Phase Behavior of Ternary Mixtures of Water–Vanillin–Ethanol for

Apr 25, 2014 - We investigated phase behavior of a mixture solution involved in the vanillin extraction via a coarse-grained simulation method, dissip...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/jced

Phase Behavior of Ternary Mixtures of Water−Vanillin−Ethanol for Vanillin Extraction via Dissipative Particle Dynamics Ga Eun Son,† Nyambayar Sugartseren,† Won-Byong Yoon,‡ and Sang Kyu Kwak*,† †

School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulsan 689-798, Republic of Korea ‡ Department of Food Science and Technology, Division of Biotechnology, School of Bioscience & Biotechnology, Kangwon National Universtiy, Chuncheon, Gangwon 200-701, Republic of Korea ABSTRACT: We investigated phase behavior of a mixture solution involved in the vanillin extraction via a coarse-grained simulation method, dissipative particle dynamics (DPD). The ternary mixture, which consists of water, vanillin, and ethanol, is chosen for the study. Four phases depending on different compositions of three species have been formed; micellar, lamellar, transitional, and columnar phases. Ternary phase diagrams have been constructed at 298 K and 333 K. From the diagrams, we identified an optimal range of compositions, which satisfy two important criteria; one is the solubility of vanillin in ethanol and the other is the FDA regulation for the ethanol usage. The ratios of volume percents for the optimal range are 40:20:40 and 50:20:30 for water, vanillin, and ethanol, respectively, and the systems mostly have lamellar phases. Interestingly, those compositions well follow an economic perspective in which there is the least usage of ethanol and the most extraction of vanillin.

1. INTRODUCTION Accumulated vanillin as a glucoside in vanilla beans is an organic compound, which possesses characteristic aroma and flavor. Natural vanillin is extracted from cured (i.e., fermented by hydrolysis) seedpods of the vanilla orchid. It is one of the most widely used aromatic compounds with approximate global-market value of ∼180 million dollars.1 Natural vanillin, however, cannot meet worldwide consumption, so vanillin is also chemically synthesized from fossil hydrocarbons or wood pulp lignins. The price for natural vanillin is as high as approximately $(1000 to 4000)/kg1,2 depending on its quality; therefore, many researchers including food companies have been making continuous efforts to improve the efficiency of its extraction and the purity of the product. Several extraction techniques such as Soxhlet extraction,3 ultrasound-assisted extraction (UAE), 3,4 and microwave-assisted extraction (MAE)4 have been developed to overcome conventional extraction methods; vanilla extract was recovered by 29.06 wt % from MAE and 14.31 wt % from UAE in ethanol/water (40:60 vol %).4 Several extractants including aliphatic alcohols have been used.5 However, ethanol has been considered for use as a suitable solvent since the maximum extraction rate (i.e., 1.86 % by MAE and 0.99 % by UAE) of pure vanillin was achieved with this solvent.4 In addition, ethanol is inexpensive, edible, and easily handled for mixing and distillation processes. For these reasons we used ethanol as an extractant to make a ternary liquid mixture including water and vanillin for this study. Note that three components are necessary for a © XXXX American Chemical Society

fundamental understanding of the liquid−liquid equilibria (LLE). Like and unlike interactions of the components at the atomistic level result in the formation of different phases, which leads to phase separations comprising micellar, lamellar, transitional, and columnar structures. Such an example has been shown in the work of Sun et al., where the self-assembly of amphiphiles in which there is an increasing solute concentration leads to micelles, normal cylinder, lamellar, inverted cylinder phases in water, and some intermediate phases between those.6 For the statistical linkage from atomistic interactions to macroscopic phenomena, a computational approach can be a powerful tool; in the case of vanillin, PinoGarcia et al. has performed molecular dynamics (MD) to study the crystallization of vanillin, which is related to structure and surface chemistry.7 Thus, we use molecular modeling and simulation methods to elucidate phase behavior of the ternary mixtures depending on composition and temperature while focusing on the characterization of efficient phases in the vanillin extraction. In consideration of mesoscopic space and time scales as well as the study of LLE, we used a coarse-grained simulation method, dissipative particle dynamics (DPD), which is introduced by Koelman and Hoogerbrugge in 1992.8 Easy Special Issue: Modeling and Simulation of Real Systems Received: February 1, 2014 Accepted: April 16, 2014

A

dx.doi.org/10.1021/je5001186 | J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data

