Role of the molecular weight and the composition on the hydrolysis kinetics of monolayers of poly(α-hydroxy acid)s

The hydrolysis kinetics of spread insoluble monolayers of poly(α-hydroxy acid)s with various molecular weights and lactic acid–glycolic acid molar ratios was followed by measuring simultaneously the decrease in the surface area at constant surface pressure and the evolution of the surface potential. The interfacial hydrolysis at alkaline pH leads to the progressive fragmentation of the polymer molecules and the appearance of charged insoluble products (detected by measuring the surface potential) and small soluble fragments (detected by measuring the decrease in the surface area). The data obtained by both approaches were interpreted in the framework of the random scission model. The rates of hydrolysis are larger for polymers with smaller initial polymerization numbers and increase with the decrease in the molar ratio of lactic acid units. Received: 7 December 1998 Accepted in revised form: 8 March 1999     

Reorganization of lipid nanocapsules at air-water interface 3. Action of hydrolytic enzymes HLL and pancreatic PLA2

The action of the hydrolytic enzymes humicola lanuginosa lipase (HLL) and pancreatic phospholipase A2 (PLA2) on monolayers formed from lipid nanocapsules (LNC) and model monolayers containing their components, Labrafac, Solutol and Lipoid, is studied by simultaneous measuring the changes in the film area and the surface potential in the "zero order" trough at constant surface pressure (pi). The kinetic models describing the hydrolysis by HLL of the Labrafac, Solutol and their mixtures have been proposed. By using the developed theoretical approach together with the experimental results the surface concentrations of the substrates, hydrolysis products and values of the global kinetic constants were obtained. The comparison between the global kinetic constants in the case of HLL hydrolysis of pure Labrafac, Solutol monolayers and those of the model mixed Labrafac/Solutol monolayers, shows that the rates of hydrolysis are of the same order of magnitude, i.e. an additively of the HLL enzyme action is observed. The composition of the mixed Labrafac/Solutol monolayer, formed after the interfacial LNC destabilization, was estimated.     

Comparative study of lipolysis by PLA2 of DOPC substrates organized as monolayers, bilayer vesicles and nanocapsules

The water-soluble lipolytic enzymes act at the interface of insoluble lipid substrates, where the catalytical step is coupled with various interfacial phenomena as enzyme penetration, solubilization of reaction products, loss of mechanical stability of organized assemblies of phospholipids molecule, etc. One biologically relevant example is the enzymatic hydrolysis of DOPC by PLA(2), which results in cleavage of phospholipids molecules into water insoluble lipolytic products, namely oleic acid and lysophospholipid. In general, the enzymatic activity depends on the substrate organization and molecular environment of the catalytic reaction. The lipolysis by phospholipase A(2) of dioleoylphosphatidylcholine substrates organized as monolayer, bilayers vesicles and lipid nanocapsules was studied by measuring the decrease of the surface area at constant surface pressure or increase of the surface pressure at constant area at air-water interface. A kinetic model describing the coupling of the catalytic act with corresponding interfacial phenomena was developed. By using the kinetic model the values for the global hydrolytic kinetic constants were obtained. The obtained value for the monolayer is five orders of magnitude higher than this obtained with small unilamellar vesicles and six orders of magnitude higher then those obtained with lipid nanocapsules. The comparison shows that the enzymatic catalytic act occurring in the lipid environment of the monolayer is more efficacious than at the vesicle and nanocapsules interfaces.     

Action of Humicola lanuginosa lipase on long-chain lipid substrates: 1. Hydrolysis of monoolein monolayers

The hydrolysis of 1-monoolein (MO) monomolecular films by Humicola lanuginosa lipase (HLL) was followed by measuring the simultaneous decrease with time in the film area and the surface potential using a ‘zero order trough’ at constant surface pressure (Verger and de Haas, Chem. Phys. Lipids 10 (1973) 127). The decrease with time in the film area reflects both the reduction in the area per molecule as well as the solubilization of the substrate and the product molecules during the transformation of the substrate MO into product of oleic acid (OA). The surface potential changes were interpreted as the results of the accumulation at the interface of negatively charged OA. The surface concentration of OA transiently present at the interface was determined by the surface pressure and the surface potential measurements on the basis of a developed kinetic model. In the proposed model we have taken into account the product and substrate solubilization rates in the presence of β-cyclodextrin (β-CD) as well as the flux supplied progressively by the moving barrier from the reservoir to the reaction compartment in order to keep the constant surface pressure. Values of the global kinetic constant Q_m were obtained. The selective lipolytic product acceptor, β-CD, accelerated considerably the hydrolytic process.     

