Supporting Online Material for

Transcription

1 Supporting nline Material for Renewable Chemical Commodity Feedstocks from Integrated Catalytic Processing of Pyrolysis ils Tushar P. Vispute, Huiyan Zhang, Aimaro Sanna, Rui Xiao, George W. Huber This PDF file includes: Materials and Methods Figs. S1 and S2 Tables S1 to S8 Published 26 November 2010, Science 330, 1222 (2010) DI: /science

2 Supporting nline Material Materials and Methods Materials Two different biooils were used in this study: Pine Wood Biooil (PWB) and DE Biooil (DEB). PWB was obtained from Mississippi State University and was produced by pyrolysis of dry pine wood in an auger reactor. DEB was supplied by the US Department of Energy (DE) and was manufactured by National Renewable Energy Laboratory (NREL), Golden, Colorado using the Thermochemical Process Development Unit (TCPDU) from white oak pellets. PWB was used in the studies involving water soluble fraction of biooil. DEB was used to study the zeolite upgrading of biooil and hydrogenated biooil. DEB was microfiltered prior to the hydrogenation and zeolite upgrading to remove fine char particles. Membralox TI70 microfiltration membranes with nominal pore size of 0.8 μm (from Pall Fluid Dynamics, De Land, Florida) were used to filter the biooil. The 5 wt% Ru/C and 5 wt% Pt/C catalysts used in this study were obtained from Strem Chemicals (Product No and respectively). The catalysts came in wet form (ca. 50% water) and were dried in an oven at 373 K for 4 hours before using in the reactions. The zeolite upgrading was done using HZSM5 catalyst obtained from Zeolyst (CBV 3024E, Si 2 /Al 2 3 = 30). All the pure compounds used in the zeolite upgrading were > 99% pure (from Fisher Scientific) and were used without further purification. Water extraction of biooil and water content analysis The separation of biooil in water soluble biooil and water insoluble biooil was carried out by addition of water to it. Biooil readily phase separated upon addition of water. ne part biooil was weighed in a centrifuge tube and four parts distilled water was added to it. Typically 28 gm distilled water was added to 7 gm of PWB. The contents of the centrifuge tube were mixed vigorously by shaking so as to extract all the water soluble compounds in aqueous fraction. The mixture was then separated in two layers in a centrifuge (Marathon 2, Fisher Scientific) at 00 rpm for 20 minutes. After centrifugation the top layer was decanted and called Water Soluble Biooil (WSB). The viscous bottom layer was called Water Insoluble Biooil (WIB). The WSB layer was weighed to determine the amount of biooil that dissolved in water. It was assumed that no externally added water would go into WIB during the extraction process. The resulting WSB solution was ca wt% water soluble biooil components in water. The water content of DEB and LTHDEB was measured using V20 volumetric KarlFischer titrator by MettlerToledo. 1

