‘SSSP’ Technology Is Finally Commercial

By Jan H. Schut

Solid State Shear Pulverization, which chills a co-rotating twin-screw extruder below the softening temperature of plastic rather than allowing it to melt, is finally commercial after more than 20 years of R&D. Zzyzx Polymers LLC, Allentown, PA (www.zpolymers.com), started up in January 2015 reportedly using the first commercial SSSP technology, according to a paper originally submitted to the Society of Plastics Engineers ANTEC 2016 conference (www.4spe.org), but not actually given. The ANTEC paper on “Commercialization of Solid-State Shear Pulverization: a Novel Polymer Processing Technology,” by Philip Brunner (co-founder of Zzyzx), Mark Tapsak, and Mike Janse, says the process has been successfully scaled up to pilot plant size for the first time.

In the SSSP process the twin-screw barrel is chilled so that tri-lobe compounding screw elements crush the polymer in its rigid state, breaking molecular bonds instead of melting the polymer. Repeated crushing turns mixed, unwashed plastics like multilayer and metalized film waste, for example, into a pinkish powder, which can be remelted and have significantly higher properties than its component polymers.

SSSP research began in the former Soviet Union in the 1980s and was continued in the 1990s by Klementina Khait, and later John Torkelson, both professors at Northwestern University, Evanston, IL (www.northwestern.edu), and by professor Katsuyuki Wakabayashi at Bucknell University, Lewisburg, PA (www.bucknell.edu). Both university labs use 25-mm twin-screw extruders from Berstorff (www.kraussmaffeiberstorff.com) with barrels modified for cooling medium (anti-freeze) to roughly -7 °C at Northwestern and -12 °C at Bucknell. But high energy cost and a two-extruder process kept SSSP from being commercialized.

SSSP Figure 1

This schematic shows Solid State Shear Pulverization in a 70-mm twin-screw extruder barrel, which is first chilled to crush mixed plastic to powder, then heated to pelletize the powder. An SSSP-modified machine doing trials at Berstorff in 2003 shows ice on screws and barrels.

Brunner’s PhD dissertation and several subsequent patent applications (U.S. Pat. Applic. # 20150131399) under Torkelson were the basis for spinning the company from Northwestern in 2013 under the name NuGen Polymers LLC. Brunner’s dissertation on “Overcoming Sustainability and Energy Challenges in Polymer Science via Solid State Shear Pulverization” models energy and cost saving from combining SSSP twin-screw extrusion directly with pelletizing in one extruder. It reportedly would use a tenth as much energy as having two extruders: one to pulverize, the other to pelletize. On a lab scale SSSP used 1-10 kw·hrs of energy per kg of material, while the scaled up, in-line extruder uses 0.25-0.40 kw·hrs/kg, Brunner’s ANTEC paper says.

The ANTEC paper describes screw speeds of 200 rpm for a mixture of PE and HMWPE with throughputs over 180 kg/hr at an energy cost of 5-7 cents/kg on a 70-mm SSSP machine. The paper exaggerates that this cost is in line with conventional melt twin-screw extrusion technologies and the SSSP process is now as energy efficient as twin-screw melt compounding.. A 70-mm co-rotating twin-screw extruder for melt compounding would produce 300 kg/hr of masterbatch and 1500 kg/hr or more of engineering plastics. So even if energy cost is comparable, output of 180 kg/hour is low for the capital cost of the machine. The ANTEC paper also describes throughput of only 145 kg/hour for PCR laminated film and 100 kg/hour for waste film containing PVC.

SSSP Figure 2

Zzyzx Polymers’ full-scale SSSP extruder successfully recycled samples of (A) PE/HMWPE at 180 kg/hr for energy cost of 0.26 kw·hr/kg; (B) postconsumer laminated bags at 145 kg/hr for 0.35 kw·hr/kg; and (C) mixed film waste with PVC at 100 kg/hr for 0.40 kw·hr/kg.

Recent patent applications from Zzyzx (U.S. # 20150065616 and especially # 20150131399) describe cooling channels in the extruder shafts and elements as well as in the barrel.  Patent application # 20150131399 says that “systems and methods for controlling the temperature of a solid-state screw extruder may include providing an extrusion screw that incorporates one or more screw shaft channels. The shaft channels may be configured to conduct a flow of a heat conducting medium along a length of the shaft.” If channels are drilled in the shafts, that would weaken them and hurt throughput. But the patent application notes that “shaft channels may be incorporated into an exterior surface” of the shaft, which wouldn’t weaken anything.

NuGen Polymers got $150,000 National Science Foundation grant in 2013, through a Small Business Innovation Research Phase I grant (Award # 1248336). In 2014, under the Zzyzx name, the company was approved for a $150,000 loan from Ben Franklin Technology Partners of Northeastern Pennsylvania, but didn’t accept it. Instead, in 2015, Zzyzx was funded by another larger $737,640 National Science Foundation grant over two years through a Small Business Innovation Research Phase II grant (Award #1434826), plus another $203,000 from the NSF in supplements. That $940,640 of NSF support allowed Zzyzx to set up a 70-mm diameter SSSP twin-screw extruder for the first time.

This pilot scale machine became “a nine thousand lb calorimeter,” Brunner’s ANTEC paper explains, allowing them to measure and control key processing parameters including coolant flow rate, coolant temperature, and heat removal for the first time. The big surprise was that the scale-up was more successful than expected, says Zzyzx CEO Michael Janse. Unexpected efficiency meant that Zzyzx didn’t need to operate cold extrusion in the same extruder with melt extrusion, which involved an extremely steep thermal gradient for a single machine and limited throughput, Janse explains. The patent application for the one-extruder process (U.S. Pat. Applic. # 20150051339) was also somewhat awkwardly shared by Northwestern and Bucknell.

Another $150,000 grant in 2016 came from the Closed Loop Foundation, part of the Closed Loop Fund (www.closedloopfund.com) set up by a group of large consumer products companies to support plastic recycling. “It will help us purchase the equipment needed to scale up our technology for testing in a plastic processing facility,” Janse said in a statement. The Closed Loop grant was primarily sponsored by S.C. Johnson & Son Inc., Racine, WI (www.scjohnson.com), which was looking for a way to support postconsumer recycling of film including its Ziploc bags from municipal collection. The Allentown Economic Development Corp., Allentown, PA (www.allentownedc.com) also incubates Zzyzx in its Bridgeworks Enterprise Center.

SSSP Figure 3

Zzyzx CTO Mark Tapsak (left) and inventor Philip Brunner (right), in its incubator site in 2014. Thanks to a $940,640 of grants in 2015-2016, Zzyzx now has a 70-mm pilot line there. Another $150,000 grant this year could support the first customer installation.

Zzyzx is definitely the first commercialization of SSSP technology, but it’s not the first time that SSSP has been scaled up. Material Sciences Corp., Canton, MI (www.materialsciencescorp.com), supported R&D on an SSSP-modified, 65-mm Berstorff twin-screw extruder in 2003, according to Material Sciences’ 10-K filing Feb. 28, 2003. Reportedly the work was successfully completed at Northwestern and available for license with a patent assigned to Material Sciences Corp., which was interested to apply the powdered plastics to metal coating, but the technology wasn’t used at the time.

Currently Zzyzx is running trials for customers and trying to determine how its technology can best be used for closed loop recycling. Since the twin-screw machine is modular, Zzyzx could either license the technology to customers with a difficult waste stream, or toll recycle difficult waste for customers.

In 2016, Brunner left the company. This leaves no one at Zzyzx with PhD level experience with SSSP, which may hamper development but for now it’s too soon to tell.

