Valorization of Tops and Branches to Textile Fibers and Biofuel: Value Chain Explored Experimentally; Environmental Sustainability Evaluated by Life Cycle Assessment

: To make biorefining more environmentally sustainable, preferably residues from forestry should be used and more than one fraction should be upgraded. A third of raw materials from forestry � tops and branches (T&B) � are either left in the forests or collected and incinerated to a low value. Herein, we apply a fast fractionation to valorize two of the fractions of this forestry residue. The cellulose is converted to textile fibers and all the lignin to hydrocarbons. The environmental sustainability of the novel value chain was studied by life cycle assessment (LCA), and benefits were found in four out of five impact categories. These are important steps to increase fiber production without affecting environmental impact, making biorefining competitive.


■ INTRODUCTION
−3 At the same time, our society is already struggling with climate change, land use change, and water scarcity. 4,5How to meet this 37.5% increase in demand due to population increase without negatively affecting environmental footprint categories is perplexing.Strategies to mitigate this comprise the use of residues that today are burnt or left to decay and to substitute products that have a negative impact on the environment such as cotton, fossil fuels, and even firstgeneration biofuels.
Forestry generates one-third timber, one-third pulp wood, and one-third tops and branches (T&B); 6 the latter stem from both thinnings and harvest.To some extent, T&B should be left in the forests 7 for recarbonization.However, the residue has a negative impact on regrowth. 8Furthermore, decay leads to greenhouse gas emissions, both carbon dioxide and methane. 9Because of the low revenue of incineration and high cost of collecting and transporting T&B, forestry owners let a substantial amount of residues decay in the forests.Thus, it would be worthwhile to find higher-value applications for this low-value residue.
Cotton production (26 Mton/year) requires 34.5 million hectares of productive agricultural land and 79 billion tons of water. 10,11To grasp these numbers, the depletion of the Aral Sea within a few decades is an illustrative yet tragic example of the water demand for cotton production.
Pulp mills currently discard T&B.One reason is the unpurity of the residue from soil in the ground.Another reason is that this wood gives weaker fibers than pulp wood, important for paper and packaging applications.−17 Central to these methodologies is the valorization of the lignin fraction in the design state of fractionation.These emerging technologies, for instance, reductive catalytic fractionation (RCF), have successfully been applied to hardwoods where half of the lignin fraction has been depolymerized to valuable monophenolic compounds.The theoretical maximum yield of monophenols is correlated to the abundance of β-ether bonds (yield of monomers = (β-ether bonds) 2 ). 18Certain hardwood species, such as birch, contain up to 70% of this motif.RCF performed on such wood species can give up to 50% yield of monophenols, i.e. the theoretical maximum yield. 19,20Soft-wood, with a lower β-ether bond content (35−40%), gives theoretical maximum yields below 15%.Thus, these wood species have received less attention from the community.There are some pioneering researchers who have taken up the challenge: Barta and co-workers reported a diol-assisted fractionation (DAF) of pine to obtain 9% monophenols; 21 Sels and co-workers reported an RCF of pine employing Pd/C to yield 16.7% of monophenols; 22 the group of Wang, Fang, and Song used Ru-based catalysis on softwoods and reported similar yields; 23 and D'Angelo et al. also performed a zeolite based stabilization and achieved 10 wt % monophenols. 24espite these advances, the 10−17% yield of monophenols corresponds to below 5 wt % yield from biomass, and further research is necessary.One strategy is to hydroprocess lignin to yield hydrocarbons.Even though there are numerous reports on the hydrotreatment of lignin, most studies report uncomplete hydrodeoxygenation, if not model compounds of lignin are used. 25,26In fact, very few studies have disclosed efficient hydrodeoxygenation of lignin using realistic reaction conditions. 27−30 However, even if full conversion of lignin to biofuel is realized, the revenue from only lignin would not bear the cost of the valorization.
