Blend Interface Characterization

1 General Aspects Concerning Polymer/Polymer Interfaces 

Helfand et al. [80-82] have developed quantitative lattice theories of the interface. From the mean-field theory, a relationship of the interfacial thickness Al o and interfacial tension coefficient CToo with the x parameter for strongly immiscible polymers, in the limit of infinite molecular weight, is derived. 

The interfacial tension at the interface between two polymers is an expression of different energetics of bulk materials. It reflects differences in thermodynamics, which are related to the x parameter, as shown by Eq. (3.4). The experimental evaluation of the interfacial tension with polymeric melts is extremely difficult due to problems associated with sample preparation and equilibration [77, 78]. Several techniques have been proposed for the measurement. The most commonly used techniques include the pendant drop method, the embedded fiber retraction technique, and the breaking thread method. Classical equilibrium interfacial tension experiments like the pendant drop technique are very difficult to apply to high polymers because of their high melt viscosities (10 -10 Pas). There are many practical problems associated with the pendant drop technique  

These problems are partially overcome by dynamic techniques such as the spinning drop method [77, 78]. 

Technique Depth resolution (nm) Sample Contrast by Lateral resolution Comments 

---

Polymer melt blends maybe miscible or immiscible. Miscible blends form solutions and there is no phase morphology to be of concern. Immiscible blends are characterized by two or more phases that are separated by interfaces. Most polymer blend systems are immiscible because of the low entropies of mixing associated with mixing chain-like molecules to produce homogeneous solution. 

This chapter covers various aspects of reactive polymer blending and compati-bilization starting with a comparison between physical and reactive blending. Issues related to chemical reactions encountered in reactive blending are considered including reaction kinetics. The inter-relation between the reaction events and morphology development is discussed. The last part of this chapter deals with the characterization of the blend interface including measurement of interfacial tension and the interface thickness. 

The last part of this chapter briefly deals with the general aspects of polymer blend interface and its characterization. 

Characterization and control of interfaces in the incompatible polymer blends were reported by Fayt et al. [23]. They used techniques such as electron microscopy, thermal transition analysis, and nonradiative energy transfer (NRET), etc. They have illustrated the exciting potentialities offered by diblock copolymers in high-performance polymer blends. 

Since they act as surfactants, copolymers are added in only small amounts, typically from a thousandth parts to a few hundredth parts. Theoretically, Leibler [30] showed that only 2% of a diblock copolymer may thermodynamically stabilize an 80%/20% incompatible blend with an optimum morphology (submicronic droplets). However, in practice kinetic control and micelle formation interfere in this best-case scenario. To a some extent, compatibilization increases with copolymer concentration [8,31,32], Beyond a critical concentration (critical micellar concentration cmc) little or no improvement is observed (moreover, for high amounts, the copolymer can act as a plasticizer). Copolymer molecular weight influence is similar to that of the concentration effect. For example, in a PS/PDMS system [8,31,32], when the copolymer molecular weight increases, domain size decreases to a certain extent. Hu et al. [31] correlated their experimental results with theoretical prediction of the Leibler s brush theory [30]. Leibler distinguishes two regimes to characterize the behaviour of the copolymer at the interface... 

A different behavior is observed [76] for bilayers composed of partially miscible polymers below their critical temperature Tc. In this case two pure blend components interdiffuse until the equilibrium of two coexisting phases is established. The above equilibrium state is characterized by the coexistence compositions ( q and (]>2 and the interfacial width w. The relaxation of the initial interface between pure constituents involves two processes (see Fig. 3) ... 

Here z(( )00) is the distance from the surface (at depth z=0) to the plateau in composition. The surface enriched/depleted in blend component A is characterized by positive/negative z (see Fig. 15). Relatively large correlation lengths for polymer mixtures (see Sects. 2.1 and 2.2.2) lead to the surface profiles ( )(z) of sufficient spatial extent that may be easily traced by current depth profiling techniques [29]. Surface enrichment has been observed at a free surface [164,165] and at a substrate [92] as well as at an interface between binary blend and a homopolymer [166]. 

The form of Eq. (54) allows us to have better insight into the problem it reflects scaling properties of a mixture between two interfaces. The behavior of such a blend is best characterized by a set of scaling parameters defined by... 

As mentioned in Sect. 2.2.2, the effective interfacial width wD characterizing the bilayer structure may be broadened beyond its intrinsic value w, yielded by a mean field theory (Eqs. 10 and 12). This is due to the capillary wave excitations causing the lateral fluctuation of the depth Ie(x,y) corresponding to the midpoint of the internal interface between coexisting phases. This fluctuation is opposed by the forces due to external interfaces, which try to stabilize the position Ie(x,y) in the center of the bilayer [6, 224, 225]. It was suggested recently [121] that the spectrum of capillary waves for a soft mode phase should be cut off by qb and y. This leads to the conclusion that the effective interfacial width wD should depend on the film thickness D as (wD/2)2= b2+ bD/4. Experimental data [121] obtained for olefinic blends (at T close to Tc) indeed show remarkable increase of the measured interfacial width from wd(D=160 nm)=14.4(3) nm to wd=45(12) nm for thickness D-660 nm, where wD levels off (because is comparable with lateral sample dimensions). This trend is in qualitative agreement with the formula due to capillary oscillations in the soft mode phase . However... 

Characterization of the interfacial regions is important to understand the mechanical properties of incompatible polymer blends. As shown, in many heterogeneous blends, the simplifying assumption of the neglect of spin diffusion between domains is reconcilable with NMR observations. In other words, most of the NMR observables are not sensitive enough to appreciate the influences of the other domains. However, it is also true that the spins are interacting with each other via the interface. To study such interactions. 

What is usually defined as multiscale modeling at this time is far less ambitious. It can best be characterized as a serial approach. It involves the physically robust use of parameters obtained as output in one scale of simulation, as input parameters in simulations at a more coarse-grained scale. There has obviously already been sufficient progress to enable one to combine the same types of modeling techniques quite effectively to predict the morphologies and properties of many types of mixtures, solutions, dispersions, blends, block copolymers and composites as well as to characterize the interfaces between different phases in such systems. 

Intense commercial and academic interest in block copolymers developed during the 1960s and continues today. These materials attract the attention of industry because of their potential for application as thermoplastic elastomers, tough plastics, compatibilizing agents for polymer blends, agents for surface and interface mo dification, polymer micelles, etc. Academic interest arises, primarily, from the use of these materials as model copolymer systems where effects of thermodynamic incompatibility of the two (or more) components on properties in bulk and solution can be probed. The synthesis, characterization, and properties of classical linear block copolymers (AB diblocks, ABA triblocks, and segmented (AB)n systems) have been well documented in a number of books and reviews [1-7] and will not be discussed herein except for the sake of comparison. 

---