By J. Coulter, P. Gao, A. Duhduh, A. Kundu
Manufacturing Science Laboratory Lehigh University, Bethlehem, PA, 18015
This research was focused on the effect of processing parameters on physical characteristics of poly-lactic acid (PLA) during vibration assisted injection molding (VAIM). In vibration assisted injection molding processes, the travel of the injecting screw is moved back and forth to create an oscillatory motion during the injection stage of the molding process. The frequency, duration and initiation point of the motion can be individually controlled. It was observed that VAIM based PLA products have higher total crystallinity than conventionally molded PLA products under identical conditions sans the vibration. Based on these results it was hypothesized that the vibration primarily affects the molecular arrangement of the polymer chains which in turn affects the nucleation density and thus the crystallinity. In addition, it was observed that the cycle time can be reduced by at least 25% when vibration was applied.
Vibration Assisted Injection Molding (VAIM)
In vibration assisted injection molding (VAIM), the standard injection molding process is augmented with a dynamic oscillatory motion of the injection screw. A schematic of this setup in illustrated in Figure 1. Instead of injecting the polymer melt directly into the mold cavity in a single motion, VAIM utilizes an additional computer system to control the hydraulic system to make the injection screw move in an oscillatory motion at specific frequencies, amplitudes, and durations as the melt is packed and solidified. Additionally, the molding parameters, such as injection temperatures, injection and packing pressure, mold temperature can be separately controlled by the injection molding machine.
VAIM has been reported to promote the molecular orientation of the polymer chain in molded samples [1]. The vibratory motion aids in untangling and orienting the molecular chains as it cools down to temperatures just above its glass transition temperature (Tg). Continuing this as the temperature dips below Tg, ensures that the induced chain orientation is maintained in the solidified polymer, and chain relaxation does not occur. Both the vibration contributed by the oscillatory shearing action (in the melt stage) and to a lesser extent, the hydrostatic pressure (as the polymer solidifies) play a role in altering certain aspects of the molded product’s material as well as physical properties. In traditional injection molding, this is not observed since solidification under a quiescent condition would have resulted in a random coil arrangement. A polarized light optical microscopy technique was utilized to investigate the molecular orientation in injection molded PLA parts and the results are shown in Figure 2 [2]. The dark colored (blue and grey in Figure 2) represents the low stress sections while the light colored(red and orange) represents the high stress sections. It can be observed that the sample with vibration have more cycles of change in stresses along the melt flow direction, which represents highly molecular orientation.
Material Selection
Polylactide or Poly(lactic acid) (PLA) is a thermoplastic polymer derived from renewable resources. This is in contrast to common commercial grade thermoplastics, such as those from the polyethylene family, and isotactic polypropylene, that are derived from nonrenewable petroleum reserves. PLA can be manufactured with different properties from primarily amorphous to largely crystalline. This is achieved by varying the relative amount of the isomers and additives. PLA has two stereoisomers namely the L-lactic and the D-lactic acid. Three forms of PLA are available commercially: pure L-lactide, pure D-lactide, and a mix of L and D-lactide. Relatively pure L-feed and D-feed PLA is referred to as PLLA and PDLA respectively [3]. Common commercial grade PLA (that requires high crystalline content) contains a majority L-feed mixture, and a minimum of 1-2% D content, whereas those requiring an amorphous product may contain up to 20% D content.
Injection molding is the primary fabrication method for producing PLA parts. PLA could be a semi crystalline polymer. The crystallinity and other physical properties of these materials can vary with the processing conditions. Presence of certain additives such as nucleants [4], and accelerants, impact modifiers, and mold flow agents can affect the crystallinity and the properties as well.. A drawback of PLA is its low Tg (approximately 60-75 °C) resulting in inferior thermal resistance to other industrial polymers. The thermal resistance can be improved by increasing the crystallinity in these materials. Higher crystallinity is also desirable for stiffness, strength, and other properties of the fabricated parts [4]. However, the crystallization rate of PLA is slow, thus obtaining sufficiently high crystallinity within reasonable manufacturing times is difficult. The typical mold-closed time is ~ 40 – 60 seconds at optimal mold temperature settings (minimum of 85 °C, and preferably above 105 °C). The high mold temperature required also increases the cost of production. Lowering the cycle time and the prescribed mold temperature could lead to parts sticking to the mold, or bend and warp upon ejection. Subsequent annealing to increase the crystallinity of molded products is a viable strategy but would add an additional step (time and cost) to the overall manufacturing process. The crystallization of PLA is of great interest to the research community and studies have focused on influences of axial stretching/strain[5], melt drawing/extrusion[6], annealing[7], and crystallization temperature[8] on PLA crystallization. This research was focused on the effect of processing conditions on the crystallinity during vibration assisted injection molding.
