Hydrolysis lignin (HL) refers to a lignin\u2010rich residue obtained after the enzymatic hydrolysis of biomass. It is recalcitrant, heterogeneous, insoluble in most common solvents, and less reactive than other lignins. To enhance the reactivity of HL, a novel environmentally friendly depolymerization approach was demonstrated to produce depolymerized hydrolysis lignin (DHL) using Kraft cooking liquor, white liquor (WL) \u2013 recoverable by the Kraft recovery cycle. The effects of various process parameters such as reaction time, WL\/HL ratio, and reaction temperature on lignin depolymerization were investigated using a 2 L Parr reactor under N2. The DHLs obtained were then characterized by Gel permeation chromatography (GPC)\u2010Ultraviolet Detector (GPC\u2010UV), 31P Nuclear Magnetic Resonance (NMR) spectroscopy, and ultraviolet visible (UV\u2013visible) spectroscopy, while the filtrates were characterized by high\u2010performance liquid chromatography (HPLC) for saccharinic acids. The DHL yield reached 45\u201370% from the treatments at 150\u2013190\u2009\u00b0C for 1\u00a0h at a WL\/HL mass ratio of 1:4\u2009~\u20092:1. The weight\u2010average molecular weight (Mw) value of the DHL obtained at 190\u2009\u00b0C after treatment for 1\u00a0h at a WL\/HL ratio of 2:1 (w\/w) was 2600\u2009Da. Moreover, a significant increase in non\u2010condensed phenolic hydroxyl and carboxylic acid group content was observed with decreasing Mw. Compared with various existing lignin modification approaches, the approach reported here is less expensive and more environmentally friendly if integrated into Kraft pulp mill operations with the residual WL from the lignin depolymerization process being recycled to the mill chemical recovery cycle. Process scale\u2010up was also demonstrated using a 20 L circulating reactor. In this case, the Mw of the DHL produced after treatment at 170\u2009\u00b0C for 2\u00a0h was 2400\u2009Da. \u00a9 2020 Society of Chemical Industry and John Wiley & Sons, Ltd.<\/p>\n
Introduction
\nIt is widely recognized that continued use of fossil resources is not sustainable, not only because of the finite amounts of these resources but also because of the toxicity of several fossil\u2010derived products and their negative impact on climate change.1 It is therefore essential to develop sustainable technology solutions that can address these growing concerns.<\/p>\n
Renewable energy is energy obtained from natural resources, which can be replenished or renewed within a human lifespan. It is viewed as a potential solution to the growing global demand for clean energy. In addition to energy, there is a need to develop chemicals and materials out of sustainable and widely available resources such as biomass. Plant biomass contains mainly three polymers: cellulose, hemicellulose, and lignin along with smaller amounts of pectin, protein, extractives, and ash.2 Cellulose is a homopolysaccharide composed of \u03b2\u2010d\u2010glucopyranose units which are linked together by \u03b2 (1\u20104) glycosidic bonds to form a linear polymer. Cellulose consists of about 40\u201345% dry wood and, as a result of its largely crystalline structure, is resistant to hydrolysis. Hemicelluloses are heterogenous and amorphous polysaccharides composed of various monomeric sugars and sugar acids. Hemicelluloses consist of 17\u201325% dry wood and can relatively easily be hydrolyzed by acids and \/ or enzymes to their monomeric components.3 Lignin is a complex and amorphous biopolymer composed of non\u2010fermentable oxygenated aromatics cross\u2010linked in three dimensions. It consists of about 28\u2009wt% softwoods and 20\u2009wt% hardwoods based on dry wood.4 Lignin is composed of three phenylpropanoid building units: p\u2010hydroxyphenylpropane, guaiacylpropane, and syringylpropane, interconnected by etheric and carbon\u2010to\u2010carbon linkages. Generally, in unprocessed lignin, 60% or more of these linkages are ether bonds, while the remaining linkages are carbon\u2013carbon bonds.5 Hydrolysis lignin (HL) is a byproduct from acid or enzymatic biomass pretreatment processes such as those employed in cellulosic sugar and \/ or ethanol plants. Hydrolysis lignin could be, for example, the solid residue from the enzymatic hydrolysis of woody biomass and is mainly composed of lignin (50\u201365\u2009wt%), cellulose and hemicellulose residues.<\/p>\n
Extensive research was undertaken in the former Soviet Union to find uses for acid hydrolysis lignins as they had several wood hydrolysis plants.6 Hydrolysis lignin’s fundamental difference from Kraft lignin is that it contains high amounts of bound residual polysaccharides (mostly insoluble cellulose) after acid or enzymatic hydrolysis, which decreases its purity, solubility, and general reactivity.7 Apart from incineration, several chemical modifications of HL were carried out at the time in an effort to come up with cost\u2010effective uses of this abundantly available polymer. However, most of the HL was disposed of, because the modifications developed were too expensive, or the products made did not find suitable applications. Similar problems are faced by researchers today in developing effective uses of HL.<\/p>\n
