Solvolytic Liquefaction of Bark: Understanding the Role of Polyhydric

Huntsman Advanced Technology Center, Huntsman International LLC, 8600 Gosling Road, The Woodlands, Texas 77381, United States. ACS Sustainable Chem. E...
1 downloads 16 Views 2MB Size
Research Article pubs.acs.org/journal/ascecg

Solvolytic Liquefaction of Bark: Understanding the Role of Polyhydric Alcohols and Organic Solvents on Polyol Characteristics Jason D’Souza,† Song Zhi Wong,‡ Rafael Camargo,§ and Ning Yan*,† †

Faculty of Forestry, University of Toronto, 33 Willcocks Street, Toronto, M5S3B3 Ontario, Canada Faculty of Applied Science and Engineering, Department of Chemical Engineering, University of Toronto, 200 College Street, M5S 3E5 Ontario, Canada § Huntsman Advanced Technology Center, Huntsman International LLC, 8600 Gosling Road, The Woodlands, Texas 77381, United States ‡

ABSTRACT: Bark was liquefied in polyhydric alcohols of various functionality, equivalent weight, and hydroxyl type, and organic solvents of varying polarity to determine how these features impact liquefaction behavior and polyol characteristics. It was found that the liquefaction yield was highly tunable with the use of polyhydric alcohols with primary hydroxyl groups, with low equivalent weight alcohols providing the highest liquefaction yield (59.3%). This showed that the highly polar hydroxyls (primary) and short chains created a highly protic solvent that improved conversion and protected the biopolymers from degradation. This was corroborated by 1H NMR analysis that indicated a greater amount of sugar degradation products were observed when polyhydric alcohols with secondary hydroxyl groups were used. Regarding organic solvents, ketonic solvents showed the greatest increase in the liquefaction yield. The composition and carbon content analysis of the residues suggested that the highly polar carbonyl group of ketonic solvents like acetyl acetone and cyclohexanone may have hindered condensation side reactions. These results have shown that selection of polyhydric alcohols and organic cosolvents can be quite impactful on the liquefaction yield and the polyol characteristics. KEYWORDS: Biomass, Bark, Liquefaction, Polyol



tree species studied.8 A high extractives content is desirable since many of these are polyphenolic in nature and tend to have a high level of hydroxyl functionality. The ease of extraction of these compounds compared to recalcitrant biopolymers like lignin and crystalline cellulose also makes them attractive. Moreover, their aromaticity can potentially impart thermal stability to their resultant PUFs.6,9 Liquefaction (acid-catalyzed glycolysis)10 has proven to be a versatile technique at digesting biomass into a low viscosity liquid polyol from a wide variety of biomass materials including bamboo,11 lignin,12 wood meal,13,14 soybean15 and wheat16 straws, sugar cane bagasse,17 and bark.4,5,18 The liquefaction of bark has previously been done at 1504,18 and at 200 °C.5 At such high temperatures, extensive degradation can occur, especially to sugars.19 Those results demonstrate the need for liquefaction under milder conditions. In our previous work, bark was liquefied at 90 and 130 °C. The polyol produced from the liquefaction of bark at 130 °C produced foams with compression strength comparable to control foams. But, the

INTRODUCTION Bark is a renewable biomass that has demonstrated a large potential as a raw material for synthesizing biobased phenolformaldehyde adhesives1 and polyols for making polyurethanes foams (PUFs).2−5 PUFs are known most notably for their versatility that enables their use in a variety of applications including cushioning and sound absorption in the automotive industry and as insulation in appliances, homes, and commercial buildings. PUFs are synthesized through an addition reaction of an isocyanate with an active hydrogen group, typically a hydroxyl or water. Polyether or polyester polymers with hydroxyl end groups are commonly used as commercial polyols. Many of these polyols have polymer backbones that lack rigidity and have poor temperature stability. In contrast, polyols derived from lignocellulosic biomass contain rigid and temperature stable aromatic polymer structures.6,7 Bark is currently an underutilized source of biomass available in large volumes suitable for industrial applications. At present, the most common usages of this low-value mill residue are as boiler fuel or garden mulch. In this study, bark from lodgepole pine (Pinus contorta) was used because it has the highest fraction of alkaline and benzene extracts among 25 different © XXXX American Chemical Society

Received: August 20, 2015 Revised: December 7, 2015

A

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Scheme 1. Condensation Reactions Involving Hydroxyl Rich Bark-Biopolymers and Polyhydric Alcohol Solvent

dipropylene glycol.14 This identifies molecular weight of the polyhydric alcohol as a factor, but no studies have investigated the impact of alcohol type or functionality. Liquefaction behavior could also be improved by the use of an organic cosolvent. Previous attempts used low-boiling solvents28 or simple alcohols like octanol.29 The use of a cosolvent is advantageous since it offers the possibility to tune the liquefaction to be more selective. Hansen et al. demonstrated that solvents with a similar solubility parameter to the biomass component will improve solubility and extraction; therefore, a lignin fragment will preferentially be extracted by hydrophobic solvents.30 Second, the solvation of reactive groups or radicals can also slow down or prevent unwanted side reactions like charring or precipitation due to extensive recondensation reactions.31 Lastly, solvents can alter the reaction kinetics through improved acid dissociation of the acid catalyst. For example, carboxylic acids in water with a relative permittivity (∈r) of 78.3 have a Ka value 106 times greater than if they were dissolved in ethanol (∈r = 24.6).32 Unfortunately, selecting a solvent based on polarity is difficult because many measures of polarity are required to be fed into solvent models.33 Some of these polarity measurements are temperature dependent, so the conditions of the liquefaction at 130 °C would make some of these polarity measurements irrelevant. This work will identify which solvent parameters under the conditions of a liquefaction reaction are most relevant and can provide a guideline for future studies. In summary, the goals of this study are aimed at understanding how solvents impact the liquefaction behavior

presence of degradation products/low functionality compounds in the bark-based polyols may have negatively affected both the closed-cell content and network formation.20 Some of these compounds may have been formed via condensation reactions, like those shown in Scheme 1. Condensation reactions like (i) and (ii) involve small molecular weight alcohols that react to improve the solubility of typically high-molecular weight, insoluble, bark biopolymers. However, bark biopolymers can also undergo condensation reactions among themselves and thereby reduce functionality or form very high molecular weight compounds that sediment into a residue. If the role of the polyhydric alcohol solvent on the extent of condensation reactions can be better understood, the liquefaction yield and the quality of the polyols can be improved. It has been shown previously in the literature how the type of biomass21 or species of bark, 22,23 acid type and acid content,24,25 temperature,16,26 biomass−solvent ratio,27 and glycerol content16 impact liquefaction yield. However, few papers discuss the role of the polyhydric alcohol structure. The most commonly used polyhydric alcohol solvent is a blend of PEG-400 and glycerol for polyurethane polyols. The very low equivalent weight of glycerol produces a reaction environment very rich in polar hydroxyl groups able to partake in the glycolysis of the bark compounds. In a study by Krzan et al. that employed microwave liquefaction, it is shown that shorter molecular weight glycols like ethylene glycol and propylene glycol were able to achieve higher liquefaction yields than their larger molecular weight counterparts, diethylene glycol and B

