Progress in polyester development for EBM applications.
For nearly 30 years, Eastman has been actively developing solutions for clear EBM applications. Market research has convincingly shown that consumers value the ability to see the contents of their food and beverage packages. Polyethylene terephthalate (PET) containers produced by the injection stretch blow molding process (ISBM) are the most common type of transparent container on the market; over 5 billion lbs of ISBM PET is sold annually in the United States. (1) Unfortunately, the ISBM process is limited to simple shapes and cannot produce bottles that contain a through-handle. With time, handles have become increasingly desirable to the consumer, particularly for larger bottle sizes where the ergonomics of gripping are critical to the user. Larger-size bottles containing a through-handle can be produced only by the EBM process.
In short, the three primary fitness-for-use (FFU) attributes desired for clear, EBM-produced, handleware are the following: bottle performance, efficient processing, and sustainability. With regard to bottle performance, drop impact (or bottle toughness) is typically the most important requirement. In some cases, hot-fill capability or chemical compatibility with a specific ingredient must be considered. Nevertheless, how a bottle behaves when dropped from a test height of 3 to 6 feet is often a critical benchmark to many brand owners. With regard to processing efficiency, a clear resin formulation must perform well on existing blow-molding machines. To do so, it must have excellent melt strength, experience minimal degradation during extrusion, and exhibit no melt fracture-induced haze, which could compromise the container clarity and gloss. Finally, sustainability has become a growing objective of all participants in the value chain. Today, the most common sustainability demand of a packaging resin for clear handleware is that it be considered compatible in the PET recycle stream (implying resin identification code 1 is acceptable). As this article will demonstrate, producing a resin with the most adequate balance of these properties is a complex and delicate act that has required many years of research and development.
The industry's earliest attempts at developing a clear EBM grade began with amorphous PETG in the 1980s. PETG is an amorphous thermoplastic copolyester still used widely for injection-molding and thermoforming applications. However, it was generally revealed to be a poor EBM resin because it had very little melt strength, exhibiting many bottle-breakage problems, and was incompatible with the PET recycle stream (because it does not crystallize). While PETG is still a material option, it is largely inadequate for the EBM market, except for only very small containers (<32 oz) where recyclability and process efficiency are of little concern.
In the early 1990s, a product referred to here as EPET was developed and promoted. EPET was produced by utilizing solid-stating (SS) polymerization to raise the inherent viscosity (IhV) of ISBM bottle-grade PET to a high level (>0.95 IhV), where melt strength becomes sufficient enough to support parison formation and blowing. The compositional similarity of EPET to ISBM PET made it highly compatible with the PET recycle stream. However, this effort suffered a number of processing-related downsides. First, the high IhV values caused significant gel and un-melt processing problems. Second, the hot processing temperatures (>280[degrees]C) often incurred degradation and subsequent melt strength loss. Finally, bottle-breakage problems proved endemic to the formulation. As such, EPET was terminated as a product in the mid-1990s, and the packaging market was left without a robust option for clear handleware.
Coincidentally, all other major producers of similar EPET products at the time also terminated production for similar reasons.
In the early 2000s, a major breakthrough occurred with the development of modified PETs having improved toughness and processing efficiency versus the previously described efforts. By employing the cyclohexanedimethanol (CHDM) glycol monomer, along with other proprietary production and formulation techniques, a robust copolyester portfolio with a balanced range of processing, performance, and recycling ability was developed. In this article, three copolyesters are generically referenced as CoPET-1, Co-PET-2, and CoPET-3, developed in 2004, 2008, and 2011, respectively. Furthermore, the three generations of resin development reflect market demand for sustainable and recyclable solutions, which concurrently minimize the drawbacks to EBM processing efficiency and bottle quality.
