RESULTS AND DISCUSSION
4.1 MODIFIED SOXHLET EXTRACTION (MSE)
4.1.1 Effects of Silica Gel to SODD Mass Ratio
A silica gel to SODD mass ratio less than 3 did not result in a successful separation.
Squalene was chosen as the indicator for the effective separation of polar and nonpolar lipids.
Table 4-1 shows that higher silica gel to SODD mass ratio results in lower recovery of FFAs,
TAGs, tocopherols and free phytosterols in NPLF. The amount of NPLF obtained per unit of SODD decreases because of the increasing silica gel to SODD mass ratio. At a silica gel to SODD mass ratio of 3, squalene and FASEs contents were significantly lower (p < 0.05) while the recovery of FASEs were significantly higher (p < 0.05) than those obtained while using a higher silica gel to SODD mass ratio. If it is desirable to concentrate squalene and FASEs into NPLF, then a lower silica gel to SODD mass ratio is preferred.Table 4-1 Effects of silica gel to SODD mass ratio on the composition of NPLF at an
extraction temperature of 65oC a.Compounds Silica gel
to SODD mass ratio
Squalene FASEs FFAs TAGs Tocopherols Free Phytosterols
Othersb
4.03±0.13c 7.17 ± 0.26 38.72±0.44 28.85±0.49 7.48±0.23 4.48±0.14 9.27±1.06 3
100.00d 97.51±0.74 41.47±1.46 43.42±0.96 48.80±0.10 20.12±0.04 57.94±8.51
4.81±0.06 7.80±0.09 30.60±0.07 15.19±0.59 7.78±0.09 3.51±0.09 30.31±0.66 4
100.0 80.79±0.21 28.68±0.59 16.02±0.22 46.74±1.32 19.40±0.03 68.00±1.90
5.61±0.23 8.50±0.21 30.37±0.59 1.85±0.22 7.94±1.32 3.12±0.03 42.61±1.90 5
100.00 71.02±1.12 20.94±0.34 3.28±0.32 35.60±5.65 18.57±0.56 72.73±3.57
7.26±0.24 9.32±0.24 18.85±0.27 3.30±0.22 7.25±1.45 4.81±0.07 49.21±2.02 6
100.00 64.03±3.82 11.71±0.23 2.31±0.08 28.75±4.82 17.61±0.32 73.22±5.45 Silica gel to SODD mass ratio
3 4 5 6 Amount, % e 47.28±3.68 33.87±0.12 31.58±0.34 22.45±0.75 Number of extractions,
cycles 40.00±1.00 40.00±1.00 70.00±1.00 80.00±1.00
Extraction times, h 2 2 3.5 4
a Each value represents the mean of three independent experiments.
bHydrocarbons, aldehydes, ketones, pesticides, herbicides, and the breakdown products of tocopherols and free phytosterols.
c Content, wt. %.
d Recovery= {(NPLF mass, g x content of the compound in NPLF, %) / (SODD mass, g x content of the compound in SODD, %)} x 100 %
e
As silica gel to SODD mass ratio decreases from 6 to 3, the recoveries of tocopherols, free phytosterols, FFAs, and TAGs increase from 28.75% to 48.80%, 17.61% to 20.12%, 11.71% to 41.47%, and 2.31% to 43.42%, respectively, while the recovery of FASE increases from 64.03%
to 97.6%. Tocopherols contents were not significantly different (p > 0.05) among all silica gel to SODD mass ratios studied. However, the content and recovery of FFAs and TAGs, along with the recovery of tocopherols and free phytosterols was significantly higher (p < 0.05) at a silica gel to SODD mass ratio of 3, compared to those obtained using higher silica gel to SODD mass ratio.
A higher silica gel to SODD mass ratio yielded a higher adsorption area available per unit mass of SODD, which resulted in a higher number of extractions, a poorer recovery of FASEs in NPLF, and less NPLF obtained. This result agrees with previous observation that the presence of a hydroxyl group on silica gel imparts a degree of polarity to the surface, so, polar molecules such as water, alcohols, phenols, and amines and unsaturated hydrocarbons are adsorbed preferentially over nonpolar molecules such as saturated hydrocarbons (Ruthven, 1984). The amount of FFAs adsorbed per unit area of silica gel was found to increase in proportional to the surface area of silica gel (Hau and Nawar, 1985), which is again consistent with the results shown in Table 4-1.
