A

spoieib©

KS

1

oipaipavmasbrt

0d

Journal of Chromatography A, 1160 (2007) 81–89

Determination of essential oil components of Artemisia haussknechtiiBoiss. using simultaneous hydrodistillation-static headspace liquid

phase microextraction-gas chromatography mass spectrometry

M. Jalali Heravi a,∗, H. Sereshti b

a Department of Chemistry, Sharif University of Technology, P.O. Box 11365-9516, Tehran, Iranb Department of Chemistry, Faculty of Sciences, University of Tehran, Tehran, Iran

Received 24 November 2006; received in revised form 7 April 2007; accepted 15 May 2007Available online 7 June 2007

bstract

A novel method for extraction and analysis of volatile compounds of Artemisia haussknechtii Boiss., using simultaneous hydro-distillation andtatic headspace liquid microextraction followed by gas chromatography–mass spectrometry (SHD-SHLPME-GCMS) is developed. SHLPMEarameters including nature of extracting solvent, headspace volume and design, extraction time, sample weight and microdrop volume wereptimized. Comparison of hydro-distillation gas chromatography–mass spectrometry and HD-SHLPME-GCMS showed that the latter methods fast, simple, inexpensive and effective for the analysis of volatile compounds of aromatic plants. By using this method, 56 compounds were

xtracted and identified for Artemisia haussknechtii Boiss. The main constituents of its essential oil that were extracted by HD-SHLPME method,nclude camphor (41.01%), 1,8-cineole (32.35%), cis-davanone (3.68%), 4-terpineol (2.99%), linalool (2.84%), beta-fenchyl alcohol (2.72%), andorneol (2.58%).

2007 Elsevier B.V. All rights reserved.

ase m)

smrmthfHetamf

eywords: Hydrodistillation (HD); Hydrodistillation-static headspace liquid phingle drop microextraction; Gas chromatography–mass spectrometry (GC–MS

. Introduction

As the sophisticated analytical instruments are not capablef handling sample matrices directly, a sample preparation steps required. Sample preparation is a critical step in an analyticalrocedure. The main aim of sample preparation is to transfer thenalyte into a form that is prepurified, concentrated, and compat-ble with the analytical system [1]. Main advantages of samplereparation procedures such as solvent microextraction (SME)nd headspace sampling (HSS) [2] are their speed and smallolume of solvents required for their performances. Single-dropicroextraction that is the advanced type of this approach issimple, inexpensive, fast, effective and virtually solvent-free

ample pretreatment technique. The major idea behind this has

een to facilitate automation, to speed up extractions, and toeduce the consumption of organic solvents. Generally, there arewo different modes of liquid phase microextraction (LPME):

∗ Corresponding author. Tel.: +98 21 66005718; fax: +98 21 66012983.E-mail address: jalali@sharif.edu (M. Jalali Heravi).

ht[

auc

021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.chroma.2007.05.096

icroextraction (SHD-SHLPME); Artemisia haussknechtii Boiss.; Essential oil;

tatic LPME consisted of a microdrop suspended at the tip of aicrosyringe needle and dynamic LPME, in which the microsy-

inge is used as a separatory funnel and featured the repeatedovement of the syringe plunger [3–9]. Headspace microex-

raction (HSME) is a combination of HSS and SME, and has theigh capabilities of them. It is a novel method that can be usedor the sample preparation in chromatographic analysis [10–12].SME is shown to be an efficient preconcentration method for

xtraction of volatile compounds from aqueous sample solu-ions [13]. Direct headspace analysis of volatile compounds inqueous or other matrices is utilized extensively for years as aean of directly determining volatile compounds without inter-

erence from the sample matrix. In such determinations, theeadspace is typically sampled directly with a microsyringe,herefore, it is applicable only to extremely volatile compounds14].

Essential oils (EOs, also called volatile or ethereal oils) areromatic oily liquids obtained from plant materials. The greatestse of EOs is in food (as flavorings), perfumes and pharma-euticals (for their functional properties) [15]. To achieve the

8 . Chr

bsToscactcGwttTaaa

tsoeebp

sf[hIaco((

2

2

BiDTnctmS

2

s

otarycwtosrGo

2

ma0f1H22

te0p3trM2

i1sps

mttopat

2

2 M. Jalali Heravi, H. Sereshti / J

est possible separation performance for essential oil analy-is, we intended to use the most effective, available technology.he result achieved then can been used to answer the researchr industrial analysis questions, which necessitated the analy-is. This may be for comparative purposes, where one oil isontrasted with other(s) for quality control or investigation ofdulteration, to discover new components, or to characterize thehemical classes of compounds present. Nowadays, the analystsurn to chromatography as a powerful separation method andombined it with mass spectrometry to aid identification. TheC–MS technique, along with improved data handling tools –ill immediately be relevant to the essential oil area. However,

he extraction of essential oils is still a problem from the dura-ion point of view and the amount of the plant material needed.he common technique of hydro-distillation needs about 3–4 hnd tens to hundreds grams of materials, which in some situ-tions, especially in cases where enough plant material is notccessible, is problematic.