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering Table 1. Polyhydric Alcohols and Their Structural Characteristics polyhydric alcohol I.D.a P-2-75 S-2-96 P-2-200 S-2-213 P-2-1025 S-2-1000 S-3-86 S-3-336 P-3-330 S-4.4-157

chemical/trade name

molecular weight (Da)

triethylene glycol tripropylene glycol PEG-400 PPG-425 PEG-2050 JEFFOL PPG-2000 JEFFOL G30-650 JEFFOL FX-31-167 JEFFOL FE-12-60 JEFFOL SD-361

150 192 400 425 2050 2000 260 1008 990 690

functionality 2 2 2 2 2 2 3 3 3 4.4

polymer structureb glycol glycol glycol glycol glycol glycol glycerol glycerol glycerol sucrose + glycol

a Naming convention: alcohol type (primary-P, secondary-S)-functionality-equivalent weight. bPolymer structure refers to the initial “seed” compounds that were polymerized to form the various polyols.

vacuum. Complete removal of a highly polar solvent like DMF could be done with even higher temperatures or multiple azeotropic distillations; however, extended heat treatments may incur further degradation. Hence, the experiments with a cosolvent did not involve polyol characterization but rather focused on residue analysis. Polyol Characterization. Hydroxyl Value Determination and Viscometry. The hydroxyl value (OHV) was determined by the standard esterification method using phthalic anhydride.34 Polyol (1 g) and the pthalation reagent (25 mL) were heated at 100 °C for 15 min, cooled to room temperature, and then pyridine (50 mL) was added, followed by water (10 mL), and then titrated with 0.5 M NaOH to its equivalence point. The phthlation reagent was a solution of phthalic anhyrdride (41.43 g) and imidazole (6.43 g) in pyridine (250 mL), where S and B1 are the milliliters at the equivalence point of the sample and blank (no polyol), respectively; N is the normality of NaOH; W is the weight of the sample; OHV is the hydroxyl value in mgKOH/g of sample.

and polyol properties of liquefied bark. First, liquefaction yield will be studied with respect to the polyhydric alcohol structure (functionality, equivalent weight, and hydroxyl type) and cosolvent polarity. The residues from the liquefactions will then be characterized to determine how a solvent impacts residue composition and to better understand which factors govern the formation of insoluble residues (condensation products). Lastly, the polyols will be characterized by 1H NMR to assess the degradation of sugars and the inclusion of aromatic compounds.



EXPERIMENTAL SECTION

Materials. Polyethylene glycol (PEG) with a molecular weight of 400 Da was purchased from Fisher Scientific. Chlorobenzene, glycerol, sulfuric acid, xylene, sodium hydroxide, dimethylformamide, dioxane, toluene, and pyridine were purchased from Caledon Laboratories. 2,4,Dimethylpentanone, acetyl acetone, cyclohexanone, imidazole, polypropylene glycol 425, phthalic anhydride, triethyelene glycol, tripropylene glycol, and polyethylene glycol (Mw. 2050) were purchased from Sigma-Aldrich. All chemicals were used as received without further purification. Mountain pine beetle infested lodgepole pine (Pinus contorta) bark was supplied by FPInnovations. It was ground into a powder using a Wiley mill and then passed through a 70 mesh sieve (0.251 mm). The powder was then dewaxed through a Soxhlet extraction with hexanes for 4 h, after which it was dried in an oven overnight to produce bark without hexane-soluble extractives. The hexane extraction removed 15% w/w relative to the pristine oven dry bark. JEFFOL G30-650, JEFFOL PPG 2000, JEFFOL FX-31-167, JEFFOL FE 12-60, and JEFFOL SD-361were supplied by Huntsman. Liquefaction. Liquefactions were conducted at 130 °C to produce bark-based polyols. A polyhydric alcohol (7.2 g), glycerol (0.8 g), sulfuric acid (0.4 g), bark (4 g), and a cosolvent (20 mL) were added to a flask fitted with a condenser. These liquefaction conditions were based upon the findings of Wang et al. that showed that a solvent− biomass mass ratio of greater than 2−3 and a sulfuric acid loading of greater than 5 wt % (relative to the polyhydric alcohols) had diminishing returns on reducing the amount of residue.16 The polyhydric alcohols used are listed in Table 1, and the cosolvent used was xylenes. The flask was then heated for 1 h at 130 °C under a nitrogen environment. Next, the solution was diluted with a dioxane− water (8:2 w/v) solution (80 mL), and then neutralized using a 5 M sodium hydroxide solution. The solutions were then filtered and washed with dioxane−water (8:2 w/v) (2 × 100 mL). The water and dioxane were then removed through rotary evaporation for a sufficient period to ensure maximal removal of water. An additional set of liquefactions were done using PEG-400 as the polyhydric alcohol, and the cosolvent was varied (chlorobenzene, acetyl acetone, cyclohexanone, dimethylformamide, and 2,4-dimethylpentanone). The cosolvents xylene and chlorobenzene were removed through rotary evaporation at 60 °C under vacuum, while the more polar solvents were removed with a short-path distillation head at 100 °C under