3 Single stage hydrogenation of water soluble biooil (WSB) & DE biooil Single stage hydrogenation of WSB and DEB was carried out at 398 K and 52 bar or bar total pressure. A continuous trickle bed reactor with both liquid and hydrogen gas moving in downward direction was used. Dry catalyst (Ru/C) was packed in a stainless steel tubular reactor with glass wool plugs on both the sides. Typical reactor was 30 cm long and with 6.35 mm outer diameter. The reactor was heated with a Lindberg Blue M tube furnace (Model No. TF55035A 1) and temperature was maintained with an inbuilt temperature controller on the furnace. The catalyst was reduced in situ prior to the reaction in flowing hydrogen (150 cm 3 min 1 ) with following temperature regime: room temperature to 533 K at 0.5 K min 1 and then hold at 533 K for 2 hours. Hydrogen flow through the reactor was maintained by a Brooks mass flow controller. After the reduction is complete, the back pressure regulator was set at the desired value and the whole reactor system was pressurized to the desired reaction pressure and heated to the desired reaction temperature. The liquid feed was then started with flow rates of 0.03 cm 3 min 1 for DEB and 0.08 cm 3 min 1 for WSB. Eldex Laboratories 1SM HPLC pump was used to feed the liquid. The reactor effluent and excess hydrogen flow to a 150 ml stainless steel gasliquid separator where liquid accumulates and gas continues to flow to the back pressure regulator. nline Agilent Gas Chromatograph (GC) (Model 7890A) was used to analyze the reactor effluent gas. The effluent gas was analyzed using a Flame Ionization Detector (FID) maintained at 573 K. Restek RTQBND (Catalog No ) column was used with helium carrier gas flowing at 4.24 ml min 1. Column oven temperature regime used was as follows: hold at 308 K for 5 min, ramp to 423 K at 5 K min 1 and hold at 423 K for 15 min. The liquid product was drained periodically to be analyzed by the GCFID and High Performance Liquid Chromatography (HPLC). At least 3 liquid samples were collected at a particular set of operating conditions to ensure the steady state. The product (and feed) compositions were measured with an Agilent GC (model 7890A) and a Shimadzu HPLC system. Flame ionization detector was used on the GC to quantify all the reactants and products except sugars, sugar alcohols and levoglucosan. The reactants and products were also verified by GCmass spectroscopy. Restek RtxVMS (Catalog No ) column was used with constant flow rate of 1.24 cm 3 min 1. Helium was used as the carrier gas. Injector and detector were both held at 513 K. The GC oven was programmed with the following temperature regime: hold at 313 K for 5 min, ramp to 513 K at 7.5 K min 1 and hold at 513 K for 15 min. n HPLC, RI detector (held at 303 K) was used to quantify sugars, sugar alcohol and levoglucosan in the feed and products. BioRad s Aminex HPX87H column (Catalog No ) was used with M H 2 S 4 as the mobile phase with the flow rate of 1 ml min 1. The column oven temperature was held constant at 303 K. The carbon content in feed and product for WSB and its products was determined by Total rganic Carbon (TC) analysis. The carbon content in DEB and its products was determined by elemental analysis. Weight Hourly Space Velocity (WHSV) used for WSB was 3 hour 1. WHSV of 1.6 hour 1 was used for DE 2

4 B. WHSV was calculated by dividing the feed (including externally added water for WSB) flow rate in gm hour 1 by weight of catalyst in gm. Two stage hydrogenation of water soluble biooil For the two stage hydrogenation a 2 nd tubular reactor (30 cm length, 6.35 mm outer diameter) was added in series after the first hydrogenation reactor. Dry Pt/C was used in this reactor and the catalyst was reduced prior to the reaction with the same temperature regime as that used for Ru/C catalyst. Same amount of catalyst was loaded in both the reactors. Two stage hydrogenation of WSB was carried out at 398 K (1 st stage), 523 K (2 nd stage) and bar total pressure. While operating with 2 reactors, the first reactor was heated using a heating tape and the second reactor was heated using the tubular furnace. A type K thermocouple was placed next to the reactor wall and reactor temperature was controlled at 398 K using an mega temperature controller. The products were analyzed as described in the previous section. We did not see any sign of reactor plugging in single stage or two stage hydrogenation of WSB for 5 days, implying that catalysts are stable. During these 5 days, the reactor was not run at one particular operating condition though. Either the space velocity or temperature was changed to study the effect of these parameters on reactant conversion and product selectivity. Zeolite upgrading of biooil, water soluble biooil and hydrogenated water soluble biooil Prior to loading in the reactor, the ZSM5 catalyst was sieved to µm size. A quartz tube (1.27 cm outer diameter) was packed with a quartz wool plug. Quartz beads (700mg, µm particle size) were placed on the quartz wool plug to act as a catalyst bed support. The sieved ZSM5 catalyst (26 mg typically) was then loaded in the reactor. The catalyst was then calcined in situ in flowing air (60 ml min 1, dehumidified by passing through a drierite tube) at 873 K for 6 hours. The reactor temperature was measured using a type K thermocouple inserted into the catalyst bed. Reactor tube was heated using a Lindberg tubular furnace and temperature was controlled using an mega temperature controller. nce the calcination was complete, helium carrier gas flow was started over the catalyst at 204 ml min 1. The catalyst was maintained at reaction temperature. All the experiments were carried out at atmospheric pressure and no significant pressure drop was observed across the catalyst bed. The liquid feed was then started at 2.7 ml hour 1 for WSB, low temperature hydrogenated WSB and high temperature hydrogenated WSB which corresponds to the WHSV of 11.7 hour 1 on the biooil content basis (excluding the added water). The liquid feed rate of 0.34 ml hour 1 (0.06 ml hour 1 for furan) was used for DEB and low temperature hydrogenated DEB, which corresponds to the WHSV of 11.7 hour 1 (1.97 hour 1 for furan). Liquid was pumped using a syringe pump (KD Scientific, Model No. 780). For pure compounds (except furan), 12.5 wt% solution in water was used as feed. This is to keep the partial pressure of water the same for pure compounds and WSB (and its hydrogenated products). Pure furan was used as feed as it is water insoluble. 3