Lignin depolymerization is one of the most promising routes recently applied to improve lignin reactivity. Several lignin depolymerization processes (via hydrolytic, reductive, or oxidative routes) have been reported in the literature. For example, Nguyen et al.8 reported a high\u2010pressure pilot process for the hydrolytic conversion of Kraft lignin (KL) into bio\u2010oils and chemicals in near critical water (350\u2009\u00b0C, 25\u2009MPa) employing a fixed\u2010bed catalytic reactor filled with ZrO2 pellets, while the lignin was dispersed in an aqueous solution containing K2CO3 (catalyst) and phenol (co\u2010solvent). Mahmood et al.9 achieved the depolymerization of Kraft lignin via hydrolysis, using aqueous NaOH as a catalyst. The process itself was very effective in achieving good\u2010quality depolymerized Kraft lignin (DKL). However, the weight\u2010average molecular weight (Mw) of the DKL was >\u20095000\u2009g\/mol after depolymerization at 250\u2009\u00b0C, 45\u2009min reaction time at 20\u2009wt% KL concentration. The Mw of DKL could be reduced to ~1500\u2009g\/mol after depolymerization at 350\u2009\u00b0C for 45\u2009min or at 250\u2009\u00b0C for 2\u00a0h at 10\u00a0wt% KL concentration. However, the reactor pressure increased from 5\u00a0MPa to 16\u2009MPa with a corresponding increase in temperature from 250\u2009\u00b0C to 350\u2009\u00b0C. In other work, Yuan et al.10 also achieved successful depolymerization of KL into oligomers in hot\u2010compressed water\u2010ethanol medium with NaOH as the catalyst and phenol as a capping agent.<\/p>\n
A promising approach was reported recently for extracting lignin from woody biomass at high yield, high purity, and low molecular weight, by using a deep eutectic solvent (DES). Deep eutectic solvents are systems formed from a eutectic mixture of Lewis or Br\u00f8nsted acids and bases. Such mixtures have a melting point much lower than either one of the individual components. Deep eutectic solvent mixtures have been prepared, for example, from choline chloride (ChCl) and four hydrogen\u2010bond donors \u2013 acetic acid, lactic acid, levulinic acid, and glycerol.11, 12 These were demonstrated to extract lignin from woody biomass in high yield (up to 78% from poplar and 58% from Douglas fir). The resulting lignin product, DESL, has several distinctive characteristics including lower and narrowly distributed molecular weight, and the lack of ether linkages, representing a new type of lignin.<\/p>\n
So far, most lignins have been de\u2010polymerized into oligomers and monomers via hydrolytic depolymerization (using water) or reductive depolymerization (using hydrogen) in various solvents and catalysts. The most commonly used solvents include water, water\u2010ethanol co\u2010solvent, water\u2010ethanol\u2010formic acid, methanol, acetone, etc. Various homogeneous, heterogonous, metallic, commercial, and industrial catalysts have also been tested for the depolymerization of lignin. Recently, Li et al.13 demonstrated a new approach for lignin depolymerization and separation in the solid state, by using an acidic lithium bromide trihydrate (ALBTH) system under mild conditions (with 40\u2009mM HCl at 110\u2009\u00b0C). Woody and grass biomass were treated with ALBTH. In this process, the cellulose and hemicellulose were hydrolyzed and dissolved, while solid, pure depolymerized lignin was separated efficiently at high purity. The depolymerized lignin product showed low molecular weight with a minimal degree of condensation. Despite any specific advantages that the lignin depolymerization processes described above might have, they also have several drawbacks including operation at high\u2010temperatures and pressures, use of expensive solvents and \/ or catalysts that are not easily recoverable, as well as issues with scalability and industrial applicability. As a result, most depolymerization processes reported in the literature are likely to be associated with high capital and operating costs.9<\/p>\n
In contrast to the above approaches, this paper presents a simple, cost\u2010effective and environmentally friendly process for the depolymerization of HL using recoverable cooking liquor from the Kraft process (white liquor) at relatively moderate temperatures, pressures, and reaction times. Furthermore, in this paper, we demonstrated the scale\u2010up of this process to the 20 L reactor level. The process developed can easily be integrated into pulp mill operations, especially in the case of Kraft pulp mills that already have a lignin recovery system such as LignoForce\u2122 or LignoBoost\u2122. It is important to note, here, that hydrolysis lignins are less resistant to depolymerization than Kraft lignins because the former contain a significantly higher number of ether linkages (weak bonds) and a significantly lower number of C\u2010C bonds. This is because, during the bulk delignification stage of the Kraft pulping process, the content of \u03b2\u2010O\u20104 linkages in both the liquor\u2010phase (Kraft) and residual lignins in the pulp significantly decreases. The depolymerization process in the current study resembles the delignification of wood chips in the Kraft pulping process in which lignin is removed from the lignocellulosic matrix via lignin depolymerization. As is the case with wood chips, raw HL contains lignin