'Stealth' corona-core nanoparticles surface modified by polyethylene glycol (PEG): influences of the corona (PEG chain length and surface density) and of the core composition on phagocytic uptake and plasma protein adsorption

Nanoparticles possessing poly(ethylene glycol) (PEG) chains on their surface have been described as blood persistent drug delivery system with potential applications for intravenous drug administration. Considering the importance of protein interactions with injected colloidal dug carriers with regard to their in vivo fate, we analysed plasma protein adsorption onto biodegradable PEG-coated poly(lactic acid) (PLA), poly(lactic-co-glycolic acid) (PLGA) and poly(-caprolactone) (PCL) nanoparticles employing two-dimensional gel electrophoresis (2-D PAGE). A series of corona/core nanoparticles of sizes 160–270 nm were prepared from diblock PEG-PLA, PEG-PLGA and PEG-PCL and from PEG-PLA:PLA blends. The PEG Mw was varied from 2000–20 000 g/mole and the particles were prepared using different PEG contents. It was thus possible to study the influence of the PEG corona thickness and density, as well as the influence of the nature of the core (PLA, PLGA or PCL), on the competitive plasma protein adsorption, zeta potential and particle uptake by polymorphonuclear (PMN) cells. 2-D PAGE studies showed that plasma protein adsorption on PEG-coated PLA nanospheres strongly depends on the PEG molecular weight (Mw) (i.e. PEG chain length at the particle surface) as well as on the PEG content in the particles (i.e. PEG chain density at the surface of the particles). Whatever the thickness or the density of the corona, the qualitative composition of the plasma protein adsorption patterns was very similar, showing that adsorption was governed by interaction with a PLA surface protected more or less by PEG chains. The main spots on the gels were albumin, fibrinogen, IgG, Ig light chains, and the apolipoproteins apoA-I and apoE. For particles made of PEG-PLA45K with different PEG Mw, a maximal reduction in protein adsorption was found for a PEG Mw of 5000 g/mole. For nanospheres differing in their PEG content from 0.5 to 20 wt %, a PEG content between 2 and 5 wt % was determined as a threshold value for optimal protein resistance. When increasing the PEG content in the nanoparticles above 5 wt % no further reduction in protein adsorption was achieved. Phagocytosis by PMN studied using chemiluminescence and zeta potential data agreed well with these findings: the same PEG surface density threshold was found to ensure simultaneously efficient steric stabilization and to avoid the uptake by PMN cells. Supposing all the PEG chains migrate to the surface, this would correspond to a distance of about 1.5 nm between two terminally attached PEG chains in the covering ‘brush’. Particles from PEG5K-PLA45K, PEG5K-PLGA45K and PEG5K-PCL45K copolymers enabled to study the influence of the core on plasma protein adsorption, all other parameters (corona thickness and density) being kept constant. Adsorption patterns were in good qualitative agreement with each other. Only a few protein species were exclusively present just on one type of nanoparticle. However, the extent of proteins adsorbed differed in a large extent from one particle to another. In vivo studies could help elucidating the role of the type and amount of proteins adsorbed on the fate of the nanoparticles after intraveinous administration, as a function of the nature of their core. These results could be useful in the design of long circulating intravenously injectable biodegradable drug carriers endowed with protein resistant properties and low phagocytic uptake.     

Colloidal stability and drug incorporation aspects of micellar-like PLA–PEG nanoparticles

