Miscibility Behavior of Polyacrylamides Poly ( Ethylene Glycol ) Blends : Flory Huggins Interaction Parameter Determined by Thermal Analysis

Blends of polyacrylamide—PAM, poly(N-isopropylacrylamide)—PNIPAAm, poly(N-tert-butylacrylamide)—PTBAA, poly(N,N-dimethylacrylamide)—PDMAA and poly(N,N-diethylacrylamide)—PDEAA with poly(ethylene glycol)— PEG were prepared by casting in methanol and water at concentrations of 20 wt%, 40 wt%, 60 wt%, and 80 wt% in PEG. The miscibility of the components was studied by Differential Scanning Calorimetry—DSC. All blend systems are characterized by a single glass transition temperature (Tg), close to the Tg of the amorphous component. The Hoffman Weeks method was used to determine equilibrium melting temperature (Tm) data. The determination of the melt point depression of the blends allowed the calculation of Flory-Huggins interaction parameter (χ12) of the two polymers in the melt, by using the Nishi Wang equation. The interaction parameters, calculated for all the blends, are slightly negative and close to zero, suggesting a partial miscibility between the components.


Introduction
Most of the proposed applications of polyacrylamides rely primarily on their behavior in aqueous solution [1][2][3][4].On the other hand, this polymer and some of its N-alkyl substituted derivatives show great potential in the preparation of miscible polymer blends due to their ability to interact through hydrogen bonding.Polyacrylamide-PAM, and some of its N-alkyl substituted derivatives, such as poly(N-isopropylacrylamide)-PNIPAAm, poly(N-tert-butylacrylamide)-PTBAA, poly(N,N-dimethylacrylamide)-PDMAA, and poly(N,N-diethylacrylamide)-PDEAA, are amorphous and water-soluble polymers with great industrial and agricultural interest [5][6][7].In previous work, Silva et al. [8] showed that their thermal behavior is strongly dependent on the alkyl bonded to the amide group.Polyacrylamides form gels, or polymer networks that are able to expand in compatible solvent, like water, exhibiting a great potential for concentration of macromolecules from solutions, water purification, enzyme immobilization, drug delivery, sensors and for several biomedical uses [1,[9][10][11][12][13].
Poly(ethyleneglycol), PEG, is a semi-crystalline polymer that shows applications as chromatographic support and polymeric electrolyte [14][15][16][17][18].In the last ten years, interest has been attracted to systems in which at least one of the components is crystallizable.In this case, a melt point depression of the crystalline phase relative to its melting point (T m ) in a non-interacting medium provides additional evidence of miscibility, since the kinetics and morphological effects over T m are eliminated [19].Measurement of the melting temperature (T m ) depression for blends allowed the determination of the Flory-Huggins interaction parameter (χ 12 ) of polymers in the melt state, by using the Nishi-Wang equation [20,21].The Hoffman-Weeks method is used to determine the equilibrium data [22].
There have been many studies on miscible blends containing poly(ethylene glycol) and poly(ethylene oxide).For example, blends of PEO/poly(epichlorohydrin) and PEO/poly(epichlorohydrin-co-ethylene oxide) were studied by Silva et al. [19].For these systems, the polymerpolymer interaction parameters are all negatives and ex-hibit dependence on the blend composition, decreasing for blends rich in PEO, suggesting that the polymer pairs are thermodynamically miscible.
The miscibility of PVC/PEO blends was studied using viscosimetry, thermal analysis and microscopy by Neiro et al. [26].χ 12 values obtained from the melting point depression are dependent on the molar mass of PVC, varying from −0.102 to −0.028, indicating that the pair is miscible in the melt.
In this paper, we report the study on the miscibility of blends of PAM, PNIPAAm, PTBAA, PDEAA e PDMAA with PEG, using differential scanning calorimetry (DSC).The Flory-Huggins interaction parameters (χ 12 ) were determined by the melting point depression method.

Materials
PAM, PNIPAAm, PTBAA, PDMAA and PDEAA were synthesized via free radical mechanism, under nitrogren atmosphere, according to the method described by Freitas and Cussler [1].
The weight-average molecular weight (M W ) of amorphous homopolymers was determined by light scattering using a Brookhaven Instruments equipment.

Blends Preparation
Binary blends of varying compositions were prepared by casting from water and methanol solutions.To ensure complete removal of the solvent, the blends were kept under vacuum at 41˚C for ten days.

Blends Characterization and Determination of Crystallization Temperatures by DSC
The DSC curves for homopolymers and blends were obtained on a DSC-50 Shimadzu module.Samples weights were maintained in the range of 10 mg.All experiments were performed under helium flow of 70 mL•min −1 .The samples were heated from ambient to 210˚C at a rate of 20˚C•min −1 , held at this temperature for 10 min to eliminate thermal history.After cooling the samples to −20˚C at a rate of 20˚C•min −1 , they were heated again to 210˚C at 10˚C•min −1 .The glass transition temperature and the melting enthalpies (H m ) values were taken from the second heating scan.The cooling scan was used to select the crystallization temperatures.

Isothermal Crystallization
Isothermal crystallization was performed on a DSC-50 Shimadzu module based on the Hoffman-Weeks method [22].Samples weights of 10 mg were heated from 25˚C to 200˚C at a rate of 20˚C•min −1 , held at this temperature for 5 min, rapidly cooled (cooling rate~ 40˚C•min −1 ) to the desired crystallization temperature (T c ), maintained at this temperature for 2 min, and then cooled to 10˚C at a rate of 20˚C•min −1 .After that, the samples were heated to 200˚C, at a rate of 20˚C•min −1 , for the measurement of the T m .This procedure was repeated for different crystallization temperatures.