mostly in its native form which is rich in \u03b2\u2010O\u20104 ether linkages.14 There are two main types of lignin reactions that occur during Kraft pulping under alkaline conditions. The first are lignin degradation reactions, which lead to lignin fragments of low average molecular weight (desirable reactions). The second are lignin condensation reactions leading to the formation of alkali\u2010stable linkages (undesirable reactions).15 Due to the high frequency of their occurrence in lignin, \u03b2\u2010O\u20134 and \u03b1\u2010O\u20134\u2010ether linkages together comprise the most abundant connections between lignin units (up to 80%).14 Two main possible chemical pathways were therefore proposed in an effort to elucidate the mechanism of delignification leading to lower molecular weight lignin \u2013 the first chemical pathway relates to phenolic lignin structures whereas the second relates to non\u2010phenolic lignin structures. These will be further discussed in the \u2018results and discussion\u2019 section of this paper as they relate to the results obtained in this work.<\/p>\n
Materials and methods<\/p>\n
Materials
\nThe hydrolysis lignin (HL) used in this study was a by\u2010product of the TMP\u2010Bio process developed by FPInnovations (Pointe Claire, Canada).10 It was derived from aspen wood and was not soluble in any solvent. Hydrolysis lignin is usually composed of 60\u201365\u2009wt% lignin, 20\u201325\u2009wt% cellulose and 5% hemicellulose. The TMP Bioprocess uses low\u2010pressure mechanical refining to disintegrate the biomass feedstock (e.g. aspen wood chips) into a fibrous mass, which is more amenable to enzymatic hydrolysis. The enzymatic hydrolysis results in two products: a sugar solution that is rich in glucose (C6) and xylose (C5) sugars, and HL. The elemental composition (on a dry basis) of HL is 49.25\u2009wt% carbon (C), 6.30\u2009wt% hydrogen (H), 0.39\u2009wt% nitrogen (N), and 44.05\u2009wt% oxygen (O) plus a negligible amount of ash. The average molecular weight of the soluble part of hydrolysis lignin after acetobromination of the original HL is believed to be around 8000\u2009g\/mol as analyzed by GPC\u2010UV. However, the average molecular weight of the whole HL is difficult to measure due to its lack of solubility in common organic solvents. Other chemicals used include: NaOH, Na2S, sulfuric acid (98% purity), acetone (99%), tetrahydrofuran (THF) (high\u2010performance liquid chromatography (HPLC) grade), butylated hydroxytoluene (BHT, 99%), glacial acetic acid (99.7%), acetyl bromide (99%), phosphoric acid (85%) and acetonitrile (HPLC grade), all ACS reagents used as received from Caledon (Georgetown, ON, Canada) without further purification.<\/p>\n
Methodology
\nIn this work, simulated white liquor was prepared by mixing 121\u2009mL of NaOH solution (at a concentration of 595\u2009g\/L as Na2O) with 224\u2009mL of Na2S solution (at a concentration of 156\u2009g\/L as Na2O) with distilled water added to the mixture to produce 1\u00a0L of solution. Titration (Brinkmann Titrino) of the solution was used to measure the effective alkali (EA) and sodium sulfide (Na2S) concentration of the solutions made. Table\u00a01 shows the adjusted EA and sulfide content of the white liquor made.<\/p>\n
The experiments for hydrolytic depolymerization of HL with white liquor were carried out in a batch 2\u2010L Parr Model 4843 reactor, equipped with a pressure gauge, thermocouple, stirrer, gas lines (in and out) and sampling line. In a typical run, 100\u2009g of HL, 100\u2009g of white liquor (1:1(w\/w)), and 500\u2009mL of deionized water were charged to the reactor. The reactor was then closed and tightened. To ensure the complete removal of any air or oxygen present, the reactor was purged two or three times with N2. Subsequently, the reactor was pressurized with N2 to a cold pressure of 2\u00a0bar and a leak test was performed. The reactor was then heated under stirring (50\u2009g) and allowed to run over a pre\u2010specified length of reaction time (e.g. 1\u2009h) after the reactor reached the set temperature (150, 170 or 190\u2009\u00b0C). Once the pre\u2010determined reaction time was reached, the reactor was immediately quenched with water to terminate further reactions. When the system nearly reached room temperature, all the reactor contents were collected and acidified using 1\u00a0mol\u2009L\u20131 H2SO4 to pH\u00a0=\u00a02, to precipitate out the depolymerized hydrolysis lignin (DHL). The liquid \/ solid mixture was then separated by filtration. Subsequently, the solid DHL cake was washed in three stages using: (a) 2\u00a0L of 0.4\u00a0N H2SO4, (b) 2\u00a0L of 0.01\u2009N H2SO4, and (c) 2\u00a0L of distilled water. All the washing filtrates were collected for further analysis. Each experiment was conducted three times to ensure that relative experimental errors are not more than 5\u201310%.<\/p>\n
31P NMR spectroscopic analysis of modified lignin product
