Introduction

Cotton knitted fabrics (CFs) are comfortable and compatible with human skin as well as offering quick sweat absorption, and thus underwear is mainly produced from cotton or cotton-based fabrics. However, slow sweat-drying properties are often a shortcoming of CFs. Some polyester fabrics and polyester/cotton blended fabrics have been developed to improve sweat-drying rates or quick-drying performance after water or sweat absorption (Yoo et al. 2000; Demiryürek and Uysaltürk 2013; Liu et al. 2013; Kuramoto 2016; Mizuhashi et al. 2016; Atasağun et al. 2018; Dai et al. 2019), despite their lesser compatibility with human skin. Furthermore, fine particles formed by abrasion during laundry treatment of polyester clothes might cause marine micro/nano-plastic problems (Napper and Thomson 2016; De Falco et al. 2018; Fontana et al. 2020; Šaravanja et al. 2022). Therefore, if underwear CFs exhibit both high sweat-absorption properties and quick sweat drying, demand for such CFs might expand as functional underwear for consumers. However, these two aforementioned functions of CFs are commonly in opposition to one another.

Polyesters are more hydrophobic than cotton cellulose. Thus, if suitable quantities of hydrophobic groups are introduced into hydroxy groups of CFs through covalent bonds, the partially hydrophobic substituent-containing CFs are expected to have both of the aforementioned functions with an optimum balance. In this study, three acetylation reactions under various conditions were applied to a CF sample to improve quick drying after water absorption. First, the original fabric structure should be maintained after acetylation; i.e., heterogeneous acetylation to the CF is required. Second, the degrees of substitution (DS) of the acetyl groups should be as low as possible for maintaining hydrophilicity or sweat absorption.

The following three acetylation systems were used in this study: (Ac-I) acetic anhydride (Ac2O), conc. H2SO4, and toluene system (used as the acetylation reagent, esterification catalyst, and organic solvent without dissolution in the system, respectively; Watanabe et al. 1968; Taira et al. 2020); (Ac-II) Ac2O, conc. H2SO4, acetic acid (AcOH), and water system (in which water helps prevent dissolution of the cotton knitted fabric in the acetylation medium; Tsuji et al. 1960a, 1960b; Shaikh et al. 2009; Tatsumi and Inoue 2016); and (Ac-III) Ac2O and sodium acetate (AcONa) system (used as the acetylation reagent/solvent and base catalyst, respectively, without using another solvent; Koroskenyi and McCarthy 2001). The quantities of chemicals and solvents, as well as the acetylation temperatures and reaction times, were varied to prepare acetylated cotton knitted fabrics (AcCFs). The yields, DS values, viscosity-average degrees of polymerization (DPv), and moisture contents at 20 °C and 65% relative humidity (RH) of the AcCFs, were measured, and their fiber surface morphologies, X-ray diffraction (XRD) crystallinities, distributions of acetyl groups, and thermogravimetric (TG) properties were analyzed. The obtained results are discussed for explaining the mechanism of quick water-drying properties of the AcCFs.

Many studies on partial acetylation of cellulose nanofibrils (CNFs) have been conducted in terms of improving their thermal stabilities and compatibilities in melt compounding with synthetic polymers (Sofla et al. 2019; Huang et al. 2019; Beaumont et al. 2020; Koso et al. 2022). Acetylation of cellulose nanocrystal (CNC) surfaces has also been investigated to add a surface-hydrophobic nature and thus improve their compatibility with hydrophobic polymers for good reinforcement. This is because in the literature, partial acetylation of the hydroxy groups of CNFs improves their thermal degradation temperatures from ~ 300 °C for the original CNFs to ~ 350 °C (suitable for melt compounding with synthetic polymers; Agustin et al. 2017; Yano et al. 2018; Lee et al. 2020; Semba et al. 2021), although the reported acetylation reactions were performed with organic solvents such as N-methyl-2-pyrolidone. However, no studies on acetylation of CFs to improve quick drying after water absorption have been reported, to the present authors’ knowledge.

