Polyurethane + polyacrylate latex mixtures: synthesis, characterization and properties.
INTRODUCTIONOne of the main directions of modern polymer technology, along with the synthesis of new compounds, is the search for rational ways of using traditional polymers by means of creating new compositions. Polymer blends have been the subject of study for decades. However, much less attention has been paid to latex blends. Such blends are prepared by mixing two polymers where each is present in the form of polymeric microspheres dispersed in a fluid medium. The mixtures of polymer water dispersions are of increasing interest due to their pollution-free status, high performance properties, and the ease of tailoring the materials to meet the specific end user requirements.
Our earlier investigation of polyurethane and polyacrylate latexes mixtures obtained by their careful mixing in a wide range of compositions showed that these systems are very promising both from practical and theoretical points of view (1, 2). Thus, it seemed worthwhile to study polyurethane + polyacrylate aqueous systems obtained not by simple mixing, but by adding polyacrylate dispersion in an acetone solution of anionic polyurethane at the stage of its extending and dispergating. It is known that the thermodynamic stability is one of the main factors for ensuring the reliability of products based on multicomponent polymer systems.
The work described herein deals with the study of colloid-chemical, rheological, thermodynamical behavior and viscoelasticity of polyurethane + polyacrylate anionic latexes, and the correlation of that behavior with mechanical characteristics of these systems.
2. EXPERIMENTAL
2.1 Preparation of the Latexes
A series of polyurethane + polyacrylate latexes with varying contents of the polyacrylate (PA) were prepared using the following initial reagents: poly (oxypropylene) glycol MW 1000 (POPG), toluene diisocyanate (TDI-blend of 2,4- and 2,6- isomers), diethyl-eneglycol (DEG), triethylamine (TEA), hydrazine hydrate (HH), sulfuric acid (SA), acetone, polyacrylate water dispersion PA-copolymer of ethylacrylate/nitryl of acrylic acid/methacrylic acid taken in proportion 85:13:2, respectively. The synthesis procedure of polyurethane + polyacrylate dispersion (with different contents of PA) was as follows: a mixture of TDI, POPG and DEG was placed in a vessel with stirring. The reaction was carried out at 75 [degrees] C under dry nitrogen until the theoretical NCO value was reached. After adding the SA, the product was neutralized by adding TEA at 50 [degrees] C. Acetone (20% of total solids content) was added to decrease the viscosity of the reaction mixture. The reaction mixture was then cooled. The dispersion in PA and chain extension with HH were carried out immediately at 15 [degrees] C in order to avoid the reaction between free NCO groups and water. Individual [TABULAR DATA FOR TABLE 1 OMITTED] polyurethane (PU) latex was obtained by dispergating the above prepolymer in a water-acetone phase in absence of PA. After latex formation, acetone was evaporated at 50 [degrees] C.
2.2 Characterization
The sample films were cast by pouring the PU + PA latexes on glass substrates, dried at room temperature for 72 hours, and then dried at 80 [degrees] C until a constant weight was reached. The mechanical properties were tested using a RM-30-1 test machine (a device constructed in Ivanovo Measure-works, Russia). Thermodynamic characteristics were determined by sorption of solvent vapors at 25 [degrees] C by the individual polymers and their mixtures. A high vacuum device with spring balance (sensitive to within 3-4 mg/mm) made of molybdenum was used. Dynamic mechanical studies were carried out with a dynamic mechanical analyzer (constructed in our Institute) at the frequency of 100 Hz and the heating rate of 1 K/min. The viscosity was measured with a rotational Rheotest-2 viscometer at room temperature. The particle size was determined from the turbidity spectrum. Surface tension was measured by means of platinum ring detachment in a device constructed by Metallist, of St. Petersburg, Russia.
3. RESULTS AND DISCUSSION
The aqueous PU + PA latexes were characterized in terms of their pH value, viscosity, solids content, particle size and surface tension (Table 1). Surface tension of mixed latexes was practically identical for all compositions except 90:10, in which case its value exceeded that of an individual and mixed latexes. It is known that the higher the surface tension, the stronger the intermolecular interaction. The increased strength of the bicomponent film of this composition is in good correlation with the data obtained [ILLUSTRATION FOR FIGURE 1 OMITTED]. The average particle size of mixed latexes almost does not change with composition variations, indicating an aggregative stability of the mixed systems. These systems are more stable than the mechanical mixtures and do not separate into individual phases within five months. Higher aggregative stability of synthesized latexes may be caused by decreased free surface of the latex particles formed.
The flow curves of our latexes are plotted in Fig. 2. The rheological behavior is typical of structural non-Newtonian liquids exhibiting thixotropic flow. The viscosity of PU + PA latexes depends on the shear rate only at the initial stage when [Gamma] = 4 [s.sup.-1] - 20 [s.sup.-1]. The character of curves is not affected by the variation of the PA content. Thixotropic flow is manifested by a high viscosity at low shear rates. However, as the shear rates increase, the viscosity drops correspondingly until it reaches a low limiting value. When the shearing force is then removed, the viscosity recovers, but not instantaneously. We have plotted the so-called apparent viscosity at [Gamma] = 6 [s.sup.-1] for mixtures with various PU/PA ratios [ILLUSTRATION FOR FIGURE 3 OMITTED]. The viscosity values are higher than those obtained assuming linear additivity up to 20% content of PA. In the same composition region the tensile strength and the elongation at break for the PU + PA latex films are considerably higher than the linearly additive values and exceed the parameters for initial components [ILLUSTRATION FOR FIGURE 1 OMITTED], indicating possible miscibility of the PU and PA components.
