Catalytic Solubilization and Conversion of Lignocellulosic ...

U N C L A S S I F I E D

Catalytic Solubilization and Conversion of Lignocellulosic Feedstocks

LANL TeamT.A. Semelsberger (P.I.) , Tony K. Burrell (Co-PI),

Rod L. Borup, Kathryn S. Lovejoy, and Kevin C. Ott

This presentation does not contain any proprietary, confidential, or otherwise restricted information

DOE Fuel Cell Technologies Program Annual Merit Review,EERE: Hydrogen, Fuel Cells and Infrastructure Technologies Program

Washington, DC June 07-11, 2010Technology Development Manager : Rick Farmer

PD042


LANL Project Overview

Timeline • Project Start Date:FY07• Project End Date: FY08• Percent Complete: 100%

2

Budget •Project End Date: FY2008• Funding:

•2008: $300K•2009: $0K*•2010: $0K

Barriers • Barriers Addressed

• Feedstock Cost and Availability• Capital Cost and Efficiency of Biomass Gasification/Pyrolysis Technology

Partners• None

*EERE Hydrogen Production and Delivery Budget Zeroed Out


LANL Project Objectives

Project Objective

Develop novel low temperature chemical routes and catalysts to produce hydrogen/syngas from lignocellulosic feedstocks

The most abundant constituent of biomass is lignocellulosic (~80%). Discovering new chemistries and catalysts that can convert lignocellulosic into

hydrogen/syngas will be critical if biomass is to be used as a feedstock for hydrogen or other alternative fuels.

Lignocellulosic depolymerization/decomposition is the most process intensive (and most challenging) constituent of biomass to convert to hydrogen/syngas

Target: By 2012, reduce the cost of hydrogen produced from biomass gasification to $1.60/gge at the plant gate (<$3.30/gge delivered). By 2017, reduce the cost of hydrogen produced from biomass gasification to $1.10/gge at the plant gate ($2.10/gge delivered).

3


LANL Project Approach

4

In general terms, LANL is in search of novel hydrogen/syngas production routes from lignocellulosics. Two approaches will be explored:

• Catalytic solubilization of lignocellulosics to generate a sugar feedstock stream for downstream APR, and • Solubilization of lignocellulosics followed by APRxn of oligomeric, soluble cellulose.

LANL will conduct screening experiments for evidence of direct aqueous-phase low-temperature reforming of lignocellulosics to hydrogen/syngas through the use of catalysts designed to cleave carbon-carbon bonds of the cellulose backbone. Tandem catalysis approaches, where two catalysts or processes are linked together in a single reaction vessel, will be explored to demonstrate “one-pot” cellulose solubilization followed by aqueous phase catalytic reforming to generate hydrogen. This is important in that if catalysts can be found that will generate hydrogen directly from soluble cellulose oligomers, this provides a ‘one-pot’ approach and offers increased utilization of residual biomass, increased efficiency and the potential for cost reductions both in feedstock and in capital equipment. LANL’s approach to producing hydrogen from lignocellulosics (i.e., middle and bottom routes) is represented by the chemical routes shown in Figure 1 (next slide).


LANL Project Approach (cont’d, Figure 1)

5

Carbohydrates

Cellulose(insoluble)

Lignin (insoluble)

Ethanol(soluble)

digestion OligomerSolution

(Cellobiose)

OligomerSolution

MonomerSolution

(Glucose)

MonomerSolution

Bio-Syngas

Figure 1. A rudimentary diagram showing LANL's approach to producing bio-syngas (i.e, hydrogen and carbon monoxide) from lignocellulosics


LANL Technical Accomplishments and Progress

• Demonstrated heterogeneous catalyzed hydrolysis of cellobiose to glucose

• Demonstrated the conversion of cellobiose to syngas [albeit at low conversions (~5%)]

Accomplishments for FY2007-2008

6

• Demonstrated catalytically enhanced decarboxylation of lignin

• Performed baseline characterization studies on model compounds (i.e., lignin and cellobiose)

