1

Optimized Method for Analysis of Commercial and Prepared Biodiesel using UltraPerformance Convergence Chromatography (UPC2)Mehdi Ashraf-Khorassani,1 Giorgis Isaac,2 and Larry T. Taylor1

1 Department of Chemistry, Virginia Tech, Blacksburg, VA2 Waters Corporation, Milford, MA, USA

IN T RO DU C T IO N

Commercial biodiesel is usually obtained by trans-esterification of vegetable

oil using methanol or ethanol as a reactant. The resulting products are fatty acid

methyl esters (FAMEs) or fatty acid ethyl esters (FAEEs). As an energy source,

biodiesel should meet certain criteria. For example, the level of impurities such

as triacylglycerol, diacylglycerol, monoacylglycerol, and free glycerol should be

0.2% to 0.8% for acylglycerols and 0.02% for glycerol. These impurities vary in

polarity, solubility, and volatility.

The standard procedure for analysis of biodiesel impurities usually involves

gas chromatography (GC) , but one must resort to either high temperature

(370 °C) or pre-derivatization for conversion to more volatile products in

order to be successful. High temperature GC usually requires harsh conditions,

causing degradation and/or rearrangement of certain analytes. Furthermore,

derivatization usually involves many steps to prepare the sample which can be

time consuming.

UltraPerformance Convergence Chromatography™ (UPC2®) is a novel technology

that applies the performance advantages of UPLC® to supercritical fluid

chromatography (SFC). Combining the use of supercritical CO2 with sub-2-µm

particle columns, UPC2 represents an analysis technique that is orthogonal to

reversed-phase liquid chromatography (RPLC) and can be used to solve many

troublesome separations that challenge conventional LC or GC analyses.

Previously, a UPC2 method that uses an ACQUITY UPC2™ HSS C18 SB Column

and evaporative light scattering (ELS) detection was developed for analysis of

biodiesel and impurities in spiked model mixtures.1 The method to analyze both a

series of biodiesels prepared in-house from tobacco seed oil2 and a commercially-

available B100 biodiesel is described here. In addition, to obtain a cleaner

biodiesel, a simple purification method employing gravity flow with a prepared

glass column packed with bare silica was applied to an in-house synthesized

biodiesel using hexane and ethanol as tandem eluents.

WAT E R S SO LU T IO NS

ACQUITY UPC2 System with

an ACQUITY UPC2 ELS Detector

ACQUITY UPC2 HSS C18 SB Column

K E Y W O R D S

SFC, convergence chromatography

(CC), UltraPerformance Convergence

Chromatography, UPC,2 triacylglycerols,

TAG, glycerol, biodiesel, biofuel, fatty

acid alkyl esters, FAMEs, FAEEs, ELSD

A P P L I C AT IO N B E N E F I T S ■■ No analyte pre-column derivatization

thus eliminating artifact formation

■■ Complete separation of all components

in less than 12 minutes

■■ No thermal degradation of the analytes

2Optimized Method for Analysis of Commercial and Prepared Biodiesel using UPC2

E X P E R IM E N TA L

Sample preparation

Octadecyl glycerol standards were purchased

from Sigma-Aldrich (St. Louis, MO). Biodiesel

B100 was obtained from a commercial source.

Synthetic biodiesel derived from tobacco seed oil

(R.J. Reynolds Tobacco Co., Winston-Salem, NC)

was prepared in- house via trans-esterification

in ethanol using three different batches of oil.

Biodiesel purification was performed using

column chromatography on bare silica via gravity

flow. Silica Gel (60Å, 200-420 mesh) was also

purchased from Sigma Aldrich.

Method conditions

System: ACQUITY UPC2

Column: ACQUITY UPC2 HSS C18 SB,

1.8 µm, 3.0 x 150 mm

Sample: 5% sample in DCM/MeOH

ABPR: 1500 psi

Column temp.: 25 °C

Injection volume: 2-8 µL

Sample solvent: DCM/MeOH (50/50)

Flow rate: 1-2 mL/min

Mobile phase A: Compressed CO2

Mobile phase B: Acetonitrile/methanol

(90/10)

Make up solvent: IPA

Make up flow rate: 0.2 mL/min

Gradient: See Figure 1 for

different methods

Detectors: ACQUITY UPC2 PDA

210 nm, Ref.