\nThe hydroxyl group content of the lignin samples was measured using quantitative 31P NMR spectroscopy. The samples (30\u201340\u2009mg) were dissolved in 500\u2009\u03bcL of anhydrous pyridine and deuterated chloroform (1.6:1, v\/v). This was followed by the addition of 50\u2009\u03bcL of chromium (III) acetylacetonate (5.13\u2009mg\/mL in anhydrous pyridine and deuterated chloroform 1.6:1, v\/v), which was used as the relaxation agent and 27.5\u00a0\u03bcL of cyclohexanol (21.07\u2009mg\/mL in anhydrous pyridine and deuterated chloroform 1.6:1, v\/v), used as the internal standard. Finally, 100\u2009\u03bcL of 2\u2010chloro\u20104,4,5,5\u2010tetramethyl\u20101,3,2\u2010dioxaphospholane (TMDP) was added to the vial for phosphorylating the cyclohexanol and lignin hydroxyl groups. The solution was thoroughly mixed and transferred to a sealed 5\u2009mm NMR tube. All NMR experiments were carried out at 298\u2009K on a Varian Inova 500 NMR spectrometer operated at a frequency of 500.13\u2009MHz and equipped with a 5\u2009mm broadband inverse probe. 31P NMR spectra were recorded with 32\u2009768 data points and a spectral width of 60\u2009606.06\u2009Hz. A relaxation delay of 5\u00a0s was used and the number of scans was 512. All chemical shifts are reported relative to the product of TDMP with cyclohexanol, which produces a sharp signal at 145.15\u2009ppm referenced from the water signal (132.2\u00a0ppm) for the phosphitylating reagent.16<\/p>\n
Carboxylic acid analysis
\nThe concentration of carboxylic acids in the filtrate was measured by HPLC. For this purpose, an Agilent 1260 system consisting of a binary pump, an autosampler, a column oven, a chromatographic column and a ultraviolet (UV) detector was used and operated with Openlab operating system software. An Allure Organic column, 4.6\u2009mm\u2009\u00d7\u2009300\u2009mm, packed with 5\u00a0\u03bcm particles from Restek was used for the chromatographic separation. Samples were diluted using a phosphoric acid solution, pH\u00a02.6, and filtered through a 0.2\u00a0\u03bcm membrane syringe filter. Twenty microliter of solution were injected into the HPLC system after system calibration with suitable standards. The HPLC analysis was carried out under the following conditions: eluent A\u00a0=\u00a0H3PO4 (0.057%, pH\u00a02.6), eluent B\u00a0=\u00a0acetonitrile (ACN), mobile phase composition at 0\u20135\u00a0min: H3PO4, 99%: acetonitrile, 1%, mobile phase composition at 20\u2009min: H3PO4 85%: ACN 15%. Run time: 40\u2009min, pump flow rate: 1\u00a0mL\/min (120\u2013130\u2009bar), column temperature: 30\u2009\u00b0C, and UV detector wavelength: 220\u2009nm.<\/p>\n
Results and discussion
\nEffects of temperature and white liquor \/ hydrolysis lignin (WL\/HL) ratio on lignin MW and reaction yield<\/p>\n
To investigate the effects of temperature and WL\/HL ratio on HL depolymerization, a pre\u2010planned set of experiments was conducted. The experiments were designed as follows: At each mass ratio of WL\/HL (1:4, 1:2, 1:1, 2:1), three temperatures were investigated (150, 170, and 190\u2009\u00b0C). Table\u00a02 presents the experimental run number for each reaction that was conducted, and the associated conditions used.<\/p>\n
Figure\u00a01 shows the molecular weight distribution (GPC\u2010 UV) of DHL obtained from the experiments conducted at various reaction temperatures for 1\u00a0h. It is clearly shown here that an increase in the reaction temperature led to a decrease in weight\u2010average molecular weight (Mw) of DHL for all the ratios tested. For example, at a 2:1 ratio of WL\/HL, at 150\u2009\u00b0C, the Mw was 4518\u2009Da, as compared to 2686\u2009Da at 190\u2009\u00b0C. These results suggest that the cleavage of ether bonds in lignin was promoted at higher temperatures as a result of overcoming the activation energy barrier of these reactions. As also shown in Fig.\u00a01, the Mw decreased with increasing WL\/HL ratio. This is likely because of the increasing quantity of nucleophilic agents in the reaction mixture such as hydrosulfide (HS\u2212), mercaptide (CH3S\u2212), and hydroxide (OH\u2212) ions. As is well known in the Kraft pulping literature, lignin is likely to be depolymerized during this process through two possible mechanisms. In the case of lignin molecules in which phenolic groups are present, the depolymerization is expected to occur through the formation of a quinone methide intermediate while in the case of lignin molecules with no phenolic groups, the depolymerization is expected to occur through an oxirane intermediate.<\/p>\n
According to the first mechanism, the lignin degradation reactions start with ionization of the phenolic groups under alkaline conditions.17 The ionization then triggers the formation of the para\u2010quinone methide intermediate. The oxygen of the quinone group then draws the electron density to the double bond thus making the \u03b1\u2010carbon more positive. This, in turn, shifts the electron densities of the other bonds in this conjugated system. As HS\u2212 is a powerful nucleophilic agent, it attacks the quinone methide intermediate at the C\u2010\u03b1 position and adds itself to the quinone methide structure. This is followed by an intramolecular attack of the thiol group at the neighboring \u03b2\u2010carbon, which causes the formation of a thiiran intermediate. Finally, elimination of elemental sulfur (formation of polysulfide), associated with re\u2010aromatization, yields coniferyl\u2010type structures.18 This reaction scheme ultimately leads to break up of \u03b2\u2010ether bonds thereby depolymerizing lignin. According to the second mechanism, the reaction proceeds to fragmentation via oxirane intermediates formed by a neighboring group\u2010assisted mechanism.18 Subsequently, the epoxide formed is attacked by nucleophiles, such as HS\u2212 and HO\u2212, or by nucleophilic sites in carbohydrates.4 The latter attack can produce stable lignin carbohydrate complexes (LCCs) \u2013 this is one of the reasons for the incomplete removal of lignin from the fibers during the bulk delignification stage.<\/p>\n
