Degumming of Hemp Fibers Using Combined Microwave Energy and Deep Eutectic Solvent Treatment


 Hemp bast fibers were degummed using combined microwave energy (MWE) and deep eutectic solvent (DES) to generate pure hemp cellulose fibers for potential textile applications. The properties of the obtained fibers were investigated and compared with those from the traditional alkali-based process using several analytical techniques. Results revealed that hemp fiber surface underwent dramatic structural disruption during the pretreatment, due to the removal of “gummy” compounds (i.e., lignin, pectin, oil, and wax) and amorphous cellulose. Ultraviolet (UV) protection factor (UPF) of DES-treated fibers with 1:20 fiber-DES ratio (i.e., 183.67) were significantly higher than those from the traditional alkali-treated (140.75) and untreated raw hemp fibers (127.47). The treated fibers also had higher thermal stability. Chemical composition analysis showed that the cellulose content in the treated fiber samples increased to 49.95% which was comparable with the cellulose content of the alkali-treated fibers (49.49%). The study demonstrates a potentially effective, less time-consuming, and environmentally sustainable protocol for manufacturing purified hemp cellulose fibers using combined MWE-DES treatment.


Introduction
Hemp has garnered great interest due to its biodegradability, low-cost, and fast-growing capability under mild and harsh weather conditions throughout the world (Dhondt 2020). Long hemp bast bers from hemp stalk have been used for a myriad of applications such as textiles, paper, and biodegradable composites ). Chemical composition of hemp bast consists of cellulose, lignin, hemicellulose, and gums including pectin, oil, wax, minerals (Liu et al. 2017; Keiller et al. 2021), and other non-cellulosic components. Gums in natural bers are complex carbohydrate polymers with long sugar chains including hydroxyproline proteins, resins, and other compounds that can be soluble, partly soluble, or insoluble. Lignin is a complex polymer of aromatic and aliphatic components of three monolignols (hydroxycinnamyl) such as coniferyl alcohol, sinapyl alcohol, and p-coumaryl alcohol. The presence of the gum and lignin in hemp bers leads to di culties in further textile processing of hemp bers such as spinning, weaving, dyeing, and other nishing processes. Thus, the degumming (i.e., removing gums, lignin, and other non-cellulosic components) of hemp bers is a crucial step for the effective utilization of the bers.
There are many methods of hemp ber degumming including chemical (Zhu et al. 2021), mechanical (Liu et al. 2017), thermal (Gedik and Avinc 2020), water retting, high energy irradiation (Stelescu et al. 2020), and other methods. Studies have been done on several chemicals and strategies including alkali treatment, acetylation, bleaching, salinization, benzoylation, and other treatments using organic and inorganic acids, peroxide, chelating agents, anhydrides, sodium chlorite, and sodium sul te (Kabir et al. 2020; Menghini et al. 2021). Even though chemical treatments are able to make the nest quality of bers, they can cause several negative impacts on the environment by generating a high amount of e uent and requiring high energy inputs, leading to increased production cost. Replacing chemical treatments with other biological treatments alone cannot produce suitable individual hemp bers as hemp raw ber contains only about 57-77% cellulosic compounds (Stevulova et al. 2014). Thus, there is a need to develop an environmentally friendly and cost-effective degumming process.
Over the last few decades, ionic liquids (ILs) have exhibited promising properties for treating lignocellulosic materials. However, certain properties of ILs including toxicity, high cost, and destructive structural impact on cellulose have limited their widespread use. Deep eutectic solvent (DES) is an alternative ionic liquid that is largely biodegradable (Abbott et al. 2004). DESs can be synthesized by the combination of two or three components, where a certain mix of hydrogen bond acceptors (HBA, e.g., quaternary ammonium salts, choline derivatives, organic acids) and hydrogen bond donors (HBD, e.g., amides, polyols, and carboxylic acids) is used. The eutectic solution consists of nonsymmetric and large ions with low lattice energy and has a melting point that is signi cantly lower than the melting points of the individual components in the mixture. The charge delocalization occurs through the formation of hydrogen bonds, leading to the suppression of melting points in the eutectic mixture compared with these of individual components. One of the main advantages of DES is that DES offers similar physiochemical properties to ILs while being more sustainable, cheaper, and having lower toxicity compared with ILs. Additionally, the recyclable and reusable traits of DES have made it more attractive for the treatment of biomass. It has been reported that DES can extract lignin without the alteration of the cellulosic structure of lignocellulosic materials (Tan, Chua, and Ngoh 2020). DES has been utilized in several sectors such as deligni cation, biodiesel production, and sugar recovery. DES has also been employed in the surface treatment of lignocellulosic materials, including ramie (Yu et al. 2020), kenaf bast (Nie et al. 2020), and apocynum bast (Song et al. 2019). It has been found that DES treatment has a comparable degumming effect to that of traditional degumming methods (alkali, silane, and bleaching treatment).
