ABSTRACT Lignin microspheres of uniform spherical shape and narrow size distribution were successfully prepared from various lignin sources using Emulsion Solvent Evaporation (ESE) technique. Four lignins were examined for this study include; isolated lignin from non-wood/hardwood biorefinery residue (L1), isolated lignin from hardwood kraft black liquor (L2), technical kraft softwood lignin (L3) and technical soda non-wood lignin (L4). Lignin microspheres were prepared either from the lignin acetate or soluble part of lignin in the organic solvent (i.e. ethyl acetate). Size, size distribution and polydispersity of lignin microspheres were determined by Dynamic Light Scattering (DLS) and the morphology of lignin microspheres were observed by Scanning Electron Microscopy (SEM). Uniform lignin acetate microspheres were successfully produced from the four lignin samples in a size range of 1.28-1.38 µm in dichloromethane and 1.31-3.43 µm in ethyl acetate. The ethyl acetate soluble fraction of L2 and L4 was formed into spherical shape with the size of 1.15 µm and 0.87 µm, respectively.
Keywords: Lignin, lignin acetate, soluble fraction, microspheres, emulsion solvent evaporation technique.
Over 70 million tons per year of lignin is produced as by-product of biomass separation process 1, 2. The major part (~95%) of the lignin is burned to produce energy, while only a small portion (~5%) is used for value added products due to unknown molecular structure with variable molecular weight and functional groups 2, 3. However, the growing interest in sustainable chemicals has lead to utilization of lignin in various value-added applications such as phenol formaldehyde resin 4, 5, polymer blends and composites 6-8 and polyurethane foam 9, 10. The interest in developing advanced materials in nano and micro scale is also increasing due to a heightened demand for advanced sustainable products. For instance, lignin nano- and micro-particles has been studied for many application such as agricultural active controlled release 11, 12, nano-sized coatings 13, filler in composites 14, controlled drug release 15, 16 and food industry fat mimetics 17.
Polymer microspheres (spherical microparticles) can be prepared through Emulsion Solvent Evaporation (ESE) method18-20. ESE method involves two phases; organic phase (organic solvent and polymer) and aqueous phase (water and surfactant) (Figure 1A). Figure 1 shows the three steps in ESE method; (B) intermix, (C) solvent removal, and (D) solidification. First, in the intermix step, an emulsion of oil droplets which contain the polymer and the organic solvent is formed by agitating the mixture at high share rate. Then the organic solvent is removed from the system by using magnetic stirrer at low share rate. The microparticles are formed in the solidification step when the organic solvent is completely removed from the system.
A suitable organic solvent for fabrication of microspheres should have low boiling point, ability to dissolve the polymer and low miscibility with water 21, 22. Dichloromethane (DCM), ethyl acetate (EA) and chloroform are recommended as suitable solvents for ESE method because these solvents have capacity to dissolve polymers, poorly soluble in water and their boiling point is lower than the boiling point of water. Among these organic solvents DCM is a commonly used solvent for fabrication of microsphere due of its low boiling point, low miscibility and good solvent for most polymers. Lignin has a very low solubility in DCM and in fact, in other organic solvents as well, due to the presence of hydrophilic groups (i.e. hydroxyl groups) in the lignin macromolecule 23, 24. Acetylation, substituting hydroxyl groups with acetyl groups using acetic anhydride, is a technique that can improve the solubility of lignin in organic solvents 25. It is important to note that DCM is a carcinogen according to the Environmental Protection Agency (EPA), so it is preferable to use a safer solvent such as EA in the ESE method 21, 26, 27.
The goal of this study was to investigate the effect of the molecular weight of lignin, type of organic solvents (i.e. DCM and EA) and solubility of lignin in organic solvents on the formation of lignin microspheres. Evaluation of the lignin microspheres was performed by measuring the size, size distribution, uniformity and morphology. These qualities were evaluated using Dynamic Light Scattering (DLS) and Scanning Electron Microscopy (SEM). The theory of solvent diffusion/evaporation was discussed in detail to explore the mechanism of lignin microspheres formation by using ESE method. In addition, soluble part of lignin (non-acetylated lignin) in EA was evaluated for its ability to form lignin microspheres.
