4.1. Influence of Fiber’s Origin and Morhplogy on Molded Pulp Products’ Initial Mechanical Properties
The first fiber’s origin comparison was between hardwood (BHKP) and softwood (BSKP) fibers that went through similar kraft and bleaching pulping process. It is widely known that hardwood fibers have shorter fibers than softwood [
1,
38,
39,
40]. These studies also showed that this difference modifies the resulting product’s mechanical properties, with lower strength at break for shorter fibers.
Fišerová et al. [
41] studied the effect of bleaching on kraft pulps and observed that softwood has a higher tensile index than hardwood, whether they are bleached or left unbleached. As for the bleaching process, it showed that it reduces the mechanical properties, as unbleached pulp showed a higher tensile index than bleached pulp, for both hardwood and softwood pulps. In our study, we observed a higher strength at break for softwood fibers but a lower difference for the bleaching process of softwood kraft pulp. With fibers having similar length and width, the resulting initial strength at break is comparable, even with a bleaching process, but we obtained a higher strain at break for unbleached fibers.
The recycled fibers tested have shorter and thinner fibers than softwood kraft fibers and a similar morphology to hardwood kraft fibers. However, their mechanical properties are different from BHKP, with a higher Young modulus for both the recycled newspapers (R-NPM) and recycled cardboard box (R-CBB) samples. The recycling process reduces the mechanical properties of fibers, due to fiber’s morphological modifications, obtained during recycling, such as length reduction, hornification, or bonding alteration, due to non-wood additives on the fibers surface [
42,
43]. As shown with FTIR analysis, the recycled samples contain minerals such as kaolinite and carbonate-based clays. These molecules are often used in the paper industry to whiten and improve the brightness of the paper, but they may also decrease the paper strength, as it may disturb the bonding in between fibers [
44]. It also seems like the recycled fibers are a mix between hardwood and softwood but also from kraft processes and mechanical processes, to obtain a higher Young modulus in this study than kraft fibers.
In other studies [
45,
46], CTMP products showed lower mechanical properties than kraft products. The fact that the MPP manufacturing process is different from papermaking also modifies the mechanical properties of the product made [
11]. When comparing the fibers’ morphology, observed with SEM images and the mechanical results, we observed that having non-cellulosic molecules on the fibers’ surface (recycled and BCTMP) improved the Young modulus of the MPP. However, for fibers with a very low amount of non-cellulosic molecules on the surface (kraft fibers), we observed a lower tensile strength and Young modulus.
A similar difference may come from the recycled fibers with a chemical modification of the fibers, due to the presence of diverse molecules with specific properties, as detailed in several studies [
47,
48,
49]. The presence of these molecules in the pulp changes the chemical behavior of the pulp, as well as the fiber’s surface morphology. These differences, thus, modify the mechanical properties of the resulting product, when compared to virgin fibers [
50].
In this study, we observed that the absence of molecules, other than cellulose fibers, reduced the fiber’s resistance to a traction mechanism, as the inter-fiber cohesion is lower than for fibers having wood molecules on the surface to bond the fibers together. As high temperature and pressure were applied to MPP samples during production, non-cellulosic molecules, such as lignin and hemicellulose, observed a thermal modification. It is known that lignin has a glass transition at about 130 °C and further softens at 170 °C [
36,
51]. This softening of lignin in the pulp may help to form a higher cohesion between the fibers and the lignin.
Bleached chemi-thermomechanical pulp (BCTMP) samples have a higher Young modulus and strength at break than recycled fibers. It is possible to see that wood molecules, such as lignin and hemicellulose, as shown in FTIR analysis, further improve the product’s mechanical resistance, as the presence of these molecules allows a higher inter-fiber cohesion. Yet recycled fibers mainly have non-wood molecules and molecules that were added in previous paper-making processes, such as minerals (bentonite and titanium dioxide), ink, glue, and other paper additives, to obtain specific paper properties [
49,
52,
53]. From the results obtained with these pulps, it seems that these synthetic molecules do not improve the inter-fiber cohesion, as they all have different and specific cohesion mechanisms.
