Improved Analytical Methods for Carbohydrate R&D

Improved Analytical Methods for Carbohydrate

Analysis for Biofuel Research

Richard Sevcik

2

Research Goals and Strategies

Compositional Analysis of

Biofuel Feedstocks

Stationary Phase

Modification and Method Validation Interlaboratory

Collaboration for Method Validation

Carbohydrate Analysis

3

Why Biofuels?• Only renewable source of liquid transportation fuel

• Renewable and sustainable energy source▫Support and expand agriculture and rural economies

• Reduce the need for oil and gas imports into the U.S.▫National Security▫Depletion of world oil reserves▫Energy Policy Act 1992

Reduce foreign oil usage by 30% 2010 ~37 billion gallons per year (energy equivalent)

4

EISA Mandated Biofuel Production Targets•Energy Independence and Security Act (EISA)

▫Cap corn derived ethanol 15 BGY

5

Ideal Biofuel Feedstock•Non-food crop

•Genetically adaptable▫Provide efficient conversion to ethanol

•Grow on marginal land

•Low water consumption

6

Feedstocks•1st Generation – food crops

▫corn starch

•2nd Generation – agricultural waste▫corn stover, forest harvest

lignocellulosic

•3rd Generation – dedicated energy crops▫sorghum, switchgrass

starch, juice, lignocellulosic

7

Feedstock Evaluation• Aqueous Extracts – gravimetric only• Cellulose / Hemicellulose

▫ complete acid hydrolysis of the polysacchrides▫quantitative measurement of resulting monosacchrides

• Lignin• Protein• Ash

• Critical information▫ Total Carbohydrate concentrations

Theoretical ethanol yield

Analytical ChallengeFeed stock Pretreatment severity

Enzyme loading

Sugars to Ethanol Conversion MediapH Temperature Time Cellulase

Glycoside hydrolase

corn stover Dilute Acid C1 Ce1 G1 Yeast

switchgrass AFEX C2 Ce2 G2 e. coli

sorghum Lime C3

lawn residue LHW C4

tree thinnings C5

corn stover Dilute acid (3) severityx 5

Enzyme loading x 4 e. coli = 60 samples

SSF: 0,1, 2,3,4,5,6, 12, 24, 48, 72, 96, 120 = 13 x 60 samples

= 780 total samples to analyze

x 3 = 2340 sample injections

GC Sugar Separation

1) fructose, 2) glucose, 3) sucrose and 4) disaccharides

Ruiz-Matute A.I.; Soria A.C.; Martinez-Castro I.; Sanz M.L. J. Agric. Food Chem., 55, 2007, 7264-7269.

• Sample Derivitization• Complex Chromatogram

9

Sample run time 40 in2340 injections x 40min = 93600 min (65 Days)To analyze all samples in 7 days requires 9 GC’s

Ligand-Exchange Pb2+

1) cellubiose, 2) glucose, 3) xylose, 4) galactose, 5) arabinose and 6) mannose

Cheng C.; Tasi H.; Chang K., J. Chromatogr. A, 1119, 2006, 188-196.

10

Sample run time 20 min 2340 injections x 20min = 46800 min (32.5 Days)To analyze all samples in 7 days requires 5 LC’s

Reverse Phase (Isocratic)

Agblevor F.; Murden A.; Hames B. Biotech. Letters, 26, 2004, 1207-1210.

11

Sample run time 26 in 2340 injections x 26min = 60840 min (42.25 Days)To analyze all samples in 7 days requires 6 LC’s

Reverse Phase (Gradient)

Agblevor F.; Murden A.; Hames B. Biotech. Letters, 26, 2004, 1207-1210.

