4 system ranges from 1 nm to 10 μm, which is smaller than the size range of starch granules. In this study, a home-made AF4 system was developed, and its capability for the size characterization of starch granules extracted from sweet potato, lotus seed, and rice was investigated. The performance of the developed AF4 system was evaluated by running a mixture of polystyrene (PS) with diameter of 2, 6, 12, and 20 μm, respectively. Baseline separation of four PS samples was achieved, and the resolution for 6 μm PS and 12 μm PS was 1.40. The detection limit of the developed AF4 system was higher than that of commercial AF4 systems. Thus, the developed AF4 system is promising for the separation and characterization of starch granules. The effect of the composition of the carrier liquid on the AF4 separation of starch granules was also studied. Moreover, the accuracy of AF4 in terms of size characterization of the starch granules was evaluated by optical microscopy (OM). The results revealed that the type of dispersant and viscosity of the carrier liquid affect the accuracy of size characterization of the starch granules. The size distribution of rice starch granules obtained using a carrier liquid containing 0.01% (w/v) sodium dodecyl sulfate (used as a dispersant), 0.02% (w/v) NaN (used as a bactericide), and 0.001% (w/v) hydroxypropylmethylcellulose (used to adjust the viscosity of the carrier liquid) was in agreement with that obtained from OM. Furthermore, a commercial AF4 system coupled with MALS and dRI detectors was employed for the separation and characterization of starch molecules. A molecularly dispersed solution is necessary for the reliable molecular characterization of starch. The effect of the starch dissolution temperature on the AF4 characterization of starch was also investigated. The optimal dissolution temperature for lotus seed and rice starch granules was 75 ℃, while that for sweet potato starch granules was 78 ℃; this difference is mainly attributed to the different botanical origins of the granules. The results showed that the ratio of the radius of gyration () to the hydrodynamic radius () of rice starch and sweet potato starch is in the range of 0.9-1.1 over the molar mass range of 10 -10 g/mol. For rice starch, the / ratio is between 1.2 and 1.4. Rice starch has the highest apparent density among the three starches, indicating that rice starch molecules have a dense structure. The results demonstrated that the AF4 system developed in this study is rapid and accurate for the size characterization of starch granules. The developed AF4 system, when combined with commercial AF4 systems coupled online with MALS and dRI detectors, can provide technical support to study the relationship between the size from the nanoscale to the microscale and functional properties of starch.