From economic and environmental sustainability points of view, the carbohydrate fraction, especially cellulose, needs a high-value application in biorefinery.Beyond enzymatic digestion, there is only one report.Sixta and co-workers employed sulfur dioxide promoted fractionation to yield dissolving pulp from soft wood; however, lignin was not valorized.Other reports have used unbleached softwood pulps. 31We recently published a holistic valorization of poplar to dissolving pulp and a biofuel. 32The value chain from this hardwood was evaluated by LCA and showed little improvement in terms of land use and climate change because plantations were used.Very recently, we reported valorization of beetle-infected spruce, 30 where environmental sustainability benefits of using a forestry residue were proven.This encouraged us to study T&B with an even higher potential.
We herein report a holistic valorization of a softwood residue to yield both textile fiber and biofuels; the environmental sustainability of the novel value chain was studied by LCA.
■ EXPERIMENTAL SECTION Materials.The biomass feedstock spruce (Picea abies) tops and branches were provided by Skogforsk, Sweden.All chemicals were purchased from Fischer chemicals, CCS Healthcare AB Sweden, Sigma-Aldrich, Honeywell, and VWR chemicals and used as received.
Feed Stock Composition Analysis.Spruce tops and branches were debarked and cut into small pieces with a size less than 5 cm.The moisture content in the biomass was reduced by drying at 60 °C for 3 days.The biomass compositions were measured by a modified procedure based on a standard protocol by the National Renewable Energy Laboratory (NREL).Sugar analysis was quantified by 1 H NMR recorded on a Bruker Avance (400 MHz) as solution in D 2 O.Chemical shifts are expressed in parts per million (ppm, δ).GC−MS and GC-FID analyses were conducted with a QP2020 system (Shimadzu, Japan) equipped with two parallel HP-5MS columns (30 m × 0.25 mm × 0.25 μm).GPC analysis was performed on a Prominence-i, LC-2030C system (Shimadzu, Japan) equipped with a UV detector a 280 nm (SI, S3 and S7).
Fast Fractionation Procedure.Tops and branches (50 g, mix size <5 cm) were loaded into stainless steel reactor (volume capacity 250 mL).The solvent system of 65% EtOH in water (400 mL) was added followed by the HCl solution (0.35 M, 6.25 mL).The reaction was heated with varied times and temperatures to optimize pulp quality with stirring at 300 rpm.After completion, the reaction was cooled down to room temperature, and then the solid residue was filtrated and washed with water (200 mL) and EtOAc (200 mL).The water phase was extracted with EtOAc (2 × 100 mL) and kept for carbohydrate analysis.The combined organic phase was dried in vacuo and kept for lignin analysis.The solid residue was placed into a 2 L beaker with 1 L of water and then disintegrated into a homogeneous pulp by using a T-25 digital ULTRA-TURRAX Homogenizer with 17.0 × 1000 rpm for 10 min.The solvent was drained, and the pulp was oven-dried at 60 °C for 12 h to yield a brown pulp (SI, S3).
Bleaching Procedure.Brown pulp (10 g) was dissolved in 200 mL of solvent system containing 1.7% NaClO 2 solution and 2.7% NaOH in 7.5% acetic acid buffer at a 1:1 ratio.The solution was heated at 85 °C for 2 h then the solvent was replaced and the bleaching was continued for 1h.After completion, the solution was cooled down to room temperature and white pulp was filtrated and washed with water (3 × 500 mL).The obtained white pulp was washed with 2 M NaOH solution at 60 °C for 2 h and washed with 0.25 wt % EDTA solution at 60 °C to reduce PDI and metal content, respectively.After washing with water, the white pulp was dried at room temperature for 2 days and stored in dry conditions for further analysis (SI, S5 and S6).
