Introduction

Currently, cellulosic materials are widely used in various industries, due to their abundance, biocompatibility and biodegradability (Alves et al. 2015; Alemdar et al. 2008; Hamawand et al. 2020). Hence, the production of microcrystalline cellulose and nanocellulosefrom various sources has garnered much attention in both scientific and industrial sectors. This is primarily due to their mechanical and thermal properties, as well as cost-effectiveness and versatility (Sá et al. 2015; Rambabu et al. 2016). The extremely ordered structure of CNCs ensures their high strength, which preconditions their use as a component in reinforcing systems (Kvien et al. 2005; Oksman et. al 2006; Bondeson et al. 2006). In addition, the application of CNCs as a filler in polymeric materials improves the mechanical properties of said materials and helps to control humidity, optical and sorption properties, and biodegradability (Belén et al. 2017; Abdul Khalil et al. 2014). Additionally, the CNC suspension can self-assemble to form left-handed chiral liquid crystals as it reaches critical concentration (Gray 2016; Sun et al. 2019), the helix structure of which can be maintained in the dried film through the slow evaporation of water (Edgar et al. 2001). As a result, the solid CNC film exhibits unique optical properties. Moreover, the presence of incident light wavelengths in the visible range causes the material to appear colorful to the naked eye (Blaise et al. 2017). To date, research on this optical film has aroused far-ranging interest, such as in the fields of optical anti-counterfeiting, sensors, and composite devices (Lin et al. 2019; Kim et al. 2020). This has lead to the application of CNC film in the pharmaceutical, food, electronic and chemical industries.

Plant-based materials are the main sources of cellulosic components, which include annual plants, agricultural wastes, bacteria and molds (Brown et al. 2000; Boerjan et al. 2003). Materials rich in cellulose, hemicelluloses and lignin are commonly referred to as lignocellulosic biomass (Carrier et al. 2011). Wastes such as straw, the husks of rice and corn, sugarcane pulp, nypafrutican stems, the fibers of pineapple and banana leaves, and coconut shells are typical sources of lignocellulosic biomass (Boerjan et al. 2003; Gibson et al. 2010; Lwako et al. 2013; Sumira et al. 2020; Song et al. 2009; Chen et al. 2015; Jayaraj et al. 2014; An et al. 2020).

Kazakhstan has an urgent need for the replacement of cellulose from trees as a raw material used in the production of paper, which would be an excellent application of lignocellulosic biomass from agricultural waste. Further, agriculture is an important part of the economy of Kazakhstan, and generates large quantities of waste every year, which is mostly neither reused nor recycled.

An important source of lignocellulose biomass is SFH, which is a by-product from the preparation of sunflower seeds for oil extraction, forming about 14% of the total seed volume of sunflower seeds of Kazakshtani production. SFH is a lignified plant tissue of cylindrical shape of up to 20–70 mm in length and approximately 8 mm in diameter, having a density of up to 1.2 thousand kg per cubic meter and a humidity level of up to 8%. According to some studies, the cellulose content in SFH is 31–42.4% of the total mass (Ovcharov et al. 2010; Khusid et al. 2015).

Sunflower farming is one of the leading branches of agriculture in Kazakhstan, and the area under cultivation for the purpose of oil extraction has been growing steadily, now exceeding one million hectares, which amounts to around 5% of all agricultural land in the country. Due to the high demand for sunflower seeds, their processing generates huge amounts of waste, the disposal of which in large quantity is both expensive and complicated. The method utilized in oil extraction is known to produce approximately 14–25 kg of husks from every 100 kg of sunflower seeds (Harkov et al. 2018). Moreover, the use of SFHs in their pure form as animal feed is undesirable, due to the high content of fiber, being approximately 50%, because it is poorly absorbed in the gastrointestinal tract (Khusid et al. 2015). Hence, the processing of large amounts of SFHs is a serious problem that needs to be resolved immediately, as much of said waste is simply burned as fuel.

