5. MAG and DAG synthesis in CO 2 saturated acetone
5.2.1 Solubility behavior in CO 2 -expanded acetone
The experimental system was designed with the purpose to easily collect samples from the expanded liquid phase. With the configuration of the studied system, 40mL of acetone were introduced as solvent. This volume was tested and confirmed to ensure the correct working of the sampling system in the whole range of investigated parameters.
Under the condition described in Figure 5.2, an increase of the input amount of palmitic acid up to 2500 mg will increase linearly the concentration of palmitic acid in the expanded liquid. As a consequence, it could be confirmed that at conditions of the previous experiment (in Chapter 4), the input amount of palmitic acid was totally dissolved in the reaction medium. On the other hand, it was observed that the concentration of fatty acid in the expanded liquid phase is ca. 120 times higher than in the gas phase (Table 5.2). For these reasons, the calculation method based on concentration in the expanded reaction medium is statistically significant. Brunner reported that the entrainer dissolves to a high degree in the liquid phase. Since the density of the
liquid phase is in general higher than that of the gas phase, stronger intermolecular forces are present which exert a greater effect on the distribution of the components than they do in the gasous phase [1]. Therefore, reactants like FFAs and their derivatives are preferentially hold in the liquid phase instead of the gasous phase in the CO2 saturated acetone system.
R2 = 0.9969
0 5 10 15 20 25 30
0 500 1000 1500 2000 2500 3000
Acid pamitic input (mg)
Concentration (mg/ml)
Figure 5.2: Loading of palmitic acid in the CO2 expanded acetone.F15
Table 5.2: Fatty acid distribution in the expanded liquid phase and the gas phase during the reaction.F16
Reaction time (min)
Molar ratio of palmitic acid (expanded liquid phase / gas phase)
120 123 180 119 240 126 300 144
15 P=65 bar, T=50°C, 40mL Acetone
16 P=65 bar, T=60°C, 40mL Acetone, 300mg palmitic acid, 300mg glycerol, 45mg Novozyme 435
b) Solubility of a mixture
A mixture of FFA, MAG and glycerol was used to test the loading capacity of a mixture in CO2 expanded acetone. Table 5.3 shows that FFA can be loaded up to 21.9-26.3 (mg/mL), while MAG and glycerol can be dissolved as 11.7-13.7 (mg/ml) and 0.5-0.6 (mg/ml) respectively. Solubility of glycerol is much less than those of MAG and FFA in CO2 saturated acetone.
Table 5.3: Loading capacity of a mixture of FFA, MAG and glycerol.17 Temperature
(°C)
Pressure (bar)
FFA (mg/ml)
MAG (mg/ml)
Glycerol (mg/ml)
40 65 26.28 13.70 0.48
60 65 25.18 11.74 0.57
60 85 21.94 11.68 0.47
c) Effect of process parameters on reactant concentration
The effect of physical properties of the medium on the reactant concentration in the reactor had been discussed in Chapter 4. It was proven that the amount of acetone, pressure, and temperature alter solubility, density, or the expanded volume, therefore, changing the concentration of palmitic acid. This could be confirmed by using the new developed system. In this study, palmitic acid in CO2 expanded acetone, which directly surrounds the active sites of the enzyme, could be withdrawn and analyzed. Within the experimental conditions, concentration of palmitic acid has a strong relationship with the expanded volume of the saturated liquid. Figure 5.3 shows that an increase in pressure from 65 to 70 bar, decreases the palmitic acid concentration from 3.1 (mg/ml) to 2.3 (mg/ml). Zevnik [145] reported that an expansion of a liquid can be controlled by pressure only to a limited expansion. Up to a certain degree, a small deviation in pressure can cause a great variation of the expanded volume. However, there is only a slight change in concentration when pressure increases from 80 to 100 bar, since the volume of the gas phase is very much reduced. The expanded liquid occupied almost 100% total volume of the reactor.