Transformation of Lignin Oil.Lignin oil obtained from fast fractionation (100 mg) was placed in a 50 mL round-bottomed flask equipped with a magnetic stirrer, and 407 mg of oleic acid was added.The mixture was heated to 120 °C while mixing.A homogeneous catalyst (4-methylpyridine, 155 mg) and acetic anhydride (157 mg) were added to the reaction mixture.The temperature of the mixture was then raised to 180 °C, and after 30 min, pressure was lowered and kept there for 2 h.Esterified lignin oil (563 mg) was dissolved in 3 mL of carrier liquid (pentane) and placed in a stainless-steel reactor.Commercially available Ni−Mo catalyst (30 mg) was added, the valve was closed, and the atmosphere was changed first to nitrogen and then pressurized with 25 bar of hydrogen.The reactor was placed in a sand bath at 360 °C for 24 h.The reactor was then cooled, and the pressure was carefully released.The reaction mixture was collected and dried over anhydrous Na 2 SO 4 .The mixture was filtered to give a yellow oil (364 mg).The resulting oil was analyzed by elemental analysis, stimulated distillation, and 2D GC (SI, S9−S12).
Viscose Fiber Spinning.The obtained dissolving pulp (viscosity 560 mL/g, 25 g) was processed with a standard viscose dope preparation (SI, S13).A wet spinning process of viscose dope was performed on Aditya Birla Spinning pilot.The viscose dope was collected in a syringe pump and extruded through a 40-hole spinneret (80 μm in diameter for each).The fiber filaments were spun at 22 °C into the coagulation bath containing 10 g/L zinc sulfate, 110 g/L of sulfuric acid (H 2 SO 4 ), and 310 g/L of sodium sulfate (Na 2 SO 4 ).The coagulation bath temperature was fixed at 48 °C.The extrusion speed was fixed at 2.6 mL min −1 with a godet speed of 12 for the first godets and 18 m min −1 for the last godet.
LCA Studies.A comparative LCA was performed according to the ISO 14040/44 standard comprising goal and scope, inventory analysis, impact assessment, and interpretation.The goal and scope of the study were to compare the environmental sustainability of viscose fiber produced from tops and branches to the conventional production of cotton fiber.A cradle to gate study was performed using 1 kg of textile fiber as a functional unit.Multifunctionality was handled by substitution.Inventory analysis was built up by combining primary data for fast fractionation and conversion to viscose fiber and hydrocarbons and secondary data from Ecoinvent v3.8.For the pulping, primary data obtained from the lab were used.For the hydroprocessing of lignin oil for biofuel production, primary data were used for the yield.For the cotton fiber at ginning stage, agriculture, harvest, separation of cotton lint, and ginning were included as the inventory was built up using the Ecoinvent v3.8 database; a global average was used.A full description of the data used can be found in the SI (S14.2).A large set of 15 impact categories was analyzed in the study in a way that avoids burden shifting phenomena.The results of five impact categories are presented and discussed in the text (climate change, land use, water scarcity, resource depletion (fossil and minerals and metals); for a complete overview, please refer to the Supporting Information.The EF 3.0 version 1.01 (2019) method was chosen for its comprehensiveness and focus on the European scenario.For completeness, a sensitivity analysis was conducted using the ReCiPe 2016 midpoint Hierarchist perspective method to assess the stability of the results; see SI (S14.4).Data processing was carried out using the LCA software SimaPro (9.2.0.2).