Traditional heating systems for the degumming process including steam, gas, and electric energy sources have several disadvantages such as longer treatment time to achieve the desired temperature and process ine ciency, and excessive cost. Recently, microwave energy (MWE) heating has gained a lot of attention as a pretreatment process of lignocellulosic materials due to its e ciency, low hazards, and reduced treatment time. MWE heating is also able to separate bers from the cell walls of hemp stem (Sun et al. 2019). Microwave is electromagnetic radiation in the spectrum between 300 and 3×10 5 MHz, a region that lies between radio frequencies and infrared, and corresponds to wavelengths of 1cm to 1m (Motasemi and Ani 2012). Due to the thermal effects of MWE, a structural change occurs in the lignocellulosic materials in the presence of a solvent as the dipole molecules of the bers are intended to align themselves with the applied electric eld direction of the microwave. Due to the electromagnetic treatment of microwave, the bers can be acted as a non-homogeneous material that allow the bers to increase polarity, hence ber material can easily absorb the MWE through the formation of hotspots within the ber. Nair et al. (Nair et al. 2015) produced high quality bers using a microwave-assisted hot water degumming, evidenced by the Near Infrared (NIR) analysis of the degummed hemp bers. The degumming performance of natural bers mainly hinges upon the effective breakage of hydrogen bonds of hemicellulose and lignin. Therefore, it is hypothesized that the combined MWE-DES treatment of hemp bers can be a promising alternative to degum the hemp bast ber for attaining high-quality hemp cellulose bers with improved mechanical and thermal properties.
The objective of this study was to use combined MWE-DES treatment for degumming hemp bast bers to produce puri ed hemp cellulose bers for the subsequent textile, nanocellulose, and composite manufacturing. DES made of choline chloride and urea was used to treat hemp bers. Scanning electron microscopy (SEM), Fourier Transform Infrared (FT-IR), Near-Infrared Magnetic Resonance (NMR), and Xray Diffraction (XRD) techniques were used to evaluate the ber surface morphology and physiochemical properties. Thermal property characterization of bers was performed by Thermogravimetric (TG) and Differential Scanning Calorimetry (DSC).
Experimental Section MATERIALS Raw hemp bast bers were obtained from a local supplier. The bers contained a small percentage of woody core (hurd). Choline chloride (pure > 98%) was purchased from TCI America (Portland, OR, US), urea (crystallized) was procured from VWR BDH Chemicals (Radnor, PA, US), sodium sulfate (anhydrous) was acquired from VWR Life Science (Solon, Ohio, US), sodium metasilicate was purchased from BTC Beantown Chemicals (Hudson. NH, US), sodium hydroxide (ACS-grade pellets) was bought from Fisher Chemical (Fair Lawn, NJ, US), sodium polyphosphate (pure) was purchased from ACROS Organics (Fair Lawn, NJ, US), and hydrogen peroxide -30% (aqueous solution of ACS reagent grade) was supplied by J.T. Baker (Radnor, PA, US).

DEGUMMING PROCESS
The woody core (hurd) from the raw hemp bers was removed manually using a small carding machine.
The processed bast bers were stored in the laboratory condition before the degumming experiments.
Traditional alkali degumming. Cleaned raw hemp samples were cooked at 90°C for one hour with a 1% NaOH solution in an oil bath and the solid-liquid ratio was 1:20 (volume/weight). After boiling, the hemp bers were washed with tap water ve times to achieve neutrality. The recipe for the alkali degumming includes 1% sodium hydroxide (NaOH), 3% hydrogen peroxide (H 2 O 2 ), 3% sodium polyphosphate (NaPO 3 ), 2% sodium sulfate anhydrous (Na 2 SO 4 ), 3% sodium metasilicate (NaSiO 3 ), and 3% urea (CH 4 N 2 O) (Jiang et al. 2018). The ber-liquid ratio of the alkali degumming was 1:20 and all the chemicals were weighed according to the weight/volume ratio. After cooking with the abovementioned recipe for one hour, the hemp bers were removed and washed with tap-water ve times to achieve neutrality and the bers were slowly dried in an oven at 60°C. The dried bers were kept for further characterizations.