There are several reports in which the extraction of essen-ial oils is based on the microextraction [16–26]. In the presenttudy, we have developed a new method that is a modificationf recently introduced extraction methods to achieve the mostffective procedure for the extraction and preconcentration ofssential oils in Artemisia haussknechtii Boiss. This method cane performed in a short time and consumes a small amount oflant material.

The genus Artemisia (compositae), with the common Per-ian name of ‘DERMANE’, includes 34 species that can beound as wild plant all over Iran of which two are endemic27,28]. In the present work chemical composition of Artemisiaaussknechtii Boiss., which was collected from Yazd province inran and extracted by hydro-distillation is reported (Table 1). Thenalysis of essential oil of this species has not previously beenhemically investigated. The main constituents of its essentialil were camphor (32.70%), 1,8-cineole (31.40%), cis-davanone7.46%), comphene (4.23%), 4-terpineol (3.00%), alpha-pinene2.59%) and borneol (2.42%).

. Experimental

.1. Reagents and material

The plant materials (aerial parts) of Artemisia haussknechtiioiss., were collected from Yazd province in the center of Iran,

n April 2006. The plant was authenticated by the herbariumepartment of the faculty of Science, University of Tehran,ehran, Iran. Chemicals, such as n-heptadecane, n-hexadecane,-pentadecane, n-dodecane, 1-dodecanol, hexamethylene diiso-yanate and anhydrous sodium sulphate with the purity higherhan 99% were purchased from Merck chemical company. Nor-

al alkanes’ standards (C-6 to C-18) purchased from ULTRAcientific, North Kingstown, USA.

.2. Isolation of essential oil

The Artemisia haussknechtii Boiss. plant was dried underhade at room temperature for 48 h. Then 50 g of aerial parts

ppbc

omatogr. A 1160 (2007) 81–89

f it, i.e. the leaves and fine stems, were separated and ground,hen this was immersed in water in a 1 l round bottom flasknd hydro-distilled in a full glass Clevenger-type apparatus asecommended by British Pharmacopeia giving transparent lightellow oils. The extraction was carried out for 3.5 h. When theondensed material cooled down, the water and essential oilsere separated. The oils were decanted to be used as essen-

ial oils. To improve the recovery and the analysis, the essentialils were taken up in n-pentane (Merck), dried over anhydrousodium sulphate (Merck) until the last traces of water wereemoved and then stored in a dark glass bottle at 4 ◦C prior toC and GC–MS analyses. The extraction yield for the essentialils of Artemisia haussknechtii Boiss. was 2.41%.

.3. Instrumentation and analysis

GC analyses were carried out on a Shimadzu-17A gas chro-atograph equipped with a flame ionization detector (FID) andDBP-5 capillary fused silica column (25 m, 0.25 mm I.D.;

.22 �m film thickness). The oven temperature was held at 40 ◦Cor 1 min then programmed at 3 ◦C/min to 250 ◦C, held for0 min. Other operating conditions were as follows: carrier gas,e (99.999%); inlet pressure, 103 kPa; with a linear velocity of8.8 cm/s; injector temperature, 250 ◦C; detector temperature,50 ◦C; split ratio, 1:50.

GC–MS analyses were performed on a HP-6890 GC sys-em coupled with a 5973 network mass selective detector andquipped with a HP5-MS capillary fused silica column (60 m,.25 mm I.D.; 0.25 �m film thickness). The oven temperaturerogram was initiated at 40 ◦C, held for 1 min then raised at◦C/min to 250 ◦C, held for 20 min. Other operating condi-

ions were as follows: carrier gas, He (99.999%); with a flowate of 1 ml/min; injector temperature, 250 ◦C; split ratio, 1:50.

ass spectra were taken at 70 eV. Mass range was from m/z0–500 amu.

Both processes of the extraction of essential oil and thenjections into GC and GC–MS were carried out using a0 �l micro-syringe model ITO MICRO SYRINGE MS-E10. Ithould be noted that the needle tip was an angled cut (to facilitateiercing of the septum in the GC injection port and headspaceeptum), not a flat cut.

At the start of the day, it is advisable to wash the microsyringeany times with the solvent solution used for the microextrac-

ion in order to clean it and to eliminate air. This is necessary ashe presence of air bubbles in the microdrop can change the ratef extraction and serve as a source of error in the analysis. Therecision of the method is improved by ensuring reproduciblend stable positioning of the sample flask and syringe needle inhe headspace, with the help of stands and clamps [3].