During the above experiments, the white liquor was gradually enriched with an extremely complex mixture of dissolved lignin and carbohydrate degradation products. As shown in Fig.\u00a01, the Mw started to decrease dramatically when the WL\/HL ratio was adjusted to 1:1. This is probably because, at this ratio, there is sufficient alkalinity in the liquor to neutralize any acidic compounds that might be generated during the cooking process, thereby maintaining the pH in the optimum range for depolymerization reactions to occur. Furthermore, additional white liquor assured the presence of an excess of HO\u2010 and HS\u2212 ions, which are the active depolymerizing agents. It is possible that the depolymerization of HL might be occurring through two chemical pathways: the first pathway might be the cleavage of linkages between lignin and carbohydrates and \/ or carbohydrate peeling reactions, thereby reducing the size of lignin carbohydrate complexes (LCC) \u2013 this was confirmed by the decreasing carbohydrate content in the final DHL product as the WL\/HL ratio was increased. The second pathway could be lignin depolymerization. For WL\/HL ratios of 1:4, 1:2, and even 1:1, no significant drop in Mw was observed, suggesting a deficiency in OH\u2212 and HS\u2212 ions needed for the cleavage of intramolecular ether linkages in the lignin. At the 2:1 ratio, however, a sharp decrease in Mw was observed (see Fig.\u00a01).<\/p>\n
Moreover, during the cooking reaction, HO\u2212 ions were consumed in the neutralization of organic acids such as formic, glycolic, lactic and acetic acids being generated from carbohydrate degradation. As Fig.\u00a02 shows, these acids reduce the pH, which can, in turn, decrease lignin solubility and \/ or affect the reactivity of the active depolymerization agents. An excess of white liquor was therefore needed to keep lignin dissolved in the liquor to achieve a lower Mw. As is the case in Kraft pulping, it is important to ensure that the liquor has a pH value >\u200911, to avoid lignin precipitation and condensation reactions in the reaction solution. In contrast to soda pulping, in Kraft pulping, as a result of the presence of hydrosulfide ions the quinone methide intermediates are largely converted to benzyl thiols, thereby reducing the opportunity for condensation reactions. This might explain the reason for the absence of solid residue in our reaction products.19 Figure\u00a02 presents the pH before and after the reaction of HL with WL for all applied WL\/HL ratios. At low ratios (1:4 and 1:2), the HO\u2212 ions were totally consumed as the final pH was acidic. On the other hand, the final pH was alkaline at higher WL\/HL ratios. The pH difference before and after the reaction is an indirect indication of white liquor consumption during the reaction. Alkali is basically consumed in three different reactions: (a) hydrolysis and degradation of carbohydrates, (b) neutralization of organic carboxylic acids generated by degradation of carbohydrates, and (c) neutralization of the phenolic groups generated by lignin depolymerization.19<\/p>\n
Figure\u00a03 shows the effects of temperature and WL\/HL ratio on depolymerized lignin yield after being recovered by precipitating it from the reaction products through the addition of H2SO4 acid to pH\u00a0=\u00a02 and filtering the lignin slurry. As this figure shows, the yield decreases with increasing temperature and WL\/HL ratio. At a WL\/HL ratio\u2009\u2265\u20091:1, the depolymerized lignin yield was significantly reduced. At the WL\/HL ratio of 1:1, the yield declined from 63% at 150\u2009\u00b0C to 57% at 190\u2009\u00b0C, while at a WL\/HL ratio of 2:1, the yield declined from 59% at 150\u2009\u00b0C to 47% at 190\u2009\u00b0C. At low WL\/HL ratios (<1:1), no significant change in reaction yield was observed with increasing temperature. Furthermore, with increasing temperature and WL\/HL ratio, the lignin: carbohydrate ratio in the final product also increased, suggesting the depolymerization \/ degradation of the carbohydrate component of the hydrolysis lignin. Whereas, the lignin content in unwashed HL was initially 48\u2009wt%, it increased to 74\u2009wt% after reaction at 190\u2009\u00b0C at a WL\/HL mass ratio of 2:1. The decrease in reaction yield can be partly attributed to carbohydrates being hydrolyzed and \/ or decomposed by alkali during cooking. Based on our analysis of the filtrates, various soluble organic compounds were generated during the reaction. It appears from these analyses that carbohydrates in HL were hydrolyzed to soluble monomeric and oligomeric sugars, as well as oxidized to various saccharinic acids, simple organic acids and alcohols. The ash content of DHL was only 0.1\u20130.2\u00a0wt% (measured by ashing at 525\u2009\u00b0C).<\/p>\n
