The effect of end groups of PEG on the crystallization behaviors of
binary crystalline polymer blends PEG/PLLA
Wei-Chi Lai
a, Wen-Bin Liau
a,b,*, Tai-Tso Lin
aa
Department of Materials Science and Engineering, National Taiwan University, 1 Roosevelt Rd. Sec.4,Taipei 106, Taiwan, ROC
b
Institute of Polymer Science and Engineering, National Taiwan University, 1 Roosevelt Rd. Sec.4,Taipei 106, Taiwan, ROC Received 30 December 2003; received in revised form 26 February 2004; accepted 3 March 2004
Abstract
The effect of end groups (2OH, 1OH, 1CH3and 2CH3) of poly(ethylene glycol) (PEG) on the miscibility and crystallization behaviors of
binary crystalline blends of PEG/poly(L-lactic acid) (PLLA) were investigated by differential scanning calorimetry (DSC) and polarizing optical microscopy (POM). A single glass-transition temperature was observed in the DSC scanning trace of the blend with a weight ratio of 10/90. Besides, the equilibrium melting point of PLLA decreased with the increasing PEG. A negative Flory interaction parameter, x12;
indicated that the PEG/PLLA blends were thermodynamically miscible. The spherulitic growth rate and isothermal crystallization rate of PEG or PLLA were influenced when the other component was added. This could cause by the change of glass transition temperature, Tgand
equilibrium melting point, Tm0: The end groups of PEG influenced the miscibility and crystallization behaviors of PEG/PLLA blends. PLLA
blended with PEG whose two end groups were CH3exhibited the greatest melting point depression, the most negative Flory interaction
parameter, the least fold surface free energy, the lowest isothermal crystallization rate and spherulitic growth rate, which meant better miscibility. On the other hand, PLLA blended with PEG whose two end groups were OH exhibited the least melting point depression, the least negative Flory interaction parameter, the greatest fold surface free energy, the greatest isothermal crystallization rate and spherulitic growth rate.
q2004 Elsevier Ltd. All rights reserved.
Keywords: Blends; Miscibility; Crystallization
1. Introduction
Binary polymer blends can be classified into amorphous/ amorphous, crystalline/amorphous, and crystalline/crystal-line systems based on the crystallizability of the constitu-ents. Most of them focused on the polymeric mixture
containing two amorphous components[1 – 3]. On the other
hand, polymer blends containing two crystalline com-ponents are more complicated and interesting since both components are able to crystallize. Furthermore, they may crystallize in different temperature regimes and within different periods of time. Thus, two crystalline polymer blends can provide various conditions to study the crystal-lization behavior and morphology in polymer blends. Recently, Liau et al.[4,5]and Qiu et al.[6 – 11]have done
some work on the miscibility and crystallization in crystal-line/crystalline polymer blends.
Both poly(ethylene glycol) (PEG) and poly(L-lactic acid) (PLLA) are very interesting and important crystalline polymers. PEG is soluble in water and many organic solvents. Meanwhile, PEG shows hydrophilicity and biocompatibility. PLLA is a biodegradable thermoplastic polyester and has attracted increasing attention due to their potential applications as biomedical and environment-friendly materials. In the present study, the binary blends of a biodegradable polymer, PLLA, with a biocompatible polymer, PEG, have been investigated. Younes and Cohn
[12] have reported the miscibility and crystallization
behavior of PEO/PLLA blends. According to them, melting point depression of PLLA occurred with increasing the PEO component, especially at higher content of PEO (. 80%
PEO). Nakafuka et al. [13,14] also has studied the PEO/
PLLA polymer blend. It was found that the melting point of PLLA was depressed and different molecular weight of PEO 0032-3861/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2004.03.003
www.elsevier.com/locate/polymer
* Corresponding author. Address: Department of Materials Science and Engineering, National Taiwan University, 1 Roosevelt Rd. Sec.4, Taipei 106, Taiwan, ROC. Tel.: þ 886-2-2362-6119; fax: þ 886-2-3363-4562.
or PLLA did affect the miscibility and crystallization behavior of PEO/PLLA blends. However, Sheth and Kumar
[15]have concluded DSC and DMA results that PEG/PLA
blends range from miscible to partially miscible, depending
on the composition. Nijenhuis et al.[16]has proposed the
thermal and physical properties of high molecular weight PLLA and PEO blends. All blends containing up to 50 wt% PEO showed single glass transition temperature. Judging from the melting point depression of the blends, this system was miscible over the whole composition range in the melt.