.4. Identification of essential oil constituents

The components of the essential oils were identified by com-

aring their retention indices and mass spectra fragmentationatterns with those stored on the Wiley 7n.l MS computer libraryuilt up using pure substances or with authentic compounds andonfirmed by comparison of their retention indices. The Kovats’

M. Jalali Heravi, H. Sereshti / J. Chromatogr. A 1160 (2007) 81–89 83

Table 1Components of Artemisia haussknechtii Boiss. extracted by HD and SHD-SHLPME methods

No. R.T. Compound RIa HDb (%) HDc-SHLPME (%) RSDd

1 9.359 Methyl isopropenyl ketone 620 0.225 tr2 16.406 2-Hexenal 820 0.018 tr3 20.117 Tricyclene 895 0.169 0.1874 20.293 alpha-Thujene 899 0.116 0.0305 20.751 alpha-Pinene 907 2.589 0.648 24.086 21.608 Comphene 924 4.232 1.344 16.907 21.824 Verbenene 928 0.090 0.010 24.398 22.816 Sabinene 946 0.973 0.561 13.549 23.067 beta-Pinene 951 0.985 0.499 25.17

10 23.548 beta-Myrcene 960 0.273 0.343 13.1911 23.694 2,3-Dehydro-1,8-cineole 963 0.323 0.06912 25.085 alpha-Terpinene 989 0.691 0.285 27.6713 25.587 Cymene 998 0.469 0.326 11.7014 25.836 Limonene 1003 0.133 0.156 11.7715 26.116 1,8-Cineole 1008 31.401 26.841 10.5716 26.499 Benzene acetaldehyde 1016 0.050 0.046 18.1917 27.295 gamma-Terpinene 1031 1.105 0.626 24.9718 27.752 cis-beta-Terpineol 1040 0.365 1.211 26.7419 28.844 alpha-Terpinolene 1060 0.282 0.267 15.2320 29.092 Rosefuran 1065 0.045 0.045 22.3821 29.399 Linalool 1071 1.556 4.442 20.7422 30.364 beta-Thujone 1089 0.022 0.054 30.6223 30.737 1-Terpineol 1096 0.234 0.30424 31.867 trans-Pinocarveol 1119 0.710 0.638 9.7625 32.24 Camphor 1126 32.702 40.830 15.6526 32.861 Pinocarvone 1139 0.429 0.603 11.8127 33.034 Borneol 1142 2.417 2.870 11.0928 33.58 4-Terpineol 1153 3.002 3.622 16.9229 34.166 beta-Fenchyl alcohol 1165 1.938 3.519 11.7530 34.453 Myrtenol 1170 0.284 0.297 13.0731 34.538 Myrtenal 1172 0.219 0.245 13.5532 34.89 Terpinene-3-ol 1179 0.061 0.133 39.8233 35.157 Verbernone 1184 0.055 0.10834 35.4 trans-Carveol 1189 0.108 0.227 10.3335 36.363 Z-Citral 1208 0.035 0.063 11.2536 36.536 alpha-Terpinene 1212 0.115 0.06837 36.699 Carvone 1215 0.217 0.327 8.0038 36.843 Geraniol 1218 0.102 0.148 27.9839 37.411 Chrysanthenyl acetate 1230 0.067 0.258 12.9940 37.709 E-Citral 1236 0.033 0.12341 38.668 Borneol acetate 1255 0.455 0.636 15.8242 41.825 Eugenol 1322 0.056 0.072 1.2443 42.672 Geranyl acetate 1342 0.100 0.29244 42.773 Methyl cinnamate 1344 0.068 0.11845 43.734 cis-Jasmone 1367 0.188 0.313 10.8946 44.998 trans-Caryophyllene 1396 0.216 0.135 12.0447 47.484 Davana oil 1454 0.363 0.140 28.6948 47.571 Germacrene D 1456 0.319 0.131 36.9549 48.209 Bicyclogermacrene 1471 0.236 0.078 21.3450 48.308 Davana ether 1474 0.731 0.214 30.2651 50.393 Farnesol 1524 0.062 0.114 26.2552 51.553 cis-Davanone 1554 7.461 4.774 29.6653 51.852 Caryophyllene oxide 1562 0.122 0.252 26.7254 52.706 Artemisia ketone 1583 0.438 0.157 20.8855 53.041 E-sesqui-lavandulol 1592 0.129 0.084 29.2156 53.791 Methyl jasmonate 1611 0.214 0.102 45.53

eTrace (<0.01).a Retention indices using a HP-5MS column.b Relative area percent (peak area relative to total peak area) for hydrodistillation method.c Relative area percent (peak area relative to total peak area except for the solvent peak) for SHD-SHLPME method.d Relative standard deviation values for SHD-SHLPME method (relative peak area).

