Solvent-Free Polymerizations and Processes - American Chemical

Chapter 16

Solvent-Free Functionalization of Polypropylene by Reactive Extrusion with Acidic Peroxides

Downloaded by UNIV OF GUELPH LIBRARY on September 6, 2012 | http://pubs.acs.org Publication Date: January 28, 1999 | doi: 10.1021/bk-1998-0713.ch016

S. C. Manning and R. B. Moore Department of Polymer Science, University of Southern Mississippi, P.O. Box 10076, Hattiesburg, M S 39406-0076

Variable quantities of functionalized peroxides bearing carboxylic acid groups were reacted with polypropylene (PP) in a twin screw extruder. Systematic variations in the molecular structure of the peroxides were found to significantly affect the efficiency of grafting the carboxylic acid groups onto PP and the polymer degradation process. This behavior was attributed to the relative reactivities of the different free radicals generated by thermal decomposition of the peroxides. Furthermore, the functionalized polypropylene (ƒ-PP) was investigated as a compatibilizing additive for 80/20 PP/PA-6,6 (polyamide 6,6) blends. With incorporation of the ƒ-PP into the blends, differential scanning calorimetry (DSC) showed an 80°C decrease in the PA-6,6 crystallization temperature. A near linear increase in the impact strength of the blends was observed with ƒ-PP incorporations up to 30% of the PP phase. Blends containing 30% ƒPP demonstrated impact properties approaching that of pure PA-6,6.

As a result of recent concerns over the environmental hazards of volatile organic compounds (VOCs), many solvent-free polymerizations and processes have been developed. For example, reactive extrusion(7,2) has emerged as an ideal low V O C technique that involves melt-phase chemical reactions performed within the configuration of an extruder. The ability to alter the down stream configuration of the extruder allows for the combination of several chemical procedures into one continuous process. Furthermore, as compared to typical solvent born batch reactors, most reactive extrusion procedures are based on a twin screw geometry(i) which creates a melt mixed environment with excellent heat and mass transfer properties. Typical reactive extrusion processes may be controlled through tailored zone architecture (i.e., variable screw design, temperature profiles, etc.) and through the rate and position at which the reactants are introduced into the extruder. Figure 1 is a schematic diagram (side view) of a twin-screw extruder illustrating variability in

248

© 1 9 9 8 American Chemical Society In Solvent-Free Polymerizations and Processes; Long, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.


249

Figure 1. Side-view schematic of a twin-screw extruder.

the screw design. Intense mixing zones may be incorporated into the screw design through the addition of kneading blocks. Furthermore, melt seals may be created through a reverse flight geometry; these seals are often used to isolate individual reaction zones along the extruder path. Strategic placement of through-barrel ports allows for injection of reactants for sequential reaction initiation as well as devolatilization of residual monomers, condensates, and/or other by-products. Within the field of reactive extrusion, organic peroxides offer the advantage of a thermally induced decomposition by homolytic cleavage of the labile 0 - 0 bonds. This cleavage forms two free radical species that may be able to abstract hydrogen atoms from polymer chains and/or add to double bonds. Furthermore, organic peroxides are generally required in low concentrations and can decompose to complete the chemical reaction during typical extrusion residence times. A s a result of these desirable characteristics, a wide variety of organic peroxides have been commercialized for reactive extrusion applications such as: bulk polymerization,^,5) cross-linking of polyethylene and elastomers,(6) production of controlled rheology polypropylene,(7-9) and grafting of functional monomers onto polymer chains.(70-74) Organic peroxides are ideally suited for applications involving modifications of polymer properties via melt-phase chemical reactions. T o optimize the chemical compatibility and/or solubility of the peroxides in a wide variety of polymeric systems, the organic character of these peroxides may be tailored by subtle changes in the molecular structure. Moreover, since applied extrusion conditions (e.g., melt temperatures and zone residence times) vary with polymer type, the numerous structures available with organic peroxides yield a wide range of decomposition kinetics and resulting free radical reactivities. Functionalization of polyolefins is often achieved by grafting a polar monomer such as maleic anhydride ( Μ Α Η ) onto the polyolefin backbone using reactive peroxides.(72-77) Compared to other initiation techniques (e.g., γ-irradiation), the use of organic peroxides initiators for Μ Α Η grafting onto P P generally results in superior Μ Α Η grafting efficiencies.(7#) This functionalization reaction can be performed in solution(79) or in the melt,(77) and may be an economic way to make these inherently

In Solvent-Free Polymerizations and Processes; Long, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