• Demonstrated low temperature catalyzed gasification of lignin

Cel

lulo

seLi

gnin


LANL Overview of Scoping Experiments Performed in FY07-08

• Flow reactor system for liquid conversion (bench-scale)• Batch reactors for liquid/solid conversion (bench-scale)

• Scoping experimental results– Liquid phase conversion

• glucose, cellobiose– Solid phase mass conversion

• lignin, pine– Residual solids analysis

• TGA (thermal gravimetric analysis)• NMR• FTIR (molecular vibrational frequencies)

– Product analysis• LC (liquid chromatograph)• gas analysis

7


Conversion of Liquid Phase Oligomeric Cellulose

• Heterogeneous catalytic conversion of soluble phase– Glucose and Cellobiose to vapor phase products

• Homogeneous catalytic conversion of model cellulose– Cellobiose as model compound to demonstrate solubization

• Operation– Flow reactor

• Well defined conditions (control of T, P, flows)• Gas analysis

– Batch reactors – closed system• Reactants loaded, put in oven

– T = 100 – 275 oC; 4 – 18 hrs• Post analysis

– Catalysts• Base metals, noble metals with Lewis acid supports (Al2O3, zeolites)• Ln Triflates, perfluorosulphonic acid as homogeneous Lewis acids

8


Analytical Tools Employed for Biomass Research

9

Back Pressure Regulator

Vapor Phase Product Stream at Ambient Conditions

Prime

Filter

Accuflow Series II Pump

Watlow Temperature

Controller

High Pressure N2 Line: Max 61 bar (900 psig)

Chiller Block

3-Way Valve

Pressure Transducer

Furnace

Liquid Feeds (H2O + Glucose, etc.)

Purge/Activation Gases (MFC 1 and 2 are manifolded)

MTI

M200HPortable Gas Chromatograph

setra

206

High Pressure 2-Phase Product Stream

High Pressure Vapor Phase Product Stream

Liquid Product Stream

Needle Valve

Prime

Filter

Accuflow Series II Pump

Watlow Temperature

Controller

High Pressure N2 Line: Max 61 bar (900 psig)

3-Way Valve

Pressure Transducer

Furnace

Liquid Feeds (H2O + Glucose, etc.)

Purge/Activation Gases (MFC 3 and 4 are manifolded)

MTI

M200HPortable Gas Chromatograph

setra

206

High Pressure 2-Phase Product Stream

High Pressure Vapor Phase Product Stream

Liquid Product Stream

Reactor 1 Reactor 2

Thermocouple Bank (Type K)

Multi-phase flow reactor• T = 20-1000°C• P = 1-60 atm

Gas Chromatograph (GC)• Gas analyses

Liquid Chromatograph (LC)• Liquid analyses

Additional Analytical Tools• Liquid NMR• Solid-state NMR• Solid-state DRIFTSAdditional Reactors•Multi-well batch reactors for rapid screening


Evolved Gas Analyzer (EGA) Setup

10

Mass Spectrometer (MS)

Gas Phase Infrared Spectrometer (IR)

Thermal Gravimetric Analyzer (TGA)

Gas Chromatogram (GC)EGA Capabilities

Measure mass changes as function of temperature

Correlate mass changes with evolved gas

Identify evolved gases with IR, MS, and GC

EGA system facilitates a deeper understanding of the reaction rates and chemistry

Our suite of analytical tools allow us to gain insight into the fundamental processes of biomass pretreatment and hydrogen production from biomass, thus allowing for tailor-made, energy efficient, cost-effective processes for biomass utilization

Analytical Tools Employed for Biomass Research


Aqueous Phase Oligomeric-Cellulose Reforming

11

Glucose conversion to vapor phase products and trace liquid phase products

Oligomeric-cellulose conversion to glucose, vapor phase products and trace liquid phase products

Cellobiose

Glucose

Cellobiose

Reac

tant: C

ellob

iose Catalyst: Pt/RhCatalyst: None

Conversion: ~ 13.8% Conversion: ~ 42%

GlucoseArea =2.5 e6

Glucose

Different Scale

Area = 7.8 e6

Reac

tant: G

lucos

e Catalyst: Pt

Catalyst: None

Conversion: ~ 68%Biosyngas

BiosyngasCellulose OligomerSolution

HOO

HOOH

O

OH

O

HOOH

OH

OH

C1(n)C1(α/β)