Materials and methods

Samples

Cotton knitted fabrics (100%) with the yarn count, fabric weight, and thickness of 40/1, 190 g/m2, and 1.1 mm, respectively. An alkali-treated CF was prepared by soaking the original CF in 18% NaOH at 60 °C for 6 min under 120–130% uniaxial stretching, followed by thorough washing with water until the washing effluent was neutral (Utsumi 1965; Wakida et al. 2002). These CFs were heated in water containing a nonionic surfactant (1 g/L; RS-400, Sanyo Kasei Kogyo, Kyoto, Japan) and Na2CO3 (5 g/L) at 70 °C for 20 min followed by thorough washing with water. Then, surfactant-treated CF samples that had not been dried were bleached in H2O2-containing water (3.8 g/L) at 70 °C and pH 10.6 (adjusted with dilute aqueous NaOH) for 20 min followed by thorough washing with water. These treatments were performed to remove impurities and knitting oils present in the original fabrics. The samples were dried at 20 °C and 65% RH for > 1 d until no change in mass was detected. Figure S1 in the Electronic Supplementary Material shows scanning electron microscopy (SEM) images of the original and alkali-treated CFs. Ac2O, 98% H2SO4, AcONa, and toluene were of special grades (Fujifilm Wako Chemicals, Osaka, Japan). Glacial AcOH (> 99.7%) was obtained from Alfar Aesar (Haverhill, MA, USA). These chemicals were used as received.

Acetylation with Ac2O/H2SO4/toluene (Ac-I) system

Cotton knitted fabrics or alkali-treated CF (9.5 g by dry mass) and AcOH (200 g) were placed in a stainless dyeing pod (440 mL). The lid of the pod was closed and the pod was rotated at 30 °C for 30 min with a mini-color dyeing system (12EL, TEXAM, Nagoya, Japan) as pretreatment. The AcOH-containing fabric was picked up from the pod and squeezed to decrease the AcOH/fabric mass to ~ 23.75 g (i.e., ~ 150% pick-up rate) with a padding machine. This AcOH-containing cotton knitted fabric was placed in the dyeing pod; then, AcOH (200 g) and H2SO4 (0.5 g) were added to the pod. The pod was rotated at 30 rpm and 30 °C for 1 min with the mini-color dyeing system for mixing. The AcOH/H2SO4-containing fabric was picked up from the pod and squeezed to decrease the AcOH/H2SO4/fabric mass to ~ 23.75 g with the padding machine, then placed in a glass reactor. Ac2O (77 g) and toluene (185 g) were added to the glass reactor, and the mixture was heated at 30 °C from 10 min to 18 h. The AcCF was picked up from the reaction mixture and washed thoroughly with ethanol, water, and hot water, in that order. The wet AcCFs were dried at 40 °C for 1 h followed by drying at 20 °C and 65% RH for > 1 d.

Acetylation with Ac2O/H2SO4/water (Ac-II) system

Cotton knitted fabrics (3.6 g by dry mass), AcOH (1.26 g), and water (52.8 g) were placed in the stainless dyeing pod, and the pod was rotated at 25 °C for 1 h with the mini-color dyeing system for mixing. The AcOH/water/fabric sample was picked up from the pod and squeezed to decrease the mass to ~ 7.2 g (i.e., AcOH/water/fabric mass ratio of ~ 0.08/3.52/3.6) with the padding machine. Then, the AcOH/water/fabric sample was added to a mixture consisting of designated quantities of Ac2O, H2SO4, AcOH, and water in the pod. The mixture was rotated for acetylation under various conditions (Table S1 of the Electronic Supplementary Material). The AcCFs were picked up and washed thoroughly with water, hot water, and water in that order. The wet AcCFs were heated at 40 °C for 1 h followed by drying at 20 °C and 65% RH for > 1 d.

Acetylation with Ac2O/AcONa (Ac-III) system

To the stainless dyeing pod (1 L), CF (15 g) as well as designated quantities of AcONa and water were added. The pod was rotated at 25 °C for 15 min with the mini-color dyeing system for mixing. The water/AcONa/fabric sample was picked up from the pod and squeezed to decrease the mass to ~ 34.5 g with the padding machine, and the water in the mixture was largely removed at 120 °C for 3 min with a pin tenter (OPT-1A, Tsuji Dyeing Machine MFG Osaka, Japan). The dried AcONa/fabric sample was soaked in Ac2O (324 g) for 20 s in a glass beaker. The Ac2O/AcONa/fabric sample was squeezed to decrease the mass to ~ 37.5 g and placed in polyethylene/aluminum double sealing bags. The mixture was reacted under various conditions to prepare AcCFs followed by washing thoroughly with water, hot water, and water in that order. Table S1 in the Electronic Supplementary Material shows detailed acetylation conditions. The samples were heated at 40 °C for 1 h followed by drying at 20 °C and 65% RH for > 1 d.

