C3Bio develops transformational knowledge and technologies for the - - PowerPoint PPT Presentation

Cen Cente ter f r for

• r Dir

Direc ect t Ca Cata talytic ytic Con Conver ersion sion

• f
• f Biomass

Biomass to to Bi Biofu

• fuels

els (C3Bio) (C3Bio)

RESEARCH PLAN AND DIRECTIONS We will maximize the energy and carbon efficiencies of advanced biofuels production by the design of both thermal and chemical conversion processes and the biomass itself. Impacts are to more than double the carbon captured into fuel molecules and expand the product range to alkanes and other energy-rich fuels.

C3Bio develops transformational knowledge and technologies for the direct conversion

• f plant lignocellulosic biomass to

advanced (drop-in) biofuels and other biobased products, currently derived from

• il, by the use of new chemical catalysts

and thermal treatments.

Mahdi Abu-Omar and Hilkka Kenttämaa / Department of Chemistry, Purdue University

Ca Cata talytic ytic co conver ersion sion of

• f li

lign gnin in

Lignin is a major component of lignocellulosic

• biomass. It is an aromatic rich polymer that is

essential for plant’s life. Lignin poses the problem of recalcitrance as well as an opportunity for making aromatic-rich liquid fuels and valuable chemicals. A desirable catalyst is one that can depolymerize lignin, remove oxygens, and retain the aromaticity. Another challenge in this area is the analysis of complex

• mixtures. We have developed a catalyst Zn/Pd/C that

cleaves aromatic ether linkages while leaving the aromatic group unscathed. We have also implemented mass spectrometry methods that enable the quantitative analysis of lignin products. We are now poised to apply these methods of catalysis and analysis to engineered lignin biomass.

OH O OH O O HO

OH O OH O

100 200 300 400 500 m/z 10 20 30 40 50 60 70 80 90 100 319 179 349 209

[M-H]-

We are involved in the synthesis and application

• f
• rganic-inorganic

hybrid materials that ultimately will become single site catalysts. Using a well developed synthetic methodology, we have created high surface area catalysts functionalized with aryl sulfonic acids. These catalysts are being tested for their ability to hydrolyze cellobiose into glucose. This is a model study that has implications for the eventual conversion of cellulose from biomass into viable fuels and other high value chemicals. In a parallel line

• f

research, we are investigating the selective oxidation of lignin models to produce quinones which may be easily transformed into value added chemicals. We are exploring a number of titanium-on-silica catalysts created through targeted synthetic methods that will allow for the determination of which active site is optimal for oxidation. Early results have been promising for conversion of the lignin models to benzoquinones in high yield and with good selectivity.

C Barnes, J Abbott, D Taylor, S Chen - Univ. of Tennessee, Knoxville

Ca Catal talyt ytic ic hy hydr drol

• lys

ysis is of

• f cellulosic ma

cellulosic mater terials ials & selectiv selective e oxida xidation of tion of lignin lignin mod models els

A Olek1, S Ding2, B Donohoe2, L Makowski3, L Paul1, and N Carpita1

Bioc ioche hemical mical mec mecha hanism nism of

• f cellulose

cellulose sy synth nthesis esis

• Synthesis of cellobiose units eliminates the steric problem of iterative

synthesis of a single unit. because the O-4 would always be in the same location in the non-reducing end of the growing chain.

• A channel of 8 x 2 = 16 membrane spanning domains would be

equivalent to callose synthase and most sugar transport proteins.

• The dimer produces two Zn-finger domains to recruit into larger

complexes.