There were four sources of lignin evaluated in this work. The first lignin (L1) was isolated from the industrial residue generated at a bioethanol plant in United States. The bioethanol plant uses hardwood and non-wood species. The second source of lignin (L2) was isolated from the black liquor of a kraft pulp mill in Botucatu, Brazil, where the main species used in the mill is eucalyptus. The method for isolation of L1 and L2 was described in previous paper, 28. The third lignin (L3) is a technical softwood kraft lignin (Indulin AT) was received from the MeadWestvaco Corporation, Charleston, SC, USA. The fourth lignin (L4) is a technical non-wood soda lignin (Protobind) was obtained from ALM, Chandigarh, India.
The organic solvents were purchased as follow: Dichloromethane (DCM) (Caledon), ethyl acetate (EA) (Caledon), Pyridine (Caledon), polyvinyl alcohol (PVA) (Sigma), acetic anhydride (Caledon).
Acetylation of lignin was carried out as described by Olarte (2011) 25. 1.0 g oven-dried lignin was added to 40 ml of pyridine-acetic anhydride (1:1) solution. After 24 hours mixing in a sealed flask, the solids were re-precipitated with 150 ml of HCl 0.1N solution. The solid was filtered and washed with HCl 0.1N solution and deionized water. The collected solid was dried at 40°C overnight and stored in a vial for further analysis.
Preparation of lignin and lignin acetate microspheres
Lignin acetate microspheres were produced following the method described by 12, 29, 30. Briefly, 10mg lignin acetate was dissolved in 1 mL of either DCM or EA. The organic phase was intermixed with 10 mL polyvinyl alcohol (0.2% w/v) by using homogenizer at 10,000 rpm for 30 second. Then, 50 mL water was added into the emulsion and stirred for 3 hours with magnetic stirrer at room temperature to remove the solvent from the system. The mixture was centrifuged for 10 min at 9000rpm, and washed with water for two times. The collected particles were freeze-dried and placed in desiccator for further analysis.
Lignin microspheres were prepared from the soluble part of lignin in ethyl acetate. The soluble part was collected after filtration the insoluble part by using medium size (10-15µm pore size) of filter crucible. The solvent from the soluble part was removed by using rotary evaporator and collected sample was dried at 50°C for overnight and sealed in a vial. The lignin microspheres were prepared in a similar way as explained for acetylated lignins.
The yield of microspheres was determined by measuring the dry mass (freeze-dried) of the microspheres. Table 1 indicates the labels for lignin and lignin acetate microspheres that were prepared by DCM and EA.
Table 1. Description of samples used in the study
Particle size determination
The mean hydrodynamic diameter (Z-average size), size distribution, and polydispersity index (PDI) of the microspheres were analyzed by using ZETASIZER NANO ZS Malvern Instrument (Malvern, UK). The samples were prepared at 0.1% w/v concentration and sonicated for 10 minutes in water bath sonicator. The intensity of the highest peak from scattered light on the hydrodynamic diameter of the microspheres was normalized to unity for all samples. PDI is a dimensionless number which indicating the width of the size distribution of microspheres, having a value between 0 and 1 (0-0.1 indicating monodispersed particles and values between 0.1 and 0.25 indicating a narrow size particle distribution) 31. At least three measurements were taken and averaged for each test.
The ζ-potential of the microspheres was obtained by using ZETASIZER NANO ZS Malvern Instrument (Malvern, UK). The samples were prepared at 0.01% w/v concentration and sonicated for 10 minutes in water bath sonicator. At least three measurements were taken and averaged for each test.
Morphology of the microspheres
The morphology of the microspheres was studied by using SEM images which captured by a JEOL field emission microscope operating at 15-20 kV. The sample was mounted onto a glass substrate and coated with a thin layer of gold (~10 nm).
The stability of the colloidal was studied by analyzing the particles at 60th day. Particle size, PDI and ζ-potential of the particles in a neutral suspension were determined on the first day of preparation and after 60 days at room temperature. The results were recorded as mean ±Standard Deviation (±SD).
One-way analysis of variance (ANOVA) was performed to determine if observed differences in particle size, PDI and ζ-potential due to the lignin from different sources were statistically significant. In all experiments, if P<0.05, the difference will be considered statistically significant. Also, one-way ANOVA was performed to analyze the stability of the lignin microspheres suspension over time with significance (p<0.05), highly significant (p<0.01) and very significant (p<0.001).
Characterization of lignin and lignin acetate microspheres
Size distribution of all lignin and lignin acetate microspheres which were produced by using either DCM or EA are illustrated in Figure 2. All four acetylated lignins that used DCM as solvent were formed in uniform and unimodal distribution in the range between 600 nm to 4000 nm (Fig. 2A). The average peak for all ACL-DCM samples was about 1300 nm. Figure 2B shows that the ACL-EA microspheres were also formed in uniform distribution, but ACL1-EA and ACL4-EA had higher average particle sizes.