17 40mL Acetone, 3000 mg mixture of FFA (~70%), MAG (~20%) and glycerol (~10%)
0 0.4 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4
0 50 100 150 200 250 300 350
Reaction time (min)
Acid palmitic (mg/ml)
65bar 70bar 80bar 100bar
average average average average
Figure 5.3: Concentration of palmitic acid in CO2 saturated acetone.F18
5.2.2 BScreening the MAG and DAG synthesis reaction
The screening process was conducted in a water bath and in a high pressure reactor. Table 5.4 shows total conversion at different amounts of enzyme (related to the amount of dissolved palmitic acid). The introduced amount of palmitic acid to the reactor was fixed at 300mg. This amount agrees with the loading capacity of CO2 saturated acetone to ensure all the reactants and final products can be dissolved in the reaction medium.
The results show that, reactions in CO2 expanded acetone are much faster than those in a water bath without support of CO2. It could be calculated from these results that the initial reaction rate can be accelerated up to more than 12 times when acetone is expanded with CO2. It required 24 hours for reactions in the water bath to reach the same level, compared to 5 hours in the high pressure CO2-acetone system.
18 T=50°C, 40mL acetone, 300mg acid palmitic, 210mg glucose
Table 5.4: Esterification of MAG and DAG with and without high pressure CO2 support.F19 Temperature
(°C)
Pressure (bar)
Enzyme (%)
Reaction Time (h)
Conversion [%]
50 0 25 3 21.2
5 28.1
24 68.0
50 0 15 3 15.7
5 21.1
24 59.2
50 0 5 3 5.5
5 7.8
24 32.9
50 85 25 1 41.6
5 71.6
24 79.7
60 85 15 1 18.1
3 56.2
5 69.5
40 75 5 1 19.7
3 29.8
5 42.5
5.2.3 Reaction progress and reaction kinetics
Reaction time curves for the esterification of glycerol and palmitic acid with Novozyme 435 in CO2 saturated acetone are presented in Figure 5.4. The first-order exponential increase in product concentration, as described in section 2.2.2, was used to model the progress curves. The results show that the experimental data are represented well with this model (Appendix B2). It is
19 40 ml acetone, 300 mg acid palmitic, 300 mg glycerol, Novozyme 435
also observed that the reaction rate decreased after 5h. Conversions after 24h is not far higher than that obtained after 5h in CO2 saturated acetone.
0 5 10 15 20 25
0 5 10 15 20 25 30
Reaction time (h) Reacted palmitic acid (Mmol/g enzyme)
Figure 5.4: Reaction progress curve at different processing condition.F20
(■) 85 bar, 50°C, 25% enzyme; (♦) 85 bar, 55 °C, 23% enzyme, (…..) modeling curves.
5.2.4 Effect of enzyme type
As mentioned in the previous parts, properties of the enzyme showed a strong effect on enzyme catalysis in acetone. In this work, different types of enzymes were screened to select the most suitable for the esterification of glycerol and palmitic acid in CO2 expanded acetone.
Lipomod 34P, Novozyme 435, RM IM, and TL IM were chosen. The total conversion of palmitic acid over the reaction time is shown in Figure 5.5. The results show that for this type of esterification, Novozyme 435 was also the best, followed by RM IM, TL IM and Lipomod 34P.
Under the described conditions, Novozyme 435 has a conversion of 50% after 5 hours, while the other enzymes can not convert more than 26%. Moreover, the initial velocity of Novozyme 435 reached 7940 (μmol/h/g enzyme) compared to 5679, 2799 and 2665 of RM IM, TL IM and Lipomod 34P respectively. Therefore, Novozyme 435 was selected as the catalyst for the esterification of palmitic acid and glycerol in the following studies.
20 40 ml acetone, 300 mg palmitic acid, 300 mg glycerol, Novozyme 435
0 10 20 30 40 50 60 70
60 180 300
Reaction time (min)
Conversion (%)
Novozyme 435 RM IM TL IM Lipomod 34P
Figure 5.5: Screening of different lipases for MAG and DAG synthesis.F21
5.2.5 BEffect of substrate ratio
Substrate ratio is one of the most important factors affecting enzymatic catalysis. Figure 5.6 shows the effect of the molar ratio of glycerol to palmitic acid on the synthesis of mono- and di-palmitin. Glycerol concentration (mol %) was varied from 1 to 6 times higher than the palmitic acid concentration, which was kept constant at 1.2 mmol (300 mg). It was originally thought that an increase in glycerol would promote the formation of MAG due to presence of excess glycerol.