Table 1 provides a summary of five copolyester (or EPET) resins developed over the last 30 years for the EBM process. The three copolyester formulations (CoPET-1,2,3) and the PETG are all melt-phase (MP)-produced. This implies that the entire polycondensation reaction required to reach the necessary molecular weight (or IhV) is carried out while the polymer is under vacuum and well above the melting temperature of the resin. On the contrary, the EPET resin was solid-stated (SS), which is a polymerization process during which the polymer chain is further lengthened (beyond the melt phase) by heat and purging with an inert gas to drive off reaction byproducts. As will be shown, there are intrinsic benefits to a melt-phase-produced resin versus the SS EPET material of the 1990s. Multiple generations of CoPET have been developed to maximize the balance of the three critical legs of the stool: bottle performance, efficient processing, and sustainability. We'll now explore each of these topics separately, highlighting important progress achieved in each key area.
No gels & un-melts
Gels (or un-melts) can best be defined as viscous inhomogeneities within a molten polymer stream. The presence of such gels in a clear blow-molded article is certainly undesirable. Un-melts are most often caused by crystalline structures, or widely varying viscosities, which are not homogenously mixed together before the end of the extruder barrel and die exit. Gels proved to be a significant problem with the initial EPET resin developed in the 1990s. As described, EPET was solid-stated, which involves heating pellets in hot stream of inert gas. As one may expect, the diffusion path for reaction byproducts being removed is much shorter for molecules located near the pellet surface. On the other hand, reactions occurring near the pellet center (or core) are much slower. This naturally generates a molecular weight (or IhV) gradient through the pellet thickness of solid-stated PET materials. This is shown clearly in Figure 1. Because melt-phase resins experience uniform IhV build throughout the polycondensation process, their pellets exhibit a constant molecular weight throughout the pellet radius.
To demonstrate this effect on processing, MP and SS versions of CoPET-3 were produced. In the SS sample, a 0.72 IhV melt-phase polymer was raised in the solid-state reactor to a 0.76 IhV, matching that of the melt-phase material (CoPET-3, Table 1). All melting temperatures, crystallization rates, and compositional variables are identical for the two materials. Once produced, the MP and SS samples were extruded on a 64-mm single-screw extruder at various RPM and temperature settings. Table 2 describes the results.
It is apparent that the SS sample produces significant gels versus the MP material, even when extruded under the same conditions. In blow molding, it is desirable to minimize the processing melt temperature so that viscosity (and melt strength) is maximized. As shown in the 15-RPM test, the MP resin exhibited good aesthetic quality at 233[degrees]C, while the SS version had to be raised to 241[degrees]C (by raising barrel zone temperatures). In the 7.5-RPM test, a good-quality melt was observed at 230[degrees]C for the MP version, whereas the SS still showed significant gels at 249[degrees]C. The reason for this is that the molecular weight gradient of the SS pellets (Figure 1) requires much more energy (and therefore hotter temperatures) to obtain a homogeneous parison stream. This is undesirable for the EBM process, and so development focus shifted to CoPET (MP) technology in the early 2000s.
No degradation during processing
Another key processing attribute for a successful blow-molding resin is that it should exhibit minimal degradation, or IhV loss, during processing. Loss of IhV has many deleterious effects on parison control, melt strength, ability to utilize regrind, and sometimes physical bottle properties. The three primary causes of degradation during polyester processing are high temperatures, long residence times, and high moisture contents. All three contribute to both thermal and hydrolytic degradation mechanisms, well known in polyester materials. (2) Because degradation of PET can be a significant concern above 280[degrees]C, it is desirable for an EBM resin to have a low melting point, which allows processing at much colder temperatures where degradation is almost nonexistent. While a low melting point is less desirable for the PET recycling process (discussed below), it offers strong processing upsides to bottle makers.
To explore this concept further, CoPET-1,2,3 and EPET were exposed to differing moisture levels (measured by Karl Fischer), and stored in drums for several weeks for equilibration. Then, each resin was extruded (64-mm extruder) at a processing temperature hot enough to ensure that no unmelts were present. A range of RPMs (5-25) was explored, to vary the residence time in the extruder barrel. The starting pellet IhV was measured, as was the extrudate IhV. Figure 2 reveals the results of this test.