Scanning Electron Microscopy (SEM)
SEM micrographs of the blends were obtained using a JEOL JSM-5410 microscope and the samples were fractured in nitrogen and covered by sputtering with a Au/Pd alloy.

Density Measurements
Density measurements of each homopolymer were carried out using a picnometer in a non-solvent (heptane and cyclohexane).

Results and Discussion
The weight-average molecular weight of PAM and its N-alkyl substituted derivatives are, as seen in Table 1, at the same order of magnitude.
Density values of PAM and its N-alkyl substituted derivatives are showed in Table 2.
It's shown in Figure 1 DSC curves of blends with 20   3. A single T g is observed for all blends and it does not change appreciably with the composition, showing values around T g of the amorphous polymer (PAM and its N-alkyl substituted derivatives), suggesting that the amorphous phase of the blend is rich in this component.The blends also show a single endothermic peak due to the melting of crystalline phase (PEG).
Figure 2 shows scanning electron micrographs of PAM, PEG and PAM/PEG blends in various compositions.It is observed that the crystal sizes are dependent on the amorphous polymer concentration.As the amorphous polymer concentration increases, PEG crystalline phase size decreases (Table 4).It is known that the kinetics of crystallization of the crystalline component in a blend is affected by the presence of the amorphous polymer [27,28].Both, the overall kinetic rate and the spherulites growth rate decrease significantly.Changes on crystallization behavior are due to low mobility of PEG related with T g of the blends and with the possibility of favorable interactions between PEG and the amorphous polymer.

Determination of Flory-Huggins Parameter
In blends in which at least one component is crystallizable, th int depressio of the crystalline p ase rel eltin oint p ides ad nal ev ce of mis .The m g p of a po er is affected not only by the thermodynamic factors but also by the hick-, the by the extrapolatio e melt po ative to its m n rov h iden g p ditio cibility eltin oint lym morphological aspects such as crystalline lamellar t ness.As described by the Flory-Huggins theory equilibrium melting point should be used to separate morphological effect from thermodynamic effect in discussing the melt point depression [29].Morphological effects are associated with changes in crystal perfection or geometry and with different thermal history of the samples.The contribution of such morphological effects can usually be removed by constructing a Hoffman-Weeks [22] plot by using melting data for PEG and for blends isothermally crystallized at different temperatures (T c ); T me is determinated n of the experimental curve of T m versus T c to theo--Wang, by Equation (1) [20,21].
retical curve, T m = T c .
Thermodynamic considerations predict that chemical potential of a crystallizable polymer will decrease caused by the addition of the miscible diluent [28].The expression to describe the dependence of the melting point depression due only to thermodynamic effects on the blends composition is given, according to the Flory-Huggins theory modified by Nishi T me e 0 me T are the equilibrium melting temperature of PEG in blends and pure PEG, respectively.Subscripts 1 e 2 are referred to the crystallizable and non-crystallizable polymers, V u is the molar volume of the repeating unit H u is the heat of fusion of the perfectly crystallizable polymer per mole of the repeat unit, m is the degree polymerization;  are the volume fractions, R is the u ve s c 0 T general, T me values showed a slight decreasing tendency relative to me value as the concentration of the amorphous component increases.In Figure 4 and in Table 7, χ 12 values are represented as a function of PEG composition.
For all systems, χ 12 values are dependent on the blend composition.Avella [25] and Yoo [30] attribute the dependence of χ 12 on the blend composition to morphological and kinetics factors such as recrystallization, phase segregation, etc. Painter et al. [31] proposed that this effect results from the strong interactions between the components as hydrogen bonding observed in poly(vinyl rsal ga onstant, and χ 12 is the polymer-polymer interaction parameter.
Figure 3 shows the Hoffman-Weeks plots for isothermally crystallized blends.
Table 5 summarizes the equilibrium melting temperatures of PEG in the blends (T me ).For pure PEG, equilibrium melting point   0 me T was determined as 57˚C.In     The ts show hat b s with er co t of p int depression was observed with increasing content of the amorphous polymer.The equilibrium melting points were estimated using Hoffmann-Weeks extrapolation.The interaction parameter was calculated and it was found to be χ 12 = −1.3.Blends miscibility was attributed to the formation of hydrogen bonds between the hydroxyl groups of PVPh and carbonyl groups of PESeb.
In the present work, χ 12 values are dependent on th end composition and are near zero or slightly negatives (Table 7), suggesting partial miscibility between PEG and the amorphous polymers in the melting, maybe as a result of interactions through hydrogen bonds.Additionally, T g behavior and SEM micrographs corroborate this result.

Conclusion
Binary blends of polyacrylamide and some of its N-alkyl substituted derivatives with PEG, obtained by co-dissolution method, are partially miscible, as demonstrated by the smooth lowering of T g values relative to the T g value of the pure amorphous component.The polymer-polymer interaction parameters obtained from thermodynamic in meltg temperature depression analysis are slightly negative and close to zero, suggesting partial miscibility among the components.Furthermore, for all blends studied, there is an endothermic peak corresponding to the melting of the PEG crystalline phase.For all blends studied, SEM micrographs indicate the presence of PEG crystalline phase.