Prof. Moshe Gottlieb
Polymers in the vicinity of surfaces and interfaces, physical gelation as result of microdomain confinement.

The research activity of Prof. Moshe Gottlieb's group is related to two main topics:
(i) Polymers in the vicinity of surfaces and interfaces. This includes the study of polymer filler interactions, effect of surface chemistry on polymer adsorption, polymers in the vicinity of interfaces in relation to polymeric surfactants and the effect of surfaces on polymer crystallization.
(ii) Gelation phenomena and especially physical gelation as result of microdomain confinement.

1. Polymer-Solid interactions (Composite Materials)
Surface Adhesion. In this area we have examined the interaction between a polymer matrix and different solid additives used in polymer composite materials. Although the study focused on polypropylene based matrices the study is aimed at obtaining a more general understanding. Composite materials demand constant improvement in mechanical properties due to more rigorous application requirements. At the same time health, safety and environmental requirements impose incorporation of additives which usually affect negatively the mechanical properties of the material. Fibers are commonly used in polymeric composites as a reinforcing agent. The fibers are introduced to bear loads above the intrinsic mechanical strength of the polymer matrix. A well-established interface between the fiber and the polymer matrix will disperse the external load to the rigid fiber and hence enhance the overall composite mechanical strength. The results are manifested in material stiffness, toughness and even in the failure mode of the composite.
In general, the adhesion between polymeric materials and fibers is weak due to poor compatibility, wettability and bonding. Coupling agents are commonly used in order to overcome the large differences between the two components and improve their compatibility by forming better interfacial forces ranging from strong chemical bonds or electric attraction to weak Van der Waals (VDW) interactions. The most widely used commercial coupling agents are silane compounds. In this work, we developed a methodology to quantify the strength of adhesion between the polymer, (Polypropylene (PP) or maleic anhydride grafted onto PP) and glass slabs treated by different silane coupling agents, as a model system for glass fibers in commercial composites.
The thickness of PP and PP-g-MA films were measured using optical phase interference microscopy (OPIM). A scratch was made to form a clear boundary between the polymer layer and the glass substrate, which allowed measuring the layer thickness. Fig 1 depicts an OPIM image of a PP film deposited on (a) untreated glass and (b) trimethoxypropylesilane (PS) treated glass.
The PP films were found to be of similar thickness, approximately 18-20nm irrespective of the type of the surface treatment. Scaling theory predicts that the thickness of a polymer layer formed from melt and rinsed in good solvent is L~aN0.83, here L~185nm. The predicted thickness is 10 times larger than the measured value. We attribute this discrepancy to the preparation method which we speculate, leaves on the surface only highly absorbed chains. The thickness of PP-g-MA films varied between ~10nm and 50nm depending on type of surface treatments. A more fundamental understanding of the behavior of polymer chains in these films will hopefully be obtained from the theoretical work of Dr. Aleksey Drozdov (see below).

 

Fig 1: A. OPIM image of: PP on untreated glass

B. OPIM images of: PP on PS treated glass

AFM was used to quantify the strength of adhesion between the polymer and the glass surface. This was achieved by gradually increasing the normal force applied by the AFM in 'contact mode' until reaching the level necessary to remove completely the polymer layer. The AFM used in non-contact 'tapping mode' was employed to determine the morphology of the structure prior and subsequent to the polymer removal. The AFM tapping-mode images shown in Fig. 8 depict the polymer film before and after film removal by the contact-AFM. The measured forces for PP are all of the same magnitude as those expected from VDW interactions and correlate well with the nature of surface treatment: the more hydrophilic the surface the weaker the interactions with PP. An opposite trend is observed in the case of the PP-g-MA which suggests that the hydrophilic MA side chains are responsible for most of the surface-polymer interaction.