8 . Chromatogr. A 1160 (2007) 81–89

rcTeAu

2

ttpscosoaatftihinwvotpswa

2m

aaoiwamitdapsAwsc

Fv

p(

3

3t

haesmitopenes or phenylpropane derivatives in which, the chemical and

4 M. Jalali Heravi, H. Sereshti / J

etention indices of all the constituents were obtained using gashromatograms by interpolation between bracketing n-alkanes.he homologous series of n-alkanes (C-6 to C-18; ULTRA Sci-ntific, Inc.; North Kingstown, USA) were used as standards.n Enhanced ChemStation G1701 DA, version D.00.01.27 wassed for the data collection and processing [15,29,30].

.5. Hydro-distillation

Hydro-distillation seems to be the best method for the extrac-ion and isolation of essential oils from the plant materials. Inhis method, the botanical material is fully submerged in water,roducing a ‘soup’, and the still is brought to boil. The resultantteam of which contains the aromatic plant molecules beingaptured and condensed. The oil will normally float on topf the ‘hydrosol’ (the distilled water component) and may beeparated off. When the condensed material cooled down, theil and hydrosol is separated and the decanted oil to be useds essential oil. This method protects the oil so extracted tocertain degree, since the surrounding water acts as a barrier

o prevent it from overheating. Hydro-distillation can be per-ormed at a reduced pressure (under vacuum) to reduce theemperature to less than 100 ◦C, which is useful in protect-ng the plant material as well as essential oil. In spite that theydro-distillation is, the most common method to extract andsolate the essential oils, but it is a time consuming process andeeds a large amounts of plant material. In the present work,e have combined hydro-distillation with static headspace sol-ent microextraction (SHD-SHLPME) to improve the efficiencyf the process. This combination would considerably shortenhe time of extraction process by using small amounts of thelant material and microliter volume of an organic solvent. Con-idering these advantages, the extraction method offered here,ould be useful especially for research centers, which work on

romatic plants.

.6. Simultaneous hydro-distillation static headspace liquidicroextraction-gas chromatography–mass spectrometry

There are three phases involved in the extraction process:queous sample mixture, headspace, and organic microdropcceptor phase. The apparatus that has been designed in our lab-ratory is a developed one used by Tellez et al. [16] and shownn Fig. 1. In this experiment, 2.5 g of the dried and ground plantas immersed in 50 ml water in a 100 ml round-bottom flask,

nd heated to boil using a heating mantle. A suitable and cleanicro-syringe was used to uptake 1 �l of n-heptadecane contain-

ng 200-ppm n-pentadecane as internal standard. The needle ofhe micro-syringe was then inserted into the headspace of hydro-istilling plant sample through a septum. After waiting for 5 minfter the reflux began, to reach a steady state, the micro-syringelunger was depressed and a microdrop of extracting solvent wasuspended from the needle tip, and the extraction was started.

fter an optimized period of time (4 min), the plunger wasithdrawn back and the microdrop was retracted back into the

yringe. The needle was removed from the headspace and itsontents were injected into the GC system. Then the relative

spat

ig. 1. The new SHD-SHLPME apparatus with schematic diagram of flowingapor.

eak areas of components with respect to the internal standardn-pentadecane) were calculated.

. Results and discussion

.1. Optimization of SHD-SHLPME: parameters affectinghe extraction

There are some parameters such as the extracting solvent,eadspace volume and design, extraction time, sample weightnd microdrop volume which directly affect the results of thextraction and control the optimum performance. In the presenttudy, we tried to optimize these parameters. All quantificationsade were based on the relative peak area of the analytes to the

nternal standard (n-pentadecane). It should be mentioned thathe split injection we used in this work may cause discriminationf some components, but the essential oils are mixtures of ter-

tructural differences between them are minimal. This wouldroduce multiple overlapping peaks whose resolution is alwaysproblem. Therefore, to reduce peak overlapping, one may use

he split mode of injection.

. Chromatogr. A 1160 (2007) 81–89 85

3

twtssoda

3

ntcgCcovconipnetostcsto

3

i

FSo

flgttlscel1uttw

M. Jalali Heravi, H. Sereshti / J

.1.1. Agitation of the sampleAgitation of the sample enhances extraction and reduces the

ime needed for thermodynamic equilibrium [3,31,32]. In thisork, the extraction takes place under the condition of simul-

aneous hydro-distillation in which both the boiling sampleolution and turbulent flow of vapor in the headspace provideome kind of stirring the sample, thus it has a positive effectn the extraction efficiency. However, when the microdrop isirectly exposed to the vapor of the sample, it may evaporate,nd its volume decreases.

