A comparison of alkaline and organosolv lignin extraction methods from coconut husks as an alternative material for green applications

Abstract

Effects of alkaline (kraft and soda) and organosolv pulping were evaluated relative to the structural properties of lignin isolated from coconut husk (CH) biomass. The various types of functional groups within the isolated lignin samples were characterized and compared using a variety of complementary analyses including Fourier-transform infrared (FTIR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, gel permeation chromatography (GPC), and high-performance liquid chromatography (HPLC). All isolated CH lignin samples contained a significant quantity of non-condensed guaiacyl-type (G) and syringyl-type (S) units but a lesser amount of p-hydroxyphenyl (H) units, as demonstrated and supported by previous research. The alkaline CH lignin produced lignin with higher molecular weight (Mw SL: 959 g mol-1 > Mw KL: 769 g mol-1) than the organosolv lignin (Mw OL: 606 g mol-1) sample, resulting in smaller fragments and a higher degree of solubility in water or other solvents. Because of significant differences in the physicochemical characteristics of the various lignin polymers, their properties and structure were improved with respect to alternative approaches in lignin-based applications.

Full Article

A Comparison of Alkaline and Organosolv Lignin Extraction Methods from Coconut Husks as an Alternative Material for Green Applications

Nur Hanis Abd Latif,^a Nicolas Brosse,^b,* Isabelle Ziegler-Devin,^b Laurent Chrusiel,^b Rokiah Hashim,^c and M. Hazwan Hussin ^a,*

Effects of alkaline (kraft and soda) and organosolv pulping were evaluated relative to the structural properties of lignin isolated from coconut husk (CH) biomass. The various types of functional groups within the isolated lignin samples were characterized and compared using a variety of complementary analyses including Fourier-transform infrared (FTIR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, gel permeation chromatography (GPC), and high-performance liquid chromatography (HPLC). All isolated CH lignin samples contained a significant quantity of non-condensed guaiacyl-type (G) and syringyl-type (S) units but a lesser amount of p-hydroxyphenyl (H) units, as demonstrated and supported by previous research. The alkaline CH lignin produced lignin with higher molecular weight (M_{w SL}: 959 g mol^-1 > M_{w KL}: 769 g mol^-1) than the organosolv lignin (M_{w OL}: 606 g mol^-1) sample, resulting in smaller fragments and a higher degree of solubility in water or other solvents. Because of significant differences in the physicochemical characteristics of the various lignin polymers, their properties and structure were improved with respect to alternative approaches in lignin-based applications.

DOI: 10.15376/biores.17.1.469-491

Keywords: Coconut husk; Alkaline lignin; Organosolv lignin

Contact information: a: Materials Technology Research Group (MaTReC), School of Chemical Sciences, Universiti Sains Malaysia, 11800, Minden, Penang, Malaysia; b: Laboratoire d’Etude et de Recherche sur le MAteriau Bois (LERMAB), Faculte des Sciences et Technologies, Universite de Lorraine, Vandoeuvre-les-Nancy, France; c: School of Technology Industrial, Universiti Sains Malaysia, 11800, Minden, Penang, Malaysia; *Corresponding author: mhh@usm.my; nicolas.brosse@univ-lorraine.fr

INTRODUCTION

Many researchers have become interested in turning abundantly available biomass waste into bio-based products as a way to help mitigate the rise in global warming induced by non-renewable petroleum/fossil-based products. Biomass wastes are typically made from agricultural wastes, such as livestock feed, which is left to decompose naturally, and still, others are left unutilized (Johari et al. 2016; Zakaria et al. 2017; Ram and Mondal 2018). In Malaysia, coconut trees (Cocos nucifera L.) are widely cultivated and known for their important use based on their various parts (Kumar 2011; Johari et al. 2016; Saari et al. 2016). According to the Department of Statistics Malaysia in 2020, the highest fruit in terms of self-sufficiency ratio (SSR) and per capita consumption (PCC) from 2015 to 2019 was coconut, which is consumed at a rate of 22.7 kg per year, followed by other fruits in selected crops (Mahidin 2020). The outer layer of the coconut fruit, known as coconut husk (CH) is a major agro-industrial waste produced in developing countries every year that is biodegradable, low in cost, and readily available in large quantities (Kaewdang and Benjakul 2015; Johari et al. 2016; Saari et al. 2016; Ram and Mondal 2018). Coconut husk is composed of 30 wt% fibre and 70 wt% pith, which is primarily made up of cellulose (34.9 to 42.3%), hemicellulose (1.1 to 21.9%), and a rich amount of lignin (29.8 to 53.5%) (van Dam et al. 2004; Israel et al. 2011; Jincy et al. 2015; Saari et al. 2016). Furthermore, coconut husk fibre has been investigated for its ethanolic extract effect relative to the gel properties of gelatin (Kaewdang and Benjakul 2015), fuel production (Ram and Mondal 2018), cement material for building applications (Lertwattanaruk and Suntijitto 2015), and performance as adsorbents for the removal of heavy metal ions (Hg) (Johari et al. 2016).

