Reduction of Visco ity o t Plastics by Entanalemen M o Enhance Process J.P. Ibar New School Polymer Physics 06/29/20 Thermoplastics Generally processed by heatingand pressurizing the material Heating softens the material. Why must the material be softened?^^^^^^^^^^^^H To Reduce Viscosity When the viscosity is reduced to liquify the nto the hen the material is forc Shape using many different proc Plastic Processing Method • 14 Most Common Ways to Process Plastics -Blow Molding (Air Pressure Expansion -bottles) -Calendering (Roller sheet formation) -Casting (Poured and cured without pressure in a mold) -Compression Molding (Heat and Pressure snue -Cold Forming (Pressure only, no heat) -Extrusion (Heated via screw, forced through die or nozzle) • From Latin ex + trudere = out + push t Winding (Winding filaments shape -M Injection Molding (Heated under pressure, melt is injected with force into sprues, runners, gates, cavities) Laminating (High pressure molding into sheet, rod, tube) atched-die ure with two molds and flash tional Molding (Liquid plastic rotated into shape) Thermoforming (Deep draw sheets) ransfer Molding (Hop nto sprues and molds) Wet Lay Up (Various layers contoured in mold and oven cured) pushe CM What do ALL Plastics End-Users Want in all their processing? Lower operating costs overall gy us< Lower capital equipment costs Less customization of resins: more generic use Use of resins in cross-functional amplications (ie: use PP in PET applications) What is the process condition that can What if an extrus like an injec hat if th er temperatu ^^orTgrade resin (MFI=1) could flow ion molding grade (MFI=12)? in could be processed at substantially ^Hres (50-100 oC)? •^Wha^^eXtrusion could be done with substantially ea fects? essures (5 to 10 times less)? What if equipment throughput could substantially, with no or less sharks What is the key to make this happen? incr kin de Li DEDUCED VISCOSITY Viscosity Viscosity is defined as the Gly onds ics = 1000 to 1 million Pa-s Water, glycerin, honey are Newtoni scosity does not f Shear Stress to Shear rate te y Molten Plastics are non-Newtonian Fluids (see next sli Viscosity decreases with higher temperature (this was the onl technology available to work w Viscoelastic fluid, has strain energy stored within, recovers olds and cavities deformation creates swell problems entering m and melt fracture due to excessive shear stress Viscosity decreases substantially with increasing shear rate (but re-increases as soon as shear rate slows down, making it unpractical or inefficient in many cases). Shear thinning is GOOD, but does not allow to save the viscosity decrease for future use of the melt/resin. This is an " in situ" viscosity change 165C Q-a 4.4 -4.2¬4.0¬3.8¬3.6-< 3.4-3.2¬3.0¬2.8¬2.6 2.4 • • • 0 I 50 Data: Dataibis_M : cafreau_simple Chi^/DoF = 0.00002 RA2 = 0,09994 P1 4.74017 ±0.01368 P2 1 ±0 P3 2 80094 ±0 17319 P4 0.736SS ±0 00277 —I 1 1— 100 150 co rad/s 200 250 Viscosity is a function of (M/Mc) a=3.4 for M>Mc and a=1 for MTg behavior. CONCLUSION: SHEAR-THINNING MANIPULATION TECHNOLOGY (step 1 of "Smart Processing") •The use of vibration technology combined with pressure or shear has been used over 30 years But this technology is not concerned with preserving in pellets the viscosity reduction induced by shear-thinning, mainly because it is not possible to do so: above Tg , the reptation time is very short, thermal history vanishes. IMPORTANT QUESTIONS: Can we induce Thermal-Mechanical History in EXTRUSION? 1-Does it change the way the melt flows? 