Yang et al. [17] has mentioned the PEG/PLLA blend
systems were miscible in the melt and the crystallization behavior of PLLA could be influenced by the addition of PEG. From the above research, it was found the different molecular weight or blend compositions would make different miscibility and crystallization behavior of PEG/ PLLA blends. In this work, the effect of end groups of PEG on the crystallization behaviors of PEG/PLLA blends was studied. Three kinds of PEG with different end groups were studied. They were PEG with two OH end groups (PEG-2OH),
PEG with two CH3end groups (PEG-2CH3), and PEG with
one OH and one CH3end groups (PEG-1OH-1CH3).
Besides, in some earlier reports they have considered primarily the effect of blending with PEG on the crystal-lization behavior of PLLA, and hence they mainly concern with a crystalline/amorphous blend system. When PEG/ PLLA blends are cooled to the temperature below the melting point (, 60 8C) of PEG, both PEG and PLLA could crystallize, and the blend would turn into a crystalline/ crystalline system whose behavior is more complicated than the crystalline/amorphous system. Since the investigation on crystalline/crystalline blends is quite limited, PEG/PLLA blends appear to be an interesting system for such a study. In this study, we will also discuss the different crystallization conditions of PEG/PLLA blends.
2. Experimental
2.1. Materials and preparing method of sample
The PLLA sample used in this study was obtained from Polysciences Co., and its weight-average molecular weight
was 200,000.PEG(2OH), PEG(1OH-1CH3) and PEG(2CH3)
obtained from Aldrich Co. had a weight-average molecular weight of 2000.
Preparation of PEG and PLLA blends was carried out by solution-casting method. The blending components were dissolved in chloroform yielding a 2% (0.4 g polymer blends/20 ml solvent) solution. The solution was sub-sequently poured onto a glass dish. A film was obtained after evaporating most chloroform solvent very slowly under ambient condition at room temperature. The film was then further dried in vacuum at 80 8C for 24 h. Also, TGA was used to check the residual solvent in the final films. The results showed no measurable residual solvent in the films.
2.2. Measurements
Thermal transitions of PEG/PLLA blends were measured with a TA instruments DSC 2010 differential scanning calorimetry (DSC). The samples were first heated up to 180 8C and held for 3 min, and then were cooled down to 2 100 8C. At a heating rate of 10 8C/min, the samples were heated from 2 100 to 200 8C, thus the glass transition temperature and enthalpy of melting would be observed.
The isothermal crystallization of PEG/PLLA blends was measured by TA DSC 2010 DSC. The samples were heated to 180 8C and held for 3 min on a Linkam THMS600 hot stage, and then quickly moved into the DSC cell, where the temperature was kept at crystallization temperature. After the crystallization peak completely showed up, isothermal crystallization of PEG/PLLA blends was observed.
The spherulitic growth of PEG/PLLA blends was observed by a Nikon HFX-DX polarizing optical micro-scope (POM). The samples were placed on cover glasses, heated to 180 8C and held for 3 min on a Linkam THMS600 hot stage. The samples were then quickly cooled to their crystallization temperature. Micrographs were taken at interval for measuring the spherulite radii at various time periods. The growth rate was calculated from the change of spherulite radius with time, dR=dt:
3. Results and discussion 3.1. Miscibility
Fig. 1(a) – (c) show DSC thermograms of PEG(2OH,
1OH-1CH3 and 2CH3)/PLLA blends. A single glass
transition temperature ðTgÞ was identified in the figure for blends with a PEG composition of under 10 wt%. For higher
compositions of PEG, the blend’s Tgcould not be observed.
