Preparation of Chitin Nanofibers from Shrimp Shell Waste by Partial Deacetylation and Mechanical Treatment

ABSTRACT

Chitin is one of the most abundant biopolymers in nature. Herein, we report the successful preparation of chitin nanofibers (ChNFs) from shrimp shell waste using a partial deacetylation process with NaOH and high-speed blending. The effects of the deacetylation reaction with NaOH concentrations (0–40 wt%) on the degree of acetylation (DA), crystallinity, zeta potential, thermal stability, and morphology of the ChNFs were investigated. With the more aggressive deacetylation reaction (higher NaOH concentration), ChNFs had the lower DA, crystallinity degree, and thermal stability, and their widths and lengths became smaller and shorter. The presence of amino groups in the chitin molecule caused by deacetylation generated repulsive forces with aids of acetic acid, efficiently leading to the individualization of ChNFs using high-speed blending. The individualized ChNFs deacetylated with 30 wt% NaOH had an average width of 8.07 ± 1.80 nm and length of less than 500 nm, whereas bundles of aggregated fibers with widths in the range of 30–100 nm and lengths of up to several μm were extracted from chitin without deacetylation. Additionally, the deacetylation with 40 wt% NaOH completely converted chitin to chitosan. The ChNFs could be efficiently used for composites, biomaterials, and packaging applications.

摘要

甲壳素是自然界中含量最丰富的生物聚合物之一. 在此，我们报道了利用NaOH脱乙酰工艺和高速共混，从虾壳废料中成功制备甲壳素纳米纤维（ChNFs）. 研究了NaOH浓度（0–40 wt%）下的脱乙酰反应对ChNFs的乙酰化程度（DA）、结晶度、ζ电位、热稳定性和形态的影响. 随着更积极的脱乙酰反应（更高的NaOH浓度），ChNFs具有更低的DA、结晶度和热稳定性，并且它们的宽度和长度变得越来越小和更短. 脱乙酰基化导致的几丁质分子中氨基的存在在乙酸的帮助下产生了排斥力，有效地导致了使用高速共混的ChNFs的个性化. 用30 wt%NaOH脱乙酰的个性化ChNFs的平均宽度为8.07 ± 1.80 nm，长度小于500 nm，而从几丁质中提取的聚集纤维束的宽度在30–100 nm范围内，长度可达数μm，而没有脱乙酰. 此外，用40 wt%的NaOH脱乙酰作用将甲壳素完全转化为壳聚糖. ChNF可以有效地用于复合材料、生物材料和包装应用

KEYWORDS:

• Shrimp shells
• protonation
• morphology
• thermal stability
• deacetylation
• blending
• mechanical treatment
• surface charges

关键词:

• 关键词
• 质子化
• 形态学
• 热稳定性
• 脱乙酰
• 混合
• 机械处理
• 表面电荷

Introduction

Every year, the global seafood industries generate an estimated 6 to 8 million tons of shell waste from crustaceans including crabs, shrimps, and lobsters. This waste can be extracted and converted into high-quality raw materials that are used in a wide range of industrials. Crustacean shells consists of three main parts: 20–50% of calcium carbonate, 15–40% of chitin, and 20–40% of protein (Yan and Chen 2015). Chitin is a linear (1,4)-β-N-acetyl glycosaminoglycan, which is the second most abundant organic compound found in nature after cellulose. It predominantly occurs in two allomorphs: α-structure (mainly found in exoskeletons of arthropods such as crabs, lobsters, and shrimps) and β-structure (found in marine diatoms such as squid pens) (Elieh-Ali-Komi and Hamblin 2016; Ifuku et al. 2009). α-type chitin is easily found in nature in comparison to β-type chitin. α-type chitin has an antiparallel chain packing mode with strong intermolecular interaction bonding between fibrils, whereas a parallel chain packing mode is found in β-type chitin. Owing to its parallel chain packing mode, β-type chitin is sensitive to enzymatic degradation and chemical treatment (Fan, Saito, and Isogai 2008b). Chitin consists of crystalline nanofibrils with widths in the range of 2.5–25 nm, depending on the origin of chitin sources (Fan, Saito, and Isogai 2008b; Wu, Mushi, and Berglund 2020). Chitin nanofibers (ChNFs) exhibit biodegradability, renewability, and high mechanical properties (Dumont et al. 2018; Li, Wu, and Zhao 2016). The elastic modulus of highly crystalline chitin filaments extracted from snow crab shells measured by synchrotron radiated X-ray diffraction was found to be ⁓60 GPa (Ogawa et al. 2011), and the tensile strength values of individualized ChNFs extracted from microalgae Phaeocystis globosa (α-type chitin) and Loligo bleekeri squid pens (β-type chitin) estimated using cavitation-induced tensile fracture in a liquid medium were ⁓1.6 GPa and ⁓3.1 GPa, respectively (Bamba et al. 2017). Owing to their promising features, ChNFs have been of great interest in various applications such as composites, packaging, and bioengineering (Svensson et al. 2022; Wang et al. 2021).

It has been reported that the individualization of ChNFs from squid pen β-type chitin was successful yielded by simple disintegration. This was owing to the low crystallinity, parallel chain packing mode, and weak interaction bonding between fibrils (Fan, Saito, and Isogai 2008b). However, owing to their greater number, the extraction of ChNFs from the α-type chitin sources is more beneficial. The high shear forces generated from mechanical treatments such as grinding (Ifuku et al. 2009, 2011) and high-pressure homogenization (Li, Wu, and Zhao 2016; Wu et al. 2014) have been used to isolate ChNFs from the α-type chitin sources. For example, to extract ChNFs from crab shells, the chitin suspension passed through the high-pressure homogenizer with a small nozzle at a pressure of 15,000 psi for 20 passes and at a pressure of 22,000 psi for another 10 passes (Wu et al. 2014). These mechanical disintegration processes require industrial equipment that consumes a lot of energy because the strong hydrogen bonding between the hydroxyl, acetamido, and amino functional groups in chitin molecules obstructs the fiber defibrillation. Nevertheless, the successful individualization of ChNFs with uniform widths in the range of 10–20 nm was obtained via high-pressure homogenization with the help of acetic acid because the cationization of amino groups on chitin could generate electrostatic repulsive forces under acidic conditions to destroy the hydrogen bonding between the functional groups of chitin, although the degree of conversion of the acetamido groups to the amino groups was only 3.9% (Ifuku et al. 2010).

Surface charges on chitin modified by chemical treatments play a key role for individualization of ChNFs (Li, Wu, and Zhao 2016; Xu et al. 2019). For example, 2,2,6,6-tetramethylpiperidine-1-oxyl radical (TEMPO)-mediated oxidation with NaOCl was studied for the selective oxidization of the C-6 primary hydroxyl groups of chitin to carboxylate groups (Fan, Saito, and Isogai 2008a; b). The presence of anionic carboxylate groups at the C-6 position on the chitin structure prepared via TEMPO-mediated oxidation could generate electrical repulsive forces between fibrils, assisting the individualization of ChNFs via sonication. Recently, Haider et al. (2022) compared morphologies of ChNFs obtained from waste shrimp shells using two different approaches: TEMPO-mediated oxidation and mechanical grinding. ChNFs prepared using TEMPO-mediated oxidation had an average length of 211 nm, a width of 8.7 nm, and an aspect ratio of 24, which were shorter and smaller than ChNFs prepared by mechanical grinding with an average length of 1,068 nm, a width of 16 nm, and an aspect ratio of 67. While TEMPO-mediated oxidation was found to produce ChNFs with a smaller width, this chemical treatment requires specific and expensive chemicals.