1Purdue University, 2NREL, 3Northeastern University/ANL

The 55 kDa catalytic domains of CesA spontaneously dimerize when a thiol-reducing agent is depleted from the reaction mixture. The dimerization is reversible and can be shown by high-performance size-exclusion chromatography, analytical ultracentrifugation, atomic-force microscopy, and X-ray scattering experiments. The 55 kDa monomer is predicted by WAXS to be 30.0Å, where the 110 kDa dimer is a more spherical 34.0Å. The ratio of the monomer : dimer estimates a distance between centers of mass to be 41.3Å

J I Kim and C Chapple, Purdue University

Under Underst standing anding cell w cell wall ass all assembl embly y using using Ar Arabido bidopsis psis lignin lignin mutan mutants ts

Lignin is a major component of the plant cell wall and understanding how, when and where it is deposited is critical to being able to catalyze its conversion to useful products such as biofuels. We have capitalized on our suite of lignin- deficient mutants

• f

Arabidopsis to generate plant lines in which lignin biosynthesis, which is normally blocked in these mutants, can be turned on by application of a chemical inducer. Normally, lignin deficiency leads to dwarfing, but when lignification is induced in these lines, they again grow normally. We are now using this system to study the early stages

• f

lignification, where lignin is first deposited and how cell wall assembly is altered when lignification is uncoupled from cell wall polysaccharide synthesis.

• DEX

+DEX

C4H-deficient Control Control C3’H-deficient C4H-deficient C3’H-deficient

“Research Goes to School” An An ou

• utlet f

tlet for

• r EFR

EFRC C scien science ce

K Clase, K Goodpaster, O Adedokun, L Kirkham, P Ertmer, G Weaver, M Abu- Omar, N Carpita, H Kenttämaa, M McCann and N Mosier, Purdue University

In June 2011, 21 in-service and pre-service teachers participated in an intensive 2-week workshop From Field to Fuel - The Science of Sustainable Energy to help educators develop biofuels curricula specifically to increase the relevance of STEM subjects for rural students. “Research Goes to School” is an NSF Innovations through Institutional Integration grant to Purdue in collaboration with the Woodrow Wilson STEM Goes Rural Initiative, National Rural Education Association, I-STEM Resources Network, and Purdue Rural Schools Network. C3Bio investigators Abu-Omar, Carpita, Kenttämaa, McCann and Mosier assisted Dr. Clase through presentations on their state-of-the-art research in advanced biofuels. McCann is a co-PI on the NSF grant. The teachers developed problem-based learning units for classroom curricula, mapped to educational standards using C3Bio content. The educators completed pre- and post- science teaching self-efficacy and content knowledge measures, and participated in a post-workshop focus

• group. Preliminary results indicate that the workshop enhanced participants’

knowledge of biofuels concepts and their beliefs that student learning can be influenced by effective teaching. Furthermore, participants expressed that the workshop enhanced their understanding of the applications of biofuels concepts to STEM content areas and enhanced their sense of purpose for teaching.

P Ciesielski, J Matthews, M Crowley, M Himmel, B Donohoe (NREL)

• Transmission electron tomography is used to obtain

3D data sets (tomograms) of thermochemically deconstructed plant cell walls. A single slice from a tomogram (top left) shows 2 intertwined cellulose microfibrils.

• The geometry of the microfibrils is determined by

fitting parametric equations to the 3D dataset (top right).

• Atomistic, macromolecular models (bottom) are

constructed by building the molecular structure of cellulose around the determined geometry of the microfibrils.

• These structures will allow for molecular dynamics

simulations that more accurately reflect the structure

• f biomass and are highly relevant to real processing

conditions.

Macr Macromolecular

• molecular modeling of

modeling of cellulose cellulose micr microf

• fibrils fr

ibrils from e

• m electr

lectron tom

• n tomog
• graphy

phy

H Kenttämaa, M Abu-Omar / Purdue Univ.

HPL HPLC/MS C/MS an anal alysis ysis of

• f de

degrad adation tion pr prod

• duc

ucts ts of

• f li

lign gnin in mod model el co compo mpoun unds ds

HPLC chromato- graph (detection by MS)

Retention Time: 6.35 min Retention Time: 7.75 min Retention Time: 10.82 min

The development of chemical methods for the direct catalytic conversion of biomass to high value organic molecules is an area of increasing interest. The plant matter component known as lignin is a polymer containing many aromatic rings. Hence, it could provide a means of obtaining aromatic chemicals currently derived solely from petroleum. We have developed a catalytic system that selectively breaks down dimeric lignin components. A high-pressure liquid chromatography tandem mass spectrometric (HPLC/MSn) method was devised for the determination

• f the products of these catalytic reactions. This

method first separates the degradation products and subsequently ionizes all components for detection by mass spectrometry, yielding molecular weight information. In MSn experiments, the ions are subjected to several consecutive collision-activated dissociation (CAD) steps to determine their structures.

