Selective Synthesis of Phenolic Compounds from Alkali Lignin in a

Feb 15, 2016 - The effect of the temperature at a constant pressure and short residence time on the selectivity and yield of phenolic products from th...
0 downloads 13 Views 2MB Size
Download PDF
Article pubs.acs.org/EF

Selective Synthesis of Phenolic Compounds from Alkali Lignin in a Mixture of Sub- and Supercritical Fluids: Catalysis by CO2 Abu M. Numan-Al-Mobin,*,† Keith Voeller,‡ Honza Bilek,‡ Evguenii Kozliak,‡ Alena Kubatova,‡ Douglas Raynie,§ David Dixon,† and Alevtina Smirnova*,† †

South Dakota School of Mines and Technology, Rapid City, South Dakota 57701, United States University of North Dakota, Grand Forks, North Dakota 58202, United States § South Dakota State University, Brookings, South Dakota 57006, Unites States ‡

S Supporting Information *

ABSTRACT: A successful selective liquefaction of lignin has been demonstrated in the presence of a H2O−CO2 mixture at 300 °C, yielding 40−50 wt % organic phenolic phase. The effect of the temperature at a constant pressure and short residence time on the selectivity and yield of phenolic products from the hydrothermal reforming of alkali lignin in a mixture of sub- and supercritical fluids (H2O mixed with CO2 or N2) has been investigated. Dependent upon the processing conditions, the lignin samples produced a homologous series of phenols, such as guaiacol, homovanillic acid, quaiacyl carbonyls, guaiacyl dimers, phenol, and cresol. Gas chromatography−mass spectrometry (GC−MS), total organic carbon (TOC), and pyrolysis−GC−MS (Py−GC−MS) were used for chemical analysis of the organic liquid and solid phases. The results from GC−MS analysis of the liquid organic phases demonstrated the trend of increasing the amounts of major guaiacol products with the temperature. The thermal carbon analysis (TCA) showed a significant increase of the readily volatile organic carbon in the liquid fractions resulting from the treatments at 300 and 400 °C at the expense of less volatile organic carbon and recalcitrant pyrolyzed carbon. Evaluated for the first time, a significant effect of CO2 versus N2 was demonstrated, providing both a higher yield of volatile products and more selective synthesis of guaiacols.

■

INTRODUCTION

To increase the amount of lower molecular weight phenolic products, lignin decomposition was carried out in supercritical water (scH2O) (400 °C) with the addition of phenol.10 This approach resulted in a lower amount of insoluble products (23 wt % without phenol versus 17 wt % with phenol) and a shift toward the lower molecular weight products (from a molecular weight of ≈1500 to a molecular weight of ≤200). At a constant water density, the molecular weight of insoluble tetrahydrofuran products was lower in the presence of phenol than without phenol. However, the amount of lower molecular weight phenolic products decreased at a higher water density. The conclusion was made that the occurrence of water in the reaction medium was insignificant and only the addition of phenol caused the formation of lower molecular weight phenolic products by reacting with the active sites pertinent to lignin depolymerization, instead of other reactive sites responsible for the formation of heavier fragments.10 Similar to lignin, biomass liquefaction in sub- and supercritical water resulted in a maximum yield of heavy oils of 53.3% at 380 °C.11,12 The biomass treated in sub- and supercritical (300−400 °C) water13 produced more carbon at higher temperatures. An increased concentration of cyclopentenones (C5H8O, 3−7%) and phenols (C6H6O, 9−14%) and a decreased guaiacol content (C7H8O2, 28−20%) were explained by hydrolysis of ether bonds and alkylation of

