Polymer 46 (2005) 10750–10757

www.elsevier.com/locate/polymer

Poly(ethylene oxide) and its blends with sodium alginate

Tuncer Çaykaraa,*, Serkan Demircia, Mehmet S. Eroğlub, Olgun Güvenc
a
Department of Chemistry, Faculty of Art and Science, Gazi University, 06500 Besevler, Ankara, Turkey
b
Department of Chemical Engineering, Marmara University, 34722 Göztepe, Istanbul, Turkey
c
Department of Chemistry, Hacettepe University, 06532 Beytepe, Ankara, Turkey

Received 1 February 2005; received in revised form 18 July 2005; accepted 2 September 2005
Available online 26 September 2005

Abstract
A series of blends based on poly(ethylene oxide) (PEO) and sodium alginate (NaAlg) were prepared by solution casting method. The blends
thus obtained were characterized by using Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA), differential
scanning calorimetry (DSC), tensile strength test, contact angle measurements and atomic force microscopy (AFM). FT-IR studies indicate that
there are the hydrogen bonding interactions due to the ether oxygen of PEO and the hydroxyl groups of NaAlg. The thermal stability of the blends
was slightly affected with increasing NaAlg content. DSC results showed that both melting point and crystallinity depend on the composition of
the blends. Mechanical properties of the blend films were improved compared to those of homopolymers. Surface free energy components of the
blend films were calculated from contact angle data of various liquids by using Van Oss–Good methodology. It was found that the surfaces both of
the blends are enriched in low surface free energy component, i.e. NaAlg. This conclusion was further confirmed by the AFM images observation
of the surface morphology of these blends.
q 2005 Elsevier Ltd. All rights reserved.

Keywords: Miscibility; Poly(ethylene oxide); Sodium alginate

1. Introduction processability, and their production or recovery cost are very

high [5]. Therefore, biologically polymeric important materials
The importance of polymeric blends has been increased in based on the blends of synthetic and natural polymers have
recent years because of the preparation of the polymeric been prepared, such as poly(N-vinyl-2-pyrrolidone)-kappa-
materials with desired properties, low basic cost, and improved carrageenan (PVP/KC), poly(N-vinyl-2-pyrrolidone)-iota-car-
processability. Polymeric blends are physical mixtures of rageenan (PVP/IC) [6], poly(ethylene oxide)-hydroxypropyl
structurally different polymers or copolymers which interact methylcellulose (PEO/HPMC) [7], poly(vinyl alcohol)-chit-
with secondary forces with no covalent bonding such as osan (PVA/C) [8].
hydrogen bonding, dipole–dipole forces and charge-transfer Alginates, a naturally occurring polysaccharide obtained
complexes for homopolymer mixtures [1–4]. from marine brown algae, comprising linear chain of (1,4)-b-D-
The blend materials from either synthetic or natural mannuronic acid and (1,3)-a-L-guluronic acid [9]. Sodium
polymers alone are not always able to meet all the complex alginate (NaAlg), a polyelectrolyte having rigid molecular
demands of the biomaterials. The success of synthetic chain [10], and good film forming ability, has been extensively
polymers as biomaterials relies on their wide range of exploited and studied in detail on biomedical applications as a
mechanical properties, transformation processes that allow a drug carrier [11,12].
variety of different shapes to be easily obtained, and at low Generally, the formation of specific intermolecular inter-
production cost. Biological polymers present good biocompat- actions through hydrogen bonding between two or more
ibility, but their mechanical properties are often poor. The polymers is responsible for the observed mixing behaviors and
necessity of preserving biological properties complicates their properties of the blends prepared from aqueous solutions [13].
The study of the blends properties is of importance to explore
further applications of the resulting blends for biomedical and
* Corresponding author. pharmaceutical devices. Poly(ethylene oxide) [PEO] is a
E-mail address: caykara@gazi.edu.tr (T. Çaykara). unique class of water-soluble, aerobically biodegradable
0032-3861/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. thermoplastic [14,15]. Due to its excellent biocompatibility
doi:10.1016/j.polymer.2005.09.041 and very low toxicity, the potential use of PEO in biomedical
 T. Çaykara et al. / Polymer 46 (2005) 10750–10757 10751