Sample Preparation and Characterization Techniques
An Ingeo TM PLA 2500HP polymer from NatureWorks was utilized in this study. To enhance the properties of the final products, we added 10% proprietary blend of additives that included a nucleating agent, accelerant, impact modifier, and a mold flow agent. PLA pellets were dried in a 40°C oven for 8 hours to reduce the amount of absorbed moisture [9].
The samples were molded into ASTM D638-01 dog bone specimens using a Nissei PS40E5A injection molding machine. Injection molding was performed with and without vibration using the modified setup presented in Figure 1. During the processing the melt was dynamically oscillated under 4 different conditions: i. without oscillation (0 Hz), ii. 1 Hz, iii. 4 Hz, iv. 8Hz, and v. 30Hz. Melt oscillation was commenced immediately as the melt was injected and packed into the cavity for the duration of 10 seconds. To ensure a positive pressure on the melt during oscillation, a 55:45 ratio of time was set for the compression stage relative to the decompression stage of each vibration cycle. The temperature of the material in the injection screw was controlled at 4 positions along the screw. The temperature at the nozzle, front, middle and rear sections of screw were maintained at 215°C, 205°C, 195°C and 175°C respectively. Injection pressures in all cases are set at 64 and 36 MPa respectively for injection period and packing period. The injection time for each sample was 15 seconds while the packing/cooling time was 20 seconds. To enhance the growth of crystal, a high temperature mold of 85°C was utilized for manufacturing all samples.
A cross sectional slice at the mid-section of the dog bone sample, weighing approximately 8-10 mg, was extracted from each sample and was utilized to investigate the crystallinity in the parts. A differential scanning calorimetry (DSC) technique using Q2000 DSC device from TA Instruments was utilized for this purpose. Only the first heating scan, from 25°C to 240°C at 10°C /min, was collected to investigate the effect of the vibrational frequency on the crystallinity in the molded material. The degree of crystallinity (XC) was calculated using Equation 1.
(1)
Where ∇Hm is the meltin enthalpy [J/g], ∇HC is the cold crystallization enthalpy [J/g], and 93 J/g is the melting enthalpy of a PLA crystal of infinite size
Results and Discussion
Table I illustrates the effect of vibration and cycle time on the product quality. The cycle time is the total time including injection and cooling times required to produce a dog bone part. A minimum cycle time of 35 seconds was required for fabricating a high-quality part without vibration. But with 1 Hz vibration, a cycle time of 21 seconds was adequate to obtain an optimal product. In contrast, dog bones fabricated without any vibration with 21 seconds cycle time were soft when demolded and deformed on further cooling. There was no visible change in the product quality when the vibration frequency was increased to higher values [10].
A series of DSC tests were performed on the dog bone samples to gain insight on the crystallinity. DSC scan profiles of injection molded samples processed under different conditions in the temperature region from 120°C to 190°C are presented in Figure 3. A DSC plot of the as received material has also been included for comparison. Plot for PLA pallets and conventional vibration samples were similar. But even with 1 Hz of vibration, the material changed significantly. The total area under the peak increased indicating a higher crystallinity in the material. An interesting observation was made when comparing the data for 1Hz and 8Hz samples. Increasing the frequency from 1 Hz to 8 Hz increases the crystallinity of the final product, but the melting point shifted to lower temperatures. It was hypothesized that vibration made the crystallization process faster by introducing shear, but the crystal structures were less perfect with higher frequency. This would indicate that the sample was subjected to higher shear rate. Hence, a lower energy was required for melting those less perfect crystal structures thus decreasing the melting temperature. The sample fabricated at 30Hz appeared to be significantly different from the other samples. The melting point shifted to higher temperatures, the melting curve is wide. A second melting peak was observed at 165°C. It was associated with the melting of the α´ phase, one of the several phases of PLA. PLA has been reported to crystallize in four distinct phases, α, α´, β and γ. These structures are different in their sizes and enthalpy. The most stable crystal structure in PLA products is the α structure, and the α´ to α conversion is irreversible. The microstructure appeared to be significantly different than the rest for the 30Hz sample.
Table II shows the percentage of crystallinity for each sample.
parameters.
Samples under the same vibration frequency but with different cycle times were also tested to understand the effect of cycle time. Figure 4 and Figure 5 show the DSC scan profiles of 1Hz VAIM samples and 30Hz VAIM samples, the degree of crystallization is presented in Table III. It is observed that the crystallinity increased from 42% for conventional injection molding to 52% utilizing 1Hz VAIM and with the same amount of cycle time. The crystallinity increased further to 54% even when the cycle time is reduced to 21 seconds instead of 35 seconds utilizing 1Hz VAIM. When VAIM was performed at 30Hz, 66% crystallinity can be achieved with 35 seconds cycle time and the crystallinity decreased to 60% when 21 seconds cycle time was employed. But in general, even with reduced cycle time, higher crystallinity can be achieved utilizing 1Hz and 30Hz VAIM.
Table III indicates that under VAIM, samples would have higher crystallinity compared with traditional injection molding samples even with reduced cycle time. This would result in reducing cycle time at the same time, keeping the same or even enhance the properties of the final products.