Lignin is one of the most abundant natural biopolymers and a major non-fossil and carbon-rich renewable source of aromatic compounds. In lignin, three main monomers (e.g., 4hydroxycinnamyl alcohol, coniferyl alcohol, and sinapyl alcohol) are linked by C−O and C−C bonds, forming arylether β-O-4 linkages (≤50% of all lignin linkages)1 that are responsible for a highly branched aromatic heteropolymer structure.2 Thus, lignin is considered as one of the most promising raw materials for synthesis of high-value organic products (e.g., vanillin, guaiacol, phenol, cresol, etc.) that are applicable for production of renewable polymers by ringopening polymerization,3 polycondensation,4 or electrophilic aromatic substitution.5 However, the selectivity in the formation of specific phenolic products has not been fully investigated, especially in the case of alkali lignin. In the past, various degradation processes of lignin and biomass have been studied, such as enzymatic degradation, pyrolysis, oxidation, and hydrothermal liquefaction, resulting in the production of aromatic aldehydes and phenolics.6 In the case of lignin oxidative treatment, various aldehydes, e.g., syringaldehyde and vanillin, as well as substituted benzoic acids have been produced.7 Hydrogenolysis in the presence of basic8 catalysts (e.g., NaOH with Ru/C) and metal catalysts9 was employed to cleave β-O-4 ether bonds, leading to lignin liquefaction (92.5% lignin was converted, yielding 12.69% phenolic monomers, 6.12% aliphatic alcohols, and less than 14.03% residual solid) based on gravimetric analyses. © XXXX American Chemical Society

Received: September 18, 2015 Revised: February 6, 2016

A

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels aromatic rings of the intermediate products. A hydrothermal treatment of lignin in the presence of small amounts (14 wt %) of formic acid as a hydrogen donor in scCO2/acetone/water fluid at 300 °C and 100 bar pressure was reported.14 However, only 12% of the aromatic compounds was produced by this method. In most of the above-mentioned and other approaches, the selectivity toward the formation of specific phenolic products in supercritical fluids has not been highlighted. However, it can be assumed that kinetics could be a dominating factor toward selective synthesis of high-value organic products. As emphasized earlier,15 as a result of fast kinetics in scH2O, the total carbon content in the liquid phase reaches its maximum at short residence times (∼1 s), causing the scission of lignin linkages and producing monomeric products, including phenol. In subcritical water,16 the maximum yield of monomers was observed at a residence time of only 2−4 s. These results were explained by a longer time required for repolymerization and either kinetic or quasi-thermodynamic mechanisms rather than a full thermodynamic control. A conclusion was made that lignin treatment aiming to shift the balance of depolymerization and repolymerization processes should be conducted at low temperatures, short residence times, and in the presence of catalysts. The goal of the present study was to validate the hypothesis that the presence of supercritical carbon dioxide (scCO2) as a second component in a mixture with sub- or supercritical water can significantly increase the yield and selectivity toward the formation of specific high-value phenolic products. To achieve the goal of lignin liquefaction in the presence of scCO2, the results obtained for a H2O−CO2 system at sub- and supercritical conditions at different temperatures were compared to a baseline system containing H2O pressurized by an inert gas (N2). The liquid- and solid-phase products were characterized using the liquid−liquid extraction (LLE) with gas chromatography−mass spectrometry (GC−MS) and pyrolysis−GC−MS (Py−GC−MS), respectively. In this work, we have employed a novel thermal carbon analysis (TCA) method, yielding a carbon mass balance closure. To the best of our knowledge, this is the first study in which a systematic analysis of the temperature effect on lignin decomposition in a mixture of sub- and supercritical environmentally benign solvents has been performed.

■

Figure 1. Block diagram of the experimental setup for the lignin hydrothermal treatment in a mixture of sub- and supercritical fluids.

the inserted thermocouple was sealed and placed into the split Carbolyte furnace for hydrothermal treatment. To reach the targeted temperature within the shortest period of time, the initial setup temperature of the Carbolyte temperature controller was adjusted to 650 °C with a ramp rate of 100 °C/min. Dependent upon the synthesis conditions, after 2−5 min, the temperature was adjusted to the required value. The pressure inside the vessel (22.063 MPa) was maintained by a Teledyne syringe pump 260D pressurized with CO2 or N2. When the pressure and temperature requirements were met, a stopwatch was used to record the residence time. During this time period, the temperature and pressure inside the vessel were continuously monitored. After 10 min, the furnace was turned off and the pressure was released. The reaction was then quenched by the immersion of the reaction vessel into cold water. The liquid that came out through the pressure valve during the pressure release was collected and combined with the liquid phase from the reactor to measure the final weight. After the LLE, the collected liquid phase was characterized using TCA, GC−MS, and total organic carbon (TOC) analysis. The solid phase was collected from the vessel after an overnight drying, weighed, and characterized using Py−GC−MS. Characterization of the Liquid Organic Phase. After the treatment of lignin in a mixture of sub- and supercritical fluids, the liquid phase was characterized using LLE with GS−MS, TOC, and TCA. The organic phase was separated from the aqueous phase by a LLE using DCM and weighed. In the first step of extraction, 10 μL of acetic acid was added to every 1 mL of a liquid sample while adjusting pH to ∼4.0. Then, the recovery standard (4-chloro-acetophenone) was added, enabling us to monitor and correct for losses during the extraction. The sample was then extracted 3 times with 1 mL of DCM with vigorous shaking. After the DCM (bottom layer) and water (top layer) phases separated completely, the bottom layers were collected and combined and then the internal standard was added. The quantification was based on the response factors of the individual standards. If the corresponding standards were not available, structurally similar compounds were used (details are provided in Supplementary Table 1 of the Supporting Information). The relative yield obtained for LLE GC−MS analysis is defined as the ratio of the area under a specific chromatogram peak to the total area of the peaks after subtraction of the area of the internal standard peak (4-chloacetophenone).

