New developments in thermoplastic foam injection molding

Lightweight parts, energy-saving processing or combining several process steps into one. It can all be summarized under one buzzword: sustainability. A technology that is highly promising is thermoplastic foam injection molding. Although in use for decades, the process offers great potential thanks to continuous development. The Institute of Polymer Injection Molding and Process Automation (IPIM) at the Johannes Kepler University (JKU) in Linz, Austria, is joining forces with ENGEL AUSTRIA and the Competence Center CHASE in Linz to leverage this potential.

Thermoplastic foam injection molding (TFIM) offered by ENGEL AUSTRIA under the name foammelt, is one of the oldest special injection molding processes. The commercialization, principally of chemical TFIM processes at first, began in the 1950s. Small amounts of baking powder were added to the melt to avoid sink marks in the part [1]. Chemical blowing agents grew in importance and led to the first series production of foamed parts in the 1970s [2].

The use of supercritical fluids (mainly nitrogen and carbon dioxide) in physical TFIM process-es, such as MuCell by Trexel today permits the manufacturer of microcellular foams with pore diameters below 100 µm or cell densities of over 109 cells/cm3. This special morphology offers advantages particularly for impact applications since the fine cells act as crack arrestors [3]. Although the foaming of polyolefins during injection molding has undergone many years of development, many questions remain open. Current developments are aimed at finding alter-natives to gas introduction, investigating the conditions in the plasticizing unit, as well as the deployment of Industry 4.0 technologies.

Minimized warpage, improved shape accuracy

Foamed parts have a three-layer structure. A porous core is surrounded by two unfoamed surface layers. This structure offers advantages since the material and resource consumption are reduced and a lower density is made possible. Classical density reductions in the TFIM process are around 10%, as new developments in the tool and process technology permit (local) density reductions of up to 50% [4], which underlines the lightweight construction potential of this technology. Another advantage for sustainability is the recyclability of the lightweight parts since the sandwich structure consists of a single material.

Regarding the mechanical performance of foamed parts, it is important to note the higher geometrical moment of inertia compared to unfoamed parts and therefore the increased specific flexural strength, as a result of the greater distance of the unfoamed outer layer from the neutral fibers. The crack-stopping properties improve the impact behavior of foamed parts [5]. Besides mechanical improvements, these parts also offer inherent functions such as acoustic or thermal insulation without further processing steps.

Technically, the homogeneous gas pressure prevailing during forming of the part minimizes warpage and therefore improves the shape accuracy. In addition, the gas acts as a plasticizer during processing. Depending on the gas and concentration, the viscosity of the material is reduced by up to 50%, which under certain circumstances permits smaller machines to be used. Primarily, this has an environmental benefit, which makes foam injection molding even more popular, particularly in regards to current developments in energy policy.

Utilize potential with high efficiency

There are also challenges that stand in the way of TFIM break through as a mass production technology. These include the increased machine and technical complexity due to the gas introduction into the polymer melt, the tool movements during forming of the part (core back technology) or the complex control of the pressures during recovery. On the other hand, the surface quality of the foamed parts is also important. Due to the pressure drop towards the flow front, the first bubbles already occur in this area during injection. The fountain flow leads to the bubbles being conveyed against the cold cavity wall, where they are sheared – this effect becomes visible as silver streaks. Methods for avoiding surface defects, such as variotherm mold temperature control or gas counterpressure, are associated with higher costs and greater technical effort.

In order to use the full potential of foam injection molding with high efficiency, it is necessary to understand the fundamentals of the technology even better – especially when it comes to specific applications - and to develop and test new process technologies. This is what the IPIM, the Competence Center Chase and ENGEL are working on together.

One of the focus areas of the joint development is the question of how the gas is introduced into the polymer melt and dissolved in it. The gas serving as blowing agent is either added physically in pure form or produced chemically directly in the polymer melt as the thermal de-composition product of a masterbatch. To obtain a homogeneous, single-phase solution, the gas must be mixed into the melt in a very short time. ENGEL has therefore developed a special plasticizing screw that meets the requirements of physical foam injection molding (Fig. 1).

The PFS (physical foaming screw) has a shear section, which homogenizes the polymer melt very effectively before the gas is introduced, with a positive effect on uniform gas distribution. In addition, the screw geometry permits the second non-return valve to be eliminated. In combination with execution in a new material, which offers an improved abrasion and corrosion resistance, the special geometry increases both the productivity and the durability of the screw.

New method for determining the solubility limit

The amount of gas that a polymer melt can absorb under certain process conditions is described by the solubility. Many methods exist for describing solubilities under static conditions. The prevailing method is to use so-called magnetic suspension balances, which have a pressure chamber for gas injection in combination with a magnetically coupled balance located outside the pressure chamber.