OHVPA = (B1 − S)(56.1)(N)/W A BYK DV-E rotational viscometer was used with a temperature controlled sample holder at 25 °C. NMR Characterization. H NMR was performed on a Varian 400 spectrometer in DMSO-d6 at a concentration of 140 mg/mL. 1H NMR used a 1 s relaxation delay, pulse angle of 30°, acquisition time of 1.7 s, spectral width from − 2 to 16 ppm, and 64 scans. Spectra were calibrated and normalized using the solvent peak for DMSO. Assessment of Degradation and Aromatic Content. From the 1 H NMR spectra in Figure 5, the following areas were integrated to represent the amount of degradation and the aromatic content. The solvent residual proton peak in deuterated DMSO (2.45−2.55 ppm) was used as the standard for integration. The degradation peaks integrated consisted of the Hf (2.05−2.15 ppm) methine peak for levulinic acid and the Ha (8−8.5 ppm) methyl peak for formic acid. The aromatic region was integrated from 6−8 ppm. The integrated values were divided by the values from the P-2-200 (PEG-400) sample to provide a relative comparison. PEG-400 is a common glycol and polyhydric alcohol used in liquefactions and served well as a basis for comparison. The peaks corresponding to PEG-400 are the methylene protons Hd (3.5 ppm), the protons of the carbon in the α position to the hydroxyl group Hc (3.4 ppm), and the hydroxyl proton Hb (3.6− 3.7 ppm). Elemental analysis was performed on a 2400 Series II CHNS Analyzer to determine the carbon content of the residue. All values were made relative to the carbon content of the initial hexane extracted bark. A value of greater than one indicates that the residue consists of condensation products that have increased in carbon content due to the release of oxygen as water. This is based on the method used by Heitz et al. where a carbon to oxygen ratio was used.31 Residue Analysis. The residue analysis was based on a single liquefaction and its respective residues; therefore, the liquefied polyol amount may not correspond with the averaged yield values reported. The extractives content of the residues was determined by the TAPPI C

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Table 2. Change in Liquefaction Yield and Polyol Characteristics Due to Variations in the Polyhydric Alcohol Liquefying solvent polyhydric alcohola P-2-75 S-2-96 P-2-200 S-2-213 P-2-1025b S-2-1000 S-3-86 S-3-336 P-3-330 S-4.4-157

liquefaction yield (%) ± SD.

bark fraction (%)c

59.3 ± 11.9 33.1 ± 1.8 27.9 ± 1.4 22.9 ± 3.0 16.0 ± 9.4 28.1 ± 5.1 29.1 ± 0.9 23.2 ± 1.4 18.3 ± 1.9 26.7 ± 4.1

22.8 14.2 12.3 10.3 7.4 12.3 12.7 10.3 8.3 11.9

hydroxyl value (mgKOH/g) 585 537 315 254 − 132 573 138 200 254

± ± ± ±

4 5 6 17

± ± ± ± ±

14 36 9 6 2

viscosity (cP) 2360 390 780 170 − 757 9770 630 1550 2900

a Naming convention: alcohol type (primary-P, secondary-S)-functionality-equivalent weight. bP-2-1025 was not analyzed for hydroxyl value or viscosity because the polyol solidified after cooling from rotary evaporation. cDetermined from the liquefaction yield and the mass of glycerol and polyhydric alcohol solvent used.

Figure 1. Effect of (a) functionality, (b) equivalent weight, and (c) hydroxyl type on the liquefaction yield using various polyhydric alcohols. Statistical Analysis. Analysis was done using the software package R, version 2.15. Correlations were done using the Pearsons correlation coefficient (Rp).

standard method T-204 cm97: Solvent Extractives of Wood and Pulp; however, benzene was substituted with toluene. The acid insoluble lignin (AISL) content was determined by the TAPPI standard method, T-222 om-02: Acid-Insoluble Lignin in Wood and Pulp. The acid soluble lignin content was found to be negligible and has been reported to be less than 1% for softwood species.35 As a result, the holocellulose content = extractive free bark − AISL.



RESULTS AND DISCUSSION Effect of Polyhydric Alcohol Structure on Liquefaction Yield. Liquefaction behavior and polyol characteristics D

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 2. Amount of bark liquefied into a polyol and the composition of the unliquefied residue depending on the polyhydric alcohol solvent used. HE-Bark is hexane extracted bark. Naming convention: alcohol type (primary-P, secondary-S)-functionality-equivalent weight. AISL, acid insoluble lignin; Holo, holocellulose (cellulose and hemicelluloses).

significantly from that of primary hydroxyls and are discussed in the following sections. To summarize, these results revealed that low equivalent weight polyhydric alcohols with primary hydroxyl groups tended to have the highest yields, when other factors were kept constant. Also, since functionality did not seem to impact the liquefaction yield, this means that liquefactions can be done in triols successfully. As a result, polyols could be designed to be more consistent with the typical glycerol-based triols or sucrose-based polyols used in commercial rigid foam formulations. Effect of Solvents on the Residue Composition. Residue analysis plays an integral role in characterizing a polyol made from biomass, especially since a polyol is difficult to characterize directly. Therefore, by characterizing the residues, one can infer what biopolymers were liquefied and converted into a polyol. The bark residue was characterized based upon the amount of extractives, AISL (acid insoluble lignin), and holocellulose. The effect of variation of the polyhydric alcohol solvent on composition is found in Figure 2. The first observation was that aside from P-2-75 all the other samples showed increased AISL content above that of the initial bark. This result was likely due to various biopolymers that underwent condensation reactions that lead to insoluble precipitates that resembled the characteristics of acid-insoluble lignin. This result was also observed in our previous work where a higher liquefaction temperature promoted condensation reactions and resulted in an increase in the AISL content.26 Although the most likely factor is the phenolic groups in the extractive compounds, this cannot be the sole contributor to the AISL fraction. This was evident from sample P-2-1025 where an AISL content of 61% was observed. This implied that in addition to the extractive compounds, compounds belonging to the holocellulose fraction are also undergoing chemical modification and rearrangement to form precipitates. Pentoses from hemicellulose are known to form furfuryl alcohol and thereafter form insoluble humic structures.36 This is corroborated by work done on the dilute acid hydrolysis of softwood that showed the AISL content increased by 135% and was described as a pseudolignin. This was ascribed to the condensation products of carbohydrates and extractives.37 Comparing all polyhydric alcohols, it appeared that highly polar polyhydric alcohols were the most effective at the