■ RESULTS AND DISCUSSION Raw Material Characterization.The composition of wood part from tops and branches was analyzed (see SI, S2). 33,34Major components comprised glucan (43.5), xylan (4.9), and lignin (29.5) and correlated to previous studies. 35,36ast Fractionation.To obtain dissolving grade pulp from spruce, the fractionation conditions were optimized. 35ractionation experiments were performed in an autoclave reactor using 50 g of wood and 400 mL solvent.After completion of the reaction, pulp was filtered, ethanol was evaporated, and organics were extracted with ethyl acetate.Poor results were obtained when fractionations were performed in the absence of acid even at 200 °C.Addition of 1% of distillable and thus recyclable HCl in 1:1 ethanol/ water solvent mixture gave good delignification of the pulp.The delignification is rapid during the first 2 h and reaches 68%; after this, the kinetics of delignification is considerably slower (Figure 1).Running the reaction for 2 additional hours only led to an increase from 68 to 79% delignification (SI, S4).A shorter reaction time has advantages for the pulp, preventing degradation leading to low viscosity.The viscosity, which is correlated to cellulose degradation, decreases from 785 mL/g (1 h) to below 200 mL/g (3 h).A viscosity above 450 mL/g is required for regeneration to yield textile fibers.The fractionation performed for 1.5 h gives a pulp within specification, i.e., a lignin fraction of 58%.Running the reaction shorter time gives a lower yield of lignin where the lignin remaining in pulp will be bleached away and discarded.The lignin from a fast fractionation is less condensed, and we were able to isolate 8 wt % monophenols from the original lignin, which is noteworthy as it corresponds to 74% of the theoretical maximum yield.The resulting pulp was bleached and washed with 2 M NaOH and EDTA solution to meet the dissolving grade pulp specifications (brightness > 90%; PDI ≤ 10; metals ≤ 50 ppm) (Table 1; see SI, S5) 37,38 Spinning of Textile Fibers.The generated pulp was dissolved using a viscose process and successfully spun to textile fibers.The viscose dope preparation was performed using the same conditions as those for commercial spruce dissolving pulp.The resulting dope had a similar ball-fall time (75.7 s) as commercial pulp.Viscose fibers were successfully spun at a pilot scale facility.In this case, no adjustments or optimization of the dope preparation or the spinning conditions for the new pulp was made.
To assess the mechanical properties of the viscose fibers, the tenacity and elongation (Table 2) of the individual viscose filaments were measured by using an Instron 5944 mechanical testing instrument (Instron, U.S.A.) equipped with a 100 N load cell in tensile mode.To calculate the tenacity, the linear density was determined by multiplying the cross-sectional area of the individual filaments with their density (=1.52 g/mL).The diameter (Table 2) was estimated from an optical microscope image analysis.Based on the literature, cotton derived viscose fibers can have a tenacity as low as 12 cN/ tex, 38,39 whereas commercial viscose fibers exhibit a tenacity of 23.9 cN/tex 38 and viscose fibers from wood dissolving pulp exhibit a tenacity in the range of 15−19 cN/tex. 38In this study, the individual filaments derived from spruce biomass waste dissolving pulp had an average tenacity of 4.84 cN/tex, whereas the conventional spruce dissolving pulp had an average tenacity of 6.37 cN/tex.The viscose filaments derived from T&B dissolving pulp had higher average linear density and lower tenacity than conventional spruce dissolving pulp, but they were in the same order of magnitude.The average elongation at break values were also higher for conventional dissolving pulp compared to the spruce waste dissolving pulp but not statistically different.The small statistical difference in the mechanical properties of filaments derived from spruce waste dissolving pulp and conventional spruce dissolving pulp paves the way to valorize waste biomass and upcycle it into textile fibers.Neither the viscose dope preparation nor the fiber spinning conditions were optimized for the new type of pulp, which would account for the differences compared to fibers from conventional pulp.Improving these conditions is likely to further enhance the quality of the filaments.