For blends with high PLLA composition, adding PEG to PLLA increased the crystallization rate of PLLA and the PLLA crystals form during quenching, resulting in the less
amorphous part and the difficulty of Tg observation. Also,
the increase of PEG resulted in the crystallization of PEG and the melting of PEG and the glass transition of blend might be overlapped. Thus, for the blends with high PLLA
composition Tg could not be easily observed by DSC. On
the other hand, for the blends with high PEG composition, the strong crystallizability and high crystallization rate of PEG during quenching resulted in the less amorphous part and the difficulty in the observation of Tg:
The miscibility of polymer blends is usually determined from the observation of a single Tg: However, in this system, only a 10/90 blend exhibits a single composition-dependent
Tg: Other blends could not be determined for sure to be
miscible in the melt. Consequently, other methods are presented to verify the miscibility. For blends containing a crystalline polymer, the melting point depression is also an indication of a miscible system. The equilibrium melting
point was determined by Hoffman – Weeks [18] analysis. The relevant equation was written in the following form:
Tm¼ 1 gTcþ 1 2 1 g Tm0 ð1Þ
where Tmand Tm0 are the experimental melting temperature
and the equilibrium melting temperature of PLLA in the
blend, respectively.g is a proportional factor between the
initial thickness of a chain-folded lamella, lpg; and the final lamellar thickness, lc:
The equilibrium melting point of PLLA, T0
m; was
obtained from the extrapolation with the Tm¼ Tc line.
Fig. 2 plots Tm0 as a function of the weight fraction of
PEG(2OH,1OH-1CH3and 2CH3)/PLLA blends. Regardless
of end groups of PEG, the equilibrium melting point of PLLA declines as the PEG content increases. However, the reduction in the melting point of PLLA depends on end
groups of PEG. The PEG(2CH3) showed the greatest
reduction in the melting-point and the PEG(2OH) showed the least reduction. The equilibrium melting point of PLLA is reasonable in comparison with the data reported by
Nijenhuis [16](196 8C) but is lower than that reported by
Tsuji[19](212 8C) and Kalb[20](215 8C).
The melting point depression of a crystalline phase with non-crystalline polymeric diluent in a miscible blend was
derived by Nishi and Wang[21]. The relevant equation can
be written as: 1 T0 m 2 1 T00 m ¼ 2RV2 DH0 fV1 lnf2 M2 þ 1 M2 2 1 M1 f1 2 RV2 DH0 fV1 ðx12f21Þ ð2Þ
where V is the molar volume of the polymer repeating unit,cis the volume fraction of the component in the blend, DHf0is the perfect crystal heat of fusion of the crystallizable polymer, M is the degree of polymerization, R is the
Fig. 1. (a). DSC traces from second run of PEG(2OH)/PLLA blends. Arrows indicate Tg: (b). DSC traces from second run of
PEG(1OH-1CH3)/PLLA blends. Arrows indicate Tg: (c). DSC traces from second run
of PEG(2CH3)/PLLA blends. Arrows indicate Tg:
Fig. 2. T0
mas a function of weight fraction of PEG(2OH, 1OH-1CH3and
universal gas constant, Tm0 is the equilibrium melting point of pure crystalline polymer, T00
m is the equilibrium melting
point of a blend, andx12is the polymer/polymer interaction parameter. The subscripts 1 and 2 denote the amorphous and crystalline components, respectively. If the molecular weights of both components of blends are large enough, the entropy of mixing can be negligible and the melting point depression is dominated by an enthalpic term, then the equation reduces to:
1 T0 m 2 1 T00 m ¼ 2 RV2 DH0 fV1 ðx12f 2 1Þ ð3Þ
It is well known that experimental factors such as scanning rate, crystallization temperature range, and time of crystallization would affect the values obtained. However, the same experimental procedures were used for all blends. Therefore, the change ofx12 with different end groups still was meaningful. The following parameters were used
[20,22 – 24]: DHf0¼ 1:883 kcal/mol, V1¼ 24:16 cm3/mol, V2¼ 44:65 cm3/mol,r1¼ 1:22 g/cm3andr2¼ 1:27 g/cm3.
Table 1presents the x12 values of different end groups of
PEG for PEG/PLLA blends. Because the value of x12 is
negative, it confirms that the polymeric mixture is thermo-dynamically miscible in the melt. Besides, Nakafuku et al.
[14]also has discussed the miscibility of PEO and PLLA
was better as the molecular weight of PEO was smaller. In this system, the molecular weight of PEG is very small so it is believed they are miscible. Smallerx12values mean better miscibility. Thus, the PEG/PLLA blends with different end
groups in order of decreasing miscibility are PEG(2CH3)/
PLLA, PEG(1OH-1CH3)/PLLA and PEG(2OH)/PLLA.
