Low viscous hydrophilic processing additives for extrusion of polyethylene at reduced temperatures.
Advantages of Extrusion of Polyolefin Resins at Reduced Temperatures
Extrusion of thermoplastic polymers is used for pelletizing, fiber spinning, sheet and film manufacturing, pipe, tube and profile forming, as well as for wire and cable insulation (1). Opposite to extrusion of metals (2), the extrusion of polymers goes ordinarily at temperatures above the melting point of the material. High molecular weight (MW) polymers can be used to produce articles with enhanced mechanical properties and chemical stability, but flow of such polymers inside the processing equipment and through the forming die is impeded by very high friction losses because of high viscosity of melts. Viscosity of thermoplastic polymers decreases with heating of the melt, and therefore processing of polymers with high MW is commonly performed at temperatures that are far above the melting point. At such temperatures, molten organic polymers rapidly react with oxygen and chemically decay with the consequence of declining mechanical and optical properties. Additionally, at high temperatures polymer flow is unstable to bubble and helical instabilities in film blowing, and to draw resonance in fiber spinning and film casting (3-9). Production rates in processing by extrusion depend very much on the temperature and the achievable rate of cooling of the extrudate. Reduced temperatures of extrusion resulting in shorter cooling times allow an increase in productivity, especially for film blowing, blow molding, and forming of pipes of large diameter to transport gas and water. Since polymers are subject to thermal degradation, reduced extrusion temperatures would also allow a smaller degree of macromolecular degradation. Crystallization of molten polymer and therefore optical and mechanical properties, as well as gas permeability of PO films depends on the rate of cooling and deformation of the polymer matrix. In many cases, extrusion at reduced temperatures would help to decrease the size of crystallites and to improve transparency of extruded films from PO resins.
PO resins are the biggest and the fastest growing polymer family of commodity polymers. Polyethylene (PE) resins represent about 40% of all polymers produced in the world, whereas the share of Polypropylene (PP) resins is about 22%. The PE resins made from modern catalysts (Ziegler-Natta and metallocen) are cheaper (price difference is about 10%), cleaner, and mechanically stronger in comparison with PE made from older catalyst (chromium oxide) (10), but extrusion rate and processability arc limited by the onset of the sharkskin instability (11), (12). Melt Index (MI) is an indication of viscosity of the polymer. The MI is measured by the grams of polymer that are extruded through a narrow orifice held at 190[degrees]C with a specified weight. If the MI is low. the apparent viscosity is high and vice versa. LDPE is characterized by strong shear thinning, and therefore it can be extruded through a forming die at relatively low pressures in comparison to LLDPE of the same MI. Therefore film-blowing machines designed for extrusion of LDPE and LDPE/LLDPE blends cannot be used for extrusion of LLDPE and LLDPE/ LDPE blends. The extrusion pressure would be reduced if the extrusion die has a wider die gap. Film blowing with such dies requires high drawdown and blowup ratios. Drawdown is defined as the ratio of the die gap to the film gauge. The blowup ratio in film blowing is defined as the ratio of the largest bubble diameter to the diameter of the die. The ultimate drawdown rate of known PE resins is limited by the onset of a melt flow instability known as draw resonance, and for LLDPE the critical drawdown ratio can be as low as 2 (13). Additionally, at high drawdown ratios long molecules are stretched and predominantly oriented along the machine direction so that the produced PE film manifests anisotropic mechanical properties. The anisotropy of the mechanical properties of PE film is considered as undesirable for many applications.
Increasing temperatures of LLDPE melt to reduce its apparent viscosity and to delay the onset of sharkskin melt fracture generally necessitates lower rates of film formation because of bubble instability as well as limitations of air-cooling and heat transfer, e.g. see (14), (15). To overcome this discrepancy, Hutchinson and Blanchard (16) proposed to heat up the tip of the extrusion die so that only the surface layer of the extrudate is heated up and therefore bubble instability is reduced. Quite opposite, Cogswell (17) proposed to cool down a surface layer of the extrudate at the die exit to temperatures not more than 15[degrees]C above its solidification point to delay an onset of the sharkskin melt fracture. In extrusion of polymers with narrow MWD, e.g. LLDPE, HDPE, and fluorinated PO resins, the extrudate starts to slip along the die surface at some critical rate of extrusion. Fields and Wolf (18) proposed a method of producing articles with smooth surface at extrusion rates above the critical extrusion rate for the onset of slip. Extrusion temperatures arc supposed to be at least 100[degrees]C above the melting point of the polymer. To our knowledge, this method to suppress sharkskin melt fracture and to reduce pressures in extrusion is practically used for processing of polymers with narrow MWD (e.g. HDPE), by blow molding, but has its limitations because of the "die drool", i.e. accumulation of polymer at the die exit.
Conventional Polymer Processing Additives
Processing of neat polymers by extrusion is rarely possible. Instead, it is a common practice to formulate compositions containing polymer and a variety of additives in relatively small, but critical amounts, sec e.g (19-21). Foremost among these additives are lubricants to reduce extrusion pressure and eliminate sharkskin melt fracture as well as the stick-slip instability (22). In the early 1960s. DuPont Canada accidentally discovered that fluorinated polymer added in a small amount to a LLDPE resin works as slip agents and postpones the onset of the sharkskin melt fracture. Mechanism of transfer of fluorinated polymer processing additive (PPA) to the die wall is investigated in (23). It was demonstrated that during the extrusion of PPA/LLDPE blends, droplets of fluoropolymer that are in close proximity to the surface of the die first coat the die entrance, and then they flow as streaks toward the die exit under the influence of shear field. Fluorinated polymers are still in use as PPA, e.g. Viton from DuPont, Dynamar from 3M, Kynar from Arkema, and Tecnoflon from Solvay Solexis (24). They are typically employed at concentrations of about 250 to 3000 parts per million (ppm) based on the mass of the thermoplastic material. During extrusion of the blend of LLDPE with PPA, the fluorinated polymer deposits at the die walls, displaces LLDPE and induces slip of the melt along the coated surface. Because of the slip the pressure that is necessary to extrude molten LLDPE through the die is reduced by about 20% from the reference values for extrusion of LLDPE without adding the fluorinated PPA.
According to Slattery and Giacomin (25), the plastics industry spends at least 2% of the cost of the resins used in extrusion on PPA to delay the sharkskin melt fracture. The total cost of the fluorinated additives used by the industry in 1993 was about $200 Million. The main problem, but not the only one, caused by the commercial use of fluorinated polymers as PPA, is a tendency for plate-out of decomposed fluorinated polymer on the extruder screw and/or the die exit, i.e. die build-up. The problem is often severe, requiring shutdown of the equipment and extensive clean-ups. Fluorinated polymers are extremely hydrophobic, and their use as PPA enhances the hydrophobic properties of the PE film, as well as the accumulation of static electrical charge at the film surface.
Deposits of fluorinated PPA inside the processing equipment can be essentially reduced and the accumulation of static electrical charge at the PE surface can be avoided as well as the amounts of fluorinated polymers needed to suppress sharkskin in extrusion of LLDPE can be diminished (by about a factor of three) if polyethylene glycol (PEG) is added to the polymeric material together with the fluorinated PPA. The use of such synergistic combinations is well known, see e.g (26-35). PPAs consisting of blends of PEG with fluorinated polymers for extrusion of PO resins are available commercially under trade names of Kynar, Dynamar, and Viton. Nevertheless, fumes from burning of the fluorinated polymers, e.g. during incineration of LLDPE film produced with fluorinated PPA, are toxic and potentially cancerous. As the fumes are Fluorine containing gasses, they deplete the ozone layer of our planet and their use, even at reduced amounts, in the composition of PPA cannot be desirable.
Mechanisms of Lubrication and Sharkskin Elimination
A cyclical instability in peeling of pressure-sensitive adhesive (PSA) tapes that involves alternate storage and dissipation of elastic energy is well known, e.g. in peeling of PSA tapes from glass surface (36). This is a so called "stick-slip" mode of peeling when smooth "adhesive" failure at the adhesive-glass interface is alternated with "cohesive" separation within an adhesive layer. It is easy to see some similarity in periodic structures at the PSA tapes in the "stick-slip" mode of peeling and sharkskin defects at the surface of an extrudate.
Conventional theories predict that adhesion of PSA to substrate should be proportional to its surface free energy but Newby and Chaudhury (37) demonstrated that adhesion on fluorocarbon surface is significantly larger than on some hydrocarbon surfaces, although the fluorocarbon surface has the lowest surface energy. Later they clearly evidenced slip motion of adhesive on clastic substrate (poly-dimethylsiloxanes) just before it peels out using tracking fluorescent particles placed at the interface, see (38), (39). Amouroux et al. (40) demonstrated that amplitude of interfacial slip motion in peeling of an acrylic tape from a silicone elastomer is correlated to the ratio of an elastic component of the complex Young modulus E' and a viscous component E" of the silicone elastomer. The slip motion is decreased for the reduced ratio E'/E" that is for reduced elasticity of the elastomer.
In accordance with ideas of Hill et al. (41) based on the apparent relation between adhesive failure and sharkskin melt fracture we believe that sharkskin instability is caused by swelling of the extrudate at the die exit, flowing of the molten polymer around the die rim, stretching of its surface layer outside the die, and periodical failures in adhesion of the molten polymer at the die rim. Due to stretching, the surface layer accumulates elastic energy and releases it during the act of adhesion failure. The adhesion failure propagates along the surface of the die as a crack. At the die rim, the crack deviates from the die wall into the melt and produces a seed crack, i.e. a notch. The seed crack grows, ruptures the surface layer and creates a valley at the surface of the extrudate. Upstream of the crack, the molten PE decelerates, flows along the die rim, and forms a ridge that will detach from the die in the next act of adhesion failure. Periodical repetitions of the melt fracture process create rough surface relief, i.e. "sharkskin", see (11) for details.
