GAS CHROMATOGRAPHY-FLAME IONISATION DETECTOR METHOD …palmoilis.mpob.gov.my/publications/joprv25april2013-Bonnie.pdf · IONISATION DETECTOR METHOD FOR DETERMINATION OF ... The follow

JOURNAL OF OIL PALM RESEARCH 25 (1) (APRIL 2013)

36

GAS CHROMATOGRAPHY-FLAME IONISATION DETECTOR METHOD FOR DETERMINATION OF CARBON CHAIN

LENGTH DISTRIBUTION OF PALM-BASED FATTY ALCOHOL

BONNIE TAY YEN PING*

* Malaysian Palm Oil Board, P. O. Box 10620, 50720 Kuala Lumpur, Malaysia. E-mail: [email protected]

INTRODUCTION

Fatty alcohols (FAL) are aliphatic alcohols comprising chains of 8 to 22 carbon atoms. They usually have an even number of carbon atoms with one alcohol group attached to the terminal carbon. Generally, they are saturated, unsaturated and some are branched. The FAL are the most important basic oleochemical used in the production of surfactants. In 2010, the total export of palm-based oleochemicals

in Malaysia amounted to 2.21 million tonnes, with 21.2% of the total as palm-based FAL. There are two routes in the commercial production of FAL. It may involve the splitting of fat or vegetable oils followed by hydrogenation to yield alcohols or reacting the oil/fat with methanol to produce methyl esters, and then hydrogenated to alcohols (Ahmad and Thambirajah, 1996). Palm oil and palm kernel oil (PKO) are useful raw materials used commercially for FAL production. This is because palm oil is a source of (C16-C18) chain lengths which produce optimal detergency. On the other hand, PKO is a source for C12-C14 chain lengths which is optimal for foamability and solubility (Ahmad and Thambirajah, 1996). The major uses (about 75%) of FAL are for the production of non-ionic and anionic surfactants

ABSTRACT

A simple and rapid gas chromatography (GC) method using a flame ionisation detector (FID) was

developed for the determination of the carbon chain length (CCL) distribution of palm-based fatty alcohols

(FAL). The method involved a single-step sample preparation, where the FAL was silylated with N,O,

-bis(trimethylsilyl)trifluoroacetamide with 1% trimethylsilyl chloride in N,N- dimethylformamide into

trimethylsilyl ethers, followed by its direct injection into the GC-FID system. The precision and long-term

precision of the GC-FID instrument performance was high as shown by coefficient of variation (CV) of

lower than 1.00% (retention time) and 1.80% (peak areas) from multiple injections of three sets of FAL

standards mixture (CCL ranging from C8 to C20) for three consecutive days. The results obtained for the

CCL distribution (as relative percentage of peak area) for commercial FAL were close to the known values

provided by the manufacturer, showing good accuracy of the method. The CV obtained from analyses of

four replicates commercial FAL and its blends was lower than 1.5% indicating that the method has good

repeatability.

Keywords: gas chromatography-flame ionisation detector, carbon chain length, fatty alcohols.

Date received: 10 July 2012; Sent for revision: 15 August 2012; Received in final form: 7 November 2012; Accepted: 19 December 2012.

Journal of Oil Palm Research Vol. 25 (1) April 2013 p. 36-43

Vol 25 Bonnie.indd 36 2/4/2013 3:20:08 PM

37

GAS CHROMATOGRAPHY-FLAME IONISATION DETECTOR METHOD FOR DETERMINATION OF CARBON CHAIN LENGTH DISTRIBUTION OF PALM-BASED FATTY ALCOHOL

such as alcohols ethoxylates, alcohols sulphates and alcohols ether sulphates derived from the mid-cut (C12-C14), and they are used in detergents, personal cares and washing products. On the other hand, short chain (C8 – C10) alcohols are used as plasticisers, perfumes intermediates, wood preservatives, and textile penetrants (Suyenty et al., 2009). The carbon chain length (CCL) distribution of a FAL is important because it determines the types of applications it can be used for. The first report on FAL CCL distributionanalyses was for Jojoba oil using gas chromatography-flame ionisation detection(GC-FID) with a stainless steel column packed with non-polar 20% Apiezon L on chromosorb W. FAL with chain length ranging from C14:0 to C24:1 was quantified by percentage weight(Miwa, 1971). The follow up study on fatty acids and alcohols chain-length analyses in Jojoba oil also reported the use of GC-FID with two packed columns, one with 5% Apiezon L on Chromosorb W and the other with 3% Silar-5 CP (Spencer et al., 1977). The sample preparation involved converting the FAL in Jojoba oil into tri-methylsilyl (TMS) ethers prior GC-FID analyses. The FAL CCL ranging from C14:0 to C26:1 was reported (Spencer et al., 1977). The CCL distribution of derivatives of FAL such as FAL polyethoxylate and its sulphated forms was also reported using GC separation (Lee and Puttnam, 1966). In the study, a packed GC Celite column containing 10% (w/w) polyethyleneglycoladipate was used for analyses (Lee and Puttnam, 1966). Recent publications on CCL distributions of vegetable-based FAL are scarce. To date, there is no published work that we know of on the determination of CCL distribution in FAL, and palm-based FAL producers rely mainly on their internal (and therefore unpublished) methods for quality control monitoring of their products. This study describes a method with a simple sample preparation for the analysis of CCL distribution for palm-based FAL using GC-FID. This new method used a commercially available and commonly used fused silica capillary GC column.