Lignin is a well-researched natural feedstock for industrial applications that is widely studied such as an emulsifiers of catalytic cracking heavy oil (Xu et al. 2011; Upton and Kasko 2016), low-cost lignin carbon fibers manufacture (Baker and Rials 2013), bioplastic (Spiridon et al. 2015), biocomposite (Privas and Navard 2013), polyurethane foam (Luo et al. 2018), and in the production of value-added chemicals. In plants, lignin acts as an adhesive by filling the gaps between cellulose and hemicellulose and binding the lignocellulose matrix together as well as providing strength and stiffness to the cell walls (Qingqing et al. 2019). It is a complex three-dimensional biopolymer composed of three phenylpropanoid monomers: p-coumaryl alcohol (p-hydroxyphenyl, H units), coniferyl alcohol (guaiacyl, G units), and sinapyl alcohol (syringyl, S units) (Fig. 1) (Jincy et al. 2015; Aziz et al. 2019; Hussin et al. 2019; Latif et al. 2019; Qingqing et al. 2019). Recently, in a cellulose-based polymer electrolyte, CH lignin was shown to play an important part as a natural plasticizer by improving the ionic conductivity of the solid polymer electrolyte along with the presence of cellulose derivatives (Chua et al. 2020), and it has also been used for dye removal in carbon cryogel beads (Ngadi et al. 2015). Because CH lignin has a similar phenolic structure to phenol, with potential to cross-link with resins, it has shown a good performance especially in mechanical properties of exterior-grade plywood sheets used in industrial manufacturing (Saari et al. 2016; Aziz et al. 2019; Hussin et al. 2019).

Fig. 1. Lignin structure with the main inter-subunit linkages (Qingqing et al. 2019)

A separation or pulping process is needed to utilize lignin and process it into high-value products. The physicochemical characteristics of lignin obtained from alkaline pulping (kraft and soda) and organosolv pulping processes were different (Aziz et al. 2019; Latif et al. 2019). According to the previous studies, the alkaline processes are used to isolate lignin from wood content in the pulp and paper industry (96% of the market). The breaking down of aryl-ether linkages occurred during alkaline pulping, resulting in the liberation of the lignin fragments as well as their dissolution and condensation reaction. During the condensation procedure, the molecular size of lignin fragments increased because of the conjugate addition of carbanions to quinine methide intermediates (Hussin et al. 2013). Due to the absence of sulfur contents, the soda lignin obtained was much purer than kraft lignin, which also improved its reactivity (Tribot et al. 2019). Previously, organosolv pulping produced more reactive lignin with higher purity, resulting in a lower molecular weight that is mostly soluble in common solvents (Hussin et al. 2013; Latif et al. 2019). Moreover, their structure contains a high proportion of oxidized groups and phenolic hydroxyl groups (such as Hibbert ketones), which facilitates chemical modification and incorporation into polymer formulations (Hussin et al. 2013).