Article

momentum and produces the correct hydrodynamics at sufficiently long time and length scales. Those forces are given by

applicability of the method is possible after the appropriate modification by Groot and Warren9 is done on the theoretical connection of a simple functional form of the conservative repulsion to the Flory−Huggins parameter. Their approach has made DPD to be more widely acceptable to study the properties of the liquid meso-structures of complex systems10−12 including the estimation of solubility parameter and surface tension. So far, there have been few studies related to the vanillin extraction via simulation approaches, and to our knowledge, the ternary mixture system has not been investigated by the DPD or similar methods. By using DPD, therefore, our goal in this study is to report the phase landscape of water−vanillin−ethanol mixtures through the construction of the ternary phase diagram based on the volume fraction, and ultimately to provide the fundamental information on the best microscopic-phase involved in the vanillin extraction.

(4)

(5)

where aii is the repulsion parameter between same type of beads. χij can be obtained from the Hildebrand solubility parameter using Vbead/kBT (δi − δj)2, where Vbead is the average volume and δi and δj are solubility parameters of beads i and j. Since the repulsion parameter is temperature-dependent as shown by the case study from Mayoral and Ghoicochea,17 aij(T) has been obtained by estimating aii(T) and χij(T), where T is the temperature. 2.2. Simulation Details. To calculate solubility parameters of components, molecular dynamics (MD) based on the COMPASS force field (i.e., the Forcite MD module from Materials Studio 7.0) has been performed to obtain the cohesive energy density (CED), which is directly related to the heat of vaporization.18−21 The following equation shows the relation, δi(T ) =

ΔH v, i(T ) − RT Vm, i

=

CED (6)

where ΔHv,i(T) is the molar enthalpy of vaporization, R is the ideal gas constant, and Vm,i is the molar volume. We used 300 molecules for each component and ran the constant pressure (i.e., NPT, where N is the number of molecules and P is the pressure) MD for 100 ps to 200 ps with the time step of 1 fs at 298 K and 333 K. Note that the high temperature is chosen to make a similar thermal environment of the extraction experiment. After the equilibration of the systems, configurations of each system at same average densities, which are from equilibrated NPT simulations, were collected for other sets of constant volume (i.e., NVT, where V is the volume) MD simulations. Each system has been run for 2 ns with the Berendsen thermostat.22 The solubility parameters estimated at each temperature are shown in Table 1. Using eq 5, DPD repulsion parameters are calculated. The data are listed in Table 2 with physical units. The bead density, which decides an

all beads are governed by Newton’s equation of motion.14 The force acting on a bead i contains conservative (FCij ), dissipative (FDij ), and random forces (FRij ) and is given by8,9,14−16

∑ (FijC + FijD + FijR ) (1)

The conservative force is soft-repulsive and is represented by ⎧ ⎛ rij ⎞ ⎪ aij⎜1 − ⎟riĵ (rij < R c) ⎪ Rc ⎠ FijC = ⎨ ⎝ ⎪ ⎪ 0 (rij ≥ Rc) ⎩

FijR = σw R (rij)ξijriĵ

aij = aii + 3.27χij

Figure 1. Schematic of coarse-grained beads of (a) 8 water molecules (red and white atoms represent oxygen and hydrogen, respectively), (b) 3 ethanol molecules (gray atom represents carbon), and (c) one vanillin molecule. Colors of beads (i.e., red, green, and blue) are kept for each component through this work and diameters (i.e., 6.214 Å) of beads are same.

j≠i

(3)

where the caret r̂ represents the normalized element vector, and ωD is a weight function of the dissipative force, σ is a noise strength, and γ is the dissipation strength, ωR is a weight function of the random force, and ξij is the random variation with zero mean and variance 1. The conservative, dissipative, and random forces are all calculated within the cutoff distance. The conservative force is used to calculate interaction between nonbonded beads. The repulsion parameter aij depends on underlying atomistic interactions and is associated with Flory− Huggins (FH) interaction parameter χij9 through the following equation,

2. SIMULATION APPROACH 2.1. Simulation Method. In DPD, coarse-graining modeling is an important concept; several atoms (or molecules) are grouped together as a bead.13 Thus, the bead may contain characteristics of matters in a broad sense. In our system, actual volumes of each molecule are different such that water is 30 Å3, vanillin is 210 Å3, and ethanol is 90 Å3. Thus, beads of water, vanillin, and ethanol contain the number of molecules of 8, 1, and 3, respectively, and they roughly have a similar average volume 240 Å3. Note that masses of each bead do not vary much (i.e., (144, 152.04, and 138.21) amu for water, vanillin, and ethanol) and we used 144 amu for the mass of one bead. Figure 1 shows a schematic of the beads. In DPD,

fi =

FijD = −γωD(rij)(vij ·riĵ )riĵ

(2)