Clearly, the lower processing temperatures of the CoPET resins provide an appreciable advantage regarding the amount of tolerable moisture before degradation becomes problematic. More than 0.03 IV loss could pose an EBM process concern. Therefore CoPET-1,2, CoPET-3, and EPET must be dried to <500, <250, and <50 ppm, respectively. The lower processing temperatures of the CoPET materials enable this advantage. Because CoPET-1 and CoPET-2 are amorphous and have low melting temperatures (<202[degrees]C), they can be processed much colder (220[degrees]C-235[degrees]C). CoPET-3 has a higher melting temperature (225[degrees]C), but is still an MP-produced polymer that can be processed cold enough, where degradation is not a major concern (235[degrees]C-255[degrees]C). EPETs have the highest melting temperature (>240[degrees]C), and must be processed even farther above that (265[degrees]C-285[degrees]C) to eliminate the gels and un-melts that are endemic to an SS resin.
One obvious drawback to resins requiring lower moisture levels for processing is the fact that the drying loop must be controlled to extremely high precision, with little margin for error. A second issue stems from the fact that once processed, virgin polymer must be reground and incorporated back into the EBM process at levels as high as 50%. A material like EPET, which is processed hot and more prone to degradation, can generate a wide range of parison viscosities (or melt strengths) with only minor variations in moisture level. These variations undoubtedly challenge day-to-day process control.
No melt fracture during processing
A final key processing attribute for the clear EBM resin is that the material must remain clear (haze-free) even at the highest output rates. Several highly viscous polymers are known to generate the condition known as sharkskin on the surface of extruded parts. This form of melt fracture is visually observable as a frosty white matte surface haze, and is undesirable for transparent bottle applications. Melt fracture is a theological flow instability that occurs as molten polymer encounters high shear stresses as it flows over a metal surface, such as the surface of an extruder die) High output rates and narrow die openings will lead to large shear stresses, which commonly manifest melt fracture on the part surface. Wheel and intermittent processes (reciprocating-screw or accumulator-head type machines) create the highest shear rates, due to the large output capability of the former and the rapid injection flow in the latter. Raising the parison temperature can reduce melt fracture, but this can cause the unfortunate consequence of reducing the material's viscosity and causing melt strength loss. As such, a material with a low tendency to melt fracture is desirable.
While CoPET-1 is a successful material on shuttle-style EBM equipment, its tendency to melt fracture has historically limited its ability to perform well on high-output machinery. Consequently, this handicap offered an opportunity for improvement, giving rise to the development of CoPET-2. While the resin formulation details are proprietary, Figure 3 clearly shows the advantages that CoPET-2 brings, regarding melt fracture elimination. In this test, each resin is dried and extruded (64-mm extruder) through a thin slit die (1.5 mm x 76 mm) at various output rates (5-50 RPMs). The temperature of each resin was processed at its recommended setting, required for sufficient melt strength (EPET = 265[degrees]C-285[degrees]C; CoPET-3 = 235[degrees]C-255[degrees]C; CoPET-1,2 = 220[degrees]C-235[degrees]C). The melt temperature, output rate, exiting IhV, and known die geometry were used to calculate the shear stress experienced by the polymer flowing through the die (viscosity times shear rate).
It is apparent from Figure 3 that CoPET-2 and CoPET-3 exhibit no melt-fracture--induced haze, compared with CoPET-1 and the EPET sample. Microscopy analysis of extrudate strands exiting a capillary rheometer (not shown) confirmed the absence of sharkskin behavior for CoPET-2 and CoPET-3. CoPET-2 has been trialed extensively on wheels and reciprocating-screw machines and has been found to produce clear bottles in every case. Lab-scale tests suggest that the newer CoPET-3 will behave identically to CoPET-2. The ability to process successfully on all EBM platforms was a formidable challenge for CoPET-1. However, resin innovation has enabled this to become reality with the newer CoPET materials.
The most commonly encountered FFU requirement of a blow-molded container is that it not fail when dropped from a specified height, relevant to its application. Other performance requirements can matter also, such as chemical compatibility with the container's contents, heat resistance at the exposure temperatures, suitable oxygen or moisture barrier, and appropriate top-load performance. For much of the market for clear handleware, the aforementioned performance needs are not stringent and the CoPET portfolio is satisfactorily robust. Drop-impact results, on the other hand, are a germane concern to almost every EBM container application. As such, maximization of bottle toughness is a vital goal in resin design.
Many factors can influence the drop-test performance of a container, including container size, container weight, unique design features (which can help or hurt), variability in the blow-molding process, mold integrity, consistency of the deflash process, drop-test temperature and test method, as well as inherent resin properties, like toughness and rigidity. Because of this large list of influencing factors, it can be difficult to determine easily just how well a particularly resin may perform. Nevertheless, years of experience have produced much data, which can be synthesized for some reasonable level of comparison.
Figure 4 is a bar chart showing the average failure height of CoPET-1,2,3 for a 59-oz and an 89-oz handleware container (23[degrees]C testing). The 59-oz data was generated from an Eastman-designed mold, run with all three generations of the CoPET family on a shuttle machine. The 89-oz data is compiled from a broadened source of container designs, produced on many different machine platforms over five years. The drop-test method used specifies that the container be dropped first at 3 ft. If the bottle survives, it is subsequently dropped at 1-ft-higher increments (4, 5, 6, etc.) until the container breaks. It is emphasized that an average failure height is being reported. Drops were conducted at room temperature, using a swing-away platform tower. To eliminate bias in the data, half of all containers were dropped with handle-in orientation and half were dropped handle-out.
The error bar heights are meant to reflect the total variability that can occur for a specific material and container size. For example, the CoPET-1 material can produce an 89-oz container that may perform well at 8 ft or higher, if everything is optimized perfectly. However, with challenging design features or a poorly optimized mold, then CoPET-1 may produce a container that survives drop heights of only 4 ft.
For CoPET-2 and CoPET-3, it is apparent that some sacrifice of toughness occurs. As durability is sacrificed, the average failure heights drop and the error deviations lessen. Nevertheless, these results indicate that all three materials should be acceptable for applications requiring a successful drop-test of a minimum of 3 ft. For large containers (> 128 oz), high design complexity, or applications needing greater sturdiness and reduced failure rate, the CoPET-1 formulation may be most preferred. While the EPET material of the early 1990s is not shown, historical experience suggests that average break heights on similar containers made from this material were less than 4 ft. Use of the CHDM monomer in CoPET production (which toughens PET) enables advantages, as shown in Figure 4. By comparison, many EPETs are copolymerized with isophthalic acid, which is known to decrease toughness.
It is undoubtedly clear that one of the largest growing FFU needs in the packaging market is for materials to have a strong sustainability story. Sustainability can mean something different to each market player. To some, lightweighting or incorporation of recycled content can be critical needs. To others, full lifecycle and energy consumption analysis are desired. While all facets are critical, there is no question that the most common sustainability demand of a packaging resin for clear handleware is that it be considered compatible in the PET recycle stream (implying resin identification code 1 is acceptable). As such, this has motivated developments of CoPET-2 and CoPET-3 to improve upon generation CoPET-1. The first objective in each development case has been to increase the resins compatibility with PET, such that it will not be problematic in the current recycling process, even at levels far exceeding expected market penetration. The secondary objective in each case was to minimize the drawbacks to process efficiency and bottle toughness, described previously.
While no single group defines PET compatibility or regulates the choice of resin identification code for all users, the Association of Postconsumer Plastic Recyclers (APR) has developed a useful testing protocol that purposes to evaluate challenges that a novel resin may pose to the existing recycle PET stream (PET Bottle Critical Guidance (CG) Document). (4) This test has become an important metric and is trusted by many participants in the value chain for assessing a new resin's potential impact on the recycle process. The CG test protocol specifies that an innovative material must be blended with a control PET standard at 25% and 50% levels. Then, the blends are crystallized, solid-stated, injection molded, and analyzed with a battery of analytical techniques. To pass the test, the blends (relative to the control) must exhibit similar IV loss during extrusion, similar solid-stating rates, minimal color shift during processing, no sticking during drying, and melting temperatures in the 235[degrees]C-255[degrees]C range.
While the full summary of test results is larger than can be presented here, the results of the melting temperature test are shown in Table 3 as a sampling.
In each case, 25% and 50% of the innovative polyester are blended with Polyclear 1101 (0.80 IhV APR-approved control PET resin). Melting temperatures are measured on a solid-stated pellet and the second-heat DSC melting temperature is reported (10[degrees]C/min scan rate). It is clear from Table 3 that CoPET-2 passes the melting temperature test (235[degrees]C-255[degrees]C) at 25% loading, but fails at 50%. CoPET-3 passes the test criteria at both 25% and 50% loadings. The final column in Table 3 summarizes how each generation of CoPET performs, when the entire CG testing protocol is considered. (Note this testing was conducted externally; except for CoPET-3). For this resin, internal testing shows "pass" at 50%, while external testing is under way at the time of this publication.
Beyond APR testing, one of the biggest practical concerns for the recycling community is how post-consumer amorphous bottle flake will perform when dried with crystalline PET at 160[degrees]C drying conditions. All EBM containers produced from any of the resins shown in Table 1 are amorphous, making them different from ISBM PET, which has crystalline sidewalls from the stretching process. Unfortunately, ground flake from amorphous containers can stick to the walls of the dryer or agglomerate with PET container flake in a dryer set at 140[degrees]C-180[degrees]C. This can pose significant problems for recyclers and can occur at levels as low as 0.1% for materials like PETG. As a result of this concern, it has become necessary to develop specific test methods to better quantify sticking for each resin in the portfolio.
To evaluate and quantify flake sticking during drying, a specially designed canister apparatus was built (10-inch diameter, 9 inches tall), and then inserted into the center of a standard Conair drying hopper. A schematic of the dryer test is shown in Figure 5.
The setup was constructed with mesh plates on top and bottom of the canister insert, to force all air flow through only the canister. In order to simulate a full-scale dryer filled with the test flake (8+ feet tall), a stack of four 25-lb weights was applied to the flake, by way of a narrow rod extending from the top of the dryer. The motivation for building this setup was to adequately represent the experience of flake in a large dryer setup (which would otherwise require 1000+ lbs of material).
To perform a dryer sticking experiment, the canister was filled with 6 lbs of the test flake blend. Then, the canister was inserted into the dryer hopper, with air temperature set to 150[degrees]C. Next, the weights were applied for a period of 2 hours (sufficiently long to assess sticking). After testing, the canister was emptied. Flake that is highly friable and loosely bound does not tend to cause major concern in a production-scale drying setup, versus flake pieces that are tightly bound. To assess this, the cooled flake was placed in a 300-1b-capacity fiber drum and tumbled for 5 minutes. Next, the flake was poured through a 1/2-x 1/2-inch wire mesh. The caught clumps were weighed, and are reported here as a percentage.
Results of the canister test are presented in Table 4.
In each case, the named polyester was blended at various levels (1%-50%) with post-consumer recycled (PCR) PET flake, obtained from Pure Tech Plastics. As anticipated, the polyester materials that crystallize the fastest (Table 1) stick the least in the dryer evaluation. Clearly, EPET and CoPET-3 should cause no significant dryer sticking at 50% loadings in most recycle drying setups, based on this study. CoPET-2 could cause concern at levels >25%, whereas CoPET-1 is expected to be problematic in the 5%-10% range. By comparison, PETG causes severe dryer sticking problems (3% or less, blended with PET PCR).
For some perspective on the APR and dryer sticking percentages reported here, it is useful to consider the expected market penetration of the CoPET resins. Given that over 5 billion lbs of PET exists on U.S. store shelves, clear handleware materials like CoPET are unlikely to ever exceed 1% of the total market. However, it is recognized that slugs of bottles can enter a recycler's production. As such, each user must carefully evaluate their individual market circumstances in assessing the degree of PET compatibility required for a specific application, along with the associated choice of resin-identification code.
This article has highlighted more than 30 years of progress in developing polyester resins for clear handleware applications. As shown, balancing the three key material requirements (process efficiency, bottle performance, sustainability) in a single formulation is challenging. There are certainly inherent tradeoffs in resin selection. CoPET-1 is an extremely tough material that processes easily on shuttle-style EBM equipment, but is only compatible in the PET recycle stream at approximately 5% levels. CoPET-3 is the newest development; it brings the highest level of compatibility in the recycle stream (50%), and is compatible on almost all EBM platforms, but it does incur some sacrifice in bottle toughness and processing ease. CoPET-2 falls in the middle. Nonetheless, the melt-phase production and compositional adjustments of all three CoPET generations mark significant progress over the solid-stated EPET efforts of the early 1990s. The three represent a robust portfolio of copolyesters.
The authors thank the Eastman Chemical Company for the opportunity to present this paper at ANTEC[R], and the many colleagues who have offered invaluable contributions to progress in this market.
(2.) S. Al-AbdulRazzak and S. Jabarin, Polymer International, 51, 164-73 (2002).
(3.) S.G. Hatzikiriakos and K.B. Migler, Polymer Processing Instabilities: Control and Understanding, Marcel Dekker, New York (2005).
(4.) http://www.plasticsrecycling.org/technical-resources/ critical-guidance.
Note: The authors presented a version of this paper at ANTEC[R] 2012.
Mark A. Treece and Thomas J. Pecorini
Molding Applications Research and Development Laboratory
Eastman Chemical Company
Kingsport, Tennessee USA
Table 1. Summary of EBM Resin Grades for Clear Handleware. Melt Phase (MP) DSC or Pellet Cryst. Melting Solid IhV Half-time [T.sub.m] Stated Polyester (dL/g) (min) [degrees]C) (SS) EPET 0.95 <2 240 SS CoPET-3 0.76 2 225 MP CoPET-2 0.73 20 202 MP CoPET-1 0.76 >200 190 MP PETG 0.75 >1000 170 MP Table 2. Comparison of a Solid-Stated vs. Melt-Phase-Produced Polyester, During Single-Screw Extrusion. Extruder CoPET-3 CoPET-3 Settings (MP) (SS) RPM Melt Temp Melt Melt (C) Quality Quality 7.5 230 Good Gels 239 Good Gels 249 Good Gels 15 233 Good Gels 241 Good Good 252 Good Good Table 3. Summary of Polyester Performance in the APR Critical Guidance Testing Protocol. Melting Point of Blend 25% 50% Passes Polyester Polyester APR in APR in APR Critical Control Control Guidance Polyester Resin Resin at: EPET > 235 (Pass) 50% CoPET-3 241 (Pass) 237 (Pass) 50% * CoPET-2 240 (Pass) 233 (Fail) 25% CoPET-1 <235 (Fail) 5% PETG * Indicates internal results only. External testing is under way. Table 4. Summary of Polyester Performance in a Dryer Sticking Test. Wt% Clumping in % Polyester in PET 50% Polyester PCR which with causes > 1 Polyester 50% PET PCR Clumping EPET <1.0% >50% CoPET-3 <1.0% 50% CoPET-2 37% 20% CoPET-1 >50% 7% PETG >50% 3% Figure 4. Drop-test results for CoPET resins in a 59-oz and 89-oz bottle. Average failure height (ft.) reported. Average Break Height (Ft.) 89 oz. 59 oz. CoPET-3 4.8 6.0 CoPET-2 5.5 7.0 CoPET-1 6.8 9.0 Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||extrusion blow molding|
|Comment:||Progress in polyester development for EBM applications.(extrusion blow molding)|
|Author:||Treece, Mark A.; Pecorini, Thomas J.|
|Date:||Jun 1, 2012|
|Previous Article:||ANTEC@NPE2012: a massive "Show & Tell" ... or rather "Tell & Show," as several NPE-exhibiting companies supplied speakers for the yearly SPE...|
|Next Article:||HDPE-TPS bottle returns double environmental benefit: more renewable material; reduced greenhouse gas emissions.|