Although DCM was found to be a good solvent for producing lignin acetate microspheres, it was not a suitable solvent for forming lignin microspheres due to very low solubility of lignin in DCM. In previous work we found that the solubility of lignin in EA is relatively higher than DCM 24. Therefore, the EA soluble lignin was also used for producing lignin microspheres. Figure 2C shows a uniform size distribution of L2-EA, and bimodal distribution for L1-EA and L4-EA. The L3-EA particles were not formed and thus no particles were detected by the DLS.
Figure 2. Particle size distribution of lignin acetate microspheres (ACL-DCM and ACL-EA) and lignin microspheres (L-EA) produced from four lignin sources
Table 2 shows the average size, PDI, ζ-potential and yield of lignin and lignin acetate microspheres. The average size of the ACL-DCM microspheres was in the range of 1280 nm to 1350 nm, and PDI showed a narrow distribution from 0.100 to 0.170. ANOVA analysis showed no significant difference between the particle size of the microspheres produced from the four lignin sources (p>0.05). The particle size of the ACL-EA microspheres were more varied than the ACL-DCM microspheres. The largest particle size was the ACL1-EA (2545 nm) and the smallest particle size was 1313nm for ACL3-EA. ANOVA showed a significant difference (p<0.05) amongst the particles sizes of ACL-EA microspheres. PDI for ACL-EA microspheres was in a range from 0.130 to 0.186. The particles sizes for lignin microspheres (L-EA) were found to be 1279nm, 1152nm and 875 nm for L1-EA, L2-EA and L4-EA, respectively. However PDI was high for L1-EA and L4-EA which indicating bimodal or high particles dispersity. The effect of organic solvent, molecular weight of lignin and the solubility of lignin in organic solvent on the size and size distribution of lignin microspheres will be discussed in the following sessions.
The ζ-potential of the four ACL-DCMs was between -21 mV to -27 mV. According to ANOVA, there was no significant difference (p>0.05) between the ζ-potential of ACL-DCM microspheres. The particle charge for all ACL-EA microspheres showed a high negative charge which was in the range from -43 mV to -48 mV. Results of one-way ANOVA revealed no significant difference (p>0.05) between the ζ-potential of ACL-EA microspheres. Lignin microspheres (L-EA) had lower charge (-22 mV to -24 mV) than lignin acetate microspheres (ACL-EA). The surface charge of the particles can be attributed to the dissociation of the lignin groups, absorption of ions or ionisable molecules from the dispersing phase and/or the remaining part of the surfactant or the solvent on the particle surface.
The yield percentage of all ACL-DCM and ACL-EA microsphere was greater than 90%, while lignin microspheres were obtained at lower yields. Only L2-EA and L4-EA with 88.8% and 55.8% were formed in microspheres through ESE process. The yield percentage for L1-EA and L3-EA was very low due to very high agglomeration.
Figures 3, 4 and 5 show the SEM images of the ACL-DCM, ACL-EA and L-EA microspheres, respectively. It is clear in Figure 3 that all four ACL-DCM microspheres were formed in uniform spherical shape. SEM images of ACL-EA microspheres shows uniform spherical shape, but with different average size and size distribution (Fig. 4). Figure 5 shows that only L2-EA and L4-EA were formed in microspheres while L1-EA and L3-EA formation was not completed. In all cases, the surfaces of the microspheres (if formed) are relatively smooth.
Figure 3. SEM images of lignin acetate microspheres using lignin from different sources (DCM was used as organic solvent in the process) (the scale bar is 5 μm)
Figure 4. SEM micrographs of lignin acetate microspheres using lignin from different sources (EA was used as organic solvent in the process) (the scale bar is 5 μm)
Effect of organic solvent on the particle size
In our previous study, a comparison between the lignin acetate microspheres (non-wood soda lignin) showed that the particle size of lignin acetates in DCM were formed in smaller size than EA 30. Therefore, in this study we attempted to make a comparison between the particle size of lignin acetate microspheres from different sources of lignin using DCM and EA as dispersing solvent. We found that all lignin acetates were completely formed in spherical shape, disregards of the lignin sources, in both DCM and EA, however, ACL-EA microspheres were formed in same or greater size than ACL-DCM microspheres. The reason for this phenomenon might be explained by the differences in the physical properties of two organic solvents. The physical properties of DCM and EA are compared in Table 3 26.