Table\u00a03 presents the lignin content (Klason and soluble lignin) as well as the carbohydrates content of the depolymerized hydrolysis lignins produced under different conditions. As seen in Table\u00a03, the carbohydrate content in the modified lignin product decreased at high temperatures and high WL\/HL ratios as compared to the control. However, there was practically no change at low WL\/HL ratios (1:4) and (1:2). This effect can be explained by the lack of sufficient WL required for lignin and carbohydrate depolymerization\/degradation. At higher WL\/HL ratios, especially at 2:1 at 150\u2009\u00b0C and 170\u2009\u00b0C, the carbohydrate content decreased to about 33% and 37% of the initial content, respectively. This is a result of carbohydrate degradation to soluble low molecular weight monomers, oligomers, and organic acids. While the corresponding decrease at 190\u2009\u00b0C was only 29%, it can perhaps be explained by the possible formation of alkali resistant LCC structures between lignin and carbohydrates as discussed above for the case of non\u2010phenolic lignin moieties.20 As a result of the carbohydrate content of hydrolysis lignin being degraded faster than the lignin content, the total % lignin (Klason and acid\u2010soluble) content in DHL increased with increasing temperature and WL\/HL ratio. At high WL\/HL ratios, the total lignin content ranged from 71 to 84% of the total mass of DHL, depending on the temperature used.<\/p>\n
Soluble lignin in the filtrate
\nThe soluble lignin in the acidic filtrate was determined using UV absorbance at 205\u2009nm using a hardwood lignin extinction coefficient of 110\u2009L\/g.cm.21 As previously reported in the literature, certain low MW lignins are soluble in acidic media (e.g. lignin model compound syringylglycerol\u2010\u03b2\u2010syringyl ether).22 Figure\u00a04 shows the concentration of soluble lignin in the acidic filtrate as a function of temperature and WL\/HL ratio. The concentration of soluble lignin clearly increased with increasing temperature and WL\/HL ratio. Lignin concentration in the filtrate dramatically increased at 190\u2009\u00b0C, with a soluble lignin concentration of about 3500\u2009ppm at a WL\/HL ratio of 1:4, as compared to 6000\u2009ppm at the ratio of 2:1. These results are consistent with the decreased yields of DHL obtained under these conditions.<\/p>\n
As mentioned above, at the end of the reaction, the products were acidified using 1\u00a0mol\u2009L\u20131 H2SO4 to precipitate DHL by adjusting the pH to 2. Liquid \/ solid separation was conducted by filtration to give a solid cake (DHL) and an acidic filtrate. The liquid (acidic filtrate) was collected and samples were submitted for carboxylic acid analysis using HPLC. During the cooking process, lignin is known to degrade into low molecular\u2010weight compounds that contain ionizable groups such as phenolate, catecholate and partly also carboxylate groups.23 The sodium salts of carboxylic acids mainly originate from carbohydrates and include, for instance, the salts of hydroxyacids such as gluco\u2010xyloisosaccharinic acid, lactic acid, and gluconic acid. Moreover, relatively large amounts of the sodium salts of formic acid, glycolic acid, and acetic acid are formed through fragmentation reactions of carbohydrates.18-22, 24 As is the case in the Kraft pulping process, at high temperatures and under alkaline conditions at the beginning of cooking, the acetyl groups in hardwood hemicelluloses (e.g. xylan) are hydrolyzed. Furthermore, in the earlier stages of cooking, the polysaccharide chains are peeled off directly from the reducing end groups (primary peeling). Because of the alkaline hydrolysis of glycosidic bonds that occurs at high temperatures, new groups are formed, following additional degradation (secondary peeling). As result, the cellulose yield is always reduced in cooking but to a lesser extent than hemicelluloses, which are degraded more extensively due to their low degree of polymerization and amorphous state. The peeling reactions are finally interrupted because of a termination reaction that converts the reducing end groups to stable carboxylic acid groups.3 The total amount of low\u2010molecular\u2010weight carboxylic acids (after cooking) was found to be around 10\u00a0wt% of the initial amount of HL. Four major carboxylic acids were detected: Glycolic, formic, lactic, and acetic acids. The carboxylic acid concentration was directly proportional to temperature and WL\/HL ratio as shown in Fig.\u00a05 for lactic acid. The composition of the alkaline reaction products is affected by many parameters such as temperature, pH, and reaction time. Whereas the formation of carboxylic acids proceeded more or less constantly during the cooking process, the most dominant acid was lactic acid with a concentration of 5500\u2009ppm for a reaction conducted at 190\u2009\u00b0C and a WL\/HL ratio of 2:1.f<\/p>\n
For other acids (glycolic, formic, and acetic), there was no clear trend that could be observed \u2013 the concentrations of these acids fluctuated between 1000 to 2500\u2009ppm at different WL\/HL ratios. It is worth noting here that, at low WL\/HL ratios (e.g. 1:4 and 1:2), the concentration of all acids did not change much with increasing temperature.