Posted in Uncategorized | Tagged , , , | Leave a comment

‘Exjection’ Molding Comes into Its Own

By Jan H. Schut

At least two commercial technologies combine extrusion with injection molding to fill molds with much lower pressure than conventional injection molding. One was recently introduced by Extrude to Fill LLC, Loveland, CO (www.extrudetofill.com), which melts plastic electrically without turning the screw (see blog September 13, 2016). The other was introduced nearly 10 years ago by IB Steiner, a plastics engineering firm in Spielberg, Austria (www.exjection.com), called Exjection, which uses a sliding mold cavity.

The two technologies are completely different. Extrude to Fill has an unusual extruder that works with conventional injection molds. Exjection has an unusual mold that works with conventional injection molding machines. The results, however, are very similar. Both technologies use less expensive, smaller barrels, lower clamp tonnage, and much less energy than equivalent conventional, high-pressure injection molding for substantially lower production cost. Exjection’s moving molds cost more than conventional injection molds, but overall production costs are 20-30% less, IB Steiner says.

But savings and efficiencies aren’t why first customers choose either one. Both were commercialized to mold parts that couldn’t be molded otherwise. IB Steiner’s Exjection (U.S. Pat. # 7910044) was originally developed for the aircraft industry to mold long thin parts for commercial planes with ribs, holes and other features out of PEI and PC to meet fire codes. But the compounds were too viscous for long flow paths, and molding with multiple nozzles would have left weak weld lines.


The solution was a long partially-enclosed mold (up to 3 meters) with all functional details like screw bosses and snap-fit hooks contained in a nearly enclosed cavity. The cavity is filled through an open slit in the top as it slides past a fixed injection nozzle in the short upper mold. The upper mold has a heated supply zone (about 130 mm), followed by the nozzle, then a cooled calibration zone (about 260 mm). The sliding cavity also splits in half length-wise for undercuts and part ejection.

“A sword-shaped bar under both the supply and calibration zones on the fixed nozzle side of the mold partially seals the otherwise open cavity to retain injection pressure. The supply zone corresponds to an extrusion die, and the calibration zone to a calibration unit,” explains Gottfried Steiner, CEO and head of engineering at IB Steiner. By the time the mold cavity moves out from under the calibration zone into open air, the exposed plastic is hard.


Exjection’s horizontal sliding cavity, with all the molded features on a part, slides under a short fixed upper mold with a heated supply zone, fixed nozzle, and cooled calibration zone. By the time the mold slides out into the open air, the surface of the plastic part is hard.

IB Steiner took three years to develop the concept with sister company Hybrid Composite Products GmbH, also in Spielberg (www.hcp0.com), which molded the first Exjection parts on an injection molding machine from Engel Austria GmbH, Schwertberg, Austria (www.engelglobal.com). Together with Engel they developed special software to control carriage speed to get the right pack pressure under the nozzle. Fill pressures range from 50 to 300 bar depending on melt viscosity and part thickness, but higher pressure can be used if part geometry and viscosity require it, IB Steiner says.

Engel introduced the technology at the K 2007 show in Germany running 1-meter long ABS strips with a Y-shaped cross section, 1.2 mm thin walls, and molded end caps on an Engel e-motion 200/55 Exjection machine. Arburg GmbH & Co. KG, Lossburg, Germany (www.arburg.com), showed Exjection next in 2008 during in-house Technology Days and at Fakuma on an Arburg A375V Exjection machine. The technology was written up at the time, but little has been reported since. Published patent applications, however, show impressive advances—notably two approaches to rotating mold blocks that can make either continuous long parts or high volume discrete parts, which Steiner described in an in-house presentation for customers last January.



IB Steiner now has nearly two dozen licensees, some in commercial production, most still in development for a variety of applications. The first licenses were for aircraft lighting and trim parts, reportedly commercial by 2009, according to an Engel press release in 2010 after IB Steiner won an Austrian government award for Exjection technology. Aircraft parts were done for Boeing Co., Chicago, IL (www.boeing.com), and Airbus Industrie, Toulouse, France (www.airbus.com), developed with Stuekerjuergen Aerospace Composites GmbH, Rietberg-Varensell, Germany (www.stuekerjuergen.com), which has one Exjection machine, and with Hybrid Composite, which has two in operation.


IB Steiner’s long sliding Exjection mold was commercialized to mold aircraft and lighting parts like these, which couldn’t be molded conventionally, but it’s now being developed for high volume applications for cost saving and better part quality. Top photo: Stuekerjuergen; Bottom photo: ENGEL.

Several European companies also commercialized long Exjection building products including Vossloh-Schwabe Lighting Solutions GmbH, Luedenscheid, Germany (www.vossloh-schwabe.com), a unit of Panasonic Lighting in Europe. Vossloh-Schwabe commercialized 1.5-meter LED lighting strips by 2014, simulating the sliding mold with an existing MoldFlow program for mold filling, which sequences multiple nozzles. Vossloh-Schwabe posted a video of Exjection molding on their website (www.youtube.com/watch?v=UV0w79IeGaA).

Besides giving licenses, IB Steiner also develops and patents technology jointly with customers like patent-applied-for (U.S. Pat. Applic. # 20150273747) “elongated hollow parts” for urinary catheters, under development for four years with Hollister Inc., Libertyville, IL. (www.hollister.com), a maker of medical supplies. The driver is the cost saving of replacing an injection molded connector and extruded thin-wall tube, which require assembly, with a one-piece molded catheter 40 cm long.

By 2008 IB Steiner had also gone beyond horizontal sliding molds with a new concept for “Endless Exjection,” which rotates cavities past one or two fixed injection nozzles. Steiner partnered with SaarGummi International GmbH, Bueschfeld, Germany (www.saargummi.com), a maker of automotive seals, to develop a continuous elastomeric seal carrier (U.S. Pat. 8900499), with cavities rotating on a chain like molds for corrugated pipe. Z-Werkzeugbau GmbH, Dornbirn, Austria (www.z-werkzeugbau.com), built the moving molds with 12 cavity segments. SaarGummi trials were done at Arburg’s customer center on a hydraulic two-component Arburg Allrounder 570S, molding the seal carrier at 7-8 m/min.

Arburg’s Selogica controls alternate the two injection units to create continuous flow into the “endless” part. “While the first injection unit is injecting melt, the second one is metering,” Steiner explains. “Then they switch over, and the second injection unit is injecting while the first is metering.” But SaarGummi’s elastomeric seal carrier program didn’t go ahead in the end because of a design change. SaarGummi declared bankruptcy in 2010 and was acquired by Chongqing Light Industry and Textile Holding Group in Chongquing, China in 2011, after which Exjection R&D at SaarGummi stopped.

To date Arburg and Engel have built more than 10 Exjection machines with IB Steiner since 2007, including two “Endless Exjection” machines. Steiner has three on-going development projects for Endless Exjection, all for mass production. So a decade after Exjection was introduced to mold long thin parts out of materials that couldn’t be molded otherwise, it’s under development for high-volume applications in medical, construction, and even packaging. Its advantages today are efficiency, cost saving, and better part quality for both individual and continuous parts.

Endless Exjection

IB Steiner developed “Endless” Exjection in 2008 for SaarGummi with 12 rotating cavities on an injection molding machine. The part wasn’t commercialized because of a design change, but Steiner now has three development projects with rotary Endless Exjection for mass production.

Posted in Uncategorized | Tagged , , , | Leave a comment

Unusual ‘Extrusion’ Molding Is Commercial

By Jan H. Schut

Low-shear, low pressure “extrusion” molding machines developed by Extrude to Fill LLC, Loveland, CO (www.extrudetofill.com), a 2014 start-up, have been commercially molding tiny acoustic parts since 2015. Extrude to Fill’s process melts plastic without motion using electric heater bands in a thin-walled extruder barrel using conduction, not compression or friction between screw and barrel. The screw turns only to fill molds.