Table 1. Solubility Parameters Used in the DPD Simulation (Unit: (J/cm3)0.5)

where aij is a maximum repulsion between bead i and j in the unit of kBT/Rc. If rij exceeds the cutoff distance (i.e., Rc), the conservative force becomes zero. Note that the subscript ij represents a pair of beads. The dissipative and random forces are treated as heat sink and source, respectively, so the combined force works as a thermostat, which conserves B

T/K

water

ethanol

vanillin

298 333

46.696 43.653

25.936 22.992

24.737 23.261

dx.doi.org/10.1021/je5001186 | J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data

Article

Table 2. Repulsion Parameter Used in DPD Simulations* (Unit: kcal/mol·Å) water water ethanol vanillin

ethanol

vanillin

13.98a 15.63b 14.00a 15.63b

13.98a 15.63b

a

13.98 15.63b 19.42a 21.01b 20.06a 20.87b

*

The superscripts a and b are repulsion parameters at 298 K and 333 K, respectively.

interaction radius, is set to be 3 so that the side length of the box, which can contain 3 beads, becomes 8.963 Å (i.e., ∼720 Å3). Length scale and cutoff radius are same as the interaction radius in DPD. The simulation box size is 200 × 200 × 200 Rc3, which can hold a total 33457 beads. Time step is set to be 136.632 fs (i.e., 0.02 in DPD unit). To obtain equilibrated phases, we have run 106 DPD steps for each system as least twice.

3. RESULTS Table 1 lists the solubility parameters of three molecules depending on the temperature. If |δi − δj| < 2(J/cm3)0.5, two beads i and j are considered to be miscible.23 On that basis, water and vanillin are immiscible due to 21.959 at 298 K and 20.392 at 333 K and the values show high immiscibility at low temperature. On the contrary, vanillin and ethanol are miscible due to 1.199 at 298 K and 0.269 at 333 K and the values show high miscibility at high temperature. Figure 2 shows

Figure 3. Self-assembled (a) micellar phase of water, (b) cylindrical phase, (c) lamellar phase, (d and e) transitional phases, and (f) micellar phase of vanillin−ethanol. To show clear phases, the other component(s) are omitted. The numbers below each figure are vol % values of water−vanillin−ethanol.

enough to overcome the hydrophobic force suppressed by the others so that the columnar phase is still formed to minimize the interfacial area of hydrophobic beads.24 It is observed that out of all phases the lamellar phase shown in Figure 3c is stable in both temperatures. We found that the lamellar phase is favored when the ratio of water and VE beads is about half. When the two amounts are unbalanced, the phase is changed from lamellar to transitional as shown in Figure 3 panels d and e. We speculate that the transitional phase of water is formed when the amount is too small to make a thick layer, which is just to overcome the repulsive effect from the other component, and when the amount is too large to stay in the layer form, where the minimization of the interfacial area of hydrophobic beads occurs and causes the formation of columnar or micelle VE beads. If water is a much larger quantity compared to VE beads, the latter aggregates together to form a micelle (see Figure 3f). Note that in all systems under our consideration vanillin is not found in water except at interfaces, where vanillin mostly exists is in the ethanol side, and this is mainly due to the strong dissolution capability of ethanol. Therefore, we assumed the complete dissolution of vanillin into ethanol, and this argument is also supported by the similarity of solubility parameters shown in Table 1. We have systematically performed the DPD simulations of a total 36 ternary systems to construct the triangular phase diagram, which is shown in Figure 4. Under the water-rich environment, the aggregate morphology of VE beads is either micellar or columnar. As the amount of VE beads increases, a lamellar phase is gradually formed. Interestingly, we observed random formation of the columnar or transitional phase in the areas of 30 vol % and 70 vol % of water (see filled-purple circles in Figure 4). When the temperature increases, a few stable

Figure 2. Phase behaviors of self-assembled binary systems of (a) water (red)−vanillin (blue) (83.3:16.7 in vol %) and (b) ethanol (green)−vanillin (blue) (83.3:16.7 in vol %), respectively. Box sizes are 100 × 100 × 100 Rc3.

