By Jan H. Schut
When you grind plastic cryogenically into powder, it has microscopic sharp edges and angles, which lower bulk density and processability. Cryogenic grinding is also expensive and energy consuming. Theoretically if you could extrude micropellets five or ten times smaller than conventional micropellets, you would have round powder with higher bulk density and better processability. That’s what Tim Osswald, a professor at the polymer engineering center at the University of Wisconsin-Madison, thought when he got an idea for extruding powder.
The idea came from the way two polymers are compounded: one polymer forms round domains or droplets in a matrix of the other polymer. Osswald thought if he could extrude a thin thread of plastic and stretch it thinner with hot moving air, the air would act like the matrix and break the polymer thread into small round domains. The phenomenon is called Rayleigh disturbances from Lord Rayleigh, a British physicist, who developed a model for water drop formation in the late 1800s.
“The idea was that with several nozzles, you could produce high-volume micropellets or powders like fiber spinning,” Osswald explains. Powders are used for rotational molding, microinjection molding, compression molding, sintering of microporous parts and selective laser sintering (SLS) processes for rapid prototyping. In injection molding higher bulk density allows faster cycle times and filling of very narrow molds. In compression molding it fills molds more efficiently with lower press height. Powders also have potential for sintering mold-less production parts.
When Osswald described his idea for a nozzle that could spray micropellet powders to colleagues, they thought it was interesting, but wouldn’t work. When he applied to the National Science Foundation in 2009 for a grant to build a prototype, four reviewers also praised the concept as interesting, but thought it wouldn’t work and turned it down.
Osswald then used a grant that accompanies his K.K. and Cindy Wang professorship (K.K. Wang is the founder of C-Mold, a developer of CAE software that is now part of Autodesk) to build the prototype. It worked from the get go. “We made little tiny droplets between 20 and 500 microns,” Osswald recalls. “Depending on how you adjust it, you can get the micro droplets all the same size.” By comparison, conventional underwater micropelletizing dies produce micropellets down to about 500 microns in size.
The patent-applied-for technology was presented for the first time in two papers given at ANTEC in Boston last May by Osswald and Martin Launhardt, a graduate student who built the prototype: “Manufacturing of Micropellets Using Rayleigh Disturbances” and by William Aquite, a graduate student who simulated the process: “Simulation of Micropelletizing Mechanisms.” A CD of all ANTEC papers is available to non-attendees for $150 from the SPE (www.4spe.org).
Grinding isn’t the only way to make powder. Polyolefin reactors produce powders, mostly irregular in shape with low bulk density,so they are pelletized. But some HDPE powders are used directly from the reactor. Size depends on catalyst type. Chromium-based catalysts make coarse 800-micron HDPE powder, which is used a lot in wood-plastic-composite lumber and masterbatches. Ziegler-Natta catalysts make finer 150-micron HDPE powder, including bimodal powders used to impart crack resistance in pipe.
Powders are also made in secondary solution and precipitation processes. The Equistar division of LyondellBasell (www.lyondellbasell.com) has made 20-micron round Microthene powders for 30 years out of EVA, HDPE, LDPE and PP in a proprietary solution process using high-speed sheer stirring (the original patent from 1973, U.S. Pat. # 3746681, has expired). But these are specialty powders used as additives in inks, coatings, lubricants, and cosmetics, not for melt processing. (An excellent description of powder markets is found in an ANTEC paper from 2005 called “Higher Performance Polyethylene Powders” by James Krohn, Brandon Hughes, and others from Equistar.) Precipitation processes are also used commercially to make powder, but don’t work with every polymer, such as POM (polyoxymethylene) and PLA (polylactic acid).
The micropelletizing setup at the University of Wisconsin starts with a ¾-in. Brabender lab extruder. Melt flows from there through a half-inch-diameter melt tube in the feed block, makes a right angle turn and flows downward into a capillary tube 1 mm diameter and 22.5 mm long. This forces the melt to stretch and speed up.
When melt exits the capillary die, it’s pulled downward and stretched again “by a stream of hot air coming out of a coaxial annular orifice with an external diameter of 3 mm and inner diameter of 1 mm,” Osswald’s paper explains. (The inner diameter is the outside of the capillary tube.)