were studied as a function of variations in the polyhydric alcohol structure. Details about liquefaction yield, bark fraction, and some polyol characteristics are given in Table 2. The wide range of liquefaction yields indicated that polyhydric alcohol structure had an influence on the liquefaction reaction. These structure effects were examined further to better understand their contributions to liquefaction. The effect of the polyhydric alcohol functionality, equivalent weight, and alcohol type were discussed with regard to their effect on the liquefaction yield. From a subset of the experiments, excluding polyhydric alcohols with low and high equivalent weights, it is shown in Figure 1a that functionality did not have a significant impact on the liquefaction yield. It was expected that higher functionality polyhydric alcohols would have led to high molecular weight compounds that became insoluble and formed a residue; however, this was not the case. The next comparison examined the role of equivalent weight. Only glycol-based polyhydric alcohols were included in the data subset and subdivided by type of alcohol. It is evident from Figure 1b that there was a negative correlation between equivalent weight and liquefaction yield. The Pearsons correlation coefficient (Rp) for primary alcohols was − 0.79 and for secondary alcohols was − 0.11. This indicated that if shorter chain molecular weight polyhydric alcohols were used, the yield could be increased and that this effect was more significant with primary alcohols. By reducing equivalent weight, the polyhydric alcohol solvent became more protic in nature and thereby increased the ability to hydrogen bond and disrupted the lignocellulosic matrix. Furthermore, smaller molecular weight polyhydric alcohols have a lower molar volume (molecular mass/density) that was beneficial for achieving greater penetration and accessibility. Finally, in order to understand the role of the hydroxyl type, all of the polyhydric alcohols were contrasted in Figure 1c with regard to their effect on liquefaction yield. Secondary hydroxyls achieved a consistent yield (average = 27 ± 3.8). However, polyols with primary hydroxyls showed a greater variation (average = 30 ± 19.9). This result reaffirmed the idea that as the equivalent weight was decreased for primary hydroxyls the yield could be improved. Since this effect was not observed for secondary hydroxyls, their mode of liquefaction might differ E

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

polarity of the solvent. The highly charge separated ketonic solvents like acetyl acetone (AA) and cyclohexanone (CH) had the lowest amount of acid-insoluble lignin/pseudolignin (condensation products). This supported the claim that AA and CH played a role in preventing condensation reactions. The final observation was that of the 57% holocellulose content of the bark, roughly two-thirds of that remained as a residue. This may be due to the large fraction of hemicelluloses that were easily broken down, while the cellulose was resistant to the treatment. Lodgepole pine bark has a sugar content of glucose (50%), galactose (7%), mannose (6%), arabinose (26%), and xylose (8%).8 With such a high percentage of hemicelluloses and the likelihood that some of that glucose is amorphous cellulose, it would be expected that more of the holocellulose would be liquefied. The likely explanation was that since these sugars existed in a hierarchical structure, their accessibility and conversion requires extended time. This was also consistent with the fact that all the residues had some extractives compounds remaining. Liquefaction for 1 h was unable to extract them from the cell walls, but a 6 h Soxhlet extraction succeeded. A finer mesh size for grinding the bark may be used to improve the yield; however, longer reaction times19 and higher temperature26 should be avoided since they increase the amount of sugar degradation. Effect of Organic Cosolvent on Liquefaction Yield. To examine the role of a cosolvent on liquefaction yield, a variety of solvents were used during liquefaction, and their effect on yield is shown in Figure 3. It was observed that ketonic solvents had the greatest polyol yield and the highest promise for increasing the liquefaction yield. To understand why this trend was observed, some of the commonly used measures of solvent polarity were correlated with liquefaction yield in Table 3. The data used for these correlations are in Table 4.

prevention of AISL/pseudolignin formation. Large molecular weight polyhydric alcohols like P-2-1025 and S-2-1000 were essentially nonpolar since their hydroxyl end groups are diluted by the long molecular weight chain. As a result, they had the highest AISL values at 61% and 45%, respectively. Secondary alcohols are known to be less polar than primary alcohols,38 and as a result, the S-2-96 (39%) residue had a higher AISL content than P-2-75 (14%). The mechanism of how the polyhydric alcohols prevent AISL formation can be explained by the formation of a protective solvation shell. Short chain primary alcohols produced a highly polar protic environment. This enabled the diol to act as a shield that hinders reactive carbocations from undergoing condensation reactions. Primary alcohols can undergo a (Sn2) nucleophilic substitution with other alcohols. Therefore, the primary alcohol solvation shell blocked condensation with other bark biopolymers that would lead to residue formation, and instead promoted the grafting of the diol through etherification as shown in Scheme 1, Reaction (i). Previous attempts to limit AISL formation focused on the addition of simple phenolics to quench reactive groups. Work done by Sealy et al.39 and Wayman et al.40 observed that the addition of simple phenolics like phloroglucinol and resorcinol, respectively, could curtail the formation of insoluble residues. This is because their high concentration and nucleophilicity enables them to condense on reactive carbocations, preventing those reactive species from undergoing condensation reactions with other high molecular weight biopolymers. The addition of phenolic groups into a polyol is undesirable since urethane linkages made from phenolic alcohols are not thermally stable.41 When only P-2-200 (PEG-400) was used as a polyhydric alcohol and the organic cosolvent was varied, the effect on the residue composition is shown in Figure 3. The primary

Table 3. Correlation between Solvent Polarity Measurements and Liquefaction Yielda solvent parameter

Rp

solvent parameter

Rp

Vm η ∈r μ ETN

−0.25 −0.04 0.15 0.36 0.50

TBP δT δd δp δh

0.58 0.28 −0.65 0.06 0.62

Vm, molar volum;, η, index of refraction; ∈r, relative permeability; μ, dipole moment; ETN, relative measure of solvent polarity based upon the change in color of a dye due to the influence of solvation; TBP, boiling point; δT, Hildebrand solubility parameter; δd,p,h, Hansen solubility parameters for dispersion, polar, and hydrogen bonding interactions. a

Figure 3. Amount of bark liquefied into a polyol and the composition of the unliquefied residue depending on the cosolvent used. HE-Bark, hexane extracted bark; CB, cholorobenzene; AA, acetyl acetone; CH, cyclohexanone; DMP, 2,4-dimethylpentanone; DMF, dimethylformamide. AISL, acid insoluble lignin; Holo, holocellulose (cellulose and hemicelluloses).