Melt temperature is controlled with sensors in the barrel and torque on the screw. A PID control loop uses those measurements to determine viscosity each time the cavity is filled. “You tell the machine what you want the output to be in terms of mold pack pressure, and let it run its algorithm to assure accurate plastic temperature and viscosity to produce the desired part,” explains Extrude to Fill CTO and co-founder Richard Fitzpatrick.

“Cycle time is the same as for conventional high-pressure molding, dictated by part cooling,” says Fitzpatrick, who worked almost 20 years on the process. “Fill time will last a second or half a second, similar to traditional injection molding. But where traditional molding pressure starts high and drops down; we start at 500 pounds and build up. Either way 500-3000 pounds is the final pressure in the mold cavity. Because of this build-up to the final pack pressure, we require less clamp tonnage than traditional machines—usually ½ to 1 ½ tons of clamp pressure per square inch (of mold surface).”


Extrude to Fill

Extrude to Fill’s machines melt plastic with electric heater bands in a thin-walled barrel without motion, not with compression and friction between screw and barrel. They come in two sizes—with ¾-inch and 1-inch screws, which turn only to fill molds, not to melt.


Extrude to Fill’s first commercial customer, Aurisonics Inc., Nashville, TN (fender.com), a maker of high-tech audio equipment, wanted to mold tips for ear buds out of temperature-sensitive Class 6 medical grade thermoplastic elastomer instead of typical two-part silicone. Plunger-shaped tips are thick at the center, thin around the edge and couldn’t be molded out of TPE conventionally, Aurisonics’ president, Dale Lott says. The advantage of TPE tips is that they “deform in response to heat from the user’s ear,” says Aurisonics’ patent application (U.S. Pat. Applic. # 20160173971), so they’re more comfortable than silicone tips, which itch.

Aurisonics bought its first machine in 2015, molding 30 and 50 durometer TPE audio tips in up to four cavities and 13-second cycles. “The machine really enabled my business to grow. Our needs went from 100s of tips a month to 1000s a week,” Lott says. Aurisonics, which was acquired in 2015 by acoustic guitar maker Fender Musical Instruments Corp., added a second machine in 2016 for up to eight cavities and ordered two more machines this year to mold larger audio components.

Extrude to Fill offers machines in two sizes with either vertical or horizontal mold opening. Both sizes were originally designed for aluminum molds. The small machine is for 6 x 9 inch molds with up to 30-gram shots and a 3/4-inch screw 20 inches long (26.6 L/D). Aurisonics uses the small machine with vertical mold opening and a customized shuttle table and mold cavity. The operator inserts parts of the ear bud device before molding. After molding, the operator removes the core with parts, removes tiny undercuts by hand, and puts a clean core in. Aurisonics builds its own aluminum molds with inserts and undercuts.

Aurisonics Rockets

Extrude to Fill’s first commercial customer, Aurisonics, has two small, vertical Extrude to Fill machines now and two more on order, molding tiny high tech acoustic parts out of TPE, which couldn’t be molded conventionally.

A year ago Extrude to Fill built the first prototypes of the larger machine with both horizontal and vertical opening platens for aluminum molds to mold parts for another customer and do additional development. The large machine is for 15 x 15 inch molds with up to 200 gram shots and a 1-inch screw 28 inches long (28 L/D).

Potential customers for the large size, however, wanted to be able to use steel molds as well as aluminum, so Extrude to Fill redesigned the platens, integrating them into the machine frame to support heavier molds. The company has just built the first commercial large machine for horizontally opening steel molds, which will be delivered in November, and has four more on order for wire and cable and electronics to be delivered in the first quarter of 2017.



Extrude to Fill machines have four temperature zones: a chiller on the initial feed zone followed by three heat zones: preheat, melting, and a heater band on the nozzle before the mold. The 2-3-inch long feed zone is chilled to control the start of melting. Then in the 3-4-inch preheat transition zone, pellets soften and merge, but don’t melt, forming a soft seal that moves forward and keeps melted plastic from flowing backward when the screw stops. Depending on polymer, plastic can be held in the barrel at melt temperature with the screw stopped. PP and PE don’t degrade if held at melt temperature, where ABS, nylon and some other resins do over time, Extrude to Fill’s Fitzpatrick says.

Next comes a 9-inch melting zone where screw geometry changes to short flights that roll melted plastic to get uniform temperature. The screw can be made of brass alloys or coated tool steel and can be heated. It’s driven by a three-phase motor: 1 h.p. for the small machine, which plugs into a 110-volt wall socket; 3 h.p. for the larger one, which needs 208 volts single-phase. The last zone has a patent-pending nozzle which closes with a needle-type valve, preventing formation of a cold slug of material. When the carriage moves forward, the screw retracts, opens the valve, and fills the mold. Instead of an injection or metering zone determining shot size like conventional injection molding, Extrude to Fill machines pump plastic continuously and on demand. There are runners and gates, but less mold cooling is required than for conventional injection molding.

Extrude to Fill describes lots of advantages to this process. Because of low pressure and sheer, it doesn’t degrade plastics. Because it doesn’t compress plastic under high force, it uses 85-90% less energy than conventional molding. Because of low friction, it can mold materials that are hard or impossible to mold conventionally like 50% glass-filled nylon or 75% stainless-steel-powder filled metal injection molding feedstock. Temperature control is more precise. It also has more control of part weight and can mold 100% regrind or recycle with consistent part weight despite variations in melt flow index.

Extrude to Fill isn’t the only electrically heated injection molding process. Hot melt adhesive and lab injection molding machines also melt with electric heat, not friction. But hot melt potting machines are limited to low temperature polymers, and lab machines are limited to very tiny shots. Extrude to Fill’s machine isn’t limited in either way.

Conventional injection molding machines can also use low pressure and resemble extrusion, like “intrusion molding,” in which plastic is first continuously extruded at low pressure into a mold, then packed with a final injection shot. (See an ANTEC 2010 paper on “Improvement of Injection Molding Machine Capacity by Intrusion Technique” by Abbas Mokhtarzadeh and Avraham Benatar of SABIC Innovative Plastics.) There is even licensed “Exjection” technology (U.S. Pat. # 7910044) from IB Steiner, a plastics engineering firm in Spielberg, Austria (www.ibsteiner.com), introduced in 2008, which can be used with conventional injection molding machines. Exjection combines extrusion with injection to mold long or continuous parts. But these processes still melt with compression and friction.

So Extrude to Fill is genuinely different, and more developments are coming. Extrude to Fill’s machines have tiebars now, but will eventually be tiebarless. Air-actuated clamp and carriage drives will be replaced by servo motors. For larger parts and higher cavitation, multiple extrusion molding machines can be used together in a patent-pending multi-port array. Extrude to Fill hasn’t built one yet, but that will be fun to watch!

Posted in Uncategorized | Tagged , , , , | 1 Comment

New Shapes for Conformal Cooling

By Jan H. Schut

Conformal cooling channels in injection molds have been built by additive manufacturing, or 3-D printing, developmentally since the late 1990s using successive thin layers of metal. Unlike conventional straight drilled cooling lines, conformal cooling channels curve around deep shapes in a mold, equidistant from the mold surface, which is the source of heat. They can only be built by additive manufacturing and are designed with cooling line requirements secondary only to part and parting line requirements, unlike traditional drilled cooling lines which are located where space is available late in the design process.

Conformal cooling, however, was limited by the capabilities of the early metal powder welding process, which used metal powders coated with polymer binder, then evaporated out the binder leaving porous metal, and infiltrated the porous space with bronze. Since about 2006, direct additive manufacturing of pure metal powders became possible with the integration of fiber lasers into metal printing vs. the original CO2 lasers. 3D Systems Inc., Rock Hill, NC (www.3dsystems.com), also added a compaction step to make a dense metal powder bed before laser melting thinner layers—20 microns vs. about 100 microns before.

