- Email: [email protected]
Micronization of vanillin by rapid expansion of supercritical solutions process
Journal of CO₂ Utilization 21 (2017) 169–176
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
Micronization of vanillin by rapid expansion of supercritical solutions process
MARK
⁎
A. Montesa, , R. Merinoa, D.M. De los Santosb, C. Pereyraa, E.J. Martínez de la Ossaa a b
Department of Chemical Engineering and Food Technology, University of Cádiz, International Excellence Agrifood Campus (CeiA3), 11510 Puerto Real, Cádiz, Spain Department of Physical Chemistry, Faculty of Sciences, University of Cádiz, International Excellence Agrifood Campus (CeiA3), 11510 Puerto Real, Cádiz, Spain
A R T I C L E I N F O
A B S T R A C T
Keywords: Vanillin RESS process Microparticles Supercritical
Uniform microparticles of vanillin were precipitated by the Rapid Expansion of Supercritical Solution (RESS) process and this reduced the commercial particle size by a factor of one hundred. A design of experiments full factorial at two levels was performed. The effects that pressure, temperature, contact time and nozzle diameter had on the particle size and yield of the RESS precipitation were evaluated. Pressure and temperature were the factors that had the most marked effects on particle size and yield. The use of higher pressure and temperature is recommended to obtain the smallest particle size and the highest yield. However, at lower pressure the temperature is a crucial factor in that the use of a lower temperature led to considerably smaller particles and good yields whereas a higher temperature gave the highest particle size and the lowest yield. A higher contact time and smaller nozzle diameter led to slight improvements in the vanillin particle size and yield according to the results of the design. The crystallinity of the RESS-processed vanillin was unaltered when compared to the unprocessed material.
1. Introduction Vanillin (4-hydroxy-3-methoxybenzaldehyde) (C8H8O3) is the main component (80–90%) [1] of vanilla orchid pods and it has an attractive flavor and fragrance. The current global demand for vanillin is estimated to be roughly 20,000 tons per year [2] and it is becoming one of the most important aromatic substances. Vanillin is often used as a flavoring agent, a food preservative, in beverages, cosmetics and drugs due to its antioxidant [3–5] and antimicrobial [6–8] activities and it is generally regarded as a safe (GRAS) substance [9]. Vanillin is also used as a crosslinker and in the preparation of a wide range of vanillin-based renewable polymers, e.g., phenolic, epoxy and benzoxazine resins, polyesters, and acrylate and methacrylate polymers are in the focus of numerous investigations [10,11]. Vanillin has been widely used as the commercial product (∼700 μm particle diameter) and this has low solubility in cold water, although the solubility increases with temperature. However, this solubility could be improved by reducing the particle size to the micrometer range. There is very little literature concerning the micronization of vanillin although composites and microencapsulates of vanillin has been obtained: microcapsules containing limonene and vanillin with a mean particle diameter of 30 μm [12] were prepared by Pitol-Filho et al. using ethyl cellulose as a coating agent by dropping the polymeric
⁎
Corresponding author. E-mail address: [email protected] (A. Montes).
http://dx.doi.org/10.1016/j.jcou.2017.07.009 Received 24 March 2017; Received in revised form 1 June 2017; Accepted 4 July 2017 2212-9820/ © 2017 Elsevier Ltd. All rights reserved.
solution into a coagulation bath, which contained water, sodium dodecyl sulfate, and acetic acid in different concentrations. Noshad et al. carried out the microencapsulation of vanillin with maltodextrin using a spray dying method and this process gave a wide particle diameter range (0.7–128 μm) [13]. A maximum encapsulation efficiency of 58.3% and minimum particle size of 6.95 μm were achieved at 184 °C with a maltodextrin concentration of 8.5% and vanillin concentration of 0.36%. The majority of the micronization methods involve the use of excessive amounts of organic solvent, high temperatures and post-processing steps and the final products often contain residues. These drawbacks could be circumvented by using supercritical CO2 (scCO2) technologies due to the low toxicity of this compound and the absence of organic solvent in the process, thus contributing to a substantial improvement in human health and environmental considerations. The application of supercritical fluid technology in micronization processes requires a study of the phase equilibrium formed by the solute and the supercritical CO2, i.e., the fluid-solid phase equilibrium. Binary data for the solubility of vanillin in CO2 have previously been reported in the literature [14,15]. Molar fractions from 0.014 to 1.295 × 10−2 were obtained in the pressure range 80–300 bar and the temperature range 313–353 K, which shows that vanillin is quite soluble in supercritical CO2. In this way, the Rapid Expansion of Supercritical Solutions
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
there is a stainless steel collection vessel (V2) in which particles are precipitated once the supercritical solution is expanded through a stainless steel nozzle with an inner diameter in the range 100–200 μm (Thar Technologies). The equipment was described in detail in a previous publication [20]. The experiments were performed as follows: Firstly, the commercial vanillin powder (0.5 g) was placed in the solubilization chamber. The system was held for a certain time once supercritical conditions for CO2 (pressure and temperature) had been achieved in order to ensure complete equilibration. Valve MV2 was then opened and the supercritical solution was expanded through a nozzle, which was pre-heated in order to compensate for the heat loss and to prevent the nozzle from clogging during the fast expansion. The precipitated particles were collected on the wall of vessel V2 for subsequent analysis.
(RESS) process could be used to micronize this compound. In the RESS process a given amount of vanillin is placed into a solubilization chamber, which is filled with scCO2 to give a supercritical solution after a certain time. The subsequent sudden expansion of this solution through a nozzle to atmospheric conditions generates a high level of supersaturation and produces rapid nucleation and the precipitation of particles on the vessel wall. The work described here concerned a study of the micro-precipitation of vanillin by the RESS process and a design of experiments was applied in order to identify the main parameters that affect the process in terms of particle size, size distribution and yield. The resulting vanillin microparticles were analyzed by X-ray diffraction to evaluate the degree of crystallinity after the supercritical process. 2. Materials and methods
2.3. Sample characterization
Vanillin (4-hydroxy-3-methoxybenzaldehyde) (C8H8O3, 99% purity) was purchased from Sigma-Aldrich (Spain). CO2 with a minimum purity of 99.8% was supplied by Linde (Spain).
