See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/222549815

Supercritical Carbon Dioxide Extraction of Andrographolide From

Andrographis paniculata: Effect of the Solvent Flow Rate, Pressure, and
Temperature

Article  in  Chinese Journal of Chemical Engineering · December 2007

DOI: 10.1016/S1004-9541(08)60018-X

CITATIONS READS

62 699

2 authors:

Andri Kumoro Masitah Hasan

Universitas Diponegoro 41 PUBLICATIONS   1,408 CITATIONS   
80 PUBLICATIONS   509 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ethanol View project

Faculty of Engineering Diponegoro University View project

All content following this page was uploaded by Andri Kumoro on 09 November 2017.

The user has requested enhancement of the downloaded file.

Chin. J. Chem. Eng., 15(6) 877—883 (2007)

Supercritical Carbon Dioxide Extraction of Andrographolide from

Andrographis paniculata: Effect of the Solvent Flow Rate, Pressure,
and Temperature*

A.C. Kumoroa and Masitah Hasanb,**

a
Department of Chemical Engineering, Faculty of Engineering, Diponegoro University, SH Road, Tembalang Se-
marang 50139, Indonesia
b
Department of Chemical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia

Abstract Andrographis paniculata Nees has been extensively used for traditional medicine and help against fever,
dysentery, diarrhoea, inflammation, and sore throat. In this study, andrographolide, the main component of this
plant was extracted from the leaves of A. paniculata using supercritical carbon dioxide. The operating pressures
were varied from 7.50 to 20MPa, the temperatures were varied from 30℃ to 60℃, and the flow rates were varied
from 0.5 to 4ml·min－1. The best extraction condition occurred at 10MPa, 40℃, and a flow rate of 2ml·min－1 for a
3g sample of A. paniculata ground-dried leaves. The measured extraction rate was found to be about 0.0174g of
andrographolide per gram of andrographolide present in the leaves per hour of operation. The future studies must
focus on the interaction between the various operating parameters such as temperature, pressure, and flow rate of
supercritical carbon dioxide.
Keywords supercritical carbon dioxide, extraction, andrographolide, Andrographis paniculata

1 INTRODUCTION Conventional production methods such as solvent

Andrographis paniculata Nees, locally known as
Hempedu Bumi and commonly called the King of
Bitter grows widely in the tropical area of South East
Asia, India, and China with a plant height of 30—
70cm. In these regions, this plant has been extensively
used for traditional medicine and help against fever,
dysentery, diarrhoea, inflammation, and sore throat[1].
Furthermore, it is a promising new way for the treat-
ment of several diseases, including HIV, AIDS, and
numerous symptoms associated with immune disor-
ders[2].
The three main diterpenoid lactones identified in
A. paniculata leaves were andrographolide,
neo-andrographolide, and deoxyandrographolide[3,4].
Andrographolide, which is grouped as an unsaturated
trihydroxy lactone has the molecular formula of
C20H30O5. The molecular structure of andrographolide
is shown in Fig.1. Andrographolide, the main compo-
nent in the leaves of A. paniculata can be easily dis-
solved in methanol, ethanol, pyridine, acetic acid, and
acetone, but slightly dissolved in ether and water. Its
physical properties are: melting point at 228—230℃,
and the ultraviolet spectrum in ethanol λmax as 223nm[4].
Figure 1 Molecular structure of andrographolide
extraction and Soxhlet, although effective for extrac-
tion, can lead to degradation of heat sensitive com-
pounds as well as leave traces of toxic solvents in the
solute. This is a concern for food and medicinal ex-
tracts[5]. Supercritical fluid extraction may be a viable
alternative to solvent extraction methods. The phe-
nomenon of supercritical fluid extraction was recorded
over 100 years ago but has been slow in finding
commercial applications, partly because of the sophis-
ticated and expensive high pressure equipment and
technology required [6].
Carbon dioxide is generally the most desirable
solvent for supercritical fluid extraction (SFE). Its low
critical temperature (31.26℃) and pressure (7.38MPa)
make it effective for the extraction of heat sensitive
(thermally labile) compounds. In addition, it is an inert,
non-flammable, non-explosive, inexpensive, odourless,
colorless clean solvent that results in no solvent resi-
due in the product; it is also non-toxic and is generally
accepted as a harmless ingredient in pharmaceuticals
and food[5]. It also has low surface tension and vis-
cosity and high diffusivity, which make it attractive as
a supercritical solvent[7]. The diffusivity of super-
critical carbon dioxide is one to two orders of magni-
tude higher than for other fluids, which permits rapid
mass transfer, resulting in a larger extraction rate than
that obtained by conventional liquid extraction[5].
Although several studies about supercritical fluid ex-
traction have been carried out in the last few decades,
only one article reported about the supercritical ex-
traction of andrographolide from A. paniculata leaves.
Ge et al.[8] reported the use of supercritical carbon
dioxide as well as carbon dioxide and ethanol mixture
to extract andrographolide in an own built extractor