250

nonpolar polymers more compatible with polar polymers. For example, enhanced minor phase dispersion and interfacial adhesion has been observed with blends of MAH-functionalized polypropylene with nylon 6(20) and when MAH-functionalized polypropylene was used as a compatibilizing agent in blends of polypropylene with nylon 6,6.(27) During a typical melt graft reaction of Μ Α Η onto polypropylene, the reactive radicals (i.e., the decomposition products from the peroxide initiators) abstract hydrogens from the polypropylene backbone to form tertiary radicals along the chains. TTiese polymeric radicals then add to the reactive double bond of Μ Α Η to form randomly distributed grafts.(72) While the initial product of this free radical addition reaction is another radical, recent C N M R studies of Μ Α Η grafted PP(22) have shown that the graft sites of the final product contain predominately single succinic anhydride rings. The possibility for continued Μ Α Η homopolymerization into block side chains has been suggested to be unlikely based on ceiling temperature considerations^.?) (i.e., at typical melt temperatures of ca. 180° C); however, a portion of the grafted Μ Α Η groups may form cross-links between polypropylene chains.(24) In addition to the desirable grafting reactions, polypropylene has also been shown to degrade during peroxide-initiated functionalization by tertiary radical βscission (i.e., chain scission at the site of the polymeric radicals). A t low Μ Α Η concentrations, the secondary anhydride radicals have been found to contribute to the formation of tertiary polymeric radicals via intramolecular chain transfer;(74) this rearrangement consequently promotes polypropylene degradation.(72,74) Recently, E l f Atochem N . A . has developed a series of asymmetric, functional peroxides (peroxyesters) bearing carboxylic groups which may be used to graft acidic functionality directly onto polyolefin chains.(25) In addition, DeNicola has demonstrated that unsaturated peroxyacids may be used to functionalize polypropylene in order to increase the interfacial adhesion to glass fiber reinforcement.(26) The original hypothesis in the use of these new peroxides in reactive processing is that highly reactive alkyl radicals from the thermal decomposition of the asymmetric peroxides abstract hydrogens from polypropylene. Subsequently, less reactive radicals (containing the acid functionality) couple with the polymeric radicals to form the grafts. Since these grafts are created during a radical termination step rather than in a chain mechanism, undesirable cross-linking and/or β-scission may be minimized. In this chapter, we compare the grafting efficiency of a series of carboxylic acid containing peroxides in a reactive extrusion process with polypropylene (PP). The focus of this study is on the subtle changes in the chemical structures of these peroxides that differ by only one methylene unit and/or one site of unsaturation. The grafting mechanism is also correlated to other possible free radical mechanisms such as βscission and cross-linking. Furthermore, the resulting / - P P w i l l be investigated as a compatibilizing agent in PP/PA-6,6 blends.


1 3

E x p e r i m e n t a l Section Materials. Polypropylene powder (Pro-Fax 6501) was obtained from Himont Inc. Polyamide 6,6 (Vydine 21) was obtained from Monsanto. The functional peroxides Luperox P M A , Luperox T A - P M A , Luperco 212-P75 and Lupersol 512 were obtained from E l f Atochem North America. Luperco 212-P75 was obtained and used as a


251


powder of the peroxide on a polypropylene carrier; Lupersol 512 was obtained as a 50 wt% solution in ethyl 3-ethoxypropionate and evaporated onto a P P powder carrier. The structures of these peroxides are represented in Figure 2. Reactive Extrusion. Polypropylene and the peroxide/PP powders were premixed to obtain samples of 0.5, 1.0 and 2.0 wt% peroxide. The mixture was then extruded in a Haake Rheochord 90 counter-rotating twin screw extruder with 3 mixing zones and a rheological slit die. The temperatures of the zones were Ti=150°C, T2=T3=180°C, T4 (die)=170°C. The screw speed was set at 15 R P M corresponding to a residence time of ca. 7 minutes, and all samples were extruded under a N2 atmosphere. Blends of P P and P A - 6 , 6 were premixed to obtain an overall 80/20 (w/w) composition of P P / P A - 6 , 6 . Polypropylene functionalized with 0.5% Luperox P M A was incorporated into the P P portion of the blends at 5, 10, 20 and 30 wt%. The premixed blends were extruded with a temperature profile of Τι = T2 = T3 = T4 = 270°C. The screw speed was set at 15 rpm, and all blends were extruded under a N2 atmosphere. Physical testing samples were obtained by injection molding the blend regrind at 270°C using a B o y 15S injection molder. Titrimetric Assay of the Functionalized Polypropylene. T o eliminate the potential contribution of unreacted peroxide, the extruded products were first purified by reprecipitation. The extrudate was dissolved in xylene at 130°C to a concentration of 5% (w/v) and reprecipitated into methanol. After filtration, the precipitate was washed with pure methanol and dried at 60 °C in vacuo. Samples of the purified polypropylene (0.5 g) were dissolved in 100 m l of xylene and titrated to the phenolphtalein endpoint with a standardized solution of benzyltrimethylammonium hydroxide in M e O H . The titrant was standardized with benzoic acid in xylene. A l l titrations were performed in triplicate at 110°C under a N2 atmosphere. Due to potential oxidation of P P during melt extrusion, the residual acidity of pure P P (extruded under identical conditions) was assayed as a blank and subtracted from the titrimetric results of the functionalized P P samples. The acid contents and sample standard deviations were calculated in units of equivalents per gram of polymer. The grafting efficiency for the peroxides were then calculated as:

^

^ . . ηΓ

„

Grafting Efficiency % =

equiv.f-PP - equiv. PP * .

„

.,

. ^

^xlOO

< λ λ

equiv. Peroxide in Feed

(1)

where equiv. f-PP is the total acidity of the functionalized P P , equiv. PP is the blank acidity, and equiv. Peroxide in Feed is the initial concentration of the peroxide/PP mixture. Characterization of Material Properties. The purified polypropylene samples were densified by melting at 190°C in vacuo and then reground into a coarse powder. The Melt Index of the products was measured with a Custom Scientific Instruments C S I Melt Indexer at 230°C and under a 2.16 k g load as per A S T M D-1238. The molecular weights of the P P samples were measured with high-temperature size exclusion chromatography (SEC) using a differential refractive index detector. The samples


252

,0 CH 3

:—CH=CH—c

CH —ί—00—

Solvent-Free Polymerizations and Processes - American Chemical

Recommend Documents