All that makes “100% metal mold inserts possible with reasonably smooth surfaces, providing a rock solid tool for mold makers,” explains Scott Young, engineering manager at Bastech Inc., Dayton, OH (www.bastech.com), a service bureau for additive manufacturing since 1994 and a reseller for 3D Systems. Bastech has built molds using additive metal manufacturing since 2000 and using pure metal additive manufacturing since 2015.



Conformal channels themselves, however, haven’t changed much in two decades of development. Primary cooling channels can transition to small capillary channels and back to primary trunk lines again to get cooling closer to the mold surface or to cool small mold details. Capillary channels can be closer together than large channels. (The cross section of the supply channel has to be equal to or greater than the sum of the cross sections of the capillaries, and the cross section of the return line has to be equal to or less than the sum of the cross sections of the capillaries.) But conformal cooling channels themselves are still typically round like drilled water lines.


Conformal Cooling Figure 1

State-of-the-art conformal cooling channels like these from 3D Systems and Renishaw show round channels like traditional drilled cooling lines. But round may not be the most efficient shape. Triangle, star and X-shaped channels all have more surface area and cool better. Left photo: 3D Systems; Right photo: Renishaw;


Graphics for advanced state-of-the-art conformal cooling from 3D Systems; EOS GmbH in Kraillingen, Germany (www.eos.info); and Renishaw PLC in Gloucestershire, U.K. (www.renishaw.com), three suppliers of metal-powder-based additive manufacturing systems, invariably show only round channels. But round may not be the best channel shape, according to a recent study by Bastech, presented at the AMUG 2016 conference (www.additivemanufacturingusersgroup.com) in April in St. Louis, MO. Bastech compared cooling efficiency of round, square, diamond, tear drop, triangular tear drop, and triangular channels, based on surface cooling area calculated from channel perimeter. So for channels with the same length and volume, the channel with the longest perimeter will have the most surface area and cool best.

Using 3D System’s Cimatron 13 software, Bastech compared a round channel with a 1.374 inch perimeter to the other shapes and found that a tear drop has a 1.454 inch perimeter; square has a 1.474 inch (rotating the square into a diamond shape has the same perimeter but is structurally stronger in a mold); triangular tear drop has a 1.574 inch; and triangular has a 1.587 inch. So a triangular channel cools 16% more efficiently than a round one. “More important, triangular channels have a long flat surface that can be oriented to parallel the mold surface,” Bastech’s Young points out.

Conformal Cooling Figure 2

Bastech simulated five cooling channel shapes—round, square, tear drop, triangular tear drop, and triangle–rotating some shapes for better mold strength and orientation to the mold surface. Triangular channels have 16% more surface area than round, which are the least efficient.


The most complex conformal channels Bastech has built so far were done this year for a mold core for an in-house promotional product, a 5-inch-tall cold drink sleeve. The 8-inch high mold core, also presented at AMUG 2016, is cooled by two inter-twined helical flow channels with X-shaped cross sections (X’s have even more surface area than triangles). The supply side channel splits to start one helix at the base of the core and the other helix and the top, creating counter flow, shorter channels, and more even cooling. The helical channels then reconnect into the return line to exit the core.


Bastech AMU page 16

Bastech built this highly unusual conformally cooled mold core with two X-shaped helical cooling channels. One runs straight to the top of the core and coils down. The other coils up, intertwined with the first for counter flow, shorter channel length, and better cooling.



Recent patent literature also mentions additive metal manufacturing of alternative cooling channel shapes. Star-shaped cooling channels are described by Siemens AG (U.S. Pat. # 8922072) and “triangular, rectangular, square, semi-circular, and ellipsoidal” channels by General Electric Co. (U.S. Pat. Applic. # 20140202163), both to cool heavy machinery, not injection molds. 3D Systems discussed the concept of star-shaped cooling channels at an electronics trade show in January this year, but without showing simulations.

Star-shaped channels, however, would be difficult to build by additive metal manufacturing “because unsupported structures shouldn’t overhang by more than 45 degrees,” notes Maximilian Boulter, manager for additive manufacturing at Renishaw’s LBC Engineering service bureau in Pliezhausen, Germany (www.renishaw.com), “though the angle is dependent on a few things like material, size, and angle toward the recoating wiper.” The X shape is also difficult to build. “The X pattern would have been doomed in a single helix, but a double helix gave us enough pitch to create the shape,” Bastech’s Young explains.

Cleaning and maintenance, however, are issues with non-round channels, notes Mads Jespersen, a partner at FlowHow ApS, Sydiylland, Denmark (www.flowhow.dk), a consultant on conformal cooling. Jespersen calculated how much steel was needed for a mold core not to collapse with wear over time and filled all the open space with cooling water. “That made some weird but very effective cooling channel shapes,” he says.

He then simulated the flow rate of cooling water using Moldex 3D software from CoreTech System Co. Ltd., Chupei City, Taiwan (www.moldex3d.com), which includes computational flow dynamics (CFD). CFD showed potential areas in the cooling channels with no flow. The risk with ‘no flow’ areas is that “these irregular channels, while effective, would also be sensitive to corrosion and deposits and impossible to clean without use of chemicals and the risk of damaging small features in the cooling channels,” Jespersen says. He concludes that “conventional round channels may be the best solution to minimize the cost of cleaning, spare parts, and break downs.”

Instead of increasing the surface area of the channel with alternative shapes that may be hard to clean later, cooling can also be improved by turbulent flow. FlowHow’s Jespersen simulated and built unusual “fish net” structures with additive manufacturing in combination with round channels to increase turbulent flow where more cooling is needed.

Like FlowHow, Bastech also used additive manufacturing to reduce the mass of metal in molds, but it wasn’t to increase cooling water volume. It was done because less metal in the mold means faster mold start up and shorter additive manufacturing time. “We only need 0.25 inches of tool steel for mold walls, plus an inch or so for the cooling channels. Everything outside of that we don’t need,” Bastech’s Young explains. For the drink sleeve core instead of a solid metal mold, Bastech built a structure of trusses and supports, leaving diamond shapes of metal out and removing roughly 25% of the metal. The drink sleeve core took only 38 hours to build vs. 42 hours for CNC programming and machining to build the same core out of solid tool steel with conventional spiral baffles.

An earlier important study on conformal cooling was done internally by Lego Group, Billund, Denmark (www.lego.com), in 2010. Lego used Moldex3D software and MSC Nastran finite element analysis software from MSC Software Corp., Santa Ana, CA (www.mscsoftware.com), to simulate and build three different cores for Lego blocks: one out of  standard tool steel, one with an Ampco bronze cooling insert, and one with conformal cooling. The study found that cycle time went down to 13.1 seconds for cores with the bronze insert and conformal cooling vs. 23.2 seconds for standard tool steel, but conformal cooling made better parts with less warpage than bronze inserts. Temperature differences between actual mold inserts and simulation were reportedly within 5%. Lego is now believed to have built more conformally cooled production molds than any other company in the world.

Conformal Cooling at LEGO

Conformal channels don’t have to be big. A 2010 study done at Lego using Moldex3D software and MSC Nastran FEA software, compared conformal cooling and bronze inserts to standard tool steel. Both inserts cut cycle time almost in half, but conformal cooling made better parts.


Posted in Uncategorized | Tagged , , , | Leave a comment

Trolling for New Technology: ANTEC 2016

By Jan H. Schut

New technologies at ANTEC 2016, the Society of Plastics Engineers Annual Technical Conference, held this year in Indianapolis, IN May 23-25 (www.4spe.org), include an array of unusual functional structures created in plastics and composites by means as different as chicken feathers, plasma, and micro fibrillation. New machine processes include two new multi-screw extruders from Chinese technical universities and a new chaotic melt mastication process from the University of Massachusetts in Amherst (www.polysci.umass.edu).

