PLA-*g*-Cellulosic Derivatives

Cellulose fibers derived from renewable biomass have attracted interest as microscale reinforcements in composite materials. Natural fibers have many advantages—low density, low cost, renewability, biodegradability—that make them excellent candidates for the design of biodegradable materials. They can advantageously replace mineral reinforcements in PLA matrices [3,4]. However, a poor compatibility between the fiber and the polymer matrix leads to materials with poor performances. In particular, nanocelluloses, due to their polar surfaces, are difficult to uniformly disperse in a non-polar medium. The consequences of this poor interfacial compatibility between polymer and filler are poor properties of the final blend. The compatibilization of PLA/cellulosic derivatives blends to improve many of the blends' properties, especially mechanical properties, while maintaining a natural source, are the most widely described in the literature [24,25]. Because PLA-*g*-MA acts as a reactive compatibilizer, a PLA-*g*-cellulose copolymer is formed as a result of the reaction between the anhydride of PLA-*g*-MA, and alcohol functions of the cellulosic derivative (Figure 6) [21]. The presence of a very low percentage of PLA-*g*-MA (<1%) in the original blend significantly improved the properties of the blend [24]. Tensile strength, tensile modulus and strain at break were increased by 55.3%, 15.45% and 30.4%, respectively, over neat PLA by adding 5 wt.% of cellulose nanofibers (CNFs), and by 169.2%, 36.3% and 139.1%, respectively, by adding 5% of PLA-*g*-MA to the blend PLA/CNF.

**Figure 6.** Reaction scheme for the grafting of a cellulosic fiber on PLA-*g*-MA (from González-López et al. [21], copyright Taylor and Francis, reproduced with permission).

Many cellulosic derivatives were introduced into PLA matrices in the presence of PLA-*g*-MA to improve the mechanical properties of blends such as Luffa [26], flax [27], coffee grounds [28], wood flour or rice husk [29,30], sisal fibers [31], straw [32], bamboo fiber [33], cassava starch [34,35], starch [36], and lemongrass fiber [37]. For example, the use of PLA-*g*-MA in a PLA/cassava starch blend has a significant impact on elongation at break but not on Young's modulus or tensile strength. However, it was noted that PLA-*g*-MA with a higher proportion of grafted MA (0.52 wt.%) had a lower molecular weight and higher dispersity value, showing some degradation of the polymer backbone [34,38].

Many other properties are improved in PLA-*g*-MA compatible blends, such as morphological, rheological, thermal, tensile and moisture sorption properties as well as thermal degradation [39–41]. For example, Figure 7 shows SEM micrographs of PLA/TPS (thermoplastic starch) blends (70/30 *w*/*w*) without (Figure 7a) and with two parts per hundred rubber (phr) PLA-*g*-MA (Figure 7b), highlighting the compatibilization of the blend of PLA-*g*-starch formed in situ.

**Figure 7.** SEM micrographs of PLA/TPS blends (70/30 *w*/*w*) (**a**) without and (**b**) with 2 phr PLA-*g*-MA (adapted from Moghaddam et al. [39], copyright Springer Science, reproduced with permission).

To obtain a PLA-*g*-starch copolymer, another method is to react maleic anhydride with starch to obtain a maleated thermoplastic starch (MTPS), which is then mixed with PLA in the presence of Luperox 101 (2,5-bis(tert-butylperoxy)-2,5-dimethylhexane) in a Brabender at 180 ◦C for 5 min. The reaction scheme is shown in Figure 8 [42].

Direct grafting of cellulose nanocrystals (CNC) on PLA, without the addition of PLA*g*-MA, is also described, following the reaction scheme of Figure 9 [43]. could his DCP was sprayed onto PLA beads, and the DCP-coated PLA pellets were mixed with CNC and extruded in a twin-screw extruder at 180 ◦C at 50 rpm for 5 min. The effective grafting of CNC onto PLA was identified by SEC, FTIR and NMR, but NMR showed a very low proportion of CNC in the copolymer. Some mechanical and structural properties were significantly impacted (increased Young modulus, decreased elongation, increased crystallinity).

**Figure 8.** Reaction scheme of the MTPS formation and coupling to PLA (from Wootthikanokkhan et al. [42], copyright Wiley-VCH GmbH. Reproduced with permission).

**Figure 9.** Reaction scheme for the synthesis of PLA-*g*-CNC (from Dhar et al. [43], copyright Elsevier, reproduced with permission).

PLA-*g*-Natural Rubber (PLA-*g*-NR)

Among the drawbacks of PLA materials, we can also highlight their fragility. Natural rubbers (NR), on the other hand, are highly flexible, environmentally friendly and derived from a renewable resource. They are good toughness agents due to their high molecular weight and very low glass transition temperature. However, due to the non-polarity of NRs, PLA/NR blends are immiscible and not compatible. To improve the interfacial interaction between PLA and NR, the reactive compatibilizer PLA-*g*-MA is used to form a graft copolymer PLA-*g*-NR [44,45]. Typically, the compatibilized blend is made in a twin screw extruder at a temperature between 160 and 180 ◦C and a screw speed of around 30 rpm. With the addition of PLA-*g*-MA, the mechanical properties of the material were significantly improved. It was found that a 3% PLA-*g*-MA was the best compatibilizer composition to achieve the best performance of the material [44]. The reverse reaction of a maleic anhydride on NR (NR-MA), followed by reaction on PLA in a radical medium, was also performed with similar results regarding mechanical properties [46,47]. However, in this case, the proposed mechanism does not involve a reaction on the PLA backbone but only the alcohol chain end, which reacts on the backbone of NR-MA (Figure 10).

**Figure 10.** Proposed mechanism for the grafting of NR-MA on PLA (from Thepthawat et al. [47], copyright Wiley-VCH GmbH. Reproduced with permission).