[cat.], H2O HOO

HOOH

OH

OH

C1(α/β)

cellobiose α/β-D-glucose


SpinWorks 2.5:

PPM 104.0 100.0 96.0 92.0 88.0 84.0 80.0 76.0 72.0 68.0 64.0 60.0 56.0 52.0

file: C:\Documents and Settings\118404\Desktop\Lanthanide_project\nmr\MS-1-14\6\fid expt: <dept135>transmitter freq.: 75.475295 MHztime domain size: 65536 pointswidth: 17985.61 Hz = 238.297995 ppm = 0.274439 Hz/ptnumber of scans: 256

freq. of 0 ppm: 75.467684 MHzprocessed size: 32768 complex pointsLB: 0.000 GB: 0.0000

anomeric carbon signals:C1(n), C1(α/β) and C1(α/β)

cellobiose only (control)

cellobiose + Gd(OTf)3

Catalyzed Hydrolysis of Cellobiose to Glucose

13C NMR

12

SpinWorks 2.5:

PPM 108.0 106.0 104.0 102.0 100.0 98.0 96.0 94.0 92.0

file: C:\Documents and Settings\118404\Desktop\Lanthanide_project\nmr\MS-1-14\6\fid expt: <dept135>transmitter freq.: 75.475295 MHztime domain size: 65536 pointswidth: 17985.61 Hz = 238.297995 ppm = 0.274439 Hz/ptnumber of scans: 256


cellobiose only (control)

cellobiose + Gd(OTf)3:

C1(n)

C1(n)

C1(β)C1(α)

C1(β) C1(α)

C1(β) C1(α)

C1(β) C1(α)

Blue = cellobioseRed = glucose

~ 75 % conversion with catalyst to free glucose without significant decomposition/carmelization

< 5 % conversionto free glucose

SpinWorks 2.5:

PPM 11.0 10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 -1.0 -2.0

file: C:\Documents and Settings\118404\Desktop\Lanthanide_project\nmr\MS-1-18\4\fid expt: <zg30>transmitter freq.: 300.131351 MHztime domain size: 32768 pointswidth: 4496.40 Hz = 14.981450 ppm = 0.137219 Hz/ptnumber of scans: 32


top: cellulose in D2O

(both spectra on same scale with identical conditions)

bottom: cellulose & Lu(OTf)3 in D2O

free aldehydes & enones

H(a) free glucose

H(b) free glucose

ring protons of free glucose and water soluble oligomers

D2O

Enhanced hydrolysis in the presence of LnX3

1H NMR

Lewis acid catalysis performs hydrolysis of cellulose to glucose


Solid Phase Conversion of Lignin

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

blank La Triflate 3% H2O2 1% perfluorosulphonicacid

Mas

s C

onve

rsio

n

LigninPine

275 oC / 18 hrs

Enhanced catalytic conversion of solids, but rates slow

13

0

5

10

15

20

25

30

35

40

H2 N2 CO CH4 CO2

Gas

Con

cent

ratio

n / %

Gd Trfilate / PineYb Triflate / PineLa Triflate / Pine

Gas Analysis post batch reactor operation(N2 from air in overhead reactor space)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Gd Trfilate / Pine Yb Triflate / Pine La Triflate / Pine

Nor

mal

ized

Des

ired

Gas

Rat

ios

H2CO CH4

Desired products:CO > H2 > CH4Major products are not alkanes

Lignin + Gd TriflateTreatment @ 275 oC for 18 hrs

Lignin + H2OTreatment @ 275 oC for 18 hrs

Significant reduction in higher residence time species when catalyzed by Gd Triflate


TGA of Lignin Residue After Various Treatments

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500 600 700 800

Temperature / oC

% In

itial

Mas

s

Lignin BlankLignin + Gd TriflateLignin + H2O2Lignin + perfluorosufonic acid

Oxidation of Lignin remains unchanged

Atmosphere: Air

14

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500 600 700 800Temperature / oC