Guangdong Industry Polytechnic (www.gdqy.edu.cn) introduced its non-identical twin and triple screw extruder for the first time with two papers reporting better mixing, finer dispersions and less breakage of fibers. South China University of Technology (www.gdqy.edu.cn) presented a physically foamed and structured terpolymer with solid domains of PS in a matrix of biodegradable PPC with in situ PTFE micro fibers that increase viscosity over 1500%. The terpolymer is made on its triple-screw extruder with vibration on the middle screw, introduced at ANTEC 2013 (see blog May 10, 2013).

The title of each highlighted paper is followed by bracketed information on when and where it was given, i.e., M, T or W indicate Monday, Tuesday or Wednesday followed by the session title and time. People who didn’t attend ANTEC can purchase a link to download all papers from the SPE (www.4spe.org) for $250 to members and $300 to non-members. Plenary speeches, New Technology Forums, and undergraduate student posters aren’t on the link.



The Mixing of Flame Retardant Polymer Materials in a Novel Co-Rotating Non-Twin Screw Extruder [M/Extrusion-Reaction&Mixing I, 10:00] by Bai-Ping Xu, professor, Guangdong Industry Polytechnic, Guangzhou, China (www.gdqy.edu.cn), presents an unusual “self-made” co-rotating nonuniform twin screw extruder for the first time in the U.S. These “fraternal” twin screws can reportedly mix compounds more efficiently than conventional identical twins. A paper published last year in “Fibers and Polymers” (SpringerLink) also describes the non-twin screw extruder mixing immiscible polymers with finer dispersed phases and fibers with less breakage than identical twins. Patented non-twin-screw technology (U.S. Pat. # 9174380) can apply to twin- or triple-screw extruders.

U.S. Pat. # 9174380

Morphology of HDPE/PS Blends along the Axial Position in a Novel Co-Rotating Non-Twin Screw Extruder [T/Alloys&Blends Compatibilization, 3:30] by Bai-Ping Xu, professor, Guangdong Industry Polytechnic, Guangzhou, China (www.gdqy.edu.cn), describes using the novel non-twin screws in a clamshell barrel, mixing an 80/20 wt% blend of immiscible HDPE and PS. The material was processed at different screw speeds and observed with the barrel open, showing smaller droplet size and narrower size distribution at higher screw speeds.

Non-twin screw configuration

Melt Mastication: A New Rheological Process to Generate High Performance Parts from Semi-Crystalline Polymers [T/Engineering Properties&Structure, 8:30] by Alan Lesser, professor, University of Massachusetts, Amherst, describes using a modified, heated static mixer for “chaotic mixing” or mastication of a semi-crystalline polymer at a temperature between its melting point and melt crystallization temperature. This new process significantly increases crystallization and gives conventional isotactic PP the strength of glass-filled PP. Patented “Melt Mastication” (U.S. Pat. # 9284388) was first mentioned at the SPE Automotive Composites conference in 2013, but ANTEC is the first presentation on how it enhances polymer stiffness, strength and toughness.



Preparation of PPC/PS/PTFE Composites with In-situ Fibrillated PTFE Monofibrillar Network and Their Supercritical Carbon Dioxide Extrusion Foaming Properties [T/Thermoplastic Materials&Foams, 9:00] by Hao-Yang Mi, professor, South China University of Technology, Guangzhou, China (en.scut.edu.cn), is done using SCUT’s triple-screw compounding extruder with vibration in the middle screw (see blog, May 10, 2013). The triple-screw creates a multi-phase composite with PTFE nano fibers created in situ in a PPC/PS matrix. Rigid PS domains and PTFE nano fibers increase PPC’s viscosity by 1576%, allowing composite foam with four times higher cell density than PPC alone. SCUT’s triple screw extruder is built commercially by POTOP Experimental Analysis Instrument Co. also in Guangzhou (www.potop-lab.com).

Three-Dimensional Hierarchical Materials by Memory-Based, Sequential Wrinkling [T/New Technology Forum-Surface Engineering, 2:00] by Teri Odom, professor, Northwestern University, Evanston, IL (www.chemgroups.northwestern.edu/odom) creates super hydrophobic 3D surface patterns on PS sheets. Pre-strained, biaxially oriented PS sheet is plasma treated to deposit a soft fluorocarbon (CFx) polymer skin. Then the strain is relaxed to create surface wrinkles. This repeatable, tunable process applies robust wrinkle patterns that can withstand subsequent bending and stretching. Oxygen treatment traps oxygen in the wrinkles to create different levels of hydrophobicity for a range of potential applications.

ODOM/NORTHWESTERN: photos of wrinkle patterns

Fabrication of Hybrid Polymeric-Metallic Foams as Scaffolds for Bone Tissue Engineering [M/Thermoplastics Materials&Foams, 1:30] by Anil Mahapatro, assistant professor, Wichita State University in Kansas (www.wichita.edu), creates an unusual foam structure by electro-depositing magnesium onto PU foams. PU is a representative example of a polymeric foam used as a bone graft material. Magnesium gives mechanical strength for bone support while improving bone binding and regrowth.




High Temperature and High Energy Density Nanolayer Film Capacitors [T/Electrical&Electronics, 3:00] by Deepak Langhe, technology director, PolymerPlus LLC, Valley View, Ohio (www.polymerplus.net), shows the latest development from PolymerPlus, a spin-off from Case Western Reserves’ high multi-layer film lab (www.stc-clips.org) in 2010 (see blog July 6, 2012). Polymer Plus’s patented developmental nanolayer co-ex film capacitors (U.S. Pat. # 8611068) with up to 257 layers offer higher energy density, higher temperature performance, and lower energy loss than current metalized and laminated capacitors. PolymerPlus also makes optic lenses with over 800,000 nanolayers (U.S. Pat. # 7002754).

Production of in situ Microfibrillar Composites as a Novel Approach Towards Improved Bio-Based Polymeric Products [W/Bioplastics, 8:30] by Chul Park, professor, University of Toronto, Ontario, Canada (www.utoronto.ca), first extrudes strands of an immiscible blend of 97/3 PLA/PA6, then draws the strands to create 200 nm PA6 micro fibrils in the PLA matrix, improving melt strength, elasticity and foamability. Strands are then chopped and molded at below the melt temperature of PA6 to retain the micro fibers. The research was published last year in the American Chemical Society’s “Biomacromolecules,” but this is its first presentation to a plastics audience.


Capillary Coextrusion: a New Process for Creating Small-Scale Coextruded Films [M/Flexible Packaging, 9:30] by Patrick Lee, assistant professor, University of Vermont, Burlington (www.uvm.edu), describes a clever device to test layer adhesion in coex films more accurately than testing cast or laminated film layers. Lee attaches a small (10 mm) coex die to a dual-bore capillary rheometer to coextrude 8-9 mm test films continuously.

PCLEE/UVM: image003.png



Green Plastics: Utilizing Chicken Feather Keratin to Improve the Thermo-Mechanical Properties of PU Composites [M/Composites-Natural/Bio, 9:30] by Firoozeh Pourjavaheri, PhD candidate, RMIT University, Melbourne, Australia (www.rmit.edu.au), enhances thermo-mechanical properties of TPU composites by adding 10% and 20% purified keratin fibers from waste feathers, mixed with urethane using solvent casting and evaporation. The resulting composites show lower glass transition temperature, % strain recovery, and mass loss of the composite than neat TPU, but higher elastic modulus, storage modulus and heat resistance, indicated by the % char.