The drug delivery properties of a series of poly(lactic acid)–poly(ethylene glycol) (PLA–PEG) micellar-like nanoparticles have been assessed in terms of their colloidal stability and their ability to incorporate a water soluble drug. These studies have focused on a range of PLA–PEG copolymers with a fixed PEG block (5 kDa) and a varying PLA segment (3–110 kDa). In aqueous media, these copolymers formed micellar-like assemblies following precipitation from water miscible solvents. There was a controlled increase in the particle size as the molecular weight of the PLA block was increased. The characteristics of the PEG corona were also highly dependent on the PLA moiety. Copolymers with a low molecular weight PLA block (3–15 kDa) formed highly colloidally stable dispersions, with a complete PEG surface coverage. However, increasing the molecular weight of the PLA block resulted in significantly less colloidally stable nanoparticle dispersions, which flocculated in solvents that were significantly better than θ-solvents for the stabilising PEG chains. This can be attributed to a reduced PEG surface coverage and the probable presence of naked PLA ‘patches’ on the particle surface. These larger PLA–PEG nanoparticles (30:5–110:5) were found to be stabilised in the presence of serum components, which are thought to adsorb into the gaps on the particle surface and prevent flocculation. All of the dispersions were found to be stable under physiological conditions and therefore suitable for in vivo administration. A reasonable loading (3.1% w/w) of the micellar-like PLA–PEG 30:5 nanoparticles with the water soluble drug procaine hydrochloride was achieved. The incorporated drug was found to have no effect on the nanoparticle structure or recovery, which can be attributed to the micellar character of these assemblies and the presence of the stabilising PEG chains.     

Action of Humicola lanuginosa lipase on long-chain lipid substrates: 2. Hydrolysis of diolein monolayers

The hydrolysis of 1,2-diolein (DO) monomolecular films by Humicola lanuginosa lipase (HLL) was studied by simultaneous measuring the decrease in the film area and the changes in the surface potential in the “zero-order trough” at constant surface pressure and in the presence of β-cyclodextrin (β-CD). The decrease with time in the film area reflects both the reduction in the area per molecule due to the transformation of substrate DO molecules into the products molecules of monoolein (MO) and oleic acid (OA) and the desorption of the soluble inclusion complexes β-CD–MO and β-CD–OA. The surface potential data were interpreted as an accumulation at the interface of negatively charged products of OA and insoluble β-CD–DO complexes. In the proposed kinetic model, the product solubilization rates in the presence of β-CD and the flux supplied progressively by the moving barrier from the reservoir to the reaction compartment in order to keep the constant surface pressure were taken into account. The surface concentrations of MO and OA transiently present at the interface were determined. The values of the global kinetic constant Q_m′ of hydrolysis of DO to MO were obtained. Comparison with the values of the global kinetic constant of hydrolysis of monoglyceride MO to OA shows that the rates of hydrolysis of diglyceride and monoglyceride by HLL are of the same order of magnitude.     

Action of Humicola lanuginosa lipase on mixed monomolecular films of tricaprylin and polyethylene glycol stearate

The hydrolysis catalyzed by Humicola lanuginosa lipase (HLL) of pure tricaprylin (TC) or stearate of polyethylene glycol 1500 (PEG-St) as well as their mixtures spread as monomolecular films were studied. The catalytic transformation of the two substrates TC or PEG-St into their respective reaction products was detected by measuring simultaneously the decrease in the film area and the surface potential using the "zero order" trough at constant surface pressure. A kinetic model describing the enzymatic hydrolysis was developed. The surface concentrations of the two substrates and their respective reaction products as well as the values of the global kinetic constants of hydrolysis were determined. The experimentally obtained global kinetic constants of the catalytic action of HLL against TC and PEG-St present in mixed monolayers of TC/PEG-St are approximately the same as in the case of pure monolayers. These obtained results give some indications that the activity of enzyme is not significantly affected by the different molecular environments in the mixed monolayers.     

Synthesis of novel graft polyhydroxyalkanoates

Graft copolymers were synthesized by direct condensation of methoxy-poly(ethylene glycol) (MePEG) or methoxy-poly(lactic acid) (MePLA) onto a reactive polyhydroxyalkanoate (PHA) backbone in organic solvent. Side carboxylic groups of the PHA were coupled with end hydroxyl groups of MePEG or MePLA in the presence of N,N′-dicylohexylcarbodiimide (DCC). Graft copolymers were characterized by ¹H NMR spectroscopy and size exclusion chromatography (SEC). NMR spectra of PHA-g-PEG and PHA-g-PLA showed the presence of significant amounts of PEG and PLA, respectively. No noticeable unreacted PEG or PLA were detected in SEC chromatograms. Grafting of hydrophilic polymers chains as PEG or biodegradable oligomers as PLA onto PHA backbone will generate polyesters with a more rapid water uptake and faster biodegradation rates. These PHA polymers conjugates could be interesting for bioactive agent delivery systems.     