4.2. The Effect of Fiber’s Morphology on Molded Pulp Products’ Moisture Uptake after Sorption Analysis
To further understand the influence of K, C
0, and τ
m values on the GAB curve, we compared a control curve with curves for which one variable was varied at a time. The result is shown in
Figure 11. K and τ
m values both have a higher influence on the resulting GAB curve. Having a lower τ
m value allows the moisture uptake to be lowered for all a
w ranges. On the contrary, with a higher τ
m value, the moisture uptake is higher on all a
w ranges. Concerning the K constant, it mostly changes the curve at higher a
w, beginning at 0.6. A lower K will decrease the curve at high a
w, whereas a higher K will increase the water uptake at a higher a
w. When comparing these values, we can also observe a higher moisture uptake on all a
w with τ
m. This means that a higher moisture content in the monolayer has a much higher impact on the sample’s ability to adsorb water in humid environments, as compared to the multilayer’s heat properties (K value).
We also observed that C0 has a low influence on the GAB curve variation, even with a high difference in C0 value (±25), compared to K (±0.05) and τm (±2) used in this comparison analysis. A higher difference in the chemical potential between the monolayer and the upper layer, given by C0 value, has a very low impact on the material’s moisture uptake. As C constant is dependent on aw, given by Equation (5), it is easier to compare C0, as it defines C in dry conditions, where aw is equal to 0.
With a negative x1, as obtained for all samples but BCTMP, the moisture uptake was reduced, whereas with a positive x1, the moisture uptake was increased. x1 was obtained with Equation (5). It is dependent on the K value. As BCTMP has the lowest K value (0.84) of all samples tested, this further increased x1 and, as a result, the sample’s moisture uptake.
Having the highest τ
m of all samples for BCTMP, as well as the lowest K value, explains the higher sample’s experimental moisture uptake. Hill et al. [
54] analyzed the moisture uptake of natural fibers and observed that with a higher monolayer content, the resulting sample’s moisture uptake was also increased. The monolayer, on the fiber’s surface, is important to better understand the moisture behavior of the samples tested.
Zhang et al. [
55] studied the effect of hemicellulose on the pulp moisture uptake and they observed that with a higher hemicellulose content, they obtained a higher moisture uptake. As hemicellulose molecules are hydrophilic, they enhance the water adsorption in the resulting pulp. In this study, the FTIR analysis showed that BCTMP contains hemicellulose, along with lignin. It seems that this presence greatly influences the sample’s moisture uptake, as opposed to kraft fibers or recycled fibers. Although hydrophobic lignin molecules are also in the BCTMP pulp, the highly hydrophilic nature of hemicellulose has a higher influence, translating in our study to a higher moisture uptake than the samples without hemicellulose and lignin.
4.3. Effect of Molded Pulp Products’ Moisture Uptake after Sorption Analysis on Samples’ Mechanical Properties
We observed that BCTMP had the highest mechanical properties, with a higher Young modulus for all a
w tested. When comparing these results with the initial mechanical properties obtained in
Section 3.4, a sample with a higher Young modulus in initial conditions maintained a higher Young modulus after sorption analysis. With BHKP having the lowest Young modulus in initial conditions, we also obtained a lower Young Modulus after sorption analysis in all a
w tested.
The Young modulus decrease, observed at high a
w, may be explained by the softening of the fibers of all samples. Salmén et al. [
56] observed a softening effect of the wood fibers after adsorption of water molecules. They also showed the material’s softening is mainly induced by hemicellulose for a
w of 0.3 to 0.8. At higher a
w, they showed that the amorphous phase of cellulose also softens the material. For BCTMP in our study, we obtained similar results with a higher water uptake, due to hemicellulose kept in the finished product, as opposed to the other fibers analyzed.
As humidity was increased, moisture uptake in the BCTMP also increased and, as a result, the tensile properties decreased, with lower Young modulus and stress at break but higher strain at break. An increasing moisture content in molded pulp product (MPP) samples may, as a result, create relaxation in the micro-compressions that may have been formed when the samples were dried during the process [
57]. The relaxation could, in this case, allow the analyzed sample to have a higher strain at break. With this phenomenon, we can observe that as the strain at break increases, the stress at break decreases, thus decreasing the samples’ Young modulus. This further shows the fiber’s softening effect when a high moisture content is adsorbed by the sample.
It is also known that hornification, mainly observed in low yield pulps, such as kraft pulps, reduces the fiber’s swelling capacity and flexibility when in contact with water, thus resulting in a mechanical strength decrease [
58,
59]. This effect modifies the fiber’s surface morphology, with a reduction in their inter-fiber bonding capacity, due to fiber stiffening [
60,
61]. This phenomenon may explain the lower strength results in the kraft pulps (BSKP, BHKP, and USKP) analyzed in this study, when compared with the recycled (R-NPM and R-CBB) and mechanical (BCTMP) pulps.
