J. Sep. Sci.

2003, 26, 97–104 Hudaib, Gotti, Pomponio, Cavrini 97

Mohammad Hudaiba), Recovery evaluation of lipophilic markers from

Roberto Gottib),
Romeo Pomponiob), Echinacea purpurea roots applying microwave-
Vanni Cavrinib) assisted solvent extraction versus conventional
a)
Department of
methods
Pharmaceutical Sciences,
Faculty of Pharmacy, University In the present study the recovery from roots of lipophilic markers (mainly alkamides)
of Jordan, Amman, Jordan of Echinacea purpurea (L.) Moench for analysis by GC/MS was evaluated by applying
b)
Dipartimento di Scienze microwave-assisted solvent extraction (MASE), as a new extraction approach, versus
Farmaceutiche, University of two other conventional methods (Soxhlet and ultrasonic extraction). A preliminary
Bologna, Via Belmeloro 6, screening of the three methods, using the best-reported parameters (solvent and
40126 Bologna, Italy extraction time) for Soxhlet (as a reference method) and ultrasonic extraction, show-
ed MASE and ultrasonic extraction (using 70% methanol as the solvent system in
both) to be superior methods to Soxhlet extraction in two solvent systems. Both
methods, MASE and ultrasound, were further evaluated applying different ratios of
methanol-water (60 to 100% methanol) as the solvent system. In these investigations,
MASE showed significantly higher recoveries than the ultrasonic technique over the
70 – 100% methanol range while comparable values were obtained at 60% methanol.
The best recovery of the individual alkamides and the whole lipophilic fraction was
obtained at 70% methanol. The MASE method could serve as good alternative proce-
dure for the preparation of more chemically potent samples and/or crude extracts
from Echinacea species.

Key Words: Sample preparation; Microwave-assisted solvent extraction (MASE); Ul-

trasonic extraction; Echinacea purpurea; Alkamides; Sesquiterpenes; GC/MS analy-
sis;
Received: March 15, 2002; accepted: July 22, 2002

1 Introduction Extracts from E. purpurea are widely used in medicine for

their immunostimulant activities [10, 11] and reports have
The main lipophilic components of Echinacea purpurea shown that the lipophilic constituents, in particular the

Original Paper
and E. angustifolia species are alkamides, typically alkyl- alkamides, play an important in vivo and in vitro immuno-
isobutylamides (# 1 – 3, 5, 6, 8 – 15, 18, 19, Figure 1.A,B) active role [1, 12 – 14]. More recently the alkamides and
or alkylmethylbutylamides (# 4, 7, 16, 17, Figure 1.A, B) polyacetylenic components of Echinacea were also found
[1 – 3]. Alkamides from E. purpurea roots possess mainly to possess light-mediated antifungal activity [9]. On
a 2,4-diene moiety (Figure 1.A), while most amides of E. account of these essential pharmacological properties,
angustifolia are characterized by a monoene moiety the search for appropriate sample preparation methods
(#12 – 14, 16, and 17, Figure 1.B) with the isomeric mix- and effective extraction techniques enabling high recov-
ture of 8/9 (dodeca-2E, 4E, 8Z, 10E-tetraenoic acid isobu- eries of these constituents is, therefore, of particular
tylamide) as the main alkamide constituent of the roots importance. Most of the analytical studies performed until
and aerial parts in both species [1]. These components recently [1 – 4, 8, 15 – 17] utilized Soxhlet extraction with
are mainly localized in roots and to lesser extent in the organic solvents (hexane, chloroform, or methanol) to
aerial parts and their concentrations, in particular the prin- recover the lipophilic or the total (lipophilic and hydrophi-
cipal components trans/cis isomers (8/9), were reported lic) fractions. Some other researchers [9] utilized mother
to be over the range 0.04 – 7.20 (mg/g, dry weight) and tinctures, prepared by maceration of roots in aqueous-
0.01 – 1.50 in the roots [1, 4 – 9] and the aerial parts [1, 8, alcoholic solution, as the starting material for subsequent
9] of E. purpurea, respectively. liquid-liquid extraction prior to analysis. Leinert et al. [18]
have evaluated the influence of three extraction methods
Correspondence: Mohammad Hudaib, Department of Phar- (maceration, Soxhlet, and supercritical fluid extraction
maceutical Sciences, Faculty of Pharmacy, University of Jor- (SFE)) on the GC patterns of Echinacea species and
dan, Amman, Jordan. Phone: +96 277 495792. recommended maceration (in dichloromethane-pentane,
Fax: +96 265 339649. E-mail: hudaib1@[Link]. 1 : 1 v/v mixture) as the favorable solvent. Recently Ber-

i 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1615-9306/2003/0101–0097$17.50+.50/0

98 Hudaib, Gotti, Pomponio, Cavrini J. Sep. Sci. 2003, 26, 97–104

Figure 1. Chemical structures of major alkamides from (A) E. purpurea and (B) E. angustifolia. Compound numbering according
to Bauer et al. [1].
J. Sep. Sci. 2003, 26, 97–104 Recovery of alkamides from E. purpurea by MASE 99

geron et al. [8] reported ultrasound extraction (in 70% 2 Experimental

methanol or ethanol) as a simple method providing better
recoveries compared to Soxhlet for both the polar and 2.1 Plant materials
apolar fractions. For the recovery of phenolic components
About 1 – 2 kg entire healthy plant materials (roots and
a modified sonication procedure, using ethanol-water
tops) of E. purpurea plants grown in the open field at the
(70 : 30) mixture, has been used by Perry et al. [19] for
Herb Garden of Casola Valsenio (Ravenna, Italy) were
standardization of Echinacea. The same researchers [5,
collected in October, 2000 at almost full growth of roots.
6] utilized homogenization with CH3CN, using an Ultratur-
Root materials were separated, washed, cut into small
rax treatment, coupled with C18 solid phase clean up pro-
parts, air-dried, and ground into small particles (0.5 mm)
cedure to recover the alkamide fraction.
just before extraction.

Microwave-assisted solvent extraction (MASE) has been 2.2 Apparatus

successfully applied to extraction of naturally derived
components [20 – 23]. Since its first application to extrac- Microwave-assisted solvent extraction was carried out
tion of organic compounds from different matrices [24], in using a Microwave Accelerated Reaction System, model
recent years many investigations have been concerned MARS Xm (CEM, Matthews, NC, USA). The system is a
with the applicability of MASE as a powerful alternative closed-vessel, provided with a GreenChem Plus vessel
technique to other conventional methods [25, 26]. The set that permit simultaneous processing of up to 14 sam-
technique, in closed and opened vessel systems [27, 28], ples under pressure (max. 350 psi) and temperature
showed many advantages over other techniques for the (max. 2108C) controlled microwave extraction.
extraction of compounds of medium to high polarity [22]
from solid matrices. Microwave assisted extraction con-
2.3 Extraction protocols
sists of heating the solvent (extractant) in contact with
sample (e.g. ground roots) with microwave energy [20, 2.3.1 Soxhlet extraction
29]. The process involves disruption of hydrogen bonds, About 1 g accurately weighed dried root materials were
as a result of microwave-induced dipole rotation of mole- extracted in a Soxhlet apparatus for 48, 3, and 2 h with n-
cules, and migration of ions, which enhance the penetra- hexane, methanol, and chloroform, respectively. The final
tion of solvent into matrix, allowing the solvation of compo- extract obtained in each solvent was filtered and the fil-
nents to be extracted [30]. The temperature and the nat- trate was concentrated under vacuum to 25 mL, for the
ure of the solvent play particular roles in partitioning of the methanol and chloroform filtrates, and below 10 mL
compounds of interest from the matrix to the solvent sys- volume for the n-hexane one. To the n-hexane filtrate, a
tem [20]. 100 mL volume of the internal standard solution (propyl p-
hydroxybenzoate, 0.1% w/v in n-hexane-ethyl acetate
(1 : 1 v/v)) was added and the volume was adjusted to
In the present study, as a part of our research on sample 10 mL with the solvent. The final solution was dried over
preparation methods to integrate GC/MS analysis of the anhydrous sodium sulphate before being subjected to
lipophilic constituents of E. purpurea roots [31], we evalu- GC/MS analysis.
ated the use of microwave-assisted solvent extraction
(MASE) in a closed vessel system as an alternative proce-
dure. To our knowledge, the use of MASE has not been 2.3.2 Ultrasonic extraction
previously reported to extract marker constituents from About 1 g accurately weighed amounts of dried ground
Echinacea roots. As a preliminary investigation, using dif- root materials were extracted in an ultrasonic bath
ferent solvent systems, a selected MASE method was (Sonorex Super RK103H Bandelin, Berlin, Germany), for
compared to both Soxhlet and ultrasonic extractions 7 min at room temperature, three times with 8 mL solvent
under the conditions reported in the literature for these lat- (60 – 100% v/v methanol in water). The extracts (n = 3)
ter methods. In order to asses the effect of solvent on the were collected, filtered through a 0.20 mm membrane fil-
recovery of the lipophilic fraction by MASE and ultrasonic ter (GyroDisc, Orange Scientific, Waterloo, Belgium), and
extraction and for the purpose of comparative evaluation the volume was completed up to 25 mL to a final concen-
of both methods, different proportions of methanol in tration equivalent to 0.04 g roots per mL.
water (60 – 70% v/v) were used as extracting solvent sys-
tems. The lipophilic (mainly alkamide) content in these
extracts was then analyzed by GC/MS, using the internal 2.3.3 MASE approach
standard method, and the recovery data were statistically About 1 g accurately weighed amounts of dried ground
evaluated. root materials were transferred into a vessel containing
100 Hudaib, Gotti, Pomponio, Cavrini J. Sep. Sci. 2003, 26, 97–104

20 mL solvent. Solvents used were either chloroform, GC/MS operation conditions and quali-quantitative analy-
absolute methanol, or methanol-water at different v/v ses were as reported in reference [31]. Briefly, a linear
ratios (60 – 100% methanol). In each run only three extrac- temperature program was adapted to separate the differ-
tions (i. e. 3 vessels containing different solvents) were ent components of the extracts as follows: the column
performed always using the control vessel. The micro- temperature was raised, from an initial value of 1008C, at
wave power was applied in a pulsed manner using 400 W 108/min to 2008C and held there for 2 min; a second ramp
(75%) as the maximum applied energy. The samples was then applied at 3 8/min up to a final temperature of
were processed for a 10 min total extraction time, being 2508C which was maintained for 5 min. For quantitative
heated by the microwave energy to 1008C final tempera- analyses, percentage contents of the various compo-
ture (150 – 200 psi extraction pressure) in 5 min, and nents, identified in the total ion current (TIC) GC/MS
maintained at that temperature for another 5 min. The sys- traces, were calculated by integrating their corresponding
tem was allowed to cool to room temperature (RT) for an peak areas (relative to the internal standard area) and
additional 5 min prior to opening the vessels. The extract assuming a uniform response for all the components.
obtained in each vessel was then filtered through a
0.20 mm membrane filter and further processed as in Sec-
tion 2.3.1, for the n-hexane solution, and as in Sec- 3 Results and discussion
tion 2.3.2 for the others. Generally the following param-
eters and conditions were maintained constant in all the 3.1 Preliminary evaluation of extraction protocols
MASE experiments performed: (a) volume of solvent
(20 mL); (b) sample weight (1 g), particle size (0.5 mm) In a previous study [31] we developed a GC/MS method
and moisture content (fully air-dried) of roots; (c) number for the analysis of the lipophilic constituents of E. purpurea
of vessels in each run (3); (d) applied energy (400 W, obtained from roots by an ultrasonic extraction method
75%); and (e) temperature program (raised to 1008C in modified from that reported by Bergeron et al. [8]. Ultraso-
5 min, isothermal for 5 min). nic extraction was chosen as it was reported to provide the
best recovery of the polar and apolar fractions from the
plant [8, 19]. The many advantages reported so far for
2.3.4 Liquid-liquid extraction (clean up procedure) microwave-assisted solvent extraction (MASE), as a mod-
To recover the lipophilic fractions a 10 mL volume of each ern extraction technique for natural products and environ-
of the solutions, obtained in methanol or chloroform by mental sample preparations [20 – 22], has prompted us to
ultrasonic, MASE, and Soxhlet extraction, was subjected apply this method to the extraction of lipophilic marker
to a threefold extraction with 20 mL n-hexane-ethyl acet- constituents from E. purpurea roots.
ate (1 : 1 v/v). The organic phases were collected, concen- As a preliminary evaluation, three extraction methods,
trated under vacuum, and a 50 mL volume of the internal including MASE, with several solvent systems were com-
standard working solution was added before making up pared. The other two methods utilized were ultrasonic and
the volume to 5 mL with n-hexane-ethyl acetate (1 : 1 v/v). Soxhlet extraction employing the general procedures
The final solution, corresponding to 0.08 g roots per mL summarized in Table 1 and as described in Section 2.
and containing 10 mg/mL of the internal standard, was Soxhlet extraction was used with two solvent systems: n-
dried over anhydrous sodium sulphate prior to GC/MS hexane (or chloroform) and methanol (absolute), which
analysis. are the solvents most commonly used to extract the lipo-
philic and total (lipophilic and hydrophilic) fractions,
respectively [2, 3, 8, 15 – 17]. Ultrasonic extraction was
2.4 GC/MS analysis
performed with 70% methanol (v/v in water), a system that
The gas chromatograph was a TRACE GC 2000 SERIES was reported to afford the best recovery of these fractions
(ThermoQuest CE Instruments, Austin TX, USA) as mentioned above. Finally, using MASE, 70% methanol
equipped with a split-splitless injector (split ratio of 20 : 1). was selected as the starting point for the preliminary
The column was an Rtxm-5MS (30 m640.25 mm, fused study. The obtained data are shown in Table 2 and Fig-
silica capillary column, 0.25 lm film thickness) consisting ure 2, where Soxhlet extraction with n-hexane is consid-
of Crossbondm (5% diphenyl 95% dimethyl polysiloxane). ered as a reference method with an assumed 100% rela-
Helium (He) was the carrier gas at a flow rate of tive recovery for each component class (alkamides, ses-
1.0 mL/min. The GC was interfaced with a GCQ plus quiterpenes, and total lipophilic fractions). As can be
(ThermoQuest-Finnigan, Austin TX, USA) mass detector observed, the MASE and ultrasonic approaches, both
operating in the EI+ mode (70 eV) with a 1.5 min filament with 70% methanol, afforded the highest recovery of the
delay. The mass spectra were generally recorded over alkamide- and the whole lipophilic-fractions, providing sig-
40 – 650 amu full-scan mode (scan time: 1.44 s) revealing nificantly higher levels of these fractions compared to con-
the total ion current (TIC) chromatograms. ventional Soxhlet extraction.
J. Sep. Sci. 2003, 26, 97–104 Recovery of alkamides from E. purpurea by MASE 101

Table 1. Conditions and schematic procedures of the extraction of E. purpurea roots by different protocols.

Extraction technique Solvent system, total volume Procedure scheme

Soxhlet n-Hexane, 25 mL 1 g dried roots treated in Soxhlet for 48 h

Methanol, 25 mL, 1 g dried roots treated in Soxhlet for 3 h
Chloroform, 25 mL 1 g dried roots treated in Soxhlet for 2 h
Ultrasonic Methanol-water (70 : 30 – 100 : 0.0, v/v) 1 g dried roots sonicated 3 times, each for 5 min, using
25 ml 7 – 8 mL solvent
MASE Methanol-water (70 : 30 – 100 : 0.0, v/v), 1 g dried roots, 400 W (75%), 200 psi max., temperature
(closed-vessel system) program (extraction time: 10 min; ramped to 1008C in 5 min,
maintained at 1008C for 5, cooled to RT in 5 min)
Chloroform and n-hexane, 20 mL as above

Table 2. Relative recovery (R.R.) of the lipophilic fractions from E. purpurea roots by different extraction protocols.

Extraction method Sesquiterpenes Alkamides Total lipophilic

R.R. R.R. R.R. (RSD)a)

Soxhlet 100 100 100 (0.93)

(n-hexane, 2 days)*
Soxhlet 55.04 69.96 67.74 (1.33)
(absolute methanol, 3 h)
Soxhlet 52.25 57.02 56.43 (2.12)
(chloroform, 2 h)
Ultrasonic 49.70 119.63 103.28 (1.42)
(methanol 70% v/v in water, 365 min)
MASE 42.88 121.57 104.15 (0.95)
(methanol 70% v/v in water, 10 min)
MASE 31.49 56.97 52.20 (1.13)
(n-hexane, 10 min)
MASE 34.59 72.85 63.99 (1.75)
(Chloroform, 10 min)

a)
Relative standard deviation (RSD) values obtained by inter-day (4 days) analysis of sample replicates (4 replicates per day).
* This method, mostly reported to extract alkamides, is considered here as a standard (100% relative recovery) reference
method for comparative purposes.
R.R.: Mean relative recovery value of 6 determinations.

Moreover, for better and more reliable comparison of the

extraction protocols and as a part of the preliminary eval-
uation study, the stability of alkamides in different solvents
has been taken in consideration. Previous studies [8, 19,
32] showed that alkamides, compared to the polar pheno-
lic fractions, are stable compounds in solution (e.g.
methanol, ethanol, and n-hexane) and also in already pre-
pared alcoholic mixtures [32] and under different storage
conditions (time and temperature). In the present study,
the stability of the lipophilic components from Echinacea
in solution in different solvents has been assessed by
inter-day analysis of random samples during the prelimin-
ary study as shown in Table 2. High stability of these frac-
tions can be concluded and confirmed by the low RSD
values obtained for each extract (in a given solvent)
repeatedly analyzed in different days. On the other hand,
Figure 2. Relative recovery of the lipophilic fractions from E.
the repeatability values (RSDs), which will be shown later
purpurea roots applying different extraction protocols. Recov-
in Table 3, obtained in the comparative evaluation of the eries are compared to that of Soxhlet extraction (n-hexane)
relative recovery of alkamides by both ultrasonic and assumed here to provide 100% relative recovery.
102 Hudaib, Gotti, Pomponio, Cavrini J. Sep. Sci. 2003, 26, 97–104

Table 3. Relative recoveries of the lipophilic fractions from E. purpurea roots extracted by ultrasonic and MASE. Relative recovery
(R.R.) values are derived as the ratio of total compounds (fraction) area to internal standard area.

Methanol% Alkamides R.R.a) P Sesquiterpenes R.R. P Total lipophilic R.R. P

(in water v/v)
MASE Ultrasonic MASE Ultrasonic MASE Ultrasonic

60% 14.39 14.97 0.0417 1.19 1.35 0.0064 15.58 16.33 0.0216
(0.26;1.83) (0.55;3.64) * (0.07;5.79) (0.09;6.89) ** (0.29;1.89) (0.61;3.74) *
70% 16.54 15.02 a0.0001 1.40 1.64 0.0077 17.94 16.66 a0.0001
(0.44;2.67) (0.19;1.24) *** (0.12;8.69) (0.13;7.87) ** (0.40;2.24) (0.17;1.00) ***
80% 12.13 8.99 a0.0001 1.74 1.50 0.0501 13.87 10.35 a0.0001
(0.52;4.25) (0.41;4.60) *** (0.11;6.60) (0.24;16.0) ns (0.48;3.44) (0.48;4.67) ***
100% 11.47 7.55 a0.0001 1.83 1.46 a0.0001 13.30 9.01 a0.0001
(0.21;1.85) (0.35;4.64) *** (0.05;2.91) (0.08;5.38) *** (0.24;1.81) (0.37;4.16) ***

a)
Mean relative recovery value of 6 determinations, between brackets (SD; RSD) are the standard deviation and the relative
standard deviation, respectively.
P: Probability value from unpaired t-test.
ns Non-significant, * significant, ** very significant, *** extremely significant relative recovery difference.

MASE can also give an indication of their stability under

extraction conditions.

3.2 MASE versus ultrasonic extraction and

influence of extraction-solvent polarity
To ascertain the potential advantages of MASE over other
techniques and for comparison between MASE and ultra-
sonic extraction, the influence of solvent polarity on the
recovery provided by both techniques was evaluated. The
MASE method utilized here was optimized from selected
applications, provided by the manufacturer, for extraction
of alkamide-related compounds from root materials. Since
the only variable between the two compared methods was
the solvent (different% methanol in water, 60 – 100% v/v)
the other experimental parameters were maintained con-
stant for all the experiments in the MASE procedure (Sec-
tion 2.3.3).

3.2.1 Recovery of volatile components

(sesquiterpenic fraction)
A cooling stage (5 min) was utilized in each MASE run in
order to minimize the possibility of losing the essential oil- Figure 3. Typical GC/MS chromatograms of the lipophilic
typical volatile components (germacrene D and beta- fractions from E. purpurea obtained from roots by MASE
extraction, showing the relative recoveries of the different
caryophyllene). The phenomenon is very common in
components using: (A) n-hexane, (B) absolute methanol
closed MASE systems [20] because of the high tempera- (100%), (C) 80% (v/v) methanol/water, and (D) 70% (v/v)
ture applied that facilitates the escape of readily volatile methanol/water, as the extracting solvent. For peak identifi-
components into the headspace compartment. Cooling cation refer to Table insert. For chromatographic and MS
before opening the vessels will, therefore, allow conden- conditions refer to Experimental. Alkamides (2, 3, 4, 7, 8, 9)
are according to Figure 1.A.
sation of these compounds and thus reliable evaluation of
extraction performance will be achieved. Representative
GC/MS chromatograms that briefly describe the effect of with increased methanol% was observed as these compo-
solvent (type and polarity) on the relative recoveries of the nents are non-polar and their extraction efficiency (solubi-
sesquiterpenes as well as the alkamides fractions are lity parameter) is directly proportional to lipophilicity of the
shown in Figure 3. Increased recovery of sesquiterpenes extraction solvent. This effect is also shown in Figure 4.A
J. Sep. Sci. 2003, 26, 97–104 Recovery of alkamides from E. purpurea by MASE 103

and is in agreement with the best recovery of these com-

pounds using Soxhlet extraction with hexane (Figure 2
and Figure 3.A). The possibility of losing these volatile
compounds in the less-controlled open extraction sys-
tems such as in ultrasonic extraction can, accordingly, be
an explanation for their lower and irregular recoveries ver-
sus increased methanol% (Figure 4.A).

3.2.2. Recovery of alkamides and the total

lipophilic fractions
The recovery values of the alkamides and the whole lipo-
philic fractions by MASE and ultrasonic methods, using
different methanol-water mixtures as extracting solvents,
are collected in Table 3 and the recovery profiles are
shown in Figure 4.B, C. Significance values (at 95% con-
fidence intervals, unpaired t-test) of the recovery differ-
ences between both methods, at each methanol%, are
also reported in Table 3. Compared to the ultrasonic
method, MASE was found to be superior, providing signifi-
cantly higher levels over the whole concentration range of
methanol, but less significant at 60%. The obtained data
suggest that the level of water in the extraction solvent
plays a critical role in determining the recovery in both
techniques. Although the lipophilic components of E. pur-
purea are considered here, the alkamides combine both
polar (amide acid moiety) and apolar (olefinic hydrocar-
bon chain moiety) properties, which largely determine
their recovery in different solvents. Two factors can be
considered, therefore, to influence the recovery of these
compounds. The first is that increasing the water content
leads to an increase in polarity and the dielectric constant
of the solvent, thus enhancing the microwave extractive
mechanism (absorption of microwave energy and conver-
sion to heat) according to Camel [20]. This factor is valid
only for MASE and not for ultrasonic extraction. The sec-
ond is that increasing the water (polar solvent) content,
while maintaining a considerable level of methanol, as a Figure 4. Comparison between MASE and ultrasonic extrac-
less polar solvent, will lead to a system with an overall tion methods showing the relative recoveries of the (A) ses-
polarity tending to enhance the solubility of alkamides. quiterpenes, (B) alkamides, and (C) total lipophilic fractions
extracted from E. purpurea roots by both methods using dif-
This last factor (solubility parameter) is applicable to both ferent methanol-water ratios.
techniques and explains the increasing alkamide recovery
by both (Figure 4.B) up to 70% methanol as the best sol-
vent. Accordingly, the MASE method, combining both
mentioned factors, provided significantly higher alkamide alkamides and the whole lipophilic fractions, in substantial
recoveries relative to the ultrasonic technique over the agreement with the results obtained by Bergeron et al. [8].
whole methanol% range, although comparable values Higher water levels (A 30%) were associated with compar-
were obtained at the 60% methanol level. The solvent able (P = 0.838) (in ultrasonic) or significantly (P a
polarity therefore influenced the efficiency of extraction by 0.0001) lower (in MASE) recoveries of either fractions and
both techniques; however, this effect was enhanced in individual components. The lipophilicity of alkamides pos-
MASE in the presence of the microwave energy. sibly played the key role in explaining the latter behaviour.
The data presented for MASE and ultrasonic extraction Most likely a 70% methanol content offers the best envir-
(Table 3 and Figure 3.B,C) showed that 70% methanol onment able to solvate alkamides, whose polarity (solubil-
was an appropriate solvent for best recovery of both the ity parameter) is close to that of the solvent system.
104 Hudaib, Gotti, Pomponio, Cavrini J. Sep. Sci. 2003, 26, 97–104

4 Concluding remarks [7] R.B.H. Wills, D.L. Staurt, Food Chemistry 1999, 67,
385 – 388.
Complex participation of many factors was found to con- [8] C. Bergeron, J.F. Livesey, D.V.C. Awang, J.T. Arnason,
trol the relative recovery of the lipophilic fractions (mainly J. Rana, B.R. Baum, W. Letchamo, Phytochem. Anal.
alkamides) from E. purpurea extracts by both MASE and 2000, 11, 207 – 215.
ultrasonic methods using methanol-water as extracting [9] S.E. Bins, B. Purgina, C. Bergeron, M.L. Smith, L. Ball,
solvent. In particular, the polarity of both the solvent sys- B.R. Baum, J.T. Arnson, Planta Med. 2000, 66, 241 –
244.
tem and the alkamide components and their relationship
to the overall solubility of these compounds (targets), the [10] S.S. Percival, Biochem. Pharmacol. 2000, 60, 155 – 158.
ability of the solvent to reach these targets, interact with [11] J. Pepping, Am. J. Health-Syst. Pharm. 1996, 56, 121 –
and liberate them from the complex matrix structures, all 122.
determine to a great extent the best recovery solvent [12] R. Bauer, H. Wagner, in: Economic and Medicinal Plant
(methanol/water ratio) by both the procedures. Moreover, Research, H. Wagner (Ed.), Academic Press, New York
1991, p. 253 – 321.
the microwave effect in MASE was an additional extrac-
tion-enhancing factor, increasing the extraction efficiency [13] R. Bauer, Z. Arztl. Fortbild. Jena 1996, 90, 111 – 115.
of the solvent, and thus providing better recoveries, rela- [14] R. Bauer, ACS Symposium Series 1998, 691, 140 – 157.
tive to ultrasonic extraction, over the different methanol/ [15] P. Pietta, P. Mauri, R. Bauer, Planta Med. 1998, 64,
water ratios used in the study (Figure 4 and Table 3). For 649 – 652.
the alkamide recovery, both methods, in particular the [16] R. Bauer, I.A. Khan, H. Wagner, Planta Med. 1988, 54,
MASE, showed good reproducibility in intra-day analysis 426 – 430.
of three separately prepared samples (each analyzed in [17] K. Glowniak, G. Zgorka, M. Kozyra, J. Chromatogr. A
duplicate) with each solvent system, as expressed by the 1996, 730, 25 – 29.
RSD values reported in Table 3 for the different lipophilic [18] D. Lienert, E. Anklam, U. Panne, Phytochem. Anal.
fractions. For the sesquiterpenic fraction, the recovery 1998, 9, 88 – 98.
reproducibility was found to be less than that obtained for [19] N.B. Perry, E.J. Burgess, V.L. Glennie, J. Agric. Food
alkamides, probably due to the higher volatility of these Chem. 2001, 49, 1702 – 1706.
components (Table 3). On the whole, under the various [20] V. Camel, Trends Anal. Chem. 2000, 19, 229 – 248.
experimental conditions, the MASE method appears to [21] L. Jassie, R. Revesz, T. Kierstead, E. Hasty, S. Matz, in:
offer the best recovery, with a good reproducibility, of the Microwave-Enhanced Chemistry. Fundamentals, Sam-
alkamide components from E. purpurea extract. ple Preparation, and Applications, H.M.S. Kingston, S.J.
Haswell (Eds.). Am. Chem. Soc., Washington D.C.
1997, pp. 569 – 609.
[22] B. Kaufmann, Ph. Christen, J.-L. Veuthey, Phytochem.
Acknowledgement Anal. 2001, 12, 327 – 331.
[23] M. Spiro, S.S. Chen, J. Microwave Power Electromag.
Thanks are due to Dr. Sauro Biffi from the Herb Garden of Energy 1994, 29, 231 – 241.
Casola Valsenio (Ravenna) for supplying the plant materi- [24] K. Ganzler, A. Salgo, K. Valko, J. Chromatogr. 1986,
als. This work was supported by a grant from MURST (Co- 371, 299 – 306.
fin, 2000), Rome, Italy. [25] K. Ganzler, A. Salgo, Z. Lebensm. Unters. Forsch. 1987,
184, 274 – 276.
[26] A. Siquin, T. Gorner, E. Dellacherie, Analusis 1993, 21,
1 – 10.
References
[27] M. Letellier, M. Butzinski, Analusis, 1999, 27, 259.
[1] R. Bauer, P. Remiger, Planta Med. 1989, 55, 367 – 371. [28] C. Demesmay, [Link], Spectra Analyse 1993, 175, 27.
[2] R. Bauer, P. Remiger, H. Wagner, Phytochemistry 1988, [29] J.R.J. Pare, J.M.R. Belanger, S.S. Stafford, Trends
27, 2339 – 2342. Anal. Chem. 1994, 13, 176.
[3] R. Bauer, P. Remiger, H. Wagner, Phytochemistry 1989, [30] K. Ganzler, I. Szinai, A. Salgo, J. Chromatogr. 1990,
28, 505 – 508. 520, 257 – 262.
[4] X.-g. He, l.-z. Lin, M.W. Bernart, L.-Z. Lian, J. Chroma- [31] M. Hudaib, J. Fiori, M.G. Bellardi, C. Rubies Autonell, V.
togr. A 1998, 815, 205 – 211. Cavrini, J. Pharm. Biomed. Anal. (in press).
[5] N.B. Perry, J.W. Van Klink, E.J. Burgess, G.A. Parmen- [32] J. Livesey, D.V. Awang, J.T. Arnason, W. Letchamo, M.
ter, Planta Med. 1997, 63, 58 – 62. Barrett, G. Pennyroyal, Phytomedicine 1999, 6, 347 –
349.
[6] N.B. Perry, J.W. Van Klink, E.J. Burgess, G.A. Parmen-
ter, Planta Med. 2000, 66, 54 – 56. [JSS 1221]