2-Does it influence the flow of pellets (on re-melting) made from "treated" melts. These are two separate questions Disentanglement Processor Fig. 1 Figure 4a Insertion of "Intelligent ribs" (i-Ribs) on both surfaces of the cone barrel FIG. 21 ROTATION and OSCILLATION play COMPLEMENTARY and SYNERGIC ROLES in producing beneficial Thermal-Mechanical history. -For instance they can shear the melt both in the cross-direction, or oscillation can be longitudinal while rotation is transversal. -Both rotation and oscillation can be used to increase shear rate. -Rotational Flow has only ONE variable: RPM -Oscillation has TWO parameters: frequency and strain amplitude: strain can produce strain softening, which, coupled with shear-thinning, is capable of modifying the stability of the entanglement network. What does OSCILLATION alone do to a polymer melt? JPI6 Melt Elasticity is increased by Frequency of oscillation. Slide 35 JPI6 the cross-over point, describing the inverse of the terminal time, is found at the intercept of these curves with horizontal line 0.5 (corresponding to G'=G"). J.P. Ibar, 11/21/2008 JPI7 • Q. 60000 -50000 -40000 -30000 -CO O O CO 20000 > 10000 -0-Newtonian Viscosity (before treatment) 57,000 ~r 0 38.000 Recovery Zone {after treatment) LLDPE 3,100 (processor's viscosity) Treatment Zone —I-20 40 -|-60 "T 1 1 80 100 Time (min) 120 Slide 37 JPI7 the first "jump" is very fast (does not show any transient behavior) and is due to shear thinning (from 57000 to 10000 Pa-s) and assumes that the initial viscosity corresponds to a stable melt. Starting at strain 20% (at 47 rad/s), viscosity becomes time dependent (transient). Each strain % would induce, at long time, a steady state viscosity value, which is strain dependent (decreases with strain). The final value is 3100 Pa-s, 3 times less than the shear-thinning value. This melt shows a strong transient tendency on recovery, i.e. when the conditions are reversed from w=47 down to 1 (and strain reduced from 25%% to 1) we observe not instantaneous response: the treatement has created a transient behavior response. In terms of EKNET theory, the instaneous behavior is essentially pure activation of the network strands, whereas transient behavior is associated with the orientation of this network strands (responding to stress), a way to dissipate the input energy into essentially enthapic modifications driven by entropic changes (compensation phenomena). J.P. Ibar, 11/21/2008 Effect of i-BIBS Ui scons Torque uith or without added VIBRATION (v) 12 Ap1P l: No Ribs 2: i-Ribs 0.0 500.0 1000.0 1500 2000 2500 Tine sec M2=60 RPM Ml =344 RPM Cpost*3) when UIB is added. 0.0 Polycarbonate T^ Z Z B 'C ADD UIBRATION . 1000 2800 Time sec 3000 4000 Tine sec PI, P2, P3: Pure EXTRUSION PRESSURE PSI P'2, P'3: EXTRUSION uith OSCILLATION+ROTATION 0.0 1000 2000 time sec Time (min) EXTRUSION AT LOWER PROCESSING TEMPERATURE PC extruded at 205 oC vs 285-315 oC normal extrusion PS extruded at 150 oC vs 220 oC (for this grade Mw=330,000) PET extruded at 135 oC vs 235 oC For Engage 8180 (Affinity 8150) LLDPE (MFI=0.5), we succeeded extruding the polymer at 40 °C!MM Normal extrusion at T= 200 oC CONCLUSION in-line "DISENTANGLEMENT" obtained with a STB works very efficiently. Viscosity is reduced by a very large factor, enhancing Processing in a spectacular way. Remaining questions: -is it really disentanglement that we have achieved? -can the disentangled melt be frozen-in into pellets that preserve the viscosity reduction ("in-pellet retention") Can the viscosity reduction of the melt be preserved in pellets? HOW and WHY? •The "HOW" opens up a new processing technology of polymer melts. •The "WHY" opens up a new discussion about the nature of ENTANGLEMENTS To achieve "in-pellet disentanglement", I used TWO STB connected through a gear pump Thermal Mechanical History is created by flowing through various PROCESS STATIONS, each combining pressure flow with cross-shear flow induced by ROTATION and OSCILLATION of Couette's inner cylinder. FEED Viscosity measurement pelletizer or RESULTS I have been able to produce pellets from "disentangled melts" which, upon reheating, have an increase of their MFI, or, show a very different shear-thinning behavior (much improved) at high strain rate. FLOW IMPROVEMENT FOR TREATED PELLETS Virgin MFI MFI after"Treatment" Flow improvement (and correction for Mw change) Polycarbonate 12 21 75% PETG 8 12 50% LLDPE 0.5 1.2 140% Polystyrene 4 6 50% PMMA 17 37 