Deacetylation with alkaline is another chemical treatment approach to convert the acetamido groups in the chitin structure to the cationic amino groups and assist the individualization of ChNFs through electrostatic repulsive forces generated by these groups (Fan, Saito, and Isogai 2010; Satam and Meredith 2021). ChNFs with an average width of 6.2 nm and length of 250 nm were isolated from chitin deacetylated with 33 wt% NaOH for 3 h using sonication (Fan, Saito, and Isogai 2010). Further, Satam and Meredith (2021) investigated the effect of deacetylation on the morphologies of ChNFs and observed an increase in fibrillation and a decrease in the fiber length with the lower degree of deacetylation (DA) of ChNFs (higher conversion of the acetamido groups to the cationic amino groups). The reduction in the length of ChNFs might be caused by strong deacetylation treatment (Fan, Saito, and Isogai 2010; Satam and Meredith 2021).

Herein, we report the relationship between deacetylation conditions with NaOH and properties of the deacetylated ChNFs extracted from shrimp shell waste as Thailand is one of the world’s largest seafood exporters, especially of shrimp products (Tapanya and Udomkit 2016). Shrimp shells were initially chemically treated to remove minerals and proteins, and deacetylation was introduced to convert the acetamido groups to the cationic amino groups in the chitin, and ChNFs were disintegrated from the deacetylated chitin via high-speed blending. Effects of the deacetylation conditions with various NaOH concentrations (0–40 wt%) on the isolation of ChNFs and their properties including chemical structure, surface charges, crystallinity, and thermal stability were investigated. The results of this study provide the suitable deacetylation condition for the defibrillation of ChNFs with simple mechanical blending.

Experimental

Materials

Shrimp shells (Penaeus merguiensis) were purchased from a local fish market in Prachuap Khiri Khan, Thailand. NaOH and Hydrochloric acid (HCl) (37%) were purchased from Union Chemical 1986 Co., Ltd. and Qchemical Co., Ltd., respectively. Acetic acid (96%) was purchased from RCI Labscan Co., Ltd.

Preparation of ChNFs

The ChNFs were prepared via chemical pretreatment (deproteinization, demineralization, and deacetylation) and mechanical disintegration, as shown in Figure 1. Shrimp shells were chemical treated to yield purified chitin using the procedure modified from previous works (Abdou, Nagy, and Elsabee 2008; Biswas et al. 2015; Wu et al. 2014). Initially, shrimp shells were ground, and the powder was subjected to deproteinization for 24 h using 4 wt% NaOH. This alkali-treated powder was subsequently washed until the pH of the powder became neutral and dried at 65°C. Then, HCl (4 wt%) was added to the dried powder and left for 24 h to eliminate minerals from the alkali-treated powder, and the treated powder was washed several times and dried at 65°C. The chemically treated chitin was deacetylated with various NaOH concentrations (0–40 wt%) at 100°C for 2 h, and the deacetylated chitin was washed with distilled water till its pH became neutral; subsequently, it was dried at 65°C. Moreover, a chitin suspension containing (1 wt%) was prepared, and the pH of this suspension was adjusted to 4 using acetic acid. ChNFs were mechanically isolated using a high-speed blender (Stromix 3500 W, Thailand) running at 42,000 rpm for 30 min. The ChNFs deacetylated with NaOH concentrations of 0, 10, 20, 30, and 40 wt% were labeled as C0, C10, C20, C30, and C40, respectively.

Figure 1. Preparation of chitin nanofibers (ChNFs) from shrimp shell waste.

Chemical structure

The chemical structures of the ChNFs treated under various deacetylation conditions were determined using Fourier-transform infrared (FTIR) spectroscopy (Nicolet 6700, USA). The samples were scanned in a frequency range of 4000–400 cm⁻¹.

