Page 478 - Biosystems Engineering

450 Cha pte r S i x tee n

Walker 2002; Andrich et al. 2005; Cantrell and Walker 2006). Badal
and Walker (2002) conducted the SFE of raw and fermented rice bran
for value-added oil. This study focused on determining the effect of
particle size (>48 mesh size and 16 to 48 mesh size) and biotreatment
on the yield and quality of rice bran oil. The results showed that
eicosapentaenoicacid (EPA) and arachidonic acid (ARA) produced
during biotreatment were extracted by SFE at 40°C and 4000 psi.

16.2 Supercritical Mass-Transfer Mechanisms
The SFE process involves removing solutes from porous matrices into
the supercritical CO via internal and external mass-transfer mechanisms.
2
Generally, the extraction process consists of three periods. The first, the
constant extraction rate period or the solubility-controlled period, is gov-
erned by the solubility equilibrium between the CO solvent and the
2
extract. The transition period starts when diffusion-controlled mass
transfer occurs and extraction rate decreases. Finally, diffusion com-
pletely controls the mass-transfer process, and the extraction is said to be
in a diffusion-controlled period. Diffusion of the extracts through the
bulk material becomes a more integral part of the extraction process;
over time, product accumulation approaches zero (Reverchon et al. 2000;
Sovova 2005). An extensive discussion on modeling techniques may be
found in McHugh and Krukonis (1994) and Walker et al. (2007).
The mass-transfer rate of high initial concentration of extracts
(such as oilseeds) from a fixed bed containing natural material typi-
cally remains constant and then declines. Supercritical fluid extrac-
tion involves controlling the solubility by manipulating temperature
and pressure. Natural materials contain multiple components mak-
ing solubility and extractability difficult to predict (Lira 1996). Vari-
ous mathematical aspects related to SFE of lipids were described by
King and List (1996) that include solubility, phase equilibria, mass
transfer, fractionation, and modeling.
The Goto et al. (1993) model for extraction of essential oil from
peppermint leaves has been successfully applied to many biological
materials. Solute is assumed to be extracted after being desorbed
from a porous biological solid substrate. During the process, diffu-
sion occurs inside its pores and the film surrounding the particle
offers mass-transfer resistance. This model was also applied to fatty
acids and carotenoid extractions from microalgae Spirulina maxima by
Canela et al. (2002). In their application, the mass of extract at the bed
outlet was described by the following equation:
⎡ ε ⎤ ε ⎧ ⎡ ⎛ t ⎞ ⎤
⎪
mt) = ⎢ p + ( −ε )⎥ p + ( −ε ) X ρ Q Aτ⎨ 1 ⎢ exp a ⎜ a ⎟− 1 ⎥
1
1
(
⎣ K p ⎦ K p 0 s CO 2 ⎩ ⎪ a ⎣ ⎝ 1 τ⎠ ⎦
1
⎡ ⎛ t ⎞ ⎤ ⎫ ⎪
+ 1 ⎢ 1−exp a ⎜ ⎟ ⎥⎬
a ⎣ ⎝ 2 τ⎠ ⎦ ⎪ (16.1)
⎦
2 ⎭