The water vapour permeability of low-density polyethylene (LDPE) films containing reactive fillers

Abstract

The food packaging industry is moving away from commodity plastics to sustainable plastics. However, these sustainable plastics have poor permeability and mechanical properties. A standard practice to extend the shelf life of goods packed in plastic is to use fillers and multi-layered films. Three strategies are available for reducing water vapour permeability. This includes the use of an impermeable polymer, the addition of an impermeable flake-like filler or a reactive filler. This study focused on the concept of incorporating reactive fillers with or without an impermeable filler in a polymer to reduce water vapour permeability. The reactive fillers considered were magnesium oxide (MgO), calcium oxide (CaO) and magnesium aluminium layered double oxide (MgAl-LDO). The metal oxides were obtained through a 1 h calcination process. Talc was selected as the impermeable flake-like filler because of its compatibility with the non-polar polymer, linear low-density polyethylene (LDPE). A feasible calcination temperature range, for each filler, was determined from thermogravimetric analysis (TGA). The calcination conditions, i.e., temperature and time, affect the reactivity of metal oxides. In principle it is primarily linked to the specific surface area. Therefore, the Brunauer-Emmett-Teller (BET) specific surface area and the reactivity (by isothermal heat flow calorimetry) were determined for all calcined samples. It was found that the specific surface area, for MgO and MgAl-LDO first increased and then decreased with calcination temperature. However, the opposite held for CaO. Isothermal heat flow calorimetry showed that CaO was highly reactive compared to MgO and MgAl-LDO. This was true irrespective of the calcination temperature. In the end the calcination temperature was selected using the maximum specific surface area rather than the reactivity in isothermal heat flow calorimetry. They were selected to be 900 ⁰C, 650 ⁰C and 475 ⁰C for CaO, magnesium aluminium layered double hydroxide (MgAl-LDH) and magnesium hydroxide [Mg(OH)2], respectively. The reactive fillers, obtained under these conditions, were used in the water vapour permeability experiments. CaO underwent a one-step hydration reaction that was autocatalytic in nature. MgO and MgAl-LDO underwent a two-step hydration reaction associated with dissolution and precipitation. The enthalpies of the dehydration were determined by differential scanning calorimetry (DSC). They amounted to 1.53 MJ·kg–1, 1.48 MJ·kg–1 and 1.80 MJ·kg–1 for calcium hydroxide [Ca(OH)2], Mg(OH)2 and MgAl-LDH respectively. Masterbatches, containing 50 wt.% filler, were prepared by twin-screw compounding. Trilayer films were made using a film blowing process. The inner-layer filler content ranged between 5 and 20 wt.% filler loading. The variability of the measured permeability values was very high. As a result, statistically meaningful conclusions related to differences between the individual films were not possible. Nevertheless, it was noted that a higher permeability relative to LDPE was recorded for the reactive fillers. A maximum increase of 15 % was found for the 40 wt.% MgAl-LDO film. Lower permeability values were measured for films containing less of these reactive fillers. The films containing talc did show a lowered permeability. The same was true for films which contained talc in combination with MgO. Α maximum reduction of 35 % was found for the film containing 20 wt.% talc in the inner layer. The hydration kinetics of the fillers in films submerged in either water or a saturated sodium bicarbonate solution were investigated. The results suggest that CaO is far too reactive to be used in water vapour permeability applications. It already partially converted to Ca(OH)2 probably even before or during the film blowing process. Complete conversion in the sodium bicarbonate solution was not reached. This may indicate that the calcium carbonate formed a dense cover around a core containing Ca(OH)2. This possibly presented an impenetrable diffusion barrier that prevented further conversion of Ca(OH)2. The high variability in the results of the Young’s modulus and tensile strength, precludes making statistically significant deductions. However, it seems that Young’s modulus increased with increasing filler loading due to the reinforcement effect whereas filler loading did not greatly impact the tensile strength.