Actually, our tentative mechanism of sharkskin melt fracture differs from a widely accepted explanation of sharkskin proposed by Cogswell (42) only by a statement that sharkskin melt fracture starts from some seed cracks and these seed cracks originate from periodical adhesion failures at the die surface. We claim that an onset of sharkskin instability would be suppressed or substantially delayed if the seed cracks are healed or prevented. In extrusion of clay paste a similar fracture process may happen and we proposed healing of seed cracks to suppress surface fracturing of an extrudate (43). In particular, we proposed a die with a core extended beyond the die exit for defect-free extrusion of clay paste. In extrusion of clay paste sliding friction of the extrudate on that core outside the die generates back-pressure at the die exit. The back-pressure heals seed cracks so that a smooth extrudate can be produced. Similarly, healing of seed cracks is used in industry to reduce shattering of glass bottles and glass windows.
Our tentative mechanism of sharkskin explains experimental observations of a "mysterious" delay of a sharkskin onset in extrusion of LLDPE with additives of platelike nanoparticles of organically modified clay (44) and Boron Nitride (45) without slip inside the die. Indeed, plate-like nanoparticles in a surface layer of the extudate are oriented predominantly parallel to the die surface and therefore they stop propagation of seed cracks in transverse direction, i.e. into the extrudate, and an onset of sharkskin melt fracture is delayed.
We believe that tiny peculiarities of the fracture process, like: development of macroscopic fracture from microscopic seed cracks; slip of polymer along the die surface just before it peels out; instability of shear cracks at the boundary of two sliding bodies; and correlation of adhesion of polymer melt with elasticity of the substrate, are important and have to be taken into account. When molten polymer peels out from the exit of a metal die and slips along the surface just before the adhesion failure, the crack that separates molten polymer from metal is unstable to deviate from the boundary into the melt. Quite opposite, the shear crack that originates at the die entrance and develops in the direction of melt flow at some high extrusion rates does not deviate to inside the melt. More over, we believe that successive propagation of many such cracks shows up as slip of polymer inside the die, i.e. "super flow" of melt.
If the die is coated by an elastic material, e.g. silicone rubber (46), the crack of adhesion failure at the die exit would not deviate from the boundary into the polymer but rather in the opposite direction. No seed crack is produced in this case and without seed cracks sharkskin melt fracture would not develop. In (47) we coated an extrusion die by Boron-cured silanol and demonstrated that sharkskin melt fracture is suppressed if elasticity (E'/E") of the coating is higher than elasticity of melt. Elasticity of the coating (accumulation of elastic energy) helps to reduce friction losses, and therefore for the higher elasticity the melt better slips and we measure lower pressure at the extrusion die. Onset of sharkskin melt fracture can be delayed also in the case when the coating is essentially viscous in comparison with the melt (48). In this case the coating works like glue and dumps propagating of a crack of adhesion failure at the die exit so that detachment of the melt from the die surface goes steadily without oscillations.
Polymer melts show rubber-like behavior when deformed rapidly, and we can expect to find similarities between the slip of polymer melts and the slip of rubbers along a solid boundary. Schallamach (49) has noticed that the lateral force required to drag a glass lens across a rubber slab is reduced, when macroscopic detachment waves are formed. The Schallamach waves detach the rubber from the hard surface and slowly propagate in the direction of relative displacement of rubber, i.e. from a leading edge of a glass lens to its trailing edge. Similar structures we observed during slip of clay (50) and molten LLDPE (46) in transparent dies. These waves of macroscopic detachment or "surface cavitation" propagate in the direction of melt flow.
Mechanism of sliding of solid elastic bodies is a subject of intensive research, e.g (51). Gerde and Marder (52) proposed that separation waves or self-healing cracks, resembling bumps on rug, run along the interface causing one solid elastic body to slip over the other. Every crack in its propagation separates an elastic body from the other one and results in a small relative displacement. Actually, separation of the bodies can be of an atomic scale. The propagation of many cracks manifests itself in continuous slip. A flux of elastic energy is going to the tip of every crack to break the contact bonds between the bodies. At its tail, the banks of the crack collapse and the bonds recover, at least partially. Low adhesion (weak contact bonding) between the contacting bodies would promote slip. In the referred model of slip, the shear cracks propagate in the direction of bulk motion of an elastic body along the rigid surface. These self-healing shear cracks are radically different from the Schallamach waves of macroscopic detachment, which propagate slowly compared with the material wave speeds and show macroscopic detachment from the boundary.
Our tentative mechanism of slip of molten polymer is described in some more details in (11). We believe that slip of molten polymer can be explained in frames of the model of Gerde and Marder. For continuous and stable slip of melt along a solid surface shear cracks should propagate in the direction of melt flow. So, development of cracks from the die entrance to the die exit along the metal can be responsible for a continuous slip at some high rates of extrusion, i.e. for "super flow" of polymer melt, while propagation of shear cracks in opposite direction, i.e. from the die exit to its entrance, is unstable. Therefore, adhesion failures at the die exit are producing sharkskin defects. For the die with an elastic coating at the inside surface of the die the situation is reversed: cracks arc stable to develop from the die exit to the die entrance and we observe stable slip of the melt but shear cracks are unstable when they propagate from the die entrance to its exit and at some elevated rates of extrusion we observe severe melt fracture of the extrudate instead of the "super flow" of melt.
A technique to reduce friction losses in transportation of very viscous fluids is long known in industry. It is a "core-annual flow" technique. Core-annual flow is the pumping through a pipeline of a viscous liquid, such as crude oil or bitumen, in a core surrounded by a lower viscosity liquid such as water (53). Normally, core-annual flow is established by injecting water around viscous oil being pumped in a pipeline. However, core-annual flow can also be established above a certain shear rate in a pipe flow that is high enough to break a water/oil emulsion and create a water-rich zone near the pipe wall that acts as a lubricating film. Critical speeds for self-lubrication of water/oil emulsions are smaller when the viscosity of the crude oil is higher. The amount of water used for core-flow can be as low as 1% if water solvable salts are used such as sodium-silicate, phosphates, borates, sulphates, carbonates, and mixtures thereof (54). Ideal core-annular flow is unstable to waves and disturbances at the interface between the core and the lubricating film (55). These waves tend to break up into droplets of the film liquid, which disperses in the viscous phase. The lubricating fluid having a low viscosity but a high elasticity suppresses inequalities in film thickness, reduces dispersion of the film into the core of the flow and reduces friction losses in the pipeline, see (56).
Gels and Greases in Lubrication
Bones are sliding along each other in joints of our skeleton with a friction coefficient as low as 0.002. For comparison, the friction coefficient of smooth steel against ice and snow lubricated by a layer of water is about 0.05. Taking a close look at a human hip-joint we see that bones are coated by elastic cartilages and the gap between the cartilages is filled by a synovial fluid. The synovial fluid lubricates better than water, showing that the low friction coefficient observed cannot result from a hydrodynamic mechanism of lubrication (57). The synovial fluid is an elastic gel and a high MW (225 kDa) glycoprotein, namely lubricin, plays an important role there. A lack of lubricin and a change of the synovial fluid into a Newtonian fluid were observed in patients with rheumatoid arthritis, see Sokoloff (58) and Pieterse (59).
By definition a gel consists of an elastic network of a solid substance with a liquid filling the inner space of the network. The network holds the liquid in place through its interaction forces and so gives the gel solidity, elasticity, and coherence, but the gel is also wet and soft and capable of undergoing large deformation. Gels can be classified into four subgroups:
* Coagulated dispersions of colloidal particles which are ~ 1-1000 nm in size at low fractions of solids in a liquid (down to a few percent).
* Rubber-like cross-linked polymers swollen in a liquid solvent.
* Jellies, such as a solution of polymer molecules in a liquid in which molecules are cross-linked due to hydrogen or ionic bonds.
* Solutions of polymeric molecules of very high MW in a liquid. Molecular entanglements work as physical bonding between the molecules.
Grease is a semi-fluid to solid mixture of a fluid lubricant, a thickener (a gelling agent), and additives. Industry widely uses greases for lubrication of ball bearings and for parts sliding along each other. Organo-clay, slaked lime, silica fume, soaps of Lithium, Aluminum, Calcium, Barium, and Sodium as well as diuretane polymers are examples of thickeners. Additives enhance performance and protect the grease and lubricated surfaces from corrosion and oxidation. Fluid lubricants used to formulate grease are normally petroleum oils or synthetic fluids. Examples of synthetic fluid lubricants are also diesters, polyolesters. polyglycols, synthetic polycarbon fluids, silicones, chlorofluorocarbon fluids, and tri-phosphate esters, see e.g. (60).
Tertiary, i.e. tri-substituted, phosphate esters were first introduced as antiwear and extreme-pressure additives for lubricants in the 1930s. They are characterized by high affinity to metals and adsorption activity on the frictional surface. Molecules of the tertiary phosphate esters deposit from a fluid lubricant to the frictional surface and form a thin layer that prevents a contact of one metal part to another and frictional welding of the parts. Nowadays, the tertiary phosphate esters are used also as a fluid lubricant, especially for aircrafts (61).
PEG in Lubricants and PPA
Polyethylene glycols (PEG) and esters of PEG are routinely used in industry as a component of lubricants, cooling and cutting fluids in machining and metal working, e.g. (62.) (63). The use of PEG, polyethylene oxide (PEO), i.e. very high MW PEG and esters of PEG as a processing aid and a release agent in injection molding is also known. DeJuneas et al. (64) disclose that additives of PEG with MW from 600 to 20,000 Da in amounts from 0.02 to 0.05 wt% to PE resins for film blowing reduce the incidence of breakdown at typical operating conditions. The authors also mentioned in a description of the patent that PEG at levels of 1-3 wt% is an antistatic agent for PE films. Wolinski (65) discloses blends of PE and low MW PEG (MW is from 1000 to 6500 Da) to provide heat sealable PE film suitable for printing. Tikuisis et al. (66) propose the use of PEG with MW from 10.000 to 50,000 Da in amounts from 100 to 2000 ppm as PPA. They report that for extrusion of LLDPE resin that comprises hindered phenol primary antioxidant and a Phosphorus-containing secondary antioxidant, the use of PEG 35,000 results in less pressure at the extrusion die in comparison with the conventional PPA, i.e. a fluoroelastomer/PEG 8000 blend with a weight ratio of 1-2. The conditioning time to "clear melt fracture" was essentially the same for both PPAs. Blong and Lavallee proposed to use PEG and PEO in amounts up to 20 wt% as PPA for extrusion of fluorinated polymers (67). They report that extrusion of fluorinated polymers with additives of PEG is possible at reduced temperatures and without surface defects. According to Duchesne and Bryce (68), the use of PEG with MW of 400 and 3350 Da as an additive in the amount 0.2 wt% for extrusion of LLDPE at 210[degrees]C does not suppress sharkskin melt fracture. Chapman et al. (69) report that the use of PEG 8000 at a concentration of 480 ppm as an additive to LLDPE does not suppress sharkskin melt fracture within the observation time of about 1 h.