Rozenberg et al.[25]examined the hydrogen bonding in
PEG-200, PEG-400, PEG(1OH)-350 and PEG(2CH3)-250
four systems using IR. They reported that, for the C – O – C
group of PEG, the band at 1104 cm21 will shift to
1127 cm21, due to the hydrogen bonding with the OH end
group. Hydrogen bonding will decrease with increasing molecular weight and fewer OH groups. Although, the molecular weight of PEG is 2000 in this study, the OH end group is expected to form the hydrogen bonding. When PEG(2OH) is blended with PLLA, the hydrogen bonding is expected to form between PEGs themselves, such that the miscibility between PEG and PLLA decreases. Therefore, it is expected the miscibility decreases with more OH end groups.
3.2. Crystallization behavior
For PEG/PLLA, four different crystallization situations could occur. First, PLLA crystallizes before the formation of PEG crystals. Second, PLLA crystallizes after the formation of PEG crystals. Third, PEG crystallizes before the formation of PLLA crystals. Last, PEG crystallizes following the formation of PLLA crystals. However, the crystallization temperature of PLLA is higher than the melting point of PEG and so the second situation is unable to study. Also, it was showed the crystallization of PLLA in blends could not be inhibited during such fast quenching from Fig. 1. Thus, the third situation is unable to study. Consequently, this study involves only two crystallization situations. One applies as the PLLA crystallization behavior is observed before PEG crystals are formed, and the other applies as PEG crystallization behavior is observed after PLLA crystals are formed.
DSC and POM instruments were used to elucidate the crystallization behavior of PEG/PLLA blends, including isothermal crystallization and spherulitic growth rate. 3.2.1. Isothermal crystallization
The kinetics of isothermal crystallization has been
analyzed in terms of the Avrami equation [26 – 28] using
the double logarithmic form:
log½2lnð1 2 XtÞ ¼ log Knþ n log t ð4Þ
where n is the Avrami exponent which is related to the geometry of the spherulitic growth and the mechanism of the nucleation. Kn is the overall kinetic rate constant. The time required to reach 50% crystallization is called half-time of crystallization and denoted as t1=2:
First, the isothermal crystallization of PLLA was
observed before PEG crystals were formed.Table 2presents
the values of n; Kn and t1=2 in the Avrami equation for
PEG(1OH-1CH3)/PLLA blends isothermally crystallized at
124 8C. Only PLLA can crystallize at this temperature. The value of t1=2first increases and then decreases as the PEG content increases. The dilution of PEG, which depresses the equilibrium melting point ðTm0Þ; reduces the driving force of crystallization ðTm0 2 TcÞ; and blending with PEG lowers the system’s glass transition temperature ðTgÞ; increasing the segmental mobility of PLLA. These two factors are such competitive that the isothermal crystallization rate of PLLA first increases and then decreases with the increase of PLLA.
Basically, if two polymers are immiscible, the values of n;
Kn and t1=2 in the Avrami equation should not be changed.
However, in this system these values change with the composition, reconfirming that PEG/PLLA blends is a
miscible system. For PEG(2OH)/PLLA and PEG(2CH3)/
PLLA blends, similar results are found, indicating that the isothermal crystallization rate of PLLA first increases and then decreases with the addition of PLLA.
With reference to the effect of different end groups of PEG on the isothermal crystallization of PLLA for Table 1
Interaction parameter of PEG/PLLA blends Polymers PEG(2OH)/ PLLA PEG(1OH-1CH3)/ PLLA PEG(2CH3)/ PLLA X12 2 0.048 2 0.144 2 0.161
PEG/PLLA blends,Table 3shows the values in the Avrami
equation for PEG(2OH, 1OH-1CH3and 2CH3)/PLLA 50/50
isothermally crystallized at 124 8C. The PEG(2OH)/PLLA blend system exhibits the fastest isothermal crystallization
rate of PLLA, followed by the PEG(1OH-1CH3)/PLLA
blend system, and the PEG(2CH3)/PLLA blend system
exhibits the slowest isothermal crystallization rate of PLLA.