MATERIALS AND METHODS

Standards, Chemical Reagent and Apparatus

Capryl alcohol (C8:0); decyl alcohol (C10:0); lauryl alcohol (C12:0); myristyl alcohol (C14:0); cetyl alcohol (C16:0); stearyl alcohol (C18:0); arachidyl alcohol (C20:0) and behenyl alcohol (C22:0) standards with assay of above 99% were purchased from Sigma Aldrich Inc. (St. Louis, MO, USA). Single FAL with 98% purity with CCL of C8, C10, C12, C14,

C16 and C18, and blends of FAL, C8-C10, C12-C14 and C16-C18 were obtained from a local reputable company in Malaysia. The silylating reagent N, O-Bis (Trimethylsilyl) trifluoroacetamide (BSTFA)with 1% trimethylsilyl chloride was purchased from Fluka Chemicals (Buchs, Switzerland). The N, N-dimethylformamide (DMF) was purchased from Mallinckrodt Baker Inc, Phillipsburg, New Jersey, USA. The Tempette TE-8D/RD-12 refrigerated water-bath used was obtained from Techne (Cambridge) Limited, New Jersey, USA and the adjustable single channel pipette (100 – 100 μl) was obtained from Brand GmbH + CO KG (Wertheim, Germany).

Fatty Alcohols Standards and Commercial Fatty Alcohol Samples Preparation for Gas Chromatography-flame Ionisation Analyses

Three sets of FAL standards mixtures (with CCL ranging from C8 – C20) were prepared to test precision of the instrument for three consecutive days. About 0.01 g of each of the individual FAL standard (CCL ranging from C8 to C20) was weighed into a 5-ml vial. Then 2 ml of DMF and 0.5 ml of the silylating reagent BSTFA were pipetted into the vial containing mixtures of seven FAL standards. The vial containing the mixtures was securely sealed, and homogenised on a vortex mixer. Then the vial was heated at 60oC for 30 min in a water bath. When the mixtures had cooled, they were transferred into a 2-ml GC vial, and 1 μl of it was injected into a GC-FID instrument. For the FAL commercial samples, about 0.01 g was weighed accurately into a 5-ml vial, and then silylated as described previously.

Gas Chromatography-flame Ionisation Analyses

The instrument used was an Agilent model 7890 GC system with FID detector and a split/splitless injection port. An Agilent Technologies (Palo Alto, CA USA) HP-5 fused-silica capillary column (30.0 m length x 0.32 mm internal diameter), coated with 5% phenyl methyl siloxane with a film thickness of 0.25 μm. The GC flowrateswere: columnflow (helium= 0.8mlmin-1) andgasesflowfortheFIDsystemwereH2=30ml min-1, air = 400mlmin-1. The split ratio was set at 100:1 with an inlet temperature of 280oC. The GC temperature programming was as follows: detector temperature 300oC; initial oven temperature 80oC with no initial holding time, ramping rate, 5oC min-1 to 250oC, and held for 16 min. The column pressure was 11.35 psi. The total run time for the FAL standards mixtures was 50 min. The CCL distributions (relative percentage) were determined from peak areas of the GC chromatogram.



38

RESULTS AND DISCUSSION

Figure 1 shows the GC-FID chromatogram for mixtures of seven types of FAL standards. The GC condition developed allowed baseline separation of the peaks for CCL ranging from C8 to C20. The intermediate precision of the GC instrument was monitored by 10 (Day 1 and Day 2) and eight (Day 3) injections of three different preparations of standards mixtures at different days (Brettell and Lester, 2004; Mcdowall, 2001; ICH, 1996). The average retention time (Table 1) and peak area (Table 2) for multiple injections of the FAL standards showed low standard deviation (SD) and has coefficientofvariation (CV)of lower than1.00%and 1.80%, respectively, over three consecutives days of analyses. This showed that the GC injection repeatability (intra-day) and long-term precision (inter-day) was good. Several types of single and blended FAL, commonly produced by local manufacturers were analysed in four replicates using this method and