12

Sample run time 40 in 2340 injections x 40min = 93600 min (65 Days)To analyze all samples in 7 days requires 9 LC’s

13

Anion-exchange

1) arabinose, 2) galactose + sucrose, 3) glucose, 4) xylose, 5) mannose and 5) fructose

0 2 4 6 8 10 12 1402468

101214161820

Time (Minutes)

Det

ecto

r R

espo

nse

(nC

)

14

3

2

5 6

Sample run time 14 min 2340 injections x 14min = 32760 min (22.75 Days)To analyze all samples in 7 days requires 3 LC’s

14

Existing Approaches Summery•Gas Chromatography

▫Sample derivitization, multiple peaks per analyte

•Ligand-Exchange Pb2+ (HPLC-RI)▫Universal Detection, Low sensitivity, hydrolysis of

sucrose

•Anion-exchange (HPAEC-PAD)▫Selective Detection, High sensitivity, co-elution of

galactose and sucrose

Challenges in Carbohydrate Analysis

•Current quantitative techniques▫long analysis times▫limited analyte resolution▫poor stability

•NREL Challenge: Improve separation of monosacchrides and sucrose▫ sucrose, arabinose, galactose, glucose, xylose, mannose, fructose▫ ~10 min or less

15

16

Interlaboratory Collaboration for Method Validation



Biofuel Feedstocks


Stationary Phase

Modification and Method Validation

Effect of Carbonate on a PA20 Column17

0 2 4

0

6

minutes108 12 14 16

1

2

3

4

5

6

7

# of

10-

μL in

ject

ions

of

40 m

M N

a 2CO

3(aq)

I.S. ara

suc+galglu

xyl

manfru

1.0 mM NaOH(aq) @ 0.5 mL/min

DO NOT prepare NaOH eluents from sodium hydroxide pellets! Carbonate in the eluent can significantly reduce retention times for carbohydrates.

From the Product manual for the Dionex ParboPac PA20

Optimized Separation18

0

5

10

15

20

0 42 6 108

mal

tose

cello

bios

e

gluc

ose

sucr

ose arab

inos

ega

lact

ose

xylo

se man

nose

fruc

toseIS

1

IS2In

tens

ity (n

C)

Time (min)

CO32--modified PA20

1.0 mM NaOH(aq)0.5 mL/min

43 C

Sevcik, R.; Mowery, R..; Becker, C.; Chambliss, C.., J. Chromatogr. A, 1218(9), 2011, 1236-43.

1. Modify Guard and Analytical column with CO3

2-

total CO32 added = 2.8 μmol;

exchange capacity = 65 μeq/column

2. Remove guard

3. Wash for 6 min with 50 mM NaOH

4. Re-equlibrate with 1 mM NaOH (pH 11)

5. Replace guard and analyze samples

19

Temperature Impact• 30 oC to 50 oC

▫ retention time decreased 4-8%

▫ minimal change in resolution

xylose, mannose, fructose


20

Column Preparation Reproducibility

Column A* Column B* Column C**

Mod 1 Mod 2 Mod 3 Mod 1 Mod 1

AnalyteRetention

Time (min)

Retention Time (min)

Retention Time(min)


Retention Time(min)