From the results, it can be observed that an increase of the glycerol ratio to 3 molar equivalents increases the conversion of palmitic acid to mono-palmitin. However, a further increase of glycerol has no effect on the formation of MAG. On the other hand, the formation of DAG increases with molar ratio. The higher the concentration of glycerol is, the more di-palmitins are formed.
21 P=75bar, T=50°C, 40 ml acetone, 300 mg palmitic acid, 300 mg glycerol, 45 mg enzyme
It can be seen that the mass-transfer was improved with the acetone expanded by CO2. Kwon reported that a high glycerol concentration resulted in a high viscosity, decreasing the mixing efficiency and therefore decreased the esterification efficiency [110].
0 20 40 60 80
1:1 1:3 1:6
Molar ratio ( palmitic acid : glycerol )
Reacted palmitic acid (%)
MAG DAG
Figure 5.6: Conversion of palmitic acid to MAG and DAG at different substrate ratios.F22
In another research, Moquin et al. show that changing the glycerol molar ratio changes phase behavior. Equilibrium concentration of MAG significantly decreased with a decrease in initial glycerol concentration [124]. Therefore, a higher glycerol molar equivalent supports a higher MAG concentration around the active sites of the enzyme and favors the condition to create more DAG. For further investigation, the substrate ratio of 1:3 was selected as an optimum to synthesize a glyceride mixture enriched in MAG.
5.2.6 BEffect of adding water
In the previous chapter, water was proven to have an effect on the degree of esterification of glucose and palmitic acid in CO2 saturated acetone. In this study, by introducing various amounts of water to the reaction medium, the effect of water content on the esterification of glycerol and palmitic acid was also investigated. It was assumed that selectivity is very sensitive to the equilibrium state, which can be shifted by changing the amount of introduced water. The
22 P=75bar, T=50°C, 5h, 40 ml acetone, 300 mg palmitic acid, 15% Novozyme 435
effect of adding water on the total conversion and the selectivity to MAG and DAG is presented in Figure 5.7. The results show that the synthesis of MAG is not favored with a high water content. A drastic decrease in conversion to MAG is observed with an increase in added amount of water. However, the conversion to DAG increases with the amount of water added. Obviously, the increase of water shifts phase equilibrium towards conditions more favorable for DAG. When the concentration of MAGs increases, there are more available MAGs to be converted to DAG.
When too much water is introduced, total conversion is reduced, because equilibrium of the reaction is shifted toward the hydrolysis of the ester, and the enzyme can get agglomerated [34, 147]. The enzymes in this case still remain in the bottom of the reactor and are therefore directly in contact with the heavy phase formed with excessive water.
0 10 20 30 40 50 60 70
0 2 4 5 8
Added water [Vwater/Vacetone , %]
Reacted palmitic acid (%)
MAG DAG
Figure 5.7: Conversion of palmitic acid to MAG and DAG at different water levels.F23
Selectivity of the synthesis reaction can be also observed in Figure 5.7. Without adding water, the molar ratio of MAG:DAG in the final products was about 11:1. However when 5%
(Vwater/Vacetone) water was added, this ratio changed to 1:11. As mentioned previously, by applying a specific pressure and temperature, the water level in the phase of CO2 expanded acetone can be kept constant, therefore helping to choose the right selectivity to MAG and DAG.
23 P=75bar, T=50°C, 5h, 40 ml acetone, substrate ratio of 1:3 with 300 mg palmitic acid, 15% Novozyme 435
For industrial application, this is of importance because the ratio of DAG to MAG in the product influences physico-chemical properties of the emulsifiers. Together with the specification of the enzyme, it can open a new way to obtain a desired DAG to MAG ratio by controlling water distribution through pressure and temperature applied.
5.2.7 Response surface analysis of MAG and DAG synthesis
The response surface method was designed to optimize the process to produce MAG and DAG. The regression result of the model is presented in Appendix B3. Statistical results indicated a very strong relationship between the model and experimental data (R2=0.95). The estimated regression coefficients from Equation 5.4 are presented in Table 5.5. The most significant factors are pressure, enzyme concentration and temperature-pressure interaction.
Table 5.5: Estimated regression coefficients from second-order model of relationship between response variable (Y) and independent variables (X1, X2, X3).