The use of esters of polyalcohols with number of carbon atoms from 2 to 6 (e.g. glycerol, pcntaerythritol, dec-aglyccrol, glycol, PEG 400) and saturated fatty acids, e.g. stearic acid, as PPA for processing of LLDPE was proposed by Williams and Geick (70). The inventors reported an 18% pressure reduction in extrusion of LLDPE by use of a 0.3 wt% blend of fatty acid esters as PPA at a temperature of 223[degrees]C and the suppression of sharkskin. Bauer et al. (71) proposed to use thermoplastic polyesters with melting points below 150[degrees]C as PPA for fiber spinning, film extrusion, and molding of PE resins. Dover Chemical recently announced Doverlube FL-599, an ester of glycols and fatty acids, for the use as PPA in processing of various polymers: High Impact Polystyrene (HIPS), Polystyrene (PS), PE, Polypropylene (PP), Acril-ButancButane-Styrene (ABS), and Polyvinylchloride (PVC). It was also reported that this PPA improves transparency of PP, reduces decomposition of polymers inside processing equipment, and can be used as a component of purging blends for cleaning of processing equipment. Esters of PEG with MW below 1500 Da and boric acid are proposed by Sato for the use as a release agent (72) and as a component of a purging blend for cleaning of processing equipment (73). The use of a reacting mixture of PEG with boric and phosphoric acids as a PPA to suppress sharkskin in extrusion of LLDPE is proposed recently in (74).
The use of a blend of PEG and fine powder of mineral particles with sizes from 3.5 to 12 [micro]m as a processing aid is also known. Corwin et al. (75) proposed the use of an antigel composition in extrusion of PE films wherein the antigel composition can be a blend of PEG with MW from 200 to 4,000,000 Da with an inorganic antiblock agent and a hindered phenolic antioxidant. Li and coworkers (76) described the use of PEG as a processing aid for extrusion of UHMW PE/PP blend. Li and coworkers (77), (78) also report a suppression of sharkskin melt fracture for additives of a binary blend of PEG and diatomaceous earth in amounts from 0.5 to 3 wt%. According to the authors, the blend works synergistically and PEG alone does not suppress sharkskin melt fracture. The use of blends and reactive mixtures of PEG and chemical substances comprising Phosphorus and oxygen, e.g. phosphoric acid, organo-phosphites, and organo-phosphates, as PPA to suppress sharkskin in extrusion of LLDPE is proposed recently in (79). Optionally, the reactive mixture can comprise fumed silica and PEO, i.e. PEG with very high MW, as a passive thickening agent.
Esters of PEG in Lubricants and PPA
Commercially available LLDPE grades comprise low MW organo-phosphites as antioxidant additives. Such antioxidants are described and classified in (80). The organophosphites deplete oxygen from the PE melt and get converted to organo-phosphates, i.e. low MW esters of alcohols and phosphoric acid. As stated earlier, organo-phosphates have a high affinity to steel alloys and the surface layer of steel parts gets in time saturated with Phosphorus oxide (PO). Saturation of steel alloys with PO, i.e. phosphating of steel, is used in industry. Actually, the first patent on phosphating by the use of phosphoric acid was granted in 1907 to Coslett (81).
Ester condensation reactions in a mixture comprising monomers, such as polyhydric alcohol and polyfunctional acids are well known (82). During the condensation reaction, if the temperature of the reactive mixture is sufficiently high and applied for a sufficient time, the composition would convert to a water stable alkyd resin. Noda et al. (83) describe in a recent patent application the use of reacting mixtures of polyhydric alcohol and a polyfunctional organic acids for manufacturing of films, fibers, foams, molded, and extruded articles.
PEG reacts under heating with carboxylic acids and anhydrides of the carboxylic acids, e.g. stearic acid, oxalic acid, succinic acid, adipic acid, citric acid, maleic anhydride, phthalie anhydride. Reactivity is higher with PEG of lower MW as well as in presence of strong mineral acids, e.g. sulfuric acid, and catalysts. If weights of the reacting components are selected close to stoichiometry and the mixture has in average more than two reacting groups per molecule, the ester-condensate is a thermoset at room temperature. In the presence of moisture and at elevated temperatures, the product of the polycondensation reaction, i.e. ester-condensate, is in equilibrium with reactants and both direct and reverse reactions occur so that covaient bonding between molecules breaks and reappears in time. Therefore at such conditions the blend of the ester-condensate and reactants behaves under applied stress not as a thermoset but as a visco-elastic fluid.
Reacting Mixtures of PEG with Polyacids as Novel PPA
We propose here novel PPA at concentrations from 0.02 to 0.5 wt% for processing of polymers at reduced temperatures. Actually, our proposal is an application and development of the "core-annual flow" transportation idea of viscous oil to the extrusion of plastics. Due to our proposal molten polymer is blended inside the extruder with a Sow-viscous lubricant. Both components do not dissolve in each other, and when such an emulsion flows through a long and narrow channel, the low-viscous component actively migrates to the wall of the channel. In contrast to the original "core-annual How" idea we propose to use as a low-viscous component a reacting mixture that comprises PEG with MW of about 300 to 4000 Da and at least one of following reactants: organic acid, anhydride of organic acid, phosphoric acid or Phosphorus oxide, and low MW ester of oxiacids of Phosphorus.
Because of the higher affinity to metal, the low-viscous component displaces molten polymer at the wall and forms a lubricating film. At the wall of the channel the reacting mixture turns into grease-like material that resists to flow, separates mollen polymer from the wall and lubricates the wall so that the melt slips along the lubricating film. The thickness of the lubricating film and the efficiency of lubrication may be enlarged if some amount of a passive thickening agent, e.g. fumed silica and PEO, is blended with PEG. Indeed, because of friction with the wall, nanoparticles of silica and long molecules of PEO accumulate at the die wall and help to turn the low-viscous reacting mixture into a viscoelastic gel. The PPA may be applied in concentrations from 0.02 to 0.5 wt% and with the advantage that there is no need to use a master batch. Instead, the additive is supplied as pellets, spray or by flow of liquid PPA directly into the extruder.
As indicated above, nowadays synergistic blends of PEG with fluorinated polymers are used in industry to improve extrusion of PO resins with narrow MWD. We show here that fluorinated polymers are not a necessary component of PPA. The use of reacting mixtures can be a cheap alternative to the existing fluorinated polymer/PEG blends, and can be especially useful for extrusion of PO resins with high amounts of fillers. The PPA can be used as a component of the master batch comprising the fillers to improve their dispersion in polymer matrix. Opposite to fluorinated polymers, the novel PPA does not accumulate inside the processing equipment and at the die exit, but rather at long exposure to heat and oxygen it disintegrates to substances of low MW, and evaporates from the equipment. The fumes they potentially generate by incineration or heating are nontoxic. Viscous fluorinated PPA stay trapped inside the polymer matrix, and only a small fraction of the PPA is used to form a lubricating layer at the die surface. In contrast, low viscous reacting mixtures made of PEG actively migrate to the outside of the melt and form a lubricating film inside the die around the viscous core of molten polymer.
Experimental Methodology and Materials Used
In a series of experiments on extrusion we used LLDPE resins from ExxonMobil Chemicals with narrow MWD, LL1201 XV (density 0.925 g per cc. melting point 123[degrees]C, and MI = 0.7 g/10 min) and LL 1001 XV (0.918 g per cc, melting point 120[degrees]C and MI = 1 g/10 min). One grade of HDPE resin was used, HPA 020 (0.952 g per cc, melting point 127[degrees]C and MI = 0.07 g/10 min). These grades have antioxidant additives. To produce a master batch with our PPA, we used LLDPE LL6301LL 6301 RQ (melting point 125[degrees]C MI = 5.0 g/10 min) from ExxonMobil Chemical supplied as a coarse powder. We also used for comparison LDPE LLI66 BA (0.923 g per cc, melting point 110[degrees]C and MI = 0.2 g/10 min) with a wide MWD from ExxonMobil Chemicals. We used the following chemicals in our experiments:
* PEG 200 PEG 300, PEG 400, PEG 600, PEG 1000. PEG 1500, PEG 2000, PEG 4000. PEG 6000, PEG 8000, PEG 10.000, and PEO with MW above 5,000,000 Da.
* Oxalic acid, adipic acid, succinic acid, citric acid, stearic acid, phthalic anhydride, and phosphoric acid from Aestar.
* Fumed silica "Aerosil 300" from Degussa.
* Fluorinated PPAs: Viton Free Flow SC-PW and Viton Free Flow Z100 from company DuPontDaPont, Kynar PPA from Arkema, Dynamar E- 15653 from 3M.
* Hostanox PAR 62, aromatic organophosphite (antioxidant) from Clariant
For extrusion of PE resins, we used a single screw extruder from the Extrudex (Germany) with flood feeding. This extruder has 4 axial grooves in its feeding zone to ensure processing of raw reactor powders of PO resins (84). The grooves are 8 mm in width and their depth is gradually decreasing from 2 mm to 0 in the machine direction. The feeding zone of the extruder is cooled by water. A big advantage of an extruder with a grooved feeding section is that the mass rate of extrusion (MRE) is less sensitive to pressure variations at the extrusion die in comparison with smooth bore extruders. All parts of the extruder that are in contact with molten polymer are made from a special steel alloy (34CrAlNi7) and nitrided, i.e. saturated by Nitrogen to harden a thin (about 0.1 mm) surface layer of the parts. The extrusion die has a diameter of 2 mm and a length of 60 mm and is also made from nitrided steel. The 2 mm bore of the die was conjugated with a 50[degrees] cone having a diameter of 8 mm at the entrance of the die. Extrudate comes out from the die in the downward direction.