There are a number of studies that have been carried out on the production of nanocellulose from agricultural waste (Boerjan et al. 2003; Gibson et al. 2010; Song et al. 2009; An et al. 2020; Bongao et al. 2020; Israel et al. 2008; Yandan et al. 2015). In addition to rice, straw and the aforementioned husks (Boerjan et al. 2003; Gibson et al. 2010; Lwako et al. 2013; Sumira et al. 2020), attention has been paid to the residue of coniferous wood, inclusive of wood chips, pine branches and needles, with a target production yield of more than 13% (Rosana et al. 2016; Kunaver et al. 2016). Furthermore, among the plants that could be cultivated in Kazakhstan, special consideration should be given to hemp, with a cellulose content in the plant stems that exceeds 50–70% (Jean et al. 2010). Analysis of the applicable literature shows that there are few studies related to the processing of SFHs into nanocellulose. Nevertheless, SFHs have been studied with regards to their use as functional animal feed additives (Khusid et al. 2015), carbon adsorbents for the removal of oil from water (Ovcharov et al. 2010), fuel (Perea-Moreno et al. 2018; Zajemska et al. 2017), fertilizers (Dmytro et al. 2021), biosorbent for the extraction of metals (Saleh et al. 2016; Tomasz et al. 2021), and agar medium (Khan et al. 2005), as well as the manufacture of ceramics from SFH ash (Quaranta et al. 2016). One of the few studies devoted to the production of nanocellulose using SFHs involved pretreatment with sodium hydroxide, bleaching with sodium hypochlorite, ethanol precipitation, acid hydrolysis with sulfuric acid, and finally dialysis in order to obtain nanocellulose (Chen et al. 2015). Of note is that alkaline treatment and bleaching with sodium hypochlorite significantly increase the formation of liquid waste.

Therefore, in the pretreatment of biomass, the use of the environmentally friendly, one-step organosolvent oxidation method is best. This method does not require any additional bleaching process, as the chlorine-containing bleaching reagents produce cellulose material via a cyclic method using the existing peroxyacetic acid (PAA), which can be reused several times (Barbash et al. 2020). In studies, high-yield cellulose materials have been obtained from annual plants using PAA (Vurasko et al. 2014). However, the use of concentrated acetic acid and 30% hydrogen peroxide to produce PAA in this method clearly prevents it from being considered as a completely environmentally safe method. Further, no systematic studies on the extraction of MCC and nanocellulose from SFHs using green technology have been conducted.

Considering these factors, the main purpose of this study is to obtain MCC from SFHs using a more environmentally friendly method, as well as to establish the optimal value of the delignification agent. The present study continued onto the creation of CNCs using acid hydrolysis of MCC, followed by the production of CNC film.

Experimental section

Materials

Hydrogen peroxide, H2O2, 15%, acetic acid (CH3COOH) ≥ 55% and ethanol 96% (C2H5OH) were obtained from Alita LLP (Kazakhstan). Sulfuric acid (98% H2SO4), sodium hydroxide (≥ 99% NaOH), hexane (99% C6H14) were obtained from Sigma-Aldrich (Bangalore, India). SFHs were obtained from a local oil refinery (Sei-Nar LLP) located in Ust-Kamenogorsk, Kazakhstan. All other reagents were of analytical grade and were used without additional purification.

Methods

Sample preparation

SFHs were washed in hexane using a Soxhlet extractor for 2 h in order to eliminate unsaturated high fatty acids, such as stearic and palmitic acids, among others. The washed SFHs were placed in a drying cabinet at 50 °C for 6 h, until any hexane had completely evaporated.

Preparation of peroxyacetic acid (PAA)

The detoxifying agent, PAA, was prepared in accordance with the method utilized by Vurasco et al. (2014), using concentrated sulfuric acid as a catalyst in a ratio of 1.5:1 (VmL/VmL) to 55 wt% glacial acetic acid (ρ = 0.5836 mg/cm3), which was then combined with 15 wt% (ρ = 1.055 mg/cm3) hydrogen peroxide.