The glass transition temperature ðTgÞ and equilibrium
melting point ðTm0Þ are main factors that influence the
crystallization rate of a polymer. In a PVAc/PEG3system, a
single composition-dependent Tg of PEG(2CH3) blended
with PVAc has been reported to be higher by 2 – 5 8C than that of PEG(2OH) blended with PVAc. Furthermore, the
difference between the Tgof these two end groups of PEG
blended with PVAc did not change with the composition. Thus, they concluded that the end groups of PEG did not affect Tg of the polymer blend too much. In this study, the
Tgs of different end groups of PEG blended with PLLA in a
10/90 ratio are very similar (Fig. 1). Hence, it is expected that Tm0 is the dominant factor on the crystallization rate of PLLA. The experimental results on the equilibrium
melting point depression of PLLA (Fig. 2) indicate that in
decreasing order of Tm0s of PLLA in blends are PEG(2OH)/
PLLA, PEG(1OH-1CH3)/PLLA and PEG(2CH3)/PLLA. As
the T0
m of PLLA is depressed, the degree of supercooling
ðT0
m2 TcÞ of PLLA is reduced. In general, the
crystal-lization rate increases first and then decreases with the increasing supercooling. The crystallization temperature (124 8C) is close to the melting point of PLLA. In this region, the crystallization rate of PLLA increases with the increasing supercooling. From the POM observation also
confirmed it. The PEG(2CH3)/PLLA blend system has the
lowest Tm0 of PLLA, and therefore the slowest crystallization rate of PLLA.
Second, the isothermal crystallization of PEG is observed after PLLA crystals are formed. Consider
PEG(1OH-1CH3)/PLLA, for example. Samples were first
heated to 180 8C, held for 3 min, and then cooled to 120 8C until the PLLA crystals formed completely. Then the
sample was quickly quenched to 35 8C to observe the isothermal crystallization of PEG.Table 4lists the values of n; Kn and t1=2 in the Avrami equation isothermally crystal-lized at 35 8C after the formation of PLLA crystals for
PEG(1OH-1CH3)/PLLA blends. FromTable 4, the
crystal-lization rate of PEG declines as the amount of PLLA increases and potentially determines three causes: (1) blending with PLLA lowers the segmental mobility of PEG as Tgincreases; (2) the dilution of PEG and the decline in the equilibrium melting point reduces the driving force of crystallization, and (3) the formation of PLLA crystal hinders the crystallization rate of PEG.
3.2.2. Spherulite growth rate
PEG is oxidized easily at moderate temperature[29 – 34]. Therefore, all the measurement is under the nitrogen atmosphere. First, the spherulitic growth rate of PLLA is observed before PEG crystals are formed. The spherulites of PLLA are not suppressed when the crystallization tempera-ture is less than 100 8C. Therefore, the experiment is carried out between 100 and 130 8C. The radial growth rate of PLLA spherulites is measured by observing the evolution of
POM images over time.Fig. 3plots the spherulitic growth
rate, G; for various compositions as a function of Tc for
PEG(1OH-1CH3)/PLLA blends. For the blend compositions
displayed inFig. 3, only PLLA crystallizes and PEG acts as
a non-crystallizable component at Tc: The variation of
PLLA spherulite radius with time (not shown) is linear in all cases, which means that the crystallization environment at Table 2
The values of n; Kn; t1=2of PEG(1OH-1CH3)/PLLA blends at 124 8C
Composition weight fraction, PEG(1OH-1CH3)/PLLA n Kn t1=2(s)
0/100 3.98 2.91 £ 10212 694.8
10/90 3.58 6.24 £ 10211 628.8
30/70 3.50 2.13 £ 10210 498.0
50/50 3.24 6.57 £ 10210 595.8
Table 3
The values of n; Kn;t1=2of PEG/PLLA 50/50 blends at 124 8C
Polymers n Kn t1=2(s) PEG(2OH)/PLLA 50/50 3.65 7.89 £ 10211 501.0 PEG(1OH-1CH3)/PLLA 50/50 3.24 6.57 £ 10 210 595.8 PEG(2CH3)/PLLA 50/50 3.16 2.32 £ 10 210