Soy Protein Isolate Films with Improved Mechanical Properties via Bio-Based Dialdehyde Carboxymethyl Cellulose Crosslinking [M/Bioplastics 8:30] by Ting Zheng, graduate student, Clemson University, Clemson, SC (www.clemson.edu), reports new glycerol-based soy protein films cross-linked with DCMC for an astonishing 218%  improvement in tensile strength. So called “edible films” for meat and poultry packaging have been a hot research topic on technical programs for decades without being commercialized because of poor strength, so a property enhancement of this magnitude could be significant.

Improvement of the Extrusion Foaming Properties of Externally Plasticized Cellulose Acetate by Reactive Melt Mixing Using a Multifunctional Reactive Oligomer [M/Extrusion-Forming I, 9:30] by Sven Hendriks, scientific research assistant, Institute of Plastics Processing at RWTH University, Aachen, Germany (www.ikv.rwth-aachen.de), uses a reactive oligomer as a chain extender to foam fine-celled cellulose acetate sheets or boards with 1,3,3,3-tetra-fluoropropene (HFO 1234ze) as physical blowing agent. The IKV has partnered with the Fraunhofer UMSICHT, Oberhausen, Germany (www.umsicht.fraunhofer.de); FKuR Kunststoff GmbH, Willich, Germany (www.fkur.de); Inde Plastik Betriebsgesellschaft mbH (www.indeplastik.de); and Jackon Insulation GmbH, Mechau, Germany (www.jackon-insulation.com), for several years foaming cellulose acetate, but this is the first mention of chain extenders.



High Precision and Repeatability in Micro Injection Molding Using the Inverse Screw [T/Injection Molding-Emerging Tech, 3:00] by Torben Fischer, chief engineer, Institute of Plastics Processing at RWTH University, Aachen, Germany (www.ikv-aachen.de), reports the first test results on the repeatability of this unusual threaded barrel/smooth shaft micro molding machine using different thermoplastics including POM, PC, PMMA, PP and 30% glass-fiber-reinforced PP. The inverse screw micro molding machine was introduced in Germany in 2013 and in the U.S. at ANTEC 2014 (see blog Nov. 11, 2014).

IKV: Geometry of the inverse screw

Development of an In-Line Plasma Treatment during the Injection Molding Process [W/Injection Molding Troubleshooting, 11:30] by Timo Nordmeyer, scientific employee, University of Paderborn in Germany (www.ktp.uni-paderborn.de), describes integrating a stationary atmospheric plasma jet inline into an injection molding process. The target is injection molding of technical parts of polymers with low surface energy like fiber-reinforced PP for automotive housings. The patent pending process was developed with Plasmatreat GmbH, Steinhagen, Germany (www.plasmatreat.com).

Cycle Time Reduction by Water Spray Cooling in Thermoforming [M/Extrusion-Forming I, 10:00] by Jonathan Martens, research assistant, Institute of Plastics Processing at RWTH University, Aachen, Germany (www.ikv.rwth-aachen.de), reports 5% faster cycle time vacuum-forming a 400-ml beaker in a female mold with atomized water mist sprayed onto the formed sheet on the open side away from the mold. Water spray cooling is used in industrial thermoforming for large parts with different stretch ratios, but not in packaging. The trick is to use just enough water to cool without leaving water residue on parts after unmolding.

Posted in Uncategorized | Tagged , , , , | Leave a comment

Competition Bubbles Up Again in Physical Foam Molding

By Jan H. Schut

Physical foam molding has been dominated for over 15 years by MuCell microcellular foam technology from Trexel Inc., Wilmington, MA (www.trexel.com), commercialized in 1999 and backed by a master part patent on “Injection Molding of Microcellular Material,” published in Europe in 1998 (EP # 952908) and the U.S. in 1999 (U.S. Pat. # 6884823). Trexel had 200 MuCell licenses by 2004 and 300 MuCell machines in operation by 2006, when it stopped requiring licenses and sold MuCell rights along with OEM machine installations. Since then Trexel hasn’t said how many MuCell machines are running, but it has licensed six OEM machine builders to sell MuCell-equipped injection molding machines directly and an additional six OEMs to install MuCell equipment purchased through Trexel.

Trexel, however, wasn’t first. Union Carbide Corp. developed physical foam molding with a continuously turning screw and melt accumulator, licensed to several machine builders in the 1960s and ‘70s, including Uniloy Springfield, now part of Milacron (www.milacron.com), which still builds them. Battenfeld GmbH in Meinerzhagen, Germany, now part of Wittmann-Battenfeld GmbH (www.wittmann-group.com), invented direct gas foam injection molding for an intermittent screw (U.S. Pat. # 4381272) in 1979 and built the machines until the late 1980s. In the 1990s Battenfeld improved control of gas injection and screw design with new patents (U.S. Pat. # 6451230) filed in 1999 in the U.S. and 1998 in Europe, just as Trexel was starting out. So Battenfeld and Trexel agreed not to sue each other’s customers.

Around 2000, two other physical micro foam technologies were developed and patented in Europe—Ergocell and Optifoam—both different from Trexel and Battenfeld, which add physical foaming into the barrel. Ergocell and Optifoam added blowing agent into the melt between the barrel and plunger, which meant dosing against much higher pressures of the injection mold vs. dosing against pressure in the barrel. A fifth microfoam technology, Profoam, developed around 2006, adds blowing agent to plastic pellets before melting with very low pressure. All five technologies—Battenfeld, MuCell, Ergocell, Optifoam and Profoam—describe using carbon dioxide and nitrogen blowing agents, but all known applications of any of them use nitrogen. All five could reportedly make microcellular foam parts that were similar or identical, which brought Ergocell and Optifoam into the shadow of Trexel’s part patent when they were commercialized.

Trexel’s microfoam part patent drove competing technologies Ergocell and Optifoam out of the market soon after commercialization. Trexel’s European part patent was later revoked, but its U.S. part patent runs until 2019, keeping most competing microfoam parts at bay.

Trexel’s microfoam part patent drove competing technologies Ergocell and Optifoam out of the market soon after commercialization. Trexel’s European part patent was later revoked, but its U.S. part patent runs until 2019, keeping most competing microfoam parts at bay.

Ergocell from Demag Ergotech GmbH in Schwaig, Germany, now part of Sumitomo (SHI) Demag Plastics Machinery GmbH (www.sumitomo-shi-demag.eu), injected physical blowing agent into the melt with a mixer and reservoir device. Ergocell was introduced at the K Show in Germany in 2001. Trexel cited its patent and insisted that molders needed a MuCell license to produce any micro foamed parts, regardless of method. By December 2001, Trexel had an agreement with Demag Ergotech to sell Ergocell machines only with a MuCell license, so Ergocell was effectively finished.

Optifoam, invented at the IKV Institute for Plastics Processing at RWTH University in Aachen, Germany (www.ikv-aachen.de), adds blowing agent to the melt using a porous spider-like torpedo in a porous sleeve (DE Pat. # 1983021). In 2003 the IKV licensed Optifoam exclusively to Sulzer Chemtech Ltd., Winterthur, Switzerland (www.sulzer.com), which commercialized it in 2005 with two automotive molders, Master Industries Inc. in Ansonia, OH, and a European molder. Sulzer, the IKV, and Huntsman Polyurethanes in Belgium also partnered to develop Optifoam thermoplastic PU shoe soles with 63% density reduction. Trexel filed a complaint in a U.S. district court against Master and in 2006 announced that Sulzer’s U.S. customer had stopped using Optifoam, while its European customer had taken a MuCell license.