Combined MWE -DES degumming. Choline chloride was used as a hydrogen bond acceptor and urea was used as a hydrogen bond donor. DES (Figure 1) was synthesized from the mixture of choline chloride and urea at a molar ratio of 1:2. The mixture was heated at 80°C and stirred magnetically for about 2 hours in the oil bath until a transparent and homogenous solution was obtained. The hurd-free hemp ber samples were submerged into the DES solution in a 90 ml Te on vessel. The microwave extraction system (model: Ethos X, Milestone Inc, Shelton, CT, US) was used for the degumming of ber samples at 1:10 and 1:20 ber-DES ratio for 60 min and 90 min, respectively. The power and stirring levels of the microwave extraction system were 1000kW and 24%, respectively. After degumming (Figure 2), bers were washed with tap water, oven-dried, and kept for further characterizations. Some of the MWE-DES treated hemp bers were further subjected to mild alkali treatment. After the alkali treatment, the obtained bers were washed with tap water and oven-dried. Raw and treated hemp bers are designated as Raw in a high vacuum condition at an accelerated voltage of 5kV. Before scanning, the ber samples were cut into small pieces and coated with a thin layer of gold using an EMS550X sputter coater.
Fourier transform infrared analysis (FT-IR). Fourier Transform Infrared analysis (FT-IR) of hemp bers (both raw and degummed) was performed to identify the presence of free functional groups. FT-IR spectra were obtained using a Bruker Alpha & Tensor 27 FT-IR spectrophotometer and OPUS software.
The operating conditions were as follows: 32 scans per sample, resolution of 4 cm -1 , with a wavelength range of 400-4000 cm -1 .
X-ray diffraction (XRD) analysis. Hemp powdered bers (both raw and degummed) were analyzed by Xray diffraction using a PANalytical Empyrean X-ray Diffractometer (Malvern, UK) equipped with the PreFIX modules (Fast interchangeable X-ray). The diffractometer was operated at 40mA and 40kV with the Ka -1 wavelength of 1.540598, Ka -2 wavelength of 1.54426, Cu-anode material, scan range of 5-50, and scan step size of 0.0131303 to get the diffraction patterns of both raw and degummed hemp bers. The crystallinity index of raw and degummed hemp bers was analyzed using the Segal equation from the Xray diffraction spectra (Segal et al. 1959).
I 200 represents the maximum intensity of the lattice peak of 2q angle at 22.6 degrees, which is imputed to the crystalline region. I am denotes the intensity of 2q angle at 18 degrees, concerning the amorphous region of cellulose. The average crystallite size from the XRD diffractograms was calculated by using the Scherrer equation (Scherrer 1912).
where d = average crystallite size in nm, K = Scherrer constant = 0.94 for spherical crystallites, cosq = Xray wavelength, Cu Ka = 1.5406 angstrom, b = Line broadening at FWHM (Full-Width-Half-Maximum) in radians, and q = Bragg's angle in degrees, which is half of 2q.
Chemical composition analysis. The Solid-State NMR test of both treated and untreated hemp bers was carried out using a 3-channel Bruker AV-400 (Bruker BioSpin, Billerica, MA, US) equipped with the 400 MHz spectrophotometer. These solid-state instruments were tted with Z-axis gradients. The conditions for one-directional (1D) cross-polarization magic angle (CP/MAS 13 C NMR) were as follows: MAS rate-10kHz, 4096 Scans, relaxation delay-2s, and CP contact time-2ms. NMR data were analyzed and processed with the Topspin Software (version-2.1) (Bruker), Origin Pro (2021 version), and dm t (NMR@CEMHTI) software. The obtained data from Topspin and Origin software were processed by utilizing Gaussian/Lorentzian models through dm t software to obtain chemical compositions of both raw and degummed hemp bers.