929.4 Fig. 3. The spherulitic growth rate G of PLLA versus crystallization temperature PEG(1OH-1CH3)/PLLA blends.
the growth front is same during the crystallization process. This result suggests that PEG and un-crystallized PLLA could be trapped in the intra-spherulitic regions.Fig. 3also shows that the spherulitic growth rate of PLLA is accelerated by increasing the amount of PEG. Although, the dilution of PLLA as well as depression in equilibrium melting point reduces the crystallization driving force, blending with PEG increases the segmental mobility of PLLA. Thus, the increased segmental mobility is assumed to be the major effect. Besides, for PEG(2OH)/PLLA and
PEG(2CH3)/PLLA blends, the spherulitic growth rate of
PLLA also increases with PEG content. The trend does not
like it showed in Table 2. However, the isothermal
crystallization kinetics investigated by DSC may represent the bulk crystallization rate, which involves both the nucleation density and the spherulitic growth rate. This can be seen from the following observation.
Fig. 4 displays the spherulitic growth rate, G; as a
function of Tcfor PEG(2OH,1OH-1CH3and 2CH3)/PLLA
blends with 50/50 composition. The PLLAs in decreasing order of spherulitic growth rate are PEG(2OH)/PLLA,
PEG(1OH-1CH3)/PLLA and PEG(2CH3)/PLLA, perhaps
determined also by the competition between Tm0 and Tg:
Again, the end group of PEG blended with PLLA does not affect Tg much. Therefore, Tm0 is the dominant factor. The
PEG(2CH3)/PLLA blend system has a greatly depressed
equilibrium melting point and the least degree of
super-cooling of PLLA. From Fig. 4, the crystallization rate of
PLLA falls on the decreasing side. Therefore, the spherulitic growth rate of PLLA is the slowest.
Besides, Fig. 5 shows the number of nuclei of PLLA
versus the weight fraction of PEG in PEG(2OH, 1OH-1CH3
and 2CH3)/PLLA blends isothermally crystallized at 120 8C
by POM. The number of nuclei of PLLA decreases with the increase of PEG regardless of end groups. Also, the number of nuclei of PLLA increases with more OH end groups when
the content of PEG is same. From Figs. 4 and 5, it is
expected that isothermal crystallization rate of PLLA first increases with PLLA and then decreases as more PLLA is added regardless of end groups. Furthermore, the
PEG(2CH3)/PLLA blend system has the fewest nuclei of
PLLA and exhibits the slowest spherulitic growth rate of PLLA, such that the isothermal crystallization rate of PLLA is also the slowest.
The spherulitic growth rate of PEG was observed after the PLLA crystals are formed. However, the birefringence of PEG interferes with that of PLLA spherulites and thus was not clearly observed using POM.
3.2.3. Fold surface free energy
The kinetic theory of polymer crystallization developed
by Hoffman et al. [35 – 37] has been used to analyze
experimental crystallization data concerning the spherulite growth rate. According to this theory, the dependence of the
growth rate Gm on the crystallization temperature Tc is
Table 4
The Values of n; Kn, t1=2of PEG(1OH-1CH3)/PLLA Blends at 35 8C After The Formation of PLLA Crystals
Composition weight fraction, PEG(1OH-1CH3)/PLLA n Kn t1=2(s)
100/0 3.53 2.59 £ 1027 69.6
70/30 3.58 8.26 £ 1028 90.6
50/50 3.65 2.72 £ 1028 110.4
30/70 3.85 6.30 £ 10210 111.6
Fig. 4. The spherulitic growth rate G of PLLA in PEG/PLLA 50/50 with different end groups of PEG.
Fig. 5. Number of nuclei of PLLA in PEG/PLLA blends with different end groups of PEG isothermally crystallized at 120 8C.