For extrusion of PE resins with PPA, we mixed pellets of PE resin and PPA in a plastic drum by shaking and rotating the drum. To ensure that PPA with low melting temperatures, e.g. made with PEG, is distributed evenly on the pellets we heated pellets of PE resins to about 90[degrees]C. Before every extrusion trial we cooled the pellets with the additives to room temperature.
A pressure transducer was arranged near the die entrance and its reading were used for data acquisition at rate about 200 Hz. Pressure reduction (PR) [bar] data are calculated by comparison of pressures in extrusions with PPA (pressure measured or PM) and without any PPA (reference pressure or RP), i.e. PR = PM--RP. The reference pressure by use of the die 2 X 60 mm and neat LLDPE is 398 bar. The data on pressure reduction (%) in the figures are presented as relative values, i.e. as a ratio of PR/RP. MRE was monitored in all experiments. In a separate extrusion trial we measured extrusion pressure of neat LLDPE vs. MRE. In the case of extrusion with PPA when the measured MRE differs from the reference value of MRE for extrusion without additives, the reference pressure (RP) for pressure reduction (PR) calculations was selected that corresponds to the measured MRE.
The "Flow Curve," which is a plot of apparent shear stress versus apparent shear rate, is commonly used for rheological characterization of molten polymers. However, we discuss here sharkskin melt fracture and slip phenomena, which may include friction and fracture. The use of reduced quantities like apparent shear stress and apparent shear rate in the characterization of friction and fracture would be confusing. Therefore we present characteristic curves of the directly measured quantities, i.e. the pressure (P) at the die entrance versus the averaged extrudate velocity (V) to characterize the flow in a circular die. The average extrudate velocity (V) is derived from the volumetric flow rate Q by
V = 4Q/[pi][d.sup.2] (1)
where d is the diameter of the die. Below we will also use a term "extrusion rate" instead of V. If needed, the apparent shear stress [[tau].sub.a] can be calculated from the measured pressure P by
[[tau].sub.a] = 4L/d P (2)
where L is the length of the die, and the apparent shear rate [[gamma].sub.a] is given by
[[gamma].sub.a] = 32Q/[pi][d.sup.3] = 8V/d (3)
Fluorinated PPA and Fluorinated PPA/PEG Blends
As a reference we prepared and extruded blends of Viton/LL1201, Dynamar/LL1201 at concentrations of 0.1 wt% of the additives, and Kynar/LL1201 at 0.2 wt% of the additive. For industry it is important to know not only the conditioning time "to clear the sharkskin melt fracture," but also the time to clean the processing equipment from the additives used by purging with neat polymer. Therefore, we first extruded 1 kg of the PPA/LLDPE blend, and then we purged the extruder with neat LLDPE, at a temperature of 165[degrees]C. The lubrication charts, i.e. percentage of pressure reduction as a function of time since start-up of extrusion in the case of Viton and Dynamar (concentration 0.1 wt%) and Kynar (concentration 0.2 wt%) are shown in Fig. 1. The time, when we changed from feeding the PPA/LLDPE blend to feeding neat LLDPE, is shown in Figs. 1, 2, and 3 by a vertical line (approximately 2 h 20 min after start of the extrusion). These and further extrusion experiments were made at an average extrudate velocity of about 40 mm/s and at a temperature of 165[degrees]C if not otherwise indicated. Residence time, i.e. the time for polymer material in the extruder, is about 15 min. With the use of fluorinated PPA as well as other PPA in following extrusion trials the sharkskin melt fracture was suppressed when Pressure Reduction was above 18-20%.
[FIGURE 1 OMITTED]
As it is possible to see from Fig. 1, the use of Dynamar results in slightly better lubrication than the use of Viton, but conditioning time is longer in this case. We observed accumulation of static electrical charge at the extrudate, when Viton and Dynamar are used as PPA. Using Kynar, which is a blend of some fluorinated PPA with PEG, no static electrical charge of the extrudate was observed in this experiment. We may also see that the use of Kynar results in a higher lubrication in comparison with the use of Viton or Dynamar, and the conditioning time is the shortest. A similar improvement in lubrication was observed when Viton FF Z100 blend was used as PPA in a concentration of 0.2%. This blend also comprises some amount of PEG. We see from the experimental curves that conventional PPA based on fluorinated polymers accumulates inside the extruder, and cannot be removed by purging of processing equipment with neat LLDPE. To clean up the extruder from Viton, Dynamar, and Kynar, we had to purge it by a blend of LLDPE and mica. The die was heated to 450[degrees]C in an electrical furnace to burn off the fluorinated PPA from its surface after extrusion trials with fluorinated PPA. A nitrided die from a steel alloy 34CrAlNi7 can stay heating up to 500[degrees]C without loosing Nitrogen.
To understand the mechanism of synergistic improvement in lubrication by the use of blends of PEG and fluorinated polymers, we prepared blends of Viton with PEG 8000 where fractions of Viton are from 10 to 80 wt%. Neat Viton and PEG 8000 were also used as additives. We made a sequence of extrusion trials with a gradual increase of the fraction of Viton in the blend used as PPA: First we extruded LLDPE with PEG 8000; next we used LLDPE with a Viton/PEG 8000 with 10 wt% of Viton, etc., till the final extrusion trial when only Viton was used as PPA. As seen from Fig. 1, it takes a while to obtain a stationary pressure at the extrusion die. Extrusion pressures and MRE at stationary conditions are presented in Table 1. Using neat PEG 8000 as PPA, it took almost 8 h to obtain a stationary level of lubrication. As seen in Table 1, MRE is stable vs. fraction of Viton in the PPA, while pressure reduction reaches a maximum at a fraction of Viton of about 30 wt%. This maximum value of pressure reduction is indeed much larger than for neat Viton, but it differs very little from the effect of neat PEG 8000 as PPA. From these measurements we may conclude that the fluorinated polymer is a not necessary component of a PPA, and the only justification for the use of fluorinated polymers as a component of a PPA is the relatively short conditioning time of the Viton/PEG blend.
TABLE 1. PPA in a concentration 0.1 wt% is made by blending Viton and PEG 8000. Fraction of Viton in PPA. wt% 0 10 20 40 60 80 100 Max. pressure reduction (%) 47 48 50 50 43 38 26 Change in mass rate (%) 0.3 -0.2 -0.3 0.3 -0.2 -1 -0.2 Temperature is 165[degrees]C. For all Tables extrusion of LLDPE with anti-oxidant (LL1201 XV) and PPA is done at averaged extrudate velocity 40 mm/s through the die 2x60 mm from nitrided steel
To estimate an impact of the die material on the efficiency of PPA and the conditioning time "to clear sharkskin melt fracture," we made extrusion trials of LLDPE with PEG 10,000 in concentration of 0.1 wt%. We used a die of diameter 2 mm and length 20 mm fabricated from nitrided steel 34CrAlNi7. The die was heated before the trials to a temperature of 450[degrees]C and then washed in water. One kilogram of the PPA/LLDPE blend was extruded at 165[degrees]C with an extrusion rate of about 40 mm/ s followed by neat LLDPE to clean the extruder. The pressure reduction curve derived from data acquisition of a pressure transducer is presented in Fig. 2 by a full line as a reference. In a similar way as for fluorinated PPA, sharkskin is suppressed when the pressure reduction exceeds about 20%. For the next experiment the die was cleaned as before, but the extruder was purged for about 10 h with neat LLDPE before the blend of PEG/LLDPE was fed to the extruder. The pressure reduction curve is presented in Fig. 2 by a dotted line. The LLDPE grade that we used for our experiments contains antioxidant, and the die surface will be saturated by organo-phosphates after long extrusion of the LLDPE.
[FIGURE 2 OMITTED]
An increase of pressure reduction in comparison with a fresh and clean die proves that the accumulation of the organo-phosphates in the extruder helps to improve lubrication. Then we treated the die by nitric acid to remove the nitrided layer from the die surface, and in this case the measured lubrication values were much less that for the nitrided die (dashed line in Fig. 2). In a similar experiment with the die treated by phosphoric acid, i.e. with a phosphated die, we observed slightly improved lubrication. The etched surface of the die is not as smooth as for the original nitrided die, so the experiments are not perfect but quantitatively we can testify that saturation of the steel alloy by Nitrogen and Phosphorus improves lubrication. In an extrusion experiments with a die of dimensions 2 X 20 mm made from aluminum alloy we obtained pressure reduction values that are compatible with the values for the nitrided die of the same sizes, whereas the use of a die made from stainless steel demonstrated considerably less lubrication than the reference experiment.
In other experiments, we charged a portion of about 1 g of urea into the extruder to "deactivate" acids inside the extruder and the die (2 X 60 mm) from nitrided steel during purging of the extruder by neat LLDPE. Urea is decomposing upon heating to ammonia and carbon dioxide, and ammonia reacts with phosphoric acid. Five hours later we started extrusion of 2 kg of LLDPE with additives (500 ppm) of PEG 1000 with silica fume (1 wt% from the PEG). Only little pressure drop was detected (about 5%) in the end of the extrusion time but after further purging of the extruder by neat LLDPE for about 15 h the same additives gave 16% of pressure reduction and partial suppression of sharkskin. Next, after 15 more hours of purging the pressure reduction in a similar experiment was about 18% with complete sharkskin suppression.
Organo-Phosphites and Organo-Phosphates
To prove the importance of having organo-phosphites and organo-phosphates as a component of PEG based PPA, we prepared mixtures of PEG 1500 with phosphoric acid (PA), sorbitol (SO) and fumed silica (SF). The pressure reduction curves are presented in Fig. 3: neat PEG 1500 (a solid line); 100 g of PEG 1500 mixed with 4 g of PA (a dashed line); 100 g of PEG 1500 mixed with 25 g of Hostanox PAR 62 (a dotted line); 100 g of PEG 1500 mixed with 8.6 g of phosphoric acid, 4 g of sorbitol and 1 g of boric acid (a dot-dashed line). The blends comprising phosphoric acid and organo-phosphite were heated to about 160[degrees]C in vacuum, and the reaction products were used as PPA in concentrations 0.1 wt%. The experimental curves presented in Figs. 1, 2, and 3 are derived from data acquisition of a pressure transducer in extruding of 1 kg of the PPA/LLDPE blend and 2 kg of neat LLDPE.