0.1 eq 5% Pd/C 0.1 ZnCl2 300psi H2 MeOH,150°C, 8h

+ +

100 200 300 400 500 600 700

Mass/Charge

10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100

Relative Abundance

10 20 30 40 50 60 70 80 90 100 180.92 122.81 10 20 30 40 50 60 70 80 90 100

Relative Abundance

10.85 7.86 6.35 8.14 11.24 164.91

• CH3
• CH2 CH2OH
• CH2OH
• CO

MSn spectra: consecutive CAD

• f ions of m/z 181, 166 and 121

Mass/Charge

80 100 120 140 160 180 200 50 100 50 100

Relative Abundance

50 100 166 181 121 166 135 136 121 93

MW 182 Da

Mass spectra

MW 124 Da MW 166 Da

L Mak Makowsk wski, H Ino Inouye ye (N (Nort rtheaste stern); rn); R Harde rder, r, J Lal (A (Argo rgonne); ); L Yang (B (Broo rookhave ven)

in in si situ tu an anal alys ysis is of

• f cellulose

cellulose cr crys ystalli tallite te str struc uctur tures es

Data collection at 34IDC (APS) Bragg peak from single crystallite Image of crystallite from maize SAXS data from X9 (NSLS)

CDI - Coherent diffraction imaging of cellulose crystals in situ has been used to demonstrate that acid pretreatments lead to changes in cellulose morphology. SAXS - small-angle x-ray scattering - provides quantitative estimates

• f the average length and diameter of the crystallites.

Comparative studies of the effect of treatments on a variety of cellulosic materials are underway.

N Mosier, M Abu-Omar, Purdue Univ.

Ultr Ultrase aselect lectiv ive ca cata talysts f ysts for

• r

hy hydr drol

• lysis o

ysis of ce cell llulose ulose

The major goal of this project is to develop catalytic processes that enable the extraction, fractionation, and depolymerization of carbohydrates from biomass into aqueous solutions. The secondary goal is to couple this catalytic fractionation/depolymerization with catalytic transformation of carbohydrates into hydrocarbons and valuable chemicals. A new reaction method (microwave heating) was validated against results obtained from previously used reaction method (sand bath heating). The data indicate that the results between the two methods were not different to any statistical significance. The new method offers advantages

• f more rapid and accurate temperature

Detailed kinetics of xylose and furfural degradation in the presence of maleic acid (250mM) allow for optimization of conditions to achieve high yields of furfural directly from biomass. Reaction of xylose with furfural was minimal when maleic acid is the catalyt, in contrast to significant coupling reactions when sulfuric acid is used.

Switchgrass Xylose Xylose Recovery (%) Selectivity (%) Xylose Conversion (%) Furfural Yield (%) HPLC 1st Run 2nd Run 3rd Run > 90 > 85 >85 67 65 60 85 80 80 57 51 48

Recycling Maleic Acid to Convert Switchgrass Xylose to Furfural

Kinetic Modeling

M Easton, J Nash, Purdue University

Le Levog

• glucosan

lucosan May May Not Not Be t Be the he Pr Product of

• duct of

Fast ast Pyr Pyrol

• lys

ysis is of

• f Cellulo

Cellulose se

For many years, levoglucosan has been thought to be the product of fast pyrolysis of cellulose. High level calculations show that a number of isomers of levoglucosan are considerably more stable than levoglucosan itself. The correct assignment of the cellulose pyrolysis product is essential for designing practical and efficient methods for its conversion to biofuel.

Note: G4MP2 relative free energies (kcal/mol) are shown in black and red.

