Towards process optimization of spar caps made of carbon fiber reinforced polyurethane composites used in wind turbine blades

In the renewable energy sector, production of wind electricity doubled between 2009 and 2013, and in 2016 wind energy accounted for 16% of the electricity generated by renewables. In this context, the demand for the wind turbines is increasingly growing and thereby the rotor blade diameter is also increased in order to raise the wind turbine capacity. This leads to the seeking of lighter materials used on the rotor blades, because with the increase of the blade size, the load on the blade goes up exponentially. As a potentially better alternative to the epoxy resin currently used on wind turbine blades, thermoset polyurethane (PU) has better load-bearing capacity in comparison with the equivalent weight of epoxy resin. In this study, we seek to investigate the applicability of using polyurethane (PU) in the production process of the wind turbine blade spar cap. The vacuum infusion (VARTM) method is commonly applied in regard to the large fiber reinforced composites because of the lower tooling costs. The goal of this study is to propose the optimum mold temperature condition during the PU infusion, thus fulfilling a minimized the processing time. To achieve this goal, this study is divided into three sections. The first section is regard to the cure kinetics model of the PU resin, while the second section is to propose the viscosity evolution, which paves the way for the simulation of the PU vacuum infusion process under different heating conditions. First and foremost, we derived cure kinetics model of the PU. The DSC (Differential Scanning Calorimetry) instrument was used to monitor the exothermic heat generation, which in turn translated to the degree of cure evolution. Moreover, the iso-conversional method was employed to calculate the activation energy dependence on the degree of cure. The PU reaction mechanism was exploited and the multi-reaction complex cure model was derived with the help of the non-linear regression on MATLAB. In addition, the PU viscosity model, as a function of degree of cure and temperature, was explored. We performed the rheology experiments on a rheometer to measure both gelation time as well as the viscosity variation under different heating conditions. The cure kinetics model derived earlier was used to predict the degree of cure in the viscosity model. As a result, the overall viscosity evolution estimated by the viscosity model achieves good correlation with the rheology experiments. Lastly, we conducted the vacuum infusion simulation by using RTM-worx software. Both cure kinetics model and viscosity model of the PU resin were employed in the simulation. The reinforcement preforms used was unidirectional carbon fiber SAERTEX 882gsm with a thickness of 38.015mm and length of 500 mm. The mold temperature was set at 30℃, 35℃, 40℃ and 70℃, while the results of infusion times are 13.29min, 12.28min, 11.43min and 8.27min respectively. This suggests that the infusion time decrease with the increase of the mold temperature. Furthermore, the cure degree of the PU during the infusion is estimated based on the cure kinetics model, we concluded that the extent of the cure is negligible during the vacuum infusion under the mold temperature studied. The post-filling simulation is then performed. The result showed that if one applies the critical temperature at 250℃, the laminate temperature will exceed this temperature for thick laminate curing process due to the excessive heat generation leading to the PU degradation. While for the criteria of 400℃, the PU application in thick laminate will be promising. By applying the practical mold temperature at around 90℃, the total processing time for thick laminate (7.86cm) is 460min, and fully cured laminate is achieved, and the maximum temperature of the laminate is reduced to 155℃ which is within the safe range.