Micro-injection moulding for production of polymeric lab-on-a-chip – simulation, production and optimization

Currently micro injection moulding is an efficient replication technology that combines mass production with a low cost. Because of its increasing application in many fields nowadays one of the main challenges is the possibility to simulate a process at microscale. This work aims to exploit the potential of process simulation to improve the production of a lab­on­a­chip with two different materials: PMMA and COC. An initial benchmark based on previous production and datasheets allowed identifying suitable process parameters by varying four factors: Mould temperature, Melt temperature, Packing pressure and In­ jection velocity. The combinations of different values of these factors were used to set up the experimental design. Then, the effects of the process parameters on the Volumet­ ric shrinkage of the chip, caused by the cooling process, were quantified with a proper metrology analysis. Finally, the results were compared with the simulation. Based on the metrology analysis of produced chips and the simulations, COC showed better results in term of volumetric shrinkage, with the dimensions of the channel closer to the mold value. Has the least shrinkage values both in practice and simulation. For each material the full factorial design was also analyzed with a statistical software (Minitab®) making it possible to determine which of the four parameters under analysis contributes more to modifying the volumetric shrinkage. The accuracy of the simulations for COC compared to the injection molded parts showed deviation within 15%, whilst PMMA has larger deviations from the real values (around 25­30 %). Even though the simulations differ a bit from the experiments a common trend could be identified: in all the simulation the parts shrink less than the real one, hence the results can be used for further production keeping in mind that the simulations underestimate actual behaviour.