Equilibrium peroxyacetic acid analysis

The concentration of PAA obtained was determined using the following formula:

$${\text{C}}_{{{\text{PAA}}}} = {\text{V}}_{{1}} {\text{N}}_{{1}} {\text{mEq1}}00/{\text{V}}_{{{\text{PAA}}}} {1}0\rho$$
(1)

where V1 is the volume of sodium thiosulfate used for sample titration (328 mL); N1 is the normality of sodium thiosulfate concentration (0.1 N); mEq is the milliequivalent of PAA (0.038 mg/mol); VPAA is the volume of the sample taken for titration (1 mL); and, ρ is the density of PAA (1.04 g/cm3) (Vurasco et. al 2014).

Extraction of microcrystalline cellulose (MCC)

In order to obtain MCC from SFH, an effective ratio of the raw material and the delignification agent had to be determined. MCC was obtained using 10 g of SFHs combined with PAA (g/mL) in the following ratios: 1/8, 1/10, 1/12, 1/14, 1/16, 1/18, 1/20, 1/22. Cellulose extraction was carried out by boiling the SFHs and PAA in a flask with a rotary condenser at 90 °C for 2 h, with continuous intensive magnetic stirring (Fig. 1). The MCC obtained was cooled to 25 °C, run through filter paper, and finally neutralized by washing with distilled water until pH was equal to 7. The neutralized pulp was dried at 60 °C for 6 h, until the mass reached a stabilized consistency. The dried material was then stored in a desiccator.

Fig. 1
figure 1

Diagram representing the synthesis of MCC via organosolvent oxidation

Full size image

Determination of MCC quality indicators

The moisture content of the fully dried MCC was determined by the ASTM D 1348–94 (2008) test protocol, β-cellulose content by ASTM D 588 (2000), residual lignin by ISO/DIS 21436, and hemicellulose content by ASTM D 5896–96 (2019) e1. In order to determine the ash content of cellulose (SiO2), 3 g MCC was fired in a muffle furnace (SNOL 8.2/1100 L Lithuania) at 700 °C for 90 min. The mass of ash (SiO2) was measured on an analytical balance (Sartogosm LV 210-A, Russian) until an average value was obtained.

MCC yield was calculated using the following formula:

$${\text{Yield }}\left( \% \right) = \, \left( {{\text{m}}_{{{\text{SFH}}}} - {\text{m}}_{{{\text{MCC}}}} } \right)/{\text{m}}_{{{\text{SFH}}}} \times {1}00\%$$
(2)

where mSFH is SFH mass; and, mMCC is the mass of the MCC obtained.

Sulfuric acid hydrolysis of nanocellulose

CNCs were obtained by way of acid hydrolysis of the resulting MCC using sulfuric acid (60 wt%) via the following procedure: 1 ± 0.01 g of the original MCC was weighed and gradually added into a sulfuric acid solution at a ratio (MCC:H2SO4) of 1:8 g/mL and a temperature of 0–5 °C.

When the cellulose and acid were completely mixed and a gel-like mass was obtained, 25 mL of deionized water was added and then heated in a water bath at 40 ± 2 °C under mechanical stirring at 500 rpm for 1 h. CNC extraction from the resulting suspension consisted of two stages. In the first step, the CNCs were stripped of phytomelanin pigment. In order to do this, the suspension was centrifuged for 5 min at 1500 rpm/min (Centrifuge 5427R Eppendorf). As a result, the phytomelanin precipitated in the form of large black particles. The cleaning process was repeated 2 times until the large particles of pigment had been completely separated (Fig. 6).

In the second stage of extraction, the CNCs were centrifuged at 8000 rpm for 15 min, followed by dialysis (Visking Dialysis Tubing MWCO-12–14,000 Daltons) for one week in deionized water until reaching a neutral pH. The CNC yield was calculated using the following formula:

$${\text{Yield }}\left( \% \right) = \left( {{\text{m}}_{{{\text{MCC}}}} - {\text{m}}_{{{\text{CNC}}}} } \right)/{\text{m}}_{{{\text{MCC}}}} \times {1}00\%$$
(3)

where mMCC is the mass of the MCC obtained by hydrolysis; and, mCNC is the mass of the CNCs obtained after sulfuric acid hydrolysis.