5 The reactor effluent is carried by the helium carrier gas to an icewater cooled condenser where heavy products are condensed. The effluent gas was then collected in a gas bag. The heavy products in the condenser were collected by washing the condenser with 10 cm 3 of ethanol. Liquid product was analyzed by an Agilent 7890A GCFID system with an Agilent capillary column (Catalog No J413). Helium was used as carrier gas with the FID detector maintained at 523 K. Following column temperature regime was used: hold at 313 K for 5 min, ramp to 523 K at 20 K min 1, and hold at 523 K for 20 min. The gaseous product was analyzed using a Shimadzu 2014 GC system. Restek RtxVMS capillary column (Catalog No ) was used to quantify aromatic hydrocarbons (with FID detector) and HAYSEP D packed column from Supelco was used to analyze C and C 2 (with TCD detector). Both FID and TCD detectors were maintained at 513 K. Helium was used as the carrier gas. Following column temperature regime was used with both the columns: hold at 308 K for 5 min, ramp to 413 K at 5 K min 1, ramp to 503 K at 50 K min 1 and hold at 503 K for 8.2 min. The coke yield was measured by burning the coke and measuring the amount of C 2 produced. After the reaction is complete, dry air (60 cm 3 min 1 ) was flown over the spent catalyst (873 K) for 2 hours to burn off the coke formed during the reaction. The resulting effluent gas was then passed in series through a copper converter (to convert C to C 2 ), a moisture trap and a C 2 trap. Copper converter (Sigma Aldrich, Part No ) contained 13 wt% Cu on alumina catalyst and was operated at 523 K. The coke yield was determined from the difference in the mass of the fresh and spent C 2 adsorbent. Thermogravimetric analysis Thermogravimetric analysis (TGA) experiments were carried out with a SDT Q600 TGA system (TA Instruments). Ultrahighpurity helium (Airgas Company) was used as the sweep gas with a flow rate of cm 3 min 1. Approximately 15 mg of sample was loaded into an aluminum pan. An aluminum cap was placed on the sample crucible to avoid any vaporization of sample prior to starting the temperature ramp. The temperature of the sample was programmed from room temperature to 973 K at 1.5 K min 1, followed by an isothermal period of 30 min at 973 K. Gel Permeation Chromatography Gel Permeation Chromatography (GPC) experiments were carried out on a Shimadzu HPLC system with an UV detector (frequency 254 nm). Varian MesoPore column (Part No ) was used with stabilized tetrahydrofuran (THF) as mobile phase flowing at 0.5 cm 3 min 1. Samples for GPC were prepared by dissolving biooil in THF at 1wt% concentration. The biooil solution in THF was then filtered with 0.45 µm filter and used for GPC. The GPC column was standardized using polystyrene molecular weight standards in the range of 162 to Da. 4