% In

itia

l Mas

Lignin BlankLignin + Gd TriflateLignin + H2O2Lignin + perflourosulphonic acidLignin + La Triflate

• Decomposition of Lignin• Most cases unchanged• different mechanism w Gd Triflate

Atmosphere: N2

Oxidation of treated pine remains unchanged

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500 600 700 800

Temperature / oC

% In

itial

Mas

s

Pine

Pine 225 oC / H2O

Pine 275 oC / H2O

Pine 1% perfluorosufonic acid / 275 oC

Pine La Triflate / 275 oC

Fresh pine shows loss of lower molecular weight HC’s

Atmosphere: Air

0

20

40

60

80

100

120

0 100 200 300 400 500 600 700 800Temperature / oC

% In

itial

Mas

s

PinePine 225CPine 275 oC / H2OPine 1% perfluorosufonic acidPine La Triflate / 275 oC

Decomposition mechanism unchanged with La Triflate

Fresh pine shows loss of lower molecular weight HC’s

Atmosphere: N2


TGA and Evolved Gas Analysis: Lignin Treated with Yb Triflate

15

DRIFTS: Fresh Lignin

Change in relative quantities of functionalities

Fresh Lignin Hydrolytic T = 25C

0.0

0.5

1.0

1.5

2.0

KM

Lignin Fresh Hydrolytic T = 600C

0.0

0.5

1.0

1.5

2.0

KM

1000 1500 2000 2500 3000 3500 Wavenumbers (cm-1)

Thermal Decomposition of Lignin

Loss of hydroxyl species

Loss of specific vibrationalfeatures

Observed changes in lignin DRIFTS spectrum after various pretreatments

chemistry is occurring


TGA and Evolved Gas Analysis: Lignin Treated with Yb Triflate

16

SS-NMR of Fresh Lignin

100 200 300 400 500 600 700 800

CO2

CH4

H2

TGA

Temperature (°C)

Mas

s S

pec

Sig

nal (

arb.

uni

ts)

60

50

40

30

20

10

0

Wt Loss (%

)Low-temperature, catalytic pyrolysis of lignin

100 200 300 400 500 600 700 800

CO2

CH4

H2

TGA

Temperature (°C)

Mas

s Sp

ec S

igna

l (ar

b. u

nits

)

60

50

40

30

20

10

0

Wt Loss (%

)

Lignin pyrolysis in the absence of a catalyst

Catalysts show a reduction in temperature required for hydrogen production

Observed changes in SS-NMR spectrum after various pretreatments

chemistry is occurring

0

100

200

300

400

500

600

150 250 350 450 550 650 750 850 950

Temp (C)

Gas

Conc

entra

tion

(ppm

)

0

10

20

30

40

50

60

70

80

90

100

Mass (%

)

H2O

Lignin Mass

CO2

H2

CH4

CO

No evolution of higher molecular weight compounds


LANL Project Summary

17

• Conversion of cellobiose to glucose is feasible, but rates currently too low• Lignin hydrophobicity is a critical challenge for APRxn processes• Recent results of low temperature catalyzed pyrolysis of lignin shows potential

Mechanism of the low temperature catalyzed pyrolysis of lignin currently unknown• Heterogeneous catalysis of glucose and cellobiose

Relatively high conversions during batch reaction (~60 – 90%) Major products appear to be gas phase for heterogeneous catalysis

• Homogeneous catalysis of cellobiose hydrolysis to glucose without significant decomposition and/or caramelization Aqueous cellulose suspension marginally hydrolyzed to free glucose

• Solid conversion of Lignin & Pine increased by Lewis Acid catalysis Gas phase products tend to syngas rather than alkanes Minimal structural change of remaining Lignin (TGA, NMR, DRIFTS)

—Some change in vibrational structure with La Triflate Lignin/Gd Triflate demonstrates different decomposition mechanism


Obstacles to Lignocellulosic Conversion

• Conversion of solubilized hydrocarbons to vapor phase• Conversion of model compounds simulating solubilization• Unknown reactivity as a function of lignin pretreatment• Lignin Solubilization