The particle sizes and zeta potential of MCC and CNCs

A 2 wt% aqueous suspension of MCC and CNCs was subjected to 30 kHz for 10 min using an ultrasonic disperser, U-sonic UZTA-0.15/22–0 (Alena, Russian). The average particle size and zeta potential were determined via dynamic light scattering (DLS) using a Zetasizer NanoZS 90 (Malvern, UK). The measured electrophoretic mobility was then converted to the zeta potential.

Preparation of CNC film

In order to prepare the CNC film, 10 mL of 96% ethanol was added to 10 mL of an aqueous solution of 15 wt% CNCs, which was then stirred in a magnetic mixer for 15 min. The fully mixed solution was poured onto a flat plastic surface and dried at 25±2 °C for 48 h. As a result, a colorless film with a thickness of 18±1 μm formed. The samples were stored in a vacuum desiccator to prevent sorption of water vapor and carbon dioxide from the air.

FTIR spectroscopy

FTIR analysis of MCC, CNCs and CNC film were performed on an FT-801 FTIR spectrometer (Simex, Russian), with a resolution of 1 cm−1 at a range of 450–4700 cm−1, in accordance with the standard method and using an accessory to measure attenuated total reflection (ATR) and specular-diffuse reflection (SDR), at a temperature of 25±1 °C.

X-ray diffraction

The crystal structures of MCC, CNCs and CNC film were studied via X-ray diffraction on a X´PertPRO diffractometer (Malvern Panalytical Empyrean, Netherlands) using monochromatized copper (CuKα) at a scan speed of 0.05° for 10 s, with a K-Alpha1 wavelength of 1.54187 Å. Measurement in reflection mode, using an aluminium rectangular multi-purpose sample holder (PW1172/01), was performed at a diffraction angle 2θ between 10° and 40°, with the X-ray tube voltage at 45 kV, current intensity at 30 mA, and a measurement time of each step of 0.5 s. The ICDD PDF-4/AXIOM database of XRD pattern was used for the analysis of the resultant XRD patterns. The POWDER CELL 2.5 software package was used for the Rietveld refinement of the unit cell parameters and CSL, run 3 times using formula 4.

The crystallinity index was calculated using the following equation (Ling et al. 2019):

$${\text{CI }}\left( \% \right) = \, \left( {{\text{A}}_{{{\text{cryst}}}} /{\text{A}}_{{{\text{crys}} + }} {\text{A}}_{{{\text{amorph}}}} } \right)*{1}00\%$$
(4)

where Acryst is the area under the calculated pattern for crystalline cellulose; and, Aamorphis the area under the pattern calculated for the amorphous content.

CSL was determined via the Scherrer formula:

$${\text{CSL}} = {\text{ k}}\lambda /\beta {\text{cos}}\theta$$
(5)

where k is a shape factor that is often 0.89 (Sintu et al. 2019); λ is the wavelength of the diffractometer (0.1542 nm); β is the FWHM (maximum half of the full width) of the (200) diffraction peak in radians; and, θ is the diffraction angle.

UV–vis analysis of CNC and CNC film

The optical absorption spectrum of the CNC aqueous solution obtained was recorded on a spectrophotometer (PE-5400UV, Russian) with a scanning speed of 240 nm/min and a wavelength of 190–1000 nm. A 10 mm thick quartz cuvette was used. The transparency of the CNC film in the optical absorption region of 600 nm was calculated using the following formula:

$${\text{Opacity }} = {\text{ Abs}}_{{{6}00}} /{\text{d}}$$
(6)

where Abs600 is the optical density of the CNC film of 600 nm; and, d is the thickness of the CNC film (µm).

Scanning electron microscopy

The surface morphologies of MCC, CNCs and CNC film were examined using a Quanta 200i 3D SEM (FEITM Netherlands). Measurements were carried out in high vacuum mode using a secondary electron detector at an accelerating voltage of 15 kV. 1 g of MCC and 1 g of CNC powders were each added into separate 50 mL solutions of deionized water and silver nanoparticles at a concentration of 10–3 mol/L and then mechanically stirred for 2 h. The prepared samples were run through filter paper and dried in an oven at 50±2 °C for 1 h. The surface of the CNC film was coated with gold nanoparticles to improve the transfer of electrons. Finally, the specimens were mounted on aluminum pins with carbon tape.