Fig. 2 – Tapping mode AFM images of:
A) polymer film prior to film removal

B) After polymer film removal

Fire Retardants. Thermal polymerization of pentabromobenzylacrylate (PBBMA) in a polypropylene (PP) composite that contains glass fibers and magnesium hydroxide has been studied using scanning and transmission electron microscopy techniques coupled with energy dispersive spectrometry. The addition of PBBMA imparts flame retardant (FR) properties to the PP composite but also affects adversely its mechanical properties. It is of practical importance to determine the spatial distribution and the extent of polymerization of the FR in the PP composite in order to understand better its role in the system. The methods we have developed allow the distinction between the monomeric and polymeric forms of the FR and to determine their spatial distributions. In a typical PP composite studied here we observe that the residual monomer is not dissolved in the composite, whereas the polymeric FR is homogeneously dispersed in PP matrix. By the combination of the electron-microscopy methods and FTIR microscopy it was determined that the polymerization reaction is highly hindered by the presence of the antioxidant and takes place only in the presence of antimony oxide. PP itself shows poor adhesion to the glass fibers (Fig. 3), which may be improved by the addition of the reactive PBBMA. The latter is polymerized during reactive extrusion through an antimony-catalyzed reaction. PP shows good adhesion to sized Mg(OH)2 as expected from a properly surface-treated filler.

Fig 3A. Scanning electron micrograph. In this image PBBPA spots, PBBMA particle, and a PP fibril are observed. PBBPA is detected at the base of the fibril near the glass fiber.

Fig 3B. Scanning electron micrograph. No PP adhesion to the glass fiber is observed. Small spots are identified as PBBPA.

Composite Optimization. Based on the insight obtained in the studies described above we set out to design an optimal composition for a composite matrix containing glass fibers, PBBMA as a primary FR and magnesium hydroxide as a secondary FR. Optimal composition is reached by means of statistical design of experiments (DOE) rather than by ‘‘trial and error’’ approach. The DOE approach allows minimization of the number of experiments, investigation of the influence of each additive and the mutual interactions between additives. It also allows prediction of optimal sample properties with improved mechanical and FR properties. Both FRs reduce the impact strength while enhancing flame retardancy. Glass fibers increase the modulus, but have only a moderate effect on the impact strength due to poor adhesion with PP. The effect of glass fibers on the flammability of the material has still to be understood.
Thermal Transitions. The effect of surfaces on thermal transitions in polymeric systems has been studied experimentally and theoretically. The enhancement of crystallization was attributed to increased ordering of chains. The work has been described in detail (see http://echem.bgu.ac.il/staff/Rachel/index.htm). The study of enhanced order in polymers in the vicinity of surfaces was complemented by the study of orientation of fibers in the vicinity of solid walls (Mor et al. J. Rheol. 2003).

2. Polymers at Interfaces
In this work we examined the properties of ABA amphiphilic triblock copolymers at the air-liquid and liquid-liquid interfaces. The interfacial properties of poly(ethyleneoxide) – poly(dimethylsiloxane) – poly(ethyleneoxide) triblock copolymer (PEO-b-PDMS-b-PEO) synthesized in our labs were investigated. Three copolymer compositions were examined as effective surfactants. The molecular weight of the PDMS block was 12K whereas the PEO end blocks were 250, 500 and 1700 g/mol respectively. The copolymer with the intermediate size PEO end blocks (500 g/mol) lowered the surface tension considerably more than those with either the smaller or larger PEO blocks. This result is in contrast with the behavior of PEO molecules for which, the surface tension decreases as the length of the chain increases. The observed behavior can be rationalized by means of packing at the interface (cf. Fig. 4). The triblock copolymer with 500 g/mol PEO block has the highest packing (~60 Å2/molecule) at the interface. Long PEO blocks may generate a higher driving force towards the interface, but it lacks the ability to pack densely, resulting in fewer molecules adsorbed at the interface. From our results it appears that there is an optimal number of EO monomers for maximum surface activity. The optimal PEO length or PEO/PDMS ratio has not yet been determined and, it may also vary for different liquid-liquid interfaces.

Fig 4: Estimated projected surface area at the decane-water interface. T = 25OC

The results obtained in this study show that PEO-PDMS-PEO triblock copolymer is highly efficient in reducing ? of the water-decane interface. The efficiency of these materials compares very favorably with those of other simple and polymeric surfactants (e. g. AOT, Pluronic) and was found to be extremely effective in long-term emulsion stabilization. The surface activity of the triblock copolymer with 11 EO monomers end-group is considerably higher than the one with longer or shorter PEO blocks. This behavior contradicts the behavior of the individual blocks, but can be rationalized by means of packing at the interface. The triblock copolymer with 11 EO monomers has the highest packing (low area per molecule) at the interface. Longer PEO blocks may generate a higher driving force for adsorption to the interface, but they lack the ability to pack densely, resulting in fewer molecules adsorbing to the interface.