Received 2006-09-29, accepted 2007-08-29.

* Supported by the Intensification of Research in Priority Areas Project (IRPA), Ministry of Science, Technology and Innovation,
Malaysia (No.09-02-03-0101-EA0001).
** To whom correspondence should be addressed. E-mail: masitahhasan@um.edu.my
878 Chin. J. Ch. E. (Vol. 15, No. 6)
unit. He used orthogonal experiment to determine the
optimum condition without explaining the effect of
each operating parameter. However, he did not study
the effect of particle size and the supercritical solvent
flow rate. Moreover, the geometry of the extractor, the
supercritical solvent flow pattern, and the ethanol
concentration in the supercritical solvent were not
reported. He reported that at high temperature and
pressure, the extraction rate and yield were higher, but
the andrographolide content in the extracts was very
low. He also qualitatively reported that the addition of
ethanol as co-solvent enhanced the extract yield and
the andrographolide content.
The objective of this study is to investigate the
effect of the supercritical solvent flow rate, pressure,
and temperature in the supercritical carbon dioxide
extraction of A. paniculata diterpenoid lactone,
namely andrographolide. Since andrographolide is a
medical substance and the use of a co-solvent in the
extraction process may affect the medicinal properties
of the extracted andrographolide and may hinder fu-
ture FDA approval for such a commercial process, this
study was focused on the use of carbon dioxide only
as a solvent.
2 EXPERIMENTAL
Dried leaves of A. paniculata were obtained from
Malaysian Agricultural Research and Development
Institute (MARDI) and were stored in a sealed air
tight container at ambient conditions. Without any
pre-treatment, the dried leaves were ground into pow-
der and were sieved to obtain a particle size of less
than 0.375mm. Carbon dioxide with a purity of 99.7%
was used as the solvent in this study and was pur-
chased in the liquid form from Air Product (M) Ber-
had. The supercritical fluid extraction apparatus used
in this study is shown in Fig.2. It consists of a 16cm3
high-pressure stainless steel extraction vessel (10mm
inner diameter×200mm length). In a typical run, the
extraction vessel was first filled with 3g of
dried-powdered A. paniculata leaves distributed in
stainless steel balls to improve the leaves-solvent
contact and to avoid channelling when CO2 flowed
through. The extraction vessel was then immersed in
Figure 2 Schematic diagram of the supercritical
fluid extraction unit
December, 2007
the water bath controlled by an electrical heater
(Thermo Haake, model DC10) to within ±0.1K and
the system was brought to the desired temperature.
Liquid CO2 was then charged into high pressure pump
(ISCO, model 260 DM) from the liquid CO2 cylinder.
The high pressure pump further compressed the CO2
to the desired pressure. The system pressure was de-
termined by a pressure transducer (Druck, model
PDCR 961). The valve between the pump and the ex-
tractor was opened and the CO2 was allowed to con-
tinuously flow through the sample at a desired volu-
metric flow rate. The extract was collected in a cold
trap by depressurising the CO2 at the back pressure
regulator (Jasco, model 880-81). The cold trap was a
glass vial immersed in a pack of ice that was main-
tained at 4℃. The samples were taken at one-hour
intervals and the experiments were terminated after
seven hours, and the total amount of extract collected
was gravimetrically determined using a balance (Met-
tler Toledo, model AG 204) with an accuracy of
±0.0001g. The amount of CO2 consumed was deter-
mined using a wet gas meter (Alaxander Wright & Co.,
model DM 3A). The glass vials were wrapped with
aluminium foil to prevent photo degradation of the
andrographolide. For each trial, at least two experi-
ments were carried out where the total amount of CO2
that passed through the cell slightly varied, hence
varying the total amount of extract collected. The rela-
tive error obtained between the duplicate experiments
was less than 5%.
HPLC analysis of andrographolides: The cali-
bration standard curve was obtained using pure crystal
of andrographolide standard compound having 98%
purity supplied by Sigma-Aldrich (M) Sdn. Bhd.
Analyses were done by high-performance liquid
chromatography (HPLC) with a system from Shima-
dzu (Japan). The column used for analysis was reverse
phase C18-Thermo Hypersil ODS 250mm×4.6mm,
with 5µm particle diameters, and a UV detector was
used. The volume injection was 20μl. The mobile
phase was methanol/water (volume ratio 54/46) at a
－
constant 1ml·min 1 flow rate. During the analysis, the
wave length of the UV detector was programmed as
223nm for 10min. Chromatographic peaks were iden-
tified by comparison with the retention time of the
standards. Linear calibrations of standards at accuracy
of more than 99% were carried out for the quantifica-
tion of the A. paniculata extracts. Single injection of
solvent (blank) was also done to determine the reten-
tion time of the solvent.
3 RESULTS AND DISCUSSION
The extracted materials collected in the cold trap
were yellowish in color and were collected in the bot-
tom part of the collection vial. The maximum yield
obtained was about 0.0176g andrographolide per gram
of initial andrographolide present in the leaves or
－
equivalent to 2.01×10 3g andrographolide per gram
of dried leaves. Ge et al.[8] reported that he obtained
－
3.81×10 3g andrographolide per gram of dried leaves.
However, it is somewhat difficult to compare his re-
sults with the results of this study since he used leaves
 Supercritical Carbon Dioxide Extraction of Andrographolide from Andrographis paniculata 879