[FIGURE 3 OMITTED]
To find the optimum amounts of organo-phosphite and organo-phosphate, we prepared mixtures of PEG 1500 with phosphoric acid and PEG 1000 with organo-phosphite Hostanox PAR 62. The mixtures were heated to 160[degrees]C in vacuum, and the reaction products were used as PPA in concentrations 0.1 wt%, and two sequences of extrusion trials were made by extruding 1 kg of the PPA/LLDPE blend and 2 kg of neat LLDPE with gradually increasing amounts of the organo-phosphites and organo-phosphates in the PPA. The maximum pressure reduction obtained during the extrusion trials is presented in Fig. 4 vs. the ratio of the number of Phosphorus atoms to the number of PEG molecules. Maximum efficiency of lubrication is achieved when the molar ratio of reactants is getting close to stoichiometry, i.e. for [P]/[PEG] ratios from 1/3 to 2/3. The molar ratio of reactants in the case of Hostanox PAR 62 has to be higher, probably because not all organo-phosphite is converted to organo-phosphite during extrusion. We also observed a higher efficiency of lubrication if the blend of Hostanox PAR 62 with PEG 1000 was heated in air to higher temperatures, i.e. 260[degrees]C instead of 160[degrees]C. It gives us a hint that for better lubrication, the organo-phosphite in the composition of the PPA has to react with oxygen.
[FIGURE 4 OMITTED]
Impact of MW of PEG and Additives of Silica Fume
To determine an impact of MW and concentration on the lubrication efficiency in extrusion of LLDPE LL1201 XV, we made several sequences of extrusion trials with PEG of various MW at increasing concentrations of PEG in the range from 250 to 4000 ppm, at 165[degrees]C and 40 mm/s extrusion rate. We prepared blends of PEG/LLDPE, and measured extrusion pressures at stationary state, i.e. in the case of 250 ppm PEG addition, the pressure readings were taken after about 10 h of extrusion, for 500 ppm after 5 h and for 1000, 2000, and 4000 ppm after 2 h 30 min of extrusion. Pressure reduction data vs. MW of PEG are presented in Fig. 5. At high concentrations of PEG, maximum lubrication is observed for PEG 4000, whereas at low concentrations of PEG, maximum lubrication is obtained for PEG with MW of 8000 or higher. We believe that the pressure reduction at high concentrations of low viscous PEG can be explained by separation from the blend and migration of the low viscous component to the die wall, i.e. the establishment of "core-annual flow," see references above.
[FIGURE 5 OMITTED]
We demonstrated already that additives of PEO and fumed silica improve lubrication with PPA from low viscous PEG, see (85). To determine the impact of fumed silica (SF), we made several sequences of extrusion trials with PPA made from of blends of PEG of various MW, fumed silica (1%) and phosphoric acid, in the range of PPA concentrations from 250 to 4000 ppm at 165[degrees]C and 40 mm/s extrusion rate. The extrusion trials were made in a similar way as for neat PEG. Pressure reduction data vs. MW of PEG are presented in Fig. 6. The maximum lubrication at high concentrations of PPA is attainable for PPA made from PEG 2000. By comparison Figs. 5 and 6 we can see that the use of silica fume improves lubrication for high concentrations of PEG with MW below 6000 but reduces it for low concentrations of PEG with MW above 6000. As a reference we made a sequence of extrusion trials of Viton/LLDPE blends at gradually increasing concentrations of Viton. In contrast to extrusion of PEG-based PPA, pressure reduction for fluorinated polymers is nearly independent of the amount of PPA under stationary conditions, but the conditioning time is shorter at higher amounts of Viton.
[FIGURE 6 OMITTED]
To investigate an impact of MW of PEG on the efficiency of PPA for extrusion at reduced temperatures, we prepared blends of PEG/LLDPE for the following MW of PEG: 200, 600, 1000, 1500, 2000, 4000, 6000, 8000, 10,000. A sequence of extrusion trials with PEG of various MW in concentrations 0.5 wt% were made by extruding 1 kg of the PPA/LLDPE blend through the extruder and the die, and then purging it by 2 kg of neat LLDPE at an extrusion rate about 40 mm/s. Pressure reduction from reference pressure without additives at a temperature of 138[degrees]C of the PPA/LLDPE blends vs. MW of PEG are presented in Table 2. We can see that the best lubrication was achieved by the use of PEG with MW from 1000 to 4000 Da. For PEG 200 we did not observe suppression of sharkskin, and we detected an 11% reduction in the MRE.
TABLE 2. PPA in concentration 5000 ppm is PEG. Temperature is 138[degrees]C. MW of PEG, 1000 Da 0.2 0.6 1 1.5 2 4 6 8 10 20 Max. pressure 11 41 67 71 69 61 53 47 45 41 reduction (%) Change in mass rate -11 -5 +2 +1 +2 +2 -1 -11 -20 -33 (%)
PEG is slightly soluble in molten LLPDE and it reduces surface tension and viscosity of the melt. In extrusion experiments with the use of PEG with molecular weights above 6000 Da local inclusions of PEG appear as shallow depressions at the extrudate surface. In our experiments with PPA from low viscous PEG (MW is below 4000 Da) the extrudate has surface without detectable depressions. Therefore, we can conclude that all this PPA is ejected from the polymer.
Reacting Mixtures of PEG and Organic Acids as PPA
PEG reacts with phosphoric acid and organic polyacids, and therefore it is possible to shorten conditioning time by the use of a reacting mixture of PEG with the acids (86). To compare the efficiency of various organic acids in the composition of PPA, we prepared reacting mixtures of PEG 6000 with 2 wt% of following acids: oxalic acid, adipic acid, succinic acid, citric acid, and stearic acid. The components of the reacting mixtures were fed to the extruder as powders and extruded through a die with 2 mm in diameter and 60 mm in length at a temperature about 60[degrees]C by the same extruder that we used for extrusion experiments. Similar to extrusion of chocolate (87), the extrudate is soft and plastic immediately after melting and solidification of PEG inside the extruder but in a short while (in about a minute) it gets solid and fragile. So, the filaments produced were crashed after solidification to small pieces (about 3 mm long) and mixed with pellets of LLDPE for extrusion in amounts of 2 g of every PPA to 1 kg of LLDPE. For comparison we also extruded neat PEG 6000 and mixed it with LLDPE. Extrusion trials were made at 165[degrees]C in a similar way as it is described above. The measured maximum pressure reduction values and conditioning times to clear sharkskin are presented in Table 3 for neat PEG 6000 and the blends of PEG 6000 with the acids. Oxalic and citric acids show the shortest conditioning time. The use of citric acid improves lubrication, whereas the use of succinic, adipic, and stearic acids worsens lubrication in comparison with the use of neat PEG 6000 for extrusion of LLDPE with antioxidants, but the differences are small.
TABLE 3. PPA in a concentration 0.2 wt% is a reacting mixture of PEG 6000 and organic acid (2 wt% of PEG). Temperature is 165[degrees]C. Name of the blend CA RF OA StA AA SA Max. pressure reduction, % 55 49 48 47 46 42 Conditioning time, min 36 77 33 36 36 37 CA is a blend of PEG 6000 with citric acid; RF is neat PEG 6000 (with anti-oxidant); OA is a blend with oxalic acid; StA is a blend with stearic acid; AA is a blend with adipic acid; SA is a blend with succinic acid. A conditioning time includes a residence time to transport pellets in an extruder, ca. 15 min.
In another set of experiments we compared efficiency of various organic acids in the composition of PPA when the molar ratio of reactants is close to stoichiometry. We prepared reacting mixtures of PEG 2000 with following acids: oxalic acid, adipic acid, succinic acid, citric acid, stearic acid, phosphoric acid, and phthalic anhydride. Silica fume was used as a passive thickening agent in concentration of 1 wt% and phosphoric acid (1.2 wt%) was used as a catalyst and adhesion enhancer. We added organic acids to PEG with fumed silica and phosphoric acid under heating to about 180[degrees]C and intensive stirring. Liquid PPA (1 g) was blended with 2 kg of hot pellets of LLDPE to ensure homogeneous distribution of the additives on the pellets. Next, the pellets were cooled down to room temperature and extruded through a die (2 X 60 mm) from nitrided steel. We used 2 kg of neat LLDPE to purge the extruder after every extrusion trial. The measured values of maximum pressure reduction and conditioning times to clear sharkskin are presented in Table 4 for PEG 2000 with silica fume and phosphoric acid as a reference, as well as for further blends with organic acids. The use of citric acid and phthalic anhydride, improves lubrication in comparison with reference composition of PPA: PEG 2000 with silica fume (1 wt% from PEG) and phosphoric acid (1.2 wt% from PEG).