described by the following equation: Gm¼f2G0e 2DE RðTc2T01Þe 2DFpm kBTc : ð5Þ
where G0is a pre-exponential factor, generally assumed to
be constant or proportional to Tc; where, DE is the activation energy for the transport of the crystallizing units across the liquid – solid interface, DFmp is the free energy required to form a nucleus of critical size, T01¼ Tg2 30; R is the gas
constant, KB is the Boltzmann, and f2 is the crystalline
polymer volume fraction. According, to Flory [38] and
Mandelkerm[39], DFmp in Eq. (5) can be expressed as: DFmp ¼ nbsuse Dhuf 1 2 Tc T0 m 2 RTcx12 Dhuf V2u V1u ð1 2f2Þ2 ð6Þ
where Dhuis the enthalpy of fusion per unit volume, Tm0 is the equilibrium melting point of the crystalline polymer in
the blend, V1uand V2uare the molar volumes of component
1 (non-crystalline) and component 2 (crystalline),x12is the polymer/polymer interaction parameter, b is the layer
thickness, su and se are lateral and fold surface free
energies, n is a coefficient depends on the growth regime
(Hoffman [37]): n ¼ 4 in regime I and III, and n ¼ 2 in
regime II, and f ¼ 2Tc=Tm0 þ Tc:
Taking Eqs. (6) into (5), the following expression is obtained: Gm¼f2G0e 2DE RðTc2T01Þe 2nbsuse=kBTc Dhuf 12Tc T0 m 2RTcV2u DhufV1ux12 ð12f2Þ2 : ð7Þ
Eq. (7) is using the double logarithmic form, and simplifying it: a¼ 2suseb ð8Þ a¼ ln Gm2 lnf22 ln G0þ DE RðTc2 T01Þ ð9Þ b¼ nb=kBTc Dhuf DT T0 m 2 RTcV2u V1u x12ð1 2f2Þ2 ð10Þ
The plot of ðaþ ln G0Þ againstbwill give thesuse in the slope. In this PEG/PLLA system, we assume the growth
regime is regime II ðn ¼ 2Þ: Table 5 gives the values of
various parameters involved in Eqs. (9) and (10). Tg is
estimated by Fox’s equation. Tm0; Gm andx12 are obtained
from the above results. Fig. 6(a) – (c) display the L – H
plot for PEG(2OH, 1OH-1CH3 and 2CH3)/PLLA blends.
Table 6lists the values ofsuseproducts in PEG(2OH, 1OH-1CH3and 2CH3)/PLLA blends.suis intrinsic property, so it would not be changed by different blend composition. From
Thomas – Stavely [40] equation: su¼abDh0f; where a is
a constant. For most polymers, a, 1;a¼ 0:25 as high
melting point polyester. Thus, a¼ 0:25 is used. b is the
layer thickness. Dh0f is the enthalpy of 100% crystalline
polymer. From Ref.[20]and[41], b ¼ 5:17 £ 10210m and
Dh0
f ¼ 1:11 £ 108J/m
3
. Therefore, we can get the lateral surface energy ðsuÞ ¼ 14:35 erg/cm
2
.
From Table 6, the fold surface free energy of PLLA decreases with the increasing PEG content regardless of end
groups. For the same composition, the PEG(2CH3)/PLLA
blend has the least fold surface free energy.
4. Conclusion
The results presented in this paper show that PEG and PLLA were miscible in the melt over the composition range investigated by DSC. As a miscible polymer blend, regardless of end groups, the crystallization of PLLA before the formation of PEG was influenced by two temperatures (Tgand Tm0). However, the crystallization of PEG after the formation of PLLA crystals was more complicate. It was found the formation of PLLA crystals would also hinder the crystallization rate of PEG besides the temperature factors Tgand Tm0:
The effect of end groups of PEG on the miscibility and crystallization behavior of PEG/PLLA blends was also investigated. The miscibility of PEO/PLLA blends in
decreasing order were PEG(2CH3)/PLLA,
PEG(1OH-1CH3)/PLLA and PEG(2OH)/PLLA. The equilibrium
melting point, number of nuclei, and fold surface free energy of PLLA in blend increased with more OH end
Table 5
Parameters used for Lauritzen – Hoffman equation
Items PEG PLLA
DE (cal/mol) 4120 b (A˚ ) 5.17 Dhu(J/cm 3 ) 111 Vu(cm 3 /mol) 24.16 44.65 ru(g/cm 3 ) 1.22 1.27 Table 6
The values ofsuseproducts in PEG/PLLA blends
Polymer blend Blend composition suse(erg2/cm4) se(erg/cm2)
PEG(2OH)/PLLA 0/100 7830 546 10/90 5400 376 30/70 3860 269 50/50 2820 197 70/30 2230 155 PEG(1OH)/PLLA 10/90 4950 345 30/70 3620 252 50/50 1940 135 70/30 1090 76 PEG(2CH3)/PLLA 10/90 4600 321 30/70 2580 180 50/50 1390 97 70/30 844 59
groups. Thus, the spherulitic growth rate and isothermal crystallization rate of PLLA in decreasing order were
PEG(2OH)/PLLA, PEG(1OH-1CH3)/PLLA and PEG(2CH3)/
PLLA.