with different andrographolide content and different

experimental setup and experimental conditions. He
performed the experiment at 25MPa, extractor tem-
perature 46℃, separator I pressure 6MPa, separator I
temperature 65℃, separator II pressure 6MPa, sepa-
－
rator II temperature 45℃, and CO2 flow rate 40kg·h 1.

3.1 Effect of carbon dioxide flow rate

In an attempt to investigate the effect of the sol-
vent flow rate on the extraction rate, a set of experi- (a) Effect of time on the andrographolide cumulative
ments was carried out at 40℃ and 10MPa, while the yield of 0.375mm particle size at 40℃ and
－ 10MPa with flow rate as a parameter
flow rate of CO2 was varied from 0.5 to 4ml·min 1. flow rate, ml·min－1: ■ 0.5; △ 1; ▲ 2; □ 3; ◆ 4
Figs.3(a) and 3(b) were constructed to show the effect
of time with the flow rate as a parameter, and the ef-
fect of supercritical carbon dioxide flow rate with time
as a parameter, on the cumulative yield of the extract.
Both figures show that the yield increases with an in-
crease in the supercritical carbon dioxide flow rate,
reaches a maximum value, and then decreases with
further increase in the flow rate. The result obtained
here can be explained as a trade-off between a mass
transfer process and a thermodynamic equilibrium
state. The interface gas phase concentration of the sol-
ute is a function of the mass transfer coefficient, or (b) Effect of the carbon dioxide flow rate on the
solvent flow rate, while the equilibrium state is fa- andrographolide cumulative yield of 0.375mm
voured by both the high mass transfer rate and the particle size at 40℃ and 10MPa with time as a parameter
long residence time[9]. At low flow rates of the sol- time, min: ◆ 60; △ 120; ■ 180; ◇ 240; ▲ 300; □ 360; ○ 420
vent, the mass transfer resistance limits the amount of
solute transported into the bulk of the solvent and the
supercritical carbon dioxide exits the extractor un-
saturated. As the flow rate is increased, the mass
transfer resistance continues to decrease until the ex-
iting solvent is saturated, and thus equilibrium is
achieved and the maximum yield is attained. Further
increase of the flow rate will reduce the residence time
causing the system to deviate from equilibrium, and
the solvent exits the extractor unsaturated despite the
high mass transfer rate[9,10]. This is because the (c) Effect of the carbon dioxide flow rate on the reduced an-
amount of solvent that is in excess to what is required drographolide yield of 0.375mm particle size as a function of
to penetrate the cellular structure of the leaves simply time at－ 40℃ and 10MPa
bypasses the extractable leaves[11]. This behavior flow rate, ml·min 1: ▲ 0.5; □ 1; ◆ 2; △ 3; ■ 4
indicates the presence of a mass transfer resistance,
which is possibly intraparticle diffusion or an external
film resistance[12]. Thus, in a semi batch supercritical
fluid extraction, the optimum solvent flow rate has to
be determined a priori. However, it should be noted
that the condition of optimum solvent flow rate de-
pends on the nature of the solvent-solute system, the
geometry of the extractor, the temperature, and the
pressure[9].
Figure 3 (c) shows the effect of the solvent flow
rate on the reduced extract yield as a function of time. (d) Effect of the carbon dioxide flow rate on the reduced an-
The reduced extract yields are plotted in Fig.3(d) as a drographolide yield of 0.375mm particle size as a function of
function of the amount of carbon dioxide consumed. the amount of carbon dioxide consumed at 40℃ and 10MPa
flow rate, ml·min－1: ◆ 0.5; □ 1; ▲ 2; ◇ 3; ■ 4
The reduced yield is defined as the ratio of yield to the
final yield. Since the final yield does not accurately Figure 3 Effect of carbon dioxide flow rate on the
andrographolide yield
reflect the same value, the reduced yield was used for
the purpose of comparison as suggested by Goto et
al.[13] and Roy et al.[7]. As seen in Fig.3(c), the ex- dioxide flow rate. The slopes of the extraction curve in
traction rate is higher at larger carbon dioxide flow the plot of reduced yield versus the amount of carbon
rate. On the other hand, the extraction rate in the plot dioxide consumed were smaller at higher flow rate.
of Fig.3(d) was affected by the supercritical carbon This figure indicates that the intraparticle diffusion
Chin. J. Ch. E. 15(6) 877 (2007)
880 Chin. J. Ch. E. (Vol. 15, No. 6)
resistance controls the extraction process[10,13,14]. If
the external mass transfer significantly affects the ex-
traction process as in the case of desorption from ac-
tivated carbon, the curve for lower flow rate becomes
gentler in Fig.3(d). All curves will coincide with each