Fucose 1.73 1.73 1.72 1.73 1.80

Sucrose 2.23 2.23 2.22 2.23 2.34

Arabinose 2.58 2.57 2.57 2.58 2.68

Galactose 2.78 2.78 2.78 2.78 2.90

Glucose 3.14 3.13 3.13 3.14 3.28

Xylose 3.58 3.57 3.56 3.58 3.73

Mannose 3.77 3.75 3.73 3.76 3.91Fructose 4.09 4.08 4.08 4.08 4.26

*Lot #006-23-017 **Lot #004-27-105


Method Validation

1. Specificity

2. Accuracy

3. Precisiona. Repeatabilityb. Reproducibility

4. Linearitya. Range

5. Robustness

Ability to assess target analyte presence

Agreement between determined and accepted values

Precision over a short period of timePrecision between laboratories

Interval between upper and lower quatitative limits

Unaffected by small, but deliberate variations in parameters

22

Experimental Design & Monitoring

0 100 200 300 400 500 600 700 800

Injection Number

Upper Action Line

Upper Warning Line

Lower Warning Line

Lower Action Line

µ

Calibration Standards

Blank

Check

Standards

Spike

Corn Stover Sampl

e 1

Corn Stover Sampl

e 2

Switchgrass Sampl

e 1

Switchgrass Sampl

e 2

LowCalibration Check Standard

High Calibration Check Standard

Repeat 3x

After 3rd repeat

Continuous system monitoring control chart

Experimental Design

23

Quantitative Performance

AnalyteInvestigated Linear Range

Line equation y = mx + b r2 LOD LOQ

(101 µg/L ) m b (101 µg/L) (101 µg/L)

Sucrose 22 - 172 0.6616 -0.0003 0.9995 1.2 3.9

Arabinose 2 - 22 0.8893 -0.0037 0.9953 0.89 2.9

Galactose 3 - 24 0.7850 -0.0011 0.9969 1.1 3.8

Glucose 46 - 371 0.9071 -0.0484 0.9996 1.2 3.8

Xylose 3 - 25 0.6632 -0.0039 0.9978 1.1 3.2

Mannose 2 - 19 0.2708 0.0004 0.9997 2.2 7.2

Fructose 27 - 223 0.4494 0.0001 0.9977 2.5 8.0


24

AccuracyAqueous extract Hydrolysate

Sugar Conc. (mg/g) Recovery (%) Sugar Conc. (mg/g) Recovery (%)Corn Stover Corn Stover

Sucrose 9.5 ± 0.5 87 ± 5 Arabinose 31.5 ± 0.2 88 ± 6Galactoseb - ± - - ± - Galactose 9.4 ± 0.2 87 ± 3Glucose 8.8 ± 0.8 82 ± 7 Glucose 379 ± 3 92 ± 1Fructose 17.5 ± 0.8 95 ± 10 Xylose 256 ± 2 79 ± 4

Switchgrass Switchgrass

Sucrose 7.04 ± 0.08 106 ± 5 Arabinose 24.5 ± 0.9 95 ± 2Galactose 0.48 ± 0.02 134 ± 15 Galactose 8.6 ± 0.5 89 ± 5Glucose 7.27 ± 0.02 91 ± 3 Glucose 386 ± 5 94 ± 2Fructose 5.2 ± 0.1 105 ± 4 Xylose 288 ± 3 99 ± 4

aConcentrations (Conc.) and Recoveries are reported as the mean value plus or minus 1 standard deviation; the number ofreplicates (n) is more than 700 independent determinations.

bAnalyte was not detected


5-Day Test of Column Stability25

1 32 4 5


1

200

800

400

600

Inje

ction

Num

ber

IS

arab

inos

ega

lact

ose

gluc

ose

xylo

se

0

Quantitative Saccharification:2-stage hydrolysis with H2SO4…

NREL LAP Determination of structural carbohydrates and lignin in biomassAqueous Extraction:ASE-200 extraction at 100 C…

NREL LAP Determination of extractives in biomass

RESULTS…

Retention time RSDs varied by less than 3% and 6%, respectively, for hydrolysate and extract samples…


26

Diverse Sample Matrix• Robustness

▫ corn stover (1st gen. feedstock)▫ poplar (2nd gen. feedstock)▫ switchgrass (3rd gen. feedstock)

• Aqueous Extract/Hydrolysates▫ diverse origins▫ complex matrix

• Near Baseline resolution

1) fucose (I.S.), 2) sucrose, 3) arabinose, 4) galactose,5) glucose, 6) xylose, 7) mannose and 8) fructose.


