Hydrocracking of Organosolv Lignin in Subcritical Water to Useful

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Hydrocracking of organosolv lignin in subcritical water to useful phenols employing various Raney nickel catalysts Jan Ole Strüven, and Dietrich Meier ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.6b00342 • Publication Date (Web): 17 May 2016 Downloaded from http://pubs.acs.org on May 21, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Sustainable Chemistry & Engineering is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sustainable Chemistry & Engineering

Hydrocracking of organosolv lignin in subcritical water to useful phenols employing various Raney nickel catalysts Jan O. Strüven* and Dietrich Meier Thünen Institute of Wood Research, Leuschnerstraße 91b, 21031 Hamburg, Germany Email address: [email protected]

KEYWORDS: Lignin conversion, Hydrothermal processes, Hydrocracking, Fractionation, Bio-chemicals, Biorefineries.

ABSTRACT: Hydrothermal conversion of organosolv lignin has been conducted under reductive hydrogen atmosphere near the critical point of water. Several Raney nickel catalysts were tested with respect to degree of liquefaction, yields of phenols, and influence on the product distribution. The application of Raney nickel catalysts is crucial for the hydrothermal liquefaction of lignin, since the liquefaction rate increased approximately 2.5 times compared to hydrothermal conversion without catalysts. In addition Raney nickel caused complete demethoxylation of guaiacyl and syringyl moieties resulting in an oil composed mainly of six highly reactive simple phenols (phenol and mono-alkylated derivatives). Depending on the catalyst type yields of these phenols were 1.3-8.0 wt % (≈ 2.6-16 % theoretical yield). Flash pressure release with subsequent condensation of the hot vapors proved to be a fast and easy

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

approach to obtain a liquid fraction with a high content (25 %) of phenol and para- or orthoalkylated derivatives, which might be used in resin formulations.

INTRODUCTION Due to the depletion of fossil resources renewable feedstocks have become into focus of research. Especially lignocellulosic feedstocks (LCF) are an important renewable resource as they are abundant and not in competition with the food supply chain, e.g. corn. LCFs are often separated into their main constituents by pulping processes for paper production or biorefinery purposes. LCFs consists of three primary polymers: cellulose, a linear glucose polymer; hemicelluloses, branched sugar polymers of mainly pentoses; and lignin, a randomly branched polymer of phenyl-propane units1. The upcoming concepts of biorefineries include full material utilization of LCFs. This may include the production of (1) monomeric molecules which could be used as fine and bulk chemicals, (2) oligomeric compounds like fibers for composite materials, and (3) the usage of sugars and lipids for fuel generation like ethanol and biodiesel 2. Most of these LCFbiorefineries will deliver lignin enriched streams by either decomposing and extracting carbohydrates or using pretreatment and pulping processes for extraction of lignin prior to enzymatic carbohydrate hydrolysis 3. These separated main components can be subjected to subsequent processing steps. While technologies for converting carbohydrates to valuable materials such as pulp or ethanol are well established, lignin degradation and conversion processes, however, still need further development and optimization. Lignin contributes about 30 wt% and 40 % to the energy content of LCF. Nowadays, lignin is usually burned in recovery boilers of pulping processes to generate power and steam. But future biorefineries will generate significant amounts of lignin which will be available for manufacture of valueadded products. The valorization of bioproducts - especially from lignin as a byproduct needs to be integrated into the biorefinery concepts to improve their overall profitability 4. M. 2 ACS Paragon Plus Environment