other if the extraction rate is not affected by the su-
percritical carbon dioxide flow rate or if the solubility
controls the extraction process[13].
3.2 Effect of pressure
The effect of pressure ranging from 7.5 to 20MPa
on the andrographolide extraction was studied using
0.375mm particle size at temperature 40℃ and su-
－
percritical carbon dioxide flow rate of 2ml·min 1. The
effect of pressure on the andrographolide yield as a
function of time and the amount of carbon dioxide
consumed are shown in Figs.4(a) and 4(b), respec-
tively. The solubility of andrographolide in supercriti-
cal carbon dioxide estimated using Vetere’s method is
shown in Fig.5[15]. The solubility of andrographolide
(a) Effect of pressure on the andrographolide yield as a
function of extraction－time at 40℃ and carbon dioxide
flow rate 2ml·min 1 with 0.375mm particle size
(b) Effect of pressure on the andrographolide yield as a
function of carbon dioxide －consumed at 40℃ and carbon
dioxide flow rate 2ml·min 1 with 0.375mm particle size
Figure 4 Effect of pressure on the andrographolide yield
pressure, MPa: ◆ 7.5; □ 10; ▲ 12.5; ◇ 15; ■ 17.5; △ 20
Figure 5 The solubility of andrographolide in
supercritical carbon dioxide estimated
using Vetere methods
temperature, ℃: ◆ 30; □ 40; ▲ 45; △ 50; ■ 60
December, 2007
and the axial dispersion coefficient increase by way of
increasing the pressure at constant temperature; how-
ever, the effective diffusivity and the mass transfer
coefficient decrease. At the same time, both internal
and external mass transfer resistance increase owing
to the decrease of the effective molecular diffusion
and mass transfer coefficients with pressure[16]. For
this reason, supercritical carbon dioxide does not dis-
perse in solid particles easily[17]. This lower value of
the effective diffusion coefficient suggests that the
mass transfer resistance is located in the solid phase.
The values of carbon dioxide density, viscosity, and
self diffusion coefficient at different temperatures and
pressures are tabulated in Table 1. At constant tem-
perature, the density of the solvent increases with an
increase in pressure, but the vapour pressure of the
solute decreases with an increase in pressure. At ele-
vated pressure, the magnitude of such a solute vapour
pressure change becomes smaller[18] and the density
change becomes more effective and can easily over-
come the effect of the solute vapour pressure change
on the extraction rate. As a result, the andrographolide
yield increases with the increase of pressure, specifi-
cally from the near critical condition to the supercriti-
cal condition. This effect is due to the increase in
solubility of the solute mixture in the fluid phase.
Owing to this fact, Figs.4(a) and 4(b) revealed that at
40℃, the elevation of pressure showed a positive ef-
fect on the extraction. At this temperature, the increase
in the solvent density with pressure overcomes the
small change of the solute vapour pressure, resulting
in favoured extraction at higher pressure. Similar
finding was also obtained by Roy et al.[19] on the
extraction of ginger oil. This crossover property owing
to the solubility is controlled by a balance between the
solvent density and the solute vapour pressure changes
with the change of pressure and temperature.
The highest yield was found at the temperature of
40℃ and pressure of 20MPa. However, it was also
found that there was no significant increase of andro-
grapholide yield when the system pressure increased
from 10 to 20MPa. Therefore, a system condition at
40 ℃ and 10MPa pressure, which roughly corre-
sponds to a carbon dioxide density value of
－
0.6297g·cm 3, can be economically considered as the
best condition. Similar result was reported by Baysal
and Starmans[23], where they found that the best con-
dition for the supercritical carbon dioxide extraction
of limonene from caraway seed was at the carbon di-
－
oxide density value of about 0.6—0.7g·cm 3.
As pressure increases, all accessible solute on the
surface of the particles dissolve rapidly in the fluid
phase owing to an increase in the solvation power of
carbon dioxide[24]. However, further increase of
pressure from 10MPa upto 20MPa does not signifi-
cantly affect the andrographolide yield although the
solubility of andrographolide increases considerably.
An increase in pressure results in an increase in the
density of the supercritical fluid and a subsequent in-
crease in its solvating power. However, as the pressure
increases, the diffusivity decreases and the molecules
have more difficulty in diffusing into the pores to
 Supercritical Carbon Dioxide Extraction of Andrographolide from Andrographis paniculata 881