H Yang, G Ma, X Liu, A Murphy, W Peer, Purdue University

Exp Expressi ession

• n of
• f metal

metal-bind binding ing pr prote

• teins

ins in in ce cell ll wall alls s as as Trojan

• jan hor

horse se ca catal talys ysts ts

Engineering peptide transporters & metal binding peptides to deliver iron catalysts to the cell wall for biomass conversion

• Transition metal accumulation in living biomass: Generate

transgenic plant materials that express genes encoding plasma membrane-localized metal transporters.

• Significance: Rice lines that lack a metal transporter have been
• generated. The lines will be tested for Fe accumulation in the cell wall.

This is enhance biomass to biofuel catalysis.

• Engineer catalysis-enhancing proteins: Construct transgenes

that encode chimeric proteins with specific metal binding peptide motifs and have affinity for particular wall components, in order to target metal ions more specifically within the structure of the cell wall.

• Significance: Iron binding peptides (IBP) that can bind Fe at cell wall

pH (pH 5.5) have been identified.

• Significance: Carbohydrate binding motifs (CBMs) targeted to the cell

wall have been identified. These motifs will be combined with the IBP and the minimal secreted Fe-binding peptide.

• Identify cellular delivery routes to enable the secretion/self-

assembly of catalyst-ready tailored biomass.

• Significance: A secreted metal-binding protein is being modified to

bind Fe instead of Zn. The minimal protein required for secretion is being identified.

Carbohydrate Binding Module (CBM)

Fluorescently tagged, targeted to cell wall

Iron Binding Peptides (IBP)

DLGEQYFKG & LAEEKREGYER

pH 5.5 pH 7.0

F Ribeiro, H Kenttämaa, Purdue University

Fast ast pyr yrol

• lysi

ysis for

• r dir

direc ect t pr prod

• duc

uction tion

• f
• f molec

molecules in ules in th the e fue fuel l ran ange ge

Pyrolysis, heating to temperatures where biomass forms a gas and then condenses to form a bio-oil, is a relatively simple process for fuel production. However, the bio-oil contains too much oxygen and is corrosive. We need to reduce the oxygen content of bio-oil as it forms, and also the huge range of undesirable products, in order to make an energy-rich fuel. We have developed a method to measure the reaction products in the gas phase by mass

• spectrometry. Using a pyroprobe instrument with which the

rate of heating and final temperature can be precisely controlled, we can evaluate the reaction products from cellulose and other model compounds in the presence of hydrogen or other gas. Instead of the thousands of products

• bserved

during conventional pyrolysis

• f

cellulose, we observe a few discrete masses using fast- hydropyrolysis (a heating rate of 1000K per second in the presence of hydrogen). We can control the types of products by varying gas temperature and flow rate. Removing the unwanted oxygen from biomass may now be feasible to produce diesel-like products.

J Madden, G Simpson, Purdue University

Comp

• mpac

act seco second ha harmon monic ic ge gene neration tion (SH (SHG) G) micr microsco

• scope

pe for

• r co

combine mbined op

• ptical

ical and and x-ray an anal alys ysis is

• Synchrotron sources provide high X-

ray intensities, enabling nanoscale characterization of biomass cellulose structure over a limited field of view. Complementary methods for rapid initial characterization over large fields of view are under development to guide positioning for X-ray analysis and improve the overall throughput.

• Second harmonic generation (SHG), or

the frequency doubling

• f

light, provides highly selective contrast for crystalline cellulose domains. X-ray CCD Detector Synchrotron X- ray radiation Cryogenic sample handling robot Goniometer for sample positioning Cryo-stream SONICC microscope

Observed areas of fiber diffraction

100 µm

• An initial low-footprint prototype SHG microscope employing an

ultrafast (<100 fs) fiber laser source has been designed, assembled, integrated into a synchrotron source, and used for combined X-ray and SHG analysis for localization of crystalline -cellulose. Regions

• f bright SHG correlated with regions of cellulose X-ray diffraction.