the percentage composition of the CCL distribution are shown in Tables 3 and 4. The CCL GC chromatogramprofileswereshowninFigures 2 to 7 (commercial FAL with 98% purity) and Figures 8 to 10 (commercial FAL blends). The commercial FAL with 98% purity showed a single sharp major peak whereas blended FAL showed two well resolved (baseline separation) major peaks. Tables 3 and 4 show the CCL distribution of commercial FAL and its blends. Commercial FAL with CCL ranging from C8 to C18 has a known composition of 98% purity (determined by a different procedure and the data provided by the supplier) was used to test accuracy of the method (ICH, 1996). The results obtained showed that percentage difference to the known values were lower than 1.6%. This showed good accuracy of themethod.TheCVobtained from four replicateanalyses of each of the sample was found to be below 1.5%. These results show that the intra-day precision/repeatability was good.

TABLE 1. PRECISION DATA OF FATTY ALCOHOL STANDARDS MIXTURES IN TERMS OF RETENTION TIME FOR GAS CHROMATOGRAPHY-FLAME IONISATION DETECTOR (GC-FID) INSTRUMENT

Carbon chain length Mean retention time (min) ± SD (% CV)

` Day 1*N = 10

Day 2*N = 10

Day 3*N = 8

C8 10.078 ± 0.001 (0.001%) 10.078 ± 0.001 (0.001%) 10.077 ± 0.001 (0.001%)C10 14.668 ± 0.001 (0.007%) 14.670 ± 0.001 (0.007%) 14.669 ± 0.001 (0.007%) C12 19.168 ± 0.001 (0.005%) 19.169 ± 0.001 (0.005%) 19.166 ± 0.001 (0.005%)C14 23.340 ± 0.002 (0.009%) 23.342 ± 0.001 (0.004%) 23.337 ± 0.001 (0.004%)C16 27.170 ± 0.002 (0.007%) 27.170 ± 0.001 (0.004%) 27.166 ± 0.002 (0.007%)C18 30.690 ± 0.002 (0.007%) 30.688 ± 0.002 (0.007%) 30.686 ± 0.001 (0.003%)C20 33.941 ± 0.001 (0.003%) 33.941 ± 0.001 (0.003%) 33.938 ± 0.003 (0.009%)

Note: * From three different preparations of fatty alcohol standard mixtures.CV–coefficientofvariation. SD – standard deviation.

Retention time (min)Note: pA: gas chromatography detector response.

Figure 1. Gas chromatogram profile of fatty alcohol standard mixtures.


39


TABLE 2. PRECISION DATA OF FATTY ALCOHOL STANDARDS MIXTURES IN TERMS OF PEAK AREA FOR GAS CHROMATOGRAPHY (GC) INSTRUMENT

Carbon chain length Mean retention time (min) ± SD (% RSD)

Day 1*N= 10

Day 2*N= 10

Day 3*N= 8

C8 14.21 ± 0.18 (1.27%) 14.99 ± 0.07 (0.47%) 15.39 ± 0.22 (1.43%) C10 13.77 ± 0.09 (0.65%) 14.97 ± 0.04 (0.27%) 15.25 ± 0.14 (0.92%) C12 13.66 ± 0.04 (0.29%) 13.99 ± 0.02 (0.14%) 13.85 ± 0.07 (0.51%)C14 14.31 ± 0.03 (0.21%) 14.78 ± 0.01 (0.07%) 13.53 ± 0.04 (0.29%)C16 15.01 ± 0.07 (0.47%) 14.42 ± 0.03 (0.21%) 14.36 ± 0.07 (0.49%)C18 14.75 ± 0.10 (0.68%) 13.23 ± 0.04 (0.30%) 13.73 ± 0.13 (0.95%)C20 14.28 ± 0.12 (0.84%) 13.62 ± 0.05 (0.37%) 14.21 ± 0.22 (1.55%)

Note: * From three different preparations of fatty alcohol standard mixtures. CV–coefficientofvariation. SD – standard deviation.

TABLE 3. CARBON CHAIN LENGTH DISTRIBUTION (relative percentage) AND STATISTICAL DATA OF 98% PURITY COMMERCIAL FATTY ALCOHOLS

Commercial fattyalcohol type

Percentage composition provided by supplier (%)

Percentage area from proposed GC methodMean (%) ± SD (CV %)

% Difference from values provided by supplier

FAL C8-98% 98 98.7 ± 0.31 (0.32%) 0.71

FAL C10-98% 98 99.5 ± 0.30 (0.30%) 1.53

FAL C12-98% 98 97.7 ± 1.06 (1.08%) 0.31

FAL C14-98% 98 98.1 ± 0.8 (0.82%) 0.10

FAL C16-98% 98 98.0 ± 1.1 (1.12%) 0.00

FAL C18-98% 98 98.3 ± 0.6 (0.61%) 0.31

Note:*SD–standarddeviation,CV–coefficientofvariation. GC – gas chromatography.