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Haase concluded that the economic feasibility of LCF-biorefineries depends strongly on the market price of lignin, which has to be at least double the revenue of the lignin’s heating value 5. Thus, conversion processes of lignin to value-added products will play a key role for the economic feasibility of biorefineries. Due to the phenolic nature of lignin it is obvious to degrade its oligomeric structure while preserving the intrinsic functional hydroxyl groups to generate reactive monomeric phenols, dihydroxybenzenes and other aromatics. As lignin is the only abundant, natural renewable resource with aromatic character it could play a key role as feedstock for phenolic and aromatic chemicals. The lignin’s intermolecular bondings consist mainly of ether-bonds or C-C linkages. Scission of these bonds can be achieved by hydrolysis and thermolysis, respectively. Hence, thermochemical and especially hydrothermal conversion of lignin for the production of aromatics seems promising 6. The liquefaction of lignin in water near the critical point (374.12 °C, 22.12 MPa) has several advantages: (1) Water changes its properties in the critical region and becomes a good solvent for organic components, (2) The ionic product of water increases, thus water can react as acid as well as base or favors at least reactions which are typically catalyzed by them 7. (3) Water can be considered as green solvent

8

as it is not toxic neither it is flammable and readily available.

Furthermore, the unambiguous allocation of organic reaction products is facilitated compared to other organic solvents which may react as well. Therefore, in recent years the conversion of biomass and lignin in sub- and supercritical water has become into focus of several studies 9-13

. A review on hydrothermal lignin conversion is given by Kang et al. 14.

First attempts at hydrogenation of lignin in aqueous solutions with the use of Raney nickel catalysts were done in 1940

15

. Hydrogenation was performed at 225-250 °C and 10.0 to

17.5 MPa hydrogen pressure. The lignin was suspended in alkalized water together with a Raney nickel catalyst. The resulting resinous water insoluble product was distilled into three low molecular weight fractions containing cyclic hydrocarbons and alcohols as well as a



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

residue which was not further characterized. Unfortunately, no yields have been reported. The water fraction was distilled as well to obtain methanol. It has to be mentioned that the terms hydrothermal and solvolysis conversion are not always clearly defined, since conversion with mixtures of water and organic solvent, e.g. ethanol or phenols were referred to as hydrothermal as well. Jiang and co-workers9 demonstrated the conversion of alkali lignin in subcritical water and determined the influence of temperature (220-340 °C), residence time (0-60 min) and amount of ethanol as co-solvent (0-100 %). The optimal condition in terms of conversion was 30 min, 310 °C and 25 % ethanol. Main products under these conditions were guaiacol (1.06 %), 2,6dimethoxyphenol (0.92 %) and phenol (0.27 %). Demethoxylation seemed to occur to a minor extent. Hydrothermal liquefaction of lignin was studied mostly without catalyst or with homogenous catalysts such as different acids and bases 16. However, only a few investigations were performed with heterogeneous catalysts

17-21

. It was found, that especially Raney nickel

catalysts seems to support the conversion of catechol to phenol, which will improve the overall product distribution17. Bembenic et al.22 concluded that adding different gases like nitrogen, hydrogen and carbon monoxide to the hydrothermal reaction of lignin in subcritical water (365 °C) had a positiv impact on the conversion rate. They also confirmed the presence of methanol in the aqueous phase. Methanol formation is due to the loss of methoxy groups. Methanol formation may also help to suppress the hydrogenation of aromatic rings on a nickel based hydrogenation catalyst

23

.In this study the generation of phenolics via a combination of hydrothermal

liquefaction and catalytic hydrodeoxygenation processes under hydrogen atmosphere is presented. We feel that radicals formed by thermal cleavage of lignin should be saturated as quickly as possible by hydrogenation reactions to prevent further recondensation reactions. Phenolics are versatile platform chemicals for the future bio-based economy. The generation 4 ACS Paragon Plus Environment

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of phenolics from lignin is desirable since lignin is the only abundant renewable resource with phenolic structure.