\nEffect of temperature and WL\/HL ratio on depolymerized lignin functional groups
\nThe development of nuclear magnetic resonance (NMR) techniques has significantly aided the understanding of lignin structure and physicochemical behavior. Phosphorus 31 is a nucleus that is 100% naturally abundant. Lignin derivatization with phosphorus\u2010containing reagents has grown dramatically to help us elucidate new analytical features in lignin.25 For example, quantitative 31P NMR has offered considerable details in identifying various aromatic groups bearing free phenolic hydroxyls, including p\u2010hydroxyphenyls, catechols, guaiacyl units, and phenols with carbon substituents at the C5 or C6 positions.26 In addition, the primary hydroxyl and carboxyl groups in lignin as well as the two diastereomeric forms of arylglycerol\u2010\u03b2\u2010aryl ether units (\u03b2\u2010O\u20104) present in DHL can also be determined using 31P NMR.27 A comprehensive analysis of hydroxyl groups in depolymerized lignin (ZHL_11) and their typical chemical shifts \/ integration ranges are summarized in Table\u00a04. Various factors make phosphorus an ideal sensor group for NMR studies of labile groups in lignin. These are: (1) Several types of organophosphorus compounds reveal signals within narrow ranges, characteristic of the oxidation state of the phosphorus nuclei; (2) The relationships have been identified between phosphorus chemical shifts and structures that, in some cases, even reveal stereochemical information.4<\/p>\n
Quantitative 31P NMR analysis was used to measure the hydroxyl group content of all DHL samples generated in this work. The results are shown in Table\u00a05, in which the hydroxyl group content is presented as a function of temperature, WL\/HL ratio, and Mw. As can be seen here, no clear trend can be observed at the WL\/HL ratios of 1:4 and 1:2. However, at higher WL\/HL ratios, the following observations can be made:<\/p>\n
Aliphatic hydroxyl groups : At higher WL\/HL ratios (e.g. 1:1 and 2:1), the aliphatic hydroxyl group content (\u03b4 150.4\u2013145.5\u00a0ppm) decreased for both ratios (see Figs\u00a06 and 7), as the temperatures increased. The same trend was also observed with Mw (the aliphatic hydroxyl group content declined, when the Mw was reduced) as previously reported by Argyropoulos et al.25<\/p>\n
Carboxylic acid groups : With increasing treatment temperature and WL\/HL ratio, a higher carboxyl group content was obtained. Figures\u00a06 and 7 show that, at high WL\/HL ratios (e.g. 2:1 and 1:1), the carboxyl group content increased as Mw decreased. For example, for Mw\u00a0=\u00a04518\u2009Da, the carboxyl group content was 0.18\u2009mmol\/g and it increased by about three times to 0.5\u2009mmol\/g as the Mw decreased to 2680\u2009Da. The explanation for this trend could be that HS\u2010 ions in the white liquor can degrade enol ether structures to produce carboxyl groups in lignin.18<\/p>\n
The phenolic hydroxyl group is one of the main functional groups of lignin, which affect its physical and chemical properties. The chemical reactivity of lignin in various modification processes is often influenced by its phenolic hydroxyl group content (e.g., in the reaction with formaldehyde for the production of lignin\u2010based adhesives).28 The two main types of phenolic groups that can be quantified by 31P NMR are non\u2010condensed and condensed phenolic hydroxyl groups.<\/p>\n
Non\u2010condensed phenolic hydroxyl groups (\u03b4 137\u2013143\u2009ppm shift): These appear to follow a similar trend to the carboxylic acid groups \u2013higher content was observed with decreasing molecular weight. Hence, it appears that, during the depolymerization process, low molecular weight DHL was produced as a result of cleavage of \u03b2\u2010O\u20104 and \u03b1\u2010O\u20104 linkages. As a result, more non\u2010condensed phenolic hydroxyl structures were generated.29 This is also confirmed in Figs\u00a06 and 7, where an increase in non\u2010condensed phenolic hydroxyls groups was observed with decreasing Mw for both WL\/HL ratios.<\/p>\n
Condensed phenolic hydroxyl groups : These groups slightly increased with decreasing Mw as seen in Figs\u00a06 and 7. This is likely to be due to repolymerization of phenolic fragments during the cooking process. An insignificant amount of condensed structures such as diphenylmethane, 4\u2010O\u20105′, and 5\u20135′ was found in depolymerized hydrolysis lignin.25<\/p>\n
The above results are in agreement with Potthast et al.29 who, in a series of sequential ultrafiltration experiments of Kraft lignins from black liquor, found that aliphatic\u2010OH group content decreased, and both carboxyl and phenolic\u2010OH group content increased with decreasing molecular weight.<\/p>\n
Hydrolysis lignin depolymerization in 20\u2010L reactor
\nBased on the results obtained from small\u2010scale experiments, an effort was made to scale\u2010up the process developed to the 20 L reactor level using a batch circulating reactor (see Fig.\u00a08). Steam was used to heat up the reactor through a jacket, and then cold water was introduced in the jacket to cool the contents after a designated reaction time. The conditions used, and the results obtained, are shown in Table\u00a06. The main differences between the experiments conducted at the 20 L level versus those conducted using the 2 L Parr reactor are the following:<\/p>\n
A double impeller agitation design was operated in the 2 L Parr reactor while a circulation mixing system was employed in the case of the 20 L reactor.
