石灰精油中成分分析检测方案

699 Mondello， Shellie， Casilli， Quinto Tranchida，Marriott， Dugo700 Luigi MondelloRobert Shellie2Alessandro Casilli1Peter Quinto TranchidalPhilip Marriott² Giovanni Dugo' 'Dipartimento Farmaco-chimico，Facoltá di Farmacia， Universitadi Messina， Viale Annunziata，98168 Messina， Italy2Australian Centre for Researchon Separation Science，Department of AppliedChemistry， RMIT University，GPO Box 2476V Melbourne，Victoria 3001， Australia Ultra-fast essential oil characterization by capillaryGC on a 50 umlD column A5m x 50 um capillary column with 0.05 um stationary phase film thickness， with acalculated efficiency of almost 20，000 plates per metre (under optimum conditions)，was used for very fast high resolution GC analysis of lime essential oil. The total anal-vsis time of this volatile essential oil was less than 90 s. Fast GC is shown to beappropriate for essential oil quality assurance analysis， and quantitative results of keycomponents are comparable with those obtained by using conventional GC analysis.The fast GC analysis is approximately 33 times faster than the conventional GCmethod. Key Words： Fast capillary GC；Essential oils； Lime oil Received： April 28， 2003； revised： July 10， 2003； accepted： July 16，2003 DOI 10.1002/jssc.200301602 1 Introduction During the past decade there have been extensive effortsto minimise analysis times in gas chromatography. Arecent discussion reviewed theoretical aspects behind themost popular contemporary approaches towards speed-ing up GC analysis [1]， and several practical aspects，including instrumental requirements and suggestions formethod development are given in reference [2]. It is generally accepted that there are three general routesto achieve the goal of faster GC. One approach (I) wouldbe to employ a column operated under non-optimal condi-tions giving lower efficiency such that it just satisfies therequired number of theoretical plates to provide resolutionof a critical pair of components. An alternative to this (II)would be to maximise the selectivity of either the chroma-tographic system or to use selective detection. A morefavourable approach (III) is to implement a method thatreduces the analysis time， but maintains (approximately)constant resolution. For multi-component mixtures such as essential oils it isimpractical to employ a method that drastically reducesthe resolution. Statistical overlap theory shows that thereis already insufficient peak capacity to resolve all of thecomponents in essential oils， even when the most efficientpresently available capillary columns are used [3]. Options for selective detection can realistically onlyinclude mass spectrometric detection for essential oils.The fast GC-TOFMS (time-of-flight MS) analysis of agrapefruit essential oil， using a 14 m×180 um capillarycolumn ensemble， which was temperature programmed ( Correspondence： Luigi Mondello， Dipartimento Fa r maco-chimi- co， Facolta di Farmacia， Universita di Messina， Viale Annunziata， 98168 Messina， Italy. Phone： +39 090 6766536. ) ( Fax： +39 090 676 6532. E-mail： Imondello@pharma.unime.it. ) at 50 K/min， was reported recently [4]. The analysis of thevolatile essential oil fraction was completed within 200 s；however， the chromatogram contained severely overlap-ping peaks， and the authors relied on pressure tuning andmass-spectral deconvolution to identify and quantify indi-vidual components. Significant speed gains can be achieved by using a shortcolumn， and by using a higher than usual carrier gas velo-city. Faster essential oil analyses have been achieved byreducing the length of 250 um ID columns [5]， but there is apenalty in that the shortened column has a reduced resolv-ing power. The appropriate choice of column is criticalwhere the goal is to maintain the highest resolution in avery short time. An excellent discussion of considerationsfor column selection for essential oils analysis is given inreference[6]， where it is shown both experimentally andtheoretically that a 25mx250um column with dr=0.25 um will have the same number of theoretical plates(n=100，000)as a 10 m×100 um column with dr=0.1 um. Theoretical and practical aspects of method translation，where the aim is to produce a scaled version of the originalchromatogram， have been developed by Blumberg andco-workers [7， 8]. The usefulness of method translationfor fast GC analysis of nutmeg and lemon essential oilshas been reported [9]. The fast GC results were acquiredusing a20 m×100 um column， and the separation qualitywas shown to be identical to the chromatograms acquiredusing the original method， which used a 60 mx250 umcolumn. The approach taken in the present investigationfalls somewhere between approaches (l) and (III). Thechoice of column was the same as if the desired resultwere to maintain constant resolution， but fast temperatureprogramming was applied (the highest linear temperatureprogram rates possible using the standard instrumenta-tion)， and the carrier gas velocity (120 cm/s) was roughly twice the optimum value. By using such conditions it isexpected that there will be some loss of efficiency (interms of number of theoretical plates)； however， we aim toshow that it is possible to obtain the necessary informationfor quality assurance purposes for essential oils withinvery fast analysis times. 