EXPERIMENTAL SECTION Chemicals. Beech wood organosolv-lignin, was procured from Fraunhofer CBP, Leuna, Germany, who operates a pilot plant on ethanol-water pulping. The pulping conditions were as follows: Cooking liquids were ethanol and water (1:1 vol. basis), liquid to wood ratio 3.2:1, 70 Kg beech wood, cooking time 90 min, temperature 170 °C, 0.5 wt % H2SO4 based on dry wood. Lignin was precipitated with twofold excess of water without further acidification and separated with a filter press. This lignin was used for all catalyst screening experiments. CHNS-analysis gave 62.1 % C, 6.2 % H, 31.4 % O, 0.3 % N, 0.0 % S. Oxygen was calculated by difference. Sugar and ash content was 2.35 % and 0.07 %, respectively. The aromatic and aliphatic hydroxyl group content was 1.95 and 3.11 mmol/g, respectively. For experiments to determine the influence of hydrogen an organosolv lignin from own pulping experiments was used. The pulping conditions were as follows: Cooking liquids were ethanol and water (1:1 vol. basis), liquid to wood ratio 4:1, cooking time 90 min, temperature 170 °C, 0.7 wt % H2SO4 based on dry wood. After cooking the extract was acidified with sulfuric acid to pH=2 and lignin precipitated with twofold excess of water. CHNS-analysis gave 62.6 % C, 6.1 % H, 31.0 % O, 0.3 % N, 0.0 % S. Oxygen was calculated by difference. Sugar and ash content was 2.15 % and 0.06 %, respectively. The aromatic and aliphatic hydroxyl group content was 2.09 and 2.24 mmol/g, respectively. Dichloromethane was procured from Fischer Scientific in analytical reagent grade and was used as delivered. Acetone was procured from BCD Chemie in analytical grade and was distilled before use to avoid impurities of diacetone alcohol. All Raney nickel catalysts were procured from Sigma Aldrich. The A-7000 catalyst was obtained from Johnson Matthey.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

Acetic anhydride was procured from Sigma Aldrich in 99.0 % purity. Anhydrous pyridine was procured from VWR in high grade and used as received. Catalytic

hydrothermal

liquefaction

experiments.

Hydrothermal

liquefaction

experiments were performed in a 250 ml batch autoclave equipped with a high speed magnetically coupled mixer (MagneDrive®) connected to a DispersimaxTM turbine (Autoclave Engineers, Burton Coblin, France). It provides high speed radial flow stirring, while drawing head space gas down a hollow shaft and dispersing the gas through the impeller ports, for effective entrainment of the gas into the liquid and to minimize gas transfer limitations. The reactor was heated with an electric heating jacket. Temperature and pressure were monitored online and logged on a PC. A 700 ml chilled condenser was connected to the autoclave exit port. To avoid distillation of products and water during the reaction time a needle valve was installed between the autoclave and the condenser (Scheme 1). Scheme 1. Experimentel setup

Typically, the reactor was subsequently loaded with 25 g of lignin, 3 g (dry matter) of catalyst and 50 ml of deionized water. The system was purged at least three times with argon to obtain an oxygen free atmosphere and then pressurized to 6 MPa for a first leak test. Afterwards, the system was purged three times with hydrogen and pressurized to 20 MPa for a second leak test. Subsequently, the pressure was released to 7 MPa and the stirrer velocity was set to 1800 rpm. The reactor was then heated to the desired final temperature (360 °C) with an average heating rate of 7 °C min—1. Reaction time was set to zero when the target 6 ACS Paragon Plus Environment

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


temperature was reached. After reaching the desired reaction time (180 min) the pressure was released by opening the valve to the condenser in order to fractionate valuable monomers from the oligomeric residue. After 20 min, permanent gases were released from the condenser to ensure that (except for permanent gases) all the products were condensed. The separated fraction was removed, filtered and extracted with DCM in a separating funnel until colorless. The reactor heating was switched off. The reactor was purged by a syringe pump (ISCO 260D) with 100 ml of DCM after reaching a temperature of approximately 200 °C. DCM was immediately in the vapor phase and transferred to the separator for the purpose of full coverage of all components, which may retain in the pipe and the separator. This DCM solution was added to the first fraction. The combined fraction is referred to as light oil after removing DCM on a rotary evaporator. A heavy oil fraction was obtained by extracting the residue with 150 ml of DCM at 100 °C directly in the autoclave. The reactor was cooled down overnight. The content of the reactor was filtered and washed with DCM. The DCM soluble components were referred to as heavy oil. The remaining residue in the filter was extracted with acetone overnight in a Soxhlet apparatus. The extraction was carried out until colorless, but at least 24 hours. The extraction thimble was dried until constant weight. The acetone soluble fraction was referred to as tar. The solid residue is referred to as coke. An overview of the work up procedure is depicted in Scheme 2