Starting in 2003 Battenfeld, Sulzer, Peguform GmbH (now part of SMP Samvardhana Motherson Peguform GmbH), a molder of car cockpit systems in Boetzingen, Germany, and several other European companies with patents for micro foam molding began a long struggle to get Trexel’s master part patent revoked in the European patent office in Munich. In 2005 Trexel’s “Microcellular Molded Article” patent was revoked in Europe. Trexel appealed, and it was reinstated in 2007, revoked again in 2008, appealed again, and finally rejected by the European patent office in September 2011.

By 2008, Battenfeld had been acquired by Wittmann in Austria. Battenfeld’s second generation physical foam molding, now named Cellmould, was announced at a Wittmann Battenfeld open house in Austria that year, but the machine wasn’t yet commercial. Sulzer spun its static mixer business and Optifoam off as a management buyout in 2012 to Promix Solutions AG, also in Winterthur (www.promix-solutions.ch), which offered medium-to-high density Optifoam only for extrusion, not for injection molding (see this blog, Nov. 9, 2010). In 2014 Sulzer acquired low density extrusion foam specialist Aixfotech GmbH in Aachen, Germany, and took the Optifoam name back, but not the technology, which Promix continues to offer for extrusion.

Battenfeld, which built physical foam injection molding machines 20 years before Trexel, had an agreement with Trexel not to sue each other’s customers. Battenfeld’s newest physical foam molding, Cellmould, was launched in 2010 and has about a dozen users, including automotive.

Battenfeld, which built physical foam injection molding machines 20 years before Trexel, had an agreement with Trexel not to sue each other’s customers. Battenfeld’s newest physical foam molding, Cellmould, was launched in 2010 and has about a dozen users, including automotive.

The IKV’s patent-applied-for Profoam (WO Pat. Applic. # 200613660) combining blowing agent with pellets hadn’t been commercialized. The IKV’s first technology (DE Pat. Applic. # 102005061053) for adding physical blowing agent before the barrel used pelletized dry ice, or frozen CO2, blended with plastic pellets. Profoam, however, is like an autoclave, impregnating plastic pellets with nitrogen under low pressure before they are fed into the barrel.

Profoam was developed after 2007 in partnership with Arburg + Co KG in Lossburg, Germany (www.arburg.com); Volkswagen AG in Wolfsburg, Germany (www.vw.com); Ticona GmbH (www.celanese.com); Bayer MaterialScience (www.covestra.com); and LyondellBasell (www.lyondellbasell.com). The IKV first presented Profoam in an ANTEC paper in 2008 by Walter Michaeli, Thorsten Krumpholz, and Domenik Obeloer, who wrote his doctoral thesis on the process. Arburg and the IKV tested the prototype successfully with PP, PC, PC/ABS, PBT, PPS, PS, PA6, and even TPE. The IKV gave a second ANTEC paper on Profoam in 2011, but no partners are named on either ANTEC paper.

The IKV Institute in Germany invented Profoam micro foaming with two autoclave chambers that impregnate plastic pellets with physical blowing agent under pressure over the feed throat of the injection molding machine. Arburg built and patented the commercial version.

The IKV Institute in Germany invented Profoam micro foaming with two autoclave chambers that impregnate plastic pellets with physical blowing agent under pressure over the feed throat of the injection molding machine. Arburg built and patented the commercial version.

Arburg built a commercial version of the Profoam device with a control interface to an injection molding machine. The IKV had applied for a patent on Profoam, but Arburg, not the IKV, patented it (DE 2009100124810 published in 2010), including a critical valve for the autoclave chambers. The patent is under the name of Arburg’s inventor and senior partner, Karl Hehl, not under Arburg’s name. The device had two pressurized chambers, one on top of the other, in which pellets are impregnated with nitrogen before being fed into a standard injection molding machine.

Pellets are loaded into the top chamber under ambient pressure. The chamber is locked; blowing agent is added under low pressure (up to 50 bar). The airlock under the chamber opens, so pellets fall into the lower chamber, which locks. The upper chamber opens to ambient air and refills, as the lower chamber opens and releases pellets still under pressure into the feed throat. When the pellets melt, blowing agent dissolves evenly into the melt.

The amount and distribution of gas is far better controlled than by any of the higher-pressure injection approaches into melt, IKV’s Obeloer says. An important modification needed for the injection molding machine is to fit the back of the screw with a seal, so gas doesn’t escape. Profoam uses more nitrogen than other processes because of venting the upper chamber, but presumably this could be recaptured. Profoam molding is lower viscosity than solid molding, so it can make glass-filled PP with longer fibers than equivalent solid parts, Arburg says. Another plus is that the Profoam device can easily be moved from one injection molding machine to another or disconnected.



In 2009 Battenfeld, which always had the least to fear from Trexel’s part patent, opened its new plant in Meinerzhagen, Germany, and showed a Cellmould machine for the first time, offering it commercially in 2010 with improved controls, an improved screw design capable of finer bubbles, and pneumatically controlled shut-off nozzles. Cellmould can be used with Battenfeld’s BFMold (ball-filled mold) technology for rapid mold temperature cycling and with other mold technologies like gas counter pressure and expanding molds.

Cellmould, which doesn’t need a Trexel license, was commercialized in 2011 for washing machine parts, using the technology more for part strength and rigidity than for light weighting per se. Battenfeld says it has about a dozen Cellmould customers now, including Tier 1 automotive suppliers.

By 2013, Arburg had commercialized the first Profoam part in-house, molding glass-filled PP instrument covers for its own injection molding machines. Arburg didn’t show Profoam publicly until 2015, first in-house at its Technology Days conference in March, then at Fakuma 2015 in October, but not at NPE 2015 in the U.S., where Arburg showed light weight composite sheet technology instead.

At Technology Days 2015 Arburg molded glass-filled PP automotive airbag housings in a two cavity mold on an Allrounder 820A in 70 second cycles. The 280-mm-long part weighed only 272 grams, 18% less than a solid part. At Fakuma 2015 Arburg molded an automotive sliding tray housing on a hydraulic Allrounder 630 S in 65-second cycles. The 20% glass fiber PC part weighed 361 grams, 13% less than a solid part.

Arburg first showed Profoam publicly at Fakuma 2015, molding this automotive tray housing. As a licensed seller of MuCell, Arburg ran Profoam side by side with MuCell physical at its in-house Technology Days in 2015 and 2016. In 2016 MuCell molded this PA6 bracket.

Arburg first showed Profoam publicly at Fakuma 2015, molding this automotive tray housing. As a licensed seller of MuCell, Arburg ran Profoam side by side with MuCell physical at its in-house Technology Days in 2015 and 2016. In 2016 MuCell molded this PA6 bracket.

At Technology Days this year, Arburg showed Profoam molding an ambitious automotive interior glove compartment door with a high gloss surface, using “variotherm” rapid mold surface temperature cycling. The PC/ABS part was molded in 60-second cycles on a hybrid Allrounder 630H with 2500 kN of clamp force. The part weighed 200 grams with wall thickness of 1.8 mm, 30% less than the previous solid part with 2.5 mm walls.

Arburg, which is a licensed seller of Trexel’s MuCell, ran Profoam and MuCell side by side at its Techology Days both years. In 2016, MuCell foamed a structural bracket out of nylon 6 using a mold designed for a solid part and achieving a 28 second cycle—20% faster than solid molding with a weight reduction of 7%, while meeting all strength and dimensional requirements.

Arburg won’t say how many Profoam devices it has sold, but Arburg builds only one Profoam size with one-liter chambers for small injection molding machines up to 200 tons, so it’s safe to assume they are for small parts and development. VW, which partnered with the IKV on Profoam in 2010-2012, has a Profoam device in its technology center, but doesn’t use it commercially, and the IKV has one. Automotive applications will presumably have to wait for the 2019 model year, when Trexel’s U.S. part patent will have expired.