2 Experimental 2.1 Materials The lime essential oil sample was diluted 1：100 (v/v) in n-hexane prior to injection for analysis by fast GC and 1：10(v/v) for analysis by conventional GC. 2.2 Instrumental Fast GC analysis was performed on a Shimadzu GC-2010gas chromatograph(ShimadzuItalia，Milan， Italy)，equipped with a FID detector operated at 290°C. The dataacquisition rate used for all analyses was 50 Hz. Analysiswas performed using an SE-52 (5% diphenyl-95%dimethyl polysiloxane) stationary phase column (Mega，Legnano， Italy). The column dimensions were 5 m x50 um (0.05 um film thickness). A 3-stage temperatureprogram was used to achieve very fast analysis： the ovenwas temperature programmed from an initial temperatureof 50°℃ to 150°C at 80 K/min； 150°C to 200°℃ at 70 K/min；and finally 200℃ to 250C at 55 K/min. The carrier gaswas hydrogen， which was supplied at a head pressure of880 kPa. An injection volume of 1.0 uL was used for allanalyses and the split ratio was 750：1. The high split ratiois required to ensure major components are not over-loaded； injector-based band dispersion is also reduced byhigh split ratios. Conventional GC analysis was performed on a ShimadzuGC-17A gas chromatograph (Shimadzu， Milan， Italy)equipped with FID detector operated at 290°℃. The dataacquisition rate used for all analyses was 5 Hz. The col-umn used was an Rtx5-MS (5% diphenyl-95% dimethylpolysiloxane) low-polarity stationary phase column. Thecolumn dimensions were 30 m x 250 um (0.25 um filmthickness). The oven was temperature programmed froman initial oven temperature of 50℃ to 250℃ at 3 K/min.The carrier gas was helium， which was supplied at a headpressure of 102 kPa. An injection volume of 1.0 uL wasused for all analyses and the split ratio was 100：1. 3 Results and discussion The h vs. u curve for n-tridecane using the 5 m x 50 umcolumn under isothermal analysis at 130℃ is given in Fig-ure 1. The experimentally determined value for the opti-mum (average) carrier gas linear velocity is in agreementwith the theoretical value of 66 cm/s. The minimum plateheight (hmin=0.052 mm) is noteworthy because this also Figure 1. H vs. u curve for n-tridecane using the 5mx50 um column under isothermal conditions at 130℃. Table 1. Trennzahl number for successive n-paraffin hydro-carbons over the elution range of Cg-C30. Cn TZ C， TZ 8-9 28 19-20 20 9-10 33 20-21 18 10-11 33 21-22 17 11-12 32 22-23 16 12-13 30 23-24 16 13-14 29 24-25 15 14-15 27 25-26 14 15-16 25 26-27 13 16-17 24 27-28 13 17-18 22 28-29 12 18-19 20 29-30 11 agrees very well with the predicted value (0.05 mm). Infact， for thin film capillary columns with a high phase ratioβ (which is calculated using β= d/4d， where d. is the col-umn internal diameter， and d is the thickness of the sta-tionary phase coating)， the resistance to mass transfer inthe stationary phase is negligible compared to the otherpeak broadening processes， and for high k values， hminapproaches dc [6]. In the present study，fast analyses were performed usingan average linear carrier gas velocity of 120 cm/s， which isalmost 2 times the optimum value as determined above，and will have ≈55-60% of the optimum efficiency of thecolumn. To characterise the separation efficiency of thecolumn under the analysis conditions described in theexperimental section， the efficiency was calculated interms of Trennzahl number， for successive n-paraffinhydrocarbons over the elution range of Cg-C30. Thesedata are presented in Table 1. The retention time of C30was only 192 s. For the majority of essential oils it isuncommon that components will be present that elute Seconds Figure 2. Fast GC analysis of lime essential oil using a 5 mx50 um (0.05 um film thickness) capillary column and fast tempera-ture programming. The identities of the numbered components are given in Table 2. above retention indices of about 2000-2100 [10]， thusthe maximum anticipated analysis time using the presentconditions would be 117-125 s respectively. Analysis oflime essential oil was achieved in approximately 90s (seeFigure 2)， which equates to a speed gain of ca. 33 times，where the elution time of the final peak using conventionalanalysis was 50 min. The peak widths of 3 componentsfrom different parts of the chromatogram are marked inFigure 2， which are included to provide another illustrationof the high efficiency of this column， even under the‘extreme'operating conditions employed. Although narrow-bore capillaries have a lower samplecapacity than their wide-bore counterparts [11]， complica-tions arising from this limitation were avoided in the pre-sent investigation by the use of a high split ratio. Thisresults in less sample being presented to the detector；however， the signal response remained high because infast GC there is typically commensurately less peak-broadening and the mass per time unit into