EXPERIMENTAL SECTION

Materials. Alkali lignin and 4-chloroacetophenone were purchased from Sigma-Aldrich. Deionized water was obtained using a Milli-Q Integral Water Purification System (EMD Millipore Corp., Billerica, MA). For the LLE, acetic acid and dichloromethane (DCM) of GC quality were obtained from Sigma-Aldrich (Atlanta, GA). The reactor, tubing, and fittings were purchased from High Pressure Equipment Company (Erie, PA) and Swagelok (Solon, OH). Sub- and Supercritical Hydrothermal Treatments of Lignin. The treatment of alkali lignin at sub- and supercritical conditions in a temperature range of 200−500 °C was carried out in a stainless-steel high-pressure vessel (316 SS) having a capacity of 12 mL with a pressure tolerance of up to 103 MPa. A type K thermocouple was inserted through the bottom of the vessel to measure the temperature inside the vessel by an AMProbe temperature meter. The internal pressure was controlled by the pressure sensor connected to the monitor. To achieve reproducible synthesis conditions, the temperature ramp was adjusted in each experiment. A complete setup for the hydrothermal procedure is shown in a block diagram (Figure 1). In each experiment, 0.1 g of lignin was placed inside the vessel and 6 mL of deionized water was added to disperse the lignin. The vessel with B

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

Figure 2. GC−MS analysis data of the phenolic products formed after lignin hydrothermal treatment at different temperatures and constant pressure (22.063 MPa) in the presence of H2O−CO2 or H2O−N2 fluid mixtures. The GC−MS analyses were performed using an Agilent 5890N GC equipped with 5971 MSD (EI). All injections (1.0 μL) were conducted in a splitless mode (0.5 min) for 1.0 min using a splitless liner with deactivated glass wool (Restek, Bellefonte, PA). A 50 m DB-5MS column (J&W Scientific, Inc., Folsom, CA) with a 0.25 mm inner diameter and a 0.25 μm film thickness was used for all separations. Ultrapure helium (99.999%) was used as a carrier gas with a constant flow rate of 1.0 mL min −1. The initial oven temperature was set to 40 °C, held for 1.0 min, then ramped to 140 °C with a rate of 20 °C/min, then ramped to 290 °C with a rate of 10 °C, and held for 12 min. The injector and transfer line temperatures were set at 250 and 280 °C, respectively. All MS data were acquired in a total ion current (TIC) mode with a mass range of m/z 50−550. The TOC analyzer employed was a Sievers InnovOx laboratory TOC analyzer with a GE autosampler. For TOC analysis, the liquid phases obtained after the lignin treatment at 200, 300, and 400 °C were filtered, centrifuged at 13 000 rpm for 10 min to remove the particles, and diluted with distilled water to 33% by volume. Distilled water (17.8 MΩ cm at 25 °C) was used as a TOC blank. The TCA was performed on a thermal optical analyzer purchased from Sunset Laboratories, Inc. (Tigard, OR). The advantage of the TCA instrument is that it enables desorption and pyrolysis of the carbonaceous species at temperatures from 200 to 850 °C under an inert atmosphere (helium). After all volatile species evolved, the heating is conducted under an oxidation atmosphere (in the presence of oxygen) to quantitatively generate CO2, which is converted to methane and quantified using flame ionization detection calibrated with methane. The use of this method ensures the comprehensive determination of all of the carbon in a sample. Characterization of the Solid Phase. The weight of the solidphase product before and after hydrothermal reforming in scCO2 was measured gravimetrically. At 300 °C, the lignin consumption was ∼50%. However, in the future, these measurements will be optimized using a bigger vessel and larger samples for accuracy. The solid phase left over after the lignin treatment was characterized using a Py−GC−MS Shimadzu QP2010 Ultra equipped with an automatic sampler from Frontier with the injection port temperature set at 250 °C and the furnace operating at 500 °C. For each sample, approximately 100 μg of a solid phase was weighed and placed in a sample holder provided by Frontier. For identification of the solidstate products, the National Institute of Standards and Technology (NIST) mass spectral library with a search program (version NIST 14) was used.