confirmatory results. In Figure 2a, vanillin is seen to aggregate into a micelle, but in Figure 2b ethanol and vanillin are well mixed. Thus, it is clearly shown that vanillin forms a micellar phase in the water-rich environment and a homogeneous ethanol−vanillin phase in the ethanol-rich environment. From Figure 3, several snapshots are obtained to show aggregate phases of vanillin extraction in ethanol−water mixtures. We categorized those to four different phases with respect to the aggregate morphology. Figure 3a shows a micelle of water beads. Under vanillin-and-ethanol (VE)-rich environment, water beads aggregate together to form a micellar phase; it is opposite to Figure 2a because of the reversed amount. Hydrophobic repulsion from a large content of the other components is very much strong in this system. When the volume ratio of water increased to 30 % water becomes columnar as shown in Figure 3b, but the amount of water is not C

dx.doi.org/10.1021/je5001186 | J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data

Article

Figure 4. Ternary phase diagrams at (a) 298 K and (b) 333 K, respectively, where filled and colored circles represent a total 36 systems under our investigation. Points A and B refer to 40:20:40 and 50:20:30 in vol %, respectively. Filled squares (black) represent experimental values of MAE and UAE methods from ref 4, and filled triangles (black) represent experimental values from ref 28.

points appear: from Figure 4b, filled-red circle at 70 vol % of water (i.e., columnar phase) and an additional filled-blue circle at 30 vol % of water (i.e., transitional phase). We have run extra repeating simulations to confirm the reproduction of data and speculated that unstable phases at these specific compositions are stabilized by thermal energy. To recognize composition range (or points) in the triangular phase diagram for an optimal extraction condition, two criteria have been set: the molarity of vanillin in ethanol (i.e., solubility of vanillin in ethanol at 298 K) is ∼2.47 M25 and the FDA regulation on the ethanol solution used in the vanillin extraction is between 35 vol % and 50 vol %.26,27 We have estimated molarities and found that the volume ratio of vanillin to ethanol should be about 1:2. Finally, two systems have been identified and indicated as A and B points in Figure 4. Their molarities are calculated to be 2.31 M and 2.77 M and the volumes of ethanol are 50 % and 37.5 % (i.e., 40:20:40 and 50:20:30 for water, vanillin, and ethanol), respectively. Note that the best point, which strictly satisfies two criteria, lies in the vicinity of the A and B points. We observed that the A point shows a lamellar phase and the B point shows a transitional phase at 298 K, but both become lamellar at 333 K. Experimental results from the MAE and UAE methods4 are also shown in Figure 4. Their vol % values are 44.3:26.2:29.5 and 52.4:12.6:35 (i.e., water, vanillin, and ethanol), respectively. Our results of points A and B are very close to those from MAE and UAE, and MAE has shown the higher extraction rate. On the basis of the vol % ratio of vanillin to ethanol, the extraction rates are 88.8 % and 36 % for MAE and UAE, respectively. Our data, which are bound by enforcing two criteria, are 50 % for A point and 66.7 % for B point, and the latter is close to the result from MAE. The difference is about 25 %. This value does not mean the error but mere comparison. Also, it must be emphasized that even though MAE shows the high extraction rate it did not follow the FDA regulation on the usage amount of the ethanol. Experimental results from the work of Kappatos et al.28 are also plotted in Figure 4, and the vol % values of ethanol used are larger than 50 %. The highest extraction rate is about 72 % when the volume ratio of 12.7:36.5:50.8 (i.e., water, vanillin, and ethanol) is used. In regard to the coarse-grained degree of components, the experimental results are well predicted by the DPD approach. Lastly, we have looked at A and B points from an economic viewpoint; more vanillin extraction with less ethanol. Figure 5 shows the volume percent of ethanol in the system and the percent ratio of vanillin to ethanol. Two data show iteratively decreasing and increasing behavior. Each point represents a system in the 36 systems. One should read the point (i.e.,

Figure 5. Plot of percentages vs point (i.e., the system number, total 36 systems). Open triangles refer to percent ratio of vanillin to ethanol in each system and open squares refer to vol % of ethanol in each system. Dashed A and B points represent the same A and B systems shown in ternary triangular diagrams in Figure 4. Filled symbols are for emphasis for the eye.

system) matching with down-to-right direction from the top point in the triangular phase diagram in Figure 4. The systems of the best economical interest might include 1, 3, 6, 10, 15, 21, 28, and 36 points from Figure 5. But, those are either unphysical or violate two criteria, which we used to find the optimal extraction condition previously. It is interesting to notice that our systems lie around the crossing sections, thus locating the crossing points may be the first guideline for the optimal extraction condition.