Consequently, the current research aimed to study the structural and physicochemical characteristics of lignin isolated from coconut husk (CH) using various pulping techniques (kraft, soda, and organosolv pulping), as well as its potential as a phenolic substitution for material applications. There have not been several studies contrasting the functionalization of alkaline and organosolv lignin isolated from CH biomass. The isolated kraft lignin (KL), soda lignin (SL), and organosolv lignin (OL) were characterized using Fourier transform infrared (FTIR) spectroscopy, gel permeation chromatography (GPC) analysis, and ¹³C- and 2D-heteronuclear single quantum coherence (HSQC) nuclear magnetic resonance (NMR) spectroscopy to investigate any changes in the structure of lignin after undergoing the pulping process. Additionally, thermal analyses were used to analyze the thermal activity of extracted lignins. Simultaneously, the phenolic acids, aldehydes, and sugar analysis of lignin were determined using high-performance liquid chromatography (HPLC) and high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). To optimize the proper applications of lignin, the physicochemical characteristic of lignin isolated from CH were compared with several previous works.

EXPERIMENTAL

Materials

Coconut husks (CH) were collected in Sungai Petani, Kedah, Malaysia (5^ᵒ39′08.5″N, 100^ᵒ31′17.5″E) in August 2019. The husks were sun-dried for 2 days to remove any excess moisture before being ground into powder using a stainless steel laboratory blender (Waring Commercial, Torrington, CT, USA). Then, the ground CH was sieved into small particles (1 to 3 mm) and further dried in an oven at 50 °C for 24 h. The chemical composition (% w/w) of the CH biomass in term of moisture content (9.26 ± 0.15%), ash content (2.50 ± 0.002%), extractives content (9.16 ± 0.14%), cellulose content (27.19 ± 0.10%), hemicellulose content (14.64 ± 0.70%), and Klason lignin (37.93 ± 0.91%) was determined based on the National Renewable Energy Laboratory (NSEL) laboratory analytical procedure (LAP) (Sluiter et al. 2005a,b, 2008a,b) and TAPPI T203 cm-09 (2009) standard. The obtained chemical composition in CH biomass was compared with previous studies and is summarized in Table 1. For this study, the chemicals used such as nitrobenzene, p-hydroxybenzaldehyde, syringic acid, syringaldehyde, vanillic acid, p-hydroxybenzoic acid, vanillin, ferulic acid, and p-coumaric acid, were obtained from Sigma-Aldrich (St. Louis, MO, USA). Toluene (grade AR), ethanol (99.7%), sulfuric acid (98%), sodium hydroxide (97%), hydrochloric acid (37% v/v), glacial acetic acid (99.8%), dichloromethane, and acetone were purchased from QReC (Rawang, Malaysia). Acetic anhydride (≥ 98.5%), acetonitrile (HPLC grade), and dimethyl sulfoxide-d6 (d6-DMSO) were obtained from Merck (Darmstadt, Germany). Chloroform (99.8%) and tetrahydrofuran (HPLC grade 99.99%) were from Fisher Chemicals (Loughborough, UK). Sodium sulfide (95%) was purchased from R&M Chemical (Essex, UK), as well as pyridine (≥ 99.5%) was purchased from Bendosen (KL, Malaysia). Sodium chlorite (80%) pure was obtained from Acros Organics (Morris Plains, NJ, USA). In all experiments, distilled water was used, except that ultrapure water was used for running HPLC and HPAEC-PAD analysis.

Table 1. Comparison of Chemical Composition of Coconut Husk Biomass with Previous Studies (Israel et al. 2011; Lertwattanaruk and Suntijitto 2015; Ram and Mondal 2018)

Extraction of Lignin Using Pulping Process

A 0.2-L TGYF-A stainless steel small high pressure chemical reactor (Henan Lanphan Industry Co. Ltd., Henan, China) was used when lignin was isolated under an alkaline (kraft and soda) and organosolv pulping process. In the alkaline pulping process, all of the conditions were the same as described in the authors’ previous works (Hussin et al. 2013; Aziz et al. 2019) with minor changes. Kraft pulping process was initiated with the addition of active alkali (20%) and sulfidity (30%) at a ratio of 1:10 (solid to liquid, S:L). Meanwhile, in the soda pulping process, 30% of active alkali was used in the same ratio as described above. The maximum temperature, time, and pressure of the two pulping processes were set to 170 °C (12 to 15 bar) for 3 h. The final products (pulp and black liquor) were collected by separation using vacuum filtration. After that, the concentrated black liquor was acidifed to pH 2 by dropwise addition of 20% (v/v) of sulfuric acid (H₂SO₄) to precipitate the kraft and soda lignin. Both precipitated lignins (KL and SL) were centrifuged at 3000 rpm for 15 min and oven-dried (50 °C) for 24 h. Subsequently, dry kraft and soda lignin were washed with excess water of pH 2 to remove any impurities or the residual of the hemicellulose compound before being dried in the oven (50 °C) for another 24 h.