TABLE 4. PPA in concentration 500 ppm is a reacting mixture of PEG 2000 and organic acid, with additives of phosphoric acid (1.2 wt% from PEG) as a catalyst and silica fume (1 wt% from PEG) as a thickening agent. Temperature is 165[degrees]C. Name of the blend RF2 PA CA PhAn OA NT SlA AA SA Max. pressure red (%) 38 40 39 36 33 33 30 27 25 Max. torque red (%) 35 23 36 26 29 34 33 28 31 Condit. Time, min 82 41 38 59 87 122 122 93 213 RF2 is a blend of PEG with silica fume (1 wt% from PEG) and phosphoric acid (1.2 wt% from PEG); PA is a mixture of RF2 with phosphoric acid (3.8 wt% from PEG); CA is a mixture of RF2 with citric acid (9.6 wt% from PEG); PhAn a mixture of RF2 with phtalic anhydride (14.8 wt% from PEG); OA is a mixture of RF2 with oxalic acid (12.6 wt% from PEG); RF is PEG with silica fume (1 wt% from PEG) without phosphoric acid; StA is a mixture of RF2 with stearic acid (14.2 wt% from PEG); AA is a mixture of RF2 with adipic acid; SA is a mixture of RF2 with succinic acid. A conditioning time includes a residence lime to transport pellets through an extruder, ca. 15
Citric acid looks as the most attractive reagent to prepare reacting mixtures with PEG. To find an optimum amount of citric acid in the composition of PPA, we used a blend of PEG 2000 with citric acid, phosphoric acid (1.2 wt%), and silica fume (1 wt%). Conditioning times to clear sharkskin and maximum pressure reduction values are presented in Table 5 for various concentrations of citric acid. We see that with increasing concentration of citric acid from 6.4 wt% to 9.6 wt% the conditioning time gets short but at high concentrations lubrication is reduced. We also prepared pellets of PEG 2000 with 2 wt% of citric acid and used the PPA in a concentration of 0.2 wt% for extrusion of LLDPE (LL1201 XV). The resulting pressure reduction is presented in Fig. 1 by a dash-doted line. For comparison we made extrusion by the use of neat PEG 2000 as PPA, and the curve of pressure reduction is also presented in Fig. 1 by a dot-dash-doted line. From Fig. 1 we see that the blend of PEG 2000 with citric acid shows a very short conditioning time similar to the use of Kynar at the same concentration, but much better lubrication than any of the fluorinated PPA. The PPA can be used in small concentrations. For example, in extrusion of LLDPE with PPA made from PEG 2000 with 6.4 wt% of citric acid, 1 wt% of silica fume and 1.2% of phosphoric acid in a concentration 250 ppm pressure reduction was about 17% with total suppression of sharkskin. This proves that fluoropolymer is not a necessary component of a PPA, and much better results in lubrication and suppression of the sharkskin during extrasion of LLDPE can be achieved by the use of novel PPA.
TABLE 5. PPA in concentration 500 ppm is a reacting mixture of PEG 2000 and citric acid (wt% from PEG) with additives of phosphoric acid (1.2 wl% from PEG) as a catalyst and silica fume (1 wt% from PEG) as a thickening agent. Temperature is 165[degrees]C. Amount of citric acid, wt% 0 (RF2) 3.2 6.4 9.6 12.8 Max. pressure reduction (%) 38 38 41 39 32 Max. torque reduction (%) 35 36 38 36 26 Conditioning time, min 82 65 74 38 66 RF2 is a blend of PEG with silica fume (1 wt% from PEG) and phosphoric acid (1.2 wt% from PEG). A conditioning time includes a residence time to transport pellets in an extruder, ca. 15 min
PEG-based PPA in Extrusion of and LLDPE
To compare an impact of PPA on the extrusion of PEs with wide and narrow MWD, we used a blend of PEG 2000 with 1% of silica fume) in a concentration 0.5 wt% for extrusion of LDPE (LD 166BA) and LLDPE (LL 1001 XV) at a temperature of 165[degrees]C and for extrusion rates from about 4 to 100 mm/s. For comparison we extruded these PE resins also without PPA. Characteristic flow curves, i.e. curves of extrusion pressure at the die versus extrusion rate are presented in Fig. 7. The best-fit curves connect experimental points for extrusion of PE resins without PPA: a full line with symbols of open circles for LLDPE, and a dashed line with open squares for LDPE. For extrusion of PE resins with PPA, the best-fit curves connect experimental points: a full line with symbols of full circles for LLDPE, and a dashed line with full squares for LDPE. At the characteristic curve for extrusion of LLDPE without PPA, the onset of sharkskin (at 4 mm/s) and the onset of stick-slip (at 52 mm/s) instabilities are marked by arrows. Opposite to extrusion of neat LLDPE, extrusion of LLDPE with PPA is stable at extrusion rates up to 80 mm/s, where the so-called elastic instability (entrance melt fracture) occurs. While extrusion of LLDPE is greatly improved by addition of the PPA and pressures at the extrusion die are reduced 4-5 fold in comparison with extrusion of LLDPE without PPA, there is only a marginal pressure reduction in the case of extrusion of LDPE with PPA. We also made an extrusion trial with a HDPE resin, HPA 020, and with 0.2 wt% of PPA composed from PEG 2000 and 2 wt% of citric acid at the temperature 165[degrees]C and for an extrusion rate 44 mm/s. The HDPE grade comprises a big portion of HDPE resin of low MW. Therefore after an extrusion time of 2 h 30 min, pressure reduction was only 39%.
[FIGURE 7 OMITTED]
Extrusion of LLDPE vs. Temperature
We compared the extrusion of LLDPE with and without PPA at various temperatures: 130, 135, 145, 165, 185, 205, 225, 235[degrees]C. Characteristic flow curves, i.e. curves of pressure at the extrusion die versus extrusion rate (expressed as average velocity of the extrudate) for extrusion trials of LL1201 XV without PPA are presented in Fig. 8 for two temperatures: 145[degrees]C (full line) and 225[degrees]C (dashed line). The best-fit curves connect experimental data: symbols of open circles (145[degrees]C) and open squares (225[degrees]C). At the characteristic curve for extrusion at 145[degrees]C, the onset of sharkskin (at 2 mm/s) and stick-slip (at 27 m/s) instabilities are marked by arrows. For extrusion at higher temperatures onsets of surface instabilities are delayed. However, for LLDPE with narrow MWD the sharkskin instability appears at low rates of extrusion (below 6 mm/s for LL1201 XV) even at temperatures as high as 225[degrees]C. For experiments with PPA we mixed under heating PEG 2000 with silica fume (1%) and added 5 g of this PPA per 1 kg of the LLDPE pellets. Extrusion trials were made at the same temperatures as without the PPA. In the range of extrusion rates from 2 to 45 mm/s and for temperatures below 185[degrees]C, the measured pressures showed little variations with temperature. Surprisingly, we observed an increase of the extrusion pressures at temperatures from 185 to 235[degrees]C. Characteristic curves of extrusion of LLDPE with PPA for two temperatures: 145 (full line) and 225[degrees]C (dashed line) are presented in Fig. 8. The best-fit curves connect experimental data: symbols of full circles (145[degrees]C) and full squares (225[degrees]C). So, the extrusion pressures at the die can be considerably less for reduced temperatures of extrusion than for elevated temperatures, if LLDPE is blended with PPA. This result is quite opposite to the extrusion of neat LLDPE, where apparent viscosity drops at elevated temperatures.
[FIGURE 8 OMITTED]
Pelletizing of LLDPE Resins with Air Cooling
In our experiments extrusion is going downward from the die, and the die is about 1 m above the floor. At reduced temperatures the extrudate quickly cools down by air convection and solidifies. To demonstrate an opportunity to produce a pelletized master batch with some compositions of PPA from viscous PEG, we blended coarse powder of LLDPE (LL6301RQ) with 3% of PEG 6000 and extruded it at temperature of about 135[degrees]C from a die 2 mm diameter. A solid strand was cut by a rotating knife of a meat grinder to produce a pelletized master batch. We also used a rotating knife of a meat grinder attached to an electrical motor to cut strands directly at the hot die face. Pelletizing of LLDPE material without PEG additives is hardly possible at low temperature of extrudate, e.g. 138[degrees]C, not only because of high pressures at the die but also because of sharkskin and pronounced die drool, i.e. accumulation of molten PE resin at the die exit. PE melt does not stick to the metal surface wetted by PPA. Pellets cut by the rotating knife at higher temperatures, e.g. 165[degrees]C, stick together and therefore pelletizing with air cooling would be not possible at such temperatures.
To demonstrate an opportunity to produce micropellets by extrusion at reduced temperatures and pelletizing in open air we used as a die set a pack of 7 tubes with thin walls, i.e. injection needles soldered by silver alloy inside a metal housing of 20 mm length. Two types of needles were used: with outer diameter 0.45 and 0.8 mm. The rotating knife of a meat grinder was used to cut strands at the hot face of the die plate to thin disks or cylinders by varying the rotation speed of the knife. Otherwise, we used a rotating blade of a wood planner to cut strands of LLDPE to pellets. In this case, extrusion was going at 132[degrees]C with a die of 2 or 4 mm diameter at averaged melt velocity from 10 to about 40 mm/s. Two rotating rolls were used to draw a strand of LLDPE and to supply it to the rotating blade. No cooling but air convection was used to solidify molten PE before it comes to the rollers that are located about half meter downward from the die. Micropellets were produced as disks of 0.8 to 1 mm diameter and 0.5 mm thickness. Such micropellets demonstrate good flow properties and obviously can be used for rotomolding or flood feeding of an extruder.
PEG-based PPA Instead of Fluorinated PPA
Industry is already using blends of fluorinated polymers with PEG as PPA. However, we demonstrated that fluorinated polymers are not a necessary component of PPA. Viscous PPA made by blending fluorinated polymers and PEG with MW about 8000 Da stay trapped inside the polymer matrix, and only a small fraction of the fluorinated PPA is used to form a lubricating layer at the die surface. In contrast, PPA made from PEG actively migrates to the outside of the melt and forms a lubricating film. It can be blended with PO resins without master batching as pellets, by spraying or flow of molten PPA into the extruder. Such PPA can be used in concentrations from 0.025 to 0.5 wt% to suppress sharkskin in extrusion of LLDPE. In some cases a composition of novel PPA that is liquid at room temperature can be recommended to be used at low concentrations, i.e. from 0.025 to 0.1 wt%, to suppress sharkskin at very moderate pressure reductions. Such compositions of PPA can be prepared from PEG with MW from 300 to 600 Da.