400 to 500 nm

ACQUITY UPC2 ELS: Nebulizer: Cooling, Drift

Tube: 50 °C, Gas Pressure:

40 psi, Gain: 10, Make up

flow was added to UPC2

column effluent, Split for

BPR and ELSD was 1:3

R E SU LT S A N D D IS C U S S IO N

Figure 1 shows the separation of a mixture of C18 triacylglycerol, diacylglycerol,

and monoacylglycerol, plus free glycerol spiked into model biodiesel, which

was composed of a mixture of FAEEs. Different gradient elution profiles and flow

rates were employed in order to achieve faster analysis with minimal loss in

resolution of all impurities. Figure 1 - Method A was used originally to develop

the separation in a previous application note.1 The analysis time for this method

was 17 minutes. However, in order to reduce the analysis time, the initial flow

rate as well as the starting and ending modifier percentages, were adjusted

to afford separation of all compounds in less than 10 minutes (Figure 1 -

Method B). Finally, the gradient time was further reduced in order to achieve

an analysis time of less than 5 minutes with still adequate resolution of all

impurities. Method C is, therefore, suitable for analysis of biodiesel with a

lesser amount of impurities. However, due to the high concentration of synthetic

biodiesel derived from tobacco seed oil in the sample, as well as elution of

various triacylglycerols close to the tail of the biodiesel peak which might

interfere with quantification, the second method (Method B) was used for all

further analyses. A faster separation with little loss in analyte resolution would

translate into higher sample throughput for monitoring product quality.

Time-0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00

LSU

0.000

100.000

200.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00

LSU

0.000

100.000

200.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00

LSU

0.000

200.000

400.000

Modelbiodiesel_M12 (4) ELSD SignalRange: 19992.37

7.72

8.99

12.10

13.45

Modelbiodiesel_M22 (4) ELSD SignalRange: 19991.32

3.39

3.55

5.44

4.28

Modelbiodiesel_M31 (4) ELSD SignalRange: 19991.31

2.41 2.90

3.263.62

Method C

Method A

Method B

Glycerol MAGs

DAGs

TAGs FAEE

Glycerol

Glycerol

MAGs

MAGs

DAGs

DAGs

TAGs

TAGs

FAEE

FAEE

Figure 1. UPC2 Single Injection of a mixture of model biodiesel, glycerol, and C18 acylglycerols with different gradients.

Gradient elution for Method A: T= 0 min, 98:2 CO2 /Modifier, T=10 min, 80:20, T=12 min, 50:50, T=15 min, 50:50, T=15.5 min, 98:2, T=17 min, 98:2, Flow: 1.0 mL/min., Oven temp.: 25 °C, Modifier 90:10 CH3CN/MeOH.

Gradient Elution for Method B: T=0 min, 90:10 CO2 /Modifier, T=10 min, 50:50, T=11 min, 50:50, T=11.1 min, 90:10, T=12 min, 90:10, Flow: 1.2 mL/min., Oven temp.: 25 °C, Modifier 90:10 CH3CN/MeOH.

Gradient Elution for Method C: T=0 min, 90:10 CO2 /Modifier, T=2 min, 50:50, T=6 min, 50:50, T=6.1 min, 90:10, Flow: 1.2 mL/min., Oven temp.: 25 °C, Modifier 90:10 CH3CN/MeOH.

3Optimized Method for Analysis of Commercial and Prepared Biodiesel using UPC2

Method B was therefore used to determine the purity of synthetic biodiesels prepared in-house derived from

three tobacco seed oil lots (Figure 2). As can be observed, all three biodiesel samples showed the presence of

multiple impurities (retention time window of 2 to 4 min), as well as monoacylglycerols at a retention time of

6.2 min. Total impurities were highest in batch 2 and lowest in batch 3. For comparison, a commercial B100

biodiesel sample was obtained and analyzed (Figure 3A). As can be observed, monoacylglycerols were also

detected in the commercial B100 biodiesel based on retention time. Other impurities were observed in the

2 to 4 minute retention window, but they were not identified. Figure 3B shows the separation of model

biodiesel (i.e., pre-mixed FAEEs) spiked with different known impurities for comparison.

Figure 2. UPC2 chromatograms of three different batches of biodiesel derived from tobacco seed oil.

Figure 3. Injection of commercial biodiesel B100 and model biodiesel mix spiked with different impurities.

