NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Fractionation of Algal Biomass for Increased Biofuel Yields and Lower Costs

Nick Nagle

Lieve Laurens

Ryan Davis

Philip Pienkos

October 2nd, 2013

2

Outline

• Potential for “whole cell” fractionation for cost-effective biofuel production

• Advantages of targeted fractionation compared to standard process

• Scale up demonstration to >100 kg of algal biomass into algal biofuel

• Gallons Gasoline Equivalents (GGE) to make comparisons between fuels and intermediates

• Techno-economic analysis (TEA) based on pilot scale data to show progress, cost reduction and reduce risk

Oil

before after

Soluble sugars Scenedesmus sp.

3

Assume national capacity for biomass production at

1.4 B tons per year (Wigmosta et al. 2011).

Biomass Fractionation

336 M tons Lipids (fatty

acids)

588 M tons Glucose

97 M tons Protein (amino

acids)

4

Feedstock Composition and Production

4

Chlorella sp. Scenedesmus sp. Nannochloropsis sp.

5

o Varies based on growth/harvest parameters

o Theoretical yields of combined biofuels process depend on carbohydrate & lipid content

o Determine best growth/harvest regime

o Secondary reactions between constituents possible

Ash Starch Carbohydrates

(no Starch) Protein Fatty acids

Scenedesmus sp.

HP (early) 5.6 6.9 17.4 34.5 6.6

HC (mid) 1.6 11.6 40.7 9 30.9

HL (late) 2 7.2 34.3 7.5 40.3

Chlorella sp.

HP (early) 4.7 3.3 9.1 40.2 13.0

HC (mid) 2.7 36.8 5.9 13.1 22.4

HL (late) 2.6 21.9 5.2 12.7 40.5

Feedstock Composition and Production

6

Process and Pathway Overview

Algal biomass production: Scenedesmus sp.

- High Carb (HC) - High Lipid (HL) - High Protein (HP)

Chlorella sp:

- High Carb (HC) - High Lipid (HL) - High Protein (HP

Lipid Conversion

and Upgrading

Zipperclave prt and

Fermentation (200-500 g)

Pilot Scale prt. and

Fermentation

Small-scale microwave prt (2-4 g)

Carbohydrate and Lipid

Yield

Carbohydrate and Ethanol

Yield

Carbohydrate, Lipid and

Ethanol Yield

Techno-economic analysis

Outputs

Data Lipids

Data

7

Dilute Acid/Base Pretreatment

Experimental Parameters – CCD Experimental Approach Whole Algae- Non Extracted

Low High

Temperature 115o C 180o C

Time (min) 1 20

Catalyst conc. (% acid) 0.0 3.0

Catalyst conc. (% NaOH) 0.0 0.1

• CEM Explorer Microwave Reactor

• Automated system w/36-positions

• 300 watt output

• Stir bar mixing-rapid heating

Acid Catalyzed Alkaline Catalyzed

• Use pretreatment to breach cell wall

• Tracking both carbohydrate and lipid

(FAME) release

Lipid Droplet

8

Small scale Dilute Acid Microwave Pretreatment

0

20

40

60

80

100

120

1%/125C/

10.5min

2%/145C/1min

2%/145C/20min

%Theore

calYield

GlucoseYield%

LipidYield%

0

20

40

60

80

100

1%/125C/

10.5min

2%/145C/1min

2%/145C/20min

%Theore

calYield

GlucoseYield%

LipidYield%

High Lipid Chlorella

High Lipid Scenedesmus

• Developed and demonstrated single-step dilute acid pretreatment to hydrolyze polysaccharides to glucose without enzymes or production of inhibitors

• Hexane extraction of pretreated solids provided good lipid yields

• Demonstrated carbohydrate release in both Chlorella sp. and Scenedesmus sp., cultivated under high lipid and high carbohydrate regimes

High Carb Scenedesmus fractionation parameter optimization

9

Impact of Process Sequence

Fatty Acids in Extract (g/kg biomass)

Fatty Acid Recovery (%)

Glucose in Liquor (g/kg biomass)

Glucose Recovery (%)

High Protein SD 9 13 129 94

High Carb SD 19 7 290 83

High Lipid SD 20 6 244 94

Fatty Acids in Extract (g/kg biomass)

Fatty Acid Recovery (%)

Glucose in Liquor (g/kg biomass)

Glucose Recovery (%)

High Protein SD 53 78 106 77

High Carb SD 236 97 254 73

High Lipid SD 268 76 186 72

Extraction >Pretreatment

Pretreatment > Extraction

Scenedesmus sp.

[Same result seen with Chlorella biomass]

10

Morphological Changes

HP

HC

HL

Scenedesmus sp. Chlorella sp.