the detector isnot much less than that for the conventional analysis.Minimal peak broadening in fast GC makes considerationof the detection capabilities very important， as the rapidelution of narrow chromatographic peaks places demandson the data acquisition rate. For modern instrumentation，this is generally not a problem； for example， FID detectorsare typically able to achieve a data acquisition rate of 50-250 Hz using the standard instrument configuration. Quality assurance of essential oils is often based uponquantitation of several important components in the sam-ple to determine if the relative abundance of these compo-nents falls within an acceptable range (this may be dictatedby quality parameters specified in international standards).The 36 components listed in Table 2 are all within accepta-ble ranges typical of quality assurance tests for this type ofsample. Quantitative results for these components，obtained from fast analysis are compared with resultsobtained from analysis using the optimised conventionalGC conditions. The quantitative comparisons (% relativeabundance) between the fast method and conventionalGC for the critical components are in very good agreement，although some loss of resolution of closely eluting compo-nents was apparent. It is worthwhile pointing out oncemore that the aim of this investigation was to operate thecolumn as fast as the instrumentation would allow. Bothmethods (fast and conventional) confirm that the essentialoil complies with the suggested quality standard. 4 Conclusion Until recently， capillary columns with de<100 um havenot been widely used， despite the fact that the theory sup-porting their use has been available for many years. Thiscan be largely attributed to the lack of equipment compati-ble with such columns， and the lack of readily availablecommercial capillary columns of these narrow diameters. Table 2. Comparison of quantitative results from conven-tional GC and fast GC analysis of the components typicallyused as routine quality indicators of lime oil. Identity is basedon conventional analysis using GC-MS and linear retentionindices using the conditions described in reference [12]. However the requirements for fast GC analysis describedhere were easily met in the present investigation with theuse of modern instrumentation， with rapid temperatureprogramming capabilities， which is able to accuratelymaintain high column-head pressure settings， reliably pro-vides a high injection split ratio， and which has suitablyfast detection capabilities. The use of a 50 um capillarycolumn with rapid temperature programming has allowedessential oil analysis in under 90 s. High resolution wasmaintained， and quantitation of key components for qual-ity assurance purposes could be performed using FID，withoutthe requirement of peak de-convolution asdescribed in previous literature. More ready access to 50 um capillary columns， of a varietyof stationary phases arising from commercial developmentwill be important in the next few years， and one may antici-pate more wide-spread use of high resolution fast GC.These columns， and the experience gathered from applica-tions as described herein， should also support the require-ments of fast GC analysis for second dimension columns incomprehensivetwo-dimensionalgas chromatography. Acknowledgements The authors gratefully acknowledge Shimadzu for the loanof the instrumentation， and Mario Galli (MeGA) for the pro-vision of the experimental capillary column. RS wishes tothank Shimadzu Italia and the University of Messina forsupport to undertake this work in Italy. References ( [1] C.A. Cramers， H.-G. Janssen， M.M. van D eursen， P.A.Leclercq， J. Chromatogr.A 1999， 856，315-329 (and refer- ences cited therein). ) ( [2] P. Korytar， H.-G. Janssen， E. M atisova， U.A . Th. Brinkman， Trends Anal. Chem. 2002，21，558-572. ) ( [3] J.M. Davis， C. Samuel， J. H igh Resol. Chromatogr. 2000， 23，235-244. ) ( [4] T . Veriotti， M . McGuigan， R. Sacks， P erfum. Flavorist 2002，27，40-49. ) ( [5] C. B icchi， C. B runelli， M. Galli， A. Sironi， J . Chromatogr. A 2001，931，129-140. ) ( [6] P. Sandra， M. P r oot， G . Diricks， F . D a vid，in ： C. Bicchi， P. Sandra (Eds.)， C apillary Gas C hromatography in Essential O il Analysis， Huthig， Heidelberg 1987. ) ( [7] L .M. Blumberg， M.S. Klee， A nal. Chem. 1998， 7 0， 3828- 3839. ) ( [8] M.S. K lee， L .M. B lumberg， J . C h romatogr. Sc i . 2002， 40 ， 234-247. ) ( [9] F. David， D.R. Gere， F. Scanlan， P. Sandra， J. Chromatogr A 1999，842，309-319. ) ( [10] R.P. Adams， I d entification of Essential Oil Components by G as Chromatography/Mass Spectroscopy. Allured Pu b lish- i ng Corporation， Carol S t ream， Illinois， 1995. ) ( [11] L.S. Ettre ， J. High Resol. Chromatogr . Chromatogr. Com- mun. 1985， 8，497-501. ) ( [12] L. Mondello， P. Dugo， A . B asile， G. D ugo， K.D. B a rtle， J . Microcol. Sep.1995，7，581-591. ) ( O 2004 WILEY-VCH Verlag GmbH & Co. KGaA， Weinheim ) @ WILEY-VCH Verlag GmbH & Co. KGaA， WeinheimJ. Sep.Sci. ww.jss-journal.de J. Sep.Sci. WILEY-VCH Verlag GmbH & Co. KGaA， Weinheimwww.jss-journal.de