Both the index of refraction and the dielectric constant correlated poorly with yield. These measures are both related to the polarizability of the solvent. This result may be explained by the fact that as temperature increased the increase in kinetic energy of the solvent molecules randomized their orientations and thus reduced their dielectric constant. Furthermore, the poor correlation with parameters related to polarizability implied that the role of induced dipole−dipole interactions was minimal. The dipole moment and ETN correlated well, which meant that the ability of the solvent to have charge separated and to engage in solvation played a role in improving liquefaction.

observation was that usage of xylene, chlorobenzene (CB), and 2,4-dimethylpentanone (DMP) resulted in an increase in the AISL/pseudolignin content. This implied that these solvents were ineffective at preventing condensation reactions. Although DMP is a ketonic solvent, the presence of so many electrondonating methyl groups in its structure greatly reduced the F

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering Table 4. Solvent Polarity Measurements Data and Liquefaction Yielda solvent xylene chlorobenzene acetyl acetone cyclohexanone DMPb DMFb

liquefaction yield (%) ± SD. 27.9 29.2 50.7 58.6 24.9 28.3

± ± ± ± ± ±

1.4 1.0 0.7 3.7 5.7 4.8

η

Vm (cm3/mol)

∈r

ETN

μ (D)

δT (MPa1/2)

TBP (°C)

δd (MPa1/2)

δp (MPa1/2)

δh (MPa1/2)

1.49 1.52 1.45 1.45 1.4 1.43

124 102 103 104 141 77

2.56 5.69 26.52 16.1 17.2 38.25

0.074 0.188 0.571 0.281 0.247 0.386

1.3 1.67 3.09 3.07 2.71 3.78

18 19.4 22.1 20.3 16.4 24.8

138 131 141 155 125 153

16.5 17.4 11.3 15.6 17.4

7 9.4 11.8 9.4 13.7

2 0 10.7 11 11.3

η, index of refraction; Vm, molar volume; ∈r, relative permeability; ETN, relative measure of solvent polarity based upon the change in color of a dye due to the influence of solvation, μ, dipole moment; δT, Hildebrand solubility parameter; TBP, boiling point; δd,p,h, Hansen solubility parameters for dispersion, polar, and hydrogen bonding interactions. All data were accrued from a mix of the following sources, refs 32, 38, 42, and 47, but the data were mostly sourced as follows: ∈r38 | Vm47 | μ, ETN, TBP, η32 | δT,δd,p,h42. bDMP: 2,4-dimethylpentanone, DMF: dimethylformamide.

a

al. conducted a similar analysis using a ratio of the carbon/ oxygen content to examine the extent of condensation reactions.31 In Scheme 1, a series of condensation reactions are depicted. These reactions released oxygen in the form of water, and consequently, the carbon content increases. Condensation reactions also increase molecular weight and thereby reduce solubility, resulting in insoluble precipitates. Therefore, by analyzing the residues with respect to their carbon content, an insight into the extent of condensation reactions can be made. The effect of different organic solvents and of different polyhydric alcohol solvents on the carbon content are shown in Table 5. The first observable trend was that polar solvents

The solvent polarity measurements that correlated best with liquefaction yield were the boiling point and the Hansen solubility parameter for hydrogen bonding (δh). The boiling point represents how strongly a solvent molecule interacts with its solvent neighbors, so it can be a simple measure of polarity. However, since the liquefaction was done at 130 °C, using a higher boiling solvent may allow the reaction to occur at higher temperature if the reaction temperature was not controlled accurately. The strong correlation with δh was likely due to the fact that only the ketonic solvents AA and CH (data was not available for DMP) have high values for δh (10.7, 11 MPa1/2), while a solvent like xylene had a value of 2 MPa1/2.42 Although they are aprotic solvents, they have highly negatively charged carbonyls and therefore are excellent proton acceptors/electron donators. The ability of a ketone’s carbonyl to donate electron density to a reactive carbocation on a biopolymer may have played a role in the prevention of condensation reactions. The condensation reaction (ii) shown in Scheme 1 illustrates how a carbocation is attracted to the electron-rich carbons in an arene.43 Ketonic solvents have highly polar carbonyl groups resulting in the oxygen of the carbonyl being electron rich. It is proposed that the ketone’s carbonyl interacted with the carbocation radical, hence blocking the radical from undergoing condensation with an arene biopolymer. The role of the ketonic solvents in blocking condensation was further discussed in the sections on composition and carbon content analysis of the residues. A liquefaction reaction was done using the polyhydric alcohol with the highest yield, P-2-75 (59.3 ± 11.9%), and the cosolvent with the highest yield, cyclohexanone (58.6 ± 3.7%), and as a result a liquefaction yield of 64.8 ± 1.6% was achieved. This would seem to indicate that the polyhydric alcohol P-2-75 and the cosolvent cyclohexanone improve liquefaction yield since their combination produced a maximal yield. Therefore, through solvent selection a high liquefaction yield was achieved even at a moderate temperature of 130 °C. Due to scarcity of solvent polarity data, especially under the conditions pertinent to a liquefaction reaction, there have been few attempts to improve liquefaction through solvent selection. Research on the conversion and dissolution of biomass using reactive solvents like carbonates44 and greener solvents like ionic liquids is expanding.45 Hopefully, this preliminary study showed that there are simple characteristics that can be used to screen these emerging solvents. Assessing the Extent of Condensation Reactions in the Residue. Elemental analysis is commonly used to verify the chemical structure of purified compounds; however, it can also be used to assess the extent of chemical reactions. Heitz et

Table 5. Extent of Condensation Reactions Based upon the Carbon Content of the Liquefaction Residues, Made Relative to Initial Bark sample

C %/CBark %

bark P-2-2-200 S-2-213 P-2-−1025 P-2-75 P-3-330 xylene chlorobenzene acetyl acetone cyclohexanone DMP DMF

1.00 1.03 1.02 1.13 0.99 1.05 1.03 0.96 0.78 0.71 0.96 0.87

seemed to reduce the carbon ratio, which implied that polar solvents prevented condensation reactions from occurring and thereby prevented the formation of insoluble residues. The residue from liquefaction in xylene, a nonpolar solvent, had nearly the same carbon ratio as that of the initial bark. Chlorobenzene, which was slightly more polar, had a lower carbon ratio, and more polar solvents had even lower carbon ratio values. Both AA and CH have the lowest carbon content values and supported the claim that these solvents played a role in preventing condensation reactions through their dipole interactions. With regard to the polyhydric alcohols, all the polyhydric alcohols that have low equivalent weights behaved more like a protic, polar solvent and have roughly a carbon ratio of one and thus showed a protective effect. Sample P-2-1025 had the highest equivalent weight, which gave it the character of a very dilute protic solvent, and therefore, it acted nonpolar and had G