Evaluation of quick drying of wet fabrics

Quick drying of the wet fabrics was measured from the volatile residual water contents (BOKEN Test Standard BQE A 028) (Kuramoto 2016). Deionized water (0.6 mL) was added to a fabric sample (10 cm × 10 cm); then, the time-dependent water/fabric masses were monitored at 20 °C and 65% RH. The residual water content of the fabric sample was obtained with Eq. (1),

$${\text{Residual water content }} = W_{{\text{t}}} /W_{0} \, \times {1}00 \, \left( \% \right),$$
(1)

where Wt is the water content (g) after time t (min), and W0 is the water content (g) immediately after water addition. The quick-drying performance of fabrics was defined as the t (min) when the residual water content decreased to 10 w/w%.

Analyses of fabric samples

The moisture content of AcCFs conditioned at 20 °C and 65% RH was measured after drying the conditioned samples in an oven at 105 °C for 2 h in accordance with JIS L-1096 8.10. Degrees of substitution (DS) of the AcCFs were determined by saponification and back-titration in accordance with ASTM 817–12. Yields of the AcCFs were calculated from the mass recovery ratios of the AcCFs, considering their DS values in accordance with Eq. (2),

$${\text{Theoretical recovery mass }}\left( {\text{g}} \right) = [{162}.{1 } - \, (DS \times {1}.0) \, + \, (DS \times {43}.0)] \times \left[ {{\text{dry mass }}\left( {\text{g}} \right){\text{ of starting cotton fabric}}/{162}.{1}} \right].$$
(2)

where 162.1, 1.0, and 43.0 are the gram/mol values of glucosyl unit, proton, and acetyl group, respectively.

The air-dried fabric sample was cut into short fibers with scissors, and the sample (~ 0.04 g by dry mass) was dissolved in 0.5 M copper ethylenediamine (EDA) hydroxide [Cu(EDA)2(OH)2, 20 mL] by stirring the mixture for ~ 30 min. The intrinsic viscosities of the solutions were measured at 25 °C with a Cannon–Fenske capillary viscometer (ISO/FDIS 5351 2010). The obtained intrinsic viscosity [η] values were calculated to the corresponding DPv values with the Mark–Houwink–Sakurada equation: [η] = 0.909 × DPv0.9 (Marx 1955; Isogai et al. 1989; Ono et al. 2023). Surface SEM images of the AcCFs were obtained for gold-spattered samples at 15 kV (TM 4000 Plus II, Hitachi-High-Tech, Tokyo, Japan). X-ray diffraction (XRD) patterns of the AcCFs were obtained with Cu Kα radiation (XRD-6100, Shimadzu, Kyoto, Japan) from 5° to 45° at diffraction angle 2θ, 40 kV, and 30 mA. Crystallinities of cellulose I for the AcCFs were calculated from the XRD patterns by the following three methods: Method A (Segal et al. 1959), Method B: the relative diffraction intensity of the (1 − 1 0) peak to the background (Isogai and Usuda 1990), and Method C: peak deconvolution/least squares/peak-area calculation methods using Gaussian and Lorentzian functions (Ono et al. 2021, 2022) (Fig. S2 in the Electronic Supplementary Material), which was slightly different from those reported by Yao et al. (2020). Thermogravimetric (TG) curves of the AcCFs (~ 8.5 mg each) were obtained (DTG-60, Shimadzu, Kyoto, Japan) from room temperature to ~ 600 °C at 10 °C/min under an N2 gas stream at 300 mL/min. PeakFit (ver. 4.12; Hulinks, Tokyo, Japan) was used for peak deconvolution. X-ray fluorescence analysis (XFA; ZSX-Primus IV, Rigaku, Tokyo Japan) was used for elemental analysis of the AcCFs (φ 30 mm).

Distribution of acetyl groups in the acetylated fabrics and fibers

A red reactive dye (5% by dry mass of sample; Cibacron scarlet LS-2G, Ciba Specialty Chemicals, Co. Ltd., Tokyo, Japan) was applied to the AcCFs with Na2CO3 (20 g/L) and Na2SO4 (80 g/L) at 60 °C for 30 min, followed by soaping with water containing a detergent (2 g/L) at 60 °C for 10 min and thorough successive washing with water. The L*a*b* color coordinates of the dye-treated cotton knitted fabrics were measured with a spectrophotometer (CM-3600d, Konica Minolta, Tokyo, Japan). The AcCF samples were thermally pressed with poly(propylene) (PP) pellets at 200 °C for 20 s to embed the AcCF in the PP matrix. Cross sections of the AcCF/PP composite were cut with a diamond knife. The degree of C = O absorption in the cross sections was measured at several points by microscopic Fourier-transform infrared (FTIR) spectroscopy (Spotlight 400-attached PerkinElmer Frontier IR, Waltham, MA, USA). An imaging attenuated total reflectance (ATR) probe was used at 8-cm−1 resolution and 1.56-μm pixel size.