118% "disentanglement" treatments also modify the parameters a, n, and tF in addition to ho, in the Carreau's equation. MFI measurements only reflect variation of ho For Injection Molding applications, it is crucial to determine viscosity at a strain rate of about 1,200 sec1. Edit Operate lools Browse Window Help »• # • II TEMPERATURE PRESSURE 300 -mm 250 -• 200 I 150-B 100 -H 50-H MFI of pellets are compared to in-line viscosity measurements on-line Viscosity vs MFI of samples 45.0 1 40.0 -jj> 35.0 -I 30.0 -I 25.0 --20.0 -_ 15.0 -| 10.0¬5.0 -0.0 -0 500 1000 1500 On-line Rheometer VISCOSITY Pa-s In this case in-line viscosity drop is retained in the pellets upon re-melting. Thermodynamics-induced re-entanglement 45000 -. 40000 -35000 -30000 -25000 -20000 15000¬10000¬5000-Re-entanglement PS 525 Recovery Curves ....«.-• T=165°C fo=0,1 rad/s T=170°C T=175 C ©=1.88 rad/s —i 1 1 ! 1 1 1 1 1 1 1 1 1 1 1 1 1 1 -1000 0 1000 2000 3000 4000 5000 6000 7000 8000 Time sec THIS RESULT IS A BIG CHALLENGE TO EXISTING THEORY OF MELT RHEOLOGY: THE MELT VISCOSITY IS NOT STABLE THE EQUATIONS TO DESCRIBE THE RHEOLOGY (G\ G", VISCOSITY) SHOULD NOT JUST BE SUCCESSFUL ACCOUNTING FOR THE INFLUENCE OF STRAIN RATE, PRESSURE, MOLECULAR WEIGHT, TEMPERATURE AND STRAIN, BUT ALSO TIME ENTANGLEMENT NETWORK RE-ORGANIZATION (ORIENTATION) IS RESPONSIBLE FOR THE TRANSIENTS. THIS IS NOT GOVERNED BY THE TERMINAL RELAXATION TIME to. ENTANGLEMENT MANIPULATION CAN INFLUENCE to (this is when we see a change of MFI). "SUSTAINED ORIENTATION" CONDITIONS (ONLY) CAN RESULT in IN-PELLETS VISCOSITY REDUCTION. This is explained in "The Great Myths of Rheology, Part III. Entanglement Network Elasticity(J. Macromol. Sci. Phys, 2011). Under Investi •ation issues are been studied kinetics, temperature, stability of J melt). disentangle 1 n-pe fie^visCosity drop retention: -possible for all polymers? -is Me variable with gap layer? WUa£jmakes network orientation MDLtr CONCLUSION: • Shear-thinning effect (viscosity decrease) can be boosted by several folds by adding a LF vibration of controlled amplitude. This creates time dependent strain softening. • The benefits of such technology to extrusion are spectacular: lower T of processing, lower P, increase throughput, decrease or eliminate defects (shark skin). • The benefits for compounding are commercially exploited. Benefits of Disentanglemen norm 3 is resto w , dissolv oducts, new processes -No thin spots, less wa Red Reduce viscosity of melt wit Viscosity is lowered during p al when the resin finally process w viscosity, itives, composites: terial u rpage, ced energy to process Lower melt temperatures: -Lower pressures: factors of Higher throughputs: + -Improve quality of e nd product -i J -/J Reduced capital equipment costs, improved operating c m sier to process very difficult resins (Metallocenes) low better orientation and treatment opportunities d to more uniformity 50 deg C §10 to 20 sheet, film, fiber extrusion) HUGE BENEFITS FOR THE COMPOUNDING INDUSTRY •Higher concentration masterbatches •Better dispersed additives (for the same concentration) •40 % more throughput for extruding masterbatches (with nanoparticles) •Improved nanoclay exfoliation (full exfoliation has been observed). •Large improvement of conductivity with CNT additives. I propose to cover these results in detail in another lecture New Generation of STB Processors -The 1st generation of disentanglers (1999-2006) was expensive to build, required long and heavy pieces of hardware which became longer and heavier as throughput increased! -The current new generation of STB processors has a much simplified design and adapts to all throughputs. This was made possible by "rethinnking the field of entanglement stability. Thank you for your attention More information on my blog and web pages: www.NewSchoolPolymerPhysics.blogspot.com http://www.WIZIQ.com/NewSchoolPolymerPhysics977161 This lecture is also available with audio as an *avi file. Inquire at newschoolpolymerphysics@gmail.com for the downloading conditions.