The structures of the deacetylated ChNFs were analyzed using a solid-state ¹³C nuclear magnetic resonance (NMR) spectroscopy (Avance III HD 400 WB, Bruker Corp., USA) at a frequency of 100 MHz in the range of 0–200 ppm. DAs of the ChNFs were calculated based on the integral (I) of the methyl groups (CH₃) and carbon atoms in the chitin structure (C_1–6) using the following equation (Ottey, Vårum, and Smidsrød 1996; Vårum et al. 1991):

(1) $\hspace{0.17em}DA=\frac{I_{\mathrm{C}{\mathrm{H}}_3}}{\left(\frac{I_{{\mathrm{C}}_1}+I_{{\mathrm{C}}_2}+I_{{\mathrm{C}}_3}+I_{{\mathrm{C}}_4}+I_{{\mathrm{C}}_5}+I_{{\mathrm{C}}_6}}{6}\right)}$(1)

Surface charges

The surface charges of the ChNFs dispersed in water with a concentration of 0.1 wt% were measured using a dynamic light scattering Zetasizer (Nano ZSP, Malvern Panalytical Ltd., UK) at 25°C.

Crystal structure and crystallinity

The crystal structures and degrees of crystallinity of the ChNFs were determined using an X-ray diffractometer (D8 Advance, Bruker Ltd., UK) operating at an acceleration voltage of 40 kV and 40 mA. The samples were scanned in the 2θ range of 5–60° at a scan speed of 0.5 s per step and an angle step increment of 0.02° with Cu Kα radiation at a wavelength of 0.154 nm. The degree of crystallinity of the ChNFs was determined from the intensity of the diffraction peaks located at 19.6° and 16.0°, corresponding to crystalline (I_C) and amorphous regions (I_A), using the following equation (Wu, Mushi, and Berglund 2020):

(2) $\mathrm{d}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{c}\mathrm{r}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{y}\left(\mathrm{\%}\right)=\left(\frac{I_{\mathrm{C}}-I_A}{I_{\mathrm{C}}}\right)\times 100$(2)

Thermal properties

The thermal properties of the ChNFs treated under various deacetylation conditions were measured via a thermogravimetric (TG) analyzer (TGA/SDTA 851^e, Mettler Toledo Ind., USA). The ChNF sample (⁓10 mg) was heated from 50°C to 700°C at the heating rate of 10°C min⁻¹ under a nitrogen atmosphere with a N₂ flow rate of 40 mL min⁻¹.

Morphology

The morphologies of ChNFs were investigated using a transmission electron microscope (TEM) (JEM-1400, JEOL, Japan) at an acceleration voltage of 80 kV. All the samples were coated with 1% uranyl acetate for 3 min. The mean widths of ChNFs were measured from 50 random fibrils using ImageJ software.

Results and discussion

Chemical structure

The chemical structures of ChNFs obtained without the NaOH treatment and ChNFs deacetylated with various NaOH concentrations were evaluated, and their FTIR spectra are shown in Figure 2. After shrimp shells were chemically treated for deproteinization and demineralization, the peaks, corresponding to calcium carbonate (1798 and 876 cm⁻¹), and the peak, corresponding to the protein in shrimp shells (1420 cm⁻¹), could not be observed (Hejazi et al. 2018; Ifuku et al. 2009; Li, Wu, and Zhao 2016). This suggested that the deproteinization and demineralization steps used in this study successfully removed calcium carbonate and protein from the shrimp shells to yield pure chitin. All ChNFs (C0, C10, C20, C30, and C40) showed similar FTIR characteristic peaks at 3435, 3260, 3100, 1658, 1622, 1560, 1420, 1310, and 1075 cm⁻¹ (Acosta et al. 1993; Hejazi et al. 2018; Salehinik et al. 2021), which are consistent with the characteristic peaks of the α-chitin structure (Acosta et al. 1993; Salehinik et al. 2021). No peaks were observed at 972 and 632 cm⁻¹, corresponding to the structure of β-chitin (Acosta et al. 1993). The characteristic peak at 3435 cm⁻¹ corresponds to stretching vibrations of the hydroxyl (–OH) group, and the peaks located at 3260 and 3100 cm⁻¹ correspond to the stretching vibrations of intermolecular hydrogen bonds between –OH and –NH in an adjacent molecular chain in chitin (Gao et al. 2017; Hejazi et al. 2018; Ifuku et al. 2009; Li, Wu, and Zhao 2016). The presence of the peaks at 1658 and 1622 cm⁻¹ suggested the existence of amide I. The peak observed at 1658 cm⁻¹ corresponds to stretching vibration of the carbonyl group (C=O), and the band at 1622 cm⁻¹ corresponds to the intermolecular interactions of C=O with the NH group in the adjacent chain or with the OH group of the inter chain. Moreover, the peak observed at 1560 cm⁻¹ corresponds to the N–H bending vibration of the amine groups of amide II, and the peak at 1310 cm⁻¹ corresponds to the C–N stretching of amide III (Dhanabalan et al. 2021; Hejazi et al. 2018; Salehinik et al. 2021). In addition, the peaks located at 1420 and 1075 cm⁻¹ correspond to the flexural deformation of CH₂ groups and C–O–C asymmetric stretching vibration in the chitin ring, respectively (Acosta et al. 1993; Hejazi et al. 2018; Xu et al. 2021).