applications has attracted a great deal of attention from both the 2.4. Thermogravimetric analysis (TGA)
industrial and scientific points of view [16–18]. Studies by
Kondo et al. have established that the primary hydroxyl group The dynamic weight loss tests were conducted on a TA
on cellulose and methylcelluloses can form a hydrogen bond to instrument 2050 thermogravimetric analyzer (TGA). All tests
ether oxygen in PEO [7]. Similarly, hydroxyl groups on sodium were conducted in a N2 purge (25 mL/min) using sample
alginate can also form a hydrogen bond to the ether oxygen in weights of 5–10 mg over a temperature range 20–600 8C at a
PEO. So, PEO as a suitable candidate blended with NaAlg was scan rate of 10 8C/min.
therefore selected.
Theoretically, the miscibility of polymer blends is mainly
determined by the chemical structure, composition and 2.5. Differential scanning calorimetry (DSC)
molecular mass of each component. In some cases, the
preparation conditions of the blends are also decisive. The glass transition temperatures of the blends were
Experimentally, various techniques have been used to determined by use of a TA instrument DSC 2010 thermal
characterize the miscibility of polymer blends, such as optical analyzer system. DSC was calibrated with metallic indium
transparency, Fourier transform infrared spectroscopy (FT-IR), (99.9% purity). All polymers were tested in crimped aluminum
electron microscopy, differential scanning calorimetry (DSC), pans at a heating rate of 10 8C/min under dry N2 gas
dynamic mechanical thermal analysis (DMTA) and high- (25 mL/min) over a temperature range from 20 to 175 8C.
resolution solid state 13C nuclear magnetic resonance (NMR) Melting temperature was taken as the peak of the melting
[19,20]. Each technique bears intrinsic limitations in terms of endoterm. The error in each measurement was estimated to be
sensitivity of detectable phase domain size. Therefore, the G0.5 8C.
phase separation observed by one of these techniques cannot
provide a perfect reflection of the thermodynamic definition of 2.6. Mechanical measurements
miscibility.
In this study, we focused on the binary blends of PEO and The stress–strain measurements of the blends were
NaAlg to investigate their miscibility as a function of blend performed by a AG-A electron tensiletester (Schimadzu Co.)
composition by the techniques of FT-IR, thermogravimetric
in the environment of 22 8C by using a crosshead speed of
analysis (TGA), DSC, mechanical testing, contact angle
5 mm/min. The rectangular samples with dimensions at
measurements and atomic force microscopy (AFM). There-
25 mm!10 mm!40G7 mm were analyzed at room tempera-
after, the miscibility, intermolecular interactions and some
ture. At least three samples were used for all mechanical
fundamental morphological and structure property correlations
measurements.
in PEO/NaAlg blends are addressed.

2.7. Contact angle measurements

2. Experimental
The contact angles of water, glycerol, ethylene glycol,
2.1. Materials
formamide and paraffin drops on the polymer films were
Sodium alginate (high viscosity) and poly(ethylene oxide) measured with a Model G-III Contact Angle Meter (Kernco
(MWZ300,000 g/mol) were obtained from Sigma Chemical Instrument Co. Inc., El Paso, TX, USA). The one-liquid
Co. and Aldrich, respectively, and used as received. method (air–liquid drop-polymer system) was used. All
measured contact angles were the average of three
measurements.
2.2. Preparation of the films

Aqueous solutions of the individual polymers (1% w/v) 2.8. Atomic force microscopy measurements
were mixed to obtain the desired proportions and stirred for
30 min at room temperature. The pure and mixture solutions The microscopic image of the blends was determined by
both transparent (wt% of NaAlg; 9, 20, 33, 43, 50) were cast on an atomic force microscopy (AFM; Nanoscope IIIa, Digital
Petri dishes by water evaporation at 25 8C. The films were Instruments, Santa Barbara, CA, USA) with a nanoprobe
dried under vacuum at 60 8C for 10 days. The dried films of 200 mm in length and a pyramidal oxide-sharpened silicon
thickness ranging 40G7 mm were obtained. nitride cantilever with a spring constant of 0.12 N/m. The
opening angle of tip was 458. The amplitudes used of
2.3. Fourier transform infrared spectroscopy (FT-IR) the drive signal applied to the cantilever oscillation were in
the between 0.5 and 2 V. The scan rates ranged from 0.8 to
FT-IR spectra of the blends were measured on a Nicolet 520 1 Hz. Tapping mode of operation was used to eliminate
FT-IR spectrophotometer. The samples were prepared by shear forces that may damage the films and reduce the
making KBr (potassium bromide) pellets containing 3 wt% of image resolution. Images (2!2 mm2) were undertaken in air
materials. at 25 8C.
10752 T. Çaykara et al. / Polymer 46 (2005) 10750–10757

Fig. 1. The FT-IR spectra of the pure and blend films (4000–2500 and 1800–700 cmK1).