4.1.2 Effect of Extraction Temperature
A lower extraction temperature is favorable because it minimizes the degradation of bioactive compounds in SODD. Our study shows that the number of extraction required achieving 100% recovery of squalene in NPLF increases from 40 to 88 cycles as extraction temperature decreases from 65oC to -6oC. It can be seen from Table 4-2 that the amount of NPLF obtained, in term of a percentage of SODD, increases with extraction temperature. While the contents of squalene and FASEs were significantly different (p < 0.05) at –6oC from those at other temperatures, the recovery of FASEs was not significantly different (p > 0.05). The important criterion for a successful separation is to recover most FASEs in NPLF. As can be seen in Table 4-2, within the temperature range studied, the recovery of FASEs varies with extraction temperature. An extraction temperature of –6oC was chosen in this study even though at this temperature the squalene content in NPLF obtained was lower than that at 3oC. Both contents and recoveries of tocopherols, free phytosterols, FFAs and TAGs were significantly lower (p < 0.05) at –6oC than those obtained at 7, 55, and 65oC. Therefore, -6oC seems to be a better choice, since we prefer most tocopherols, free phytosterols, FFAs and TAGs to be concentrated in PLF.
Adsorption is usually an exothermic process, a decrease in extraction temperature yields a better separation of polar and nonpolar components, as can be seen in Table 4-2. These results agree with previous observations that the ultimate capacity of silica gel is generally higher at low temperature (Ruthven, 1984). In another study, Chu et al. (2004) found that a lower temperature led to higher vitamin E uptake at equilibrium, indicating that vitamin E adsorption by silica is an exothermic process. Moreira and Baltanás (2004) have shown that separation of free phytosterols and tocopherols is usually performed through a fractional crystallization because free phytosterols tend to precipitate at low temperature.
Our results suggest that under the following operation conditions: silica gel o SODD mass
ratio of 3; extraction temperature of –6oC; number of extractions of 88 cycles, the modified soxhlet extraction operated on SODD could yield a 3.4- and 3.3-fold increase in squalene and FASEs content in NPLF and a 1.3-fold increase in tocopherols content and 1.4-fold increase in free phytosterols content in PLF. These results are comparable with those obtained by employing molecular distillation. Table 4-3 shows that by using molecular distillation, a 3.5-fold increase in FASEs content could be concentrated in a high boiling point fraction, and a 1.4-fold tocopherols increase and 1.2-fold free phytosterols increase in contents, could be concentrated in a low boiling point fraction.
Table 4-2 Effects of extraction temperature on the composition of NPLF at silica gel to
SODD mass ratio of 3 aCompounds Extraction
temperature (oC)
Squalene FASEs FFAs TAGs Tocopherols Free Phytosterols
Othersb
4.03±0.13c 7.17±0.26 38.72±0.44 28.85±0.49 7.48±0.23 4.48±0.14 9.27±1.06 65
100.00d 97.51±0.74 41.47±1.46 43.42±0.96 48.80±0.10 20.12±0.04 57.94±8.51
5.16±0.19 8.95±0.39 43.32±3.37 20.03±1.43 5.82±0.31 2.27±0.15 8.98±1.11 55
100.0 95.11±1.82 36.28±3.93 23.54±0.93 29.71±2.59 7.96±0.79 70.37±9.55
5.77±0.33 9.29±0.91 44.64±4.61 18.09±1.84 6.08±1.89 0.78±0.31 15.37±4.24 7
100.00 94.99±0.69 34.08±2.68 26.92±3.46 24.99±5.80 3.12±0.51 66.46±9.67
6.29±0.05 12.19±0.30 35.05±2.77 3.49±1.80 2.39±0.51 0.41±0.14 40.19±0.06 -6
100.00 94.32±3.01 20.41±1.46 5.89±3.08 9.83±2.18 2.09±0.65 75.91±1.78 Extraction temperature, oC
65 55 7 3 -6 Amount, % e 47.28±3.68 34.76±0.93 35.00±2.23 26.32±4.76 28.84±0.21 Number of extractions,
Cycles 40.00±1.00 40.00±1.00 66.64±1.00 72.00±1.00 88.00±1.00
Extraction times, h 2 2 8 9 11
The composition of SODD sample I is shown in Table 3-1.
a Each value represents the mean of three different experiments .
bHydrocarbons, aldehydes, ketones, pesticides, herbicides, and the breakdown product of tocopherols and free phytosterols.
c Content, wt. %.