Table 1 Density, viscosity, and self diffusion coefficient of carbon dioxide at various temperatures and pressures
Temperature, ℃ Pressure, MPa Density①, g·cm－3 Viscosity②, g·cm－1·s－1 Self diffusion coefficient③, cm2·s－1
－2
7.5 0.6618 5.17×10 3.15×10－4
10.0 0.7722 6.72×10－2 2.70×10－4
12.5 0.8170 7.70×10－2 2.55×10－4
30
15.0 0.8476 8.03×10－2 2.46×10－4
17.5 0.8714 8.60×10－2 2.39×10－4
20.0 0.8911 9.04×10－2 2.34×10－4
－
7.5 0.2318 2.09×10 2 9.29×10－4
10.0 0.6297 4.82×10－2 3.42×10－4
12.5 0.7323 6.12×10－2 2.94×10－4
40
15.0 0.7811 6.88×10－2 2.76×10－4
－
17.5 0.8146 7.46×10 2 2.64×10－4
－
20.0 0.8406 7.96×10 2 2.56×10－4
7.5 0.2093 2.09×10－2 1.05×10－3
10.0 0.4997 3.61×10－2 4.38×10－4
12.5 0.6786 5.41×10－2 3.23×10－4
45
15.0 0.7430 6.29×10－2 2.95×10－4
17.5 0.7834 6.93×10－2 2.79×10－4
20.0 0.8135 7.45×10－2 2.19×10－4
－ －
7.5 0.1939 2.01×10 2 1.15×10 3
10.0 0.3852 2.85×10－2 5.77×10－4
12.5 0.6147 4.69×10－2 3.62×10－4
50
15.0 0.7009 5.71×10－2 3.17×10－4
17.5 0.7502 6.41×10－2 2.96×10－4
－
20.0 0.7851 6.96×10 2 2.83×10－4
7.5 0.1728 1.73×10－2 1.33×10－3
10.0 0.2898 2.42×10－2 7.91×10－4
－2
12.5 0.4729 3.47×10 4.85×10－4
60 －2
15.0 0.6055 4.63×10 3.79×10－4
－2
17.5 0.6774 5.44×10 3.38×10－4
－2
20.0 0.7244 6.06×10 3.16×10－4
① Taken from literature[20].
② Predicted using correlation proposed by Jossi, Stiel and Thodos[21].
③ Predicted using Liu-Selva-Macedo correlation[22].

