Cassava starch nanoparticles with an average size of (273.8 ± 26.41) nm were prepared by ultrasonic crushing followed by ethanol sedimentation and were esterified with acetic anhydride to prepare hydrophobic nanoparticles. Fourier transform infrared (FTIR) spectroscopy and nuclear magnetic resonance (NMR) spectroscopy showed that the degree of substitution (DS) of the esterified products was as high as 0.53 without adding any catalysts during esterification. X-ray diffraction (XRD) analysis showed that the crystal structure of starch nanoparticles did not change significantly after modification, and both the original and modified nanoparticles were amorphous. Scanning electron microscopy (SEM) showed that the morphology of the nanoparticles did not change significantly after modification. The three-phase contact angle increased with increasing DS, reaching a maximum of (109 ± 3.56)°, and the higher the DS, the longer the contact angle was maintained. The modified product was easier to disperse in the oil phase, leading to the formation of microemulsions. Surface modification changed the hydrophilic/hydrophobic properties of starch nanoparticles, and improved its potential for emulsification at the oil-water interface. In contrast to the unmodified sample, the modified sample showed similar enzymatic digestion curves and no fundamental change in eating quality, but lower enzymatic digestibility at the same time points, suggesting stronger resistance of the modified product with higher DS to enzymatic hydrolysis.

In order to study the feasibility of applying cassava starch-fatty acid complexes as a Pickering emulsion stabilizer, complex nanoparticles with complexing index (CPI) of 2.74%, 9.17% and 27.66% were prepared by mixing cassava starch paste containing 78.65% amylopectin at 95 ℃ and stearic acid followed by alcohol precipitation. The three complexes had an irregular spherical-like shape under field emission scanning electron microscopy (FESEM), and their average particle sizes, determined by a laser particle size analyzer, were 315.35, 348.19 and 427.60 nm, respectively. The X-ray diffraction pattern of each of the complexes showed two peaks at 13° and 21°, which were characteristics of the V type crystal structure, and the crystal content increase with increasing CPI. Their deconvoluted infrared spectra exhibited changes in short-range ordering at 1047, 1022 and 995 cm^-1. The contact angle of the particles with the highest CPI was 60.30°. The three complex nanoparticles stabilized Pickering emulsions for more than seven days compared to less than two days with starch nanoparticles. The complex nanoparticles with CPI of 27.66% stabilized emulsions best. The addition of the complex nanoparticles with CPI of 27.66% at levels above 0.1 g/100 mL resulted in the formation of an emulsion with an oil-to-water ratio of 1:9 (V/V). The emulsion with this nanoparticle at 7 g/100 mL exhibited an improved stability for 60 days without creaming or phase separation. Moreover, no significant changes in the droplet size distribution were observed. The emulsion was stable at pH 5.6-9.0 and not affected by NaCl concentration in the range of 0.01-0.1 mol/L. The emulsion maintained its morphology well after being heated to 80 ℃. These results suggest that the complex nanoparticles are a potential Pickering emulsion stabilizer.