In organosolv pulping, the addition of ethanol: water (65:35, v/v) and 0.5% (w/w) of sulphuric acid as a catalyst was mixed with the same S:L ratio as mentioned in alkaline pulping. As outlined in the method by El Hage et al. (2010), the cooking temperature used was at 190 °C for 1 h with continuous stirring. After the process, the fibrous residue was washed three times (3 × 50 mL) with 80% warm ethanol (v/v), and the concentrated black liquor was secured using vacuum filtration. Then, the concentrated black liquor underwent precipitation with three volumes of water and the lignin was collected using the same centrifuge conditions as described above for alkaline pulping. Finally, the wet solid organosolv lignin (OL) was oven-dried at 50 for 24 h before being stored in a vial for further analysis (Hussin et al. 2013; Latif et al. 2019). The percentage yields of all collected lignin (KL, SL, and OL) were calculated using Eq. 1 below (Aziz et al. 2019):

(1)

Klason Lignin (HPAEC-PAD Analyses)

All the CH pulp of kraft, soda, and organosolv pulping of extractive-free samples were powder-ground, and the moisture content was measured using a Kern MRS 120-3- Mettler Toledo moisture analyzer (Greifensee, Switzerland). Based on the method illustrated by Tahri et al. (2016), approximately 0.175 g of dried extractive-free pulp samples were measured and put inside the 50-mL centrifuge tubes. Then, 72% of sulfuric acid (H₂SO₄) was added to approximately 1.5 mL in the tubes and the mixture was stirred continuously around 5 min using a glass rod. After stirring, the tubes were incubated in a 30 °C water bath for 1 h before the addition of distilled water (42 mL) and kept in an autoclave for 1 h. After the tubes were cooled down, the samples were filtered using glass microfibers in a Buchner funnel. The sample residue was washed with distilled water and left to dry in a 105 °C oven for Klason lignin content. As for the filtrate (soluble fraction), it was transferred into a round bottom flask and diluted to 100 mL with distilled water for HPAEC-PAD analysis (Dionex ICS-3000, Dionex Corp; Thermo Scientific, Sunnyvale, CA, USA).

From the Klason lignin hydrolysis, the monosaccharide contents of soluble fractions can be separated and analyzed using HPAEC-PAD (ICS-3000 Dionex) equipped with a Dionex CarboPac PA-20 (3 x 150 mm) analytical column, as illustrated in Tahri et al. (2016) and Obame et al. (2019). Approximately 20 μL of filtered samples were eluted at 35 °C with a flow rate (0.4 mL/min) under the following conditions: 99.2% pure water and 0.8% 250 mM NaOH (0 to 20 min); 75% pure water, 20% 250 mM NaOH, and 5% 1 M CH₃COONa/20 mM NaOH (20 to 37 min); and 40% pure water, 20% 250 mM NaOH, and 40% 1 M CH₃COONa/20 mM NaOH (37 to 41 min). After each elution, there was a wash and a period of equilibration. The external sugar and uronic acid standards were used for calibration with seven points per curve. All the sugar-containing monosaccharides (such as fucose, glucose, xylose, galactose, mannose, rhamnose, arabinose, galacturonic acid, and glucuronic acid) standards were supplied by Sigma-Aldrich (St. Louis, MO, USA).

Lignin Characterization

Fourier transform infrared analysis

A Thermo-Nicolet IR 2000 (Thermo Scientific; Waltham, MA, USA) FTIR spectrometer coupled with attenuated total reflectance (ATR) head of Perkin Elmer model (Waltham, MA, USA) was used to analyze the isolated alkaline and organosolv lignin (KL, SL, and OL) samples under the indirect transmittance mode. The spectra of the vibration bands were recorded across the 600 to 4000 cm^-1 region with a resolution of 4 cm^-1 and 20 scans per sample.