C Staiger & J Henty, Purdue University

Fast filament dynamics remodel the cortical actin cytoskeleton

The cytoskeleton provides a filamentous framework that serves as “tracks” for a variety of intracellular organelle movements, including trafficking of membrane bound, polysaccharide precursor delivery to the cell wall. In the cortical array of epidermal cells, these tracks are of at least two types: massive filament bundle superhighways, and fine individual filaments. The latter population of tracks is under constant rearrangement by a process of rapid growth balanced by stochastic severing events and disassembly. To study the interplay between these populations and to dissect the molecular mechanism, we examined cytoskeletal dynamics in a homozygous mutant for an adf4 knockout using state-of-the-art live cell imaging. The adf4 knockout has a 3-fold reduction in severing frequency, longer filament lengths and lifetimes, as well as increased number of filament bundles. This provides compelling evidence for the contribution of a key actin-binding to filament dynamics and reveals a mechanism for the interplay between single filaments and bundled actin arrays. Future work will examine whether both populations support movement of secretory vesicle cargo to the cell wall.

• Figure. Actin filament dynamics in the cortical array
• f Arabidopsis epidermal arrays imaged with

variable angle epifluorescence microscopy. (A) Wild-type cell. (B) adf4 knockout mutant cell. Henty et al., (2011) The Plant Cell, submitted

Deliv elivering ering meta metal l co co-ca catal talys ysts ts to plan to plant t ce cell ll walls alls for t

• r the

he dec deconstr

• nstruction

uction of

• f engineer

engineered ed biomas biomass Incorporating iron ions into dilute acid pretreatment of biomass is a promising technology for increasing sugar

• yields. We are developing approaches to express metal-

binding or storing proteins into plant cell walls to enhance biomass deconstruction. One technique being developed is to down regulate an oligopeptide transporter gene (OPT3), leading to the accumulation of iron and other metals in stem and likely apoplast of model plants (Stacey et al., 2008). In addition, we have tested 6 cellulose binding modules (CBMs), among which CBM11 was the most efficient in attaching to plant cell walls, making it a good candidate for combining with iron binding peptides for precise delivery of metal ion co-catalysts into cell walls during plant growth. One concern with this approach is that in some metal- storing proteins such as ferritin, iron exists as ferric oxide nanoparticles which are stable up to 300-400 oC, raising a question about their bio-availability. We have confirmed that ferritin-Fe3+ can be released, and at a concentration of 2 mM, it’s incorporation into corn stover significantly enhances both glucose and xylose monomer releases by 14% and 29%, respectively, in dilute acid pretreatment.

• H. Wei1, H. Yang2, J. Cox2, P.N. Ciesielski1, B.S. Donohoe1, A.S. Murphy2,
• W. Peer2, M.E. Himmel1, M. McCann2, M.P. Tucker1 1NREL, 2Purdue Univ.

CBM11-IBPs fused genes are being engineered into plants to deliver iron catalysts to cell walls.

Iron co-catalyst pretreatment using an iron storage protein (ferritin) increases biomass digestibility

ferritin protein (~4500 iron ions /molecule)

(modified from Masuda et

• al. 2010)

impregnated with ferritin

corn stover

pretreated in dilute acid 160 oC, 20 min

* *

CBM11 was shown to be the most effective at attaching to cell walls of live plants.

Iron binding peptide (IBP) CBM11 Binding curve Corn Red fluorophore Arabidopsis Cellulose binding module (CBM)

CBM3-mCherry bright field

Metal catalysts delivery to plant cell walls

Fabio Ribeiro, rocket scientists, Purdue Univ.

Fast ast hy hydr drop

• pyr

yrol

• lysis

ysis – closing losing th the e mass mass ba balanc lance

In February, we highlighted a breakthrough in measuring the primary reaction products of pyrolysis of cellulose using mass spectrometry. Instead of the thousands of products observed during conventional pyrolysis, we observed only a few discrete masses using fast-hydropyrolysis (a heating rate of 1000K per second in the presence of hydrogen). However, we were limited to micrograms of material by our commercial pyroprobe, making it impossible to conduct a mass balance. In collaboration with rocket scientists at Purdue, our chemical engineers have built and tested a high-pressure, low-residence time, hydropyrolysis reactor at a safe distance from main campus buildings. Good news – at this much larger scale of tens of grams, we can close the mass balance to within 20%.

Product distribution from cellulose pyrolyzed in the rocket reactor