Reduced water vapour sorption in cellulose nanocomposites with starch matrix

Abstract

The effects of microfibrillated cellulose nanofibers from wood on the moisture sorption kinetics (30% RH) of glycerol plasticized and pure high-amylopectin starch films were studied. The presence of a nanofiber network (70 wt% cellulose nanofibers) reduced the moisture uptake to half the value of the pure plasticized starch film. The swelling yielded a moisture concentration-dependent diffusivity. Quite surprisingly, the moisture diffusivity decreased rapidly with increasing nanofiber content and the diffusivity of the neat cellulose network was, in relative terms, very low. It was possible to describe the strong decrease in zero-concentration diffusivity with increasing cellulose nanofiber/matrix ratio, simply by assuming only geometrical blocking using the model due to Aris. The adjusted model parameters suggested a “simplified” composite structure with dense nanofiber layers oriented in the plane of the film. Still, also constraining effects on swelling from the high modulus/hydrogen bonding cellulose network and reduced amylopectin molecular mobility due to strong starch–cellulose molecular interactions were suggested to contribute to the reductions in moisture diffusivity.

Introduction

Nanocomposites, i.e. composites containing fillers that have at least one nanosized dimension, represent a new class of materials that have interesting penetrant diffusion properties [1], [2], [3]. The presence of nanoparticles can significantly enhance the gas barrier, as was shown for polyimide films containing only 2 wt% clay [2]. The improved barrier was explained by the geometrical blocking due to the wide impermeable clay platelets. Because of their small size, the surface-to-volume ratio of the nanoparticles is significantly greater than that of the corresponding microsized particles at the same mass content [1], [3]. Provided that strong particle–polymer molecular interactions exist, the smaller particles have a greater ability to bond to the adjacent polymer material, thereby reducing the chain segmental mobility and thus the penetrant diffusivity [3]. The presence of nanoparticles may also lead, in some cases, to a reduced chain packing adjacent to the filler surface and thereby an increase in the free volume. This will then yield faster solute diffusion through the interfacial particle–polymer layer [1]. Additional factors that are likely to be important for the diffusivity properties of nanocomposites are the interactions (including aggregation) between the nanoparticles.

The present paper reports the diffusion of moisture in pure and glycerol plasticized high-amylopectin starch, hereafter referred to as amylopectin or AP, reinforced with different amounts of microfibrillated cellulose nanofibers. This material has interesting mechanical properties [4] and mimics biological-plant structures in several ways. Processing is possible at room temperature in a water medium. The matrix and reinforcement phases are both polysaccharides, where the cellulose is nanostructured as in plants. The matrix phase is very soft with nearly viscous behavior and the mechanical material integrity derives from the cellulose nanofiber network. Previous studies on glycerol plasticized starch materials, reinforced by cellulose nanofibers [4], [5], have also shown favourable interactions between cellulose and the polymer matrix. Microfibrillated cellulose (MFC) nanofibers should not be mistaken for microfibrils. Cellulose microfibrils are the 3–10 nm thick fibrils formed during cellulose biosynthesis in higher plants [6], whereas the microfibrillated cellulose nanofibers in the present study consists of aggregates of microfibrils. The cellulose nanofibers are obtained by disintegration of the plant fibre cell wall and they typically have a width and length of, respectively, ca. 5–50 nm and several micrometers and they have a high specific surface area [7]. In this study, the source of nanofibers was bleached sulphite softwood cellulose pulp that was disintegrated by combining mild enzymatic hydrolysis with mechanical shearing and high-pressure homogenization [8]. The cellulose nanofiber is a highly attractive organic biodegradable reinforcement in polymer nanocomposites, due to its high aspect ratio, good inherent mechanical properties and its ability to form a network [4], [5]. The structure of the nanofibers is of interest. The polymer chains are in highly ordered extended chain conformation. The general structure of cellulose has been reviewed [9] and more recent studies based on neutron diffraction provide details of the crystal structure [10], [11]. The pulping process involves removal of non-cellulosic wood biopolymers at elevated temperature, and the structure of nanofibers from pulp has been studied [12]. Wood crystallites are very small with typical crystallite dimensions of 4 nm (lateral) by 20 nm (length). The cellulose aggregates making up the nanofibers have a typical lateral dimension of about 15 nm. A significant portion of the chains are in a paracrystalline extended chain conformation. X-ray diffraction data of the cellulose nanofiber in the present study show an estimated degree of crystallinity of 66% [13] although the distinction between crystalline and amorphous phase from the data is somewhat ambiguous. Disordered regions are preferentially at aggregate surfaces [14], and possibly also in localized amorphous regions along the chains [9]. Such regions are expected to be accessible to water in contrast to the interior of cellulose aggregates.

Native starch consists of amylose and amylopectin. Both polymers are composed of α-d-glucose units, but amylose is a linear molecule whereas amylopectin is an extensively branched macromolecule. In the present study, the starch is from potato tuber and contains predominantly amylopectin. The main disadvantage with starch materials is that they readily absorb water, which significantly deteriorates their mechanical performance [5].

Diffusion of water in cellulose nanofiber reinforced starch composites was reported in previous studies [5], [15], [16], [17]. In these studies the moisture diffusivity was considered to be constant, independent of moisture content. This is also assumed in the modelling of thermoset based fiber composites [18]. However, water vapour sorption in the present starch matrix systems necessitates evaluation of more complex models. In fact, Russo et al. [19] showed that the moisture diffusivity in high-amylose starch, blended with a small amount of a water soluble polyol (1–10%), increased as an exponential function of the water content. A moisture dependent diffusivity was therefore evaluated and implemented in the present study. The objective was to assess the effects of cellulose nanofibers on the moisture transport properties of amylopectin films. A strong reduction in moisture diffusivity is observed and two different theoretical models are used to predict the reduction in zero-concentration moisture diffusivity with increasing cellulose content.