With competition on the horizon Trexel’s business is also becoming more mature and competitive. In April Trexel introduced its T Series of new user friendly “smart” dosing control, which requires an operator to input only two parameter—shot weight and percent of supercritical nitrogen—setting the rest automatically. Previously MuCell dosing was much more difficult, requiring a highly trained technician to enter six parameters.

Trexel is also becoming a broader based foam company, offering engineering design for foam parts with up to 30% weight reduction instead of the usual 10%. In September 2015, Trexel even introduced a chemical foaming agent, called TecoCell, licensed for injection and automotive blow molding from Polyfil Corp., Rockaway, NJ. Polyfil continues to offer its EcoCell foaming agent for extrusion. So Trexel can now offer foaming for smaller production runs without capital investment.

Posted in Uncategorized | Tagged , , , , , | Leave a comment

New Process Makes PLA Block Copolymers

By Jan H. Schut

A new two-stage catalytic process from NatureWorks LLC, Minnetonka, MN (www.natureworksllc.com), the world’s largest producer of polylactic acid (PLA) biopolymer, takes an unusual mixed lactic acid feed stream, invented at Nagoya University in Japan (www.agr.nagoya-u-ac.jp), and makes it into semi-crystalline PLA block copolymers. The new copolymers, which are still in R&D, could potentially offer substantially higher melt temperatures than PLA homopolymer.

All commercial PLA so far is based on bacterial fermentation to a single lactic acid isomer (L-HLA), which is condensed into L-lactide monomer, which is polymerized into PLLA homopolymer. PLLA homopolymer was introduced in the 1990s by Cargill Inc., Minnetonka, MN (www.cargill.com) and commercialized in 2003 by NatureWorks, which is 50% owned by Cargill.

Instead of L-lactide, the feedstock for NatureWorks’ developmental PLA block copolymers is a mixture of “racemic-lactide” and “meso-lactide.” Lactides are ring compounds made by dehydration and condensation of hydroxyl and carboxyl groups from lactic acid and come in three shapes. L-lactide is a ring made of two L lactic acids; D-lactide is a ring made of two D lactic acids; and meso-lactide is a ring made of one of each. A racemic mixture means a 50/50 mixture of left- and right-handed molecules, so rac-lactide has equal amounts of L-lactide and D-lactide.

Lactic Acid

The feedstock for NatureWorks developmental process for PLA copolymer is a lactide mix invented at Nagoya University with equal amounts of LLA, DLA and meso-LA rings, made chemically from glycerin. Commercial PLAs are homopolymers made by fermentation of sugar.


The rac-lactide mixture is made by a process invented by Professor Nobuyoshi Nomura of the graduate bio-agricultural polymer chemistry department of Nagoya University and reported in a series of Japanese patents and technical papers in 2003-2011. Nagoya University’s patented process (JP Pat. Applic. # 2003-64174) uses a purely chemical reaction of glycerin, or glycol, a byproduct from bio diesel production, with sodium hydroxide, or caustic soda.

The process was interesting, but not commercially significant at the time because the racemic lactic acid mixture yielded mixed lactides that didn’t make PLA with useful properties. Conventional ring-opening catalyst polymerized mixed lactides into atactic amorphous PLA, which has no melt temperature. Rac-lactide (half LLA and half DLA) could be separated from meso-lactide because rac-lactide has a higher melt temperature, but then one-third of the feedstock would be wasted. So the cost of separation and waste would have made Nagoya University’s chemical route to PLLA copolymer too expensive vs. existing fermentation routes to PLLA homopolymer.

Test Properties of NatureWorks’ PLA Block Copolymers in R&D

Polymerization time

rac-LA/ meso-LA Ratio

Step 1 Step 2 Mol. wt. Melt temp. °C Tacticity type
Reference example 3 100/0 2 hrs 0 109,000 183 Semi- crystalline
Working example 6 90/10 1 hr 3 hrs 143,000 177 Semi- crystalline
Comparative example 7 90/10 3 hrs 0 136,000 N/A Amorphous
Source: NatureWorks/Nagoya Univ. JP Patent #50689; U.S. Pat. Applic. #20150087799)

Dozens of researchers since the mid-1990s tried to find a catalyst process to polymerize a mixture of racemic-LA and meso-LA. Nagoya University’s Nomura tested a wide range of site-selective catalysts, including chiral salen- and homosalen aluminum complexes and conventional ring-opening catalysts with chain-end polymerization trying to create isotactic PLA copolymers. But with both site selective and chain end catalysts, the meso-lactide in the mix created “errors” in the PLA chains that made the material amorphous and unusable with low isotacticity. The isotacticity level of a polymer is the average number of molecules strung together with regularly alternating “thumbs” without errors in the sequence.


NatureWorks’ invented a two-stage polymerization process that converts Nagoya University’s glycerin-based mixture of lactides (33% L-LA, 33% D-LA, and 33% meso-LA) into stereo regular, semi-crystalline PLA copolymers without separation. The patent (JP# 5806890; U.S. Pat. Applic. #20150087799) names NatureWorks, Nagoya University, and two companies, Hitachi Zosen Corp., Tokyo, Japan (www.hitachizosen.co.jp), and Tohoku Electric Power Co., Sendai, Japan (www.tohoku-epco.co.jp), which were on the original Nagoya University patent.

In the first stage, the higher solubility of meso-LA at lower temperature allows it to melt and dissolve into a small amount of solvent and be polymerized with ring-opening catalyst into amorphous polymeso-LA blocks, while the lower solubility of rac-lactide leaves it unreacted in the solid phase. In one patent example, first stage polymerization of meso-LA is done at 90 degrees C for one hour. Since the first polymerization stage allows meso-LA to be separated from racemic-LA, NatureWorks could test several different ratios of racemic-LA to meso-LA including 80/20, 90/10, and 100% racemic-LA with no meso-LA.

JP Patent image

NatureWorks two-stage catalytic process polymerizes low temperature meso-LA into amorphous blocks. Then site-selective catalyst polymerizes long blocks of LLA and DLA onto polymeso-LA, forming semi-crystalline PLA copolymers with melt temperatures up to 220 °C.

In the second polymerization stage, rac-lactide (50% L-LA and 50% D-LA) is heated to a higher temperature, allowing it to dissolve and react in the presence of Nagoya University’s single-site salen aluminum catalyst. Second-stage polymerization is described in one patent example as taking three hours at 130 degrees C without solvent. In the second polymerization stage, rac-lactide is therefore grafted onto polymeso-lactide.

“The final architecture is a block copolymer with one block of meso-lactide followed by large blocks of L-lactide followed by large blocks of D-lactide and vice versa,” NatureWorks’ chief scientist, Joseph Schroeder explains. “The catalyst is capable of generating blocks of L and D that are long enough to form stereo complex crystals that melt at 220 °C.” NatureWorks’ patent describes PLLA and PDLA blocks of eight molecules or more. PLLA and PDLA homopolymers have lower melt temperatures of 162-180 °C.

NatureWorks’ R&D on glycerin-based PLA copolymers so far is only on a bench top scale, and no timing has been set for when it might be scaled up. In the meantime, NatureWorks more immediate R&D priority is PLA based on methane, which could come from a variety of sources. Methane, which is produced by natural decomposition of plant matter, is a component of natural gas and is also generated from landfills and waste water treatment.

PLA from methane is part of NatureWorks’ joint R&D program with Calysta Inc., Menlo Park, CA (www.calysta.com), supported by a U.S. Department of Energy grant, for bacterial fermentation of methane to lactic acid, which would feed NatureWorks’ existing PLA homopolymer. Both glycerin and methane based research programs are environmentally interesting because they create PLA from industrial bi-products, not from sugar.

Posted in Uncategorized | Leave a comment