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

lower OHV. A chi-squared test of all the samples yielded a Pvalue of 0.25, which does not meet the 95% confidence interval typically used; however, a confidence of 75% would still seem to imply a trend. This was an important finding since it linked the extent of condensation reactions to the type of alcohol in the polyhydric alcohol solvent used. Assessment of Aromatic Content and Sugar Degradation in the Polyol. Determination of the viscosity and hydroxyl value are key parameters for the practical use of a polyol. However, in order to understand a polyols structure, especially one derived from liquefied natural materials, a greater depth of analysis is needed. 1H NMR is highly practical since its high sensitivity enables quantification of structural features through the integration of certain regions. This was shown in Figure 5 where the proton spectrum of P-2-200 was magnified to depict the regions related to the degradation of sugars and aromatic compounds. The degradation of high functionality sugars into low molecular weight, low functionality degradation products have negative implications for the polyol properties and resultant foam properties. The protons associated with formic and levulinic acids were integrated to assess the amount of degradation, and the values were made relative to the degradation observed from the P-2-200 sample (PEG-400, the standard polyhydric alcohol solvent). In Table 6, it is shown that there was a significant difference between the amount of degradation and the type of polyhydric alcohol used during liquefaction. This trend was more evident in the box plot in Figure 6, where both the mean amount of degradation and the range for secondary alcohols (1.34, 1−1.6) was higher than primary alcohols (0.9, 0.6−1.1). This was consistent with the above findings from the residue analysis that showed that primary alcohols were more effective at preventing AISL/ pseudolignin formation. It was proposed that the greater polarity of primary alcohols enabled hydrogen bonding and the formation of a protective solvation shell that prevented degradation reactions from occurring. The greatest amount of degradation was observed for the sample S-4.4-157. Since S-4.4-157 was a blend of propoxylated sucrose and diethylene glycol, it would be expected that those sugars might have degraded under the liquefaction conditions and contributed to the amount of degradation products. The aromatic region was integrated to assess the amount of aromatic structures in the polyol. The inclusion of phenolic groups introduces weak polyurethane linkages in a PUF, but the inclusion of an aromatic backbone will impart rigidity and temperature stability to a PUF.41 The presence of aromatic structures in the polyol indicated that the conditions used were favorable for preserving their structures unscathed. From Table 5, the aromatic content may have been influenced by the alcohol type. P-2-200 and S-2-213 have similar functionality and equivalent weight, and yet the primary alcohol has a higher aromatic content. P-2-75 also had a higher aromatic content than S-2-96. This observation may have been biased by the fact that a higher liquefaction yield would have inherently incorporated a greater fraction of aromatic groups. P-2-75 had the highest liquefaction yield and also had the highest aromatic content.

the highest carbon ratio. The solvation of bark biopolymers by highly polar solvent molecules could explain why the formation of insoluble condensation products was reduced. To summarize, low equivalent weight polyhydric alcohols and polar organic solvents showed a protective effect against condensation. The first may be attributed to the formation of a solvation shell through hydrogen bonding, while the latter could involve dipole interactions. These interactions of the biopolymer’s hydroxyl group with either the polyhydric alcohol solvent or a cosolvent may alter its reactivity and thereby hinder condensation reactions. Effect of Polyhydric Alcohols on Polyol Viscosity and Hydroxyl Value. Viscosity is a key characteristic of a polyol since it is intrinsically linked to molecular weight and is a determinant of how well a polyol mixes with an isocyanate and bubble growth kinetics. The viscosity of the polyols had a strong correlation (Rp = 0.91, P = 0.03) with the functionality of the polyhydric alcohol solvent used during the liquefaction. In Figure 4, only polyols with medium equivalent weight were

Figure 4. Effect of functionality and hydroxyl type on polyol viscosity.

shown for a simpler comparison and revealed a clear trend that as the solvent’s functionality increased, the liquefied product had a higher viscosity. This is simply explained since a greater amount of functional groups would have led to a greater amount of grafting and thus increased the molecular weight and therefore the viscosity. Another observation was that the methyl group in the PPG chains plays a role in preventing intermolecular interactions. This was demonstrated by the fact that S-2-1000 had a low viscosity of 757 cP while P-2-1025 had solidified after cooling from rotary evaporation. The hydroxyl value is a critical characteristic of a polyol since it represents the number of reactive hydroxyl groups. Polyols for rigid foam formulations generally have an hydroxyl value (OHV) from 300−600 mgKOH/g.46 From Table 2, it is shown that liquefactions done in polyhydric alcohols with lower equivalent weights were able to meet this criteria, while the higher equivalent weight samples would be better suited for flexible foam formulations. There also appeared to be a decrease in the OHV when using polyhydric alcohols with secondary alcohols. This suggested that when secondary alcohols are present more condensation reactions occur, resulting in a lower OHV. This was most evident with S-3336 and P-3-330, where their equivalent weights were nearly the same and yet their OHVs were substantially different, 138 versus 200 mgKOH/g. However, a small part of this difference can also be attributed to the methyl group of the secondary alcohol that resulted in a higher equivalent weight and therefore



CONCLUSIONS This study of bark liquefaction in polyhydric alcohols with varying structural characteristics and with solvents of varying polarity and structure have produced insights into the H

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 5. 1H NMR spectra of bark polyol using P-2-200 as a polyhydric alcohol solvent. Integration areas used for determining the amount of degradation (Hf-levulinic acid and Ha-formic acid) and the aromatic content are shown.

Table 6. 1H-NMR Was Used To Assess the Relative Amount of Degradation and Aromatics Content of the Bark-Based Polyols polyhydric alcoholb

relative degradationa

relative aromaticsa

P-2-75 S-2-96 P-2-200 S-2-213 P-2-1025 S-2-1000 S-3-87 P-3-330 S-3-336 S-4.4-157

0.6 1.0 1.0 1.6 1.1 1.1 1.4 0.9 1.6 2.3

1.2 0.7 1.0 0.7 0.3 0.4 0.9 0.6 0.9 0.9

Figure 6. As determined by H NMR, the relative amount of sugar degradation products based on the type of hydroxyl in the polyhydric alcohol solvent. All samples were made relative to P-2-200 as the basis of comparison.

a

Determination of the relative amounts of degradation and aromatic content was based upon a comparison to the PEG-400 sample (P-2200), and the methodology used was described in the Experimental Section. bNaming convention: alcohol type (primary-P, secondary-S)functionality-equivalent weight.