Formation of ACL-DCM microspheres through ESE method occurs in shorter time than ACL-EA microspheres due to lower boiling point and higher evaporation rate of DCM than EA. In addition, DCM is immiscible in water but EA is slightly soluble in water. Therefore, DCM rapidly evaporates from the emulsion after diffusion from the droplets into the continuous phase, while EA is slightly dissolved in water before evaporation.
In general, the use of DCM is preferred due to its desirable physical properties ability to dissolve large amount of polymer (i.e. lignin acetate) and extremely low solubility in water. Birnbaum and his group (2000) stated that lower solubility of the organic solvent in water and higher solubility of the polymer in the organic solvent normally leads to more spherical and better size distribution of polylactic glycolic acid (PLGA) microparticles 27. They reported 57μm and 74μm for an average particle size of PLGA microparticles (with 5% drug loading) in DCM and EA, respectively.
Table 3. physical properties of DCM and EA 26
Mathematical models have been proposed for the solvent diffusion/evaporation from an open vessel system 22, 32, 33. The model of diffusion/evaporation helps to understand the process of lignin microspheres formation. Li et. al., (2008) in their review paper discussed the two main steps in the process of microspheres formation (Figure 6a): solvent diffusion from the dispersed phase (drops) to the continuous phase (F1) and solvent evaporation from continuous phase into the air (F2) 22. Microspheres will be solidified by diffusion of the organic solvent from the drops to the continuous phase, and evaporation from the continuous phase. In the first step (F1), organic solvent diffuses inside the drop and then it diffuses at the boundary to the continuous phases (Fig. 6b). At the boundary, the mass transfer of solvent causes the decrease in the size of the drop. Mainardes and Evangelista (2005) stated that the rate of diffusion of the organic solvent through the boundary is inversely proportional to the particle size 34. Therefore, the smaller particles are obtained at higher diffusion rates.
Figure 6. Schematic of a) solvent diffusion/evaporation steps and b) mass transfers during solidification of microsphere (Adopted from 22)
Li et al., (2008) also explained the three stages of the solvent concentration in the solvent diffusion/evaporation process 22. At the first stage, the dispersed phase is rich in solvent therefore the solvent is rapidly diffused into the continuous phase. This stage is very short (few seconds), but it is involved in the processes of droplet coalescence and agglomeration due to high volume of dispersion phase 34. In the second stage, the concentration of solvent in the continuous phase (Cs) remains constant due to replacing of the evaporated solvent by diffusing of the solvent from dispersed phase into the continuous phase. During the final stage, the polymer concentration in the continuous phase increases by decreasing the diffusivity of solvent from the dispersed phase to the continuous phase 22.
Based on Fick’s law and by assumption of zero solvent concentration above the surface of the continuous phase, the solvent evaporation would be 22:
where M is the total mass of solvent in the vessel (kg), Awa is the surface area of water-air interface (m2), t is time (s), K is evaporation constant (m/s), Cs is concentration of solvent in the continuous phase (kg/m3). In the second stage, Cs would be equal to the solubility of solvent in the continuous phase (Csol) 22. Equation 2 shows that the solvent evaporation is directly proportional to the solubility of organic solvent in continuous phase.
Therefore, during the solvent evaporation process, the viscosity of the dispersed droplets is increased by decreasing the dispersion volume 35. It is important to note that smaller particles are obtained in the shorter time of the solvent evaporation by reducing the possibility of coalescence, especially at the beginning of the solvent removal 34. Since EA takes longer time than DCM to evaporate from the reactor, solidification takes longer time, as a result, ACL-EA microspheres are formed in bigger size than ACL-DCM microspheres.
In general, higher molecular weight of the polymer results in higher rate of solidification due to low solubility in organic solvent and higher viscosity of polymer dispersion 19, 36. Li and workers (2008) stated that the particle size is exponentially increased by increasing the molecular weight of the polymer due to increasing the viscosity of the organic phase 22. For instance, the particle size of PLGA microspheres tended to increase from 3.3μm to 5.1μm with increasing the molecular weight of PLGA polymer from 53KDa to 64KDa 37. Soppimath and Aminabhavi (2002) stated that higher Mw polymers produce microspheres with larger size than the low Mw 21. They found that the PLGA particle size increased from 8μm to 14μm when the Mw increased from 12KDa to 34KDa. In another study, the mean particle size of PLGA microspheres from a series of PLGA with different Mw; RG 502 H (Mw 13.6KDa), RG 502 (Mw 17.2KDa) RG 503 H (Mw 35.7KDa), and RG 504 (Mw 48.3KDa) were found to be 3.98, 4.56, 6.53 and 8.74 μm, respectively 19. Ravi et al., (2008) stated that the molecular weight of the polymers was the main factor related to the particle size of the microspheres. They found that there was a statistically significant difference in the size of the microspheres among the polymers but no statistically significant difference in the particle size of microspheres between RG502 and RG502H due to the small difference in their molecular weight of polymers 19.