TABLE 4. CARBON CHAIN LENGTH DISTRIBUTION (relative percentage) AND STATISTICAL DATA OF COMMERCIAL BLENDED FATTY ALCOHOLS

FAL C8-C10 ± SD (CV%)N= 4

FAL C12-C14 ± SD (CV%)N = 4

FAL C16-C18 ± SD (CV%)N = 4

C8:0 C:10 C12:0 C14:0 C16:0 C18:0

53.9 ± 0.36 (0.67%) 44.2 ± 0.57(1.28%)

73.5 ± 0.36(0.49%)

23.2 ± 0.24(1.03%)

28.0 ± 0.38 (1.36%)

67.1 ± 0.88(1.31 %)

Note:*SD–standarddeviation.CV–coefficientofvariation.

Retention time (min) Figure 2. Gas chromatogram profile of fatty alcohol C8 (98% purity).



40



Retention time (min)Figure 5. Gas chromatogram profile of fatty alcohol C14 (98% purity).


41



Retention time (min)Figure 7. Gas chromatogram profile of fatty alcohol C18 (98% purity).

pA

Retention time (min)Figure 8. Gas chromatogram profile of fatty alcohol C8:C10 blend.

pA



42

Retention time (min)Figure 10. Gas chromatogram profile of fatty alcohol C16:C18 blend.

pA

Retention time (min)Figure 9. Gas chromatogram profiles of fatty alcohol C12:C14 blend.

CONCLUSION

The proposed method for determining the CCL distribution for single and blended FAL can be used for the routine analysis of FAL CCL composition because the sample preparation is simple, the test method is repeatable, and it uses commonly available silica capillary GC column. The GC profilesobtained can alsobeused tomonitor thepresence of other cross-over contamination from other FAL or other impurities sensitive to the FID detector. There may be internal methods for CCL developed by some local FAL manufacturers for quality control purposes, however the method is proprietary to the company concerned and is not available in the public domain. Therefore, this method can be a useful alternative for local manufacturers who still do not have the method for FAL CCL analyses.

ACKNOWLEDGEMENT

The author would like to thank Ms Bahriah Bilal and Ms Khomsatun Telepok for their technical assistance and the Director-General of MPOB for permission to publish this article.

REFERENCES

AHMAD, S and THAMBIRAJAH, J J (1996). Downstream products from oils/fats and basic oleochemicals. Proc. of the 1996 International Palm Oil Congress (Soaps and Detergents). p. 560.

BRETTELL, T A and LESTER, R E (2004). Chapter 17:ValidationandQA/QCofgaschromatographicmethods. Modern Practice of Gas Chromatography (Robert L Grob and Eugene F Barry eds.). Fourth edition. John Wiley & Sons, Inc. p. 978-979.


43


I N T E R N A T I O N A L C O N F E R E N C E O F HARMONISATION (1996). Guidance for Industry Q2B Validation of Analytical Procedures: Methodology. United States Department of Health and Human Services, Food and Drug Administration. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM073384.pdf, p. 1-10, accessed on 10 October 2012.

LEE, S and PUTTNAM, N A (1966). GLC determination of chain length distribution in fatty alcohol polyethoxylates and sulfated derivatives. J. Amer. Oil Chem. Soc. Vol. 43(12): 690.

MCDOWALL, R D (2001). Chapter 6: Method validation in gas chromatography. Gas Chromatographic Techniques and Applications (Handley, A J and Adlard, E R eds.). SheffieldAcademic Press CRC. p. 200.

MIWA, T K (1971). Jojoba oil wax esters and derived fatty acids and alcohols: gas chromatographic analyses. J. Amer. Oil Chem. Soc. Vol. 48(6): 259-264.

SPENCER, G F; PLATTNER, R D and MIWA, T (1977). Jojoba oil analysis by high pressure liquid chromatography and gas chromatography/mass spectrometry. J. Amer. Oil Chem. Soc. Vol. 54(5): 187-189.

SUYENTY, E; SUTANTO, E and LIE, A (2009). Applications of basic oleochemicals: old and new. Proc. of the PIPOC 2009 International Palm Oil Congress - Oleo & Speciality Chemical Conference. MPOB, Bangi. p. 79-82.


GAS CHROMATOGRAPHY-FLAME IONISATION DETECTOR METHOD …palmoilis.mpob.gov.my/publications/joprv25april2013-Bonnie.pdf · IONISATION DETECTOR METHOD FOR DETERMINATION OF ... The follow

Documents