Figure 2. Fourier-transform infrared (FTIR) spectra of ChNFs deacetylated with various NaOH concentrations.

With increasing NaOH concentration during the deacetylation process, the intensity of the 1658 cm⁻¹ peak corresponding to amide I became weaker, indicating the successful deacetylation of chitin (Liu et al. 2017). This was because of the partial replacement of the acetamido groups by the amino groups in the chitin structure (Salehinik et al. 2021; Tsai et al. 2019). The FTIR spectrum of C40 was marginally different in comparison with that of C0, C10, C20, and C30. The peak at 1554 cm⁻¹, corresponding to N–H bending vibration from amino groups (–NH₂), was noticeable (Salehinik et al. 2021; Tanpichai et al. 2020, 2022), and peaks with lower intensities at 1658 and 1622 cm⁻¹ were observed. Notably, C40 presented a broad peak in the range of 3600–3000 cm⁻¹, which might be owing to the interference from water absorption during measurements (Mat Zin, Jimat, and Wan Nawawi 2022; Paulino et al. 2006). Additionally, the spectrum of C40 was found to be similar to that of chitosan (Tanpichai et al. 2020, 2022). This could be because of the conversion of chitin to chitosan with a high concentration of NaOH (40 wt%) (Kumari and Rath 2014; Paulino et al. 2006). A previous study reported a successful conversion of chitin to chitosan using a strong alkaline solution (50%) (Lin et al. 2021; Pereira, Muniz, and Hsieh 2014).

The DAs of ChNFs deacetylated with various NaOH concentrations are shown in Table 1, and Figure 3 presents ¹³C NMR spectra of ChNFs. The eight main peaks, positioned at ⁓175 (C=O), ⁓105 (C₁), ⁓83 (C₄), ⁓75 (C₅), ⁓73 (C₃), ⁓61 (C₆), ⁓55 (C₂), and ⁓23 ppm (CH₃) in the NMR spectra of all ChNFs, correspond to the carbon atoms in the chitin structure (Li, Wu, and Zhao 2016; Pereira, Muniz, and Hsieh 2014). The DA of C0 was calculated to be 0.88 and decreased to 0.84 with the 10 wt% NaOH. With increasing NaOH concentration, a significant reduction in DA was noticed. The DAs of C20, C30, and C40 were 0.81, 0.71, and 0.17, respectively. This meant that a large number of acetamido groups were converted to amino groups when the higher concentration of NaOH was applied during the deacetylation reaction. Moreover, C40 became chitosan as its DA was < 0.50 and was soluble in diluted acetic acid (Pereira, Muniz, and Hsieh 2014; Wu et al. 2014), which was agreed with the FTIR measurements.