An external collar contains and controls the flow of hot air and powder into a “cooling device,” actually a commercial bag-less vacuum cleaner with chilled air at the intake. Chilled air at –20 C combines with hot air at 200 C to lower air temperature to 50 C. The vortex in the canister then cools and separates micropellet powder from the air stream.
Initial research was done with HDPE and LDPE. For HDPE the extruder was set at 150 C. Hot air was tested at temperatures from 130 C, close to HDPE’s crystallization temperature, up to 160 C, above HDPE’s melt temperature. Airflow was tested from 345 cu m/sec, where the polymer thread starts to break up, to 730 cu m/sec, corresponding to air speeds of 55 to 120 m/sec. Maximum air pressure in the nozzle was 0.6 MPa. Screw speed varied between 1 and 20 rpm for throughputs of between 55 and 350 g/hr.
Different polymers seem to make different powder shapes and sizes, but all are consistent and rounded. HDPE makes micropellet powders that range in size from 65 to 400 microns and are smooth and lemon-shaped under a microscope with tails at one or both ends caused by the Rayleigh disturbances. LDPE makes micropellet powders that are slightly larger (169 to 505 microns), rougher and bean-shaped, sometimes without tails.
Osswald has since received a $50,000 grant from the University of Wisconsin (Madison Draper Technology Innovations Fund) to scale the prototype up and characterize the process better. Material testing will try to figure out what influences the breakup into droplets and model optimum process conditions for a variety of polymers. Simulation is being done using OpenFOAM, an open fluid dynamics software program developed at the Inter University Centre of Dubrovnik, Croatia (www.iuc.hr).
The initial focus is on PLA. This year researchers successfully made PLA micropellets, which are somewhat elongated like little worms. Powdered PLA, made by grinding, is currently available only from one supplier, ICO Polymers in Houston, a division of A. Schulman (www.icopolymers.com). ICO sells PLA in several screened sizes and in both amorphous and crystalline form.
Osswald also wants to be able to separate die and air temperatures. In the prototype the capillary tube in the die is used to guide the hot air, so air and die temperatures are linked. Having independent air and die temperatures might allow finer tuning with some materials, Oswald expects.
OTHER NEW MICROPELLET TECHNOLOGIES
There is another patented micro-droplet technology in the literature (German patent PCT/EP 2004/000506) belonging to Zapf Creation AG, a doll and toy maker in Rodental, Germany. Zapf’s patent for a “Method and Device for Producing a PVC-Free Powder that is Essentially Made of Plastic” describes extruding a curtain of polymer downward from a slot die, then breaking the curtain up into droplets using a supersonic air stream from a De Laval nozzle. The work was done ten years ago by inventor, Gerhard Barich, a professor from Munich University of Applied Sciences in Germany, to produce non-PVC compounds for toys. The patent describes making spherical micropellets 20 to 500 micron in size out of styrene ethylene butylene styrene (SEBS) rubber at 270-290 C along with other materials. So far it hasn’t been commercialized.
RTP Company, a compounder in Winona, Minn. (www.rtpcompany.com), recently commercialized another interesting approach to micropellets. They aren’t as small as powder, but they also target higher bulk density competing against ground powders. RTP’s patent-applied-for technology for “controlled geometry composite micropellets” (U.S. Pat. # 20100291388) extrudes a variety of strands with different cross-sectional geometries including circles, triangles, rectangles, and diamonds and pelletizes them in lengths of 1.2 to 2 mm.
“When micropellets with a mixture of geometric shapes are poured into a mold, they stack with very little empty space, so you don’t have to overfill a compression mold,” says RTP market manager Eduardo Alvarez, one of the inventors. CGP (controlled geometry pellets) also reportedly wet out fiber better in compression molding than cryogenically ground powder. RTP commercialized CGP micropellets late last year primarily for high temperature resins. Strand pelletizing can be used for higher temperature polymers than underwater micropelletizing.
Conventional under-water micropelletizing dies are also making ever smaller micropellet sizes for rotomolding. Gala Industries Inc. in Eagle Rock, Va. (www.gala-industries.com) has made micropellets as small as 450 microns, where typically the smallest commercial micropellet size is 500 micron. Gala also invested in a full-scale rotomolding machine with multiple arms for comparison trials in its labs.