dissolve the solute. In addition, an increase in pressure
causes the solid matrix to become more packed and
the void fraction decreases also reducing the extrac-
tion. The solubility of the solute in carbon dioxide is a
non-uniform function of temperature and pressure as
discussed below. Finally, the selectivity of solute ex-
traction is a function of pressure, i.e., as the pressure
increases, different solutes are preferentially ex-
tracted[25]. Therefore, an increase in pressure results
in opposing forces and it is reasonable to expect that
an optimum pressure may exist. It seems that the in-
ternal and external mass transfer resistance signifi-
cantly control the extraction rate at high pressure. At
high pressure, supercritical carbon dioxide does not
disperse in solid particles easily owing to the decrease
of the effective diffusion and mass transfer coeffi-
cient[25]. This finding can be supported by the fact
that intraparticle diffusion controls the desorption rate
of the solute from plant matrices as mentioned earlier
in the study of the effect of solvent flow rate on the
extraction rate. The local dissolving kinetics may de-
pend on the ratio of the solute content to the saturation
concentration (solubility) in the fluid phase, the bound
state of the solute in the matrix, and the interaction
with coexisting material[14]. In this system, it was
found that the extracted andrographolide amount is
considerably smaller than the solubility of pure an-
drographolide in supercritical carbon dioxide, indicat-
ing that the interaction between the solute and the
solid matrix is very strong.
The other possible reason for the andrographolide
yield extract not increasing at pressure 10MPa and
above is that the extraction of andrographolide from
plant is related to the extraction capabilities of other
matrix compounds. Then, a competitive extraction of
other diluting materials should occur with increasing
pressure. Brunner at al.[26] and de Lucas et al.[27]
observed a similar behavior in the supercritical extrac-
tion of tocopherol from palm leaves at 90℃ and pres-
sures ranging from 30 to 50MPa and of tocopherol
Chin. J. Ch. E. 15(6) 877 (2007)
882 Chin. J. Ch. E. (Vol. 15, No. 6)
from olive tree leaves at 40℃ and pressures ranging
from 25 to 45MPa.
3.3 Effect of temperature
The variation of temperature during supercritical
carbon dioxide extraction affected the density of car-
bon dioxide, the volatile property of the analytes, and
the desorption of the analytes from the plant matrices.
At higher temperature, the analytes become more
volatile but the supercritical carbon dioxide density
decreases[28]. Generally, at high pressure, the influ-
ence of temperature on the solubility is predominated
by the solid vapour pressure effect more than the su-
percritical carbon dioxide density variation, while at
low pressure, the effects of density predominate[29].
In this study, a magnitude pressure of 10MPa is con-
sidered to be low as compared to 20MPa. Therefore,
the effect of density change on the solubility of an-
drographolide dominates the effect of the solid vapour
pressure change. The evidence for this hypothesis can
be seen in Table 2 given below where the solubility of
andrographolide decreases with increasing tempera-
ture at 10MPa owing to the sharp decrease of the su-
percritical carbon dioxide density.
Table 2 Carbon dioxide density and solubility of
andrographolide at 10MPa
Andrographolide
Temperature,℃ Density, g·cm－3 ① －