TGA analysis

The thermal characteristics of MCC, CNCs and CNC film were studied with a LabSys evo differential thermogravimetric analyzer (Setaram, France), in an argon atmosphere. The temperature range was 30 ± 5–700 ± 5 °C, with a heating rate of 10 ± 1 °C/min. The mass of the samples was approximately 20 ± 2 mg.

Results and discussion

Theconcentrationof PAA obtained

According to the calculation based on formula 1 (above), the concentration of the PAA obtained was 12%.

Characterization of materials

Figure 2 shows the raw material and the product obtained in the present study. In Fig. 2a, the raw SFHs are of a pale color in the interior region and a black appearance on the outside region, due to the presence of phytomelanin pigment. Further, the unprocessed SFHs have a size range of 8–12 mm in length and 4–6 mm in width. Figure 2b shows an image of MCC, which, due to the presence of residual pigment, is gray in color, with a cotton-like fibrous appearance.

Fig. 2
figure 2

a Raw SFHs and b MCC

Full size image

Determination of microcrystalline cellulose yield and quality

The results of MCC yield and quality obtained at different ratios of SFH:PAA via the organosolvent oxidation method are shown in Table 1 and Fig. 3. At a ratio of 1:16 g/mL the first noticable increase in the activity of the delignification process occurs. At 1:20 g/mL, the pulp yield reaches a high level, 47.8±2%. Further increases in PAA raise the cellulose yield by only 0.04% (Table 1). Therefore, the ratio of 1:20 g/mL is considered the most effective and shall be used in further studies, taking into account the consumption of reagents. Finally, MCC obtained at the ratio of 1:20 g/mL was used in CNC synthesis.

Table 1 Effect of SFH:PAA ratio on MCC quality.
Full size table
Fig. 3
figure 3

SEM images of MCC

Full size image

In terms of quality analysis, the SFH:PAA ratio of 1:20 g/mL was found to produce the highest value of α-cellulose (72 ± 3%). The minimum content of residual lignin and hemicellulose was also observed at 1:20, resulting in 2.1 ± 0.5% and 13.02 ± 2%, respectively (Table 1). The moisture and ash content (SiO2) of the MCC obtained (Table 1) did not differ significantly from previous studies (Ritesh et al. 2019). The unprocessed biomass of the annual plants used in the study showed a content of β-cellulose ranging from 30 to 60% and a content of residual lignin from 12 to 22%. Comparing these values to those of MCC, the effectiveness of the method of organosolvent oxidation in increasing the quantity of cellulose is clearly demonstrated. Of note is that oxidative-organosolvent technology allows extraction of MCC from SFHs under mild conditions and at high output without excessive pressure or temperature. In the process of obtaining MCC from SFHs, this delignification method was found to be quite effective (Table 1).

Surface morphology, particle size and suspension stability of MCC

The surface morphology of MCC is shown in Fig. 3. MCC fibers have an average length of 200–300 µm and a width of 3.73–21.30 µm, and a surface that is clean and relatively smooth. This indicates that cellulose microfibers obtained from SFHs during organosolvent oxidation are separate, with limited quantities of hemicellulose and lignin.

The average particle size of MCC fibers was found to be 1971 ± 100 nm in length and 266 ± 35 nm in width (Fig. 4 and Table 2). Figure 5a shows an image of the freshly prepared MCC aqueous suspension and 5b shows an image of the MCC aqueous suspension after one month. Following a month, the MCC can be seen to fall out of suspension under the influence of gravity.