Figure 3. ¹³C₋nuclear magnetic resonance (NMR) spectra of ChNFs treated with various NaOH concentrations during the deacetylation treatment.

Table 1. Degree of acetylation (DA), degree of crystallinity, temperature at 10% weight loss (T_10%), and maximum decomposition temperature (T_max) of chitin nanofibers (ChNFs) treated with various NaOH concentrations during the deacetylated reactions.

Material	DA	Degree of crystallinity (%)	T10% (°C)	Tmax (°C)
C0	0.88	94.6	293.2	377.5
C10	0.84	93.9	296.2	379.3
C20	0.81	92.8	278.8	379.3
C30	0.71	90.2	224.0	376.3
C40	0.17	53.7	185.2	270.3

Surface charges

The positive charges on the surface of ChNFs as a function of NaOH concentrations during the deacetylation reaction are presented in Figure 4. C0 had a mean zeta potential of −11.9 mV, and the average zeta potential value increased to ⁓15 mV when 10% and 20% NaOH were used for the deacetylation reaction. The increase in the mean zeta potential was attributed to the protonation of amino groups on chitin surfaces under acidic conditions (Li, Wu, and Zhao 2016). Protonation could occur when the conversion of surface acetamido groups to amino groups was high (Li, Wu, and Zhao 2016), with C30 having the mean zeta potential of 42.1 mV. However, when 40% NaOH was used for deacetylation, the zeta potential value of C40 decreased to 29.2 mV. This decrease was because of the formation of aggregates of chitosan molecules due to inter/intramolecular hydrogen bonding, which generated negative surface charges (Chang et al. 2015; Tsai et al. 2019). Moreover, the zeta potential of ChNFs depends on the pH of the suspension (Biswas et al. 2015; Wijesena et al. 2020). With increasing pH, the zeta potential of ChNFs decreased owing to the presence of anionic groups on their surface (Fan, Saito, and Isogai 2008b; Wijesena et al. 2020).

Figure 4. Zeta potential of ChNFs deacetylated with various NaOH concentrations.

Crystal structure and crystallinity

The crystal structures of the ChNFs treated with various NaOH concentrations are presented in Figure 5. The X-ray diffraction (XRD) curves of all ChNF samples presented two characteristic peaks at 2θ of ⁓9° and 19°, attributed to the diffraction of (020) and (110) planes, and small peaks at 2θ of ⁓20°, 23°, and 26°, corresponding to the planes of (120), (130), and (013) (Deng et al. 2014; Dhanabalan et al. 2021). These curves were in agreement with the XRD patterns of α-chitin extracted from crab shells (Mat Zin, Jimat, and Wan Nawawi 2022), acetes shells (Dhanabalan et al. 2021), and spent pupal shells of black soldier fly (Lin et al. 2021). Calcium carbonate in chitin was reported to present a strong diffraction peak at 29.6° (Ifuku et al. 2009); however, this peak could not be observed for all XRD patterns in this study. This indicated the successful removal of minerals from shrimp shells. After the deacetylation reaction, the two main characteristic peaks at 9° and 19° shifted marginally toward a lower degree. This demonstrated the deacetylation caused a change in the crystal structure of ChNFs (Liu et al. 2017). Table 1 presents the degree of crystallinity of the ChNFs calculated based on the ratio of the crystalline peak to the amorphous region (Wu, Mushi, and Berglund 2020). The crystallinity index of C0 was 94.6%. After the deacetylation reaction, the crystallinity index decreased marginally. The degree of crystallinity of C10 was 93.9%, and the degree of crystallinity decreased with increasing NaOH concentration for deacetylation. C20 and C30 had the degree of crystallinity of 92.8 and 90.2%, respectively. This decrease in the degree of crystallinity was because of the destruction of the crystalline regions during the deacetylation treatment (Fan, Saito, and Isogai 2010; Liu et al. 2017). A substantial decrease in the crystallinity degree to 53.7% was observed for C40, which was owing to the damage to the main chains and effect of the N–H bond on the arrangement of molecular chains (Liu et al. 2017).