6 Process block flow diagram Light olefins, C x Aromatics Biomass Water Char C x Fast Pyrolysis Reactor Biooil ption 1 Zeolite Upgrading ption 1a HZSM5 873 K ption 2c H 2 Pt/C K Light Alkanes (to steam reformer or to fuel furnace) H H High Temperature Hydrogenation H CH 3 H H H H H High Temperature Hydrogenated Water Soluble Biooil H H H Water LL Extractor ption 2 Water Insoluble Biooil to further processing ption 1b ption 2a ption 2b H 2 Water Soluble Biooil Ru/C 398 K Light Alkanes (to steam reformer or to fuel furnace) Fig. S1. Production of olefins, aromatic hydrocarbons, diols, and gasoline range alcohols from the integrated catalytic processing of pyrolysis oil. H H Low Temperature Hydrogenation H CH 3 H H H H H H H H H H Low Temperature Hydrogenated Water Soluble Biooil H H H H H H 5

7 Chemical analysis of DEB and low temperature hydrogenation product of DEB Following table depicts the detailed composition of DEB feed. Table S1 Composition of DEB feed Compound mmol carbon Classification min 1 Methyl acetate Hydroxyacetaldehyde Ester Aldehyde Acetic acid Acid Hydroxyacetone Ketone 2Furanone Ketone 3Methyl1, Ketone cyclopentadione Phenol Phenolic 1Hydroxy2butanone Ketone Furfural Aldehyde 2Cyclopenten1one γbutyrolactone Ketone Ketone 5Hydroxymethylfurfural Aldehyde Levoglucosan Sugar Sugars Sugar Methanol Alcohol Total carbon identified Total carbon as determined by elemental analysis Total carbon fed to the reactor was mmol min 1 and composition is given in mmolc min1 for each component. Fraction carbon contribution of each compound can be found by dividing mmol carbon min 1 for that compound by mmol carbon min 1. 6

8 Following table depicts the detailed composition of low temperature hydrogenation product of DEB (LTHDEB). Table S2 Composition of low temperature hydrogenated DEB Compound mmol carbon min 1 Methanol Methyl acetate Hydroxyacetaldehyde Acetic acid Hydroxyacetone Furfural Furanone Phenol Levoglucosan Sugars Sorbitol Total carbon identified Total carbon as determined by elemental analysis Low temperature hydrogenation of DEB was carried out over 5 wt% Ru/C catalyst at 398K, bar and at WHSV of 1.6 hour 1. See footnote of Table S1 for concentration units. 7

9 Gel Permeation Chromatography (GPC) of DE Biooil and low temperature hydrogenation product of DEB Following is the GPC chromatogram of DE Biooil and low temperature hydrogenated DE Biooil. 6E+05 Intensity 5E+05 3E+05 2E+05 0E Molecular Weight (Da) Fig. S2. Gel permeation chromatogram of DE biooil and low temperature hydrogenated DE biooil. Dotted line: DE biooil, solid line: low temperature hydrogenated DE biooil, Hydrogenation carried out at 398 K, bar and at the WHSV of 1.6 hour 1. 8