27

Caveats•Column modification

▫~2 hours

•High dilutions due to detector sensitivity

1 32 4 5

0

15

20

10

5Inte

nsity

(nC

)

1

43

5

6

1

43

5

6

1

43

5

6

0

Corn Stover

1:1000 1:600 1:400

1 32 4 50 1 32 4 50


Switchgrass Poplar Wood

Analtical Challenge(reduced)28

0

5

10

15

20

0 42 6 108m

alto

se

cello

bios

e

gluc

ose

sucr

ose

arab

inos

ega

lact

ose

xylo

sem

anno

sefr

ucto

seIS1

IS2In

tens

ity (n

C)

Time (min)


Sample run time 5 min 2340 injections x 5 min = 11700 min (8.125 Days)To analyze all samples in 7 days requires 2 LC’s

J. Chromatogr. A, 1218 (2011) 1236-1243

29

30



Stationary Phase




Biofuel Feedstocks

31

Feedstock Composition Analysis

•Goal ▫Characterization of water-soluble materials extracted

from selected sorghum samples 3rd generation feedstock

dedicated energy feedstocks sweet Sorghum – accumulates a high level of sugar in the

stalk (similar to sugar cane) bioenergy – structural carbohydrate

32

Why Sorghum?

•3rd generation energy crop

•Similar carbohydrate storage as sugar cane

•Drought tolerant

•Genetically Adaptable

33

Aqueous Extracts• Aqueous extracts “Extractives” materials that are soluble in water

▫ non-structural components (indirect determination) sucrose nitrate/nitrites protein

• Interferences

▫ error in structural sugar / lignin values

▫ inhibit hydrolysis

▫ inhibit fermentation

34

Notable Compound Classes in Extracts

•Corn Stover / Switchgrass

▫Free monomeric sugars

▫Non-structural oligomeric sugars

▫Phenolics

Chen, S. -F.; Mowery, R. A.; Scarlata, C. J.; Chambliss, C. K. Compositional Analysis of Water-Soluble Materials in Corn Stover. J. Agric. Food Chem. 2007, 2007, 5912-5918. Chen, S. -F.; Mowery, R. A.; Sevcik, R. S.; Scarlata, C. J.; Chambliss, C. K. Compositional Analysis of Water-Soluble Materials in Switchgrass. J. Agric. Food Chem. 2010, 58, 3251-3258.

35

Sample DescriptionSample Sample ID Production

EnvironmentBiomass Sample

Sorghum Type

Sample Maturity Comments

Bioenergy-1 08CMP-0580 College Stn. stem only Bioenergy Full season vegetative

Photoperiod sensitive

breeding line



breeding line



breeding line

Sweet-1 08CMP-0909 College Stn. whole plant Sweet Physiological maturity

Sweet sorghum variety

Sweet-2 08CMP-0903 College Stn. whole plant Sweet Physiological maturity

Sweet sorghum hybrid

Sweet-3 07CMP-1172 Weslaco whole plant Sweet Physiological maturity

Seed parent sweet sorghum

parental line

36

Percent Water-soluble Materials

Sample mean (n=3) (% dry weight) RSD (%)

Bioenergy-1 39.1 3.8



Sweet-1 39.5 0.3

Sweet-2 26.4 0.5

Sweet-3 36.9 0.3

Corn Stover 18.4 5.3

Switchgrass 13.2 1.3

Chen, S. -.; Mowery, R. A.; Scarlata, C. J.; Chambliss, C. K. Compositional Analysis of Water-Soluble Materials in Corn Stover. J. Agric. Food Chem. 2007, 2007, 5912-5918. Chen, S. -.; Mowery, R. A.; Sevcik, R. S.; Scarlata, C. J.; Chambliss, C. K. Compositional Analysis of Water-Soluble Materials in Switchgrass. J. Agric. Food Chem. 2010, 58, 3251-3258.

37

Sample Fractioning

38

Representative Chromatograms

(A) sugars, (B) alditols, (C) inorganic cations and (D) organic acids and inorganic anions in sample fractions derived from an aqueous extract of sorghum. See text for details. Peaks: 1) fucose (I.S.) 2) sucrose, 3) glucose, 4) fructose, 5) glycerol, 6) arabitol, 7) sorbitol, 8) mannitol, 9) lithium, 10) sodium, 11) ammonium, 12) magnesium, 13) potassium, 14) calcium, 15) lactic acid, 16) acetic acid, 17) propionic acid, 18) methysulfonic acid (I.S.), 19) chloride, 20) nitrate, 21) sulfate, 22) maleic acid, 23) phosphate, 24) cis-aconitic acid, 25) trans-aconitic acid, *) unknown.