4. CONCLUSION Vanillin is a highly profitable product depending on its purity, thus process system-driven experimental studies have been intensively done for its best extraction. In this study, however, we have narrowed studies focused on macroscopic extraction processes down to a microscopic exploration into atomistic behaviors (i.e., phase behaviors) involved in the vanillin extraction in terms of composition and temperature. A mesoscale simulation method, DPD, with coarse-grained models, has been used for obtaining peculiar phase equilibria of the ternary mixtures of water, vanillin, and ethanol. To our knowledge, the DPD or similar methods on the vanillin-related study has not been done. Also, this work is uniquely D

dx.doi.org/10.1021/je5001186 | J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data

Article

(9) Groot, R. D.; Warren, P. B. Dissipative particle dynamics: Bridging the gap between atomistic and mesoscopic simulation. J. Chem. Phys. 1997, 107, 4423−4435. (10) Yuan, S.-L.; Cai, Z.-T.; Xu, G.-Y.; Jiang, Y.-S. Mesoscopic simulation study on phase diagram of the system oil/water/aerosol OT. Chem. Phys. Lett. 2002, 365, 347−353. (11) Dzwinel, W.; Yuen, D. A.; Boryczko, K. Mesoscopic dynamics of colloids simulated with dissipative particle dynamics and fluid particle model. J. Mol. Model. 2002, 8, 33−43. (12) Soto-Figueroa, C.; Rodríguez-Hidalgo, M.-d.-R.; Vicente, L. Mesoscopic simulation of micellar-shuttle pathway of PB−PEO copolymer in a water/[BMIM][PF6] system. Chem. Phys. Lett. 2012, 531, 155−159. (13) Qian, H.-J.; Lu, Z.-Y.; Chen, L.-J.; Li, Z.-S.; Sun, C.-C. Dissipative particle dynamics study on the interfaces in incompatible A/B homopolymer blends and with their block copolymers. J. Chem. Phys. 2005, 122, 184907. (14) Español, P.; Warren, P. Statistical mechanics of dissipative particle dynamics. Europhys. Lett. 1995, 30, 191−196. (15) Groot, R. D.; Madden, T. J. Dynamic simulation of diblock copolymer microphase separation. J. Chem. Phys. 1998, 108, 8713− 8724. (16) Groot, R. D.; Madden, T. J.; Tildesley, D. J. On the role of hydrodynamic interactions in block copolymer microphase separation. J. Chem. Phys. 1999, 110, 9739−9749. (17) Mayoral, E.; Goicochea, A. G. Modeling the temperature dependent interfacial tension between organic solvents and water using dissipative particle dynamics. J. Chem. Phys. 2013, 138, 094703. (18) Rigby, D.; Sun, H.; Eichinger, B. E. Computer Simulations of poly(ethylene oxide): Force Field, PVT diagram and cyclization behaviour. Polym. Int. 1997, 44, 311−330. (19) Sun, H. COMPASS: An ab initio force-field optimized for condensed-phase applications-overview with details on alkane and benzene compounds. J. Phys. Chem. B 1998, 102, 7338−7364. (20) Sun, H.; Ren, P.; Fried, J. R. The COMPASS force field: Parameterization and validation for phosphazenes. Comput. Theor. Polym. Sci. 1998, 8, 229−246. (21) Accelrys Materials Studio. http://accelrys.com/products/ materials-studio/index.html (Accessed January 31, 2014). (22) Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; DiNola, A.; Haak, J. R. Molecular dynamics with coupling to an external bath. J. Chem. Phys. 1984, 81, 3684−3690. (23) Gupta, J.; Nunes, C.; Vyas, S.; Jonnalagadda, S. Prediction of solubility parameters and miscibility of pharmaceutical compounds by molecular dynamics simulations. J. Phys. Chem. B 2011, 115, 2014− 2023. (24) Guo, X.; Zhang, L.; Qian, Y.; Zhou, J. Effect of composition on the formation of poly(DL-lactide) microspheres for drug delivery systems: Mesoscale simulations. Chem. Eng. J. 2007, 131, 195−201. (25) Bradley, J.-C.; Neylon, C.; Williams, A.; Guha, R.; Hooker, B.; Lang, A. S.; Freisen, B.; Bohinski, T.; Bulger, D.; Federici, M. Open Notebook Science Challenge: Solubilities of Organic Compounds in Organic Solvents. Nature 2010, DOI: 10.1038/npre2010.4243.3. (26) Waliszewski, K. N.; Ovando, S. L.; Pardio, V. T. Effect of hydration and enzymatic pretreatment of vanilla beans on the kinetics of vanillin extraction. J. Food Eng. 2007, 78, 1267−1273. (27) Waliszewski, K. N.; Pardio, V. T.; Ovando, S. L. A simple and rapid HPLC technique for vanillin determination in alcohol extract. Food Chem. 2007, 101, 1059−1062. (28) Kappatos, T.; Gordon, M. H.; Birch, G. G. Solution properties of vanillin and diacetyl in aqueous−ethanol solutions. Food Chem. 1996, 57, 275−282.