Acknowledgements
The authors acknowledge with gratitude financial sup-port from the National Science Council, Taiwan, ROC, through Grant No. NSC89-2216-E-002-032.
References
[1] Paul DR, Barlow JW, Polymer science and technology, vol. 11. New York: Plenum Press; 1980.
[2] Runt JP, Martynowicz LM. Multicomponent polymer materials. Washington: American Chemical Society; 1986.
[3] Wu WB, Chiu WY, Liau WB. J Appl Polym Sci 1997;64:411. [4] Liu AS, Liau WB, Chiu WY. Macromolecules 1998;31:6539. [5] Liau WB, Liu AS, Chiu WY. Macromol Chem Phys 2002;203:294. [6] Qiu Z, Ikehara T, Nishi T. Macromolecules 2002;35:8251. [7] Qiu Z, Ikehara T, Nishi T. Polymer 2003;44:2503. [8] Qiu Z, Ikehara T, Nishi T. Polymer 2003;44:2799. [9] Qiu Z, Ikehara T, Nishi T. Polymer 2003;44:3101. [10] Qiu Z, Ikehara T, Nishi T. Polymer 2003;44:7519. [11] Qiu Z, Ikehara T, Nishi T. Polymer 2003;44:7749. [12] Younes H, Cohn D. Eur Polym J 1988;24:765. [13] Nakafuku C, Sakoda M. Polym J 1993;25:909. [14] Nakafuku C. Polym J 1996;28:568.
[15] Sheth M, Kumar RAJ. Appl Polym Sci 1997;66:1495. [16] Nijenhuis AJ, Pennings AJ. Polymer 1996;37:5849. [17] Yang JM, Chen HL. Polym J 1997;8:657.
[18] Hoffman JD, Weeks JJ. Res Nat Bur Stand A 1962;66:13. [19] Tsuji H, Ikada Y. Polymer 1995;36:14.
[20] Kalb B, Pennings AJ. Polymer 1980;21:607. [21] Nishi T, Wang TT. Macromolecules 1975;8:909. [22] De Santis P, Kovacs AJ. Biomaterials 1968;6:299.
[23] Roitman DB, Marand H, Miller RL, Hoffman JD. J Phys Chem 1989; 93:6919.
[24] Vasanthakumari R, Pennings AJ. Polymer 1983;24:175.
[25] Rozenberg M, Loewenschuss A, Marcus Y. Spectrochimica Acta Part A 1998;54:1819.
[26] Avrami MJ. J Chem Phys 1939;7:1103. [27] Avrami MJ. J Chem Phys 1940;8:212. [28] Avrami MJ. J Chem Phys 1941;9:177.
[29] Afifi-Effat AM, Hay JN. Eur Polym J 1972;8:289.
[30] Cameron GG, Ingram MD, Qureshi MY, Gearing HM. Eur Polym J 1989;25:779.
[31] Scheirs J, Bigger SW, Delatycki O. Eur Polym J 1991;27:1111. [32] Scheirs J, Bigger SW, Delatycki O. Polymer 1991;32:2014. [33] Han S, Kim C, Kwon D. Polym Degrad Stab 1995;47:203. [34] Han S, Kim C, Kwon D. Polymer 1997;38:317.
[35] Hoffman JD. Soc Plast Engng Trans 1960;4:315.
[36] Hoffman JD, Frolen LJ, Ross GS, Lauritzen JI. J Res Natl Stand US 1975;79A:671.
[37] Hoffman JD. Polymer 1983;24:3. [38] Flory PJ. J Chem Phys 1949;17:223. [39] Mandelkerm L. J Appl Phys 1955;26:443.
[40] Thomas DG, Staveley LAK. J Chem Sci 1952;4569.
[41] Fischer EW, Stergel HJ, Wegner G. Kolloid Zu Polym 1973;251:980. Fig. 6. (a) Growth rates analyzed using Lauritzen – Hoffman equation of
PEG(2OH)/PLLA blends. (b) Growth rates analyzed using Lauritzen – Hoffman equation of PEG(1OH-1CH3)/PLLA blends. (c). Growth rates