39

Bioenergy-1 Bioenergy-2 Bioenergy -3 Sweet-1 Sweet-2 Sweet-3 Switchgrass* Corn Stover**0%

10%

20%

30%

40%

50%

60%

70%

80%

Free Sugars

Glycans

Organic Acids

Inorganic Anions

Alditols

Inorganic Cations

Red-Brown

Sample

% o

f tot

al E

xtra

ctiv

e

Cross Feedstock Fraction Comparison

(96%) (95%) (79%) (92%) (87%) (79%) (86%) (88%)

**Chen, S. -F.; Mowery, R. A.; Scarlata, C. J.; Chambliss, C. K. Compositional Analysis of Water-Soluble Materials in Corn Stover. J. Agric. Food Chem. 2007, 2007, 5912-5918. *Chen, S. -F; Mowery, R. A.; Sevcik, R. S.; Scarlata, C. J.; Chambliss, C. K. Compositional Analysis of Water-Soluble Materials in Switchgrass. J. Agric. Food Chem. 2010, 58, 3251-3258.

40

Percent Fermentable Sugars

Sample Type Water-Soluble Sugars(% mass in extract )

Lignocellulosic-Sugar(% mass in extract)

Bioenergy-1 74 0.8

Bioenergy-2 57 11

Bioenergy-3 48 1

Sweet-1 45 30

Sweet-2 8 61

Sweet-3 62 0.7

Corn stover 43 0

Switch grass 23 6

41

0 10 20 30 40 50-10

-5

0

5

10

15

20

25

30

35

45*33*

10*

46*

28*

36*

Bioenergy-1Bioenergy-2Bioenergy-3Sweet-1Sweet-2Sweet-3

Free Sugars (gal ethanol/ton dry feedstock)

Olig

omer

ic S

ugar

s (g

al e

than

ol/to

n d

ry fe

edst

ock)

Potential Ethanol Yield

*Theoretical gallons of ethanol per ton of dry feedstock

42

Theoretical Ethanol YieldsFeed Stock Gal Eth./Dry Ton feedstockCorn Grain 124

Mixed Paper 116

Corn Stover 113

Bagasse 112

Rice Straw 109

Hardwood Sawdust 101

Forest Thinnings 82

Cotton Gin Trash 57

Sweet-1 46

Bioenergy-1 45

Sweet-3 36

Bioenergy-2 33

Sweet-2 28

Bioenergy-3 10

43

Summary

• Studied the water-soluble materials▫Sweet and Bioenergy sorghum▫Utilizing previously developed methodologies

• Performed cross comparison▫Sweet vs bioenergy sorghum

Carbohydrates bioenergy > sweet▫Studied sorghum vs corn stover and switchgrass

Carbohydrate fraction significantly larger Red-brown fraction significantly smaller

• Carbohydrates in extracts▫ Impact ethanol yield▫ Influence production paradigms

44



Biofuel Feedstocks


Stationary Phase



45

CarboPac SA-10 Carbohydrates Column Thermo Scientific (Dionex, Corp)

Carbonate Modified CarboPac PA-20 CarboPac SA-10


Inte

nsity

(nC

)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

2

4

6

8

10

12

14

16

fuco

se

sucr

ose

arab

inos

ega

lact

ose gl

ucos

e

xylo

sem

anno

sefr

ucto

se

0 1 2 3 4 5 6 70

2

4

6

8

10

12

14

16

fuco

se

sucr

ose

arab

inos

ega

lact

ose

gluc

ose

xylo

sem

anno

se

fruc

tose

46

Collaborative CarboPac SA10 Evaluation

WasteH2O

Pump

Eluent Generator

AutoSampler

Column Oven Detector Oven

CarboPac SA10 Guard and Analytical

Columns

High VolumePulse

AmperometricDetector

1 mM NaOHpH 11

0.5 mL/min

40oC45 oC

Low VolumeSample

Injection

•Compare analytical performance against industry standard (ligand-exchange Pb+)• 21 samples of opportunity