Recent advances in the field of lactic acid/glycolic acid polymer-based therapeutic systems

The mechanism of degradation of bioresorbable lactic acid- and glycolic acid-based aliphatic polyesters (PLA/GA) is still far from being totally understood although a majority of authors agree to consider in vivo degradation as essentially hydrolytic. In the past years, we have shown that hydrolysis of large size devices is heterogeneous, i.e. faster inside than at the surface because of reaction-diffusion phenomena involving acid-catalyzed ester cleavage reactions and water-soluble macromolecular fragments. The proposed mechanism is recalled together with some direct or indirect consequences, namely induced crystallization at body temperature, size- and formulation-dependence of the degradation rates. From these features, one can conclude that it is now possible to increase or decrease the degradation rate and to turn on or off heterogeneous degradation by using accelerating or braking factors, however, accurate piloting of the hydrolytic degradation of PLA/GA is still not feasible.     

Savinase proteolysis of insulin Langmuir monolayers studied by surface pressure and surface potential measurements accompanied by atomic force microscopy (AFM) imaging

The mechanism of the enzymatic action of Savinase on an insulin substrate organized in a monolayer at the air-water interface was studied. We followed two steps experimental approach classical surface pressure and surface potential measurements in combination with atomic force microscopy imaging. Utilizing the barostat surface balance, the hydrolysis kinetic was followed by measuring simultaneously the decrease in the surface area and the change of the surface potential versus time. The decrease in the surface area is a result of the random scission of the peptide bonds of polypeptide chain, progressively appearance of amino acid residues, and their solubilization in the aqueous subphase. The interpretation of the surface potential data was based on the contribution of the dipole moments of the intact and broken peptide groups which remain at the interface during the proteolysis. An appropriate kinetic model for the Savinase action was applied, and the global kinetic constant was obtained. The application of the AFM revealed the state of the insulin monolayers before and after the Savinase action. The comparison of the topography of the films and the roughness analysis showed that insulin Langmuir-Blodgett (LB) films transferred before the enzyme action were flat, while at the end of hydrolysis, roughness of films has increased and the appearance of 3D structures was observed.     

Physicochemical evaluation of PLA nanoparticles stabilized by water-soluble MPEO-PLA block copolymers

Different water-soluble MPEO-PLA diblock copolymers with various alpha-methoxy-omega-hydroxyl polyethylene (MPEO) and poly(lactic acid) (PLA) block lengths have been synthesized. Their surface-active properties were evidenced by surface tension (water/air) measurements. In each case the surface tension leveled down above a critical polymer concentration, which was attributed to the formation of a dense polymer layer at the liquid-air interface. The applicability of copolymers as emulsion stabilizers in the preparation of PLA nanospheres by an o/w emulsion/evaporation technique was then investigated. Four copolymers presenting sufficient water solubility and good surfactive properties were used to prepare PLA nanospheres with MPEO chains firmly anchored at the particle surface. The effect of polymer concentration in emulsion on particle size and surface coverage was examined. Whatever the copolymer characteristics, it was found that the optimal concentration to obtain a large amount of MPEO at the particle surface was similar (around 2 g/l). The effect of the copolymer composition on MPEO layer characteristics and on colloidal stability was also evaluated. The conformation of MPEO blocks at the PLA particle surface is discussed in relation to the layer thickness and the surface area occupied per molecule.     

Role of the electrostatic interactions on the basic or acidic hydrolysis kinetics of poly-( , -lactide) monolayers

The hydrolysis kinetics of insoluble poly-( , -lactide) monolayers spread on basic or acidic aqueous subphase were followed by measuring simultaneously the decrease in the surface area at constant surface pressure and the evolution of the surface potential. An approach to analyse the role of the electrostatic interactions during the hydrolysis at alkaline pH, interpreting the surface potential data was developed. The theoretical predictions based on the idea of a random fragmentation of polymer molecules leading to the interfacial accumulation of charged insoluble products and solubilisation of small fragments describes well the experimental results. The reversibility of the hydrolysis/esterification reaction at acidic pH is taken into account.     