Scanning electron microscopy (SEM) images of the powder precipitated on the wall of the vessel were obtained using a NOVASEM scanning electron microscope. Prior to analysis the samples were placed on carbon tape and then covered with a coating of gold using a sputter coater. Particle size distribution (PSD) and mean particle size of the processed samples were measured using DLS technology (Nano-ZS, Malvern Instruments, UK). This technique measures the diffusion of particles moving under Brownian motion, and converts this to a size and a size distribution using the Stokes–Einstein relationship. NonInvasive Back Scatter (NIBS) technology was incorporated to give the highest sensitivity along with the highest possible size and concentration range. The measurement range was from 0.3 nm to 10 μm. Prior to analysis, the samples of vanillin were weighed and suspended in hexane (1 mg/mL) with one drop of surfactant (Tween 80). The PSD and mean particle size of commercial vanillin could not be processed by this technique due to its large size, which gave rise to an unreliable measurement. In this case, an LS 13 320 laser diffraction particle size analyzer from Beckman Coulter was used. This system was able to measure particles up to 2000 μm. Polarization Intensity Differential Scattering (PIDS) technology was employed in which the particle size distribution could be determined by measuring the variations between the horizontally and the vertically scattered light for each wavelength. In this case the sample was introduced as a solid powder using a module (Tornado) that could handle dry powder. X-ray diffraction (XRD) analysis was performed on Bruker D8 Advance diffractometer to determine the amorphous or crystalline nature of the precipitate obtained by both processes. All diffraction patterns were scanned from 10° to 80° in 2θ angle with a step size of 0.02° and one second as the step time.
2.1. Experimental design A design of experiment (DOE) was carried out in order to identify the critical factors in the vanillin miconization by the RESS process. A full factorial design at two levels was performed. Four factors (24) were studied but the complete design consisted of 19 experimental points that included 16 factor points and three replications at the center point (experiments 17–19). The critical factors were selected with appropriate ranges for this design. The responses of the design were mean particle size and yield. Yield was calculated by subtracting the weight of powder precipitated from the initial weight of vanillin placed into the solubilization chamber (0.5 g). Other full factorial designs have been successfully applied for screening purposes [16–19]. Pressure (P), temperature (T), nozzle diameter (Øn) and contact time (t) were identified as possible parameters that could influence the particle size and size distribution of vanillin in the RESS process. The two levels for each factor are shown in Table 1. The levels of pressure and temperature were selected to evaluate the influence of different supercritical CO2 densities and therefore solvent power. Nozzle diameters were selected according the those available from the equipment supplier. The contact times were selected in order to ensure that there was sufficient time to form a homogenous supercritical solution and to keep the costs reasonable. 2.2. RESS process The experiments listed in Table 2 were carried out in a pilot plant developed by Thar Technologies® (model RESS250). A schematic diagram of this equipment is shown in Fig. 1. Pressures of 100–300 bar, temperatures of 313–343 K, nozzle diameters of 100–200 μm and contact times of 1–2 h were evaluated. The RESS250 equipment comprises a high-pressure pump (P1) to fill a 250 mL stainless steel solubilization vessel (V1) with CO2. This vessel is heated by an electrical heating jacket (V1-HJ1) that surrounds the vessel. Supercritical CO2 and vanillin were continuously mixed with a magnetic stirrer (maximum speed 2500 rpm). At the end of the line
2.4. Statistical analysis Each experiment was carried out following the order established by the design. Influence of each factor on the particle size and yield of vanillin was statistically assessed by ANOVA with 95% confidence level. Data obtained from the partial least square (PLS) regression were statistically analyzed using ANOVA for the response variable in order to test the model significance (p < 0.05) (Table 3). The Modde (Version 5.1, Unmetrics, USA) software was employed to analyze the results and optimize the experimental conditions. Significance of pressure, temperature, nozzle diameter and contact time on particle size and yield were established at 95% of confidence level (Table 4).
Table 1 Two – Level assessment for each factor and calculated effects on PS. Factors
Low level
High level
PS Effects
Yield Effects
P (bar) T (K) t (h) Øn (μm)
100 313 1 100
300 343 2 200
−6.49 5.29 −1.37 2.41
53.37 −15.87 −3.62 −1.12
3. Results and discussion 3.1. Analysis of the design of experiments The values for the coefficient of determination R2 (adjusted and experimental), degree of freedom (DF), sum of squares (SS) and mean 170
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Table 2 Experimental design and observed responses. Run name
Run order
P (bar)
T(°C)
Nozzle diameter (μm)
Contact time (h)
Yield (%)
Particle size (μm)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
13 2 10 3 1 12 8 18 5 16 19 6 15 14 11 17 7 4 9
100 300 100 300 100 300 100 300 100 300 100 300 100 300 100 300 200 200 200
313 313 313 313 343 343 343 343 313 313 313 313 343 343 343 343 328 328 328
100 100 200 200 100 100 200 200 100 100 200 200 100 100 200 200 150 150 150
1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 1.5 1.5 1.5
62 85 58 81 14 98 12 96 61 83 56 79 14 99 11 94 75 72 74
7.90 ± 0.30 3.85 ± 0.06 5.48 ± 0.05 5.25 ± 0.08 12.45 ± 0.48 6.23 ± 0.07 26.67 ± 0.56 3.89 ± 0.06 6.39 ± 0.08 5.36 ± 0.07 5.57 ± 0.05 4.34 ± 0.09 10.44 ± 0.26 3.03 ± 0.03 18.2 ± 0.44 5.58 ± 0.05 6.00 ± 0.07 6.48 ± 0.05 6.23 ± 0.06
combines high and low levels of factors and, as a consequence, some experiments could be in the low solubility region, e.g., when the higher temperature and the lower pressure were set. The effects on particle size and yield were positive and negative, as can be seen from the results in Table 1. The sign of each effect indicates whether the particle size or yield increase when the experiments change from a low to a high level. For instance, an increase in pressure from the low to the high level led to a decrease in particle size and an increase in temperature led to an increase in particle size. With regard to the yield, the most marked effect was observed for pressure followed by temperature. These effects had different signs, which indicates that yield increased with pressure and decreased with temperature.
square (MS) are given in Table 3. The results of this analysis showed that the model is significant through the values of F and p [21,22]. A high F value and a low p value indicate that the corresponding variables are significant. The p values given by the model (Table 3) are less than 0.004 (F = 41.61) and 0.008 (F = 6.17), respectively, and this indicates that the model is significant for both responses. It can be seen from the results in Table 3 that the probability of a lack of fit is not significant at 95%, so statistically the model does not have a lack of fit. This analysis is suitable to determine the optimum operating conditions to achieve a higher particle size reduction and higher yield of vanillin. The experimental design and the main particle sizes and yields of the micronization experiments are provided in Table 2. The design
Fig. 1. Schematic diagram of RESS250 pilot plant.