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

Scheme 2. Work up procedure

Product Definition and Calculations. All obtained yields were calculated on moisture and ash free (maf) intake of lignin. The dry amount of catalyst was calculated on maf lignin as well. Raney nickel catalysts were stored under water to avoid oxidation and deactivation, thus the weighing of catalyst was performed in a wet state. The dry content of the catalyst sample weight was measured five times and an average content of 70.74 wt% with rel.std.dev. (RSD) of 1.46 % was calculated. The coke amount was calculated as acetone-insoluble minus dry catalyst intake. Analysis. GC/MS-measurements were performed on an Agilent 6890N / 5975B MSD equipped with VF-1701ms column 60 m x 0.25 µm (Agilent) and FID for quantification. Fluoranthene was used as internal standard. For the purpose of GC-measurements an aliquot was taken from the DCM-solution before removing the solvent on a rotary evaporator to avoid losses of volatile components. For quantitative GC analyses of catechols, these components need to be derivatized. In this study acetylation was chosen to ease volatilization, improve the separation and the detector response. Acetylation was done as follows: 40 mg of analyte was dissolved in 500 µl pyridine and 500 µl of acetic anhydride were added to the solution. Acetylation reaction proceeds 8 ACS Paragon Plus Environment

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


completely within 24 h at room temperature and permanent shaking. The internal standard was added right bevor measurement. GPC was performed on an Agilent 1100 series equipped with PolarGel-L Guard 50 x 7.5 mm and 2 x Varian PolarGel-L; each 300 mm, I. D. 7.5 mm using dimethyl sulfoxide with 0.1 wt % LiBr as eluent. 100 µl of solutions containing 2 mg ml-1 of analyte were injected. The system was calibrated using polyethyleneglycol (PEG) standards (194 to 21,030 g/mol). Ash content was determined according to TAPPI-standard (T211 om-02). Sugars were determined according to Lorenz et al.24 Reproducibility was determined for one triplicate experiment under standard conditions: Raney 3202, 7 MPa cold initial hydrogen pressure and 180 min reaction time at 360 °C. Rel. std. dev. were 1.87, 1.89, 17.49 and 17.35 % for light oil, heavy oil, tar, and coke fractions, respectively. The poorer std. dev. in tar and coke can be attributed to uncontrolled condensation reaction in the autoclave during the work up procedure and to uncertainties in the initial weight of catalyst.

RESULTS AND DISCUSSION Catalysts Screening. The use of catalysts for the thermochemical conversion of lignin to phenols plays an essential role for the yields, product distribution and selectivity of the process. The yield of monomeric components as well as mass balances for solids, tars, heavy and light oils from organosolv lignin in water near its critical point were determined. Gas composition and water soluble components were considered for an experiment using the most promising catalyst with respect to formation of phenols. The selectivity of the hydrothermal liquefaction of organosolv lignin towards phenols can be improved by the use of Raney Nickel catalysts 17. In this study organosolv lignin from ethanol-water pulping was chosen for several reasons: (1) The lignin is sulfur-free, thus no catalyst poisoning by sulfur occurs and