d Recovery= {(NPLF mass, g x content of the compound in NPLF, %) / (SODD mass, g x content of the compound in SODD, %)} x 100 %
e Amount = (NPLF mass, g /SODD mass, g ) x 100
The effectiveness of this separation was confirmed in subsequent experiments, carried out under the same conditions except with different SODD compositions (Table 4-4). From results obtained in this study, it can be seen that modified soxhlet extraction, which is much easier to operate than molecular distillation, can concentrate hydrocarbons and FASEs into NPLF. This method is capable of eliminating tocopherols, free phytosterols, FFAs, and acylglcerols from SODD by 87.99 – 95.68%, 97.26 – 99.76%, 76.13 – 82.11%, and 91.03 – 97.25%, respectively.
Table 4-3 Comparison of fractionation of SODD obtained by modified soxhlet
extraction and molecular distillationSeparation method
Compounds Modified soxhlet extraction a Molecular distillation b NPLF PLF Low b.p. fraction High b.p. fraction
Squalene 6.29c 100.00d NDe 0.00 N/Af N/A
FASEs 12.19 94.32 0.30 5.80 ND 0.00 45.4 100.73
FFAs 35.05 20.41 55.42 79.62 41.40 95.75 1.4 1.32
TAGs 3.49 5.89 22.71 94.55 ND 0.00 32.1 95.96
Tocopherols 2.39 9.83 8.88 90.17 14.80 99.05 0.70 1.90 Free phytosterols 0.41 2.09 7.67 98.09 12.80 86.41 4.00 11.03 Othersg 40.19 75.91 5.03 23.42 25.40 91.60 4.80 7.06
Amount, % h 28.84 71.16 69.60 28.40
The composition of SODD sample I is shown in Table 3-1.
The composition of SODD was used in Hirota et.al’s work is shown in Table 1-2
a Each value represents the mean of three independent experiments. Operating conditions: silica gel to SODD mass ratio = 3;
T = -6oC; and number of extractions = 88 cycles.
b Hirota et al. (4). Operating conditions: T = 250oC at 0.02 mmHg.
c Content, wt.%
d Recovery= {(fraction mass, g x content of the compound in fraction, %) / (SODD mass, g x content of the compound in SODD, %)} x 100 %
e Not detected
f Data not included in study
g Hydrocarbons, aldehydes, ketones, pesticides, herbicides, and the breakdown products of tocopherols and free phytosterols.
h Amount = (Fraction mass, g /SODD mass, g ) x 100
Nitrogen adsorption-desorption isotherm analysis was used to examine the characteristic of silica gels. As seen in Table 4-5, the specific surface area derived using Langmuir model is higher than that obtained by BET method. Since Langmuir and BET models are not based on the same assumptions, we would not expect to find an agreement between values determined by these two models. A lower BET surface area yielded a lower adsorption area available per unit mass of SODD, which resulted in a poorer recovery of FFAs in NPLF. This result agrees with previous observation that the amount of FFAs adsorbed per unit area of silica gel was found to decrease with decreasing surface area of silica gel (Hau and Nawar, 1985). Also, total specific pore volume of used silica gel dropped to 65 % of that of the initial silica gel. These results are not due to uncompleted removal of absorbed compounds during regeneration. No compounds were remaining in silica gel, as confirmed by the FTIR analysis.
Table 4-4 Compositions of NPLF obtained from SODD sample I, II, III, and IV.
aAmount, % e 28.84±0.21 27.03±0.90 28.00±1.06 25.16±1.10
The composition of SODD sample I, II, III and IV is shown in Table 3-1.
a Average of three independent experiments
b Content, wt.%
c Recovery= {(NPLF mass, g x content of the compound in NPLF, %) / (SODD mass, g x content of the compound in SODD, %)} x 100 %
d Hydrocarbons, aldehydes, ketones, pesticides, herbicides, and the breakdown products of tocopherols and free phytosterols.
e Amount = (Fraction mass, g /SODD mass, g ) x 100
Table 4-5 Characteristics of silica gel used in modified soxhlet extraction
Specific surface area (m2/g) Specific pore volume (cm3/g) Samplea Fresh silica gel without pretreatment (Data is given by company)
b Fresh silica gel with pretreatment at 150oC for 1 h (as starting material for 1st stage)
c Used silica gel with pretreatment at 150oC for 1 h (as starting material for 2nd stage)
d Used silica gel with pretreatment at 150oC for 1 h (as starting material for 3rd stage)