mL). The second component (carbon dioxide) that is always present in the vessel in the supercritical state was used for pressurization. To identify the specific effects of scCO2, the results were repeated with N2. In the chosen temperature range, nitrogen inside the vessel, like CO2, was always present in a supercritical state. At 22.063 MPa in the experimental temperature range of 200−500 °C, both N2 and CO2 occur as supercritical fluids; however, water becomes supercritical only at 374 °C, whereas in the lower temperature range (250−350 °C), it is present in a subcritical phase. Complex interactions between two fluids and the properties of the scCO2 in sub- or supercritical water affect the mutual solubility of the reagents,17 chemical reactions, and kinetics of the catalytic processes that take place on the lignin surface. Furthermore, the gaseous and phenolic species produced during hydrothermal treatment are known to catalyze lignin depolymerization,10 which makes the analysis of lignin depolymerization even more complicated. The gravimetric analysis shows that about 50% of lignin is converted into the organic extractable products in a temperature range of 300−500 °C, with the rest being a solid-phase product. Considering that the mass of the evolved gases during the lignin treatment is negligible in comparison to the weight of the solid and liquid products, it was assumed that this mass of the gas phase should have a minimal effect on the total mass balance. LLE GC−MS Analysis of the Liquid Organic Phases. The comparative GC−MS results for the lignin decomposition products formed in H2O−CO2 and H2O−N2 fluid mixtures within a temperature range of 300−400 °C, constant pressure (22.063 MPa), fixed residence time (10 min), and experimental yield equal to or greater than 5% are presented in Figure 2. Focused mainly on selective production of specific phenolic compounds, Figure 2 shows only representative classes of compounds that were produced. These experiments were reproduced several times, and the standard deviation did not exceed 3%. The lignin degradation starts at lower temperatures (e.g., 200 °C), but the guaiacol derivatives as main phenolic products are observed only at higher temperatures for both H2O−N2 and H2O−CO2 near critical mixtures, which is in good correlation with the previously published results.18 As a general trend, guaiacyl acids represented primarily by a homovanillic acid are the main products at lower temperatures, but their amounts decline in the samples obtained at higher temperatures and are

■

RESULTS AND DISCUSSION Two fluid mixtures, specifically H2O−CO2 and H2O−N2, were chosen for the hydrothermal degradation of lignin. In those mixtures, one of the components (water) in the sub- or supercritical state was present at a constant concentration defined by the volume of water in the supercritical vessel (6 C

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels Table 1. Tentatively Identified Lignin-Related Derivatives from GC−MS Library

a

name of the compounda

RTb

m/z

quantification ion

phenol phenol, 2-methoxycreosol phenol, 4-ethyl-2-methoxyeugenol phenol, 2-methoxy-4-propylbenzaldehyde, 3-hydroxy-4-methoxy- (vanillin) trans-isoeugenol phenol, 2-methoxy-4-propyl1,2,4-trimethoxybenzene apocynin 1,2-dimethoxy-4-n-propylbenzene 2-propanone, 1-(4-hydroxy-3-methoxyphenyl)methyl-(2-hydoxy-3-ethoxy-benzyl)ether benzeneacetic acid, 4-hydroxy-3-methoxy-, methyl ester 2-propenal, 3-(4-hydroxy-3-methoxyphenyl)benzeneacetic acid, 2,5-dimethoxyhomovanillyl alcohol retene 4′-methoxy-2-hydroxystilbene methyl dehydroabietate 1,1′-biphenyl-3,4,4′-trimethoxy-6′-formylcarinol homovanillic acid