Lignin Oil Valorization.The organic fraction comprising lignin oil (18.5 wt %) was isolated by extraction and analyzed by gel-permeation chromatography (GPC) and gas chromatography (GC).By GC, monophenols were identified (SI, S7).Derivatives with both propyl and ethyl side chains were identified in a total yield of 2.4 wt % from wood.By GPC, the molecular weight of the polymer was determined to be below 6770 Da.Given the state-of-art, vide supra; to increase the yield of the valuable fraction from lignin, another strategy is required for softwoods.We proposed that hydrotreatment of the lignin fraction could effectively achieve two objectives: lignin depolymerization and hydrodeoxygenation, resulting in the production of valuable hydrocarbons suitable for biofuel applications.One of the primary challenges in hydrotreating lignin is its low solubility in carrier liquids.By utilization of carrier liquids, exothermic reactions leading to coke formation can be minimized.Success in solubilizing lignin in a carrier liquid enables using the existing refinery infrastructure.Recently, we have published a method in which kraft lignin is esterified using tall oil fatty acid, enabling the solubilization of lignin in hydrocarbon-based carrier liquids. 28hus, an esterification process by utilizing oleic acid, which is the primary constituent of tall oil derived from forestry, was envisioned.The esterification reaction was facilitated by the addition of acetic anhydride using distillable and recyclable 4methylpyridine as organocatalyst.The reaction was conducted at 180 °C for 2 h, resulting in the formation of a mixture of esterified lignin and fatty acid derivatives.The esterification process proved successful even considering that the lignin used in this study possessed a higher molecular weight (>6 kDa) compared to previous studies that employed kraft lignin (3−4 kDa). 28The successful esterification of lignin was observed through GPC (gel permeation chromatography) analysis (Figure 2); full solubility in pentane.The GPC analysis of the esterified lignin displayed a shift in the signal toward higher molecular weight ranges (6700−6800 Da).This indicates the successful esterification of lignin with an increase in its molecular weight.It is noteworthy that acetic acid, a byproduct of the reaction, was continuously distilled and could be recycled for the regeneration of the anhydride.Additionally, the organocatalyst methylpyridine employed could be distilled and reused (Figure 3).HDO of esterified lignin was performed at 360 °C and 25 bar initial hydrogen pressure in a batch setup using a commercial NiMo catalyst and pentane as inert carrier liquid (SI, S9).Commercial NiMo catalysts are recyclable after run times of 3−6 years (up to 97%), important for sustainability. 40 esterified mixture was fully deoxygenated to yield liquid hydrocarbons in 72 wt % yield.By gravimetric analysis of the hydrotreated mixture, the yield of hydrocarbons from lignin was determined to be 9 wt %, which corresponds to 50 wt % yield based on the initial lignin loading before its esterification.
The composition of the hydrotreated product mixture was analyzed using two-dimensional gas chromatography (2D GC).The product distribution disclosed by 2D GC confirmed that the primary component in the product mixture was linear alkanes derived from fatty acids (Figure 4, SI, Table S11).Additionally, cycloalkanes and alkenes were observed, which could be attributed to lignin-derived products.Notably, the analysis revealed the presence of deoxygenated aromatic compounds comprising 4 wt % of the hydrotreated oil (SI, S12).The presence of these deoxygenated aromatics holds significant value, as they can considerably enhance the quality of the resulting blend, making it a desirable option for utilization as a biofuel.
Furthermore, successful hydrodeoxygenation was confirmed by elemental analysis.A significant decrease in the oxygen content was observed compared to the organosolv lignin.This transformation resulted in a change in the C/H/O ratio, shifting from 65:6:29 (lignin oil) to 84:16:0.1 (hydrogenated oil).To assess the energy potential, the higher heating value (HHV) of the hydrogenated oil was calculated based on the elemental composition.HHV was estimated to be within the range of 48−52 MJ•kg −1 , comparable to commercially available fuels (SI, Tables S8 and S9).The boiling point profile analysis of the hydrogenated oil revealed that the majority of the fuel (approximately 62%) falls within the diesel boiling point range of 180−360 °C (SI, Table S10, and Figure S5).This indicates that the hydrogenated oil possesses desirable properties for use as a diesel fuel substitute.