– Interactions with catalysts limited– Hydrophobicity– Steric hindrance

• Conversion chemistry– Reaction mechanisms not understood

• Innovation in chemistry and catalysis• Innovation in reactor design and reaction engineering• Current approaches use highly corrosive bases (>10 molar) requiring costlymaterials of construction

18


LANL Future Work (FY10/11)

19

• Continue screening for novel low-temperature biomass gasification catalysts• Explore conversion chemistry of oligomeric cellulose in phase transfer media• Explore lignin solubilization and catalytic conversion chemistry of lignin in phase transfer media (PRIMARY FOCUS)

Obstacles addressed:• Lignin Solubilization

– Interactions with catalysts limited– Hydrophobicity– Steric hinderance

• Conversion chemistry– Reaction mechanisms not understood

• Innovation in chemistry and catalysis• Innovation in reactor design and reaction engineering• Eliminate highly corrosive solvents and/or reactants• Reduce process cost• Increase process efficiency


LANL Future Work: Envisioned Process

20

Biomass Digestion Tank: Complete dissolution of raw biomass

Non-Aqueous Phase Reactor: Soluble biomass is cracked into lower molecular weight, water-soluble species

Aqueous Phase Reactor: Water-soluble, lower molecular weight species react in aqueous phase producing biosyn gas or liquid biofuels

1. A non-corrosive, cheap solvent required to dissolve/digest raw lignin, making lignin tractable [Lignin Solublization, Reduce process cost ]

2. An active, durable, cheap water-insoluble catalyst required for cracking lignin into water-soluble oligomers[Innovation in chemistry and catalysis, Innovation in reactor design and reaction engineering, & Increase process efficiency]

3. Reaction chemistries and mechanisms must be understood to optimize process viability and reduce cost [Innovation in chemistry and catalysis , Innovation in reactor design and reaction engineering, & Increase process efficiency]

Innovation in reactor design and reaction engineering

Requirements for process viability


LANL Future Work: Biomass Digestion Tank Chemistry

21

Biomass digestion tank

LANL has demonstrated experimentally the dissolution of lignin, keratin, cellulose, and pine dust in various non-aqueous media

• Lignocellulosics are cross-linked by extensive intra-and inter-chain hydrogen bonds

• Solvents that can break up the hydrogen bond network are known to solubilize lignocellulosics

• Ionic liquids are known to have the ability to solubilize lignocellulosics in this way

Non-Aqueous Solvent

Hydrophobic Biomass Solid Hydrophobic Biomass Solution

Literature precedents

Cellulose film containing entrapped laccase (2.78% w/w) formed using the IL-dissolution and reconstitution treatment, before (left) and after (right)

*M. Turner, et al Biomacromolecules. Vol: 5, 1379-1384 (2004)


LANL Future Work: Biomass Digestion Tank Chemistry

22

Biomass digestion tank LANL has demonstrated experimentally the dissolution of lignin, keratin, cellulose, and pine dust in various non-aqueous media

0

20

40

60

80

100

120

600160026003600

%Tr

ansm

issi

on

Wavenumbers (cm-1)

methyl cellulose

Cyphos 101

1% wt solution, methyl cellulose in CYPHOS 101

10% slurry, methyl cellulose in CYPHOS 101

50% slurry, methyl cellulose in CYPHOS 101

0102030405060708090

100

140015001600170018001900

% T

rans

mis

sion

Wavenumbers (cm-1)

PR4 (CYPHOS 164)1% keratin in PR42.5% keratin in PR45% keratin in PR47.5% keratin in PR410% keratin in PR420% keratin in PR4solid keratinred wool

0

20

40

60

80

100

600110016002100260031003600

% T

rans

mis

sion

Wavenumber (cm-1)

CYPHOS 101

solid lignin, hydrolytic

solid lignin, organosolv

2.5 wt pct lignin, hydrolytic in CYPHOS 101

2.5 wt pct lignin, organosolv in CYPHOS 101

IR Spectra of dissolved keratin in PTM

IR Spectra of dissolved lignin in PTM

IR Spectra of dissolved cellulose in PTM

Liquid phase IR capable of detecting and quantifying extents of dissolution of keratin, lignin, and celluloseWe will also employ this technique to track the cracking of lignin and

cellulose into lower MW oligomers


LANL Future Work: LT Phase Transfer Chemistry

23

LT Phase Transfer Reactor (LT-PTR)

