A Comparative Study of Nanofibers from Regenerated Cotton and Jute

ABSTRACT Nanofibers electrospun from cellulose sources have received substantial attention. Using trifluoro acetic acid (TFA) as the solvent, dissolved cotton and jute were comparatively electrospun to cotton nanofibers (CNFs) and jute nanofibers (JNFs), respectively. The viscosity and conductivity of the as-prepared polymer solutions were measured. The morphology, chemical structure, thermal stability and decomposition pattern together with the crystallinity index of the JNFs and CNFs were characterized by scanning electron microscopy (SEM), Fourier transform infrared (FTIR) analysis, thermogravimetric analysis (TGA) and X-ray diffraction (XRD), respectively. SEM analysis showed no major differences in surface morphology between JNFs and CNFs. The TGA results confirmed that the as-spun nanofibers have excellent thermal stability up to 210 ºC temperature before cellulose decomposition. The JNFs showed a similar degradation pattern as the CNFs. The results of this work will contribute to proposing appropriate applications of electrospun nanofibers from jute.

Both cotton and jute are naturally abundant cellulose fibers, and nanofibers can be regenerated from dissolved polymer solution from them. However, their chemical components except for cellulose are different. Thus, their regenerated nanofibers may exhibit different properties. This work reports on a comparative study of the structure and properties between electrospun cotton nanofibers (CNFs) and jute nanofibers (JNFs) using the TFA solvent. The solution viscosity and conductivity, spinnability, chemical components, morphology, thermal stability and crystalline structure were tested/ characterized and compared. The findings of this work will shed a light on the potential applications of JNFs in the same application areas of CNFs.

Materials
Jute fibers (Bangla Tossa C grade) were collected from Khulna jute mills, Bangladesh. Cotton fibers were collected from local market. Both fibers were used without any pre-treatment. TFA was bought from Sigma Aldrich. TFA was of analytical grade and used without further purification.

Fabrication
The schematics of the experiments are shown in Figure 1. Dissolution of cotton fibers into TFA solution was the same as reported in our previous work (Ara et al. 2021). Briefly, cotton fiber pieces (1-5 mm) were directly added into TFA solution with different concentrations (1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5 & 5 cotton wt.%) under magnetic stirring at room temperature. The stirring process lasted for 72 h to make a completely dissolved cotton-TFA solution. The raw jute was dissolved in TFA using the same process as reported previously (Ara et al. 2021). Raw jute took 120 h to fully dissolve under continuous magnetic stirring. It also took almost 7-8 days to fully dissolve the 5 wt.% jute fibers into a highly viscous solution. Both solutions were used as the polymer solution to perform electrospinning, as shown in Figure 1. A conventional needle-based electrospinning setup was used with experimental parameters listed in Table 1. Proper experimental parameters for electrospinning were chosen after a few trials, and electrospun nanofibers were collected on the aluminum foil collector for further characterization and measurements.

Characterization and measurements
Solution conductivity was measured by a Palintest 146 Waterproof 800 pH/Conductivity/TDS Meter. The conductive electrode was completely immersed in the solution at room temperature, while the displayed readings were recorded. Viscosity was measured by a Brookfield KF40 falling ball viscometer at a temperature of 25°C.
The Fourier transform Infrared (FTIR) spectra were measured by a Perkin Elmer Spectroscopy (400 series, UK) device with a universal attenuated total reflectance (ATR) sampling accessory. The spectra were collected in the range of 650-4000 cm −1 . Peak heights of absorptions band were measured by the software OriginPro70.
A Hitachi tabletop scanning electron microscope (TM4000 series) was used to observe the surface morphology of as-collected nanofibers with an accelerating voltage of 15 kV (in the case of JNFs) and 5 KV (in the case of CNFs) under mixed vacuum mode. Samples were coated for 75 s (3.75 nm) with gold by a SC7620 mini sputter coater before observation. The fiber diameter and surface area were measured from the SEM images using ImageJ 1.46 image analysis software.
Thermogravimetric analysis (TGA) was performed with a Perkin Elmer TGA Pyris 1 instrument to measure the thermal stability and decomposition patterns of JNFs and CNFs. Approximately 3 mg of nanofibers were used for each measurement. The samples were heated from 30°C to 800°C at a heating rate of 10°C/min under nitrogen protection.
The X-ray diffraction patterns of as-spun nanofibers were measured by a Bruker D4 Endeavor powder X-ray diffractometer using Mo/Cu radiation source at 40 kV and 30 mA over the scanning angle range 2θ of 5°-50° at a scanning speed of 0.5°/min.