Mass Balance Calculation.The mass balance of the value chain from T&B to textile fibers and biofuel was evaluated by combining the results obtained from fast fractionation at 200 °C for 1.5 h and hydrotreatment of lignin oil (Figure 5).The fast fractionation yields carbohydrate-rich lignocellulose, which contains 82.0% glucan and 2.7% xylan, in 42 wt % from biomass.After bleaching, the remaining lignin and hemicellulose were removed to yield dissolving grade pulp in 38.5 wt % from biomass.The lignin oil from fast fractionation contains monophenolics of 2.4 wt %; hydrotreatment of the lignin oil generates hydrocarbons in 9 wt % yield from biomass.During pulping, hemicellulose was hydrolyzed to yield monosaccharides and their derivatives in 8.7 w% in water phase (see SI, S3.2).Overall, 83 wt % of cellulose in spruce wood fraction was processed viscose fiber, and 50 wt % of available lignin was converted into biofuel.The hydrolyzed  monosaccharides are used for bioethanol production and as an internal makeup solvent for the fast fractionation. 41,42CA Studies.To assess the environmental sustainability of the value chain from T&B to textile fiber and biofuel, a cradleto-gate LCA focused on footprint categories was performed (Figure 6).The textile fiber was benchmarked to cotton fiber; thus, 1 kg cellulose-based textile fiber was chosen as a functional unit.Viscose fiber can substitute cotton fiber in certain applications 43 where the predicted increase in the demand of textile fiber cannot be sustainably supplied by cotton. 31Primary data from experiments were used for pulping and hydrodeoxygenation, and secondary data were used for pretreatment steps (collection, transportation, debarking− chipping), bleaching, and viscose spinning.Byproducts such as reject fiber, bark, and biofuel were treated by substitution approach crediting incineration to produce heat and power and transportation fuels.The study was carried out at a preliminary stage of process development; therefore, the data used cannot   all be primary and derived from an industrial scale.In particular, some key process parameters, i.e., conversion efficiencies, are derived from laboratory measurements (other assumptions in the inventory construction are discussed in S14.2).To assess the impact of the quality of these data, a sensitivity analysis was performed to analyze a worst case scenario for the pulping process (+10% of input data).The results of this analysis, available in the Supporting Information (S14.4),show the stability of the conclusions even when this parameter is varied.
The developed value chain disclosed a lower burden in four out of five impact categories compared to cotton.As mentioned, our previous study using poplar plantations also showed lower burdens in water use and fossil resource depletion.This value chain additionally shows improvements in climate change and land use due to the fact that a forestry residue is used.It should be noted that we have not modeled additional benefits in climate change from avoiding decay of residue.The poor performance in resource depletion, minerals, and metals depends on the viscose process, emphasizing the need to further improve dissolution processes to produce textile fibers.

■ CONCLUSIONS
The developed fast fractionation of tops and branches, a forestry residue, can contribute to meet a future demand in textiles without harvesting more trees or increasing land use by cotton plantation.Key findings and thus the novelty of this report that enable this value chain comprise both the fast fractionation of softwood residue and the upgrading of both the cellulose to produce textile grade pulp and the whole lignin fraction to biofuel by first esterification and then HDO.Whereas previous studies have employed expensive catalysts and hydrogen gas, i.e., below 5 wt % of lignin from biomass has been transformed to monophenols, this study shows that a similar yield of monophenols could be obtained without a catalyst.However, by esterification of lignin followed by HDO, the whole lignin fraction was transformed to hydrocarbons, i.e., biofuel mostly in the diesel fuel range.Taking into account that tops and branches usually contain more ashes than wood, this fast fractionation approach has advantages compared to employing RCF where precious metals would be contaminated.The LCA evaluation of the whole process has shown a low environmental impact in four of five footprint categories.This study makes progress in the fractionation of softwoods with respect to both the carbohydrate fraction and the lignin fraction.We hope that this study will inspire other researchers to develop value chains from residues.
■ ASSOCIATED CONTENT * sı Supporting Information

Figure 4 .
Figure 4. Two-dimensional gas chromatogram of the product mixture.

Figure 5 .
Figure 5. Overall mass balance of viscose fiber production from spruce tops and brunches.

Figure 6 .
Figure 6.Results from impact assessment: (a) comparison of viscose fiber from T&B compared to cotton and (b) contribution analysis of viscose fiber from T&B.

Table 1 .
Obtained Dissolving Grade Pulp Properties