Thermodynamic investigations have been performed also. Isotherms of sorption for the PU ionomer, PA copolymer and their mixtures are similar to those for the polymers in the elastic state and are not affected by variations of the PA contents [ILLUSTRATION FOR FIGURE 4 OMITTED]. The isotherm of sorption of ethanol by the polyurethane ionomer is located above the curve for the acrylic copolymer. The isotherms of sorption for our latexes are located between the curves of the initial polymers, but at a high relative vapor pressure (above 0.8) the inversion of the curves takes place. Isotherms of sorption for mixtures with 5% and 10% content of PA are below the curve for the initial PA. This is probably due to higher packing density in the latexes and also may indicate their thermodynamic compatibility.
Application of thermodynamic methods in the study of sorption of solvent vapor allows to evaluate a number of the system characteristics in a quantitative way. Variations of the chemical potential (the partial Gibbs function of ethanol ([Delta][[Mu].sub.1]) have been determined from experimental isotherms of sorption, namely
[Delta][[Mu].sub.1] = [M.sup.-1]RT ln(P/[P.sub.0]) (1)
where: M is the molecular mass of the solvent; P/[P.sub.0] is the relative solution vapor pressure with respect to the pure solvent.
Variations of the chemical potentials of the constituents with composition during sorption have been calculated from the Gibbs-Duhem equation:
[W.sub.1][Delta][Delta][[Mu].sub.1]/[Delta][W.sub.1] + [W.sub.2][Delta][Delta][[Mu].sub.2]/[Delta][W.sub.1] = 0 (2)
where [W.sub.1] and [W.sub.2] are the mass fractions of the solvent and the polymer (the latter pertains to either the original polymers or to the sum for their mixtures).
The Gibbs function of mixing has been calculated as
[G.sup.M] = [W.sub.1][Delta][[Mu].sub.1] + [W.sub.2][Delta][[Mu].sub.2] (3)
The results are shown graphically in Fig. 5. It can be seen that all studied systems (PU + ethanol, PA + ethanol, PU + PA + ethanol) are thermodynamically stable ([[Delta].sup.2][G.sup.M]/[Delta][[W.sub.2].sup.2] [less than] 0) (3). The affinity of ethanol to the polyurethane ionomer is considerably higher (Fig. 5, curve 1) than to the pure acrylate copolymer (Fig. 5, curve 8).
Based on concentration dependencies GM of ethanol with individual PU, PA and their mixtures using thermodynamic cycles (4), the values of Gibbs function of mixing of polymers (per 1 g of mixture) were calculated. [G.sup.M] values for the polyurethane ionomer and the acrylate copolymer are negative in the region of concentrations up to 20% of PA [ILLUSTRATION FOR FIGURE 6 OMITTED], demonstrating the thermodynamic compatibility. At a further increase of the PA concentration [greater than] 20%, the sign of [G.sup.M] changes and we observe small positive values. This region is characterized by a two-phase state of the mixtures.
Because of the high viscosity of the latexes, macroseparation of the components does not occur, and the two-phase state may coexist for a long time. We note the arguments of Brostow and co-workers that non-equilibrium phases with high longevity deserve to be included in phase diagrams (5, 6). The true equilibrium state in our latexes exists at concentrations up to 20% content of PA - as characterized by negative value of the Gibbs function.
According to the concept of "intersupplementing differences" (7), polymers with different kinds of groups are able to create hydrogen bonds; EDA- and [Pi]-complexes exhibit good miscibility. This concept is used for selection of thermodynamically compatible polymers. In our blends the formation of hydrogen bonds between the urethane groups of PU and ether grouping of PA is probable. Strong interaction between the different kinds of macromolecules usually results in negative values of the enthalpy of mixing [H.sup.M]. However, the entropy of mixing polymers [S.sup.M] does not play a smaller role (8), especially its noncombinatorial part, arising as a result of disintegration of initial structures of individual polymers and of the formation of a new structure of the blend.
Thus, compatibility of polymers depends on the competition of these two effects according to the standard formula
[G.sup.M] = [H.sup.M] - [TS.sup.M] (4)
Up to a certain polymer concentration the enthalpy of mixing prevails and the blends are compatible. An increase of concentration of the other component (the acrylic copolymer in our case) leads to an increase of the value of the entropy of mixing. Then both the enthalpy and the entropy of mixing are negative, but if [absolute value of [H.sup.M]] [less than] [absolute value of [TS.sup.M]], then [G.sup.M] [greater than] 0, and blends become incompatible.