171
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Table 3 Analysis of variance for design model of process variables. variables PS
Model Residual Error Model Residual Error
Yield
DF
SS
MS
R2
AdjR2
Q2
Significance F
Lack of fit p
18 8 2 18 8 2
598.02 68.55 0.11 16528.60 311.01 4.66
33.22 8.56 0.06 918.25 38.87 2.33
0.91
0.79
0.42
6.17
0.008
0.98
0.96
0.74
41.61
0.004
considering runs 1 and 2 (Fig. 4). If the pressure is increased in the solubilization chamber, the density of the supercritical fluid increases and this enhances the solvating power of the supercritical fluid and increases the solute concentration in the supercritical solution. According to the classical theory of nucleation, higher super saturation induces a higher nucleation rate during the expansion period and thus the particle size of precipitated powder is reduced. Similar results were obtained by other authors with other solutes and the same process [23–25]. In the case of solubilization, the effect of temperature seems to be the opposite to that of pressure, i.e., a lower temperature leads to a lower particle size. An increase in the solubilization temperature leads to a decrease in the density of CO2 and a concurrent increase in the solute vapor pressure, which leads to two competing phenomena: a decrease in the solvent strength and an increase in the solute solubility [23]. In this work the increase in solute solubility prevails and higher supersaturations are achieved, thus giving a smaller particle size. However, when the high level of pressure is held the trend for temperature is opposite in that a higher temperature led to a smaller particle size. Thus, pressure and temperature are two factors that interact with one another, as can be seen in Fig. 5. A low level of pressure and a lower temperature are recommended to reduce further the particle size, but at high pressures a high temperature should be set. A surface response plot of pressure and temperature is shown in Fig. 6 and this enables the interaction to be visualized more clearly. Pressure and temperature define a critical state with a particular density, which has a strong influence on the level of supersaturation and thus the analysis as a whole is more accurate. An increase in temperature should be accompanied by an increase in pressure to achieve the optimal particle size reduction. It was also observed that at lower temperature the pressure does not seem to be an important factor. This can be seen in Fig. 6, where the particle size areas are wider at the bottom of the plot, which indicates that the particle size of vanillin is relatively similar regardless of pressure. However, when the temperature is increased the particle size areas become lower so the influence of pressure is at a maximum. In any case, from an economic point of view, it is desirable to work at the lowest levels of pressure and temperature assayed because the vanillin particle sizes do not differ significantly from the
Table 4 Coefficients list of linear model for particle size and yield and their significances. Coeff. (PS) Constant P Ө T t P·Ө P·T P·t Ө·T Ө·t T·t
7.86 −3.27 1.14 2.18 −0.65 −1.01 −2.66 0.71 1.62 −0.03 −0.49
p (PS) −6
1.24 × 10 7.95 × 10−4 0.11 8.39 × 10−3 0.33 0.12 1.98 × 10−3 0.26 0.02 0.95 0.42
Coeff. (Yield)
p (Yield)
64.42 25.16 −1.70 −7.48 −0.53 −0.05 13.61 −0.05 0.27 −0.27 0.27
6.52 × 10−11 1.37 × 10−7 0.28 9.38 × 10−4 0.73 0.97 9.69 × 10−6 0.97 0.85 0.85 0.85
P, pressure; T, temperature; t, contact time; Ө, nozzle diameter.
In this way the coefficients of the linear model for both particle size and yield showed in Table 4 probe that pressure and temperature are the only significative variables for both responses (p < 0.05). 3.2. RESS process The RESS process led to powder precipitation under all of the experimental conditions employed due to the relatively high solubility of vanillin in supercritical CO2. These experiments were performed in the region where a solid/fluid two-phase equilibrium exists for each pressure, i.e., conditions on the left-hand side of the SLG line of vanillin [14]. The size of the raw material (700 μm as main particle size) was reduced by a factor of one hundred in the RESS process (3–26 μm s mean particle size), as can be seen by comparing the results in Table 2 and Fig. 2, where the PSD of the raw material is compared with run 14 as an example. Not only was the particle size reduced but the uniformity of the particles was greatly improved and the morphology seemed to be unaltered. The main effects on the particle size of vanillin are shown in Fig. 3. It can be seen that higher pressures are recommended to reduce the particle size of precipitated vanillin. This effect can clearly be seen by
Fig. 2. Curves of the probability density function for particle size distribution of a) commercial, and b) RESS processed sample (run 14).
172
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Fig. 3. Main effects on particle size and yield of vanillin precipitated particles.
With regard to the yield, the effects of the main parameters separately are shown in Fig. 3. Higher pressures and lower temperatures are recommended to increase the yield [29] and these effects are more pronounced than particle size effects, as can be seen from the results in Table 1. In this case, the interaction between pressure and temperature on the yield was also studied and, as can be seen in Fig. 3, higher pressure and higher temperature produced the highest yield of precipitated powder. Under these conditions the solubilization of the vanillin is higher and this affects the yields obtained in the trials. Nozzle diameter and contact time had less effects on the yield and were not significative variables as can be seen in Table 4. The crystallinity of the samples processed by RESS was evaluated. The crystallinity is directly related to the activity of the substance and loss of crystallinity can therefore lead to loss of activity. In this case, the RESS-processed samples had identical diffractograms to the commercial vanillin, which indicates that the crystallinity remained unaltered after processing (see Fig. 7). The diffractograms of the products obtained under the lowest (run 1) and the highest (run 14) conditions (temperature and pressure) are shown in Fig. 7.
particle size obtained on working with the best conditions identified. A priori, it can be seen from the results in Fig. 4 and Table 1 that a smaller nozzle diameter and higher contact time are recommended to reduce the particle size, although these effects on particle size are less pronounced than the pressure and temperature effects. SEM images of the powder precipitated with different nozzle diameters are shown in Fig. 4, runs 2 and 4, and runs 14 and 16. It can be seen that when the nozzle was changed from 100 to 200 μm the mean particle size increased to almost double the diameter. Increasing the nozzle diameter leads to lower mass flow rate and therewith a higher residence time of the particles in the expansion so more time available for particle growth producing bigger particles. On the contrary, it could be considered that when the nozzle diameter decreases, particles have less space to move without cross, though the probability of coagulation increases [26]. In our case the first effect prevails so decreasing of the nozzle diameter causes a decrease in the mean particle size as others authors concluded [27,28]. However, according to the significance study of the variables (Table 4), nozzle diameter would not be a significative variable so could be discarded to the optimization design. SEM images of powders precipitated at different contact times are shown Fig. 4 (runs 4 and 12 and 5 and 13). It was observed that a higher contact time led to the production of particles that were 15% smaller than those obtained with a low contact time. Again, according to results showed in Table 4 these differences are not significatives, at least in the assayed levels, and are not important to optimize the micronization process.
4. Conclusions Uniform vanillin microparticles have been precipitated by an RESS process. The particle size was reduced by a factor of one hundred when compared to that of commercial vanillin. A full factorial design at two 173
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Fig. 4. SEM images of commercial and precipitated vanillin at different conditions.
contact time and nozzle diameter also affected the particle parameters, albeit to a lesser extent, although a smaller nozzle diameter and higher contact time led to a slight reduction in particle size and increase in the yield. The RESS process did not have any discernable effect on the morphology and crystallinity of the precipitated vanillin when compared to the unprocessed commercial starting material.
levels was applied in order to evaluate the main parameters that affect the particle size and yield of vanillin. Pressure and temperature were the main factors that dictated particle size and yield. The use of higher pressure and temperature led to smaller particles and higher yields in accordance of the results of the design. In any case, at the lowest temperature the pressure effect did not appear to be relevant for the particle size of vanillin and the results did not differ significantly. The
Fig. 5. P-T Interactions on particle size of vanillin precipitated particles.