3. Results and discussion at 1611 and 3500 cmK1 were observed in all spectra of the
blends. The spectrum of the PEO/NaAlg blend films shows a
3.1. FT-IR significant difference in the region of the C–O–C asymmetric
stretch at 1100 cmK1. The blend films that have undergone the
Fourier transform infrared (FT-IR) spectroscopy of blend step transition show a broader C–O–C band compared with the
films was carried out in order to detect any peak shift that could pure components that have not. This broadening results in a
be attributed to weak interactions between the two polymers, band shift to lower wavenumber. The change in the C–O–C
such as hydrogen bonding or complexation. The FT-IR spectra band in the spectrum, suggests that hydrogen bonding is the
of the interpolymer complexes show spectral features similar to underlying mechanism in the interaction. In addition, hydrogen
those for the homopolymers, but the bands appear at shifted bonding has the strongest influence on the donor (in our case
positions. Hydrogen bonds are formed between the proton- the –OH of NaAlg) and the absorption maxima of stretching
donor and proton-acceptor molecules. The intensity of the vibration shifts toward lower wavenumbers compared to that
hydrogen bond band depends on the acidity of the hydrogen in for the pure NaAlg. It is also noticed that the hydroxyl
the proton-donor, the alkalinity of the proton-acceptor and stretching bands became much more broad with increasing
possibility of their close contacts. As a consequence of NaAlg content. This strongly supports the idea that a hydrogen
hydrogen bonding, the covalent bonds in the donor and bonding can form between ether oxygen atoms of PEO and
acceptor are weaker, while the energy barrier for angle hydroxyl groups of NaAlg (Scheme 1).
deformation becomes higher. Hence, in the groups which are
involved in the hydrogen bonding formation, frequency of the
valence vibrations decrease with the simultaneous increase in
the frequency of the deformation vibrations.
Fig. 1 shows the FT-IR spectra of the pure and blend films in
the wavelength ranges of 4000–2500 and 1800–700 cmK1. The
characteristic band of PEO was observed at 843 cmK1 due to
the C–O–C bending. On the other hand, the bands of NaAlg
appeared at 3500 cmK1 for the hydroxyl groups and at 1613
and 1415 cmK1 for the asymmetric –COOK stretching
vibration and symmetric –COOK stretching vibration, respect-
ively. The spectrum of the PEO/NaAlg blend films was
characterized by the presence of the absorption bands typical of
the pure components, with the intensity roughly proportional
the blending ratio. The characteristic bands of NaAlg appeared Scheme 1.
 T. Çaykara et al. / Polymer 46 (2005) 10750–10757 10753

3.2. TGA

Typical weight loss (TG) and derivative of weight loss

(DTG) curves of PEO, NaAlg and PEO/NaAlg blends were
presented in Figs. 2 and 3. From the TG curves initial
degradation temperature and final degradation temperature
were determined. From DTG curves, the maximum tempera-
ture of weight loss was also noted.
The mass loss of pure PEO begins at 360 8C and reaches to
maximum at 395 8C. However, the mass loss of pure NaAlg
starts at 229 8C and reach to maximum at 243 8C. The PEO
shows better thermal stability than that of NaAlg. The TG
curves of both NaAlg and PEO also indicate one reaction stage
(Fig. 2) which is reflected as single peak in the DTG curves
(Fig. 3). However, PEO/NaAlg blends degrade in two steps.
This is evidenced by the appearance of distinct peaks in DTG
thermograms. Two distinct reaction peaks at around 240 and
408 8C are identified in the DTG thermograms of PEO/NaAlg
blends. These peaks were attributed to thermal degradation of

Fig. 3. The DTGA curves of the pure and blend films.

NaAlg and PEO, respectively. This behavior showed that the

thermal degradation reaction mechanism of PEO/NaAlg blends
is the same as pure PEO and NaAlg homopolymers. On the
other hand, the initial thermal degradation temperatures of the
blends were slightly affected with increasing NaAlg. This may
be corresponded to the formation of hydrogen bonding from
the ether oxygen of PEO and the hydroxyl groups of NaAlg in
the PEO/NaAlg blends.

3.3. DSC

Fig. 4 shows the DSC curves of NaAlg, PEO and their

blends. From these curves, crystallinity (Xc) of PEO in the
blends was calculated by the following equation.
DH
Xc Z (1)
Fig. 2. The TGA curves of the pure and blend films. fw DH0
10754 T. Çaykara et al. / Polymer 46 (2005) 10750–10757

72 67

66 66

60 65

%Xc

Tm
54 64

48 63

42 62

36 61
0 10 20 30 40 50
wt% NaAlg

Fig. 5. The effect of the NaAlg content on melting point and crystallinity.

