6972
Cellulose (2020) 27:6961–6976
brittle so that the fractures formed at higher compres- sive strains caused a decay of the compression stress. At small strains, the addition of LBG or similar hydrocolloidal natural polysaccharide can improve the strength and stiffness of fibre foams made with SDS tremendously. Values that are similar to or improved compared with foaming with PVA can be achieved. Figure 11 shows the stress–strain behaviour for unrefined kraft and hemp with added fine components. Earlier without added fines, this furnish (TP3) led to significant deviation from the theoretical behaviour of Eqs. (1, 2) at strains below 40% (Fig. 8c). Curiously, with added fine components, the agreement was much better over the whole strain range of 10–50%, as shown in Fig. 11b–d. The same was observed with added cationic starch (Fig. 11a). As shown by Fig. 12, both the fine components and the starch made the network more uniform and tightly bonded, which might have brought the distribution of segment lengths closer to the assumed exponential distribution of the theoretical model. The same theoretical form was well maintained, even when the fitted factor r 0 of Eq. (1) increased significantly when finer particles were used. This increase was probably at least partly caused by the higher local density (Ketoja et al. 2019) achieved with more effective bonding. In Table 3, the scaled slope of stress–strain curve 1 r d r d at 50% compression varies in the range of 3.5–4.5. Thus, the deviation from the theoretical value 3.8 is rather small. Moreover, the experimental slope is not correlated with material density (also compare with Table 2). Similar quantitative agreement was observed with a broader range of densities earlier (Ketoja et al. 2019). A correlation with material
LBG (TP11) with addition level of 25% of the long fibre amount. With its high molecule weight, this component could be viewed to describe the bonding behaviour for an additive having size scale between PVA and TCNF. Assuming complete retention, the amounts of fine components (V-fines, CMF, and TCNF) and LBG in the whole material were 7% and 18%, respectively. Table 3 shows the measured material properties with densities varying in the range of 38–51 kg/m 3 . The addition of fine components caused an apparent increase in all the measured strength properties, as shown in Fig. 10. Within experimental accuracy, the finest and highly charged TCNF did not increase compression stress much more than CMF. The clearly coarsest fine component, fibrillar V-fines, also had potential as a strength additive. The addition of LBG clearly led to the highest compression modulus and compression stress at the small 10% deformation level (see Fig. 10). The formed material was very hard and Fig. 9 Recovery from 50% compression 1 min after the load was removed. The recovery depended more on the fibre properties than on network bonding. TP1–3 are samples made with SDS and TP4-6 samples made with PVA foam
Fig. 10 a Specific compression stress at 10% and 50% deformation, and b compression modulus for the samples with added fine components (Table 3). Note that the modulus obtained with added LBG exceeded the scale of the y-axis
123
Made with FlippingBook Online newsletter maker