Role of the electrostatic interactions on the basic or acidic hydrolysis kinetics of poly-(d,l-lactide) monolayers

The hydrolysis kinetics of insoluble poly-(d,l-lactide) monolayers spread on basic or acidic aqueous subphase were followed by measuring simultaneously the decrease in the surface area at constant surface pressure and the evolution of the surface potential. An approach to analyse the role of the electrostatic interactions during the hydrolysis at alkaline pH, interpreting the surface potential data was developed. The theoretical predictions based on the idea of a random fragmentation of polymer molecules leading to the interfacial accumulation of charged insoluble products and solubilisation of small fragments describes well the experimental results. The reversibility of the hydrolysis/esterification reaction at acidic pH is taken into account.     

Bulk and surface modifications of polylactide

This article reviews various methods of modifying the bulk and surface properties of poly(lactic acid) (PLA) so that the polymer may be used as a drug carrier in a drug delivery system (DDS) and as a cell scaffold in tissue engineering. Copolymerization of lactide with other lactone-type monomers or monomers with functional groups such as malic acid, copolymerization of lactide with macromolecular monomer such as poly(ethylene glycol) (PEG) or dextran, as well as blending polylactide and natural derivatives and other methods of bulk modification are discussed. Surface modifications of PLA-type copolymers, such as surface coating, chemical modification, and plasma treatment are described. Cell culture technology proves the efficiency of bulk and surface modification and the potential application of PLA in tissue engineering.     

FT‐IR characterization and hydrolysis of PLA‐PEG‐PLA based copolyester hydrogels with short PLA segments and a cytocompatibility study

A series of the biodegradable copolyester hydrogels was prepared using a redox‐initiated polymerization with a constant 1:9 mole ratio of the Boltorn‐based acrylate and diacrylate triblock comacromonomers. The Boltorn® macromonomer was derived from the hyperbranched polyester Boltorn H20, which was functionalized at each terminus with poly(ethylene glycol) acrylate, and the diacrylate triblock macromonomer was poly (lactide‐b‐ethylene glycol‐b‐lactide) diacrylate. The hydrolysis of the copolyesters at pH 7.4 in a phosphate buffered saline solution at 37 °C was studied using ATR‐FTIR spectroscopy. It was found that the presence of the Boltorn, the PEG, and lactide block lengths both play vital roles in determining the structure‐property relationships in these materials. The ATR‐FTIR studies showed that with increasing lactide segment length, the rate of ester hydrolysis increased due to the increased concentration of the hydrolytically sensitive poly(lactic acid) (PLA) ester groups in the network. However, incorporation of Boltorn into the PLA‐PEG‐PLA copolymer did not significantly change the kinetic rate constant for hydrolysis of the PLA segments. The cytocompatibility of a typical one of these materials in the presence of its degradation by‐products was assessed using cultured osteoblasts from the rat. The hydrogel was degraded for 28 days and found to be cytocompatible with osteoblasts over days 23 to 28 of the hydrolysis period. © 2013 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2013 , 51, 5163–5176     

Synthesis and pyrolysis of ABC type miktoarm star copolymers with polystyrene,poly(lactic acid) and poly(ethylene glycol) arms

An ABC type miktoarm star copolymer possessing polystyrene (PS), poly(lactic acid) (PLA) and poly(ethylene glycol) (PEG) arms was synthesized by combining Atom Transfer Radical Polymerization (ATRP) and Ring Opening Polymerization (ROP) with two click chemistries, namely thiol–ene and copper catalyzed azide–alkyne cycloaddition (CuAAC). For this purpose, a core 1-(allyloxy)-3-azidopropan-2-ol with allyl and azide functionalities was synthesized in two steps. Then, clickable polymers, polystyrene with thiol functionality (PS–SH) and poly(ethylene glycol) with alkyne functionality (PEG–acetylene) were independently prepared. As the first step of the grafting onto process, PS–SH was thiol–ene clicked onto the core to yield PS–N₃–OH. The second arm was then incorporated onto the core by the Ring Opening Polymerization (ROP) of l-(?)-Lactide (LA) using as PS–N₃–OH initiator and tin(II) 2-ethylhexanoate as catalyst. Finally, alkyne–PEG–acetylene was bonded to the resulting PLA–PS–N₃ using CuAAC click reaction. All intermediates, related polymers at different stages and final PS–PLA–PEG miktoarm star copolymer were characterized by ¹H NMR, FT-IR, SEC and DP-MS analyses. Direct pyrolysis mass spectrometry, (DP-MS) analyses of PS–PLA–PEG and all intermediate polymers indicated that the decomposition of PS and PEG chains occurred almost independently, following the degradation mechanisms of the corresponding homopolymers. On the other hand, during the pyrolysis of PS–PLA–PEG, elimination of H₂O during the decomposition of PEG chains at the early stages of pyrolysis caused hydrolysis of PLA chains and increased the yields of CO₂, CO and units involving unsaturation and/or crosslinked structure.     