Acetylation of lignin

According to Faris et al. (2015), all lignin samples need to undergo an acetylation process to enhance their solubility in organic solvents, especially for NMR spectroscopy and GPC techniques. During the acetylation process, all the hydroxyl (-OH) groups are replaced with acetyl (-COCH₃) groups. This process was done by dissolving approximately 1.2 g of lignin in an acetic anhydride/pyridine mixture (1:1), stirring continuously for 24 h at ambient temperature. After cooling the mixture to 0 °C with ice cubes, approximately 150 mL of 1% (w/v) hydrochloric acid (HCl) was added dropwise until precipitation developed. Finally, the solid precipitation was collected by vacuum filtration and dried in a 60 °C oven for 24 h after being washed with an excess of water until the pH of filtrate became neutral (pH 6 to 7).

For liquid-state ¹³C and 2D-HSQC NMR spectroscopy, a volume of 0.5 mL d6-DMSO solvent was added to the acetylated lignin samples (KL, SL, and OL), and the samples were lightly heated with continuous stirring until completely dissolved. All NMR spectra were acquired using a Bruker Avance 500 MHz NMR spectrometer (Fallanden, Switzerland) with 22000 scans at 50 °C and interpreted with Bruker TopSpin software (v.3.5, Oxford, MS, USA). Next, the measurement of the average molecular weight (M_w) and average molecular number (M_n) were measured using an Agilent GPC model 1200 series analyzer connected to a UV detector (Agilent Technologies, Santa Clara, CA, US). An HPLC grade of mobile phase (tetrahydrofuran, THF) was used with a flow rate of 1.0 mL min-1 in a Waters 1525 Binary HPLC pump. In brief, approximately 15 mg of acetylated lignin was mixed in 1.5 mL of THF solvent, and elimination of impurities was done using a 0.40 μm filter. Following that, the polydispersity index (PDI) was calculated using the ratio (M_w/M_n) after about 20 μL of solution (filtered) was injected into the GPC instrument.

Thermal analysis

The thermogravimetric analyses (TGA) of lignin samples (KL, SL, and OL) were studied using a Perkin Elmer TGA 7 analyzer (Waltham, MA, USA). The TGA analysis was performed at a continuous heating rate of 20 °C min^-1 in a nitrogen atmosphere (N₂) across a temperature range of 30 to 900 °C. Differential scanning calorimetry (DSC) analysis was carried out with a Perkin Elmer Pyris 1 (Waltham, MA, USA) to determine the glass transition temperature (T_g) of each lignin sample. Each lignin sample was heated at a constant heating rate of 20 °C min^-1 over a temperature range between 30 and 200 °C.

Phenolic acids and aldehydes composition

A simple nitrobenzene oxidation method was used to determine the S/G ratio of coconut husk lignin by analyzing using high-performance liquid chromatography (HPLC) with minor modifications, as implied by the previous works (Hussin et al. 2013; Latif et al. 2019). Ground-dried lignin (approximately 50 mg) was placed in a steel autoclave with a mixture of 2 M NaOH and 5 mL of nitrobenzene before undergoing oil bath heating at 165 °C for 3 h. Then, the solution was cooled down in ice water before being transferred to a glass-liquid separation funnel. The extraction was done three times with the addition of chloroform in the separation funnel to get rid of the nitrobenzene reduction product. The oxidation mixture (upper layer) was acidified by adding concentrated HCl solution until reaching a pH of 2 and 3 before the second extraction was done using chloroform. After the second extraction, the excess chloroform solvent inside the oxidation mixture was removed by an evaporation process using a rotary evaporator heated between 48 and 50 °C. The final volume of an organic mixture (as a stock solution) obtained was dissolved with dichloromethane solvent up to 10 mL. A volume of 0.4 mL of stock solution was pipetted into a 50-mL volumetric flask with subsequent dilution up to the mark with acetonitrile/ultrapure water mixture (1:2, v/v). To avoid any impurities and insoluble particles before being injected into the HPLC system, the mixture of stock solution was filtered using a 0.40-μm Millipore membrane filter. Without any doubt, the classification of phenolic acids and aldehydes contents in CH lignin was detected using the Shidmazu HPLC system (Kyoto, Japan) coupled with Thermo Hypersil Gold C18 column (particle size 5 μm, length 250 mm, inner diameter 64.6 mm) by injecting approximately 20 μL of the solution. For the mobile phase, a mixture of acetonitrile/ultrapure water (1:8, v/v) together with 1% (v/v) of acetic acid was used in the HPLC system and operated at a flow rate of 1 mL min^-1. Each chromatogram of the samples was supervised at 280 nm using an ultraviolet (UV) detector (Shidmazu HPLC system; Kyoto, Japan).