\nThe HL\/deionized water ratio in the Parr reactor was set to 1\/5 (w\/w), whereas in the circulating reactor it was 1\/8 (w\/w). Furthermore, the white liquor to HL ratio in the 20 L reactor was 3.2\/1. These high ratios were selected for the circulation reactor to insure continuous flow of the constituents through the reactor without clogging.
\nWhile the residence time in the PARR reactor was 1\u00a0h, the residence time in the circulating batch reactor was 2\u00a0h to guarantee good mixing, which can often prevent heat losses and improve reaction performance.<\/p>\n
The reaction rate and degree of mixing are usually closely related to each other. For a reaction to occur, molecules need to come into close contact with each other. Having a sufficient degree of mixing is important in terms of achieving the chemical kinetics needed to produce new species, especially in mass transfer\u2010controlled reaction systems.30 The results of scale\u2010up experiments (LS_23) were compared with the results obtained for experimental run ZHL_12, since both samples were produced at the same temperature (170\u2009\u00b0C). The Mw of the final product of experiment LS_23 was 2392\u2009Da, while that for ZHL_12 was 3200\u2009Da. The main reason for this is likely to be the longer reaction time and higher WL\/HL ratio used in the former experiment compared to the latter. These results are in agreement with previous studies.10 In particular, the high WL\/HL ratio maintains a high concentration of HS\u2010 and HO\u2010 (the active depolymerization reagents) and a high pH throughout the duration of the reaction. This leads to the cleavage of a higher number of ether bonds, which, in turn, results in a lower Mw.19<\/p>\n
In both above experiments, the HL starting material contained 54.5% lignin, and 27% carbohydrates. After the treatment in ZHL_12, the composition of the DHL changed to 81% lignin and 16.5% carbohydrates. A further increase in lignin content was achieved in LS_23, for which we obtained 84.2% lignin and 15.8% carbohydrates. However, the percentage yield for LS_23 was 40% whereas in ZHL_12 it was around 50% (see Table\u00a07). 31P NMR analysis was used to determine the main functional hydroxyl groups of the DHL for ZHL_12 and LS_23. The contents of carboxylic acid groups, aliphatic hydroxyl groups, phenolic hydroxyl groups, and condensed units, are presented in Table\u00a07. As expected, the lower molecular weight of LS_23 resulted in higher free phenolic and carboxyl group content compared to ZHL_12. Given the high content of free phenolic groups in DHL produced from both processes, they are projected to be quite suitable for use in the manufacture of phenolic resins. The same applies to the use of these DHLs in the manufacture of polyurethane foams, where high primary aliphatic hydroxyl group content is required.<\/p>\n
Conclusions
\nIn this work, we developed a novel, environmentally friendly approach to produce DHL using recoverable WL from the Kraft recovery cycle. The weight\u2010average molecular weight (Mw) of the lignin was significantly reduced from non\u2010measurable levels to 2600\u2009Da at 190\u2009\u00b0C for 1\u00a0h at a WL\/HL mass ratio of 2:1. Moreover, a significant increase in non\u2010condensed phenolic hydroxyl and carboxylic acid group contents was observed with decreasing Mw. We found that effective hydrolysis lignin depolymerization can be achieved at WL\/HL ratio\u2009\u2265\u20091:1 \u2013 below this ratio, minimal lignin depolymerization was observed. The approach developed could be significantly more cost\u2010effective if integrated into Kraft pulp mill operations where fresh WL can be utilized with the residual WL recovered using the existing chemical recovery system. Process scale\u2010up was demonstrated using a 20 L circulating reactor, for which the Mw of the DHL produced after treatment at 170\u2009\u00b0C for 2\u00a0h was 2400\u2009Da. Further investigation is needed to understand better the effect of mixing hydrodynamics on lignin depolymerization.<\/p>\n","protected":false},"excerpt":{"rendered":"
Hydrolysis lignin (HL) refers to a lignin\u2010rich residue obtained after the enzymatic hydrolysis of biomass. It is recalcitrant, heterogeneous, insoluble in most common solvents, and less reactive than other lignins. To enhance the reactivity of HL, a novel environmentally friendly depolymerization approach was demonstrated to produce depolymerized hydrolysis lignin (DHL) using Kraft cooking liquor, white […]<\/p>\n","protected":false},"author":59,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_seopress_robots_primary_cat":"","nova_meta_subtitle":"","footnotes":""},"categories":[5572],"tags":[5838,5842,6162,11828],"supplier":[16614,2139,16641],"class_list":["post-71357","post","type-post","status-publish","format-standard","hentry","category-bio-based","tag-bioeconomy","tag-biomass","tag-cellulose","tag-lignin","supplier-fpinnovations-inc","supplier-university-of-western-ontario","supplier-zhengzhou-university"],"_links":{"self":[{"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/posts\/71357","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/users\/59"}],"replies":[{"embeddable":true,"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/comments?post=71357"}],"version-history":[{"count":0,"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/posts\/71357\/revisions"}],"wp:attachment":[{"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/media?parent=71357"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/categories?post=71357"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/tags?post=71357"},{"taxonomy":"supplier","embeddable":true,"href":"https:\/\/renewable-carbon.eu\/news\/wp-json\/wp\/v2\/supplier?post=71357"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}