Non-Aqueous Phase Catalyst

Hydrophobic Biomass Solution

Low Molecular-Weight, Water soluble Lignocellulosic Oligomers

• PTM catalyst in presence of trace water partially hydrolyses cellulose

• Produces water-soluble lower MW oligomersof cellulose and lignin

Biphasic IL-metal cation/water systems• Known and demonstrated at LANL

Water-immiscible ILs• Known and demonstrated at LANL

Quaternary PTM immiscible mixture

PTM Phase 1

PTM Phase 2

PTM Phase 3

PTM Phase 4

Water phase

PTM phase Phases have remained immiscible for greater than seven months



24

LT Phase Transfer Reactor (LT-PTR)• Cellulose and lignin oligomers converted

to biosyn gas or other engineered products

Catalytic hydrolysis of cellulose• Demonstrated in FY2007-2008 research

funded by Hydrogen ProductionAqueous

Phase Catalyst

Low Molecular-Weight, Water soluble Lignocellulosic Oligomers

HOO

HOOH

OH

OH

CO2 + CO + H2

Liquid Fuels

Engineered Product Selectivity

Glucose

Cellobiose

Catalyst: Pt/Rh

Conversion: ~ 42%

*S.K. Hanson, et al J. Am. Chem. Soc. 131, 428, (2009)

β O-4 linkage

A primary building unit of lignin

LANL LDRD-funded research* has demonstrated catalytic selective oxidation at the β O-4 linkages of a series of model compounds that generate low molecular weight phenols, benzoic acids, aldehydes, among others


LANL Future Work: Cost Estimates

25

Cost Estimates using Phase Transfer Media (PTM)Catalyst assumptions:• Mass is equal to SMR plant• Equivalent lifetimes as SMR catalyst • Pt loading = 0.5%

Sizing assumptions:• PTM mass/volume based on a solubility of0.5 g biomass/gPTM• Equivalent lifetimes as SMR catalyst • Reactor residence time= 10 min

Basis:SMR Plant• H2 Production capacity = 2.8 x 105 kg H2/day • Catalyst volume = 20.5 m3

• Catalyst mass = 1.9 x 104 kg • Catalyst lifetime = 5 yrs

Catalyst and PTM Costs• Raw Pt catalyst cost = $5.5Ma

• Assumed catalyst cost = $19M• Phase transfer media (@ $45b/kg) = $0.32M

a Pt catalyst loading 0.5%, Stock Price = $55/gb PTM quoted priceCatalyst cost includes:

• Precious metal recycling cost• 10% Pt loss• Interest

Note: assumed catalyst cost is extremely high compared to current industrial prices; Proposed catalysts do not contain precious metals

MAXIMUM COST CONTRIBUTION OF PTM AND CATALYST

$ 2 2(Solvent and Catalyst) / kg H Produced 0.02 -0.04 $ / kg H=Costs reflect worst case scenario



26

Issues to be resolved Individual steps known independently, but not in one system – need to demonstrate If APR of short-chain cellulose oligomers is slow, then we will focus on cracking

lignocellulosics all the way to glucose

Advantages/Uniqueness of LANL Project One pot reactor capable of solubilizing and catalyzing both lignin and cellulose

• Phase transfer catalysis• Water soluble fractions fed into APR process

Extremely flexible process capable of producing various chemical feedstocks for further APR processing

Maximum cost contributions of PTM and catalyst are on the order of $0.02-0.04 per kg of H2 produced

Reactor and plumbing materials can be carbon steel


Acknowledgements

Hydrogen, Fuel Cells & Infrastructure Technologies Program: Hydrogen Production and DeliveryProgram Manager: Richard Farmer

Catalytic Solubilization and Conversion of Lignocellulosic ...

Documents

Transcript of Catalytic Solubilization and Conversion of Lignocellulosic ...