174
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Fig. 6. P-T response surface plot of particle size and yield of vanillin precipitated particles.
for analyses. References [1] G. Romero de la Vega, M.A. Salgado Cervantes, M.A. Garcia Alvarado, A. RomeroMartínez, P.E. Hegel, Fractionation of vanilla oleoresin by supercritical CO2 technology, J. Supercrit. Fluids 108 (2016) 79–88. [2] M.M. Bomgardner, Following many routes to naturally derived vanillin, Chem. Eng. News 92 (6) (2014) 14. [3] A. Tai, T. Sawano, F. Yazama, Evaluation of antioxidant activity of vanillin by using multiple antioxidant assays, Biochim. Biophys. Acta 1810 (2) (2011) 170–177. [4] K. Ho, N. Yazan, M. Ismail, Toxicology study of vanillin on rats via oral and intraperitoneal administration, Food Chem. Toxicol. 49 (1) (2011) 25–30. [5] L.F. Dalmolin, N.M. Khalil, R.M. Mainardes, Delivery of vanillin by poly(lactic-acid) nanoparticles: development, characterization and in vitro evaluation of antioxidant activity, Mater. Sci. Eng. C 62 (2016) 1–8. [6] S. Rakchoy, P. Suppakul, T. Jinkarn, Antimicrobial effects of vanillin coated solution for coating paperboard intended for packaging bakery products, Asian J. Food Agro-Ind. 2 (4) (2009) 138–147. [7] H.P.V. Rupasinghe, J. Boulter-Bitzer, T. Ahn, J.A. Odumeru, Vanillin inhibits pathogenic and spoilage microorganisms in vitro and aerobic microbial growth in fresh-cut apples, Food Res. Int. 39 (5) (2006) 575–580. [8] M. Stroescu, A. Stoica-Guzun, G. Isopencu, S.I. Jinga, O. Parvulescu, T. Dobre, M. Vasilescu, Chitosan-vanillin composites with antimicrobial properties, Food Hydrocoll. 48 (2015) 62–71. [9] I. Mourtzinos, S. Konteles, N. Kalogeropoulos, V.T. Karathanos, Thermal oxidation of vanillin affects its antioxidant and antimicrobial properties, Food Chem. 114 (3) (2009) 791–797. [10] M. Fache, B. Boutevin, S. Caillol, Vanillin, a key-intermediate of biobased polymers, Eur. Polym. J. 68 (2015) 488–502. [11] S. Walke, G. Srivastava, M. Nikalje, J. Doshi, R. Kumar, S. Ravetkar, P. Doshi, Fabrication of chitosan microspheres using vanillin/TPP dual crosslinkers for protein antigens encapsulation, Carbohydr. Polym. 128 (2015) 188–198. [12] L. Pitol-Filho, V. Kokol, B. Voncina, Synthesis and characterization of ethyl cellulose microcapsules containing model active ingredients, Macromol. Symp. 328 (2013) 45–55. [13] M. Noshad, M. Mohebbi, A. Koocheki, F. Shahidi, Microencapsulation of vanillin by spray drying using soy protein isolate–maltodextrin as wall material, Flavour Fragr. J. 30 (5) (2015) 387–391. [14] M. Skerget, L. Cretnik, Z. Knez, M. Skrinjar, Influence of the aromatic ringsubstituents on phase equilibria of vanillins in binary systems with CO2, Fluid Phase Equilib. 231 (2005) 11–19. [15] J. Liu, Y. Kim, M. McHugh, Phase behavior of the vanillin–CO2 system at high pressures, J. Supercrit. Fluids 39 (2006) 201–205. [16] A. Montes, C. Pereyra, E.J. Martínez de la Ossa, Screening design of experiment applied to the supercritical antisolvent precipitation of quercetin, J. Supercrit. Fluids 104 (2015) 10–18. [17] P.B. Nalawade, A.K. Gajjar, Microencapsulation of lutein extracted from marigold flowers (Tagetes erecta L.) using full factorial design, J. Drug Deliv. Sci. Technol. 33 (2016) 75–87. [18] L. Azouz, F. Dahmoune, F. Rezgui, C. G’Sell, Full factorial design optimization of anti-inflammatory drug release by PCL–PEG–PCL microspheres, Mater. Sci. Eng. C
Fig. 7. XRD patterns of commercial and SAS processed vanillin.
Acknowledgments We gratefully acknowledge the European Regional Development Funds (UNCA10-1E-1125 and 18INIA1103. 2011) for financial support, and Central Services of Science and Technology of University of Cádiz 175
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al. 58 (2016) 412–419. [19] B. Süto, S. Weber, A. Zimmer, G. Farkas, A. Kelemen, M. Budai-Szucs, S. Berkó, P. Szabó-Révész, E. Csányi, Optimization and design of an ibuprofen-loaded nanostructured lipid carrier with a 23 full factorial design, Chem. Eng. Res. Des. 104 (2015) 488–496. [20] A. Montes, A. Bendel, R. Kürti, M.D. Gordillo, C. Pereyra, E.J. Martínez de la Ossa, Processing naproxen with supercritical CO2, J. Supercrit. Fluids 75 (2013) 21–29. [21] F. Dahmoune, G. Spigno, K. Moussi, H. Remini, A. Cherbal, K. Madani, Pistacia lentiscus leaves as a source of phenolic compounds: microwave-assisted extraction optimized and compared with ultrasound-assisted and conventional solvent extraction, Ind. Crops Prod. 61 (2014) 31–40. [22] A.Z. Hezave, F. Esmaeilzadeh, Micronization of drug particles via RESS process, J. Supercrit. Fluids 52 (2010) 84–98. [23] E. Reverchon, G. Donsi, D. Gorgoglione, Salicylic acid solubilization in supercritical CO2 and its micronization by RESS, J. Supercrit. Fluids 6 (1993) 241–248. [24] I. Asghari, F. Esmaeilzadeh, Formation of ultrafine deferasirox particles via rapid
[25] [26]
[27] [28]
[29]
176
expansion of supercritical solution (RESS process) using Taguchi approach, Int. J. Pharm. 433 (2012) 149–156. A.Z. Hezave, F. Esmaeilzadeh, The effects of RESS parameters on the diclofenac particle size, Adv. Powder Technol. 22 (2011) 587–595. H.R. Satvati, M.N. Lotfollahi, Effects of extraction temperature, extraction pressure and nozzle diameter on micronization of cholesterol by RESS process, Powder Technol. 210 (2011) 109–114. J. Huang, T. Moriyoshi, Fabrication of fine powders by RESS with a clearance nozzle, J. Supercrit. Fluids 37 (2006) 292–297. R. Salinas-Hernández, F.A. Ruiz-Treviño, C.-H. Ortiz-Estrada, Chitin microstructure formation by rapid expansion techniques with supercritical carbon dioxide, Ind. Eng. Chem. Res. 48 (2009) 769–778. G.E. Oliveira, J.F. Pinto, Evaluation of the potential use of laminar extrudates on stabilizing micronized coumarin particles by supercritical fluids (RESS)—study of different RESS processing variables and mode of operation, AAPS Pharm. Sci. Technol. (2017) 1–16.