Figure 5. X-ray diffraction (XRD) patterns of ChNFs deacetylated with various NaOH concentrations.

Thermal properties

Figure 6 presents the TG and derivative TG (DTG) curves of ChNFs, and the temperature at 10% weight loss (T_10%) and the maximum decomposition temperature (T_max) of ChNFs are presented in Table 1. The first decomposition stage occurs between 50°C and 100°C owing to moisture evaporation (Paulino et al. 2006), and the second decomposition stage occurs between at 300°C and 400°C. This is attributed to chitin degradation, including the dehydration of the saccharide rings and the decomposition of the acetylated and deacetylated units of the N-acetylglucosamine chains (Abdou, Nagy, and Elsabee 2008; Liu et al. 2017; Paulino et al. 2006; Sebestyén et al. 2020). C0 and C10 had similar characteristic thermal stabilities (T_10% and T_max). This suggested that a low NaOH concentration (≤10%) could not affect the conversion of the acetamido groups to the amino groups in the chitin structure, which was supported by FTIR and DA results. However, C20 and C30 exhibited lower thermal stability with T_10% of 278.8°C and 224.0°C and T_max of 379.3°C and 376.3°C, respectively. This was because of the high degree of deacetylation in the chitin structure. However, C40 exhibited three decomposition stages. The first and third transition stages (moisture loss and chitin degradation) were similar to those of C0, C10, C20, and C30 (Paulino et al. 2006), whereas the second decomposition stage of C40 occurred in the temperature range of 220–300°C owing to the degradation of deacetylated molecules (Paulino et al. 2006). Hence, C40 has inferior thermal stability compared to other ChNFs. The T_10% and T_max values of C40 were 185.2°C and 270.3°C, respectively. Additionally, C40 exhibited the highest content of carbon residue at 700°C compared to other ChNFs. The highest char residue of C40 was owing to the high nitrogen content in its structure (Sebestyén et al. 2020).

Figure 6. (a) Thermogravimetric (TG) and (b) derivative TG (DTG) curves of ChNFs treated with various deacetylation reaction conditions.

Morphology

The morphologies of ChNFs are presented in Figure 7. Bundles of aggregated fibers with widths of 30–100 nm were observed for C0 (without deacetylation). The defibrillation of C0 was not possible using only high-speed blending because the limited force generated by this device could not damage the strong hydrogen interaction between nanofibers (Biswas et al. 2015). This was in agreement with a work of Ifuku et al. (2009). The fiber widths in the range of 10–100 nm were derived from crab shells using grinding. With the introduction of deacetylation with 10 wt% NaOH, ChNFs with an average width of 11.38 ± 4.3 nm and lengths of up to several μm with the presence of fewer bundles with widths in the range of 30–70 nm were observed. The disintegration of ChNFs was attributed to electrostatic repulsive forces generated by the protonation of the amino groups (NH₃⁺) on chitin surface at a pH of 4 (Biswas et al. 2015; Ifuku et al. 2009). C20 had the mean fiber diameters of 9.24 ± 2.64 nm, and their morphology was combined with short (<500 nm) and long nanofibers (>600 nm). The successful individualization of nanofibers without aggregation of fibrils was observed for C30 owing to its high content of NH₃⁺, supported by its DA value, which generated strong repulsive forces. The mean fiber diameters of C30 were significantly reduced to 8.07 ± 1.80 nm, and the C30 length was less than 500 nm. More deacetylation occurred in the amorphous region of chitin than the crystalline part, and the highly deacetylated amorphous region could be soluble in an acidic solution, resulting in the shortening in length of ChNFs (Ji et al. 2022). Therefore, a decrease in length was observed for ChNFs with lower DA. Satam and Meredith (2021) similarly found that the lengths of ChNFs with DAs of 0.89 and 0.79 were similar, but a decrease in length from ⁓890 nm to 428 nm was observed for ChNFs with DA of 0.72. Therefore, the electrostatic repulsive forces from protonation of the amino groups on the chitin surface with acetic acid could facilitate individualization of ChNFs using the high-speed household blending. In addition, Ji et al. (2022) investigated the effects of deacetylation conditions (NaOH concentration, temperature, and treatment time) and found that when the deacetylation reaction became more severe (an increase in NaOH concentration, temperature, and treatment time), the average length of ChNFs significantly reduced. ChNFs extracted from chitin deacetylated with 25 wt% NaOH at the temperature of 110°C for 60 min had the average width and length of 5.3 nm and 417 nm, respectively, while ChNFs with the average width and length of 4.5 nm and 169 nm were obtained from chitin deacetylated with 35 wt% NaOH at 140°C for 100 min. However, no nanofibers were observed for C40, possibly because the severe deacetylation reaction with 40 wt% NaOH mostly converted the acetamido groups to the amino groups in the chitin structure, which made C40 soluble in acidic conditions. To avoid corrosive and hazardous reagents causing the environmental damage, our further study will focus on the preparation of ChNFs from chitin extracted by solvent-free methods such as microbial fermentation or sub- and supercritical water pretreatment (Mohan et al. 2022; Osada et al. 2015).