Thermogravimetric analysis (TGA). Thermogravimetric analysis was performed for both raw and degummed hemp bers in a nitrogen atmosphere with a ow of 40ml/min by using a TG analyzer (Model-Q50, TA Instruments Inc., New Castle, DE, US). The heating rate was 10°C/min up to 600°C at room temperature. All TG samples were about 20 mg and kept in the platinum crucible. TG and derivative TG curves were obtained as a function of weight and temperature to investigate the thermal degradation of hemp bers.
Differential scanning calorimetry (DSC) analysis. DSC measurement was conducted with a Q10 DSC (TA Instruments Inc., New Castle, DE, USA) and operated at a 40 ml/min sample purge ow rate under a nitrogen atmosphere. Fiber samples were heated from 30°C to 540°C at a heating rate of 20°C/min. All DSC samples were about 5 mg.
UV-Vis spectrophotometry and UV shielding property analysis. A UV-vis spectrophotometer (Evolution 350, Thermo Scienti c, Waltham, MA, US) was used to investigate the UV-shielding performance of both raw and degummed bers according to the slightly modi ed Australia/New Zealand (AS/NZS 4399:2017) standard (AS/NZS 2017). Sample ber lms of treated and untreated hemp bers were made at 100°C using a small hot press. The obtained ber lms were scanned with the spectrophotometer equipped with an integrating sphere at a wavelength range of 290-400nm and a slit bandwidth of 1 nm. Transmittance was recorded at 5 nm intervals. The UV protection factor (UPF) is regarded as the protection e ciency of bers or fabric from UV radiation. A higher or lower UPF value is de ned as bers or fabric's higher or lower protection level from the UV radiation, respectively. UPF was calculated as where E λ = CIE Erythemal spectral effectiveness, S λ = Solar spectral irradiance for a typical summer day,  Figure 3i and j of DES-treatment at 1:20 DES-ber ratio), the hemp bers had a clean surface with a white glowing appearance which was comparable with alkali-treated ber surfaces. MWE-DES treated bers (D) (Figures 3i and j) had very few cracks and voids even after increasing the process temperature and time up to 120°C and 1.5 hours; this might be caused due to the uniform dispersion of microwave heating throughout the hemp ber structure and low freezing temperature of DES (12°C). On the other hand, subsequent alkali treatment of MWE-DES-treated bers (C and E) revealed some pits and cracks (Figure 3g, h, k, and l) in the ber surface due to the direct attack of alkali into the cellulose of the ber structure since DES treatment removed most of the gummy matters. In addition, lignin is strongly linked by the hydrogen bonds and formed lignin-carbohydrate complexes. However, these strong hydrogen bonds can be weakened due to the formation of competing hydrogen bonds between the hydroxyl groups of lignin-carbohydrate complexes and the chloride ions of the DES network; thus, lignin could be removed from the MWE-DES treated ber structures. However, the H + protons (from the hydrogen bond donor of DES) can selectively cleave the ester and/or ether linkages of lignin-carbohydrate complexes without the destruction of C-C bonds of cellulose, removing hemicellulose and lignin from the ber structure.  Crystallinity of hemp bers. Figure 5a represents the XRD spectra of treated and untreated hemp bers. The peaks at 14.6° and 16.4° were assigned to the characteristic peak of cellulose-I and appeared as a wide peak at raw hemp XRD spectrum due to the presence of lignin, pectin, and hemicellulose (Perel 1990). Bou 2011). The crystallinity % of the DES-treated bers was increased as compared to that of raw hemp ber (Raw). It was an obvious indication of the removal of non-cellulosic components (non-crystalline elements) from the hemp ber structure. It was reported from previous research that the removal of gummy materials facilitates the cellulose chains to be relaxed (increased stress relaxation) (Viscusi, Barra, and Gorrasi 2020). Therefore, micro brils of the cellulosic components might get rearranged; the polymeric chain and surface morphology after alkali or DES treatment could be altered due to the reaction of Na + and hydroxyl groups of bers (for alkali-treated), and cationic component (choline + ) of the DES network and hydroxyl groups of bers (for MWE-DES treated bers).