Fig. 4
figure 4

The particle size of MCC

Full size image
Table 2 CI, CSL and average particle values of CNCs obtained from MCC via acid hydrolysis
Full size table
Fig. 5
figure 5

MCC suspension obtained from SFH. a fresh MCC and b after one month

Full size image

Preparation, surface morphology, particle size and suspension stability of CNC

Figure 6 is a diagram of CNC extraction from MCC using acid hydrolysis, and shows the resulting CNCs. The CNC yield from the MCC was 35 ± 1%. In a previous study (Chen et al. 2015), nanocellulose was obtained using acid hydrolysis from SFHs, and had a yield of 31.67%. Compared to that study, the present one yielded 3.33% more CNC volume.

Fig. 6
figure 6

Diagram of CNC synthesis via acid hydrolysis

Full size image

In Fig. 7, the SEM microphotographs show that the surface of the CNC obtained using sulfuric acid hydrolysis consists of agglomerated fibrils. Acid hydrolysis breaks down the amorphous parts of the MCC and converts the microfibers into nanofibrils. The process of hydrolysis significantly reduces the size of microcrystalline fibers, resulting in a length of 450 ± 50 and width of 50 ± 15 nm (Table 2). Furthermore, additional agglomeration of nanofibrils may be due to the effect of interfacial hydrogen bonding. Onkarappa et al. (2020) and Radakisnin et al. (2020) have shown that nanofibrils obtained through acid hydrolysis from various rawmaterials are agglomerated and achieve an average diameter of 50–150 nm. Thus, the CNC surface morphology obtained from SFHs did not differ from the CNCs obtained from other raw materials.

Fig. 7
figure 7

SEM image of CNCs

Full size image

The average length of the CNCs reduced by 4 times to 450 ± 50 nm, as compared to the particle size of MCC, and the width shrunk by around 5 times to 50 ± 15 nm (Fig. 8 and Table 2). Figures 9a, b show that the CNC aqueous suspension remains unchanged after one month. A white homogeneous solution without any precipitation is shown in Fig. 9b, which indicates its stability. Figure 9a also shows the unimpeded passage of laser light through the CNC suspension. This is due to the Tyndall effect, which is caused by the impact of charges on the surface of cellulose nanoparticles (Zhan et al. 2018; Elanthikkal et al. 2010; Mohamed et al. 2015). In order to clarify the cause of the above effect, the zeta potentials of MCC and CNC aqueous suspensions were determined, and were found to be negative in both cases. The zeta potential value of the CNC suspension in 80 wt% phosphoric acid was -21.2 mV. The zeta potential results showed the formation of an electric charge between the shear plane of the final external layer and the bulk solution. This shows that during acid hydrolysis negative charges accumulate on the CNC surface, leading to an increase in the zeta potential.

Fig. 8
figure 8

The particle size of CNC

Full size image
Fig. 9
figure 9

CNC suspensions. a fresh CNC and b CNC after one month

Full size image

UV–vis analysis of CNCs and CNC film

Figure 10 shows that the layer of 18 ± 1 µm thick CNC film is both colorless and transparent. CNCs and CNC film were recorded in the ultraviolet spectrum at a wavelength of 200 to 1000 nm (Fig. 11). The spectra of the two samples did not show the absorption of the electron transitions of the π−π* (C=O) bond of the aromatic ring at a wavelength of 260–280 nm (Fig. 11a). This purity is explained by complete purification of lignin and hemicellulose (Lamia et al. 2009; Claudia et al. 2015), which can be clearly seen in the FTIR spectrum shown in Fig. 13.

Fig. 10
figure 10

CNC film production

Full size image
Fig. 11
figure 11

a UV–vis spectrum of CNC and CNC film and b optical transmittance of CNC film

Full size image

The CNC film optical transmittance became smaller as the wavelength decreased, and the maximum absorption in the visible region was 50 ± 1% (Fig. 11b). The CNC film has high clarity and a good transparency. A study by Trifol et al. (2017) found that CNC film obtained from sisal fibers has good optical properties. Compared to the results of previous studies, the optical properties of the CNC film obtained here are generally better.

Morphological aspects of CNC film

The surface microstructure of CNC film is shown in Fig. 12. The surface of the CNC film appears homogeneous and consists of rod-shaped nanofibrils. Due to the shrinkage of fibers, the crystal is compacted and in rod form. In addition, the presence of a surface charge, caused by the sulfo groups, on the nanocrystallites has a positive effect on the homogeneity of the film surface (Martha et al. 2014; Trifol et al. 2017).