Results and discussion

Acetylation of cotton knitted fabrics

The CFs were heterogeneously acetylated to partially introduce hydrophobic acetyl groups into cellulose, maintaining the original fabric structure. Three acetylation methods [i.e., Ac-I) the Ac2O/H2SO4/toluene, Ac-II) Ac2O/H2SO4/AcOH/water, and Ac-III) Ac2O/AcONa systems] were applied to the CFs (Figs. 1 and S3, and Table S1 in the Electronic Supplementary Material). The quick water-drying properties of the AcCFs were studied in terms of their yields, DS values, DPv values, moisture contents, fiber surface morphologies, distributions of acetyl groups, XRD crystallinities, and TG behavior. In the Ac-I and Ac-III systems, the original fabric structures were maintained after acetylation. In the Ac-II system, the CFs were dissolved in the acetylation medium when the CF/water mass ratio was 1/0–1/0.9. When the CF/water mass ratio was 1/1.2–1/6, the AcCFs were fine powders. Thus, the CF/water mass ratio was adjusted to 1/7.5 and 1/9 in the Ac-II system (Fig. 1).

Fig. 1
figure 1

Scheme to prepare acetylated cotton fabrics from knitted cotton fabric or alkali-treated cotton knitted fabric with three acetylation systems including characterization of products

Full size image

The morphologies of the CF surfaces were observed by SEM (Fig. 2). Fine cotton cellulose fibrils were observed on all of the fiber surfaces. The AcCF sample with DS 1.32 prepared with the Ac-I system exhibited smooth fiber surfaces almost similar to those of the original CF. However, the fiber surfaces of the AcCF sample with DS 2.10 were partially damaged. The AcCF sample with DS 0.07 prepared with the Ac-II system exhibited smooth fiber surfaces, whereas that with DS 0.10 exhibited substantially damaged fiber surfaces. The use of conc. H2SO4 and water in the Ac-II system may have caused such fiber damages with substantial depolymerization, as discussed in the following section. The fiber surfaces of the AcCF sample with DS 0.28 (prepared with the Ac-III system) were mostly smooth. Thus, the fiber surfaces were damaged depending on the acetylation system and the obtained DS value.

Fig. 2
figure 2

SEM images of fiber surfaces of original cotton knitted fabric and acetylated cotton fabrics prepared with the three systems shown in Fig. 1

Full size image

Yields as well as DS and DP v values of the acetylated cotton fabrics

The mass recovery ratios of the AcCFs were plotted against their DS values (Fig. S4 in the Electronic Supplementary Material). The mass recovery ratios roughly increased with increasing DS value. The yields were calculated from the mass recovery ratios of the AcCFs and their DS values in accordance with Eq. (2), and thus should be ≤ 100% (Fig. 3a). The yields of the AcCFs prepared with the Ac-I system were almost 100% for the entire range of DS values. However, the AcCFs samples that were prepared with the Ac-II and Ac-III systems were prone to decreasing yield with increasing DS. Cellulose acetates with DS values of 0.3–0.9 are water-soluble (Kamide et al. 1981, 1987; Miyamoto et al. 1985; Iijima et al. 1992; Gomez-Bujedo et al. 2004; Cao et al. 2016; Pang et al. 2016). Thus, cellulose acetate molecules with low DS values in the AcCFs might have become increasingly water-soluble, resulting in decreased yield.