3. “Soft” Microdomain Interactions
This study is targeted at gels formed by a spatially-confined gelation process. . This type of gels is of relevance to thin film technology, membrane formation, and surface phenomena. Gelation takes place in a spatially confined environment as result of the presence of microdomains. The microdomains studied here are either 1) micelles formed by short chain surfactants interacting with amphiphilic polymers or 2) microphase separation domains as result of hydrophobic interactions. The imposed confinement affects the dynamic properties of these systems as manifested by the rheological properties of the resulting gels. The work involves the study of two classes of gels: a)amphiphilic block mesogels based on ABA copolymers of the type discussed above b)hydrophobically substituted biopolymers (e.g. cellulose derivatives). Block copolymers are known to form well ordered morphologies being further enhanced by the contradictory solvation in amphiphilic systems. Such systems have potential to serve as nanocomposite templates. Amphiphilic ABA triblock copolymer with the different parts of the polymer residing in different solvents may offer even a greater degree of complexity. As result of the presence of short chain surfactants the solvent which is dispersed in the second continuous-phase solvent may adopt micellar, cylindrical, or lamellar mesophase morphologies. The copolymer is capable, depending on chain and segment size and on system composition, to bridge between adjacent dispersed aggregates. Gelation will occur when a critical bridging is achieved resulting in the formation of a weak physical mesogels. A systematic study of structure, morphology and rheology of the gelation process and the resulting materials is required. Recent theoretical work by Zilman et al. (Zilman A; Kieffer J; Molino F; Porte G; Safran SA. PRL (2003) 91 (1): art. no.-015901) predicted a phase separated region and two rheologically distinct phases depending on the composition of the system. We found this description incomplete and highly dependent on minor changes in the nature of system constituents. A typical example of the results obtained in this ongoing study are presented in Fig. 5.

Fig. 5. Phase diagram of a system composed of isooctane/water/AOT and ABA copolymer. The rheological behavior of a system residing in the phase separated region indicates a gel-like behavior.

The gelation behavior of hydrohpobically modified biopolymers such as methyl cellulose show the peculiar thermal gelation, i.e. the system gels reversibly as result of heating. Gelation is attributed to the formation of microphase separated domains in which the hydrophobic groups assemble. In reversible physical gels of most natural and synthetic polymers the process of gelation occurs upon cooling of the polymer solution. In contrast, aqueous solutions of methylcellulose have the unusual property of forming gels on heating and reverting to the solution state on cooling. Gelation of methylcellulose solution is believed to be caused primarily by the hydrophobic interaction between molecules containing methoxyl substitution. The purpose of this paper is to investigate the role of the chemical structure on the physical properties of methylcellulose gels.
Two different processes for cellulose methylation are available: homogeneous and heterogeneous. The heterogeneous process, which is carried out in aqueous environment, is the one commonly used in industry. The cellulose is not dissolved but rather dispersed in the medium. As result, the methylation is believed to be unevenly distributed along the polymer chain resulting in large hydrophobic and hydrophilic zones along the polymer backbone. Homogeneous methylation is carried out in a solvent capable of molecularly dissolving cellulose bundles and is expected to yield a more uniformly distributed methylation. In this work, methylcellulose was prepared by homogeneous reaction with different degree of substitution (0.9<DS<2.3) and molecular weight. The structural characterization of the samples was carried out using 13C-NMR and HPLC. Microstructures of semi-dilute solution and gel have been tested by SAXS measurement. Rheological studies were performed over the range temperature 25-800C and the properties of the samples were compared with those of commercial samples. Thermodynamic as well as kinetic effects have been examined. Large differences were observed between the commercial (heterogeneous) sample and the samples prepared by homogeneous reaction. To demonstrate the highly complex morphologies involved in this process a phase diagram complemented by cryo-TEM imaging of systems under different conditions is presented in Fig. 6.

Fig. 6: Morphologies observed by cryo-TEM in MC solutions. (In collaboration with Y. Talmon, Technion)