Chemical compositions of hemp bers. CP-MAS/ 13 C-NMR is an effective technique to illustrate the structure and chemical composition of polymeric materials. The peaks in the 60-110 ppm (Figure 6c) range revealed the solid-state NMR spectra of cellulose and hemicellulose. The chemical shift at 105 ppm corresponds to the cellulose C 1 carbons, and the up-eld shoulder peak at 103 ppm is attributed to the hemicellulose or xylan carbons (Terrett et al. 2019). The C 4 carbon of cellulose showed a peak in the spectra at 89 ppm, which is ascribed to the carbons of the crystalline region. The peak of 72 ppm to 75 ppm corresponds to the cellulose carbons superimposed to the nearby xylan peaks. Carbon C 6 of crystalline and amorphous cellulose has a similar type of observations and interpretations at 63 ppm to 65 ppm by Bonatti et al. (Bonatti et al. 2004). Carbonyl and methoxy groups of lignin showed peaks at 174 ppm and 56 ppm, respectively (Moussa et al. 2020). As shown in Figure 6b,d, raw hemp ber (Raw) had a broad peak at 174 ppm, and 56 ppm that gradually became absent in the Microwave-DES treated bers (as the temperature, time, and solid-liquid ratio increased) and alkali-treated bers. It was a clear indication of the removal of lignin and breakage of hemicellulose to be turned into glucose molecules and contributed to increased crystallinity and chemical composition rate of cellulose in the ber structure. This phenomenon has a closer agreement with the XRD spectra (Figure 5a) of the raw and degummed hemp bers. In addition, the broad peak at 21 ppm and the small peak at 32 ppm in the raw hemp are assigned to the methyl groups and methylene carbons of lignin and xylan, respectively. However, after DES treatment, those peaks were nowhere to be found (Figure 6b). The disappearance of the peaks is indicative of the removal of gummy compounds in the treated bers. Lignin, a complex biopolymer can be derived from the polymerization of phenyl propane monomers, synapsyl alcohol, coniferyl alcohol, and p-coumaryl alcohol (Hansen et al. 2016). These carbons of monolignol units exist in the aromatic lignin region as amino acid groups in the raw hemp bers, disappearing after the alkali and DES treatment (Figure 6d). The signals in the NMR spectra of raw hemp bers at 21 ppm and 174 ppm are the consequence of agglomeration and interaction of hemicellulose and methyl moieties associated with acetyl groups (Simmons et al. 2016). The signal at 81 ppm of C 4 is also generated due to the interaction of cellulose and xylan present in the raw hemp bers. This wide-ranging signal disappeared after DES and alkali treatment. This is caused by the removal of hemicellulose and lignin from the ber structure, which is also supported by the previous literature (Hosseinmardi et al. 2018). It is possible to nd the carbon ratios which existed in the lignin structure by an integration of the area between 110 ppm-160 ppm (Fu et al. 2015). According to the literature, the chemical shift at 64 ppm could be assigned to the C 5 of xylan or hemicellulose (Simmons et al. 2016), and this signal was attened after DES and alkali treatment, indicating the removal of hemicellulose. A similar explanation can be used for the NMR band at 84 ppm. Therefore, the integration of the area under those two peaks (64 ppm and 84 ppm) can be an estimation of hemicellulose content in the ber structure. Figure 6e represents an estimation of the chemical composition of hemp bers obtained from Solid-state NMR spectra using the Gaussian -Lorentzian model and peak deconvolution.