Miscible polymer blends are usually characterized by improved mechanical properties as compared to the pure components (9, 10). The region of compatibility of PU and PA is characterized by a maximal strength of the mixtures; see again Fig. 1. Tensile strength and elongation at break at PA contents up to 20% are higher than the additive values and also exceed the parameters for pure initial polymers.
The viscoelastic characteristics (the temperature dependence of the storage modulus [E.sup.I], the loss modulus [E.sup.II], and the mechanical loss tangent, tan [Delta]) of the initial polyurethane and polyacrylate are plotted in Fig. 7a. The behavior of the polyacrylate is typical for amorphous linear polymers. Clear maxima of EII at 12 [degrees] C and of tan [Delta] at 31 [degrees] C in the glass transition region are observed. As for the polyurethane, a clearly visible maximum of [E.sup.II] at 0 [degrees] C is close to the glass transition temperature for typical linear polyurethanes based on POPG-1000. However, the temperature dependence of tan [Delta] is not typical. After an increase in the glass transition region (-15 to 20 [degrees] C), instead of its expected decrease, an insignificant rise is observed. The reason for this behavior of tan [Delta] may be a forced mixing of hard and soft segments of polyurethane, although two peaks of loss corresponding to glass transition of hard and soft segments are usually observed (11). Viscoelastic behavior of the system is only slightly affected by adding 1% of PA [ILLUSTRATION FOR FIGURE 7B OMITTED]. At 5% PA content, [T.sub.g] shifts to a lower value of -10 [degrees] C. The well-defined shoulder in the temperature dependence of tan [Delta] curve occurs with the increase of the PA content in the blend up to 10%. At 20% of PA, this shoulder becomes more distinct at -20 [degrees] C, probably because of a more complete separation of soft segment phase of the PU. The second maximum on the curve may be attributed to PA regions that become separated out at that composition. Apparently, the addition of relatively small amount of PA (up to 20%) initiates the process of microphase separation in PU; this in turn causes a relaxation transition of the soft segments. The increase of the PA concentration up to 30-50% again creates a single glass transition peak corresponding to the PA component. It is likely that, beginning from 20% of PA in the latex, PA forms its individual phase. This is reflected in the thermodynamic data: the Gibbs function of mixing acquires a positive value and a two-phase system is formed. Probably the phase inversion takes place in this region. The monotonous increase of the loss peak intensity at 50% PA content indicates an increase of the amount of the phase inversion.
CONCLUSIONS
Polyurethane + polyacrylate latexes with different component ratios based on anionomer polyurethane and polyacrylate water dispersions have been prepared and characterized by colloid-chemical, rheological, thermodynamic, and viscoelastic properties. We have found that the latexes possess the same level of dispersity as the individual polymers. The latexes are non-Newtonian liquids with thixotropic properties. The boundaries of thermodynamic compatibility of polyurethane and polyacrylate components have been defined; up to a 20% of PA the Gibbs function of mixing is negative, indicating component miscibility in this region. The broad glass transition region of pure PU indicates a strong interaction between soft and hard segments, causing a partial phase mixing of various segments. Low-level additions of PA initiate a process of microphase separation between the hard and soft segments of PU, resulting in the improved mechanical properties.
REFERENCES
1. A. P. Grekov, S. A. Sukhorukova, and T. V. Travinska, Ukrain. Khim. Zh., 56, 96 (1990).
2. S. A. Sukhorukova, T. V. Travinska, and A. P. Grekov, Ukrain. Khim. Zh., 55, 89 (1989).
3. A. Tager, Vysokomol. Soed., A. 14, 2690 (1972).
4. A. Tager, T. Sholokhovich, and M. Tsilypotkina, Vysokomol. Soed., A, 6, 1423 (1972).
5 W. Brostow, M. Hess, and B. L. Lopez, Macromolecules, 27, 2262 (1994).
6 M. Hess and B. L. Lopez, in Polymer Liquid Crystals. Mechanical and Thermophysical Properties, W. Brostow, ed., Thomson Science, London (1991).
7. O. Olabisi, L. Robertson, and M. T. Shaw. Polymer-Polymer Miscibility, Academic Press, New York (1979).
8 A. Tager, Kompozyts. polym. mater., 33, 3 (1987).
9 V. M. Kuleznev, Polymer Blends, Moscow (1980) (in Russian).
10. D. R. Paul and S. Newman eds., Polymer Blends, vols. 1 and 2, Academic Press, New York (1978).
11. J. A. Manson and L. H. Sperling, Polymer Blends and Composites. Plenum, New York (1976).
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Title Annotation: | 5th International Conference on Polymer Characterization |
|---|---|
| Author: | Travinska, Tamara V.; Lipatov, Yuri S.; Maslak, Yuri V.; Rosovitsky, Valeri F. |
| Publication: | Polymer Engineering and Science |
| Date: | Mar 1, 1999 |
| Words: | 2526 |
| Previous Article: | Laser light-scattering studies of soluble high performance fluorine-containing polyimides. |
| Next Article: | Thermal stability of poly(methyl methacrylate-co-butyl acrylate) and poly(styrene-co-butyl acrylate) polymers. |
| Topics: | |