Advances in materials and processing refinements make new medical device designs possible and add to the toolboxes of competent design engineers. This article explores key factors of the multilumen extrusion process as they are used by processors and engineers specifying the development of new medical products.
The development of new medical products is both a time consuming and expensive process that is often extended by a lack of information, which, in hindsight, could have been utilized to reach the solution sooner. This hindsight is essentially what is known as experience and while there may be no known substitute for direct experience, there is much to be gained by learning from other’s experience.
The multilumen extrusion process is one that depends heavily upon experience. This isn’t because the process is too complex to be understood scientifically or that the process can’t be controlled by automation equipment and closed-loop feedback control systems (because it is). Control equipment and systems are used commonly by many manufacturers, including computational fluid dynamics in the design of tooling. In addition, an in-line ultrasonic wall measurement and laser micrometer may be used, providing closed loop feedback control to operate the puller, air controllers, and screw automatically and continuously, making adjustments according to measured output. Regardless, solving product development challenges in the area of multilumen medical extrusion remains an iterative approach, always taking place within a finite amount of time and budget. As such, the extrusion processor’s experience producing similar designs from similar materials will be the best predictor of achieving technical success on budget and on schedule.
The extrusion process is essentially a thermodynamic system with input variables producing certain outputs. As with other thermodynamic systems, a particular output can be obtained through a number of different input variable combinations. A state-space system defines the relationship between the current state, its input, and the future state of the system, which is being controlled by the current output. The closed-loop control system automatically adjusts based on the variable dependence of the system’s current state, its input, and the output (as a function of time). It is this aspect of thermodynamic systems not having a unique solution or, in other words, being able to “get the same result” a number of different ways that often leads those less familiar with the process to describe it as an art or pseudo-science. However, there are scientific approaches that are taken to isolate and understand the effect that particular variables have on the process output and this type of scientific experimentation is performed during extrusion validation activities.
Identifying the effect that individual input variables have on the output can be done by constructing and performing a design of experiments (DOE) to isolate and decouple the interdependence of the variables. A properly planned DOE enables process understanding and results in stable operating conditions that are repeatable and reliable day after day through anticipated raw material lot-to-lot variations and establishes consistency across multiple machine operators.
Since there are many variables that can affect the output of the multilumen extrusion process, a reduced multilevel factorial experimental design is often selected for economy when validating an extrusion process. Considering a full factorial design, the number of individual extrusion runs becomes prohibitively expensive. For example, a two level, five variable DOE fractional design could result in eight independent extrusion runs to investigate the variable interaction effects of barrel temperature, screw speed, air pressure, puller speed, and air gap distance, whereas in the full factorial design, the number of extrusion runs would be thirty two.
Of all the variables, the first consideration in the design of multilumen extrusions is the particular resin that will be extruded. The reason for the overlying importance of the material choice is that the multilumen extrusion tooling will direct the material to form the part geometry by forcing the molten material through and around the tooling set, which is known as a tip and die. With challenging multilumen designs where there are exceptionally tight orifices or sharp corners, some materials can accommodate being forced into tight specific geometry and others will not. Even within the class of peba resins, which are known to have favorable processing characteristics, certain durometers will not produce conforming parts without the use of specific tooling to obtain identical multilumen geometry from different durometers of the same resin type. During the design phase, it can be beneficial to initiate early discussions even prior to having formal drawings ready for quote with the extrusion processor who can provide guidance on dimensioning and tolerancing schemes based upon tool design and melt flow considerations.
An example of an area of application where this understanding becomes important is in the creation of ultra-thin wall, deflectable tip catheters. These types of catheters typically have wall thicknesses in the range of 0.008 to 0.012 in. depending upon the French size and may incorporate a PTFE liner for lubricity on the main internal diameter. Additionally, there may be one, two, three, or four lumens within the wall of the catheter, each of which may also be lined with PTFE and contain a pull wire to actuate the deflection. One approach to constructing such a design is to assemble the component extrusions that are multilumen liners of various durometers along with the other component materials and fuse the assembly together with the outer jacket extrusions in one or more post extrusion processing steps, also known as reflow or lamination. The resultant product tolerances can be as tight as ±0.001 in. on the ID and OD and, as such, are highly dependent upon holding close tolerances of the extruded multilumen liners and jackets in a variety of durometers.
Of the various material properties that are measured and can influence a material’s ability to “run well” with a given tool set are the melt flow index, the melt range, and the molecular weight. Since these material properties do not consider the shearing and polymer chain shortening that takes place during the extrusion process, the information is not predictive to the success of obtaining a desired result for extrusion designs with a given material. Success remains dependent upon the actual conditions of the run to be attempted.
The plastic pellets enter the feed throat of the extruder and pass through three independently controlled temperature zones while being sheared by a mixing screw, which melts the plastic. The material will have to conform to the tooling while in the dynamically flowing melted state, and maintain that shape as it is cooled and collected. The physical properties of material are degraded within the mixing that occurs within the barrel, and subsequently, the melt strength of the material is reduced. While the melt strength is not typically measured during the extrusion process, it is a property that is dependent upon the input process variables that include the mixing screw flute design (which has a direct relation to the degree of shear the material undergoes), screw speed, temperature of the zones, dwell time within the barrel, cross head design, area draw down ratio, and the presence or absence of a gear pump, among other factors. As the material degrades, the melt strength is reduced, which increases the tendency of the material to “break the line” while being strung up through the various downstream equipment of the water bath and puller. With the multilumen extrusion process, as the line is initially strung and subsequent adjustments are made to center the tip and orient the die, adjustments are made to the puller speed, the screw speed (or pressure control), and the individual air inputs to each of the lumens. With overly prolonged set up times or inputting process parameters that prematurely degrade material properties, insufficient melt strength can result. Certain materials provide more challenges to “dial in” the extrusion to the required dimensions, which can be frustrating, time consuming, and ultimately lead to the abandonment of the run to try again at a subsequent time after purging the material, cleaning the tooling, and starting over.
Due to the lack of published information on the processing conditions that extrusion processors use to produce specific geometries across multiple materials, the best way for a design engineer to specify a material for a particular multilumen design without his or her having direct extrusion experience is to discuss the dimensional and functional performance requirements of the tubing with the processor so that relevant processing experience can be relied upon to shorten the cost and timeline to produce unique solutions to product development initiatives.
For more information, visit www.dukeempirical.com.