Factor Coefficient Std. Error Significant
B0 59.25 4.96627816 0.001
B1 10.30 2.48313908 0.0143
B2 -2.81 2.48313908 0.3210
B3 14.19 2.48313908 0.0046
B11 -6.91 3.92618762 0.1531
B22 -6.16 3.92618762 0.1915
B33 -3.93 3.92618762 0.3731
B12 12.02 3.51168896 0.0267
B13 4.42 3.51168896 0.2768
B23 2.66 3.51168896 0.4906
As a result, the final Equation 5.5 for the total conversion of palmitic acid is obtained as:
Y= 59.25 + 10.30 * X1 - 2.81*X2 + 14.19 * X3 - 6.91*X12- 6.16*X22
- 3.93*X32 + 12.02*X1*X2 + 4.42*X1*X3 + 2.66*X2*X3 (5.5)
The response surface developed from this equation is presented as Figure 5.8. The developed response surfaces are curved surfaces in a multi-dimensional space, which describe the dependence of the chosen response variable on all the process variables. In this case, it means that the response Y (the conversion of palmitic acid) can be observed as a function of pressure (X1), temperature (X2), and enzyme concentration (X3). The actual experimental data and the predicted values are presented in Table 5.6. The results showed that the synthesis reactions were in favor for MAG. It was observed that MAG molar content could be up to 80 – 100% in the final glyceride mixtures.
Table 5.6: Experimental data vs. predicted values
Exp Nr.
Pressure [bar]
Temp.
[°C]
Enzyme [%]
MAG Yield (%)
DAG Yield (%)
Conversion [%]
Predicted Conversion
[%]
1 85 (1) 50 (0) 5 (-1) 33.47 0 33.47 40.10
2 75 (0) 40 (-1) 25 (1) 42.43 7.94 58.31 63.48 3 85 (1) 60 (1) 15 (0) 53.67 6.73 67.14 65.68 4 75 (0) 50 (0) 15 (0) 50.96 4.47 59.90 59.25 5 85 (1) 40 (-1) 15 (0) 44.16 3.83 51.82 47.26 6 85 (1) 50 (0) 25 (1) 58.68 9.62 77.91 77.30 7 65 (-1) 40 (-1) 15 (0) 49.23 0 49.24 50.70
8 65 (-1) 50 (0) 5 (-1) 27.72 0 27.72 28.33
9 65 (-1) 50 (0) 25 (1) 52.74 0.88 54.49 47.86 10 65 (-1) 60 (1) 15 (0) 10.91 2.79 16.47 21.03 11 75 (0) 50 (0) 15 (0) 52.48 3.06 58.59 59.25 12 75 (0) 40 (-1) 5 (-1) 42.50 0 42.50 40.44 13 75 (0) 60 (1) 5 (-1) 33.36 0.65 34.66 29.49 14 75 (0) 60 (1) 25 (1) 55.55 2.79 61.12 63.19
Figure 5.8: 3D–Plot for the conversion of palmitic acid (PA) vs. pressure, temperature and enzyme concentration.F24
a) Effect of process variable
Effect of pressure: As mentioned in the previous section, pressure affects the physico- chemical properties of the reaction system. Therefore, changing pressure will change the conversion of palmitic acid. Figure 5.9 shows that, at a low temperature, pressure has not a strong effect. However, the pressure effect is more pronounced at a higher temperature. In Figure 5.10, pressure also shows a strong effect at a high enzyme concentration, but not at a low one. In general, the synthesis requires a high pressure for a better reaction.
Effect of temperature: Figure 5.9 shows the combined effect of temperature and pressure.
At a high pressure, higher temperature will give a better conversion. Effect of temperature is more pronounced at low pressure than at a higher one. However, at low pressure, a low temperature is more favorable to the synthesis. It could be observed that a high temperature is not always required to obtain a good reaction. In some cases, the optimum temperature is not the highest one (Figure 5.11).