RESULTS AND DISCUSSION

Proximate Analysis

Table 2 summarizes the percentage yields of lignin samples (KL, SL, and OL), as well as the sugar analysis of the resulting hydrolysates with HPAEC-PAD. Kraft and soda lignins produced higher percentage yields (%) from the alkaline pulping than organosolv lignin. This revealed that alkaline pulping had more severe reaction conditions, which stimulates the delignification process of lignin from CH biomass (Hussin et al. 2013; Domínguez-Robles et al. 2017). At the same time, organosolv pulping produced a low percentage yield of lignin due to the use of ethanol solvent and low severity reaction conditions, which dissolves the small molecular weight fraction of lignin (Hussin et al. 2013; Latif et al. 2019). According to the literature on the monosaccharide composition of coconut husk (CH), all the pulp in this study after alkaline and organosolv pulping contained a higher proportion of glucose and xylose.

Table 2. Sugar Analysis of Kraft, Soda, and Organosolv Lignin from Coconut Husk (% w/w on Dry Matter)

All the glucose comes from cellulose, a linear macromolecular polysaccharide that is made up of β-1,4-glycosidic bonds that connect a long chain of glucose units. Meanwhile, hemicellulose monosaccharide units primarily consist of xylose, arabinose, galactose, mannose, and some low-content saccharides (rhamnose and fructose) (Wang et al. 2017). The cellulosic composition in alkaline substrates pulp (KL and SL) was lower in glucose composition than that of organosolv substrate pulp (OL), resulting in higher removal of lignin because alkaline lignin produced a higher percentage yield of lignin (Sangian and Widjaja 2018).

Structural Characterization of Lignin

FTIR analysis

Fourier-transform infrared spectroscopy was used to observe any structural differences in the peak positions of vibrational bands between the three types of lignin compounds (KL, SL, and OL). Figure 2 depicts absorption band functionalities within the expanded region (600 to 1800 cm^-1) of FTIR spectra in all CH lignin samples, and the assignment bands are compiled in Table 3. The KL and SL spectra revealed moderately intense G bands (around 1271 and 1032 cm^-1) and weakly intense S bands (around 1331, 1112, and 826 cm^-1) while the OL spectra revealed quite strong G and S bands. The absorbance signal of S unit bands (1327, 1114, and 826 cm^-1) in OL spectra was noticeably higher than that of G bands in KL and SL, suggesting that the OL sample contained more syringyl content, as reported in the previous literature (Hussin et al. 2013).

Fig. 2. FTIR spectra of KL, SL, and OL isolated from CH

Table 3. FTIR Assignment Bands of Coconut Husk Lignin

The use of mixture solvent (ethanol-water) in organosolv pulping leads to the presence of free acetic acid from hemicelluloses, which contributes to lignin dissolution. As a result, it stimulates acid hydrolysis of the lignin structure, resulting in more intense peaks of the phenolic groups (1367 cm^-1) and carbonyl groups (1700 cm^-1) than alkaline pulping (Latif et al. 2019). In comparison, KL and SL samples had remarkably higher G bands intensities than OL, which was primarily because of the demethoxylation during the alkaline pulping process. It has been stated that demethoxylation reactions occur under severe reaction conditions during the alkaline pulping process, resulting in more stable G-type structures from S-type groups within the lignin molecules (Hussin et al. 2013; Aziz et al. 2019). According to Aziz et al. (2019), the presence of more G-type units in the lignin structure demonstrated that lignin had a better potential as an active site for polymerization. Furthermore, the typical absorption signal of HGS units for all lignin isolated from CH is comparable to wood lignin rather than occurring plant lignin due to a lack of hydroxyl phenyl propane (H) band in the spectra around 1160 cm^-1 (Hussin et al. 2018). Hereby, it was confirmed that all the CH lignin structures produced are primarily made up of S-type and G-type units.