7.05 8.89 10.50 11.75 12.85 12.99 13.45 14.13 14.18 14.30 14.57 14.80 15.08 16.53 17.39 17.53 18.53 19.30 22.51 23.16 23.49 25.81 27.28 28.03

94 124 138 152 164 164 152 164 166 168 166 180 180 182 196 178 196 168 234 226 314 272 378 182

C6H6O C7H8O2 C8H10O2 C9H12O2 C10H12O2 C10H12O2 C8H8O3 C10H12O2 C10H14O2 C9H12O3 C9H10O3 C11H16O2 C10H12O3 C10H14O3 C10H12O4 C10H10O3 C10H12O4 C9H12O3 C18H18 C15H14O2 C21H30O2 C16H16O4 C20H26O7 C9H10O4

The names of compounds are based on the identification using the MS NIST library. bRT = retention time.

highest yield, as high as 75% at 500 °C. However, as expected for high-temperature processes, other phenolic derivatives were observed, such as phenol and its homologues, signifying lower selectivity. TOC Analysis. The results of the TOC evaluation for H2O−CO2 in comparison to H2O−N2 fluid mixtures at 200, 300, and 400 °C are presented in Figure 3. At 200 °C, the total

replaced with guaiacol derivatives, i.e., guaiacol itself and its methyl, ethyl, and propyl homologues (Figure 2). Table 1 showing mass spectral data used in identification of the products given in Figure 2 is provided. A similar trend is also observed for quaiacyl carbonyls (represented mainly by vanillin) and quaiacyl dimers tentatively identified on the basis of mass spectra similar to quaiacol and molecular ions of 372 and 372 amu. The most significant difference between the CO2- and N2treated samples is observed at 300 °C. Namely, guaiacols are recovered in higher abundance (Figure 2) in the presence of scCO2 with relative yields of 60% in contrast to merely 37% in the presence of N2. The guaiacyl acids are the second most abundant group of products. In contrast, guaiacyl acids are the main products when the process was conducted at 300 °C in the presence of N2. When the treatment temperature is increased to 400 °C, this difference between the CO2- and N2treated samples becomes less pronounced. The observed difference between H2O−CO2 and H2O−N2 at 300 °C with respect to a higher selectivity in the formation of guaiacols can be explained by the presence of carbon dioxide as an acidic homogeneous catalyst that affects the pathways of lignin depolymerization.19 The catalyst-enabled selectivity is expected to diminish at higher temperatures, as observed. The difference between the 300 and 400 °C CO2-treated samples can be explained by the decrease in solubility of scCO2 in subversus supercritical water from 2.9 mol % at 300 °C to 2.0 mol % at 400 °C, respectively.17 This observation is in correlation with the earlier published data for kenaf stems in subcritical water explained by the formation of carbonic acid in CO2pressurized hydrothermal treatment that caused enhanced solubilization.20 The selected experiments were also conducted at 500 °C in the H2O−N2 system, and their results corroborate these conclusions. As expected, guaiacol was detected at the

Figure 3. TOC concentration (mg/L) for different treatment temperatures in H2O−CO2 and H2O−N2 fluid mixtures.

carbon content is almost the same in both H2O−N2 and H2O− CO2 systems. In contrast, the amount of carbon is higher in H2O−CO2 than in H2O−N2 at 300 and 400 °C. These observations are in correlation with the GC−MS results and also provide new information about the process, because TOC allows for a comparison of the total carbon in the liquid phase for varied process conditions. TCA of the Liquid-Phase Products. The comparative results of the TCA of the liquid mixed organic/aqueous phase produced in H2O−CO2 or H2O−N2 fluid mixtures at 200, 300, D

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

Figure 4. TCA of the liquid organic phase produced from lignin decomposition in H2O−CO2 or H2O−N2 fluid mixtures at (a) 200 °C, (b) 300 °C, and (c) 400 °C.