24 40 ml acetone, substrate ratio of 1:3 with 300 mg palmitic acid, 5h
Temperature Pressure
Enzyme concentration
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1-1 -0.85 -0.7 -0.55 -0.4 -0.25 -0.1 0.05 0.2 0.35 0.5 0.65 0.8 0.95
Temperature
Pressure
20-27 27-34 34-41 41-48 48-55 55-62 62-69 69-76
Figure 5.9: Conversion of palmitic acid vs. pressure and temperature.F25
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1-1
-0.85 -0.7 -0.55 -0.4 -0.25 -0.1 0.05 0.2 0.35 0.5 0.65 0.8 0.95
Enzyme
Pressure
20-28 28-36 36-44 44-52 52-60 60-68 68-76 76-84
Figure 5.10: Conversion of palmitic acid vs. pressure and enzyme.26
25 15% enzyme, 5h, 40 ml acetone, substrate ratio of 1:3 with 300 mg palmitic acid
26 T=50°C, 5h, substrate ratio of 1:3 with 300 mg palmitic acid
(%)
(%)
Effect of enzyme concentration: Within the experimental range, enzyme concentration has a positive effect on total conversion (Figure 5.10 and 5.11). The higher the concentration of the enzyme is, the better is the reaction. This effect is more pronounced at high pressure and high temperature than at lower values.
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1-1
-0.85 -0.7 -0.55 -0.4 -0.25 -0.1 0.05 0.2 0.35 0.5 0.65 0.8 0.95
Temperature
Enzyme
20-27 27-34 34-41 41-48 48-55 55-62 62-69 69-76
Figure 5.11: Conversion of palmitic acid vs. enzyme concentration and temperature.F27
b) Optimum condition
As observed, the response surfaces allow to obtain the optimum combination of process variables. The result is that there are many combinations of pressure, temperature, and amount of enzyme to obtain a good conversion. Table 5.7 presents two possible conditions, which could be used as optimum points of the predicted values from the response surface model and the real experimental data. A pressure of 85 bar, a temperature of 50°C, and an enzyme concentration of 25% are the final suggested optimum conditions for the esterification of glycerol and palmitic acid in CO2 expanded acetone.
27 P=75bar, 5h, 40 ml acetone, substrate ratio of 1:3 with 300 mg palmitic acid (%)
Table 5.7: Predicted and experimental data at optimum points.
Pressure (bar)
Temperature (°C)
Enzyme concentration (%)
Predicted conversion (%)
Experimental conversion (%)
85 55 20 74.7 71.1
85 50 25 77.9 77.3
5.2.8 Screening reactions with other types of fatty acids
The esterification in CO2 expanded acetone is a base to develop diversified MAG and DAG products. The fatty acid moiety can be defined from different number of carbon atoms and degree of saturation, different in functional and nutritional properties.
Products from separation of free fatty acids are a mixture of short and long chain, saturated, and unsaturated acids. Therefore, it is more practical to prove the application of esterification in CO2 expanded acetone with a fatty acid mixture.
Figure 5.12 shows the degree of esterification of a fatty acid mixtureF28 vs. reaction time. It is observed that the optimum condition reported in Table 5.7 for esterification of pure palmitic acid could not be applied for a fatty acid mixture. Yang [116] showed different effects of the FFA chain length (C8 to C22) and the number of double bonds (0 to 6) on MAG synthesis. As a result, it is recommended that an experiment must be conducted to obtain the optimum conditions for an individual fatty acid substrate in the high pressure CO2-acetone system. However, based on the presented results, it is suggested that the esterification in CO2 saturated actone has a potential to synthesise valuable esters from a mixture of fatty acids.
5.2.9 Enzyme stability
As proven in section 4.2.11, the enzyme can be re-used after processing with acetone-CO2
at 65 bar. In this study, the enzymes recycled from reaction at a pressure of 85 bar, were checked by comparing their activity to fresh enzymes. Table 5.8 shows the conversion of fresh and recycled enzymes. The results show that the activity and stability of the enzyme were retained after the reaction in CO2 expanded acetone. It was calculated that under the operating conditions,
28 fatty acids w. 1 chain > C18, ≤4.0%; linoleic acid, ≤18.0%; linolenic acid, 4.0%; margaric acid, ≤4.0%; myristic acid, ≤5.0%; oleic acid, 65.0-88.0%; palmitic acid, ≤16.0%; palmitoleic acid, ≤8.0%; stearic acid, ≤6.0%