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
Micronization of vanillin by rapid expansion of supercritical solutions process
MARK
⁎
A. Montesa, , R. Merinoa, D.M. De los Santosb, C. Pereyraa, E.J. Martínez de la Ossaa a b
Department of Chemical Engineering and Food Technology, University of Cádiz, International Excellence Agrifood Campus (CeiA3), 11510 Puerto Real, Cádiz, Spain Department of Physical Chemistry, Faculty of Sciences, University of Cádiz, International Excellence Agrifood Campus (CeiA3), 11510 Puerto Real, Cádiz, Spain
A R T I C L E I N F O
A B S T R A C T
Keywords: Vanillin RESS process Microparticles Supercritical
Uniform microparticles of vanillin were precipitated by the Rapid Expansion of Supercritical Solution (RESS) process and this reduced the commercial particle size by a factor of one hundred. A design of experiments full factorial at two levels was performed. The effects that pressure, temperature, contact time and nozzle diameter had on the particle size and yield of the RESS precipitation were evaluated. Pressure and temperature were the factors that had the most marked effects on particle size and yield. The use of higher pressure and temperature is recommended to obtain the smallest particle size and the highest yield. However, at lower pressure the temperature is a crucial factor in that the use of a lower temperature led to considerably smaller particles and good yields whereas a higher temperature gave the highest particle size and the lowest yield. A higher contact time and smaller nozzle diameter led to slight improvements in the vanillin particle size and yield according to the results of the design. The crystallinity of the RESS-processed vanillin was unaltered when compared to the unprocessed material.
1. Introduction Vanillin (4-hydroxy-3-methoxybenzaldehyde) (C8H8O3) is the main component (80–90%) [1] of vanilla orchid pods and it has an attractive flavor and fragrance. The current global demand for vanillin is estimated to be roughly 20,000 tons per year [2] and it is becoming one of the most important aromatic substances. Vanillin is often used as a flavoring agent, a food preservative, in beverages, cosmetics and drugs due to its antioxidant [3–5] and antimicrobial [6–8] activities and it is generally regarded as a safe (GRAS) substance [9]. Vanillin is also used as a crosslinker and in the preparation of a wide range of vanillin-based renewable polymers, e.g., phenolic, epoxy and benzoxazine resins, polyesters, and acrylate and methacrylate polymers are in the focus of numerous investigations [10,11]. Vanillin has been widely used as the commercial product (∼700 μm particle diameter) and this has low solubility in cold water, although the solubility increases with temperature. However, this solubility could be improved by reducing the particle size to the micrometer range. There is very little literature concerning the micronization of vanillin although composites and microencapsulates of vanillin has been obtained: microcapsules containing limonene and vanillin with a mean particle diameter of 30 μm [12] were prepared by Pitol-Filho et al. using ethyl cellulose as a coating agent by dropping the polymeric
⁎
Corresponding author. E-mail address: [email protected] (A. Montes).
http://dx.doi.org/10.1016/j.jcou.2017.07.009 Received 24 March 2017; Received in revised form 1 June 2017; Accepted 4 July 2017 2212-9820/ © 2017 Elsevier Ltd. All rights reserved.
solution into a coagulation bath, which contained water, sodium dodecyl sulfate, and acetic acid in different concentrations. Noshad et al. carried out the microencapsulation of vanillin with maltodextrin using a spray dying method and this process gave a wide particle diameter range (0.7–128 μm) [13]. A maximum encapsulation efficiency of 58.3% and minimum particle size of 6.95 μm were achieved at 184 °C with a maltodextrin concentration of 8.5% and vanillin concentration of 0.36%. The majority of the micronization methods involve the use of excessive amounts of organic solvent, high temperatures and post-processing steps and the final products often contain residues. These drawbacks could be circumvented by using supercritical CO2 (scCO2) technologies due to the low toxicity of this compound and the absence of organic solvent in the process, thus contributing to a substantial improvement in human health and environmental considerations. The application of supercritical fluid technology in micronization processes requires a study of the phase equilibrium formed by the solute and the supercritical CO2, i.e., the fluid-solid phase equilibrium. Binary data for the solubility of vanillin in CO2 have previously been reported in the literature [14,15]. Molar fractions from 0.014 to 1.295 × 10−2 were obtained in the pressure range 80–300 bar and the temperature range 313–353 K, which shows that vanillin is quite soluble in supercritical CO2. In this way, the Rapid Expansion of Supercritical Solutions
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
there is a stainless steel collection vessel (V2) in which particles are precipitated once the supercritical solution is expanded through a stainless steel nozzle with an inner diameter in the range 100–200 μm (Thar Technologies). The equipment was described in detail in a previous publication [20]. The experiments were performed as follows: Firstly, the commercial vanillin powder (0.5 g) was placed in the solubilization chamber. The system was held for a certain time once supercritical conditions for CO2 (pressure and temperature) had been achieved in order to ensure complete equilibration. Valve MV2 was then opened and the supercritical solution was expanded through a nozzle, which was pre-heated in order to compensate for the heat loss and to prevent the nozzle from clogging during the fast expansion. The precipitated particles were collected on the wall of vessel V2 for subsequent analysis.
(RESS) process could be used to micronize this compound. In the RESS process a given amount of vanillin is placed into a solubilization chamber, which is filled with scCO2 to give a supercritical solution after a certain time. The subsequent sudden expansion of this solution through a nozzle to atmospheric conditions generates a high level of supersaturation and produces rapid nucleation and the precipitation of particles on the vessel wall. The work described here concerned a study of the micro-precipitation of vanillin by the RESS process and a design of experiments was applied in order to identify the main parameters that affect the process in terms of particle size, size distribution and yield. The resulting vanillin microparticles were analyzed by X-ray diffraction to evaluate the degree of crystallinity after the supercritical process. 2. Materials and methods
2.3. Sample characterization
Vanillin (4-hydroxy-3-methoxybenzaldehyde) (C8H8O3, 99% purity) was purchased from Sigma-Aldrich (Spain). CO2 with a minimum purity of 99.8% was supplied by Linde (Spain).
Scanning electron microscopy (SEM) images of the powder precipitated on the wall of the vessel were obtained using a NOVASEM scanning electron microscope. Prior to analysis the samples were placed on carbon tape and then covered with a coating of gold using a sputter coater. Particle size distribution (PSD) and mean particle size of the processed samples were measured using DLS technology (Nano-ZS, Malvern Instruments, UK). This technique measures the diffusion of particles moving under Brownian motion, and converts this to a size and a size distribution using the Stokes–Einstein relationship. NonInvasive Back Scatter (NIBS) technology was incorporated to give the highest sensitivity along with the highest possible size and concentration range. The measurement range was from 0.3 nm to 10 μm. Prior to analysis, the samples of vanillin were weighed and suspended in hexane (1 mg/mL) with one drop of surfactant (Tween 80). The PSD and mean particle size of commercial vanillin could not be processed by this technique due to its large size, which gave rise to an unreliable measurement. In this case, an LS 13 320 laser diffraction particle size analyzer from Beckman Coulter was used. This system was able to measure particles up to 2000 μm. Polarization Intensity Differential Scattering (PIDS) technology was employed in which the particle size distribution could be determined by measuring the variations between the horizontally and the vertically scattered light for each wavelength. In this case the sample was introduced as a solid powder using a module (Tornado) that could handle dry powder. X-ray diffraction (XRD) analysis was performed on Bruker D8 Advance diffractometer to determine the amorphous or crystalline nature of the precipitate obtained by both processes. All diffraction patterns were scanned from 10° to 80° in 2θ angle with a step size of 0.02° and one second as the step time.