Thermal stability of hemp bers. Figures 7a, b, c, d, and e represent the TG and DSC curves of treated and untreated hemp bers. The heterogeneous chemical composition of natural bers makes thermal degradation complex. The thermal degradation of hemp bers from TGA can be divided into three thermal stages including water evaporation at 50°C -150°C, hemicellulose decomposition at 220°C -300°C, and cellulose degradation at 300°C -400°C. However, it is thought that the thermal degradation of lignin begins at roughly 280°C and ends at about 450°C. The complex aromatic nature of lignin enables lignin to keep thermal stability at a higher temperature; therefore, it can slow down the decomposition of cellulose to some extent in terms of the thermal degradation event of cellulose (Viscusi et al. 2021;Fisher et al. 2002). MWE-DES treated samples and alkali-treated sample both demonstrated a small amount of water evaporation as compared to the raw hemp sample because of the reduction of OH groups. It is evident from the TG thermograms (Figure 7b) that hemicellulose has lower thermal stability as the degradation of hemicellulose observed at 256°C for raw hemp and sample B (on the other side, the absence of the hemicellulose peak in the derivative TG curve indicates the removal of hemicellulose from the samples C, D, and E), which would facilitate the second stage thermal degradation (lignin thermal degradation) for a longer period (Viscusi, Barra, and Gorrasi 2020) which might help lignin to have higher thermal stability. At the third degradation stage, the peak at 300°C -375°C represents the cellulose degradation, and 375°C -400°C could be assigned to the oxidative degradation of cellulose. However, DES-treated bers, especially the samples C, D, and E had higher thermal stability (shifting the degradation peak toward higher temperature in TG curves for all the degummed samples) as the degradation starts from 320°C to 375°C. The main weight loss of raw hemp bers was recorded at about 225°C -390°C while the treated hemp bers showed their main weight loss in the 250°C -390°C region as can be seen in Figure 7a. The onset points of DES-treated hemp bers including alkali-treated bers moved from 287°C to 310°C -318°C, as similarly found by Fan et al. (Fan 2010) ( Table 2). The thermal degradation of cellulosic components usually occurs at higher temperatures compared to that of noncellulosic components. The increased onset temperatures in MWE-DES treated bers indicated the removal of non-cellulosic components. The removal of non-cellulosic components helps cellulose to be rearranged in such a way to increase crystallinity with the densely packed rearrangement of glucose molecules, and it renders high thermal stability to cellulose since densely packed crystalline rearrangements slow down thermal decomposition. Table 2. Summary of TG and DSC data of both raw and degummed hemp bers (T D , T max , and T g refer to thermal degradation, peak maximum, and glass transition temperatures in the TG and DSC curves) Bond water (OH groups) to cellulose and hemicellulose have higher thermal stability (Kabir et al. 2013), thus water evaporation occurred in the raw hemp in the range of 50°C -150°C as seen in the DSC curves (Figure 7e). On the other hand, the range in MWE-DES treated bers was 50°C -140°C. Water evaporation for the alkali-treated bers occurred at 50°C -135°C (Figure 7a). It can be assumed that there was therefore a reduction of hydroxyl groups in the DES-treated and alkali-treated samples. The removal of hemicellulose and improved hydrophobicity of the treated bers results in higher thermal stability for cellulosic compounds. According to TG data (Table 2), after 350°C in the rst heating run, more than 75% of the hemp samples was decomposed. However, in the DSC Figure 7e, the glass transition of dry cellulose occurs at 200-250°C depending on its molecular weight, crystallinity, and structure (Kubát and Pattyranie 1967). The T g of 70% crystalline cellulose containing 2% water (playing the role of a plasticizer) drops to 160°C (Szcześniak, Rachocki, and Tritt-Goc 2008). The T g of the raw hemp sample is clearly observed right around 250°C, but this transition is absent in all other alkali and MWE-DES treated samples, obviously due to the plasticizing effect of some remnants after heating to 150°C ( Table 2) since below T g , all materials become rigid, one may conclude that the rigidity of alkali and MWE-DES treated samples is practically absent and these bers might exhibit a certain degree of elasticity. An exothermic peak was observed at around 370°C for the raw hemp ber in the DSC curve ( Figure 7e). The exothermic peak for the MWE-DES treated bers was shifted towards a higher temperature and sample D demonstrated an exothermic peak around 450°C, which con rmed the higher thermal stability of the cellulose in the treated bers. Hemicellulose decomposition occurred between 50°C -320°C with an exothermic peak at 287°C for raw hemp bers. The disappearance of this exothermic peak in the treated bers validated the removal of hemicellulose. MWE-DES treated and alkali-treated bers showed a small endothermic peak at around 376°C -390°C (Figure 7e), thus, it can be concluded that the removal of hemicellulose resulted in higher thermal stability of the treated bers as the broad exothermic peak of cellulose in the raw hemp bers at about 372°C moved toward 435°C for the alkali-treated bers, while MWE-DES treated bers had a peak between 450°C -473°C. Lignin in natural bers normally has a decomposition range between 250°C -450°C (Gargol et al. 2021), which is followed by the raw hemp bers in this study. The aromatic structure of the lignin (aromatic hydrocarbons, hydroxy phenolics, guaiacyl, and syringyl compounds) enables lignin to have a broader range of degradation temperatures, which is supported by the MWE-DES treated bers and alkalized bers. However, DES treated bers and alkalized bers showed degradation of residual lignin up to 520°C (Figure 7e), indicating higher thermal stability of the treated bers. In summary, it can be concluded that puri cation and less variation occurred with MWE-DES treated hemp bers, which resulted in a higher thermal stability of the treated bers.