NMR spectral analysis

The structural properties of all acetylated CH lignin were further characterized via NMR techniques. All the spectra were interpreted and compared with the other previous research as a reference on acetylated lignin (Sa’don et al. 2017a,b; Aziz et al. 2019; Latif et al. 2019). As shown in Fig. 3 and summarized in Table 4, the ¹³C NMR spectra of all acetylated lignin were qualitatively and quantitatively measured and placed into different categories of hydroxyl groups (-OH) and their derivatives. As the reference, the integral of the 162 to 102 ppm region was used, with six aromatic carbons and 0.12 vinylic carbons assumed. The integral value was divided by 6.12, which corresponded to one aromatic ring (Hussin et al. 2013). Within the integral region between 172 and 165 ppm chemical shift, the number of hydroxyl (-OH) groups in acetylated lignin samples were calculated. In all acetylated lignin spectra, the presence of guaicyl, syringyl, and p-hydroxyphenyl units was associated with the aromatic region at 100 to 160 ppm (Fig. 3). According to the previous studies, the signals of S units were revealed around 104, 139, and 154 ppm (carbon positions C2/C6, C4, and C3/C5), while G units signals were found at 113, 119, 135, and 146 ppm (carbon positions C2, C6, etherified C1, and non-etherified C3). The signal of 127 ppm was attributed to the H units in both alkaline and organosolv lignin samples. Furthermore, the carbon in carbonyl and carboxyl groups produced by the aliphatic esters and aliphatic carboxyls in lignin structures resulted in 172 ppm and 170 ppm showing a strong and moderate signal, respectively (El Hage et al. 2009; Hussin et al. 2013; Aziz et al. 2019). Moreover, the higher G unit signals in all acetylated lignin samples implied that CH lignin consists of a higher content of guaiacyl (G) units compared to the syringyl (S) units, as indicated by FTIR analysis and reported by Aziz et al. (2019).

Based on the integration of acetylated lignin samples, Fig. 3 and Table 4 show the amount of primary aliphatic (172.0 to 169.6 ppm), secondary aliphatic (169.6 to 168.6 ppm), phenol hydroxyl groups (168.6 to 165.0 ppm), and methoxy groups (58 to 54 ppm). The methoxy groups (-OCH₃) in G and S units in the lignin structures were assigned to the sharp peak observed at signal 57 ppm. It was detected that OL had a higher quantity of primary aliphatic – OH (0.49/Ar) than KL (0.13/Ar) and SL (0.16/Ar). For the amount of secondary aliphatic – OH, the KL (0.28/Ar) and SL (0.27/Ar) were nearly unchanged, but the OL (0.18/Ar) was remarkably lower. According to the previous study, the value of primary aliphatic-OH groups driven from integrals at 172 to 165 ppm and 65 to 58 ppm (0.97/Ar) had a good correlation for OL, indicating a higher amount of primary aliphatic-OH (0.49/Ar) functional groups in lignin structures (Hussin et al. 2013).

Fig. 3. ¹³C NMR spectrum of acetylated coconut husk KL, SL, and OL

Moreover, the amount of phenolic-OH in SL (0.25/Ar) and KL (0.19/Ar) were higher than in OL (0.15/Ar) because during alkaline pulping, the CH biomass undergoes a significant cleavage of aryl-ether bonds, resulting in a greater amount of phenolic hydroxyl groups. According to Aziz et al. (2019), in a study on the utilization of alkaline lignin from coconut husk biomass, the phenolic-OH content of SL (0.82/Ar) was also higher than that of KL (0.76/Ar), as shown in Table 4. As a result, it has been demonstrated that isolation of lignin from CH biomass using soda pulping produced lignin structures with a high amount of phenolic content. Furthermore, an increased amount of phenolic-OH content in the lignin structure allows for the activation of free ring position during the polymerization reaction, making it more reactive especially during the resinification process of phenolic resins (Hussin et al. 2018; Aziz et al. 2019).