3.4. Mechanical properties

The results of the stress–strain measurements of NaAlg and

PEO/NaAlg blend films were presented in Fig. 6. By casting
from aqueous solutions, pure PEO rendered slightly opaque
and brittle films and it was not possible to measure its
mechanical properties. NaAlg, on the other hand, is a very
strong material with a great stress and Young modulus. For the
blends, the incorporation of NaAlg made the films stronger.
The trend of the stress–strain values of the blend films is in
good agreement with the results from FT-IR, TGA and DSC
and hydrogen bonding is advantageous to improving the
mechanical properties. When the mechanical properties of pure
components were compared with their blends, it was observed
that, the Young modulus and the stress and the elongation at
break increased with increasing NaAlg content. This may be
due to the intermolecular hydrogen bonding between NaAlg
and PEO, such as those reported for the blend films of silk
fibroin/NaAlg [22] and poly(acrylamide)/NaAlg [23].

3.5. Surface free energy analysis

Contact angles are characteristic constants of liquid–solid

systems and provide valuable information on the surface
energies of solids. The contact-angle values of paraffin oil,
water, glycerol, ethylene glycol, formamide drops on the
surface of PEO, NaAlg and their blend films with different
Fig. 4. The DSC curves of the pure and blend films.

where DH0Z18,810 J/g is the heat of fusion for 100%

crystalline PEO [21], DH is the heat of fusion of the blend
and fw is the weight fraction of PEO in the corresponding blend.
The Xc and melting points (Tm) values were shown in Fig. 5. It
was notable that both the Tm and Xc values decreased with
increasing NaAlg content when NaAlg content is less than 33%
and then increased. Both of these parameters have minimum
values when the blend contains 33% NaAlg. The decrease of
both the Tm and Xc values may be due to the stiff molecular
chain of NaAlg, which has a significant effect on the overall
chain mobility in the mixture and retards the rate of crystal
growth. Fig. 6. Stress–strain curves of the pure and blend films.
 T. Çaykara et al. / Polymer 46 (2005) 10750–10757 10755

Table 1
Contact angle results of PEO/NaAlg

NaAlg (wt%) Paraffin Water Glycerol Ethylene glycol Formamide

0 10.0G0.0 48.0G0.4 50.0G0.0 35.0G0.0 60.0G0.0
9 20.8G0.4 52.8G0.4 53.7G0.5 35.9G0.7 47.8G0.8
20 25.0G0.5 57.0G0.0 60.9G0.9 38.2G0.4 43.0G0.6
33 29.0G0.6 62.3G0.5 65.9G0.4 39.1G0.6 34.3G0.5
43 34.0G0.0 66.2G0.4 70.0G0.0 45.0G0.0 30.0G0.0
50 39.2G0.8 70.0G0.0 74.0G0.6 47.1G0.7 27.8G0.4
100 46.1G0.4 78.0G0.5 68.0G0.8 47.2G0.4 25.0G0.6

NaAlg content are shown in Table 1. The surface free energy are used, two equations of the form of Eq. (1) constitute a set of
components of these blend films containing various weight two simultaneous equations which can be solved for the two
percentage of NaAlg were determined from the contact angle unknown properties of solid gK s and gs . Then the gs and
C K

data of polar and apolar liquids by using the following gs results are averaged for a single value. Water drop contact-
C