2.1. Experimental design A design of experiment (DOE) was carried out in order to identify the critical factors in the vanillin miconization by the RESS process. A full factorial design at two levels was performed. Four factors (24) were studied but the complete design consisted of 19 experimental points that included 16 factor points and three replications at the center point (experiments 17–19). The critical factors were selected with appropriate ranges for this design. The responses of the design were mean particle size and yield. Yield was calculated by subtracting the weight of powder precipitated from the initial weight of vanillin placed into the solubilization chamber (0.5 g). Other full factorial designs have been successfully applied for screening purposes [16–19]. Pressure (P), temperature (T), nozzle diameter (Øn) and contact time (t) were identified as possible parameters that could influence the particle size and size distribution of vanillin in the RESS process. The two levels for each factor are shown in Table 1. The levels of pressure and temperature were selected to evaluate the influence of different supercritical CO2 densities and therefore solvent power. Nozzle diameters were selected according the those available from the equipment supplier. The contact times were selected in order to ensure that there was sufficient time to form a homogenous supercritical solution and to keep the costs reasonable. 2.2. RESS process The experiments listed in Table 2 were carried out in a pilot plant developed by Thar Technologies® (model RESS250). A schematic diagram of this equipment is shown in Fig. 1. Pressures of 100–300 bar, temperatures of 313–343 K, nozzle diameters of 100–200 μm and contact times of 1–2 h were evaluated. The RESS250 equipment comprises a high-pressure pump (P1) to fill a 250 mL stainless steel solubilization vessel (V1) with CO2. This vessel is heated by an electrical heating jacket (V1-HJ1) that surrounds the vessel. Supercritical CO2 and vanillin were continuously mixed with a magnetic stirrer (maximum speed 2500 rpm). At the end of the line
2.4. Statistical analysis Each experiment was carried out following the order established by the design. Influence of each factor on the particle size and yield of vanillin was statistically assessed by ANOVA with 95% confidence level. Data obtained from the partial least square (PLS) regression were statistically analyzed using ANOVA for the response variable in order to test the model significance (p < 0.05) (Table 3). The Modde (Version 5.1, Unmetrics, USA) software was employed to analyze the results and optimize the experimental conditions. Significance of pressure, temperature, nozzle diameter and contact time on particle size and yield were established at 95% of confidence level (Table 4).
Table 1 Two – Level assessment for each factor and calculated effects on PS. Factors
Low level
High level
PS Effects
Yield Effects
P (bar) T (K) t (h) Øn (μm)
100 313 1 100
300 343 2 200
−6.49 5.29 −1.37 2.41
53.37 −15.87 −3.62 −1.12
3. Results and discussion 3.1. Analysis of the design of experiments The values for the coefficient of determination R2 (adjusted and experimental), degree of freedom (DF), sum of squares (SS) and mean 170
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Table 2 Experimental design and observed responses. Run name
Run order
P (bar)
T(°C)
Nozzle diameter (μm)
Contact time (h)
Yield (%)
Particle size (μm)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
13 2 10 3 1 12 8 18 5 16 19 6 15 14 11 17 7 4 9
100 300 100 300 100 300 100 300 100 300 100 300 100 300 100 300 200 200 200
313 313 313 313 343 343 343 343 313 313 313 313 343 343 343 343 328 328 328
100 100 200 200 100 100 200 200 100 100 200 200 100 100 200 200 150 150 150
1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 1.5 1.5 1.5
62 85 58 81 14 98 12 96 61 83 56 79 14 99 11 94 75 72 74
7.90 ± 0.30 3.85 ± 0.06 5.48 ± 0.05 5.25 ± 0.08 12.45 ± 0.48 6.23 ± 0.07 26.67 ± 0.56 3.89 ± 0.06 6.39 ± 0.08 5.36 ± 0.07 5.57 ± 0.05 4.34 ± 0.09 10.44 ± 0.26 3.03 ± 0.03 18.2 ± 0.44 5.58 ± 0.05 6.00 ± 0.07 6.48 ± 0.05 6.23 ± 0.06
combines high and low levels of factors and, as a consequence, some experiments could be in the low solubility region, e.g., when the higher temperature and the lower pressure were set. The effects on particle size and yield were positive and negative, as can be seen from the results in Table 1. The sign of each effect indicates whether the particle size or yield increase when the experiments change from a low to a high level. For instance, an increase in pressure from the low to the high level led to a decrease in particle size and an increase in temperature led to an increase in particle size. With regard to the yield, the most marked effect was observed for pressure followed by temperature. These effects had different signs, which indicates that yield increased with pressure and decreased with temperature.
square (MS) are given in Table 3. The results of this analysis showed that the model is significant through the values of F and p [21,22]. A high F value and a low p value indicate that the corresponding variables are significant. The p values given by the model (Table 3) are less than 0.004 (F = 41.61) and 0.008 (F = 6.17), respectively, and this indicates that the model is significant for both responses. It can be seen from the results in Table 3 that the probability of a lack of fit is not significant at 95%, so statistically the model does not have a lack of fit. This analysis is suitable to determine the optimum operating conditions to achieve a higher particle size reduction and higher yield of vanillin. The experimental design and the main particle sizes and yields of the micronization experiments are provided in Table 2. The design
Fig. 1. Schematic diagram of RESS250 pilot plant.