In our previous papers we reported the molecular weight of lignin from the four sources and their solubility in EA as well as the solubility of lignin and lignin acetate in DCM and EA (Table 4) 24, 28. Although the lignins from the four sources have different molecular weights, the formation of lignin acetate microspheres in DCM was independent of the molecular weight of the lignin and the size of the particles was found in narrow range (1280-1350 nm). On the other hand, ACL-EA microspheres were formed in a larger and wider range of size (1313-2545 nm) than ACL-DCM microspheres. It is can be seen that the particles size of ACL-EA microspheres depends on the Mw of the lignin (except ACL3-EA). The particles size of ACL1-EA was obviously larger than other ACL-EA microspheres due to greater molecular weight of L1 than other lignins. Therefore, molecular weight has an important influence on the formation of the microspheres when EA use as dispersing solvent in ESE method. It is also important to note that all L2 microspheres were successfully formed in all three cases (ACL2-DCM, ACL2-EA and L2-EA) due to lowest molecular weight of L2 than other lignins.
All ACL-DCM and ACL-EA microspheres were apparently formed in a uniform shape, narrow size distribution and high yield, while L-EA microspheres were partially obtained with uniform shape and low yield. Our previous study showed higher solubility of all lignins in EA than DCM (Table 4). Therefore, an effort has been made to fabricate lignin microspheres from the EA soluble portion of lignin. Formation of the lignin microspheres is directly proportional to the solubility of the lignin in EA. A major drawback of the L1-EA and L3-EA microspheres was aggregation due to low solubility and low concentration of L1 and L3 in organic phase.
Stability of the lignin microspheres suspension
Table 5 illustrates the particle size and PDI of ACL-DCM, ACL-EA and L-EA microspheres after 60 days in aqueous suspension. A very significant increase in particle size was observed for almost all samples (p<0.001), except L2-based microspheres which exhibited the most stable formulations; ACL2-EA (P>0.01), ACL2-DCM (p>0.001) and L2-EA (p>0.001). ACL2-EA was found to be the most stable colloid in this study due to lowest ζ-potential and smallest particle size than other lignin microspheres.
The stability of colloidal system is related to the surface charge of the particles. Particles with high ζ-potential value lead to a stable system, whereas a low ζ-potential value results in particle aggregation. The ζ-potential represents an index for colloid stability where aggregation is prevented by electrostatic repulsion of the particles. Generally, ζ-potential from 0 to ±5mV indicates rapid coagulation or flocculation, from ±10 mV to ±30 mV represents incipient instability, and from ±30 mV to ±60mV (and more) represents sufficient repulsion energy to result in stable condition with no agglomeration 38. In this study, the ζ-potential of ACL-DCM produced in the range of ±21 to ±27 and ACL-EA in the range of ±43 to ±48 and L-EA was ±22 to ±24. In some cases, a higher PDI could be attributed to propensity to aggregate or swelling of the microspheres. When aggregation of microparticles occurred in a solution, the resulting values for the average particle size were artificially high 27.
Uniform lignin acetate microspheres can be successfully prepared from a number of different lignin sources through ESE method by using either DCM or EA. Molecular weight of lignin is a critical parameter when lignin acetate microspheres are prepared in EA. In addition, lignin microspheres (non-acetylated lignin) can be prepared from the EA soluble portion of hardwood kraft lignin and non-wood soda lignin. Formation of lignin microsphere depends on the solubility of lignin in the organic solvent which is an essential factor in lignin microsphere formation. Lignin microspheres were successfully prepared from L2 (lignin isolated from hardwood kraft black liquor) in all three conditions (ACL2-DCM, ACL2-EA and L2-EA) due to its low molecular weight and highest solubility in EA. Particle size of the microspheres shows greater value over 60 days in neutral aqueous solution due to swelling and agglomeration of the microspheres.
The authors greatly acknowledge Ontario Research Fund- Research Excellence (ORF-RE) for providing financial support.