complete Young equation comprising both the apolar and polar angle values are always used in these sets in order to avoid
interactions [24,25] large discrepancies between simultaneous equation solutions
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffi [26].
gL ð1 C cos qÞ Z 2 gLWs gs C
LW gK s gs C
C gC s gs
K (2) The contact-angle values of the liquids used were evaluated
to determine the surface free energy components of
In the above equation, the superscript LW represents the PEO/NaAlg blend films with various NaAlg content by using
apolar Lifshitz–Van der Waals forces. Subscripts L and S refer Van Oss–Good methodology thorough Eqs. (1) and (2). The
to liquid and solid, respectively. gL is the total surface free surface free energy component values of the liquids used were
energy of the liquid, q is the contact angle of liquid drop on the taken from Ref. [26] and given in Table 2.
solid surface, and gLW
L and gs
LW
are the apolar Lifshitz–Van der The LW component of the surface free energy of the blend
Waals components of liquid and solid, respectively. gC s and gL
C
films were calculated by using Eq. (2) and the paraffin oil drop
are the electron acceptor surface free energy components of contact-angle values and given in Table 3. As seen in this table,
solid and liquid, respectively. gKs and gL are the electron donor
K
the gLW values of the blend films decreased with increasing
s
surface free energy components of solid and liquid, respect- NaAlg content. This can be explained on the basis that the
ively. gLW
L can be determined first by using an apolar liquid NaAlg units of the blends have a higher dispersion force than
LW
(paraffin). For an apolar liquid, gC L Z gL Z 0 and gL Z gL
K
that of the PEO units. It is well known that there is an inverse
hence, the last two terms of the right hand side of Eq. (1)
relationship between intermolecular distance and the dis-
become zero, this equation can therefore be written in the form
persion forces present.
gL ð1 C cos qÞ2 s and gs values of these blends were calculated using the
gC K
gLW
s Z (3) general contact-angle equation (Eq. (1)), where previously
4
found gLW s values were inserted. Water–glycerol, water–
Consequently, the gLW s value can be determined directly from
ethylene glycol, and water–formamide sets were simul-
apolar liquid (paraffin) contact angle. When two polar liquids
taneously solved and the average results were also given in
Table 2 Table 3. As shown in this table, the PEO has a highly basic
Surface free energy component values of the liquids used [18] (mJ/m2) 2 2
character (gK s Z 40:99 mJ=m O gs Z 1:37 mJ=m ), whereas
C

Liquid gTOT
L gLW
L gC
L gK
L
surface of the NaAlg has a slightly acidic character
Paraffin 28.9 28.9 0 0 1.100
Water 72.8 21.8 25.5 25.5
Glycerol 64.0 34.0 3.92 57.4
Ethylene glycol 48.0 29.0 1.92 47.0 1.075
Formamide 58.0 39.0 2.28 39.6

1.050
γCalc/ γExp

Table 3
Surface free energy components of PEO/NaAlg blend films (mJ/m2)
1.025
NaAlg (wt%) gLW
s gC
s gK
s gTOT
s

0 28.46 1.37 40.99 80.26

1.000
9 27.05 1.49 30.72 78.52
20 26.26 1.81 26.14 75.59
33 25.29 3.09 19.38 72.74
43 24.17 3.91 16.49 69.70 0 10 20 30 40 50 60
50 22.76 4.85 13.76 66.83 wt% NaAlg
100 20.72 7.71 5.40 63.48
Fig. 7. The change of gCalc/gExp ratios of the blends with wt% NaAlg.
10756 T. Çaykara et al. / Polymer 46 (2005) 10750–10757

Fig. 8. AFM images of (a) pure PEO, (b) PEO/NaAlg (9 wt% NaAlg), (c) PEO/NaAlg (50 wt% NaAlg).
2
(gC
s Z 7:71 mJ=m O gK
s Z 5:40 mJ=m2 ). Generally, the PEO/ 4. Conclusions
NaAlg blends have also a highly basic character and the gK s
values decreased with increasing NaAg in the blends. The Blend films of PEO and NaAlg could be easily obtained
decrease of both gLW s and gs values also reflected in the
K
over the whole composition range from water solutions by
decrease of total surface energy (gTOT s ) of the blends with solution blending and casting onto the glass plate. Mechanical
increasing NaAlg content. In addition, the PEO has a higher properties of the blends were enhanced relative to those of PEO
gTOT
s value than that of NaAlg (Table 3). Thus, it is the reason and NaAlg. The enhancement is caused by the existence of
for NaAlg enhanced on the surface of all blend systems. It is specific intermolecular interactions between PEO and NaAlg in
well known that some surface enrichment of the low surface the blend. FT-IR analysis revealed this interaction from the
free energy component takes place in the blend systems [27]. shift and change of intensity of –OH and C–O–C bands. The
We have also tried to calculate the total surface free energy thermal stability of the blends was slightly affected depending
(gTOT
s ) of the blends from the following equation, assuming on NaAlg component. NaAlg was found to hinder the
additivity rule; crystallization of PEO during casting, and to reduce the
stability (i.e. melting temperature) of crystals. The results of
gTOT ðBlendÞ Z f1 gTOT ð1Þ C f2 gTOT ð2Þ (4) AFM showed microscopic phase separation and islands on the
s s s
surfaces of the blends because of the aggregation of NaAlg
content. The phase separation increased with an increase of
where f1 and f2 denote the weight percentage of PEO and NaAlg content. The surface free energy of blends decreased
NaAlg in the blend, gTOT
s ð1Þ and gTOT
s ð2Þ are the total surface with increasing NaAlg component to a considerable extent. It
energies of pure PEO and NaAlg, respectively. The ratios of was determined that the casting from water solutions had an
the calculated total surface energy values to the experimental effect on localizing the remaining blend compositions on the
total surface energy values (gCalc/gExp) increased with surface of the films different from the bulk composition. Both,
increasing NaAlg in the blend, which is plotted in Fig. 7. As
contact angle measurements and AFM investigations have
shown in this figure, the gCalc/gExp values of these blends are
proved that the surface of the blends is enriched in NaAlg
always higher than 1. This is mainly due to increasing of the
component.
component with the low surface free energy of blends at the
surface.
Acknowledgements