171
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Table 3 Analysis of variance for design model of process variables. variables PS
Model Residual Error Model Residual Error
Yield
DF
SS
MS
R2
AdjR2
Q2
Significance F
Lack of fit p
18 8 2 18 8 2
598.02 68.55 0.11 16528.60 311.01 4.66
33.22 8.56 0.06 918.25 38.87 2.33
0.91
0.79
0.42
6.17
0.008
0.98
0.96
0.74
41.61
0.004
considering runs 1 and 2 (Fig. 4). If the pressure is increased in the solubilization chamber, the density of the supercritical fluid increases and this enhances the solvating power of the supercritical fluid and increases the solute concentration in the supercritical solution. According to the classical theory of nucleation, higher super saturation induces a higher nucleation rate during the expansion period and thus the particle size of precipitated powder is reduced. Similar results were obtained by other authors with other solutes and the same process [23–25]. In the case of solubilization, the effect of temperature seems to be the opposite to that of pressure, i.e., a lower temperature leads to a lower particle size. An increase in the solubilization temperature leads to a decrease in the density of CO2 and a concurrent increase in the solute vapor pressure, which leads to two competing phenomena: a decrease in the solvent strength and an increase in the solute solubility [23]. In this work the increase in solute solubility prevails and higher supersaturations are achieved, thus giving a smaller particle size. However, when the high level of pressure is held the trend for temperature is opposite in that a higher temperature led to a smaller particle size. Thus, pressure and temperature are two factors that interact with one another, as can be seen in Fig. 5. A low level of pressure and a lower temperature are recommended to reduce further the particle size, but at high pressures a high temperature should be set. A surface response plot of pressure and temperature is shown in Fig. 6 and this enables the interaction to be visualized more clearly. Pressure and temperature define a critical state with a particular density, which has a strong influence on the level of supersaturation and thus the analysis as a whole is more accurate. An increase in temperature should be accompanied by an increase in pressure to achieve the optimal particle size reduction. It was also observed that at lower temperature the pressure does not seem to be an important factor. This can be seen in Fig. 6, where the particle size areas are wider at the bottom of the plot, which indicates that the particle size of vanillin is relatively similar regardless of pressure. However, when the temperature is increased the particle size areas become lower so the influence of pressure is at a maximum. In any case, from an economic point of view, it is desirable to work at the lowest levels of pressure and temperature assayed because the vanillin particle sizes do not differ significantly from the
Table 4 Coefficients list of linear model for particle size and yield and their significances. Coeff. (PS) Constant P Ө T t P·Ө P·T P·t Ө·T Ө·t T·t
7.86 −3.27 1.14 2.18 −0.65 −1.01 −2.66 0.71 1.62 −0.03 −0.49
p (PS) −6
1.24 × 10 7.95 × 10−4 0.11 8.39 × 10−3 0.33 0.12 1.98 × 10−3 0.26 0.02 0.95 0.42
Coeff. (Yield)
p (Yield)
64.42 25.16 −1.70 −7.48 −0.53 −0.05 13.61 −0.05 0.27 −0.27 0.27
6.52 × 10−11 1.37 × 10−7 0.28 9.38 × 10−4 0.73 0.97 9.69 × 10−6 0.97 0.85 0.85 0.85
P, pressure; T, temperature; t, contact time; Ө, nozzle diameter.
In this way the coefficients of the linear model for both particle size and yield showed in Table 4 probe that pressure and temperature are the only significative variables for both responses (p < 0.05). 3.2. RESS process The RESS process led to powder precipitation under all of the experimental conditions employed due to the relatively high solubility of vanillin in supercritical CO2. These experiments were performed in the region where a solid/fluid two-phase equilibrium exists for each pressure, i.e., conditions on the left-hand side of the SLG line of vanillin [14]. The size of the raw material (700 μm as main particle size) was reduced by a factor of one hundred in the RESS process (3–26 μm s mean particle size), as can be seen by comparing the results in Table 2 and Fig. 2, where the PSD of the raw material is compared with run 14 as an example. Not only was the particle size reduced but the uniformity of the particles was greatly improved and the morphology seemed to be unaltered. The main effects on the particle size of vanillin are shown in Fig. 3. It can be seen that higher pressures are recommended to reduce the particle size of precipitated vanillin. This effect can clearly be seen by
Fig. 2. Curves of the probability density function for particle size distribution of a) commercial, and b) RESS processed sample (run 14).
172
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Fig. 3. Main effects on particle size and yield of vanillin precipitated particles.
With regard to the yield, the effects of the main parameters separately are shown in Fig. 3. Higher pressures and lower temperatures are recommended to increase the yield [29] and these effects are more pronounced than particle size effects, as can be seen from the results in Table 1. In this case, the interaction between pressure and temperature on the yield was also studied and, as can be seen in Fig. 3, higher pressure and higher temperature produced the highest yield of precipitated powder. Under these conditions the solubilization of the vanillin is higher and this affects the yields obtained in the trials. Nozzle diameter and contact time had less effects on the yield and were not significative variables as can be seen in Table 4. The crystallinity of the samples processed by RESS was evaluated. The crystallinity is directly related to the activity of the substance and loss of crystallinity can therefore lead to loss of activity. In this case, the RESS-processed samples had identical diffractograms to the commercial vanillin, which indicates that the crystallinity remained unaltered after processing (see Fig. 7). The diffractograms of the products obtained under the lowest (run 1) and the highest (run 14) conditions (temperature and pressure) are shown in Fig. 7.
particle size obtained on working with the best conditions identified. A priori, it can be seen from the results in Fig. 4 and Table 1 that a smaller nozzle diameter and higher contact time are recommended to reduce the particle size, although these effects on particle size are less pronounced than the pressure and temperature effects. SEM images of the powder precipitated with different nozzle diameters are shown in Fig. 4, runs 2 and 4, and runs 14 and 16. It can be seen that when the nozzle was changed from 100 to 200 μm the mean particle size increased to almost double the diameter. Increasing the nozzle diameter leads to lower mass flow rate and therewith a higher residence time of the particles in the expansion so more time available for particle growth producing bigger particles. On the contrary, it could be considered that when the nozzle diameter decreases, particles have less space to move without cross, though the probability of coagulation increases [26]. In our case the first effect prevails so decreasing of the nozzle diameter causes a decrease in the mean particle size as others authors concluded [27,28]. However, according to the significance study of the variables (Table 4), nozzle diameter would not be a significative variable so could be discarded to the optimization design. SEM images of powders precipitated at different contact times are shown Fig. 4 (runs 4 and 12 and 5 and 13). It was observed that a higher contact time led to the production of particles that were 15% smaller than those obtained with a low contact time. Again, according to results showed in Table 4 these differences are not significatives, at least in the assayed levels, and are not important to optimize the micronization process.
4. Conclusions Uniform vanillin microparticles have been precipitated by an RESS process. The particle size was reduced by a factor of one hundred when compared to that of commercial vanillin. A full factorial design at two 173
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Fig. 4. SEM images of commercial and precipitated vanillin at different conditions.
contact time and nozzle diameter also affected the particle parameters, albeit to a lesser extent, although a smaller nozzle diameter and higher contact time led to a slight reduction in particle size and increase in the yield. The RESS process did not have any discernable effect on the morphology and crystallinity of the precipitated vanillin when compared to the unprocessed commercial starting material.
levels was applied in order to evaluate the main parameters that affect the particle size and yield of vanillin. Pressure and temperature were the main factors that dictated particle size and yield. The use of higher pressure and temperature led to smaller particles and higher yields in accordance of the results of the design. In any case, at the lowest temperature the pressure effect did not appear to be relevant for the particle size of vanillin and the results did not differ significantly. The
Fig. 5. P-T Interactions on particle size of vanillin precipitated particles.
174
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al.