and 400 °C are presented in panels a−c of Figure 4. All of the samples were tested a few times for reproducibility, and the error bars indicate the measurement deviation. The benefit of TCA is in obtaining a comprehensive distribution of carbonaceous species in the liquid sample, for both the untreated (raw) lignin and the products of its decomposition obtained as a result of hydrotreatment, thus allowing for their comparison. The temperature fractions reported from TCA can be differentiated as those desorbing small-molecular-weight compounds (200 and 300 °C), pyrolyzed large-molecularweight compounds (400−850 °C), and the final fraction that evolved in the presence of oxygen only after all volatile compounds are evaporated. This last fraction represents recalcitrant pyrolyzed carbon (char) formed during the TCA heating process. It is necessary to note that the raw lignin shows as much as 60 wt % in this recalcitrant “char” fraction. For the lignin samples produced in CO2−H2O or CO2−N2 fluid mixtures at 200 °C, no difference was detected, including an abundant pyrolyzed carbon fraction (Figure 4a), meaning that this temperature is not sufficient for lignin depolymerization. In contrast, the TCA profiles of lignin samples produced by the thermal lignin treatment in CO2−H2O or N2−H2O fluid mixtures at 300 °C show a major difference compared to TCA of the raw lignin (Figure 4b). Apparently, as a result of such a high temperature, most of the lignin is converted into volatile products and provides the highest observed organic carbon content at the expense of “char” carbon. The catalytic effect of the scCO2 treatment at 300 °C is revealed as the difference in the amounts of carbon evolving at 200 °C with and without

scCO2 (Figure 4b), validating the GC−MS data. A significantly higher concentration of pyrolyzed carbon (41%) is observed in the samples treated at the same temperature in the N2−H2O fluid mixture in comparison to the samples produced in the CO2−H2O system (16%). At 400 °C (Figure 4c), the highest amount of carbon evolving at 200 °C is detected, also validating the above GC− MS data. Concurrently with volatile monomer accumulation, the char carbon fraction further declines compared to lower treatment temperatures. In contrast to lignin liquefaction at 300 °C, only a small increase of the relative yield (1.4%) in the H2O−CO2 system compared to the H2O−N2 system is observed at 400 °C. A significant difference is also detected for the H2O−CO2 and H2O−N2 systems when water changes its physical state from subcritical at 300 °C to supercritical at 400 °C. At this temperature (400 °C), both water and carbon dioxide are present in the supercritical state. However, water is much more active at 400 °C than at 300 °C, thus minimizing the difference between H2O and CO2 as supercritical catalytically active fluids. As a result of the TCA in the produced liquid phase in a temperature range of 200−400 °C, a conclusion can be made that lignin depolymerization is more efficient in H2O−CO2 than H2O−N2 fluid mixtures, but only at moderately high temperatures, e.g., 300 °C. In the presence of scCO2, higher concentrations of the monomers are produced at this temperature in combination with lower amounts of pyrolyzed carbon. E

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels Py−GC−MS Data. As a result of lignin depolymerization in H2O−CO2 and H2O−N2, significant amounts of solid-phase products are produced, yielding up to 50 wt % of the initial lignin amount. At up to 300 °C hydrotreatment, the solid phase, appearing only slightly darker than the raw lignin, was collected as a powder without visible tar or char formation. The samples hydrothermally treated at higher temperatures (400 and 500 °C) were much darker, indicating a partial char formation. However, in comparison to previous reports,8 no tar was produced. The results of Py−GC−MS of the solid-state phase produced in H2O−CO2 at two temperatures of 250 and 350 °C (around the key temperature of 300 °C) are presented in Figure 5.

Figure 6. Simplified scheme of lignin degradation, adapted with permission from ref 21. Copyright 2011 Elsevier.

In a sub- or supercritical condition, water is a rather strong oxidant.6,21 However, the presence of scCO2 increases the acidity of the reaction media, promoting the β-O-4 scission as a homogeneous acidic catalyst.

Figure 5. Py−GC−MS data for the solid phase obtained for raw alkali lignin and after its exposure to the sub- or supercritical conditions in the presence of scCO2 at 250 and 350 °C.

