*2.4. Measurements*

1H-NMR spectra were recorded using a JEOL ECX 400 spectrometer (JEOL, Akishima, Tokyo, Japan). Microscopic laser images were obtained using a Keyence VK-8500 laser microscope (Keyence, Osaka, Japan). SEM images were obtained using a Hitachi S-4100H electron microscope (Hitachi High-Technologies Corporation, Tokyo, Japan) with an accelerating voltage of 5 kV. IR spectra were recorded on a PerkinElmer Spectrum Two spectrometer (PerkinElmer Japan Co., Ltd., Yokohama, Japan).

#### **3. Results and Discussion**

Prior to performing Pickering emulsion polymerization, we prepared SD-ChNF dispersions in aqueous acetic acid according to a previously reported procedure as follows (Figure 1a) [27,28]. The self-assembled ChNF film was first fabricated by regeneration from the chitin/AMIMBr ion gel using methanol to produce the ChNF dispersion and subsequent filtration [18]. In Figure 2a, the SEM image of the ChNF/methanol dispersion shows the same nanoscale morphology as that reported in a previous study [18]. The film

was then subjected to treatment with 30 wt% aqueous sodium hydroxide at 80 ◦C for 5 h to achieve partial deacetylation. A DDA value of 23% was estimated by 1H-NMR analysis after the acidic hydrolysis and dissolution in DCl/D2O. The produced film was mixed with 1.0 mol/L aqueous acetic acid through ultrasonication using a homogenizer (20 kHz, 400 W) to produce the SD-ChNF/aqueous acetic acid dispersion. The sizes of the resulting nanochitin observed in the SEM image of the dispersion (Figure 2b) were clearly reduced as compared with those of the parent ChNF/methanol dispersion (Figure 2a), indicating the successful preparation of the desired SD-ChNF dispersion.

**Figure 2.** SEM images of the (**a**) parent ChNF/methanol dispersion and (**b**) SD-ChNF/aqueous acetic acid dispersion.

The Pickering emulsions were prepared by mixing styrene with the resulting dispersion (weight ratios of SD-ChNFs to styrene = 0.1:1–1.4:1, runs 1–5 in Table 1)) (Figure 1b). The laser microscopic images of the obtained mixtures (images of Pickering emulsions of runs 1 and 5 are representatively shown in Figure 3) supported the adequate formation of the emulsions at all weight ratios, even at higher ChNF contents (runs 3–5) than those that did not induce the formation of clear emulsions using the maleylated ChNFs [21]. These results sugges<sup>t</sup> the practice of scaling down the parent ChNFs into SD-ChNFs for improving dispersibility in aqueous media in order to effectively act as stabilizers in Pickering emulsions.

**Figure 3.** Microscopic laser images of Pickering emulsions for (**a**) run 1 and (**b**) run 5.

Pickering emulsion polymerization of styrene was then performed using the same procedure as that used for the maleylated ChNFs reported in our previous studies [21,22]. After nitrogen gas was bubbled into the emulsions, radical polymerization of styrene was conducted in the presence of potassium persulfate (0.35 mol% with styrene) at 70 ◦C for 24 h while stirring in a nitrogen atmosphere (Figure 1b). The products were isolated by centrifugation and lyophilized to obtain SD-ChNF/polystyrene composite particles. The IR spectrum of the product (run 2, Figure 4c) exhibited characteristic absorptions from chitin at 3437 cm<sup>−</sup><sup>1</sup> (O–H), 1656 and 1618 cm<sup>−</sup><sup>1</sup> (C=O of amide I), 1552 cm<sup>−</sup><sup>1</sup> (C=O of amide II), and 1072 cm<sup>−</sup><sup>1</sup> (C–O–C), as well as at 697 cm<sup>−</sup><sup>1</sup> for polystyrene (C–H of aromatic

ring), respectively (Figure 4a,b). The 1H-NMR spectrum of a solubilized fraction of the product of run 4 in CDCl3 is shown in Figure 5, where the observed signals are assignable to polystyrene, but the chitin component was not analyzed by NMR measurement due to its insolubility in common NMR solvents. These data indicate that the particles consisted of both chitin and polystyrene. After the products were re-dispersed in water, SEM images of the spin-coated samples from the re-dispersions were captured to evaluate the morphologies and sizes of the composite particles.

**Figure 4.** IR spectra of (**a**) the partially deacetylated (PDA-)ChNF film, (**b**) polystyrene, and (**c**) composite particles for run 2.

**Figure 5.** 1H-NMR spectrum of solubilized fraction of composite particles for run 4 in CDCl3.

The SEM images of the samples obtained using all the SD-ChNF/styrene weight ratios clearly show the particle morphologies (Figure 6). The particle sizes (average diameters) were calculated based on the vertical and horizontal lengths of 50 particles in each SEM image (Table 1). The standard deviation values for all the average diameters were relatively small, indicating the formation of composite particles with uniform sizes in all cases. The average diameter of the particles obtained by the SD-ChNF/styrene weight ratio of 0.1:1 (run 1) was comparable to that obtained by the same weight ratio of the maleylated ChNFs with an upper hierarchical scale as reported in our previous study (259 and 266 nm, respectively) [21]; however, at a higher weight ratio (0.2:1, run 2), the average diameter of

the particles using the SD-ChNFs was smaller than that using the maleylated ChNFs (167 and 199 nm, respectively). Moreover, the average diameters were smaller with increasing SD-ChNF/styrene weight ratios and reached 84 nm with an SD-ChNF/styrene weight ratio of 1.4:1. These results strongly indicate that the SD-ChNFs efficiently acted as stabilizers for the present Pickering emulsion polymerization to fabricate composite particles that were smaller than 100 nm. The weight yields of composite particles increased when increasing the SD-ChNF/styrene weigh ratios, indicating the formation of more stable emulsions in accordance with increased SD-ChNF ratios.

**Figure 6.** SEM images of SD-ChNF/polystyrene composite particles for (**<sup>a</sup>**–**<sup>e</sup>**) runs 1–5.