• Analyzed by 3 independent researchers in 3 independent labs utilizing HPAE-PAD method

• Same 21 samples analyzed using industry standard method

Minimizing the Dilution Requirement47

2 mil 62 mil gasket

10 μL 400 nL sample loop

Dionex ICS-3000, AS50 autosampler, EG, SA10

hydrolysate sample analysis at 1:20 dilution…

Method Validation

1. Specificity2. Accuracy3. Precision

a. Repeatabilityb. Reproducibility

4. Linearitya. Range

5. Robustness

Ability to assess target analyte presenceAgreement between determined and accepted values

Precision over a short period of timePrecision between laboratories

Interval between upper and lower quatitative limitsUnaffected by small, but deliberate variations in parameters

3 Analysts in 3 Laboratories:▫ Thermo Scientific (Dionex Corp) [Industry]▫ NREL (National Renewable Energy Lab)

[Government]▫ Baylor University [Academia]

Specificity: Pb2+ vs SA10

0 5 10 15 20 250

20000

40000

60000

80000

100000

120000


Res

ponc

e (R

Iu)

13.5

- s

ucro

se

19.3

- a

rabi

nose

17.9

- g

alac

tose

15.7

- g

luco

se

16.8

- x

ylos

e

19.8

- m

anno

se

20.8

- f

ruct

ose

13.7

- c

ello

bios

e

0 5 10 15 20 250

5

10

15

20

25

30


Res

ponc

e (n

C)

3.4

- su

cros

e3.

7 -

arab

inos

e

3.9

- ga

lac-

tose

4.2

- gl

ucos

e4.

6 -

xylo

se

4.8

- m

anno

se

5.1

- fr

ucto

se

7.6

- ce

llobi

ose

49

Pb2+ Industry Method - 60 min total run time

SA10 - ~5 min monosacchrides / sucrose - ~8 min monosacchrides / sucrose / cellobiose

High-volume Detector Linearity50

0 1 2 3 4 5 6 7 8 9 100

1

2

3

4

5

6

7

Concentration (g/L)

Res

pons

e Fa

ctor

0.0 0.5 1.0 1.5 2.0 2.5 3.04.0

4.5

5.0

5.5

6.0

Concentration (g/L)

Res

pons

e Fa

ctor

(●)xylose ( ) arabinose

Initial linear range experiment: 15 concentrations0.05 – 10 g/L

IUPAC defines the linear range of a chromatographic detector as the range over which sensitivity is constant within 5%.

Narrowed linear range experiment:15 concentrations 0.01 – 3.0 g/L

0.0 0.5 1.0 1.5 2.00

2

4

6

8

10

12

14

f(x) = 4.35999538561798 x + 0.138936466156869R² = 0.999246019235652

R² = 0.997406069791648

Concentration (g/L)

Are

ax/A

reaI

S

Calibration: Least-squares regression6 calibration standards 0.2 – 2.4 g/L

RF = response factor = peak area

concentration

+5%

-5%+5%

-5%

Representative Biofuel R&D Samples Sample No. Sample Type Dilution

Quantitative Saccharification1,4 corn stover ND3 miscanthus ND

2 NIST bagasse† ND

Dilute-Acid Pretreatment5,8,9,12 vertical reactor 1:50

6,10 steam gun 1:107,11 slurry 1:50

Simultaneous Saccharification and Fermentation13-15 T0 1:1016-20 Tf 1:10

Surrogate Hydrolysate‡

21 dilute acid 1:50 † NIST Standard Reference Material‡ glucose, xylose, acetic acid, furfural, and 5-HMF in 0.7% H2SO4