Mechanisms of Lubrication
A tentative mechanism of slip is discussed shortly in introduction. Commercially available LLDPE grades comprise low MW organo-phosphites as antioxidant additives. The antioxidants deplete Oxygen from the PE melt and get converted to organo-phosphates, i.e. low MW esters of alcohols and phosphoric acid. Organo-phosphates are well known for their high affinity to metal surfaces and for their ability to saturate surface layers of steel alloys with Phosphorus oxide (phosphating of steel). If PEG is used as an additive to LLDPE resins, it reacts with these esters at the surface of the die, and high MW esters appear in the reaction of transesterification. Reaction rates depend on the presence of catalysts. We used in our experiments a nitrided die that is saturated by Phosphorus and Oxygen in its surface layer. Therefore we observed relatively short conditioning times, whereas with dies having no active centers for hydrogen and covalent bonding of the esters, e.g. a not-oxidized Chromium- or a Platinum-coated die, the conditioning times would be too long for practical use. The esters of PEG and phosphoric acid accumulate at the die inside and form a layer of a plastic lubricant, i.e. grease. The grease is a blend of PEG and high MW organo-phosphates that are thickening agents. Due to high affinity to steel the organo-phosphates decelerate flow of the grease along the die inside and, therefore, a thicker lubricating film can be formed with less concentrations of the lubricant. Passive, i.e. nonreacting, thickening agents are used in industry as a component of greases, and the use of them in the composition of PPA with low viscous PEG would further improve lubrication and shorten conditioning times "to clear sharkskin melt fracture."
The use of simplified compositions of PPA that include PEG and organo-phosphites has the obvious disadvantage of long conditioning times. At least two chemical reactions are involved in the formation of a lubricating layer with the use of PEG and organo-phosphites: conversion of the organo-phosphites to organo-phosphates, and a reaction of transesterification. The conditioning time can be considerably reduced if organo-phosphates are added to the PE resin together with PEG. In our experiments, the PPA that comprises organo-phosphates made from sorbitol shows a short conditioning time.
Some organic acids readily react with PEG, and the conditioning time would be short if the PPA is a reacting mixture of PEG and such organic acid or anhydride of organic acid. By use of citric acid as an ingredient of PPA, we observed not only short conditioning times but also better lubrication. We may speculate that citric acid reacts at the die surface with PEG and forms a gel-like substance that resists to flow, and therefore accumulates at the die surface as a lubricating film of larger thickness and of better lubricating performances.
Esters of PEG and citric or phosphoric acid are of amber color and if the additives are used in high concentration, i.e. 0.5 wt%, it is possible to see them at the die exit as beads of a dark viscous fluid. It wets metal surface and flows under forces of surface tension so that it can coat a considerable part of the die. In time, PEG decomposes and evaporates from the die surface but deposits of Phosphorus oxide remain. Such deposits at the die exit are reduced or avoided at smaller concentrations of the additives.
The proposed PPA is a development of a "core-annual flow" idea that was originally used to reduce friction losses in transportation of viscous bitumen and crude oils in long pipelines. We believe that some reacting mixtures of PEG with acids and comprising thickening agents can be used in low amounts, i.e. 0.1-1%, instead of water also to improve transportation of crude oils and bitumen during a cold season in areas of northern climates.
Processing Equipment for the Use with PEG-Based PPA
The use of novel PPA is a challenging opportunity to simplify processing of PO resins. The use of low viscous PEG in the composition of PPA with existing machinery has to be combined with changes in the die design for film blowing, blow molding, and pipe extrusion to extract and distribute evenly low viscous PPA: elongated dies, circumferential grooves in the mandrel and housing near the die exit, as well as tapered ridges of the die and an elongated core. Kurtz in several patents (88-90) described various designs of the extrusion die to delay the onset of sharkskin instability that would also be beneficial for the use with our PPA. There is no explanation for the geometrical solution of the sharkskin problem in the patents of Kurtz. We may speculate that accumulation of lubricants like Zinc stearate in the grooves and tapered areas of the die can contribute to the delay of sharkskin melt fracture. In a film blowing experiment by use of a die with elongated core we observed accumulation of low viscous PPA between the PE film and the core. Obviously, the pool of low viscous liquid between the core and the PE film ensures homogeneous lubrication and take-up of the PPA by the PE film. Otherwise, because of variations in the PPA content and locale variation of friction losses within a ring and slot dies, we may observe sporadic zones of locale spurt of the extrudate and variations of the film thickness.
In processing of pelletized PO resins that comprise synergistic blends of fluorinated polymers with PEG as PPA as well as our PEG-based PPA, conveying of pellets in smooth bore extruders in conditions of flood-feeding can get unstable or even terminate when a lubricant is fed to the extruder. This phenomenon is known in industry as "screw-slip," while it is actually a barrel slip. To ensure stable conveying of pellets, a single-screw extruder with grooves in its feeding zone is recommended as well as starve-feeding (91) of the extruder, i.e. the extruder has to be fed at a rate less than the capacity of the screw.
Novel Opportunities for Polymer Processing with PEG-based PPA
Extrusion at reduced temperatures potentially simplifies a design of processing machines for pelletizing, film blowing, and film casting as well as for tube and pipe production. It helps to suppress bubble and helical instabilities in film blowing, as well as draw resonance in film casting and fiber spinning, see (92-94). For extrusion at reduced temperatures with the use of novel PPA less antioxidants in the polymer can be used. Indeed, most efficient antioxidants are aromatic and mixed alkyl-aril phosphites. When these substances react at high temperatures with oxygen they form organo-phosphates that are toxic. PEG itself readily reacts with oxygen and therefore it is an antioxidant additive itself. So, the use of novel PPA and reduced amounts of the organo-phosphites would not only reduce material cost for polymer processing but also make our environment cleaner. It has to be clearly understood that composition of PPA has to be optimized for the polymer grade and technology of processing. For example, for film blowing of LLDPE the optimal composition of PPA is far from the composition that is optimal for high lubrication.
We may speculate about an advantage to produce pellets and micropellets by extrusion of PO resins at reduced temperatures in conditions of air cooling. Some amount of PEG will stay on the pellets and would be useful for further processing of the polymers. Micropellets of oval shape with sizes from 0.3 to 0.8 mm provide superior handling and processing performances in rotomolding (95). Surprisingly, we observed shorter times to sinter LLDPE particulates with additives of a composition that is identical to the proposed PPA for extrusion of LLDPE resins. Details of these experiments are published in (96).
We propose to use low viscous additives as lubricants in processing of polyolefin (PO) resins with narrow molecular weight distribution (MWD) and low melt index (MI) by extrusion at reduced temperatures, and we present first experimental results illustrating the improved performance of the novel polymer processing additives (PPA). The PPA proposed is a reacting mixture of polyethylene glycol (PEG) of MW from 300 to 10 000 Da with reactants from a following list: citric and phosphoric acids, phtalic anhydride, polyesters of oxiacids of Phosphorus. Silica fume and polyethylene oxide of high MW are used in composition of PPA as thickening agents. The PPA can be blended with pellets or powder of polymer without master-batching. Such PEG-based PPA would be a cheap alternative to existing blends of fluorinated polymers and PEG that are used as PPA nowadays. Surprisingly, we discovered that at reduced temperatures friction losses in extrusion with our PPA can be less than at elevated temperatures.
Further experimental and theoretical research is necessary to optimize compositions of PPA for particular processing technologies and grades of PO resins, as well as to get better insight into the physical and chemical phenomena underlying the observed improvements in extrusion. Experiments were made with LLDPE and HDPE, but we believe that the PPA proposed can improve processing of many other polymers with narrow MWD and low MI. Some compositions of the proposed PPA, e.g. comprising PEG and citric acid, would easily comply with pharmaceutical and foodstuff contact regulations.
The authors thank Dr. Kalman Migler for presenting a copy of his recent book on Instabilities in Polymer Processing and Prof. Hatzikiriakos for sending his publication (97) related to their findings. They also thank Dr. Chris Rauwendaal for discussion and his valuable comments, ExxonMobil Chemicals for supplying polyethylene resins, Clariant for samples of organo-phosphites, as well as 3M and Arkema for samples of fluorinated PPAs.
(1.) Z. Tadmor and C.G. Gogos, Principles of Polymer Processing, 2nd ed., Wiley, New York (2006).
(2.) B. Avitzur, Metal Forming. Encyclopedia of Physical Science & Technology, Academic Press, San Diego (1987).
(3.) R.E. Christensen, S. P. E. J., 18, 751 (1962).
(4.) J.C. Miller, S. P. E. Trans., 3, 134 (1963).
(5.) S. Kase and T. Matsuo, J. Polym. Sci Part A., 3, 2541 (1965).
(6.) M.M. Denn, Ann. Rev. Fluid. Mech., 12, 365 (1980).
(7.) T. Kanai and J.L. White, Polym. Eng. Sci., 24, 1185 (1984).
(8.) W. Minoshima and J.L. White, J. Non-Newt. Fluid Mech., 19,275 (1986).
(9.) J.L. White and H. Yamane, Pure Appl. Chem., 59, 193 (1987).
(10.) H.F. Mark and J.I. Kroschwitz, Ethylene Polymers. Encyclopedia of Polymer Science and Technology, 2, Wiley, New York (2003).
(11.) O.L. Kulikov, K. Hornung, and M. Wagner, in Advances n Polymer Processing, Ch. 15, S. Thomas and W. Yang, Eds., Wood Head Publ. (2009).
(12.) J.-F. Agassant, D. Arda, C. Combeaud, A. Merten, H. Munstedt, M. Mackley, R. Laurent, and V. Bruno, Intern. Polym. Proc, 21, 239 (2006).
(13.) H.W. Jung and J.C. Hyun, in Polymer Processing Instabilities: Control and Understanding, Ch. 7, S.G. Hatzikiriakos and K.B. Migler, Eds., Marcel Dekker, New York (2005).
(14.) CD. Han and J.Y. Park, J. Appl. Poly. Sci, 19, 3291 (1975).
(15.) CD. Han and R. Shetty, IEC Fundam., 16, 49 (1977).
(16.) L.B. Hutchinson and R.R. Blanchard, U.S. Patent 4,080,138 (1978).
(17.) F.N. Cogswell, U.S. Patent 3,920,782 (1975).
(18.) R.T. Fields and C.F.W. Wolf, U.S. Patent 2,991,508 (1961).
(19.) J.T. Lutz Jr. and R.F. Grossman, Polymer Modifiers and Additives, Marcel Dekker, New York (2001).
(20.) H. Zweifel, Plastic Additives Handbook. 5th ed., Hanser, Munich (2001).
(21.) E.W. Flick, Plastics Additives: an Industrial Guide, 3rd ed., Plastics Design Library, New York (2001).