It can be questioned how relevant such statically determined data is for the highly dynamic foam injection molding processes, since any impact of movement, especially shear, is neglected. In addition, static solubility measurements have measurement times in the range of hours, while injection molding processes only permit seconds up to a few minutes at most to absorb the gas into the polymer melt. The scientific literature does not clear on the influence of shear on the gas solubility. While some authors assume that shear motions do not influence the solubility [6], others report an increase of solubility of up to 40% [7].

Because of this discrepancy, the IPIM has developed a new measurement method that describes the gas solubility under dynamic conditions - inline during the injection molding process. The starting point for the theoretical preliminary work was the Sanchez-Lacombe equation of state (1) for describing the pvT behavior [8] and consequently the bulk modulus K of polymer-gas mixtures (2):

With increasing gas content in the melt, there is a reduction in the bulk modulus of the mixture (Fig. 2).

However, a possible solubility limit – the static solubility of nitrogen in polypropylene is 2-4%, depending on the pressure and temperature [9] – cannot be described.

For this reason, tests were performed on an injection molding machine. The bulk modulus characterizes the volume changes as a result of a pressure rise. This state can also be produced in front of the screw tip by leaving a shut-off nozzle closed at the beginning of the injection process in order to compress the material. The pressure rise and volume contraction can be determined by machine sensors and then evaluated.

The assumption it described as the following: Gases have a bulk modulus that is lower than that of polymer melts by about a factor of 10. As long as the gas can be dissolved in the melt, it moderately reduces the bulk modulus of the mixture. Once the solubility limit is exceeded, the gas forms a separate – and far more compressible – phase, which should cause the bulk modulus of the mixture to drop significantly. It was therefore expected that there would be a deflection in the compression module when the solubility limit was reached.

Measurements on the machine confirmed this behavior (Fig. 3). For validation purposes: interruptions in the ultrasound signal at high gas contents are an indication of undissolved gas bubbles. A new, very process-related measurement method for determining the solubility limit based on the compression behavior was found.

The latest findings based on this measurement method falsify a widespread misconception. In the course of measuring the solubility limit with nitrogen and carbon dioxide on the injection molding machine, it could be shown that the gases are not completely dissolved in the polymer melt, but for the most part are only finely divided. In the very short interaction time be-tween melt and gas mentioned above, the function of the intensive shear is to produce a polymer-gas mixture that is as homogeneous as possible [10].

By means of the bulk-modulus method, not only material characterizations but also process optimizations are possible. From an economical point of view, it is particularly interesting to automatically reduce the backpressure to the technically necessary minimum, which significantly reduces the wear and energy consumption and improves the conveying behavior. This development is based on the circumstance that gases do not dissolve in the material when the pressures are too low, and consequently form a separate phase. When the bulk modulus is plotted against backpressure with constant gas loading, a distinct bend in the curve is expected again, which is indeed seen in the experiment. This method is also confirmed by ultra-sound measurements (Fig. 4). The bulk modulus has therefore proved to be an important tool for process optimization and for gaining a better understanding of the process in foam technology.

In the future, the injection molding machine will be able to automatically determine and set the minimum necessary backpressure. Intelligent assistant systems are already an important topic in thermoplastic foam injection molding. “iQ weight control,” successfully used in many chemical and physical foammelt and MuCell foam injection molding processes, achieves bet-ter process stability and a constant part weight. The software from ENGEL’s “inject 4.0” port-folio adjusts the injection profile and changeover point to the current conditions shot by shot during injection molding, and thereby fully automatically compensates external influences such as batch fluctuations and changes of environmental conditions in real time.

Foaming with Artificial Intelligence

Artificial Intelligence (AI) opens up even greater potential for optimizing foam processes. Specifically, the three development partners are working on the use of convolutional neural networks for optimizing part surfaces in TFIM processes. The goal is to find optimized process settings for unknown processes without having to perform complicated test series. In its present form, the convolutional neural network (CNN) developed and trained at the IPIM (Fig. 5) can assign process settings to the parts and thereby make recommendations for parameter adjustments. In future, the scanning of the part will be automated using an autonomous optimization mechanism. In correlation with the setting parameters, the system will automatically correct for deviations from the intended process.

The wide variety of developments of the process, the software and the machine are important steps on the way to applying state-of-the-art foam technology that also unlocks its full potential concerning sustainability. In this regard, TFIM is less and less to be considered as a special process.