Fig. 12
figure 12

SEM images of CNC film

Full size image

FT IR spectroscopy of MCC, CNCs and CNC film

Comparative FTIR spectra of the chemical structures of MCC, CNCs and CNC film obtained under the established optimal conditions (1:20 g/mL) are shown in Fig. 13. IR spectroscopy shows the absorption regions in all spectra: 655.9 cm−1, at the C–OH bond due to out-of-plane bending (Zghari et al. 2018; Stefan et al. 2020); 891 cm−1 and 1153.6 cm−1 β- (1,4), caused by the glycoside C–O–C, an amorphous region (Zghari et al. 2018; Stefan et al. 2020); 1014.7 cm−1 and 1153.6 cm−1 high-intensity signals, from C–O and C–C bonds in the aromatic ring and C–O in the pyranose ring; 1053.3 cm−1, showing chronic oscillations of the C bond; 1253.9 cm−1, 1307.9 cm−1 and 1365.8 cm−1, due to the C–H bond; 1412.1 cm−1, from the C–H2 groups (Zghari et al. 2018; Stefan et al. 2020; Kacuráková et al. 2002; Kian et al. 2017; Haafiz et al. 2014); 1635.9 cм−1, caused by the O–H in the water molecule sorbed from the air (Zghari et al. 2018); and, 2889.8 cm−1 and 3337.8 cm−1, due to the CH and OH groups (Trifol et al. 2017; Kondo et al. 1996; Cao et al. 2004) (Fig. 13). In the MCC spectrum, 1516.2 cm−1 is associated with C=C in the aromatic ring of lignin (Sun et al. 2005), and 1724.6 cm-1 shows a chronic oscillation of the acetyl and ester group, C=O, in the hemicellulose molecule (Trifo et al. 2017). This confirms the accuracy of the numerical values of hemicellulose given in Table 1, and indicates the effective extraction of lignin and hemicellulose during the production of CNCs via acid hydrolysis. IR spectra show that the chemical structure of CNCs is similar to that of MCC, and that acid hydrolysis does not destroy the chemical structure of MCC (Djalal et al. 2014; Benyoussif et al. 2015). The result obtained is in good agreement with Kian et al. (2017). Finally, the chemical structure of CNC film shows no significant differences from the structure of CNCs themselves (Fig. 13).

Fig. 13
figure 13

FTIR spectra of MCC, CNCs and CNC film

Full size image

X-ray diffractometry of MCC, CNCs and CNC film

In general, the crystallinity of cellulose is an important characteristic that determines its thermal and mechanical properties. Figure 14 shows a comparative X-ray diffractogram of MCC, CNCs and CNC film. A survey of the relevant literature was done to assess the crystallinity of cellulose (Podgorbunskikh et al. 2018; Mondragon et al. 2014). Four diffraction peaks are observed, equal to 2ϴ = 15.6 (1–10), 16.5 (110), 22.2 (200) and 34.5 (004), which demonstrate that the crystal structure of the molecule is characteristic of a two-chain monoclinic unit cell. Theyare the counterparts to the peaks of the cellulose I β pattern (French 2014).The diffractogram ofthe CNC film showed a loss of peak intensities of 2Ɵ = 15.60 (1–10) and 16.50 (110), which maybe due to higher oxygen ingress into the crystallite layers during air drying of the CNC film (Kazi et al. 2021). At the aforementioned ideal ratio of 1:20 g/mL, MCC has a CI of 72.9 ± 4%, a CSL at 2.9 ± 0.1 nm, and an average fiber particle size of 1971 ± 100 nm in length and 266 ± 35 nm in width (Table 2).