10 Chemical analysis of water soluble fraction of pine wood biooil (WSB) and single & twostage hydrogenation product of WSB Following table depicts the detailed composition of WSB feed. Table S3 Composition of WSB feed * Compound mmol Classification carbon L 1 Hydroxyacetaldehyde Aldehyde Acetic acid Acid Hydroxyacetone Ketone 2Furanone 37.6 Ketone Phenol 2.5 Phenolic 3Methyl1, Ketone cyclopentadione Guaiacol 10.3 Phenolic Catechol Phenolic 1Hydroxy2butanone 20.2 Ketone Furfural 20.9 Aldehyde 2Cyclopenten1one 21.9 Ketone 5Hydroxymethylfurfural 63.9 Aldehyde 4Methyl catechol 47.5 Phenolic Levoglucosan Sugar Sugars Sugar Methanol 24.4 Alcohol Total carbon Identified Total carbon as measured by TC * made by mixing 7 gm pine wood biooil with 28 gm water. The WSB has mmol carbon L 1, hence the carbon concentration of each component is given in mmol carbon L 1 from than compound in WSB. Fraction carbon contribution of each compound can be found by dividing mmol carbon L 1 for that compound by mmol carbon L 1. 9

11 Following table depicts the detailed composition of products of low temperature (single stage) and high temperature (2stage) hydrogenation of WSB. Table S4 Composition of the WSB hydrogenation products Low temperature hydrogenation reaction conditions: over 5 wt% Ru/C catalyst ; T: 398 K; P: 52 bar; WHSV: 3 hour 1, high temperature hydrogenation reaction conditions: first over 5wt% Ru/C catalyst (398 K) then over 5 wt% Pt/C catalyst (523 K); P: bar, WHSV: 3 hour 1 mmol carbon L 1 Compound LTH WSB* HTH WSB Pentane Hexane Acetic acid Levoglucosan Sugars Methanol Ethanol Propanol Tetrahydrofuran Butanol Methyltetrahydrofuran ,5Dimethyltetrahydrofuran Butanol Pentanol Pentanol Ethylene glycol Cyclopentanol Hexanol Propylene glycol ,3Butanediol Cyclohexanol ,2Butanediol Tetrahydrofurfuryl alcohol ,4Butanediol γbutyrolactone γvalerolactone Glycerol ,2Cyclohexanediol Hydroxymethylγbutyrolactone Sorbitol Methylcyclopentanol ,2,3Butanetriol ,4Pentanediol methylcyclohexanol methylcyclohexanol ,2Hexanediol ,2,6Hexanetriol Total carbon identified Total carbon in liquid as measured by TC

12 *low temperature hydrogenation product of WSB, high temperature hydrogenation product of WSB, see footnote of Table S3 for concentration units Detailed composition of zeolite upgrading products Following table depicts the detailed product composition obtained upon zeolite upgrading of water soluble biooil, its hydrogenated products, DE biooil, its hydrogenated product and pure biomass model compounds. Feed Table S5 Carbon yields (%) for HZSM5 upgrading of biomassderived feedstocks. Catalyst: HZSM5 (Si 2 /Al 2 3 = 30), Reaction temperature = 873 K, Helium carrier gas flow rate: 204 cm 3 min 1. WSB is water soluble fraction of the pine wood biooil, LTHWSB is low temperature hydrogenated WSB, HTHWSB is high temperature hydrogenated WSB, LTHDEB is low temperature hydrogenated DEB, THF is tetrahydrofuran. Carbon Yield (%) or Carbon Selectivity (%) WSB LTH WSB HTH WSB DE B LTH DE B Glucose Sorbitol Furan Glycerol THF Methanol wt% WHSV * (hour 1 ) H/C eff ratio C C Coke lefins Carbon Selectivity (%) Ethylene Propylene Butylene Aromatics Carbon Selectivity (%) Benzene Toluene Xylenes Ethylbenzene Styrene Indene Naphthalene Total identified carbon (%) * WHSV for WSB, LTHWSB, and HTHWSB is on the water soluble biooil content basis, excluding the water that is added externally in the feed. 11