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

(2) this type of lignin has less condensed phenolic structures retaining much of its ß-O-4 interunit linkages and has less ash content compared to lignin from Kraft pulping 25. To calculate the C-balance, composition and volume of the gas phase as well as the TOC content of the aqueous phase were determined based on the standard experiment with Raney 2400, since this catalyst was the most promising in terms of phenol formation. Analyses were performed in triplicate. The gas composition was as follows: Carbon dioxide 51.47 % (RSD 2.8 %), methane 45.56 % (RSD 2.75 %), ethane 1.32 % (RSD 2.6 %), propane 0.69 %(RSD 2.4 %), n-butane 0.36 % (RSD 2.4 %), n-pentane 0.19 % (RSD 3.2 %), carbon monoxide 0.17 % (RSD 0.8 %), isobutane 0.11 % (RSD 2.5 %), propene 0.08 % (RSD 124.5 %) , n-hexane 0.03 % (RSD 173.2 %), and ethene 0.02 % (RSD 50 %) representing 35.6 wt % or 29.5 C % based on maf lignin. Complete C-balance is presented in Table 1. Table 1. C-balance of standard experiment conducted with Raney 2400

Fraction Light oil Heavy Oil Tar Coke Gas Water Total

C% 38.8 21.7 3.3 6.3 29.5 1.3 100.8

Figure 1 displays the yields of solid and liquid fractions. Hydrothermal degradation of organosolv lignin by hydrolysis will play a key role due to the abundance of ether linkages. However, without catalysts the process resulted in 47.5 wt % coke formation (see Figure 1).


Page 11 of 29

Coke Yield wt% maf lignin

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Tar

Heavy oil

Light oil

70 60 50 40 30 20 10 0

Figure 1. Yields (wt%) of the fractions coke, tar, heavy oil, and light oil. The introduction of Raney nickel catalysts into the process improved the overall liquefaction of organosolv lignin considerably, regardless the type of catalyst. The yields of coke for Raney 4200, 3202, 2800, A-7000, and 2400 were 4.8 wt %, 1.9 wt %, 5.2 wt %, 5.8 wt %, and 0.65 wt %, respectively. The coke yield was significantly lowered by any Raney nickel used in this study, but no statistical significant differences between the catalysts can be deduced. The lowered coke yields could be due to rapid hydrogenation of reactive components prone to repolymerisation or the possible ability of Raney catalysts to liquefy and/or gasify the formed coke as mentioned by Forchheim et al.17. The yields of the target fraction “light oil” can be increased by Raney nickel as well. It can be seen in Figure 1 that the yield of light oil at least doubles compared to the conversion without catalysts. Without catalyst

the

yield

of

light

oil

was

11.8 wt %

whereas

26.7 wt %,

27.2 wt %,

28.8 wt%, 29.4 wt %, and 30.2 wt% were obtained with use of Raney 4200, 3202, 2800, A7000, and 2400, respectively. The yields of tar were about 4 wt % for all catalysts with the exception of Raney 3202 and 2400 with yields of 5.2 wt % and 2.5 wt %, respectively. The highest yields (52 wt %) in terms of soluble products (acetone and DCM) were achieved with the sponge type nickel catalyst A-7000. All other tested Raney nickel catalysts showed 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

slightly lower yields (44 wt % - 47 wt %). The catalysts specifications according to the supplier are shown in Table 2. Elemental composition of catalysts A-7000 is almost equal to Raney 2800 with the exception of a low iron content of the latter, thus presumably, the surface of the A-7000 is decisive for the higher yield. The oil yields were affected to a small extent by the type of Raney Ni. Nevertheless, beside the A-7000 highest yields of light as well as heavy oil were achieved with Raney 2400 under these conditions. With respect to the reproducibility for all experiments tar and coke yields were at the same level. However, also best performance with respect to low tar and coke yields was obtained with Raney 2400 (2.51 wt %, 0.65 wt %, respectively). Table 2: Catalyst specifications according to the supplier and Wu et al.26 Raney

Sponge Metal

2400

3202

4200

2800

A-7000

Nickel [%]

>81.0

>92.0

>93.0

>89.0

≥90

Aluminum [%]

6.0 - 13.0

Hydrocracking of Organosolv Lignin in Subcritical Water to Useful

Recommend Documents