For the mechanical softening of BCTMP, it is maintained up to 0.98 aw. This means that the mechanical decrease, observed for all MPP samples, was obtained due to the softening of amorphous cellulose at high aw.
The curve in
Figure 12 was used to further analyze the influence of moisture uptake on the mechanical properties of MPP samples. With this curve, we obtained an affine regression in the form “y = a·x + b”, with y as the Young modulus (in GPa), x as the water uptake, a as the curves’ slope giving information on the speed of change of E depending on a
w, and b as the Young modulus in the sample’s dry state (E
dry).
The theoretical samples of dry Young modulus obtained by the curve give us the same order as for Young modulus in the initial conditions (
Section 3.4). The curve’s slope (a) differentiates the fibers used in two categories. The first one is for kraft fibers with a plot around −0.016 and the second is for BCTMP and recycled fibers with a plot around −0.031. This disparity may be explained by the morphological difference between the two categories. In kraft fibers, most of non-cellulosic molecules were removed (lignin, hemicellulose, and pectin), whereas most of wood molecules were maintained in BCTMP, and other types of molecules were added in recycled fibers, as shown in our FTIR analysis.
As there is a great potential variety of molecules that may be present in the recycled fibers, the effect of each of them on the resulting properties of the MPP samples will be complicated to precisely define. Several studies and reviews have performed research on the contaminants and particles in recycled paper and board [
48,
62] and found a high variety of both synthetic and natural molecules. Some of these molecules may increase the sample’s hydrophilicity, such a starch, whereas other molecules are added to reduce the paper’s hydrophilicity, such as waxes and wet strength agents. FTIR analysis showed us that clays, such as kaolin, were present in the recycles samples. These molecules may have an important role in the hygroscopic and mechanical properties of the resulting recycled samples (R-NPM and R-CBB). Naijian et al. [
52] observed that the addition of kaolin in the paper reduced the water absorption. However, the addition of a high amount of kaolin also reduced the mechanical resistance of the paper. In our study, the presence of kaolin in the paper was beneficial to reduce the MPP sample’s moisture uptake. As the mechanical properties of the recycled samples were kept higher than kraft fibers, it seemed that the amount of kaolin in the pulp was sufficiently low to avoid a decrease in the mechanical resistance.
Having non-cellulosic molecules on the cellulose fibers surface, thus, allowed the MPP samples to have higher initial and dry Young modulus but also a higher Young modulus loss with increasing moisture uptake.
When comparing the molecules covering the fibers’ surface, it seems that wood molecules, naturally existing and maintained in BCTMP, allowed the resulting MPP to obtain a higher Young modulus on all conditions tested in this study. With recycled fibers, the origin and amount of molecules covering the cellulose fibers were complicated to analyze. Moreover, the bonding and hydrophilic properties between these molecules and the fiber surfaces, as well as other molecules, may be significantly different. Each recycled paper previously underwent specific chemical pretreatments and an addition of agents (molecules) to obtain the desired properties of the paper. These molecules were artificially added and mostly synthetic or minerals. Their cohesion to cellulose fibers, as well as all other additives that may be in contact, may be much lower than the cohesion between molecules that were originally in the wood, such as lignin and hemicellulose.
In our study, the MPP made with a pulp having non-cellulosic molecules on the cellulose fiber’s surface allowed the product to have the desired mechanical properties for packaging applications.
We also observed that, even with longer fibers, as given by morphological analysis, kraft fiber based MPPs showed lower mechanical properties. The impact of the fiber’s length was lower than the effect of preserving the wood molecules in the pulp or even adding molecules in the pulp (recycled fibers). However, when comparing the fiber’s origin, obtained with the same pulping process as used for BSKP and BHKP, we observed that using softwood fibers (BSKP) allowed the resulting MPP to gain a higher Young modulus, both in the initial conditions and after moisture uptake, when compared to hardwood fibers (BHKP).
The effect of bleaching was also observed with higher initial properties for USKP, compared to BSKP. A higher moisture uptake was observed for the unbleached fibers for all aw tested, as compared to bleached fibers. It seems that the bleaching process, performed on chemical pulps, further removed non-cellulosic molecules that could have been left after the kraft process. Thus, it changed the properties of the resulting MPP made with an increase in hydrophilicity. With this difference between both fibers, MPP, made with USKP in this study, gave higher mechanical properties under humid conditions in all aw tested than BSKP samples.