.1.2. Nature of extracting solventThe experiment was first conducted to study the effect of the

ature of the extracting solvent. Choosing most suitable extrac-ion solvent is very critical especially for the analysis of volatileomponents of essential oils in SHLPMD method because of thereat differences of the compounds in polarity and volatility [33].onsidering the principle ‘like dissolves like’, several importantriteria such as ability to extract the components of essentialil, stability of the drop under the experiment conditions, lowolatility, selectivity, extraction efficiency, a favorable partitionoefficient for volatile components, low volatile impurities, ratef drop dissolution, non-overlapping peak with sample compo-ents in GC, and level of toxicity should be met [3]. Taking thesento account, several different solvents with a diverse range ofolarity such as n-heptadecane, n-hexadecane, n-pentadecane,-dodecane, 1-dodecanol and hexamethylene diisocyanate werexamined. According to the results shown in Fig. 2, amonghe tested solvents, after a detailed comparison, n-heptadecaneffered the highest extraction efficiency, lowest volatility with atable microdrop at the tip of the syringe needle over the extrac-ion period. In addition, this solvent showed nonoverlappinghromatographic peaks and thus was selected as the extractionolvent. Other examined solvents, not only had lower extrac-ion efficiency, but also their peaks overlapped with the peaks ofther components [3,33–35].

.1.3. Optimization of the headspace volumeFig. 1 shows schematic diagram of the flow of sample vapor

n the headspace. The bubble creates more space and a turbulent

Fig. 2. Relative extraction efficiency for different solvents.

(tb

3

itettiapOasisIa

ig. 3. Comparison of extraction efficiency resulted by two systems: (S1) SHD-HLPME with new apparatus and (S2) SHD-SHLPME with old apparatus basedn the relative peak area.

ow of gas phase sample so that the microdrop receives moreas phase sample, thus the equilibrium is reached in a shorterime and the efficiency increases. On the other hand, due to theurbulent flow of the vapor, the microdrop cannot have a volumearger than 2 �l. This would increase the efficiency of extraction,ince large volume injections result in broadening of the peaks inapillary GC. The results of experiments show that the extractionfficiency from 1 �l drop in the new apparatus is considerablyarger (in average about six times larger) than that of that from�l in the old design (Table 2). The extraction efficiency resultedsing these two systems is compared in Fig. 3. It is clear thathe efficiency of the method presented in this work is higher andhe number of components detected is larger than the previousork. Therefore, the new apparatus has two main advantages:

1) the extraction efficiency has increased considerably and (2)he problems raised from the large volume injections to GC haseen removed.

.1.4. Optimization of the extraction timeThe amount of the analyte transferred into the microdrop

s expected to increase with increasing its exposure time tohe headspace of the sample. However, the SHLPME is not anxhaustive extraction method and the analyte partitions amonghe bulk sample phase, the headspace and the microdrop. Thus,he amount of the analyte transferred into the microdrop reachests maximum when the equilibrium is established [34]. However,lthough maximum sensitivity is attained at equilibrium, com-lete equilibrium is not needed for accurate and precise analysis.nly when sufficient mass is transferred into the microdrop in

n exact reproducible extraction time is adequate [4]. Fig. 4hows the extraction-time profile which was achieved by vary-

ng the exposure time of the microdrop in the headspace of aample from 2 to 6 min, and monitoring the analytical signal.t shows an increase in sampling time in the range of 2–4 min,nd a decrease after 4 min. This decrease can be attributed to the

86 M. Jalali Heravi, H. Sereshti / J. Chromatogr. A 1160 (2007) 81–89

Table 2Comparision of extraction efficiency of SHD-SHLPME with and without bubble system based on relative peak areas

No. R.T. Compound RIa SHD-SHLPMEb SHD-SHLPMEc

1 9.359 Methyl isopropenyl ketone 620 ndd 0.012 16.406 2-Hexenal 820 nd 0.033 20.117 Tricyclene 895 nd 0.544 20.293 alpha-Thujene 899 nd 0.095 20.751 alpha-Pinene 907 0.29 2.036 21.608 Comphene 924 0.88 4.177 21.824 Verbenene 928 nd 0.038 22.816 Sabinene 946 0.35 1.749 23.067 beta-Pinene 951 0.35 1.56