Figure 7. Morphologies of ChNFs deacetylated with various NaOH concentrations taken by transmission electron microscopy.

Conclusions

This study demonstrates a successful approach for individualization of ChNFs from shrimp shell waste using a combination of the deacetylation reaction and mechanical treatment by high-speed blending. The greater replacement of acetamido groups by the amino groups in the chitin structure, induced by the deacetylation conditions with a higher NaOH concentration, led to a decrease in crystallinity and thermal stability of ChNFs. The morphology of ChNFs also became smaller and shorter. ChNFs deacetylated with 30 wt% NaOH (DA of 0.71) had the degree of crystallinity and T_10% of 90.2% and 224.0°C, respectively, and had widths of 8.07 ± 1.80 nm and lengths of less than 500 nm; however, bundles of aggregated fibers with widths of 30–100 nm were observed for non-deacetylated ChNFs (DA of 0.88) with the degree of crystallinity of 94.6% and T_10% of 293.2°C. The electrostatic repulsive forces generated from protonation of the amino groups on the chitin surface with the help of acetic acid facilitated individualization of ChNFs using high-speed household blending and could reduce the requirement of industrial machines for disintegration. The obtained ChNFs are useful for various applications including composites, packaging, and bioengineering. Nevertheless, it should be noted that the use of 40 wt% NaOH for the deacetylation resulted in the transformation of chitin to chitosan, which was soluble in acidic solutions.

Highlights

• Chitin nanofibers (ChNFs) with an average width of 8.07 ± 1.80 nm and length of less than 500 nm were successfully prepared from shrimp shell waste.

• Effects of deacetylation conditions with various NaOH concentrations (0–40 wt%) on properties of ChNFs were investigated.

• A decrease in crystallinity degree and thermal stability was observed for deacetylated ChNFs.

• Deacetylation facilitated individualization of ChNFs using high-speed household blending.

• The aggressive deacetylation reaction with a higher concentration of NaOH caused a reduction in the length of ChNFs.

Ethical approval

The research does not involve any human or animal-welfare-related issues.

Acknowledgements

This project is funded by National Research Council of Thailand (NRCT) (Grant No. N41A640090). The authors also gratefully acknowledge the support from the department of Materials Science, Faculty of Science, Chulalongkorn University and Learning Institute, King Mongkut’s University of Technology Thonburi. S. Tanpichai would like to thank Thailand Toray Science Foundation for the financial support of the science and technology research grant (28^th TTSF, 2021) and express gratitude to Mstr. Skoravit Tanpichai and Mstr. Monapicha Tanpichai for their invaluable curiosity and support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Research Council of Thailand (NRCT) under Grant No. N41A640090