Later, the system was started, turning on the radial blower to supply air find more to the system, and also, turning on the electrical heating until the set point temperature was reached. The behavior of the bed was found through the pressure drop for each increase of air velocity. Maximum pressure drop and the minimum spouting velocity (point where the bed collapse occurred)

were found. The fluid dynamics was carried out in all temperatures used (90, 100 and 110 °C). In each geometry, chitosan was dried in three inlet air temperatures (90, 100 and 110 °C), and air velocity used in the experiments was 100% over minimum spouting velocity, as recommended by Mathur and Epstein (1974) for pastes drying. When a steady velocity regime was established, the feeding system was set in motion and the chitosan paste with solid content of 4 g 100 g−1 (wet basis) was fed (0.18 kg paste kg inert−1 h−1) into the cell, through atomization with peristaltic pump and air compressed at pressure of 105 Pa gauge. Spouted bed chitosan drying occurred by fluid-particle contact, and also by friction

between inert particles caused by the Selleck BLU9931 high rate of circulation of the inert in the spouted bed interior. Dried chitosan in powder form was transported pneumatically by the drying air stream and collected in a cyclone. The dry and wet bulb temperatures of air drying were measured. The drying spouted bed experiments were carried out in 3 h, later the dried product was analyzed. Dryer performance was evaluated through determination of the accumulated mass in the bed and product recovery. Accumulated mass and product recovery were estimated by mass balance in the drier using Eqs. (1) and (2): equation(1) AC=(mFB−mIB)(1−UFB)mI×100 equation(2) R=mc(1−UF)mI×100where, AC is the mass accumulated in the bed (g 100 g−1), R is product recovery (g 100 g−1),

mFB and mIB are total bed mass in the end and in the begin of operation, respectively, UFB is final moisture content of the powder accumulated in the bed (g 100 g−1), UF is final moisture for content of powder (g 100 g−1), mI and mC are total solid mass introduced into the drier and collected in the cyclone, respectively. Chitosan paste was characterized according to centesimal chemical composition (A.O.A.C., 1995), molecular weight and deacetylation degree. Chitosan powder was characterized according to molecular weight, deacetylation degree, color and particle size. In the best drying condition, TG and DTG curves, FT-IR analysis and SEM were carried out to verify the powder quality. Chitosan molecular weight was determined by viscosimetric method (Cannon-Fenske capillary viscosimeter, model Schott Gerate, GMBH-D65719, Germany). Reduced viscosity was determined by Huggins equation, and converted into molecular weight through Mark-Houwink-Sakurada equation (Eq. (3)), using K = 1.81 × 10−3 mL g−1 and α = 0.93 ( Weska et al., 2007).