informative because of the construction of the triangular phase diagram depicted by micellar, columnar, lamellar, and transitional as well as oscillatory phases. From this study, characteristics of the chosen components are representative enough to capture major phenomena in phase separation, evidently shown by comparing computational results with those from experimental extraction results. We found that the best extract, based on the vanillin solubility in ethanol, the FDA regulation, and economical perspective, has the lamellar phase, which is mostly formed when water and VE beads are comparable in vol % (i.e., 40:60 to 60:40). Different ratios lead to other phases; the larger difference leads the system to the micellar phase. We finally remark that this study focused on the system in equilibrium, yet the exact interpretation of the extraction process may require the computational treatment on the nonequilibrium, which is reserved for future studies.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Funding

This work was supported by the basic science research program (Grant No.: 2013R1A1A2007491) through the National Research Foundation of Korea in the Ministry of Education, Science and Technology and the Korea CCS R&D Center (KCRC) grant funded by the Korea government (Ministry of Science, ICT & Future Planning) (NRF2013M1A8A1039968). Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS Computational resource has been provided by the High Performance Computing Center at UNIST (HPC-UNIST). REFERENCES

(1) Hansen, E. H.; Møller, B. L.; Kock, G. R.; Bünner, C. M.; Kristensen, C.; Jensen, O. R.; Okkels, F. T.; Olsen, C. E.; Motawia, M. S.; Hansen, J. De novo biosynthesis of vanillin in fission yeast (Schizosaccharomyces pombe) and baker’s yeast (Saccharomyces cerevisiae). Appl. Environ. Microbiol. 2009, 75, 2765−2774. (2) Cláudio, A. F. M.; Freire, M. G.; Freire, C. S. R.; Silvestre, A. J. D.; Coutinho, J. A. P. Extraction of vanillin using ionic-liquid-based aqueous two-phase systems. Sep. Purif. Technol. 2010, 75, 39−47. (3) Jadhav, D.; B.N, R.; Gogate, P. R.; Rathod, V. K. Extraction of vanillin from vanilla pods: A comparison study of conventional soxhlet and ultrasound assisted extraction. J. Food Eng. 2009, 93, 421−426. (4) Sharma, A.; Verma, S. C.; Saxena, N.; Chadda, N.; Singh, N. P.; Sinha, A. K. Microwave- and ultrasound-assisted extraction of vanillin and its quantification by high-performance liquid chromatography in Vanilla planifolia. J. Sep. Sci. 2006, 29, 613−619. (5) Reis, I. A. O.; Santos, S. B.; Santos, L. A.; Oliveira, N.; Freire, M. G.; Pereira, J. F. B.; Ventura, S. P. M.; Coutinho, J. A. P.; Soares, C. M. F.; Lima, Á . S. Increased significance of food wastes: Selective recovery of added-value compounds. Food Chem. 2012, 135, 2453−2461. (6) Sun, S.; Wong, J. T. Y.; Zhang, T.-Y. Effects of external electric fields on the self-assembly of phospholipids/water mixturesCoarsegrained molecular dynamics simulations. Soft Matter 2011, 7, 9307− 9310. (7) Pino-García, O.; Rasmuson, Å. C. Influence of additives on nucleation of vanillin: Experiments and introductory molecular simulations. Cryst. Growth Des. 2004, 4, 1025−1037. (8) Hoogerbrugge, P. J.; Koelman, J. M. V. A. Simulating microscopic hydrodynamic phenomena with dissipative particle dynamics. Europhys. Lett. 1992, 19, 155−160. E

dx.doi.org/10.1021/je5001186 | J. Chem. Eng. Data XXXX, XXX, XXX−XXX