UV shielding performance of hemp bers. It is evident that all the hemp samples had excellent UV protection properties (all the hemp ber samples have the transmittance % at a range of 0% -0.06%) ( Figure 8a). Among all the hemp samples, sample C had the lowest UPF value, even though other MWE-DES treated samples also had excellent UPF values (Figure 7b). It is obvious that there is a tendency to have a lower value of UPF after treating with NaOH since aggressive alkali treatment removes most of the lignin content from the ber surface. The aromatic structure of lignin aids to block the UV radiation in the raw hemp bers. The phenolic hydroxyl, carboxyl, and carbonyl groups in the lignin structure are responsible for generating heat when the ber is subjected to the photon energy of UV light. However, these phenolic groups, especially hydroxyl groups in the lignin could absorb the generated heat and also are able to quench the active radicals through the mechanics of electron transfer (Dean et al. 2013). The presence of lignin in the ber structure facilitates improved UV blocking performance. For instance, MWE-DES treated bers at 1:10 ber-DES ratio (sample B) had a UPF value of about 118 and after NaOH treatment (sample C), the UPF value was reduced to 102.43. A similar trend was observed for samples D and E (Figure 8b). Since the viscosity plays a signi cant role in the deligni cation of biomass, the low viscosity of choline chloride and urea-based DES with MWE treatment accelerates the deligni cation process. A similar result was found from this work as the rate of lignin removal was higher than the hemicellulose removal. Therefore, DES made of choline chloride and urea is preferred to treat the hemp bers.

MWE
Urea as a hydrogen bond donor in the deep eutectic mixture with choline chloride formed a complex hydrogen bond network between components. It was found that the DES network acts as a three dimensional (3D) structure of complex ions in which involves molecules of two urea, one choline, and one chloride (Hammond, Bowron, and Edler 2016). Therefore, it can be hypothesized that urea helps to stabilize the deep eutectic behavior of the mixture by the formation of complementary hydrogen bond with chloride and choline. Thus, the degumming performance of hemp bers is dependent on the urea concentration in the DES and especially the molar ratio of urea and choline in the DES. From the NMR analysis of MWE-DES treated bers, it was con rmed that there was a signi cant reduction of lignin content in the treated bers. Therefore, DES made of urea and choline chloride at a 2:1 molar ratio was proven to be an effective deep eutectic mixture in terms of hemp ber degumming.

Conclusions
Degumming of hemp bers is crucial for further ber processing with respect to textile applications or composite manufacturing. A novel method was developed to degum the hemp bast ber in our study by utilizing DES and MWE. There was no signi cant difference between MWE-DES treated bers and traditional alkali-treated bers in terms of chemical structures, crystallinity, and thermal stability. One step MWE-DES treatment of hemp bers at 1:10 and 1:20 ber-DES ratio exhibited effective deligni cation with clean ber morphology as lignin content reduced to 6.16% in the treated bers at 1:20 ber-DES ratio.
MWE-DES treatment was carried out without the use of water and even combined MWE-DES treatment needs fewer chemical consumptions as compared to the traditional alkali treatment. Therefore, MWE-DES treatment can be a novel path for degumming of natural bers, including hemp bers, for the purpose of improving ber qualities in all distinct aspects of sustainable subsequent ber processing. Future research can be conducted on the reusability and recovery of used DES liquid.

Author Contributions
All authors contributed to the study conception and design. The study was supervised by Qinglin Wu. Material preparation, data collection and analysis were performed by Bulbul Ahmed. The rst draft of the manuscript was written by Bulbul Ahmed and all authors commented on previous versions of the manuscript. All authors read and approved the nal manuscript. Schematic showing reaction mechanism of DES system made of choline chloride and urea  FT-IR spectra of both raw and degummed hemp ber samples    UV-vis spectra (a) and UPF values (b) of both raw and degummed hemp bers treatment (sample C), the UPF value was reduced to 102.43. A similar trend was observed for samples D and E (Figure 8b).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.