ments, the highest correlation was achieved with the Hansen solubility parameter for hydrogen bonding (δh). The ability of the ketone’s carbonyl to behave as a proton acceptor/electron donator may have facilitated interactions with carbocations that effectively blocked condensation reactions with biopolymer arenes. The composition and carbon content analysis of the residues also suggested that ketonic solvents like acetyl acetone and cyclohexanone may have played a role in preventing condensation reactions. Analysis of the composition of the liquefaction residues showed that short chain primary hydroxyl alcohols were most effective at preventing AISL/pseudolignin formation. It was proposed that under these conditions the solvent was highly protic, and the ability to hydrogen bond created a protective solvation shell that inhibited condensation reactions with other

effectiveness of liquefaction, the composition of the unliquefied residues, and the solvent factors that influenced the condensation reactions responsible for AISL/pseudolignin formation. Liquefaction yield was found to increase with decreasing equivalent weight. Polyhydric alcohols with secondary hydroxyls had a consistent yield, while those with primary hydroxyls showed a stronger dependence on equivalent weight. This showed that the highly polar hydroxyls (primary) and short chains created a highly protic solvent that improved conversion. Regarding organic cosolvents, ketonic solvents showed the greatest increase in the liquefaction yield. When liquefaction yield was correlated with various solvent polarity measureI

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

(6) Hirose, S.; Kobashigawa, K.; Izuta, Y.; Hatakeyama, H. Thermal degradation of polyurethanes containing lignin studied by TG-FTIR. Polym. Int. 1998, 47, 247−256. (7) Hatakeyama, T.; Matsumoto, Y.; Asano, Y.; Hatakeyama, H. Glass transition of rigid polyurethane foams derived from sodium lignosulfonate mixed with diethylene, triethylene and polyethylene glycols. Thermochim. Acta 2004, 416, 29−33. (8) Harkin, J. M.; Rowe, J. W. Bark and Its Possible Uses; U.S. Department of Agriculture, Forest Service, Forest Products Laboratory: Madison, WI, 1971; p 56. (9) Chattopadhyay, D. K.; Webster, D. C. Thermal stability and flame retardancy of polyurethanes. Prog. Polym. Sci. 2009, 34, 1068− 1133. (10) Hu, S.; Luo, X.; Li, Y. Polyols and polyurethanes from the liquefaction of lignocellulosic biomass. ChemSusChem 2014, 7, 66−72. (11) Gao, L.; Liu, Y.; Lei, H.; Peng, H.; Ruan, R. Preparation of semirigid polyurethane foam with liquefied bamboo residues. J. Appl. Polym. Sci. 2010, 116, 1694−1699. (12) Jin, Y.; Ruan, X.; Cheng, X.; Lü, Q. Liquefaction of lignin by polyethyleneglycol and glycerol. Bioresour. Technol. 2011, 102, 3581− 3583. (13) Zheng, Z.; Pan, H.; Huang, Y.; Chung, Y. H.; Zhang, X.; Feng, H. Rapid liquefaction of wood in polyhydric alcohols under microwave heating and its liquefied products for preparation of rigid polyurethane foam. Open Mater. Sci. J. 2011, 5, 1−8. (14) Kržan, A.; Ž agar, E. Microwave driven wood liquefaction with glycols. Bioresour. Technol. 2009, 100, 3143−6. (15) Hu, S.; Wan, C.; Li, Y. Production and characterization of biopolyols and polyurethane foams from crude glycerol based liquefaction of soybean straw. Bioresour. Technol. 2012, 103, 227−233. (16) Wang, H.; Chen, H.-Z. A novel method of utilizing the biomass resource: Rapid liquefaction of wheat straw and preparation of biodegradable polyurethane foam (PUF). J. Chin. Inst. Chem. Eng. 2007, 38, 95−102. (17) Hakim, A. A. A.; Nassar, M.; Emam, A.; Sultan, M. Preparation and characterization of rigid polyurethane foam prepared from sugarcane bagasse polyol. Mater. Chem. Phys. 2011, 129, 301−307. (18) Ge, J.; Zhong, W.; Guo, Z.; Li, W.; Sakai, K. Biodegradable polyurethane materials from bark and starch. I. Highly resilient foams. J. Appl. Polym. Sci. 2000, 77, 2575−2580. (19) Yamada, T.; Ono, H. Characterization of the products resulting from ethylene glycol liquefaction of cellulose. J. Wood Sci. 2001, 47, 458−464. (20) D’Souza, J.; Camargo, R.; Yan, N. Polyurethane foams made from liquefied bark-based polyols. J. Appl. Polym. Sci. 2014, 131, 40599. (21) Lee, S. H.; Teramoto, Y.; Shiraishi, N. Biodegradable Polyurethane Foam from Liquefied Waste Paper and its Thermal Stability, Biodegradability, and Genotoxicity. J. Appl. Polym. Sci. 2002, 83, 1482−1489. (22) Mun, S.-P.; Hassan, E.-B. M. Liquefaction of Lignocellulosic Biomass with Mixtures of Ethanol and Small Amounts of Phenol in the Presence of Methanesulfonic Acid Catalyst. J. Ind. Eng. Chem. 2004, 10, 722−727. (23) Kurimoto, Y.; Koizumi, A.; Doi, S.; Tamura, Y.; Ono, H. Wood species effects on the characteristics of liquefied wood and the properties of polyurethane films prepared from the liquefied wood. Biomass Bioenergy 2001, 21, 381−390. (24) Mun, S. P.; Gilmour, I. A.; Jordan, P. J. Effect of Organic Sulfonic Acids as Catalysts during Phenol Liquefaction of Pinus radiata Bark. J. Ind. Eng. Chem. 2006, 12, 720−726. (25) Jasiukaitytė, E.; Kunaver, M.; Strlič, M. Cellulose liquefaction in acidified ethylene glycol. Cellulose 2009, 16, 393−405. (26) D’Souza, J.; Yan, N. Producing Bark-based Polyols through Liquefaction: Effect of Liquefaction Temperature. ACS Sustainable Chem. Eng. 2013, 1, 534−540. (27) Yao, Y.; Yoshioka, M.; Shiraishi, N. Combined Liquefaction of Wood and Starch in a Polyethylene-Glycol Glycerin Blended Solvent. Mokuzai Gakkaishi 1993, 39, 930−938.