■

CONCLUSION The effect of the temperature at a constant pressure, short residence time, and fixed amount of alkali lignin on the selectivity and yield of the phenolic products from the hydrothermal reforming was investigated in a mixture of suband supercritical fluids (H2O−CO2 or H2O−N2). Dependent upon the hydrothermal treatment conditions, the lignin samples produce different phenolic compounds, such as guaiacol and its homologues, homovanillic acid, quaiacyl carbonyls, guaiacyl dimers, phenol, and cresols, as shown by GC−MS analysis. Verifying this information, the TCA of the liquid phases from the pressure vessel without LLE demonstrated the increase in the volatile compound concentration with the temperature and dependence upon the presence of scCO2. The increase of the volatile fraction of the liquid phase at a higher temperature occurs at the expense of the recalcitrant pyrolyzed carbon or “char”. The phenolic organic products of the alkali lignin degradation in the presence of a H2O−CO2 fluid mixture are significantly different in comparison to those obtained in H2O− N2 at 300 °C, apparently as a result of a catalytic effect of CO2 as an acidic catalyst in a supercritical state. At this temperature and pressure, a higher abundance of the guaiacol derivatives is observed in the liquid phase. Similarly, the TCA confirms an increased content of the volatile fraction in the scCO2-treated samples. Finally, the solid phases obtained with scCO2 at higher temperatures show a significant content of phenolics, which were not detected in either raw lignin or the solid products at lower temperatures. The effect of CO2 versus N2 pressurization evaluated for the first time should have significant implications regarding the selective synthesis of phenolic compounds and their use in the synthesis of polymers with desirable properties.

These data show that phenolic products make up the solid phase formed as a result of lignin depolymerization; therefore, the solid-phase product can be used as an additional source of valuable phenolics. However, a more detailed analysis demonstrates that the composition of these solid-phase products is similar to the original lignin only at 250 °C, whereas the solid phase obtained at 350 °C shows a significant increase in the content of both phenols and benzenediols at the expense of guaiacols, guaiacol carbonyls, and guaiacyl dimers (Figure 5). Representative phenolic compounds are cresol and creosol derivatives (the details are provided in Supplementary Table 2 of the Supporting Information), which amounts increase with the treatment temperature. Thus, the solid phase produced in the presence of scCO2 is significantly different in its monomer composition from the liquid-phase monomeric products (Supplementary Table 1 of the Supporting Information). Pathway for CO2 Catalysis. Despite the lower TOC yield, the use of 300 °C rather than 400 °C as a treatment temperature is more economically feasible considering the energy requirements. Given that the lignin treatment at this temperature appears to be efficient and also most selective only in the presence of scCO2, a potential explanation of its role may be as follows. As an aromatic heteropolymer, lignin in the process of hydrothermal treatment forms various phenols and methoxyphenols through the hydrolysis of the ether β-O-4 bond.1,7 Degradation of lignin can proceed further through hydrolysis of methoxy groups, without compromising the stability of the benzene ring itself (Figure 6).21 F

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

■

biomass from a phytoremediation trial in sub- and supercritical water conditions. Biomass Bioenergy 2011, 35, 872−883. (14) Gosselink, R. J.; Teunissen, W.; Van Dam, J. E.; De Jong, E.; Gellerstedt, G.; Scott, E. L.; et al. Lignin depolymerisation in supercritical carbon dioxide/acetone/water fluid for the production of aromatic chemicals. Bioresour. Technol. 2012, 106, 173−177. (15) Yong, T. L.-K.; Matsumura, Y. Kinetic analysis of lignin hydrothermal conversion in sub- and supercritical water. Ind. Eng. Chem. Res. 2013, 52, 5626−5639. (16) Yong, T. L.-K.; Matsumura, Y. Reaction kinetics of the lignin conversion in supercritical water. Ind. Eng. Chem. Res. 2012, 51, 11975−11988. (17) Tabasinejad, F.; Moore, R. G.; Mehta, S. A.; Van Fraassen, K. C.; Barzin, Y.; Rushing, J. A.; et al. Water solubility in supercritical methane, nitrogen, and carbon dioxide: Measurement and modeling from 422 to 483 K and pressures from 3.6 to 134 MPa. Ind. Eng. Chem. Res. 2011, 50, 4029−4041. (18) Brebu, M.; Vasile, C. Thermal degradation of ligninA review. Cellul. Chem. Technol. 2010, 44, 353. (19) Jessop, P. G. Applications of CO2 in Homogeneous Catalysis. Prepr. Pap.Am. Chem. Soc., Div. Fuel Chem. 2004, 49, 1. (20) Ö ztürk, I.̇ ; Irmak, S.; Hesenov, A.; Erbatur, O. Hydrolysis of kenaf (Hibiscus cannabinus L.) stems by catalytical thermal treatment in subcritical water. Biomass Bioenergy 2010, 34, 1578−1585. (21) Toor, S. S.; Rosendahl, L.; Rudolf, A. Hydrothermal liquefaction of biomass: A review of subcritical water technologies. Energy 2011, 36, 2328−2342.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.energyfuels.5b02136. Standards used for quantification of guaiacol derivatives (Supplementary Table 1) and tentatively identified lignin-related derivatives in a solid residue following Py−GC−MS (Supplementary Table 2) (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study is supported by the National Science Foundation Experimental Program to Stimulate Competitive Research (EPSCoR) Track II “Dakota BioCon” Project for North Dakota and South Dakota (Grants IIA-1330840 and IIA1330842).