High-Toughness Poly(lactic Acid)/Starch Blends Prepared through Reactive Blending Plasticization and Compatibilization

In this study, poly(lactic acid) (PLA)/starch blends were prepared through reactive melt blending by using PLA and starch as raw materials and vegetable oil polyols, polyethylene glycol (PEG), and citric acid (CA) as additives. The effects of CA and PEG on the toughness of PLA/starch blends were analyzed using a mechanical performance test, scanning electron microscope analysis, differential scanning calorimetry, Fourier-transform infrared spectroscopy, X-ray diffraction, rheological analysis, and hydrophilicity test. Results showed that the elongation at break and impact strength of the PLA/premixed starch (PSt)/PEG/CA blend were 140.51% and 3.56 kJ·m⁻², which were 13.4 and 1.8 times higher than those of pure PLA, respectively. The essence of the improvement in the toughness of the PLA/PSt/PEG/CA blend was the esterification reaction among CA, PEG, and starch. During the melt-blending process, the CA with abundant carboxyl groups reacted in the amorphous region of the starch. The shape and crystal form of the starch did not change, but the surface activity of the starch improved and consequently increased the adhesion between starch and PLA. As a plasticizer for PLA and starch, PEG effectively enhanced the mobility of the molecular chains. After PEG was dispersed, it participated in the esterification reaction of CA and starch at the interface and formed a branched/crosslinked copolymer that was embedded in the interface of PLA and starch. This copolymer further improved the compatibility of the PLA/starch blends. PEGs with small molecules and CA were used as compatibilizers to reduce the effect on PLA biodegradability. The esterification reaction on the starch surface improved the compatibilization and toughness of the PLA/starch blend materials and broadens their application prospects in the fields of medicine and high-fill packaging.     

Poly(Lactic Acid)-Based Graft Copolymers: Syntheses Strategies and Improvement of Properties for Biomedical and Environmentally Friendly Applications: A Review

As a potential replacement for petroleum-based plastics, biodegradable bio-based polymers such as poly(lactic acid) (PLA) have received much attention in recent years. PLA is a biodegradable polymer with major applications in packaging and medicine. Unfortunately, PLA is less flexible and has less impact resistance than petroleum-based plastics. To improve the mechanical properties of PLA, PLA-based blends are very often used, but the outcome does not meet expectations because of the non-compatibility of the polymer blends. From a chemical point of view, the use of graft copolymers as a compatibilizer with a PLA backbone bearing side chains is an interesting option for improving the compatibility of these blends, which remains challenging. This review article reports on the various graft copolymers based on a PLA backbone and their syntheses following two chemical strategies: the synthesis and polymerization of modified lactide or direct chemical post-polymerization modification of PLA. The main applications of these PLA graft copolymers in the environmental and biomedical fields are presented.     

Preparation and characterization of biodegradable poly(l-lactide)/poly(ethylene glycol) microcapsules containing erythromycin by emulsion solvent evaporation technique

In this work, the producing of a biodegradable poly(l-lactide) (PLA)/poly(ethylene glycol) (PEG) microcapsule by emulsion solvent evaporation method was investigated. The effect of PEG segments added to the PLA microcapsules on the degradation, size distribution, and release behavior was studied. According to the results, PLA/PEG copolymer was more hydrophilic than PLA homopolymer, and with lower glass transition temperature. The surface of PLA/PEG microcapsules was not as smooth as that of PLA microcapsules, the mean diameters of prepared PLA and PLA/PEG microcapsules were 40 and 57 microm, respectively. And spherical forms were observed by the image analyzer and the scanning electron microscope (SEM). Drug release from microcapsules was affected by the properties of PLA/PEG copolymers determined by UV-vis spectra. It was found that the drug release rates of the microcapsules were significantly increased with adding of PEG, which explained by increasing hydrophilic groups.