Fig. 3
figure 3

Relationships between degree of substitution (DS) of acetylated cotton fabrics and a yield or b viscosity-average degree of polymerization (DPv)

Full size image

The cellulose acetates prepared in this study have heterogeneous chemical structures consisting of cellulose and partially acetylated cellulose molecules with various ratios. Furthermore, the acetate ester groups might have been partially or mostly removed from the cellulose acetate molecules during dissolution in strongly alkaline 0.5 M Cu(EDA)2(OH)2. Thus, the DPv values obtained in this study exhibited approximate DP values. All the samples were dissolved in 0.5 M Cu(EDA)2(OH)2, and reliable viscosities of the solutions with small standard deviations were obtained (Fig. 3b). The AcCFs roughly decreased in terms of DPv with increasing DS. AcCFs prepared with the Ac-III system gave higher DPv values than those of the other AcCFs at similar DS values. The DPv values of AcCFs with DS values lower than 0.14 prepared with the Ac-II system substantially decreased from 2300 for the original CF to 300–400. Thus, H2SO4 (the acetylation catalyst) corresponded to depolymerization. In particular, when both H2SO4 and water were in the acetylation system, substantial acid hydrolysis was unavoidable.

Figure S5a in the Electronic Supplementary Material shows the relationship between the reaction time at 30 °C in the Ac-I system and the DS values of the AcCFs. The DS value increased with increasing reaction time. The alkali-treated cotton fabric gave higher DS values at the same reaction time. Figure S5b shows the relationship between the mass ratio of water/CF in the Ac-II system and the DS values of the AcCFs. At low water/CF mass ratios of < 0.9, the DS values of the AcCFs were higher than 1.2. However, these samples were soluble in the reaction media, and the original fabric structure was lost. When the water/CF mass ratios were > 7.5, the original fabric structures were maintained. Figure S6 shows relationships between the AcONa/CF mass ratios, reaction time at 80 °C, or reaction temperature for 4 h versus the DS values of the AcCFs prepared with the Ac-III system. As the AcONa/CF mass ratio increased or the reaction temperature increased, the DS value of the products increased. The reaction time at 80 °C had almost no influence on the DS values within the tested reaction times.

Quick drying of acetylated cotton fabrics after water absorption

Figure 4 shows the moisture contents of the AcCFs prepared under various conditions after conditioning at 20 °C and 65% RH. The moisture contents decreased with increasing DS of the AcCFs. Partial introduction of hydrophobic acetyl groups into the CF resulted in partial reduction of the hydrophilic nature of the original CF. Figure S7 in the Electronic Supplementary Material shows the relationships between the DPv values of the AcCFs and their moisture contents. No clear relationship between these two factors was obtained.

Fig. 4
figure 4

Relationship between DS of acetylated cotton fabrics and moisture contents at 20 °C and 65% RH

Full size image

Figure 5 shows the relationship between the DS values of the AcCFs and the drying times to 10% water content after water absorption, used as an index of quick drying. The target drying time was 58 ± 3.3 min, obtained from a polyester fabric. Some AcCFs with DS values of 0.06–0.12 prepared with the Ac-I and Ac-II systems exhibited drying times of < 65 min, which is close to that of the polyester fabric. Thus, quick drying similar to that of the polyester fabric after water absorption was achieved by partial acetylation of the CF under suitable conditions. AcCFs prepared with the Ac-III system exhibited longer drying times or inferior quick drying after water absorption than the other AcCFs.

Fig. 5
figure 5

Relationships between DS values of acetylated cotton fabrics versus drying times to 10% water content at 20 °C and 65% RH in drying rate tests of wet fabrics

Full size image

Figure S8a shows the relationship between the moisture content of the AcCFs at 20 °C, 65% RH, and a drying time to 10% water content after water absorption. Some of the AcCFs prepared with the Ac-I and Ac-II systems exhibited moisture contents of 6.5–7.4%, similar to that of the original CF. Thus, AcCFs prepared with the Ac-I and Ac-II systems under suitable conditions exhibited quick drying after water absorption, whereas they absorbed moisture similarly to that of the original CF. These properties might be beneficial as underwear for good comfort and quick sweat absorption, although their low DPv values (and probably low mechanical strengths) prepared with the Ac-II system (Fig. 3b) should be considered during production.

Figure S8b in the Electronic Supplementary Material shows the relationship between the DPv values of the AcCFs versus drying time to 10% water content after water absorption. It was difficult to obtain a conclusion from the results in Fig. S8b because of the insufficient number of data points.

Distribution of acetyl groups in AcCFs and their fibers

Figure S9 shows photographs of reactive red dye-treated AcCFs, and Fig. S10 shows the relationship between the optical a* values of the AcCFs versus either DS values or drying time to 10% water content after water absorption. The a* value decreased with increasing DS of acetyl groups of the samples, because the reactive red dye molecules cannot react with or adsorb on acetyl groups. However, no clear relationships were obtained in Fig. S10. This is probably because the acetylation used in this study was a heterogeneous liquid/fabric reaction, and the water drying time might be influenced not only by acetyl groups on the fabric surface but also those inside the fabric.

Then, the distributions of acetyl groups across each fiber width for the AcCFs were analyzed in terms of the fiber cross sections by microscopic FTIR–ATR, in which the infrared (IR) absorption ratios of A1730/A1030 were used as markers of degrees of acetylation. The IR absorption intensities at 1730 and 1030 cm−1 were regarded as peaks ascribed to the acetyl C = O group and cellulose backbone in the AcCFs, respectively (Fig. S11 in the Electronic Supplementary Material). The distribution was not straightforward; AcCFs prepared with the Ac-I and Ac-III systems exhibited rather heterogeneous distributions of acetyl groups between one surface and the opposite surface sides. In contrast, AcCFs prepared with the Ac-II system exhibited homogeneous distributions of acetyl groups between one surface and opposite surface sides. The aforementioned homogeneous distributions might correspond to high quick-drying performance (Fig. 5).

XRD patterns, crystallinities, and crystal widths of acetylated cotton fabrics

The alkali-treated CF was used for acetylation only in the Ac-I system. This is because, in the initial stage of this study, the alkali-treated CF was expected to have higher DS values than the original CF by acetylation. The DS values for the alkali-treated CF after acetylation were higher than those of AcCFs prepared from the original CF at the same reaction time (Fig. S5a), yet the differences in DS between the two CFs were insubstantial. This is probably because the crystallinity of cellulose I of the alkali-treated CF was ~ 81%, which was close to that (~ 85%) of the original CF, when calculated by the Method A (Fig. S2). However, a small diffraction peak owing to the (1–10) plane of cellulose II was detected in the XRD pattern of the alkali-treated cotton fabric (Fig. S12). The original CF was then used in the other acetylation systems.

Figure 6 shows XRD patterns of the AcCFs prepared with the Ac-I system. Figures S13 and S14 in the Electronic Supplementary material show those prepared with the Ac-II as well as Ac-III systems, respectively. The original cellulose I crystal structures were mostly maintained for AcCFs with DS values of ≤ 1.3; heterogeneous acetylation of the CF mostly proceeded on the crystalline cellulose microfibril surfaces over the aforementioned DS range. The AcCF sample with DS 2.20 mainly exhibited a cellulose I-type crystal structure, although a small diffraction peak at ~ 9° because of a cellulose triacetate I crystal structure was detected (Watanabe et al. 1968; Sikorski et al. 2004; Wada and Hori 2009; Taira et al. 2020).

Fig. 6
figure 6

XRD patterns of acetylated cotton fabrics with various DS values prepared with the Ac2O/H2SO4/toluene (Ac-I) system

Full size image

Figure 7a and b show the crystallinities of cellulose I for the AcCFs (measured by the three methods) and the crystal widths calculated from the full width at half height of the (2 0 0) peak with Scherrer’s equation, respectively. The crystallinities measured by the three methods gave similar results. The original crystallinities were mostly maintained up to DS 0.28, and then linearly decreased with increasing DS. Thus, some of the crystalline regions in the original CF became disordered with increasing DS of acetylation. In contrast, the crystal widths were mostly unchanged over a wide DS range of ≤ 1.32.

Fig. 7
figure 7

Relationships between the DS values of acetylated cotton fabrics versus a crystallinities of cellulose I, calculated by three methods, or b crystal widths of cellulose I

Full size image

Figure S15 in the Electronic Supplementary Material shows the relationship between the DPv values of the AcCFs and their crystallinities with cellulose I. No clear relationship was obtained. Based on the results for the AcCFs (Fig. 7), Fig. 8 shows a schematic of a single cellulose microfibril in the cotton fabric during heterogeneous acetylation. Over the DS range of ≤ 1.32, the original crystal width of ~ 6 nm was maintained, whereas crystalline regions of cellulose decreased for acetylated fabrics with a DS of > 0.28. The DPv value decreased with increasing DS of the AcCFs.

Fig. 8
figure 8

Schematic of heterogeneous acetylation of a single cotton cellulose microfibril consisting of crystalline and disordered regions periodically present along the longitudinal direction in a cotton fabric, maintaining the original crystal width but decreasing crystallinity and DPv by acetylation

Full size image

TG of acetylated cotton fabrics

TG analysis was applied to the CF and alkali-treated CF sample (Fig. 9a). The CF and alkali-treated CF exhibited almost the same TG curves, and exhibited a temperature of ~ 330 °C at 5% mass loss by thermal degradation.

Fig. 9
figure 9

TG curves of a original cotton and alkali-treated cotton fabrics; and acetylated cotton fabrics prepared with the b Ac2O/H2SO4/toluene (Ac-I) system, c Ac2O/H2SO4/AcOH/water (Ac-II) system, and d Ac2O/AcONa (Ac-III) system

Full size image

Regarding AcCFs prepared with the Ac-I system, the thermal degradation temperatures clearly decreased to ~ 250 °C for DS values from 0.06 to 1.32 (Fig. 9b). Thus, these AcCFs were thermally unstable. When AcCFs with DS values of ≤ 0.14 were prepared with the Ac-II system, the TG curves were almost the same as that of the original CF. The AcCF sample with DS 1.18 was somewhat unstable to thermal degradation (Fig. 9c). In contrast, AcCFs prepared with the Ac-III system clearly exhibited increased thermal degradation temperatures (Fig. 9d). Figure S16 shows the relationships between the DS values, moisture contents, crystallinities of cellulose I, or DPv values of the AcCFs versus their temperatures at 5% mass loss in the TG curves. No clear relationships were obtained between any of the four factors in Fig. S16 and the thermal degradation temperatures.

It has been reported that CNCs containing protonated sulfate ester groups exhibit lower thermal degradation temperatures of < 200 °C (Kargarzadeh et al. 2012; Vanderfleet et al. 2019, 2022; D’Acierno et al. 2020), as in the case of the AcCFs prepared with the Ac-I system in this study. Thus, the relative sulfur contents of the original CF, two AcCFs with DS values of 0.06 and 1.32 prepared with the Ac-I system, as well as two AcCFs with DS values of 0.10 and 0.11 prepared with the Ac-II system were analyzed by XFA. The two AcCFs prepared with the Ac-I system had sulfur contents of 4.1 and 3.5 µg/cm2, whereas the other two AcCFs and the original CF had lower sulfur contents of 0.1–0.3 µg/cm2 (Fig. S17). Thus, the results of the low thermal degradation temperatures of the AcCFs prepared with the Ac-I system correspond to the presence of thermally unstable protonated sulfate ester groups. The Ac-I system introduced not only acetyl ester groups but also sulfate ester groups into CF, whereas the Ac-II system caused no sulfate ester formation, irrespective of the presence of H2SO4 in the acetylation system. The higher DPv values of the AcCFs with the Ac-III system might correspond to the higher thermal degradation temperatures compared with those prepared with the Ac-II system, although these samples contained almost no sulfate ester groups. As a consequence, AcCFs prepared with the Ac-III system at 80 °C for 4 h have higher thermal degradation temperatures than that of the original CF, and have high DPv values that are close to that of the original CF.

Conclusions

AcCFs with DS of ≤ 0.5 and yields of > 80% were prepared with three heterogeneous acetylation systems. The AcCFs prepared with the Ac-III system exhibited high DPv values of > 1500, whereas those prepared with the Ac-II system exhibited low DP values of ≤ 400. Acid hydrolysis of the cellulose molecules was evident during acetylation with H2SO4 and water at 40 or 80 °C. The moisture contents of the AcCFs at 20 °C and 65% RH decreased from 7.1% to ~ 4.7% with increasing DS up to 0.46; introducing hydrophobic acetyl groups into the CFs decreased their hydrophilic nature. Quick-drying performance was achieved for some AcCFs with DS values of < 0.2, prepared with the Ac-II system, similar to that of a polyester fabric. The homogeneous distributions of acetyl groups across each fiber width in the AcCFs prepared with the Ac-II system might correspond to substantial quick drying. However, the DPv values of these AcCFs substantially decreased to < 400, and their mechanical properties might be insufficient for use in commercial underwear. The crystal widths of cellulose I for AcCFs with DS values of ≤ 1.38 were ~ 6 nm, and were unchanged from that of the original CF. The crystallinities of cellulose I were unchanged for AcCFs with DS values of ≤ 0.28. Neither the crystallinities nor the crystal widths of cellulose I for the AcCFs corresponded to quick drying after water absorption. Thermal degradation of the AcCFs depended on their DPv values and/or the presence of protonated sulfate ester groups in the AcCFs.