■

REFERENCES

(1) Sturgeon, M. R.; Kim, S.; Lawrence, K.; Paton, R. S.; Chmely, S. C.; Nimlos, M.; Foust, T. D.; Beckham, G. T. A mechanistic investigation of acid-catalyzed cleavage of aryl-ether linkages: Implications for lignin depolymerization in acidic environments. ACS Sustainable Chem. Eng. 2014, 2, 472−485. (2) Vanholme, R.; Demedts, B.; Morreel, K.; Ralph, J.; Boerjan, W. Lignin biosynthesis and structure. Plant Physiol. 2010, 153, 895−905. (3) Ning, X.; Ishida, H. Phenolic materials via ring-opening polymerization: Synthesis and characterization of bisphenol-A based benzoxazines and their polymers. J. Polym. Sci., Part A: Polym. Chem. 1994, 32, 1121−1129. (4) Firdaus, M.; Meier, M. A. Renewable co-polymers derived from vanillin and fatty acid derivatives. Eur. Polym. J. 2013, 49, 156−166. (5) Wang, C.; Sun, J.; Liu, X.; Sudo, A.; Endo, T. Synthesis and copolymerization of fully bio-based benzoxazines from guaiacol, furfurylamine and stearylamine. Green Chem. 2012, 14, 2799−2806. (6) Kang, S.; Li, X.; Fan, J.; Chang, J. Hydrothermal conversion of lignin: A review. Renewable Sustainable Energy Rev. 2013, 27, 546−558. (7) Lange, H.; Decina, S.; Crestini, C. Oxidative upgrade of lignin− Recent routes reviewed. Eur. Polym. J. 2013, 49, 1151−1173. (8) Long, J.; Xu, Y.; Wang, T.; Yuan, Z.; Shu, R.; Zhang, Q.; et al. Efficient base-catalyzed decomposition and in situ hydrogenolysis process for lignin depolymerization and char elimination. Appl. Energy 2015, 141, 70−79. (9) Konnerth, H.; Zhang, J.; Ma, D.; Prechtl, M. H.; Yan, N. Base promoted hydrogenolysis of lignin model compounds and organosolv lignin over metal catalysts in water. Chem. Eng. Sci. 2015, 123, 155− 163. (10) Saisu, M.; Sato, T.; Watanabe, M.; Adschiri, T.; Arai, K. Conversion of lignin with supercritical water-phenol mixtures. Energy Fuels 2003, 17, 922−928. (11) Diono, W.; Sasaki, M.; Goto, M. Decomposition of Lignin Alkaline and Chemicals Recovery in Sub- and Supercritical Water; International Society for Advancement of Supercritical Fluids (ISASF): Nancy, France, 2004. (12) Qian, Y.; Zuo, C.; Tan, J.; He, J. Structural analysis of bio-oils from sub- and supercritical water liquefaction of woody biomass. Energy 2007, 32, 196−202. (13) Carrier, M.; Loppinet-Serani, A.; Absalon, C.; Marias, F.; Aymonier, C.; Mench, M. Conversion of fern (Pteris vittata L.) G

DOI: 10.1021/acs.energyfuels.5b02136 Energy Fuels XXXX, XXX, XXX−XXX