Thermoplastic foam injection molding since 1914

Thermoplastic foam injection molding is based on methods that were originally developed out-side injection molding. The first foamed polymer-based products came onto the market in 1914 [11] and were made of natural rubber. Ammonium carbonate was widely used, which, when an acid was added, released the gases that enabled foaming. In the following two decades the two processes still used today were developed: Dunlop and Talalay. Dunlop is based on the physical mixing of air into the polymer melt together with a gelling agent (often sodium hexafluorosilicate). Talalay uses oxygen as a decomposition product of hydrogen peroxide [12]. It is comparable with modern chemical foam injection processes regarding both the foaming mechanism and the discontinuous process control.

Foamed polystyrene represents a milestone. The first patent, by Carl Munters and John Gudbrand Tandberg, dates from 1932 [13]. In 1947, the Dow Chemical Company followed suit with an extrusion process. A mixture of polystyrene, nucleating agent and volatile liquid blow-ing agents permitted the manufacture of larger foamed parts. The material is now protected under the name Styrofoam. Just four years later, BASF developed a process for manufacturing expandable polystyrene particles (EPS), in which the blowing agent is added already during polymerization.

The foaming of polyolefins took another decade. Here, too, the gases were added to the polymer melt at the beginning. In 1941, Frederick L. Johnston patented two different processes [14]. In the first, nitrogen is dissolved in a polyethylene melt in a pressure chamber while in the second, thermal decomposition of a blowing agent is used. Both processes already have a strong similarity to the physical and chemical processes that are used today in injection molding, although another four years would pass until carbon dioxide was used [15].

The Authors

Dr. Clemens Kastner is a postdoc at the Institute of Polymer Injection Molding and Process Automation (IPIM) at the Johannes Kepler University in Linz, Austria, and project manager at the Competence Center Chase in Linz; [email protected]

Dipl.-Ing. Wolfgang Kienzl is Product Manager Technologies at ENGEL AUSTRIA GmbH in Schwertberg, Austria; [email protected]

Dipl.-Ing. Eva Kobler is a research assistant at the IPIM of Johannes Kepler University and in the Competence Center Chase in Linz; [email protected]

Univ.-Prof. Dr. Georg Steinbichler is chairman of the IPIM and head of the I4.0 pilot LIT Factory in the field of plastics technology at Johannes Kepler University in Linz; [email protected]

References

[1]. Altstädt, V. und Mantey, A.: Thermoplast-Schaumspritzgießen. München: Hanser, 2011

[2] Handschke, A., und Mitzler, J. (2012). Physikalisches Schäumen ganz leicht gemacht. Kunststoffe, 10, 151-156.

[3] Bledzki, A.K., Rohleder M., Kirschling H., und Chate A. (2010). Correlation Between Morphology and Notched Impact Strength of Microcellular Foamed Polycarbonate. Journal of Cellular Plastics, 46(5), 415-440.

[4] Heim H.-P., und Tromm M. (2016). Injection molded components with functionally graded foam structures – Procedure and essential results. Journal of Cellular Plastics, 52(3), 299-319.

[5] Bureau, M. N. und Kumar, V. (2006). Fracture Toughness of High Density Polycarbonate Microcellular Foams. Journal of Cellular Plastics, 42(3), 229-240.

[6] Lee, S. T. (2004). Introduction: Polymeric Foams, Mechanisms, and Materials. Polymeric Foams: Mechanisms and Materials, CRC-Press, Boca Raton, Florida, USA, Kap. 1, S. 11-27.

[7] Chen, L., Sheth, H. und Kim, R. (2001). Gas absorption with filled polymer systems. Polymer Engineering & Science, 41, 990-997.

[8] Sanchez, I., und Lacombe, R. H. (1978). Statistical Thermodynamics of Polymer Solutions. Macromolecules, 11, 1145-1156.

[9] Li, G. et al. (2007). “Solubility Measurements of N2 and CO2 in Polypropylene and Ethene/Octene Copolymer”. In: Journal of Applied Polymer Science 103, 2945–2953.

[10] Kastner, C., und Steinbichler, G. (2020). Theoretical Background and Automation Approach for a Novel Measurement Method for Determining Dynamic Solubility Limits of Super-critical Fluids in Injection Foam Molding. Polymer Engineering & Science, 60, 330-340.

[11] Frisch, K.C.: “History of Science and Technology of Polymeric Foams”. In: Journal of Macromolecular Science, Part A: Chemistry 15.6 (1981), 1089–1112.

[12] Joseph, R.: Practical Guide to Latex Technology. Smithers Rapra, 2013.

[13] Munters, C.G. und Gudbrand, T.J.: “Heat insulation”. Pat. US2023204A. 1932.

[14] Johnston, F.L. (du Pont): “Synthetic spongy material”. Pat. US2256483A. 1941. [15] Hamilton, G.M. (Callenders Cable & Const Co.): “Improvements in the manufacture of cellular polymerised ethylene”. Pat. GB573306A. 1945.