2. Lievonen, M.; Valle-Delgado, J. J.; Mattinen, M.-L.; Hult, E.-L.; Lintinen, K.; Kostiainen, M. A.; Paananen, A.; Szilvay, G. R.; Setälä, H.; Österberg, M., A simple process for lignin nanoparticle preparation. Green Chemistry 2016, 18, (5), 1416-1422.
5. Zhang, W.; Ma, Y.; Wang, C.; Li, S.; Zhang, M.; Chu, F., Preparation and properties of lignin–phenol–formaldehyde resins based on different biorefinery residues of agricultural biomass. Industrial Crops and Products 2013, 43, 326-333.
8. Thakur, V. K.; Thakur, M. K.; Raghavan, P.; Kessler, M. R., Progress in green polymer composites from lignin for multifunctional applications: a review. ACS Sustainable Chemistry & Engineering 2014, 2, (5), 1072-1092.
11. Fernandez-Perez, M.; Villafranca-Sanchez, M.; Flores-Cespedes, F.; Daza-Fernandez, I., Ethylcellulose and lignin as bearer polymers in controlled release formulations of chloridazon. Carbohydr. Polym. 2011, 83, (4), 1672-1679.
16. Chen, N.; Dempere, L. A.; Tong, Z., Synthesis of pH-Responsive Lignin-Based Nanocapsules for Controlled Release of Hydrophobic Molecules. ACS Sustainable Chemistry & Engineering 2016, 4, (10), 5204-5211.
17. Stewart, H.; Golding, M.; Matia-Merino, L.; Archer, R.; Davies, C., Manufacture of lignin microparticles by anti-solvent precipitation: Effect of preparation temperature and presence of sodium dodecyl sulfate. Food Research International 2014, 66, 93-99.
19. Ravi, S.; Peh, K.; Darwis, Y.; Murthy, B. K.; Singh, T.; Mallikarjun, C., Development and characterization of polymeric microspheres for controlled release protein loaded drug delivery system. Indian journal of pharmaceutical sciences 2008, 70, (3), 303-309.
21. Soppimath, K.; Aminabhavi, T., Ethyl acetate as a dispersing solvent in the production of poly (DL-lactide-co-glycolide) microspheres: effect of process parameters and polymer type. Journal of microencapsulation 2002, 19, (3), 281-292.
26. Sah, H., Microencapsulation techniques using ethyl acetate as a dispersed solvent: effects of its extraction rate on the characteristics of PLGA microspheres. Journal of Controlled Release 1997, 47, (3), 233-245.
27. Birnbaum, D. T.; Kosmala, J. D.; Henthorn, D. B.; Brannon-Peppas, L., Controlled release of β-estradiol from PLAGA microparticles:: The effect of organic phase solvent on encapsulation and release. Journal of Controlled Release 2000, 65, (3), 375-387.
29. Silva, A. L.; Rosalia, R. A.; Sazak, A.; Carstens, M. G.; Ossendorp, F.; Oostendorp, J.; Jiskoot, W., Optimization of encapsulation of a synthetic long peptide in PLGA nanoparticles: Low-burst release is crucial for efficient CD8+ T cell activation. European Journal of Pharmaceutics and Biopharmaceutics 2013, 83, (3), 338-345.
30. Sameni, J.; Krigstin, S.; Sain, M., Effect of Preparation Parameters on the Formation of Lignin Acetate Microspheres. International Journal of Engineering and Innovative Technology 2015, 4, (8), 102-113.
33. Li, W.-I.; Anderson, K. W.; Deluca, P. P., Kinetic and thermodynamic modeling of the formation of polymeric microspheres using solvent extraction/evaporation method. Journal of controlled release 1995, 37, (3), 187-198.
34. Mainardes, R. M.; Evangelista, R. C., PLGA nanoparticles containing praziquantel: effect of formulation variables on size distribution. International Journal of Pharmaceutics 2005, 290, (1–2), 137-144.
35. Lamprecht, A.; Ubrich, N.; Yamamoto, H.; Schäfer, U.; Takeuchi, H.; Lehr, C.-M.; Maincent, P.; Kawashima, Y., Design of rolipram-loaded nanoparticles: comparison of two preparation methods. Journal of Controlled Release 2001, 71, (3), 297-306.
37. Witschi, C.; Doelker, E., Influence of the microencapsulation method and peptide loading on poly (lactic acid) and poly (lactic-co-glycolic acid) degradation during in vitro testing. Journal of controlled release 1998, 51, (2), 327-341.