Fig. 14
figure 14

XRD diffractions of MCC and CNCs and CNC film

Full size image

The CI content of the CNCs obtained via acid hydrolysis increased to 82 ± 4%, and the CSL value decreased to 2.7 ± 0.1 nm. This demonstrates that during the chemical treatment, non-cellulose components are effectively removed from the amorphous part of the lignin, hemicellulose and cellulose, further increasing β-cellulose content (Martins et al. 2011). This is also evident from the results of SEM images of CNCs and CNC film (Figs. 7 and 12). The crystal structure of the film obtained from CNCs was not greatly changed in comparison with the CNCs themselves.

Thermo gravimetric analysis of MCC, CNCs and CNC film

Figure 15 is of the relative thermograms showing the thermal stability of MCC, CNCs and CNC film. The mass reduction of all samples can be divided into three stages. The first stage involves a decrease in mass in all samples, due to evaporation of water molecules at a temperature range of 35–125 °C (Voronova et al. 2015; Zhan et al. 2018). At the same time, MCC lost around 11% by weight, while the maximum reduction in the mass of CNCs and CNC film was approximately 6%, which is clearly evident from the XRD results (Fig. 14 and Table 2).

Fig. 15
figure 15

TGA curves of MCC, CNCs and CNC film

Full size image

At the second stage, a significant loss of sample mass occurred in the temperature range of 200–358 °C. MCC, CNCs and CNC film lost 63%, 50% and 51% of their mass, respectively. The CNCs and CNC film lost a comparatively smaller amount of mass, which may be due to the high number of crystallites in the CNCs (Augusta et al. 2018). At this temperature range, cellulose is strongly depolymerized, while volatile components, such as CO, CO2 and CH4, are formed (Roman 2004). A sharp decrease in MCC mass is observed to start at a temperature of 254 °C, whereas the mass reduction of CNCs and CNC film begins at a temperature of 226 °C, with most decrease in the latter occurring in the range of 226–366 °C. The thermal stability of CNCs and CNC film was lower than that of MCC. This is due to the rapid degradation of the sulfate groups on the surface of the CNCs (Roman et al. 2004; Kusmono et al. 2020). In the final stage, thermal decomposition of intermediate volatile products, such as levoglucosan, coke, hydrogen, ethylene, ethane and resin, occurs in the temperature range of 360–700 °C (Roman et al. 2004).

Conclusions

The present study has shown that an effective method for obtaining MCC from SFHs is conversion via the organosolvent oxidation method, with an ideal SFH to PAA ratio of 1:20 g/mL. The MCC quality indicators support this assertion, with β-cellulose at 72 ± 3%, residual lignin at 2.1 ± 0.5%, hemicellulose at 13.7 ± 2%, humidity at 2.75 ± 0.5%, and ash content (SiO2) at 2.2 ± 0.5%.

CNCs were synthesized at a yield equal to 35 ± 1% from MCC via sulfuric acid hydrolysis (60 wt%), which could later be made into CNC film using the previously outlined method. FTIR analysis of all products showed that their chemical structures are similar to each other. Analysis of average particle size showed that CNC crystals were several times smaller than MCC particles. Further, the average CI value increased from 72.9 ± 4% (MCC) to 82 ± 4% (CNCs) and CSL decreased from 2.9 ± 0.1 nm to 2.7 ± 0.1 nm, respectively.

The relative crystal structure of the CNC film was shown not to differ in any significant way from the original CNCs. Additionally, UV spectroscopy demonstrated that the dispersion of the CNCs in an aqueous medium was very high. Further, the Zeta potential was measured at -21.2 mV. Finally, the transmission coefficient of the CNC-film was equal to 50 ± 1%.

The SEM results show that MCC fibers have a smooth surface, while the morphology of CNCs consists of agglomerated fibrils. The surface of the CNC-film shows homogeneous rod-like nanofibrils. All synthesized cellulose materials were shown to undergo degradation in the temperature range of 200–358 °C, and lose 51–63% of their mass.

The present study demonstrates that by adopting the modified organosolvent oxidation process, MCC can be easily extracted from SFHs, which can then be further processed into nanocellulose. Hence, the present method can be employed successfully for the cost-effective production of nanocellulose. Moreover, the method is relatively environmentally friendly, highly efficient and an excellent protocol for the extraction of nanocellulose from SFHs.