Viscosity and conductivity
Viscosity and conductivity were measured to determine the suitability of the as-prepared solutions for electrospinning. The jute-TFA solutions of 4.5 wt.% and 5 wt.% were too dark due to the presence of lignin, hence the falling ball was not visible to measure their viscosity accurately. For both cotton and jute, a highly viscous solution can be obtained when dissolved in TFA due to their high molecular weight. Table 2 shows that the viscosity value of polymer solutions increases with the increase in concentration for both jute and cotton. The jute-TFA solution shows a slightly higher viscosity than the cotton-TFA solution due to the presence of lignin and hemicellulose.
The conductivity of both solutions suggests the feasibility of electrospinning ( Table 2). The relationship between the concentration and conductivity is not clear as seen from Table 2. Electrospinning involves stretching of a solution caused by repulsion of charges at its surface, and more charges can be carried by the electrospinning jet if the conductivity of the solution is increased C=concentration. C=concentration. (Ramakrishna 2005). Hence, a small amount of salt or co-solvents can be added to increase the conductivity of the cotton-TFA and jute-TFA solutions to gain more stable electrospinning jets.

Spinnability
For cotton fiber dissolved in TFA solvent, up to 5 wt.% solution can be made. Cotton-TFA solution with the concentration of 1.5, 2, 3.5 and 5 wt.% can produce visible nanofibers through electrospinning. Difficulties were experienced during electrospinning of 2.5, 3 and 4 wt.% solutions, as no jets were found and there were no nanofibers obtained on the aluminum foil from the collector. This may be due to the low conductivity of 2.5 and 3 wt.% solution. During electrospinning of 5 wt.% solution, droplets and discontinuous jets were formed. Only 1.5, 2, 3.5 and 5 wt.% solutions could be electrospun properly with nanofibers deposited on the collector for further characterization. In both cases of JNF and CNF electrospinning, the spinning became difficult with the increase in solution concentration. This might be due to the presence of lignin in jute fiber and the increased viscosity of the jute and cotton solutions. Figure 2 shows the FTIR spectra of dissolved cotton and jute solutions together with their resultant nanofibers. In Figure 2(a), the spectral bands of 3499 cm −1 in cotton-TFA solution and 3131 cm −1 in jute-TFA solution are ascribed to alcoholic O-H stretching from the vibrations of the hydrogenbonded hydroxyl group (Lakshmanan and Chakraborty 2017). Peaks at 1150 cm −1 and 1149 cm −1 represent C=C ring stretching bands respectively for cotton and jute solutions (Soni, Hassan, and Mahmoud 2015). The spectral bands of 1778 cm −1 and 1776 cm −1 are due to the absorption of the carbonyl of trifluoroacetyl group resulting from the esterification of -OH group of cellulose, hemicellulose, and lignin (Hasegawa et al. 1992). Absorption bands around 811 cm −1 in jute-TFA solution, 810 cm −1 and 782 cm −1 in cotton-TFA also show the presence of trifluoroacetic acid (Hasegawa et al. 1992).

Chemical components
In Figure 2(b), it can be clearly seen that strong absorption peaks appear at 3320 cm −1 and 2895 cm −1 , corresponding to the O-H stretching in the carboxylic acid group and C-H stretching of cellulose, respectively (Mayandi et al. 2016). O ̶ H bending of adsorbed water at 1654 cm −1 and C-O stretching vibration of the cellulose backbone at 1020 cm −1 are also observed . The absence of peaks at 1778 cm −1 , 1776 cm −1 and in the range 810-782 cm −1 clearly proves the evaporation of TFA from both CNFs and JNFs. Figure 2(b) distinctly shows no prominent peaks of lignin or hemicellulose, indicating that there are some minor differences in the chemical structure of cotton and jute nanofibers. Figure 3 shows the SEM images of JNFs and CNFs with different concentrations. The associated statistics on fiber diameter are shown in Table 3. In 1.5 wt.% JNFs, the as-collected sample shows a visible nanofibrous web with an average fiber diameter of 118 ± 27 nm, but beads can be observed occasionally. On the other hand, well-separated and web-like nanofibrous membrane with an average fiber diameter of 215 ± 43 nm can be seen from 1.5 wt.% CNFs. It was also observed that regenerated cotton is more spinnable than regenerated jute under a low concentration.

Morphology
About 2 wt.% JNFs exhibit an even nanofibrous web with an average diameter of 123 ± 57 nm. Similarly, nanofibrous web can be seen for 2 wt.% and 3.5 wt.% CNFs with an average diameter of 178 ± 41 nm and 269 ± 82 nm, respectively. As shown in Table 3, the average diameter of JNFs is lower than that of CNFs under the same concentration. In the spinnable concentration range, e.g., 2-3.5 wt.%, uniform nanofibers can be electrospun with a relatively low standard deviation in fiber diameter, suggesting that electrospinning of regenerated jute is comparable to that of regenerated cotton.
The SEM image of 5 wt.% JNFs shows no obvious nanofibrous structure because the jet emission failed. Similarly, no visible nanofibers can be observed in the case of 5 wt.% CNFs, and this may be due to the high viscosity of the solutions. It is obvious that both solutions are not spinnable when the concentration is 5 wt.% or higher.