10 23.548 beta-Myrcene 960 0.15 1.0611 23.694 2,3-Dehydro-1,8-cineole 963 nd 0.2212 25.085 alpha-Terpinene 989 0.18 0.9013 25.587 Cymene 998 0.22 1.0114 25.836 Limonene 1003 0.07 0.4815 26.116 1,8-Cineole 1008 14.52 83.4016 26.499 Benzene acetaldehyde 1016 nd 0.1417 27.295 gamma-Terpinene 1031 0.40 1.9718 27.752 cis-beta-Terpineol 1040 0.47 3.7719 28.844 alpha-Terpinolene 1060 0.16 0.8320 29.092 Rosefuran 1065 nd 0.1421 29.399 Linalool 1071 0.34 13.9022 30.364 beta-Thujone 1089 nd 0.1723 30.737 1-Terpineol 1096 0.44 0.9124 31.867 trans-Pinocarveol 1119 nd 1.9825 32.24 Camphor 1126 19.57 127.0926 32.861 Pinocarvone 1139 0.35 1.8827 33.034 Borneol 1142 1.57 8.9228 33.58 4-Terpineol 1153 1.60 11.3629 34.166 beta-Fenchyl alcohol 1165 1.46 10.8930 34.453 Myrtenol 1170 0.29 0.9231 34.538 Myrtenal 1172 nd 0.7532 34.89 Terpinene-3-ol 1179 nd 0.4033 35.157 Verbernone 1184 nd 0.3434 35.4 trans-Carveol 1189 nd 0.7035 36.363 Z-Citral 1208 nd 0.1936 36.536 alpha-Terpinene 1212 nd 0.2137 36.699 Carvone 1215 0.19 1.0138 36.843 Geraniol 1218 nd 0.4439 37.411 Chrysanthenyl acetate 1230 nd 0.7840 37.709 E-Citral 1236 nd 0.3641 38.668 Borneol acetate 1255 0.18 1.9142 41.825 Eugenol 1322 0.11 0.2243 42.672 Geranyl acetate 1342 0.12 0.8644 42.773 Methyl cinnamate 1344 nd 0.3745 43.734 cis-Jasmone 1367 0.38 0.9546 44.998 trans-Caryophyllene 1396 0.10 0.4147 47.484 Davana oil 1454 nd 0.4348 47.571 Germacrene D 1456 nd 0.3849 48.209 Bicyclogermacrene 1471 nd 0.2350 48.308 Davana ether 1474 nd 0.6651 50.393 Farnesol 1524 nd 0.3552 51.553 cis-Davanone 1554 4.11 14.3653 51.852 Caryophyllene oxide 1562 0.43 0.7654 52.706 Artemisia ketone 1583 nd 0.4855 53.041 E-sesqui-lavandulol 1592 nd 0.2556 53.791 Methyl jasmonate 1611 nd 0.31

a Retention indices using a HP-5MS column.b Relative area percent (peak area relative to total peak area except for solvent and internal standard peak) for SHD-SHDLPME method (without bubble system).c Relative area percent (peak area relative to total peak area except for solvent and internal standard peak) for SHD-SHLPME method (with bubble system).d Undetectable compound.

M. Jalali Heravi, H. Sereshti / J. Chromatogr. A 1160 (2007) 81–89 87

Fw

sdmuicia

3

absoccatsitie

Fm

3

bppoiwtva[hlbsc

3

of Artemisia haussknechtii Boiss. by SHD-SHLPME-GCMS

ig. 4. Effect of extraction time on the efficiency. Extraction conditions: sampleeight, 2.5 g; microdrop volume, 1.0 �l.

olvent evaporation and to the back-extraction from the micro-rop into the headspace [36]. As prolonged extraction timesay result in drop dissolution and drop loss, extraction time is

sually matched to the chromatography run, thus allowing max-mum sample throughput. However, when an extraction time ishosen in the rising portion of the profile, choosing precise tim-ng becomes essential for a good precision [3]. Based on Fig. 5,n exposure time of 4 min was selected in this work.

.1.5. Optimization of the sample weightThe effect of sample size was investigated by varying the

mount of the ground plant material sample in a 100 ml round-ottom flask containing 50 ml water. Increasing sample weighthould enhance the extracted analytes in the drop [37]. The effectf sample weight on the extraction efficiency is shown in Fig. 5. Itan be seen from this figure that the extracted analytes increasedontinuously with the increasing sample size from 0.5 to 2.5 g,nd then decreased. This is due to the fact that after saturatinghe microdrop with volatile analyte compounds, increase in theample has no further effect on the mass transfer into the extract-ng solvent. This behavior might also be due to the solubility of

he volatile compounds in water [36] and the microdrop vapor-zation. As a result, the sample weight of 2.5 g was used for thextraction in this work.

mcH

ig. 5. Effect of sample weight on extraction efficiency. Extraction conditions:icrodrop volume, 1.0 �l; extraction time, 4 min.

.1.6. Optimization of the microdrop volumeIt is important to find a suitable solvent volume for

oth the extraction and the chromatographic analysis. In theresent study, a drop of n-heptadecane containing 200 ppm n-entadecane (as internal standard) was exposed to the headspacef boiling sample solution for different solvent volumes prior tonjection into the GC system. Solvent volumes from 0.5 to 2.0 �lere tested and the results are shown in Fig. 6. It can be seen that

he peak area of the analyte has increased with increasing solventolume from 0.5 to 1.0 �l and then has decreased. The decreasefter 1.0 �l can be attributed to the insufficient equilibration time36]. On the other hand, although a bigger organic drop givesigher signal response, its manipulation is more elaborate andess reliable. Further, large injection volumes bring about largerand broadening in capillary GC [38]. Taking these into con-ideration, a drop volume of 1.0 �l was chosen in the presentontribution.

.2. Evaluation of the results

Fifty six components were obtained for the essential oil

ethod in this work (Table 1). Retention indices using HP-5MSolumn together with the relative areas obtained for HD andD-SHLPME are shown in this table. Comparing the results

88 M. Jalali Heravi, H. Sereshti / J. Chromatogr. A 1160 (2007) 81–89

Fs

olctHt((mimd(2anstcboui

Bi

Fig. 7. Total ion chromatograms of essential oil components of Artemisiahaussknechtii Boiss. extracted by (a) hydro-distillation, (b) SHD-SHLPMEusing new apparatus, (c) SHD-SHLPME using old apparatus. (1) 1,8-Cineole,(h(

apltepdcsflmti

tsingle drop of an extracting solvent is suspended at the tip of

ig. 6. Effect of drop volume on the extraction efficiency. Extraction conditions:ample weight, 2.5 g; extraction time, 4 min.

btained by these methods shows a higher efficiency for theatter. Obviously, by applying HD-SHLPME not only allomponents of the essential oil have been extracted but alsohe amounts of 29 compounds are larger than those usingD method. In addition, the process has a shorter period of

ime (4.0 min.) and consumed small amounts of plant material2.5 g). In contrast, the HD method took a long period of time3.5 h: 52.5 times longer), consuming a large amount of plantaterials (50 g: 20 times larger). Therefore, HD-SHLPME

s much more efficient than HD, which is the most commonethod for extracting the essential oils. Efficiencies of the old

esign [16] and the new apparatus designed in our laboratoryFig. 1) have been compared in Table 2. As can be seen, only8 compounds were extracted using the old design and theirmounts were considerably smaller than those extracted by theew apparatus in Fig. 1. Therefore, the results in Tables 1 and 2how that SHLPME is a fast and more efficient method forhe extraction of the essential oils, and generally volatileompounds from aqueous solutions at high temperatures up tooiling points. For comparison purposes, the chromatogramsbtained for the extracted components of the essential oilssing hydro-distillation, new and old apparatus are shownn Fig. 7.

The RSD for the main components of Artemisia haussknechtiioiss., (camphor and 1,8-cineole) is between 17 and 19%, which

s relatively satisfactory, but for some minor components it is not

awp

2) linalool, (3) camphor, (4) borneol, (5) 4-terpineol, (6) beta-fenchyl alco-ol, (7) cis-davanone, (8) solvent (n-heptadecane), and (9) internal standardn-pentadecane).

s good as that for the main components. There could be fiveossible reasons for this discrepancy. First, the partition equi-ibrium has not reached in the experiments; therefore, a preciseiming is essential for a good precision. Second, the organic dropvaporates slightly with increasing the exposure time. Third, thelunger was stopped based on visual inspection when the organicrop was withdrawn back into the syringe after sampling wasompleted. This conceivably contributed to the error [4]. Fourth,ince the sample boils and the microdrop is exposed to a hotowing steam, it is not as stable as ordinary LPME, and theicrodrop evaporates more rapidly. Fifth, the turbulent flow in

he bubble headspace increases extraction efficiency, however,t results in a higher RSD.

Under the extraction conditions, the volatile compounds ofhe sample evaporate and transfer into the headspace where a

syringe needle. Over a time period, a dynamic equilibriumill be established among the matrix, headspace and the solventhase [34].

. Chr

4

uhdhtpottt(oicSf

ihiten

R

[

[[[

[

[[

[

[[

[

[

[

[

[[[[

[[

[

[[[

[

[

M. Jalali Heravi, H. Sereshti / J

. Conclusions

The method of SHD-SHLPME-GC–MS, which has beensed in this work for analysis of the components of Artemisiaaussknechtii Boiss., is a combination of simultaneous hydro-istillation and static headspace liquid phase microextraction viaeadspace sampling followed by GC–MS analysis. The aim ofhis study was to offer a fast, low cost, facile, low consuminglant material and efficient method for analysis of the essentialils. In comparison with the hydro-distillation (HD) which isime consuming and needs large amounts of botanical material,he proposed method offers advantages such as: (1) high extrac-ion speed; (2) does not need a large amount of plant material;3) inexpensive apparatus; (4) simplicity, (5) low consumptionf a nontoxic solvent, and (6) small injection volume. Compar-ng the results of the two methods, it is obvious that most of theomponents of the essential oil has been determined using SHD-HLPME. Therefore, it can be considered as a useful methodor fast analysis of the essential oils.

Increasing the extraction efficiency for the method presentedn this work due to the turbulent flow of gas phase sample in theeadspace, allows using small volumes of the microdrop, whichs enriched by analytes in a short time. Therefore, problems dueo large volume injection in GC system could be removed, thextracted compounds would be increased and the trace compo-ents can be identified.

eferences

[1] G. Shen, H.K. Lee, Anal. Chem. 75 (2003) 98.[2] B. Kolb, J. Chromatogr. A 842 (1999) 163.[3] E. Psillakis, N. Kalogerakis, Trends Anal. Chem. 21 (2002) 53.[4] Y. He, H.K. Lee, Anal. Chem. 69 (1997) 4634.[5] S. Pedersen-Bjergaard, K.E. Rasmussen, J. Chromatogr. B 817 (2005) 3.[6] M. Palit, D. Pardasani, A.K. Gupta, D.K. Dubey, Anal. Chem. 77 (2005)

711.[7] T.S. Ho, S. Pedersen-Bjergaard, K.E. Rasmussen, J. Chromatogr. A 963

(2002) 3.[8] M.A. Jeannot, F.F. Cantwell, Anal. Chem. 68 (1996) 2236.[9] G. Ouyang, W. Zhao, J. Pawliszyn, Anal. Chem. 77 (2005) 8122.

[

[[

omatogr. A 1160 (2007) 81–89 89

10] Y. Yamini, M. Hojjati, M. Haji-Hosseini, M. Shamsipur, Talanta 62 (2004)265.

11] A. Przyjazny, J.M. Kokosa, J. Chromatogr. A 977 (2002) 143.12] D.C. Wood, J.M. Miller, I. Christ, LC–GC 22 (2004) 516.13] S. Shariati-Feizabadi, Y. Yamini, N. Bahramifar, Anal. Chim. Acta 489

(2003) 21.14] A.L. Theis, A.J. Waldack, S.M. Hansen, M.A. Jeannot, Anal. Chem. 73

(2001) 5651.15] S. Burt, Int. J. Food Microbiol. 94 (2004) 223.16] M.R. Tellez, I.A. Khanb, B.T. Schaneberg, S.L. Crockett, A.M. Rimando,

M. Kobaisy, J. Chromatogr. A 1025 (2004) 51.17] L. Jirovetz, G. Buchbauera, M.B. Ngassoumb, M. Geisslerc, J. Chromatogr.

A 976 (2002) 265.18] C. Deng1, N. Yao, A. Wang, X. Zhang, Anal. Chim. Acta 536 (2005) 237.19] P. Lı̌opez, M.A. Huerga, R. Batlle, C. Nerin, Anal. Chim. Acta 559 (2006)

97.20] C. Deng, X. Xu, N. Yao, N. Li, X. Zhang, Anal. Chim. Acta 556 (2006)

289.21] P.Q. Tranchida, M.L. Presti, R. Costa, P. Dugo, G. Dugoa, L. Mondello, J.

Chromatogr. A 1103 (2006) 162.22] M. Adam, M. Juklovı̌a, T. Bajer, A. Eisner, K. Ventura, J. Chromatogr. A

1084 (2005) 2.23] C. Deng, A. Wang, S. Shen, D. Fu, J. Chen, X. Zhang, J. Pharm. Biomed.

Anal. 38 (2005) 326.24] C. Deng, N. Li, X. Zhang, J. Chromatogr. A 1059 (2004) 149.25] S. Pedersen-Bjergaard, K.E. Rasmussen, J. Sep. Sci. 27 (2004) 1511.26] C. Deng, X. Yang, X. Zhang, Talanta 68 (2005) 6.27] W. Mozaffarian, A Dictionary of Iranian Plant Names, Farhang Moaser,

Tehran, Iran, 1996.28] L. Ahmadi, M. Mirza, F. Shahmir, Flavour Fragr. J. 17 (2002) 141.29] R. Oprean, M. Tamas, R. Sandulescu, L. Roman, J. Pharm. Biomed. Anal.

18 (1998) 651.30] R. Oprean, L. Oprean, M. Tamas, R. Sandulescu, L. Roman, J. Pharm.

Biomed. Anal. 24 (2001) 1163.31] J. Zhang, T. Su, H.K. Lee, Anal. Chem. 77 (2005) 1988.32] E. Psillakis, N. Kalogerakis, Trends Anal. Chem. 22 (2003) 565.33] L. Fang, M. Qi, T. Li, Q. Shao, R. Fu, J. Pharm. Biomed. Anal. 41 (2006)

791.34] A. Besharati-Seidani, A. Jabbari, Y. Yamini, Anal. Chim. Acta 530 (2005)

155.35] J. Cao, M. Qi, Y. Zhang, S. Zhou, Q. Shao, R. Fu, Anal. Chim. Acta 561

(2006) 88.36] A.R. Fakhari, P. Salehi, R. Heydari, S. Nejad Ebrahimi, P.R. Haddad, J.

Chromatogr. A 1098 (2005) 14.37] E. Psillakis, N. Kalogerakis, J. Chromatogr. A 907 (2001) 211.38] P.J. Marriotta, R. Shellie, C. Cornwell, J. Chromatogr. A 936 (2001) 1.