13 Following table depicts the effect of temperature on the product yield in zeolite upgrading of high temperature hydrogenated water soluble biooil (HTHWSB). Table S6 Effect of temperature on the product carbon yields (%) for zeolite upgrading of HTH WSB on HZSM5. WHSV: 11.7 hour 1, Helium carrier gas flow rate: 204 cm 3 min 1, HTH WSB was obtained by 2stage hydrogenation of WSB. Hydrogenation reaction conditions, 1 st stage: 5 wt% Ru/C catalyst, 398K, bar, WHSV: 3 hour 1, 2 nd stage: 5 wt% Pt/C catalyst, 523K, bar, WHSV: 3 hour 1. Product Temperature (K) C C Coke Ethylene Propylene Butylene lefins Benzene Toluene Xylene Ethyl benzene Styrene Indene Naphthalene Aromatics Total identified carbon (%) Hydrogen to carbon effective ratio calculation The H/C effective ratio (H/C eff ) is defined as, 12 H 2 H / C eff = (S1) C Where H,, and C are moles of hydrogen, oxygen and carbon respectively, calculated from the elemental composition of the substance in consideration. The H/C eff ratios for DEB and low temperature hydrogenated DEB were calculated based on the elemental analysis done at

14 Galbraith Laboratories, Knoxville, Tennessee. The exact elemental composition cannot be determined for water soluble fraction of pine wood biooil (WSB) (and its hydrogenated products) as it contains a large amount of water, the amount of which could not be determined accurately. Hence the H/C eff ratio for WSB, low temperature hydrogenated WSB and high temperature hydrogenated WSB was estimated using their composition determined by GCFID and HPLC. A small error in the H/C eff ratio calculation will be introduced due to the unidentified carbon in WSB and its hydrogenated products. Homogeneous thermal coking Following table depicts the homogeneous coke yields from the thermogravimetric analysis of various biomass derived feedstocks. Table S7 Homogeneous coke yield for different feedstocks Feed H/C eff ratio Coke (%wt) Glucose DEB LTHDEB WSB Sorbitol Furan Glycerol LTHWSB HTHWSB Tetrahydrofuran Methanol Theoretical yield in zeolite upgrading In zeolite upgrading of an oxygenated hydrocarbon C6C8 aromatic hydrocarbons, C2C4 olefins, C, and water are the major products. xygen is removed from the feed in the form of C or H 2. In aromatics, toluene is the major component, whereas in olefins, propylene is the major component produced. Equations S1 and S2 show the overall stoichiometry for the conversion of glucose to toluene and propylene respectively. C 6 H 12 6 (12/22) C 7 H 8 + (26/22) C + (84/22) H 2 (S1) C 6 H 12 6 C 3 H 6 + 3C + 3H 2 (S2) 13

15 Hence the theoretical carbon yield of toluene and propylene from glucose is 63.6% and 50.0% respectively. Similarly the stoichiometry for the conversion ethylene glycol over zeolite catalyst is shown in Equations S3 and S4. C 2 H 6 2 (6/22) C 7 H 8 + (2/22) C + (42/22) H 2 (S3) C 2 H C 3 H C H 2 (S4) The theoretical carbon yields of toluene and propylene from ethylene glycol are 95.5% and 75.0% respectively. Based on the elemental composition of DEB, low temperature hydrogenated DEB and the measured composition of WSB and its hydrogenated products, similar maximum theoretical yield equations can be written. Table S8 depicts the theoretical yields to toluene and propylene from various oxygenated hydrocarbons of interest in this study. The table also depicts the observed aromatic + olefin yield as a percentage of theoretical toluene yield. Table S8 Theoretical toluene and propylene yields and percentage of theoretical toluene yield for different feedstocks Feed H/C eff ratio Theoretical carbon yield (%) Toluene Propylene Experimental aromatic + olefin carbon yield (%) Percentage of theoretical toluene yield* Glucose DEB LTHDEB WSB Sorbitol Furan Glycerol LTHWSB Ethylene glycol HTHWSB Propylene glycol Tetrahydrofuran Methanol Ethanol 1Propanol *Percentage of theoretical toluene yield is calculated by dividing the experimental aromatic + olefin carbon yield (%) by theoretical carbon yield to toluene (%). 14