Time-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

10.000

20.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

10.000

20.000

30.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

10.000

20.000

30.000

3rdbatch (4) ELSD SignalRange: 19991.03 1.55

1.79

6.162.49 3.31 4.85

2ndbatch1 (4) ELSD SignalRange: 19991.03 1.32

1.72

2.77

3.23 6.16

1stbatch1 (4) ELSD SignalRange: 19991.04 1.46

1.782.56

2.78

6.163.18 4.21 4.96

Synthetic biodieselUnknown impurities

MAGs

Batch 3

Batch 2

Batch 1

Time-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

20.000

40.000

60.000

80.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

20.000

40.000

60.000

80.000

100.000

MixstdBiodieselF5 (4) ELSD SignalRange: 19991.32

1.23

3.67

1.49 4.05

8.786.17

B100Biodiesel (4) ELSD SignalRange: 19991.02 1.29

1.78

2.03

2.54 6.18 6.62

Biodiesel

MAGs

Model mix

Commercial B100

A

B

TAGs

DAGs

MAGs Glycerol

4Optimized Method for Analysis of Commercial and Prepared Biodiesel using UPC2

It is important for manufacturers and industries to have pure biodiesel with no glycerol or acyglycerols as

long as their removal is commercially viable. Therefore, it is essential to efficiently remove these impurities.

Currently, the industry makes use of the density difference to separate the bulk of the glycerol impurity. In this

application, we have developed a method to remove all impurities, including glycerol, from synthetic biodiesel

via a simple, two-step column chromatographic process. More specifically, biodiesel derived from tobacco seed

oil and the associated impurities were passed through a bare silica column and eluted via gravity first using

hexane and then using ethanol.

Each fraction was then evaporated and the resulting sample was dissolved in MeOH/DCM (1:1). Analysis of

the fractions was performed using the UPC2 method described in Figure 1B. Figure 4 shows the separation

of the synthetic biodiesel before purification, after purification with hexane as the eluting solvent, and after

purification with ethanol as the eluting solvent. As can be observed, nearly pure synthetic biodiesel was

obtained via the hexane fraction. Peaks eluting in the same retention window as monoacylgycerols were

separated and detected in the ethanol fraction of each biodiesel sample. This purification process should be

suitable for industries that are required to have pure biodiesel (FAEE or FAME). Using Method C, UPC2 can

easily be used to show the purity of sample in less than 5 minutes. The ability to inject hexane and ethanol

fractions directly onto UPC2 allows for rapid assessment of the purity of biodiesel.

Time-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

20.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

10.000

20.000

30.000

-0.00 2.00 4.00 6.00 8.00 10.00 12.00

LSU

0.000

10.000

20.000

30.000

PurifiedTE2nddfrac10mlTSO1 (4) ELSD SignalRange: 542.59

2.291.19

6.25

4.272.80 3.99 6.73

PurifiedTE1stdfrac10mlTSO1 (4) ELSD SignalRange: 19991.13

0.00

1.62

2.032.15

2.39

TEtobaccooil1 (4) ELSD SignalRange: 19991.02 1.45 6.162.55

1.78

2.764.14

3.204.46 6.62

10.78

Before purification

After purification using hexane

After purification using ethanol

Figure 3. Injection of commercial biodiesel B100 and model biodiesel mix spiked with different impurities.

Waters Corporation 34 Maple Street Milford, MA 01757 U.S.A. T: 1 508 478 2000 F: 1 508 872 1990 www.waters.com

References

1. Ashraf-Khorassani M, Taylor L.T. Analysis of Impurities in Model Biodiesel using UltraPerformance Convergence Chromatography (UPC2). Waters Application Note no. 720004872EN, 2013.

2. Mukhtar H, Ullah H, Mukhtar A. “Fatty acid composition of tobacco seed oil and synthesis of alkyd resin.” Chin. J. Chem., 25 (2007) 705.

CO N C LU S IO NS

A single ACQUITY UPC2 HSS C18 SB Column packed with 1.8-µm

particles was successfully used to separate a mixture of glycerol,

acylglycerols, and FAEEs (i.e., model biodiesel). A gradient

of CO2 and acetonitrile/methanol served as the mobile phase

and detection was performed by ELS detection. The optimized

method provided fast separation and detection of all impurities in

biodiesel without employment of sample derivatization or sample

preparation. The method was used to detect impurities in three

different synthetic biodiesels derived from tobacco seed oil and a

commercial biodiesel. A simple, two-step, column chromatographic

method using bare silica afforded nearly pure biodiesel fractions.

This level of biodiesel purity can be determined using UPC2 in less

than five minutes.

Waters, T he Science of What’s Possible, UPC2, and UPLC are registered trademarks of Waters Corporation. ACQUITY UPC2 and UltraPerformance Convergence Chromatography are trademarks of Waters Corporation. All other trademarks are property of their respective owners.

©2013 Waters Corporation. Produced in the U.S.A. March 2014 720004896EN AG-PDF