Table 4. Comparison of Signal Assignment for ¹³C NMR Spectroscopy of Acetylated Coconut Husk of KL, SL, and OL with the Previous Study (Aziz et al. 2019)

The 2D-HSQC NMR spectrum was used to obtain details on the structural composition of all lignin samples. In the aromatic region of the HSQC spectrum, the main signals (δ_C/δ_H 160 to 100/6.0 to 8.0 ppm) are depicted in Fig. 4 (a, b, and c). According to the previous studies (Hussin et al. 2014; Sa’don et al. 2017a; Latif et al. 2019), the correlation at δ_C/δ_H (104.8/6.73 ppm) was assigned to C₂/C₆ position of syringyl (S) units, while the correlation for guaiacyl (G) units was assigned at δ_C/δ_H 107.7/6.9 to 7.2 ppm (C₂/H₂) and δ_C/δ_H 115.4/6.8 to 6.9 ppm (C₅/H₅). Furthermore, the signal for p-hydroxyphenyl (H) units was assigned at a correlation of δ_C/δ_H (130.5/7.25 ppm). This information was in strong agreement with the FTIR and ¹³C NMR results, which confirmed kraft, soda, and organosolv lignin as natural HGS units lignin. The p-coumarate (pCE 2/6) correlations were encountered at the δ_C/δ_H (126.8/7.4 to 7.6 ppm) region.

Fig. 4. 2D-HSQC NMR spectrum of acetylated coconut husk a) KL, b) SL, and c) OL

GPC analysis

Gel permeation chromatography was used to investigate the changes in molecules size and the distribution of alkaline and organosolv acetylated lignin samples. Table 5 shows the polydispersity value (PD) of the lignin fraction, which was calculated by dividing the value of average molecular weight (M_w) by the average molecular number (M_n). Based on Table 5, the SL sample had the highest M_w (M_{w SL}: 959 g mol^-1 > M_{w KL}: 769 g mol^-1 > M_{w OL}: 606 g mol^-1) and polydispersity values (PD) (PD _SL: 2.44 > PD _KL: 2.32 > PD _OL: 1.37) when compared to KL and OL samples. This observation could be attributed to the different lignin compositions produced by the various pulping processes. According to the previous studies, the molecular weight of lignin may be related to some amounts of C-C bonds between phenylpropanoic units (involved in the C5 aromatic ring), even though β-O-4 linkages are the most common linkages in all types of lignin. It was reported that when guaiacyl-type (G) units were compared to syringyl-type (S) units, the G-type units were more adequate to form these C-C bonds because S-type units contained both substitution of C3 and C5 positions of methoxy groups. This explains why during the pulping process, this kind of C-C bond does not cleave due to its high stability. Thus, lignin with more guaiacyl-type (G) units was anticipated to result in an increased amount of molecular weight than lignin with syringyl-type (S) units (Hussin et al. 2013; Aziz et al. 2019). The difference in M_w values between KL and SL could be associated with the repolymerization reaction that occurs during the alkaline pulping method. During the kraft pulping method, the hydroxide and hydrosulfide anions react with lignin molecules, resulting in smaller water/alkali soluble-fragments of the polymer. Because of the production of a smaller molecular weight of KL, this fragmentation was extended through the cleavage of α-aryl ether and β-aryl ether bonds, which is the primary process in alkaline medium (Ibrahim et al. 2011). The results obtained for alkaline lignin (KL and SL) were further reviewed with the previous research (in terms of the same source of raw material and pulping process). The M_w value of SL (M_{w SL}: 2076 g mol^-1), and KL (M_{w KL}: 1639 g mol^-1) (Aziz et al. 2019), and an ethanol soluble fraction of coconut shell (M_{w EtOH-F}: 1598 g mol^-1) (Avelino et al. 2019) were higher than M_w values obtained in this study. As previously reported, the OL would result in low M_w value with smaller fragments due to the breaking of carbohydrate-lignin α-O-4 bonds and β-O-4 bonds to a limited extent during the organosolv pulping process. Because OL has a lower polydispersity (PD) value than KL and SL, it is predicted that OL will have a higher solubility in some organic solvents than other lignin derivatives (Hussin et al. 2013; Sa’don et al. 2017a, 2017b; Latif et al. 2019). Many studies focused on the preparation of phenolic resins with a higher molecular weight of lignin samples (KL and SL) due to the activation of more free ring positions in G-type units (Hussin et al. 2018; Aziz et al. 2019; Hussin et al. 2019).

Table 5. Comparison of GPC Analysis of All Acetylated Alkaline and Organosolv Lignin with the Previous Studies (Avelino et al. 2019; Aziz et al. 2019)