(22.) E.C. Achilleos, G. Georgiou, and S.G. Hatzikiriakos, J. Vinyl Addit. Technol., 8, 7 (2002).
(23.) S.B. Kharchenko, P.M. McGuiggan, and K.B. Migler. J. Rheol, 47, 1523 (2003).
(24.) S.B. Kharchenko, K.B. Migler, and S.G. Hatzikiriakos, Polymer Processing Instabilities: Control and Understanding, Ch. 8, S.G. Hatzikiriakos and K. Migler, Eds., Marcel Dekker, New York (2005).
(25.) J.C. Slattery and A.J. Giacomin, U.S. Patent 5.637,268 (1997).
(26.) D.E. Priester, U.S. Patent 5,587,429 (1996).
(27.) R. Chiu, J.W. Taylor, D.L. Cooke, S.K. Goyal, and R.E. Oswin, U.S. Patent 5,550,193 (1996).
(28.) T.J. Blong, M.P. Greuel, and C Lavallee, U.S. Patent 5,830,947 (1998).
(29.) K. Focquel, G. Dewitte, and S.E. Amos, U.S. Patent 6,294,604 (2001).
(30.) S.S. Woods, U.S. Patent 6,734,252 (2004).
(31.) J. Briers, J.J. Cernohous, and R.R. Nuyttens, U.S. Patent 20,050,101,722 Al (2005).
(32.) G.R. Chapman Jr. and S.R. Oriani, U.S. Patent 6,894,118 (2005).
(33.) S.R. Oriani, S. D'Uva, and V.P. Trilokekar, U.S. Patent 6,906,137 (2005).
(34.) B. Barriere, A. Bonnet, J. Laffargue. and G. Marot, U.S. Patent 20,060,025,523 Al (2006).
(35.) G.R. Chapman Jr. and S.R. Oriani, U.S. Patent 20,060, 116,477 Al (2006).
(36.) D.W. Aubrey, G.N. Welding, and T. Wong. J. Appl. Polym. Sci, 13, 2193 (1969).
(37.) B.Z. Newby, M.K. Chaudhury. and H.R. Brown, Science, 269, 1407 (1995).
(38.) B.Z. Newby and M.K. Chaudhury, Langmuir, 13, 1805 (1997).
(39.) B.Z. Newby and M.K. Chaudhury, Langmuir, 14. 4865 (1998).
(40.) N. Amouroux, J. Petit, and L. Leger, Langmuir, 17, 6510 (2001).
(41.) D.A. Hill, T. Hasegawa, and M.M. Denn, J. Rheol., 34, 891 (1990).
(42.) F.N. Cogswell, J. Non-Newtonian Fluid Mech., 2, 37, (1977).
(43.) O. Kulikov and K. Hornung, J. Non-Newtonian Fluid Mech., 98, 107 (2001).
(44.) S.G. Hatzikiriakos, Polymer Processing Instabilities: Control and Understanding, Ch. 9. S.G. Hatzikiriakos and K. Migler, Eds., Marcel Dekker, New York (2005).
(45.) S.G. Hatzikiriakos, N. Rathod, and E.B. Muliawan, Polym. Eng. .Sci., 45, 1098 (2005).
(46.) O. Kulikov and K. Hornung, J. Non-Newtonian Fluid Mech., 124, 103 (2004).
(47.) O. Kulikov, K. Hornung, and M. Wagner, Rheol. Acta, 46, 741 (2007).
(48.) O. Kulikov, J. Vinyl Addit. Techn., 11, 127 (2005).
(49.) A. Schallamach, Wear, 17, 301 (1971).
(50.) O. Kulikov and K. Hornung, J. Non-Newtonian Fluid Mech., 107, 133 (2002).
(51.) G. Lykotrafiti and A.J. Rosakis, Int. J. Fracture, 140, 213 (2006).
(52.) E. Gerde and M. Marder, Nature, 413. 285 (2001).
(53.) D.D. Joseph and Y.Y. Renardy, Fundamentals of Two-Fluid Dynamics. Springer, New York (1993).
(54.) D.E. Broussard, P.R. Scott, and V.R. Kruka, U.S. Patent 3,977,469 (1976).
(55.) D.D. Joseph, J. Non-Newtonian Fluid Mech., 70, 187 (1997).
(56.) J.L. Lummus, U.S. Patent 3,434,485 (1969).
(57.) LP. Herman, Ed., Physics of the Human Body, Springer, Berlin (2007).
(58.) L. Sokoloff, Ed., The Joints and Synovial Fluid. I. Academic Press, New York (1978).
(59.) N. Pieterse, Development of a Dynamic Hip Joint Simulation Model, M.Sc. Thesis. University of Pretoria, Pretoria (2006).
(60.) R.E. Booser, Ed., Tribology Data Handbook, CRC Press, Boca Raton, Florida (1997).
(61.) L.R. Rudnick, Synthetics, Mineral Oils, and Bio-based Lubricants. CRC Press, Boca Raton, Florida (2006).
(62.) N. Gaylord, Polyalkylene Oxides and Other Polyethers, Interscience, New York (1963).
(63.) L.E. Mirci, S. Boran, V. Pode, and D. Resiga, J. Synth. Lubr., 2491, 51 (2007).
(64.) J.V. DeJuneas, G.L. McIntyre, and J.F. O'Horo Jr., U.S. Patent 4,013,622 (1977).
(65.) L.E. Wolinski. U.S. Patent 3,222,314 (1965).
(66.) T. Tikuisis, G. Arnould, J. Bayley, P.S. Chisholm, and S. Marshall,
U.S. Patent 20,050,070,644 Al (2005).
(67.) T.J. Blong and C. Lavallee, U.S. Patent 5,527,858 (1996).
(68.) D.J. Duchesne and V. Bryce, U.S. Patent 4,855,360 (1989).
(69.) G.R. Chapman Jr. and S.R. Oriani, U.S. Patent 7,001,951 (2006).
(70.) J.B. Williams and K.S. Geick, U.S. Patent 20.020,063,359 Al (2002).
(71.) P. Bauer, U. Seeliger. and U. Faller, U.S. Patent 6,048,937 (2000).
(72.) S. Sato, U.S. Patent 20,040,083,925 Al (2004).
(73.) S. Sato, U.S. Patent Appl. 20,040,132,878 Al (2004).
(74.) O. Kulikov, R.U. Pat. 2,288.095 CI (2006).
(75.) M.A. Corwin and G.N. Foster. U.S. Patent 4,540,538 (1985).
(76.) M. Xie, X. Liu. and H. Li, .J. Appl. Polym. Sci., 100. 1282 (2006).
(77.) J. Chen, X. Liu, and H. Li, .J. Appl. Polym. Sci., 103, 1927 (2007).
(78.) X. Liu and H. Li, Polym. Eng. Sci.. 45, 898 (2005).
(79.) O. Kulikov, R.U. Patent Appl. 20,08,109,571/04(010359) (2008).
(80.) D.R. Stevenson, T.C. Jennings, M.E. Harr, and M.R. Jakupca, U.S. Patent 7,320,764 (2008).
(81.) T.W. Coslett, U.S. Patent 870,937 (1907).
(82.) G.W. Becker and D. Braun, Kunststoff-Handbuch, 3(1), 15, Hanser Verlag, Munich (1992).
(83.) 1. Noda, W.M. Allen, J.T. Knapmeyer, and M.M. Satkowski, U.S. Patent 20,080,200,591 (2008).
(84.) E. Gruenschloss, Intern. Polym. Processing, 3, 226 (2003).
(85.) O. Kulikov, K. Hornung, and M.H. Wagner, PPS-24 Abstracts, Salerno, Italy (2008).
(86.) O. Kulikov, K. Hornung, and M.H. Wagner, 2nd Int. Conf on Advanced Tribology: Abstracts, Singapore (2008).
(87.) Y.W. Chen and M.R. Mackley, Soft Matter, 2, 304 (2006).
(88.) S.J. Kurtz, U.S. Patent 4,267,146 (1981).
(89.) S.J. Kurtz, U.S. Patent 4,348,349 (1982).
(90.) S.J. Kurtz, U.S. Patent 4,360,494 (1982).
(91.) C. Rauwendaal, Polymer Extrusion, 4th ed., Hanser Verlag, Munich (2001).
(92.) M. Wishman and G.E. Hagler, in Handbook of Fiber Chemistry, 2nd ed., Ch. 3, M. Lewin and E.M. Pearce, Eds., Marcel Dekker, New York (1998).
(93.) R.J. Fisher and M.M. Denn, AIChE .J., 22. 236 (1976).
(94.) B.M. Devereux and M.M. Denn, Ind. Eng. Chem. Res., 33, 2384 (1994).
(95.) R.J. Crawford and M.P. Kearns, Practical Guide to Rotational Moulding, Rapra Technology Shawbury, UK (2003).
(96.) O. Kulikov, K. Hornung, and M.H. Wagner, Int. Pol. Proc, 5, 452 (2009).
(97.) S.G. Hatzikiriakos and J.M. Dealy, Int. Pol. Proc, 8, 36 (1993).
Oleg Kulikov, (1) Klaus Hornung, (1) Manfred Wagner (2)
(1) University of the Federal Armed Forces Munich, LRT-7 and Polymerphysik, 85577 Neubiberg, Germany
(2) Technical University Berlin, LRT-7 and Polymerphysik, Fasanenstr. 90, D-10623 Berlin, Germany
Correspondence to: Oleg Kulikov: e-mail: firstname.lastname@example.org
Contract grant sponsor: German Science Foundation (Deutsche Forschungsgemeinschaft).
Published online in Wiley InterScience (www.interscience.wiley.com).
[C]2010 Society of Plastics Engineers
|Printer friendly Cite/link Email Feedback|
|Author:||Kulikov, Oleg; Hornung, Klaus; Wagner, Manfred|
|Publication:||Polymer Engineering and Science|
|Date:||Jun 1, 2010|
|Previous Article:||Modeling and prediction of morphology and crystallinity for cylindrical-shaped crystals during Polymer Processing.|
|Next Article:||Characterization of highly filled wood flour-polyvinyl chloride composites: dynamic mechanical and dielectric analysis.|