Poly-paper: a sustainable material for packaging, based on recycled paper and recyclable with paper

Specimen preparation

Composites were melt blended in a corotating twin-screw extruder (Lab-Compounder KETSE 20/40D EC; Brabender, Duisburg, Germany). The extruder line was equipped with a side feeder (MT1-12; Brabender) to feed powder directly into the melt and a volumetric feeder (DRS28) to feed the pellet into the extruder barrel. The sequence of compounding was as follows: modified PVA pellets were fed through the input hopper with the volumetric feeder; finally the cellulose was dosed from the extruder opening side with the side feeder. The temperature pattern of the extruder was 190°C-190°C-195°C-195°C-190°C-190°C-180°C from input to output zones. The composites were pelletized from the extruded filament, to be processed by injection molding. Composites pellets were dried (24 hours at 50°C under vacuum) before injection molding. From all composites, pellet dog bone specimens with dimensions according to ASTM D638 were fabricated using a 12 mL microinjection molder (DSM Xplorer) at 190°C melt temperature and 60°C mold temperature with injection and holding pressure of 16 bar. The specimens were allowed to cool in the mold for 5 minutes before extraction. This allowed us to obtain homogeneous and strong postextrusion materials with cellulose fiber rates ranging from 30% to 50% (Fig. 1B, C). As shown, without fibers, the post-extrusion modified PVA only matrix had a thin texture (Fig. 1A), and became unsubstantial with 70% cellulose fibers (Fig. 1D).

The possibility to obtain filaments by extrusion was finally explored by checking, with positive results (data not shown), whether the material could be used to produce pellets (Fig. 2) in view of injection molding.

Experimental characterization techniques

Thermal stability of cellulose fibers was studied using a TGA-500 V6.7 instrument (TA Instrument), coupled with TA Instrument Explorer operating software. The analyses were performed under dynamic heating conditions, from 50°C to 800°C under nitrogen flow (60 ml/min), at a heating rate of 10°C/min, using about 2 mg of sample. Data recorded show the thermal behavior of cellulose fibers in terms of weight loss percentage increasing the test temperature.

The morphology of the dried cellulose fibers was observed by EVO Scanning Electron Microscope (Zeiss, Cambridge, UK) at room temperature. The fibers were stuck on the sample stub. The samples were gold sputtered up to a thickness of 20 nm by means of a Emitech K-550 sputter coater (Emitech, Ashford, Kent, UK). An accelerating voltage of 15 kV was used to collect the micrographs. The surface of the tensile-fractured specimens was also analysed after gold sputtering. The same observation conditions used for the dry fibers were used to analyze composite samples.

Tensile properties of the injection molded specimens were measured by using an Instron 5985 universal testing machine, equipped with a load cell of 10 kN in accordance to ASTM D638 standard. The tensile specimens had a length, width and thickness of narrow section of 165, 13 and 3.2 mm, respectively. These dimensions are in accordance with specimen Type I as reported in the ASTM D638 standard. Five specimens were tested for each composite, with a constant speed of 5 mm/min, while compliance correction was used. System control and data analysis were performed using Instron’s Blue Hill software.

The ability of the composite material (with 50% cellulose fiber) to melt by soaking with water was then evaluated by simulating maceration conditions. Briefly, tests were performed on specimens (40 × 40 × 5 mm), both at room temperature and at 45°C. Some specimens were superficially coated with a varnish film to check for possible water resistance of the material in view of specific applications.

Characterization of cellulose fibers

Thermal stability of the cellulose fibers was investigated by thermogravimetric analysis (TGA) under nitrogen flow. In Figure 3, the TGA result for dried cellulose fiber is reported. The TGA curve showed 3 thermal degradation steps ranging from 200°C to 800°C. In particular, the first degradation step at 267°C was due to the degradation of cellulosic substances, such as hemicellulose and cellulose. The second degradation (T = 383°C) of the decomposition was related to the degradation of noncellulosic materials in the fibers, while, the final degradation steps between 620°C and 800°C were due to minor components in the fibers. The residual char accounted for 19% of the initial weight. The thermal stability of the cellulose fibers up to 267°C ensured the processability of the composites up to 195°C which was the maximum selected temperature in the extruder for compounding.

The morphology of the cellulose fiber obtained by the paper recycling process was investigated with SEM (Fig. 4). The recycled paper appeared in a fibrillary form, with fibers ranging from 50 to 200 µm. The surface of the fibrils appeared rough probably due to the recycling method used.

Tensile properties of the composites

The tensile properties of the modified PVA/cellulose composites were compared with neat modified PVA. The stress–strain curves of the neat modified PVA and of the composites are reported in Figure 5. The stress–strain curves showed a necking extension for the neat modified PVA sample. The yield stress for the composites decreased slightly with increasing the cellulose fiber content. The tensile modulus showed sharp increases with tensile modulus varying from 3.57 to 5.19 and 7.17 GPa, with an increase of 45% and 101% for the composites with 30 wt% and 50 wt% of cellulose, respectively (Tab. I). This indicates that the stress would be expected to be transferred from the polymer matrix to the stronger cellulose fiber, indicating also a good interfacial adhesion. Similar results were obtained by Huda et al (6) with the addition of 30 wt% recycled newspaper fibers to PLA. In addition, the mechanical properties here obtained were in the range of, or even higher than, those reported by Graupner for PLA-reinforced composites for use in the automotive sector (10). In addition, the use of natural or recyclable constituents makes the composites studied interesting as environmentally friendly materials (11).

The tensile-fractured surface of the specimens considered was analysed by SEM (Fig. 6). Neat PVA (Fig. 6A) showed a rough surface, confirming the mechanical data for a ductile behavior, while the surface appeared less rough for the sample modified with 30 wt% cellulose fibers (Fig. 6B). The fibers were covered by matrix in both of the samples at 30 wt% (Fig. 6B) and 50 wt% (Fig. 6C, D). However, in the sample with 50 wt% of fibers, the higher fiber volume fraction resulted in a more complex surface topology.

Stability test

The stability tests showed that the composite material (Fig. 7) was completely dissolved within about 2 hours when uncoated (cellulose 50 wt% in Fig. 7) and coated with 2 different percentages (3.3% and 10%) of ketone-aldehyde resin. When the composite was coated with a varnish, either non-water-soluble or water-soluble, the coated composite specimens showed good stability up to the end of the test. The obtained results showed that if appropriately coated, the composite can be stable for a longer time, hence it can be used also in an environment with a high percentage of humidity.