11

Fermentation Results-Shake Flasks

0

20

40

60

80

100

120

0 5 10 15 20 25 30

%Yield

Time(Hr)

HPCZ

HLCZ

HCCZ

0

20

40

60

80

100

0 5 10 15 20 25 30

%Yield

Time(hr)

HPSD

HLSD

HCSD

• Fermentation completed at 6-21 hr. • HMF concentration ranged from 0.5-1.5 g/L • No furfural measured • Shake flask setup using neutralized prt

hydrolyzates • Yeast peptone addition at 0.5X (3-4g/L) • Yeast (D5a) fermentation organism • No pH control

Scenedesmus sp.

Chlorella sp.

Dilute acid pretreatment at 100 g (dwt.) scale in Zipperclave reactor.

12

Fuel Yields Per Ton of Biomass

HP SD HC SD HL SD HP CZ HC CZ HL CZ

Combined fermentable sugars (kg) 193 324 220 5 234 182

Ethanol yield (%) 82 100 81 73 91 100

Ethanol-fermentation (kg) 81 165 91 2 109 93

Ethanol (Gallons) 28 55 30 1 36 31

Gasoline Gallon Equivalent (GGE) 19 36 20 0 24 20

Btu equivalent (x10e3) 2206 4224 2329 43 2778 2378

Fatty Acids (FAME in ext) (kg) 48 214 243 46 43 163

Hydrocarbon (kg) 37 167 189 36 33 127

Diesel Equivalent (gallon) 13 57 65 12 11 43

Btu equivalent (x10e3) 1573 7005 7954 1514 1395 5342

Total Fuel Energy (x10e3 Btu) 3779 11229 10283 1557 4173 7721

Total GGE 32 97 88 13 36 66

13

Solid/Liquid Separation (NREL)

Sugar Fermentation (NREL)

Biomass Production (ASU)

Integrated Scale-Up Demonstration

Pretreatment (NREL) Solids Lipid Extraction (NREL)

Nutrient Recycle

Protein Fermentation (SNL)

Protein

Lipids

14

Integrated Algal Biomass Processing

Outcome:

• High carb Scenedesmus sp. (highest

biofuel potential and shortened growth)

• Dilute acid pretreatment > solids liquid

separation > hexane extraction

• Established biofuel potential of fractions:

• Lipid yields based on FAME analysis

• Ethanol fermentation of carbohydrates

• Biobutanol fermentation of protein fraction

• Established value of ethanol fermentation

broth for nutrient recycle

15

Fermentation Results- Demonstration Scale

0

20

40

60

80

100

0 5 10 15 20 25 30

Ethan

ol%

Timehr.

Final Yield (%)

1 Based on initial composition 2 Based on 0.51g EtOH/g glucose 3 Recovery based on initial composition

Key Findings

• Demonstrated scalable process

from mg to g to kg scale

• Minimal fermentation inhibitors

produced from dilute acid

pretreatment

• Carbohydrate release and

FAME recovery demonstrated

• Ethanol fermentation is a good

proxy for other fuels/chemicals

• Pretreatment of algae followed

by extraction offers a new

process paradigm.

Glucose1 Ethanol2 Lipid3

68 80 45-73

16

Theoretical Yields HP SD HC SD HL SD HP CZ HC CZ HL CZ

Total Carbohydrates 220 435 356 112 378 246

Glucose/Mannose 190 420 343 52 333 214

Ethanol 97 214 175 27 170 109

Ethanol (gallon) 32 72 59 9 57 36

Gasoline equivalent (gallon) 21 47 39 6 37 24

Btu equivalent (x10e3) 2478 5481 4476 678 4344 2787

Total FAMEs 60 240 371 118 200 367

Hydrocarbon 47 187 289 92 156 286

Diesel equivalent (gallon) 16 64 99 31 53 98

Btu equivalent (x10e3) 1959 7865 12139 3858 6559 12021

Amino Acids 266 69 60 304 104 82

Butanol 106 28 24 122 42 33

Butanol (gallon) 34 9 8 39 13 11

Gasoline equivalent Btu equivalent (x10e3) 3425 890 775 3914 1341 1057

Total fuel energy (x10e3 Btu) 7861 14236 17390 8451 12244 15865

Total GGE/short ton 68 123 150 73 105 137

Algal Biofuel Potential

170 B GGE per year

17

Harmonization Results*: Year-Average

Updated baseline: $19.80/GGE |13.2 g/m2/day

*Davis et al. 2012 www.nrel.gov/docs/fy12osti/55431.pdf

18

Potential Cost Savings from Biomass Fractionation

Updated baseline: $16.16/GGE demonstrated

19

Conclusions

• Fractionation an alternative to current model using lipid extraction

• Demonstrated scale-up of fractionation approach

• Results from pretreatment, fermentation and extraction processes have common denominator for comparison

• Use GGE to benchmark energy produced and process efficacy

• TEA demonstrates basecase data reduces process cost by $3/gal

Questions?

Thanks to:

Stefanie Van Wychen

Nicholas Sweeney

Bob McCormick

John McGowen (ASU)