3.6. AFM O.G. acknowledges the support of TÜBA, the Academy of

Sciences of Turkey. The authors thank to Dr M.M. Demir from
The AFM images of PEO and PEO/NaAlg films are shown Sabancı University for AFM measurements. This work was
in Fig. 8. The surface morphology characteristics of PEO film supported by State Planning Organization of Turkey (2003 K
and its blends with NaAlg are observed to depend on 120 470-31 DPT).
composition. As shown in Fig. 8, the PEO film has relatively
uniform surface structure, while its blends with NaAlg start to
References
show two separated phases. This observation agrees very well
with the contact angle observations. Although the micro- [1] Krause S. Polymer blends. New York: Academic Press; 1979.
separation could not be observed for low NaAlg containing [2] Varnell DF, Coleman MM. Polymer 1981;22:1324.
blends, the separation becomes more evident at 50% mixtures. [3] Varnell DF, Runt JP, Coleman MM. Polymer 1983;24:37.
The boundary between two phases developed with increasing [4] Woo EM, Barlow JW, Paul DR. J Appl Polym Sci 1986;32:3889.
[5] Maria GC. Polym Int 1997;43:55.
concentrations of NaAlg and at 50% composition small islands
[6] Relleve L, Yoshii F, Rosa DA, Kume T. Angew Makromol Chem 1999;
of NaAlg formed at the surface. This is in good agreement with 273:63.
the results obtained by contact angle measurements, where an [7] Kondo T, Sawatari C. Polymer 1994;35(20):4423.
enrichment of NaAlg on the blend surface was observed. [8] Miya M, Iwamoto R. J Polym Sci, Part B: Polym Phys 1984;22:1149.
 T. Çaykara et al. / Polymer 46 (2005) 10750–10757 10757

[9] Moe ST, Skjak-brek G, Ichijo H. J Appl Polym Sci 1994;51:1773. [19] Cowie JMG. Encylopedia of polymer science and engineering. New
[10] Hirano S, Mizutani C, Yamaguchi K, Miura O. Biopolymers 1978;17: York: Wiley; 1989.
805. [20] Walsh DJ. Comprehensive polymer science. New York: Pergamon Press;
[11] Yuk SH, Cho SH, Lee HB. J Controlled Release 1995;7:69. 1989.
[12] Hari PR, Chandy T, Sharma CP. J Appl Polym Sci 1996;59:1795. [21] Kim KM, Chiou JS, Barlow JW, Paul DR. Polymer 1987;28:1721.
[13] Sun Y, Shao Z, Hu P, Liu Y, Yu T. J Polym Sci, Part B: Polym Phys 1997; [22] Liang CX, Hirabayashi K. J Appl Polym Sci 1992;45:1937.
35:1405. [23] Xiao C, Lu Y, Liu H, Zhang L. J Macromol Sci Pure Appl Chem 2000;37:
[14] Bailey FE, Koleske JV. Poly(ethylene oxide). New York: Academic 1663.
Press; 1976. [24] Van Oss CJ, Good RJ. Langmuir 1992;8:2877.
[15] Ramachandran R. Plast Eng 1996;52:31. [25] Erbil HY. Langmuir 1994;10:276.
[16] Desai NP, Hubbell JA. Biomaterials 1991;12:144. [26] Van Oss CJ. Interfacial forces in aqueous media. New York: Marcel
[17] Desai NP, Hubbell JA. Macromolecules 1992;25:226. Dekker; 1994.
[18] Lee JH, Kopecek J, Andrade JD. J Biomed Mater Res 1989;23:351. [27] Çaykara T, Eroğlu MS, Güven O. J Appl Polym Sci 1998;69:1551.