Conclusion
The primary conclusions of this research are as follows:
1. VAIM appeared to affect the structure of molded PLA part and resulting in the increase of crystallinity as inferred from the DSC results. Even with 1Hz VAIM, a 24% increase can be observed on degree of crystallinity. If frequency was increased to 30Hz, a further 27% increase in crystallinity was achieved compared to 1Hz VAIM samples. It is also noted that the microstructures of the crystals also changed with different vibration frequencies.
2. Experiments show that the cycle time to manufacture a dog bone sample can be reduced from 35 seconds to 21 seconds using VAIM without inducing any negative impacts to overall product quality. This could lead to a large reduction in production time and enhanced production speeds during mass-scale production of the PLA parts.
References
[1] D. C. Angstadt and J. P. Coulter, “Investigation of melt manipulation phenomena during injection molding via in situ birefringence observation,” Polym. Eng. Sci., vol. 46, no. 12, pp. 1691–1697, 2006.
[2] Q. Li, “An Investigation of the Native and Manipulated Effects of Shear Imbalanced Melt Flows during Molding Processes,” 2017.
[3] S. Saeidlou, M. A. Huneault, H. Li, and C. B. Park, “Poly(lactic acid) crystallization,” Prog. Polym. Sci., vol. 37, no. 12, pp. 1657–1677, 2012.
[4 E. C. L. Angela M. Harris, “Improving Mechanical Performance of Injection Molded PLA by Controlling Crystallinity,” Polymer (Guildf)., no. June, 2008.
[5] R. Zhuo et al., “Structural evolution of poly(lactic acid) upon uniaxial stretching investigated by in situ infrared spectroscopy,” Vib. Spectrosc., vol. 86, pp. 262–269, 2016.
[6] B. Hu et al., “Influence of melt-draw ratio on the structure and properties of poly(vinylidiene fluoride) cast film,” J. Plast. Film Sheeting, vol. 30, no. 3, pp. 300–313, 2014.
[7] J. Zhang et al., “Crystal modifications and thermal behavior of poly(L-lactic acid) revealed by infrared spectroscopy,” Macromolecules, vol. 38, no. 19, pp. 8012–8021, 2005.
[8] P. J. A. Lohmeijer, J. G. P. Goossens, and G. W. M. Peters, “Quiescent crystallization of poly(lactic acid) studied by optical microscopy and light-scattering techniques,” J. Appl. Polym. Sci., vol. 134, no. 10, pp. 2–10, 2017.
[9] “Ingeo TM Biopolymer 2500HP Technical Data Sheet,” no. 4, pp. 1–5.
[10] P. Gao, A. Duhduh, A. Kundu, and J. P. Coulter, “Vibration Assisted Injection Molding for PLA with Enhanced Mechanical Properties and Reduced Cycle Time” TechConnect Briefs 2019, pp. 446-449, 2019.
About the Author
John P. Coulter is a professor in the Department of Mechanical Engineering and Mechanics at Lehigh University who also serves as the Senior Associate Dean for Research for the P.C. Rossin College of Engineering and Applied Science. On two separate occasions, the latest taking place from 2015 through 2016, Dr. Coulter has served as the Interim Dean of the College of Engineering and Applied Science at Lehigh. In that capacity he was responsible for the operation, coordination, financial management, advancement, and oversight of all engineering related activities at the University. He holds Bachelor of Science and Master of Science degrees in mechanical and aerospace engineering from the University of Delaware, and completed his doctoral studies in mechanical engineering at Delaware in 1987. His graduate studies were supported by a prestigious and nationally competitive DoD Fellowship awarded through the Office of Naval Research. Coulter has 29 years of teaching and research experience at Lehigh, as well as several years of industrial experience with Lord Corporation, a multi-national company specializing in materials and devices for vibration and acoustic control. During his time at Lehigh, he has taught several thousand undergraduate students, mentored 25 doctoral students and 65 master’s students, and won several awards for curriculum innovation that incorporates K-12 students from diverse backgrounds. His accomplishments at Lehigh have been recognized through continuous federal, state, and industrial research support as well as numerous awards for teaching and research including a prestigious NSF National Young Investigator (i.e. CAREER) award, Lehigh’s first-ever NSF Presidential Faculty Fellow (i.e. PECASE) award, a Future Technology Award from the Society of Plastics Engineers, and two Innovative Curriculum Awards from the American Society of Mechanical Engineers.
As an individual scholar, he has more than 200 publications and several patents in the areas of material processing, manufacturing science, and intelligent materials/systems. He has led funded research projects involving faculty Co-PIs from all eight departments in the college of engineering, and generated more than $9 million in individual research and teaching grants along with intellectual property donations to Lehigh valued at approximately $100 million. He has also organized a number of international conferences, workshops and symposia, served on three international journal editorial boards, and currently serves on the Board of the ASEE Engineering Research Council.