Fig. 6. P-T response surface plot of particle size and yield of vanillin precipitated particles.
for analyses. References [1] G. Romero de la Vega, M.A. Salgado Cervantes, M.A. Garcia Alvarado, A. RomeroMartínez, P.E. Hegel, Fractionation of vanilla oleoresin by supercritical CO2 technology, J. Supercrit. Fluids 108 (2016) 79–88. [2] M.M. Bomgardner, Following many routes to naturally derived vanillin, Chem. Eng. News 92 (6) (2014) 14. [3] A. Tai, T. Sawano, F. Yazama, Evaluation of antioxidant activity of vanillin by using multiple antioxidant assays, Biochim. Biophys. Acta 1810 (2) (2011) 170–177. [4] K. Ho, N. Yazan, M. Ismail, Toxicology study of vanillin on rats via oral and intraperitoneal administration, Food Chem. Toxicol. 49 (1) (2011) 25–30. [5] L.F. Dalmolin, N.M. Khalil, R.M. Mainardes, Delivery of vanillin by poly(lactic-acid) nanoparticles: development, characterization and in vitro evaluation of antioxidant activity, Mater. Sci. Eng. C 62 (2016) 1–8. [6] S. Rakchoy, P. Suppakul, T. Jinkarn, Antimicrobial effects of vanillin coated solution for coating paperboard intended for packaging bakery products, Asian J. Food Agro-Ind. 2 (4) (2009) 138–147. [7] H.P.V. Rupasinghe, J. Boulter-Bitzer, T. Ahn, J.A. Odumeru, Vanillin inhibits pathogenic and spoilage microorganisms in vitro and aerobic microbial growth in fresh-cut apples, Food Res. Int. 39 (5) (2006) 575–580. [8] M. Stroescu, A. Stoica-Guzun, G. Isopencu, S.I. Jinga, O. Parvulescu, T. Dobre, M. Vasilescu, Chitosan-vanillin composites with antimicrobial properties, Food Hydrocoll. 48 (2015) 62–71. [9] I. Mourtzinos, S. Konteles, N. Kalogeropoulos, V.T. Karathanos, Thermal oxidation of vanillin affects its antioxidant and antimicrobial properties, Food Chem. 114 (3) (2009) 791–797. [10] M. Fache, B. Boutevin, S. Caillol, Vanillin, a key-intermediate of biobased polymers, Eur. Polym. J. 68 (2015) 488–502. [11] S. Walke, G. Srivastava, M. Nikalje, J. Doshi, R. Kumar, S. Ravetkar, P. Doshi, Fabrication of chitosan microspheres using vanillin/TPP dual crosslinkers for protein antigens encapsulation, Carbohydr. Polym. 128 (2015) 188–198. [12] L. Pitol-Filho, V. Kokol, B. Voncina, Synthesis and characterization of ethyl cellulose microcapsules containing model active ingredients, Macromol. Symp. 328 (2013) 45–55. [13] M. Noshad, M. Mohebbi, A. Koocheki, F. Shahidi, Microencapsulation of vanillin by spray drying using soy protein isolate–maltodextrin as wall material, Flavour Fragr. J. 30 (5) (2015) 387–391. [14] M. Skerget, L. Cretnik, Z. Knez, M. Skrinjar, Influence of the aromatic ringsubstituents on phase equilibria of vanillins in binary systems with CO2, Fluid Phase Equilib. 231 (2005) 11–19. [15] J. Liu, Y. Kim, M. McHugh, Phase behavior of the vanillin–CO2 system at high pressures, J. Supercrit. Fluids 39 (2006) 201–205. [16] A. Montes, C. Pereyra, E.J. Martínez de la Ossa, Screening design of experiment applied to the supercritical antisolvent precipitation of quercetin, J. Supercrit. Fluids 104 (2015) 10–18. [17] P.B. Nalawade, A.K. Gajjar, Microencapsulation of lutein extracted from marigold flowers (Tagetes erecta L.) using full factorial design, J. Drug Deliv. Sci. Technol. 33 (2016) 75–87. [18] L. Azouz, F. Dahmoune, F. Rezgui, C. G’Sell, Full factorial design optimization of anti-inflammatory drug release by PCL–PEG–PCL microspheres, Mater. Sci. Eng. C
Fig. 7. XRD patterns of commercial and SAS processed vanillin.
Acknowledgments We gratefully acknowledge the European Regional Development Funds (UNCA10-1E-1125 and 18INIA1103. 2011) for financial support, and Central Services of Science and Technology of University of Cádiz 175
Journal of CO₂ Utilization 21 (2017) 169–176
A. Montes et al. 58 (2016) 412–419. [19] B. Süto, S. Weber, A. Zimmer, G. Farkas, A. Kelemen, M. Budai-Szucs, S. Berkó, P. Szabó-Révész, E. Csányi, Optimization and design of an ibuprofen-loaded nanostructured lipid carrier with a 23 full factorial design, Chem. Eng. Res. Des. 104 (2015) 488–496. [20] A. Montes, A. Bendel, R. Kürti, M.D. Gordillo, C. Pereyra, E.J. Martínez de la Ossa, Processing naproxen with supercritical CO2, J. Supercrit. Fluids 75 (2013) 21–29. [21] F. Dahmoune, G. Spigno, K. Moussi, H. Remini, A. Cherbal, K. Madani, Pistacia lentiscus leaves as a source of phenolic compounds: microwave-assisted extraction optimized and compared with ultrasound-assisted and conventional solvent extraction, Ind. Crops Prod. 61 (2014) 31–40. [22] A.Z. Hezave, F. Esmaeilzadeh, Micronization of drug particles via RESS process, J. Supercrit. Fluids 52 (2010) 84–98. [23] E. Reverchon, G. Donsi, D. Gorgoglione, Salicylic acid solubilization in supercritical CO2 and its micronization by RESS, J. Supercrit. Fluids 6 (1993) 241–248. [24] I. Asghari, F. Esmaeilzadeh, Formation of ultrafine deferasirox particles via rapid
[25] [26]
[27] [28]
[29]
176
expansion of supercritical solution (RESS process) using Taguchi approach, Int. J. Pharm. 433 (2012) 149–156. A.Z. Hezave, F. Esmaeilzadeh, The effects of RESS parameters on the diclofenac particle size, Adv. Powder Technol. 22 (2011) 587–595. H.R. Satvati, M.N. Lotfollahi, Effects of extraction temperature, extraction pressure and nozzle diameter on micronization of cholesterol by RESS process, Powder Technol. 210 (2011) 109–114. J. Huang, T. Moriyoshi, Fabrication of fine powders by RESS with a clearance nozzle, J. Supercrit. Fluids 37 (2006) 292–297. R. Salinas-Hernández, F.A. Ruiz-Treviño, C.-H. Ortiz-Estrada, Chitin microstructure formation by rapid expansion techniques with supercritical carbon dioxide, Ind. Eng. Chem. Res. 48 (2009) 769–778. G.E. Oliveira, J.F. Pinto, Evaluation of the potential use of laminar extrudates on stabilizing micronized coumarin particles by supercritical fluids (RESS)—study of different RESS processing variables and mode of operation, AAPS Pharm. Sci. Technol. (2017) 1–16.