ND = Not dilution

51

quantitative saccharification:-2 step hydrolysis-Total sugars

Dilute Acid Pretreatment:-~1% H2SO4-"deconstruct" the plant cell walls, disrupting the lignin wrapping surrounding the biomass

SSF:-Delinking of cellulose/hemicellulose into monomers-Conversion of monomers to ethanol

Surrogate:- Sample of known compounds

Visual Comparison of Data52

0 5 10 15 20 25 30 35

0

5

10

15

20

25

30

35

f(x) = 0.997446351377466 x

Site 3 (g/L)

Site

1 (g

/L)

A

0 5 10 15 20 25 30 35

0

5

10

15

20

25

30

35f(x) = 1.08193781324816 x

Site 3 (g/L)

Site

2 (g

/L)

B

0 5 10 15 20 25 30 35

0

5

10

15

20

25

30

35

f(x) = 1.08544931708684 x

Site 1 (g/L)

Site

2 (g

/L)

C

0 10 20 30 40 50 60 70 80 90 100

0102030405060708090

100

f(x) = 0.954766325105191 x

Site 3 (g/L)

Site

1 (g

/L)

D

0 3 5 8 10 13 15

0

3

5

8

10

13

15

f(x) = 1.02638070330919 x

Site 3 (g/L)

Site

1 (g

/L)

G

0 3 5 8 10 13 15

0

3

5

8

10

13

15

f(x) = 1.01612806849733 x

Site 3 (g/L)

Site

2 (g

/L)

H

0 3 5 8 10 13 15

0

3

5

8

10

13

15

f(x) = 0.990233576035499 x

Site 1 (g/L)Si

te 2

(g/L

)

I

0 10 20 30 40 50 60 70 80 90 100

0102030405060708090

100

f(x) = 0.980630757886847 x

Site 3 (g/L)

Site

2 (g

/L)

E

0 10 20 30 40 50 60 70 80 90 100

0

10

20

30

40

50

60

70

80

90100

f(x) = 1.02668261624221 x

Site 1 (g/L)

Site

2 (g

/L)

F

Glucose (A – C) Xylose (D – F) Arabinose (G - I)

Method Repeatability RSD CV HPLC-RI HPAE-PAD Site 1 Site 2 Site 3

Glucose (n = 17; 1.4 - 50.0 g/L)max 2.22% 2.24% 2.52% 6.22% 13.25%min 0.04% 0.08% 0.12% 0.49% 3.51%median 0.48% 0.98% 0.64% 2.54% 8.53%

Xylose (n = 13; 0.8 - 90.7 g/L)max 0.99% 1.85% 2.29% 6.89% 16.06%min 0.04% 0.16% 0.07% 0.44% 0.55%median 0.29% 0.56% 0.29% 2.22% 6.04%

Arabinose (n = 11; 4.5 - 11.0 g/L)max 1.99% 2.72% 6.15% 6.06% 8.85%min 0.07% 0.20% 0.13% 1.06% 0.48%median 0.37% 1.61% 0.94% 3.25% 4.00%

Galactose (n = 8; 2.2 - 2.9 g/L)max 3.13% 3.04% 22.56% 5.27% 13.34%min 0.16% 0.36% 0.86% 0.39% 4.17%median 0.44% 1.02% 2.43% 3.30% 9.88%

Fructose (n = 4; 2.6 - 2.7 g/L)max 1.50% 6.07% 3.70% 22.30% 41.87%min 0.31% 0.29% 1.22% 2.50% 11.95%median 0.67% 2.31% 2.41% 13.58% 18.77%

53

Repeatability

Reproducibility

Site 1(RSD)

Site 2(RSD)

Site3(RSD)

0.56 – 2.31% 0.29 – 2.41% 2.22 – 13.58%

Interlaboratory (CV)