biopolymers. Finally, elemental analysis of the residues was used to identify the extent of condensation reactions since the loss of water will have increased the carbon content. It was found that polar organic solvents and low equivalent weight polyhydric alcohols had a lower carbon ratio, and therefore, less condensation reactions occurred. Determination of the hydroxyl value of the polyols also proved insightful. A slight trend was observed of lower OHV values for secondary polyhydric alcohol compared to the primary. This indicated that condensation reactions may occur to a greater extent when secondary alcohols are present. This is consistent with the 1H NMR analysis that showed a greater amount of sugar degradation products when secondary alcohols were used. These results have shown that the selection of polyhydric alcohols and organic cosolvents have had a significant impact on the liquefaction yield and the polyol characteristics. The effect of the polyhydric alcohol solvent was especially important since it remains as the dominant constituent of the bark-based polyols and will largely dictate the PUF behavior. With a better understanding of how solvent impacts liquefaction, it will be possible to have greater control over the properties of the polyol and improve the viability of bark-based PUFs.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (416) 946 8070. Fax: (416) 978 3834. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding

The authors acknowledge the financial support from the ORF“Bark biorefinery” partners and the Mitacs-Accelerate Scholarship Program. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank FPInnovations for supplying lodgepole pine bark and Huntsman for providing their Jeffol polyols (JEFFOL® is a registered trademark of Huntsman Corporation or an affiliate thereof in one or more but not all countries). Funding support from ORF-“Bark Biorefinery” partners is greatly acknowledged.



REFERENCES

(1) Zhao, Y.; Yan, N.; Feng, M. W. Biobased phenol formaldehyde resins derived from beetle-infested pine barks - Structure and composition. ACS Sustainable Chem. Eng. 2013, 1, 91−101. (2) Hartman, S. Polyurethane Foams from the Reaction of Bark and Diisocyanate. In Wood Technology: Chemical Aspects; Goldstein, I., Ed.; ACS Symposium Series 43; American Chemical Society: Washington, DC, 1977; Vol. 43, pp 257−269. (3) Ge, J. J.; Sakai, K. Compressive Properties and Biodegradabilities of Polyurethane Foams Derived from Condensed Tannin. Mokuzai Gakkaishi 1993, 39, 801−806. (4) Zhao, Y.; Yan, N.; Feng, M. Polyurethane foams derived from liquefied mountain pine beetle-Infested barks. J. Appl. Polym. Sci. 2012, 123, 2849−2858. (5) Ueno, T. Preparation and properties of polyurethane foam derived from liquefied cedar bark. Mokuzai Kogyo 2007, 62, 358−363. J

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering (28) Mun, S.-P.; Hassan, E.-B. M. Liquefaction of Lignocellulosic Biomass with Dioxane/Polar Solvent Mixtures in the Presence of an Acid Catalyst. J. Ind. Eng. Chem. 2004, 10, 473−477. (29) Zou, X.; Qin, T.; Wang, Y.; Huang, L. Mechanisms and Product Specialties of the Alcoholysis Processes of Poplar Components. Energy Fuels 2011, 25, 3786−3792. (30) Hansen, C. M.; Bjorkman, A. The Ultrastructure of Wood from a Solubility Parameter Point of View. Holzforschung 1998, 52, 335− 344. (31) Heitz, M.; Brown, A.; Chornet, E. Solvent Effects on Liquefaction - Solubilization Profiles of a Canadian Prototype Wood, Populus-Deltoides, in the Presence of Different Solvents. Can. J. Chem. Eng. 1994, 72, 1021−1027. (32) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry; 3rd ed.; Wiley-VCH, 2003. (33) Katritzky, A. R.; Fara, D. C.; Yang, H.; Tamm, K.; Tamm, T.; Karelson, M. Quantitative Measures of Solvent Polarity. Chem. Rev. 2004, 104, 175−198. (34) Carey, M. A.; Wellons, S. L.; Elder, D. K. Rapid Method for Measuring the Hydroxyl Content of Polyurethane Polyols. J. Cell. Plast. 1984, 20, 42−48. (35) Maekawa, E.; Ichizawa, T.; Koshijima, T. An Evaluation of the Acid-Soluble Lignin Determination in Analyses of Lignin by the Sulfuric Acid Method. J. Wood Chem. Technol. 1989, 9, 549−567. (36) Saeman, J. F. Kinetics of wood saccharification - hydrolysis of cellulose and decomposition of sugars in dilute acid at high temperature. Ind. Eng. Chem. 1945, 37, 43−52. (37) Nguyen, Q. A.; Tucker, M. P.; Keller, F. A.; Beaty, D. A.; Connors, K. M.; Eddy, F. P. Dilute acid hydrolysis of softwoods. Appl. Biochem. Biotechnol. 1999, 77, 133−142. (38) Wohlfarth, C. Permittivity (Dielectric Constants) of Liquids. In CRC Handbook of Chemistry and Physics; Haynes, W. M., Ed.; CRC Press, 2011; pp 187−208. (39) Sealyfisher, V. J.; Pizzi, A. Increased Pine Tannins Extraction and Wood Adhesives Development by Phlobaphenes Minimization. Holz Als Roh-Und Werkst. 1992, 50, 212−220. (40) Wayman, M.; Lora, J. H. Simulated Autohydrolysis of Aspen Milled Wood Lignin in the Presence of Aromatic Additives. Changes in Molecular Weight Distribution. J. Appl. Polym. Sci. 1980, 25, 589− 596. (41) Chattopadhyay, D. K. K.; Webster, D. C. Thermal stability and flame retardancy of polyurethanes. Prog. Polym. Sci. 2009, 34, 1068− 1133. (42) Barton, A. F. M. CRC Handbook of Solubility Parameters and Other Cohesion Parameters; CRC Press: Boca Raton, FL, 1983. (43) Shimada, K.; Hosoya, S.; Ikeda, T. Condensation reactions of softwood and hardwood lignin model compounds under organic acid cooking conditions. J. Wood Chem. Technol. 1997, 17, 57−72. (44) Kühnel, I.; Podschun, J.; Saake, B.; Lehnen, R. Synthesis of lignin polyols via oxyalkylation with propylene carbonate. Holzforschung 2015, 69, 531−538. (45) Kilpeläinen, I.; Xie, H.; King, A.; Granstrom, M.; Heikkinen, S.; Argyropoulos, D. S. Dissolution of Wood in Ionic Liquids. J. Agric. Food Chem. 2007, 55, 9142−9148. (46) Ionescu, M. Chemistry and Technology of Polyols for Polyurethanes; Rapra Technology Limited: Shawbury, Shrewsbury, Shropshire, U.K., 2005. (47) Physical Constants of Organic Compounds. In CRC Handbook of Chemistry and Physics; Haynes, W. M., Ed.; CRC Press, 2015.

K

DOI: 10.1021/acssuschemeng.5b00908 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX