Extrusion, Problems

Reduce Gel Formation in Medical Extrusion

Reduce Gel Formation in Medical Extrusion

Thomas Black
Merritt Davis Corporation
Medical/Pharmaceutical Conference
Las Vegas, NV Dec 2001

Introduction:

More so than in almost any other industry, plastic processors of medical products are particularly sensitive to concerns of extrudate quality. Whether the product be flexible PVC drip tubing or PET balloon catheters, processors must take a “zero tolerance” approach toward defects in their end product and vigorously combat the formation of gels wherever present in the process. While “gel free” processing may be wishful thinking, with careful material evaluation and persistent process sleuthing we can clearly reduce the number of gels, which wind up in the end product. This paper will review the more common materials used in medical applications, their potential for gel formation and recommend strategies to improve extrudate quality by reducing the occurrence of gels in the final product.

Polymer Compounds:

All polymers materials undergo chain scission followed by some cross-linking when heated for too long a period of time at too high of a temperature. What makes these materials prone to gel formation is their basic molecular architecture. Consider the following;

Polyethylene is free radical polymerized from ethylene monomer and can be produced into a number of different products including, branched HPLDPE, LLDPE and HDPE among others. Owing to their aliphatic structure, polyethylene’s tendency to crystallize is well known with semi-crystalline PE morphology consisting of crystalline structures dispersed within an amorphous domain. Interestingly enough, the free radical polymerization mechanism of PE is initiated with the use of a peroxide catalyst. Further, the degree of branching is often controlled with the addition of comonomers such as butene, hexene or octene, with branching often occurring where tertiary carbons are found. These are carbon atoms with 3 valences attached to other carbons and a single valence connected to hydrogen. Known as the weak points in the chain, these positions are particularly susceptible to degradation from atmospheric oxygen. Given this scenario, it becomes apparent that all grades of polyethylene require the use of stabilizers in order to protect against oxidative chain degradation.

Like polyethylene, flexible PVC is polymerized via free radical mechanisms, and as such, is prone to many of the same degradation schemes. Beginning with vinyl monomer, unsaturation present in the main chain is used for polymerization with peroxide initiators. In light of this, residual unsaturation found in the polymer becomes an opportunity for thermo-oxidation during melt processing leading to the development of cross-linked species or gel formation. Again, tertiary carbon atoms are points for attack and PVC heat stabilizers must be used to trap hydrogen chloride generated during degradation, eliminate weak points through selective substitution and provide anti-oxidant protection against thermo-oxidative degradation.

PEBAX or elastomeric nylon is a condensation polymerized polyether block polyamide, produced from the reaction of polyether diols or polyesters and polyamides containing carbonyl end groups. The connection between hard and soft segments, made with amide linkages, are susceptible to hydrolysis (more so with polyester based PEBAX) and therefore, the polymer must be pre-dried prior to processing.   Degradation of PEBAX is typically through chain scission mechanisms, along with cross-links formed from the creation of carbonyl groups, resulting from the attack of atmospheric oxygen along points of unsaturation in the main chain.

TPU’s (thermoplastic urethanes) are also block copolymers, albeit of a hard diisocyanate crystalline segment coupled with a urethane linkage to a soft polyether or polyester amorphous segment. Like the amide linkage of nylons, the urethane linkage is capable of hydrogen bonding and is affected by the presence of oxygen. Based on either polyester or polyether co-polymerization with diamines, TPU’s are hygroscopic and require pre-drying prior to processing. Moreover, polyether based TPU’s are more resistant to hydrolytic and oxidative degradation, and therefore should be less prone to cross-linking or gel formation. As well, TPU’s tend to be shear sensitive and are typically high in viscosity, which can lead to high melt temperatures, high discharge pressures and high shear stresses resulting in the potential for degradation during processing.

(Gels) What are They?

As a working definition, gels are small solid masses that cannot be melted which result from cross-links formed when a polymer is exposed to high temperatures for a long period of time. As discussed above, these degradation products are the result of polymer architecture coupled with process conditions, and are catalyzed in the presence of atmospheric oxygen. For materials such as polyethylene, the cross-linking reaction predominates the degradation scheme due to the lack of steric hindrance within the polymer molecule, whereas for polypropylene chain scission is the prevailing degradation scheme based on the steric hindrance created from pendant methyl groups. This thermo-oxidative degradation mechanism is catalyzed by the presence of oxygen and follows the generalized Arrenhius expression for chemical reactions, which states that the rate of a reaction is approximately doubled for every 100C increase in temperature.   This implies that reducing the melt temperature by as little as 100C allows the polymer to spend twice as long at a given temperature prior to the onset of degradation.

(Gels) Where Do They Come From?

As mentioned earlier, gels are the result of thermal degradation which can form anywhere along the process lifecycle, from polymerization, to processing, to end product use, and even storage. In some cases, high molecular weight cross-linked species are developed in the polymerization reactor due to un-wiped surfaces resulting in long residence times. In other cases, gels are the product of high melt temperatures and long residence times developed in the extrusion process due, perhaps, to the presence of dead spots or stagnation along the melt train or a broad residence time distribution (RTD). Evidence of PVC cross-linking or degradation is clearly demonstrated in the graph below of viscosity verus residence time taken at several different values of melt temperature (Figure 1).

It’s well understood that polymer degradation is both time and temperature dependent, suggesting that polymers can be processed for either a short duration at a high temperature or for a longer duration at a lower temperature. Moreover, based on the understanding of residence time distribution, it becomes clear that a narrow RTD is far more favorable than a broad distribution in reducing the opportunity for gel formation.

Recognizing that the polymer temperature is highest in the melt state, it becomes apparent that the melt residence time is an important parameter not to be overlooked. Described in the expression below, it’s noted that the residence time of the polymer melt is inversely related to extruder drag flow and linearly related to pressure flow. This leads to a reduction in residence time as extruder throughput is increased and an increase in residence time as extruder discharge pressure is raised.

This suggests that a high output screw design is preferable when attempting to avoid gel formation.   However, extruder throughput should be balanced against the potential for melt temperature generation and product output requirements so as not to result in an excessively low screw speed and long residence time. As well, successive polymer heat histories, even when below the resin decomposition temperature, can result in resin degradation and gel formation.   This is clearly demonstrated in the figure below of successive extrusions depicting nylon 6/6 at various temperatures (Figure 2).


Furthermore, it is well recognized that potential causes of gel formation include poor resin quality, poor shut-down procedures, excessively high temps, long heat up times, poor resin stabilization, the use of regrind, contaminated resin or concentrate incompatibility among others. All potential causes should be evaluated on their merits and reduced or eliminated whenever possible.

(Gels) How Do We Get Rid of Them?

If understanding the potential for gel formation is the first step, then determining how best to eliminate or reduce gels comes is a close second. Clearly the most obvious and beneficial approach in reducing gel formation is to address those contributions which lead to either an increase in residence time, an increase in melt temperature or an increase in both.   In terms of the compound, we should assure the material is adequately pre-dried per the manufacturers recommendations, evaluate the use of processing aides to reduce melt viscosity, check all material handling equipment for contamination or fines, consider a nitrogen hopper blanket to exclude atmospheric oxygen, and evaluate stabilizers which interfere with the degradation mechanism. In this context, phosphites have proven effective as processing stabilizers for high temperature applications of short duration where the polymer melt is in the presence of oxygen, while the incorporation of phenols have demonstrated their effectiveness in acting as long-term antioxidants.

Considering process improvements, appropriate start-up and shut-down procedures should be followed using only stable compounds, careful control of extruder operating set points should be implemented to minimize residence time and melt temperature, careful inspection, cleaning and maintenance of all melt train components should be considered, and efforts to increase the gel capture capability of the screen pack evaluated by increasing the screen pack mesh density.

The Screw Design Connection

As evidenced by the discussion thus far, the extruder screw contributes to gel formation by providing the opportunity for degradation due to high melt temperatures, potentially long residence times and processing which is carried out in the presence of oxygen. As a rule, extruder screws should be free from any nicks, scrapes, scratches or gouges and should include generous radii of the leading and trailing flight flank to prevent potential dead spots and the opportunity for degradation. As well, screw surfaces should be well polished to a finish of 16 RMS or better with the addition of chrome plating as an added insurance policy against melt stagnation. Moreover, screws should be periodically measured for OD screw wear to assure the screw diameter is within tolerance and thereby does not provide excessive shear stress or residence time to the polymer compound.

Furthermore, a screw profile which provides high output, low residence time, low shear rate and a narrow RTD is preferred in the battle against gel formation; provided, it is not oversized for production rate requirements. Over-sizing of the screw profile to achieve a higher throughput may result in grossly limiting the extruder screw speed based line speed requirements and result in an increase in residence time (albeit at potentially cooler melt temperatures). Moreover, as barrier screws have been shown to provide a favorable reduction in melt temperature, consideration should be given to meticulous purging procedures and screw cleaning to avoid degradation of the compound which may occur on the additional un-wiped surface perimeter of the leading and trailing edge of the secondary barrier flight. In addition, the barrier entrance geometry should assure continued conveyance of both the solid and melt fraction, while the barrier flight clearance is designed to permit continuous melt film removal.

Whether in addition to a barrier profile or as an adjunct to a general-purpose design, dispersive mixing elements have proven beneficial in reducing agglomerates and the average non-melt particle size. However, as with any screw element, all mixer channels should be streamlined and dead spots eliminated. Coinciding with the screw theory which considers a deep screw profile to provide cool melt temperatures at the expense of melt quality, a dispersive mixing device can be used to augment a mixing function which has been compromised due to the lower cumulative shear stress history of the deeper flighted screw.

This is accomplished by forcing the high viscosity compound through the tight tolerance of the mixer geometry (expression 2).

Die Design Contributions

When discussing die design, it should be noted that the largest contribution to gel formation is the long residence time resulting from improper head design, poor plating and maintenance techniques, excessive heat generation and stagnation. As a general rule, the die should consist of as few individual parts as is practical, and when possible, the melt should be supplied centrally to the die. All dead spots and corners should be eliminated and sudden or sharp transitions avoided. All radii should be 1/8” minimum along with a surface roughness <0.2lm for all polished and chrome plated surfaces. At all times, large cross sections should be avoided as they result in a decrease in fluid velocity and an increase in residence time. This implies that a low volume head design which minimizes flow lines is preferred, as “melt healing” requires an increase in die land length and residence time.

Tips and dies should be chosen based on product dimensions and process requirements including draw down ratio, draw balance and sizing ratio, with typical land length to gap ratios in the range of 10:1 to 20:1. Moreover, while long land lengths have many benefits such as improved shape definition and reduction of die swell, the penalty for a long land length is an increase in pressure drop and consequently residence time, thereby increasing the potential for degradation.

Often times land length may be the limiting factor as die pressure may become excessive and the need to truncate land length when running high viscosity compounds should be considered.

Taper angles in the converging region of tubing dies vary for self-centered versus adjustable dies with a range of 300’s to 400’s for the former and 80’s to 150’s for the latter. Land length and the angle of converging sections should not exceed the critical tensile deformation rate. Tensile stress has been shown to increase in the converging flow region and reaches a maximum at the narrow end of the taper.   When a polymer flows from a large to a small channel it forms a natural streamline angle; and when the taper angle of the tooling exceeds this angle, dead spots will form. This raises the potential for gel formation leading to a recommended taper angle, which is equal to, or less than the natural streamline angle of the polymer.

Conclusion

While it is not always clear where gels have originated, they can conceivably develop anywhere along the polymer lifecycle; from the moment of polymerization and resin storage, to extrusion processing and end use performance. Furthermore, melt degradation greatly affects polymer morphology, which in turn affects end-use properties and product performance. In approaching the problem of gel formation, no stone can be left unturned, and evaluation of polymer properties and stabilizers, resin production and storage, along with material handling and melt processing should be considered. Stabilizers and processing aides have proven particularly beneficial, and a thorough evaluation of all melt train components is mandatory. And finally, screw and die design parameters can assist in reducing gel formation when designed properly and considered early in the development of process constraints.

References:

“Tooling Design for Medical Tubing,” C. Rauwendaal, Plastic Equipment News, Mar 1993

“Extrusion Dies; Design and Engineering Computations,” W. Michaeli, Hanser Publishers, 1984

“Understanding Extrusion,” C. Rauwendaal, Hanser Publishers, 1988

“Plastics Materials and Processing,” B. Strong, Prentice Hall, 1996

“International Plastics Handbook,” H. Saechtling, Hanser, 1987

“A System Approach to Screw Design,” T. Black, Plastics Engineering, 1991

“Handbook of Plastics, Elastomers and Composites,” C. Harper, McGraw-Hill, 1996

“Modern Plastics Encyclopedia Handbook,” McGraw-Hill, 1994

“Plastics Chemistry and Technology,” W. Driver, Van Nostrand Reinhold, 1979

“Polymer Materials, Structure-Properties-Applications,” G. Ehrenstein, Hanser Publishers, 2001

“BASF Thermoplastic Urethane Processing Guide,” Internet Resource, www.basf.com

“Fundamentals of Polymer Processing,” S. Middleman, McGraw Hill, 1977

“Kayeness Rheology Seminar Materials,” J. Reilly, 1998

“The Macrogalleria,” Internet Resource, www.psrc.usm.edu

Tom Black