Thermal stability
TGA was used to characterize the decomposition rate and thermal stability of materials during controlled heating (Lakshmanan and Chakraborty 2017). Figure 4 shows the thermogravimetric (TG) and derivative thermogravimetric (DTG) curves of JNFs and CNFs under different concentrations. The JNF curves show a similar weight loss profile with four major weight loss stages in response to moisture evaporation, hemicellulose decomposition, cellulose decomposition, and lignin decomposition, respectively (Rangan et al. 2017). The earliest average weight loss of 3.99% due to the water evaporation occurred in JNFs between 30°C and 126°C. All tested CNFs showed an average 8.31% weight loss in the temperature range 30-126ºC due to evaporation of absorbed and intermolecular water. In the first weight loss stage, JNFs show less weight loss than that of CNFs, as seen from Figure 4(a).
A large weight loss appears in the temperature range 210-300ºC for JNFs, and this is due to the thermal depolymerization of hemicellulose and the degradation of cellulose (which usually takes place between 275 and 400ºC) (Deepa et al. 2015). Lignin decomposition took place between 397 and 713ºC for all JNF samples with an average weight loss of 7.45%. The amounts of char residues at 800ºC from the three JNF samples are 19.49%, 19.20% and 16.83%, respectively. According to Figure 4(b), a sudden weight loss of 32% in CNF samples started at around 170-210ºC and 258-270ºC due to the decomposition of hemicellulose. Cellulose decomposition occurred in the temperature range 210-415ºC with a major weight loss of around 56%. There was very little (1.17%, 0.17% and 0.66%, respectively) residue left at 800ºC. Complete pyrolysis occurred at around 560 ºC for all CNF samples. The TGA results show that both JNFs and CNFs have excellent thermal stability. The tested JNFs show a degradation pattern like CNFs. The nanofibers can be utilized in the application areas up to 210ºC temperature before cellulose decomposition.

XRD analysis
To assess the crystalline and amorphous regions before and after electrospinning, X-ray diffraction studies were carried out for raw cotton and jute fibers and for the as spun JNFs and CNFs with different concentrations. The XRD curves are shown in Figure 5 and the crystallinity and amorphous percentages are statistically listed in Table 4.
X-ray diffraction peaks were observed at 2θ = 22-23° and 15-17° for all tested samples, which are the characteristic peaks of cellulose indicating the existence of crystalline regions. Some sharp peaks at 2θ = 20.1°, 38.6°, 44.3° and 44.9°, and 44.1° represents the presence of aluminum foil in the tested sample, which is a major drawback while collecting nanofibers from aluminum foil. Due to this drawback, the results collected for the JNF and CNF samples are not sufficiently clear for comparative analysis.
Nevertheless, from the crystallinity index given in Table 4, there is an increase in crystallinity from raw cotton (57.5%) to 2 wt.% CNFs (66.3%), but a decreasing pattern can be observed from raw jute (67.8%) to 3.5 wt.% JNFs (28.6%). Regeneration of jute by dissolution in TFA may have transformed the crystalline structure from cellulose I to cellulose II with a large amount of amorphous region.

Conclusions
Nanofibers can be electrospun from both dissolved cotton and jute solution using TFA as the solvent. TFA facilitated the dissolution of cellulose and lignocellulose in jute by esterifying the cellulose, lignin, and hemicelluloses to trifluoroacetyl esters, as confirmed in the FTIR analysis. The jute-TFA solution showed a slightly higher viscosity than the cotton-TFA solution due to the presence of lignin and hemicellulose. The FTIR spectra of JNFs and CNFs showed no major differences in chemical composition between the nanofibers and their raw materials. An average diameter of 125 ± 35 nm JNFs was obtained for raw jute, whereas for cotton 220 ± 55 nm CNFs were collected. The SEM images showed no significant difference between CNFs and JNFs with a spinnable range of 1.5-3.5 wt.%, but  the average diameter of JNFs was lower than that of CNFs under a given concentration. The TGA results showed that both the JNFs and CNFs have excellent thermal stability up to 210 ºC temperature before cellulose decomposition, and the JNFs showed a degradation pattern almost identical to the CNFs. The present results as obtained from this work indicate that JNFs have the potential to be applied as regenerated cellulose nanofibers with structure and properties similar to CNFs. The as-spun nanofibers from jute could find applications in several areas, such as biomedical, pharmaceuticals, electronics, barrier films, nanocomposites, membranes for water and wastewater treatment and supercapacitors.

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