4.00 – 18.77%

Accuracy: HPAE-PAD vs. HPLC-RI54

0 5 10 15 20 25 30 35

0

5

10

15

20

25

30

35

f(x) = 0.937617664116298 x

Site 1 PAD (g/L)

Site

1 R

ID (

g/L)

0 10 20 30 40 50 60 70 80 90 100

0

10

20

30

40

50

60

70

80

90

100

f(x) = 0.922387792020627 x

Site 1 PAD (g/L)Si

te 1

RID

(m

g/m

l)

0 3 5 8 10 13 15

0

3

5

8

10

13

15

f(x) = 1.1166258763179 x

Site 1 PAD (g/L)

Site

1 R

ID (

g/L)

glucose xylose arabinose

Student’s t test at the 95% confidence level - solid squares not statistically different- open squares statistically different

Q-Q plot- negative deviation for glucose and xylose- positive deviation for arabinose

Co-eluting Interferencessucrose arabinose galactose glucose xylose Mannose fructose cellobiose

g/L g/L g/L g/L g/L g/L g/L g/L

Example 1Site 1 - - - 1.50(1) 1.08(1) - - -Site 2 - - - 1.270(2) 0.83(2) - - -Site 3 - - - 1.45(4) 0.99(3) - - -Pb2+ - 0.129(2) 0.081(1) 1.523(1) 0.939(3) - - -

Example 2Site 1 - 11.0(1) - 25.3(2) 80.5(5) - - -Site 2 - 10.81(1) - 27.3(1) 80.86(7) - - -Site 3 - 11.3(5) - 24.8(5) 78.0(5) - - -Pb2+ - 12.5(2) 5.9(3) 24.31(9) 74.2(4) - 4.2(2) 2.55(2)

Example 3Site 1 - 5.97(7) 2.64(2) 13.6(1) 37.0(4) a - 2.45(4) -Site 2 - 6.09(2) 2.90(2) 15.28(8) 39.4(2) a - 3.20(4) -Site 3 - 5.6(2) 2.3(1) 13.0(5) 43(2) a - 2.39(6) -Pb2+ - 6.69(3) 3.50(3) 13.96(2) 44.287(7) - 2.36(4) 1.2(2)

Example 4Site 1 - - - 10.3(1) 82.5(3) - - -Site 2 - - - 10.7(2) 83.4(7) - - -Site 3 - - - 9.5(2) 83.2(4) - - -Pb2+ - 9.1(1) 4.925(8) 9.62(5) 75.8(5) - 1.3(2) 1.66(2)

56

0 50 100 150 200 250 300 350-5

-3

-1

1

3

5

Injection Number

Tim

e (M

in)

0 10 20 30 40 50 60 70 80 90 100-5-4-3-2-1012345

Ret

entio

n Ti

me

(%∆)

Column Stability

30-min wash: 100 mM NaOH @ 1.00 mL/min30-min equilibration: 1 mM NaOH @ 1.5 mL/min

Improvements57

1. Long analysis times – SA10 (or CO32-modified PA20) stationary phase offers a

significant reduction in analysis time (two-fold or more) compared to current practice…

2. Dilution…a caveat of PAD – Dionex has engineered commercially-available hardware that minimizes the need for sample dilution (i.e., low-volume injector and high-volume detection cell)…

3. Method transfer – reasonable agreement with sugar concentrations determined by the Pb2+ column method, excellent to good intra- and interlab repeatability

58

SA10 Method Caveats•Stability

▫Steady loss of analyte retention time Wash

60 min effectively doubles run time

59

Summary• Developed and validated a robust method, carbonate-

modified PA20 anion-exchange column, to analyse carbohydrates (monosaccharides and sucrose) ~5 min.

• Utilized the CM-PA20 to analzye aqueous extracts from potential bioenergy sorghum feedstocks

• Lead the interlaboratory method validation of the CarboPac SA10 carbohydrate separation method (which was inspired by the CM-PA20)

Documents

Improved Analytical Methods for Carbohydrate R&D