solubility , mol·mol 1
30 0.7722 1.08×10－5
40 0.6297 4.39×10－6
45 0.4997 1.61×10－6
50 0.3852 6.88×10－7
60 0.2898 5.42×10－7
① Predicted using Vetere correlation[30].
Figures 6(a) and 6(b) show the increase of the
extraction rate with increasing temperature from 30℃
to 40℃ at 10MPa although the solubility of andro-
grapholide at 40℃ is lower than that at 30℃. The
increase in extraction efficiency with increasing proc-
ess temperature can be dependent on the molecular
mass, which is related to the binary diffusion coeffi-
cient of solute molecules in supercritical carbon diox-
ide[20]. The fluid mass transfer coefficient and the
effective diffusion coefficient increase with tempera-
ture owing to the increase in the binary molecular dif-
fusivity. The desorption kinetic constant also increases
more obviously with temperature when compared to
the adsorption kinetic constant according to Arrhenius
dependence[16]. In addition, analytes can be strongly
bound to the real-world samples, a certain amount of
energy is required to the extraction system, greatly
increasing the rate of the desorption process at high
temperatures[28].
Since the extraction of solute from plant matrices
is basically a desorption process, the increase of de-
sorption kinetic constant leads to an increase of the
extraction rate and the extract yield. However, further
increase of temperature reduces the extract yield. This
December, 2007
(a) Effect of temperature on the andrographolide yield as a
function of extraction time
－
at 10MPa and carbon dioxide
flow rate 2ml·min 1 with 0.375mm particle size
(b) Effect of temperature on the andrographolide yield as a
function of carbon dioxide consumed at 10MPa and carbon
dioxide flow rate 2ml·min－1 with 0.375mm particle size
Figure 6 Effect of temperature on the
andrographolide yield
temperature, ℃: ▲ 30; □ 40; ◇ 45; ■ 50; △ 60
is because the carbon dioxide density reduces signifi-
cantly when the temperature increases from 40℃ to
60℃ resulting in reducing the solubility of androgra-
pholide. The density of carbon dioxide is the pre-
dominant factor for the dissolution of organic mole-
cules in supercritical carbon dioxide. The density of
carbon dioxide determines strongly the number of
interactions between carbon dioxide and the solute
molecules. If sufficient interactions occur, the cohe-
sive forces between the individual solute molecules
are broken down rapidly and solubilization will occur.
Therefore, solubilization of solute in carbon dioxide
will be a function of the molecular mass of the solute
and the level of interaction between carbon dioxide
and the solute molecules[23]. At constant pressure, the
density of the solvent decreases with an increase in
temperature, but the vapour pressure of the solute in-
creases with an increase in temperature[21]. The sol-
vent density change becomes more effective and can
easily overcome the effect of the solute vapour pres-
sure change on the extraction rate[7]. This finding
agrees well with supercritical carbon dioxide extrac-
tion of poly-cyclic aromatic hydrocarbons from envi-
ronmental samples reported by Yang et al.[28].
Supercritical fluid extraction is basically a de-
sorption process. Generally, at constant pressure, the
adsorption equilibrium constant decreases with the
increase of temperature, and at constant temperature,
its value decreases with the increase of pressure[31].
However, the results found here show that the desorp-
tion rate decreases with the increase of temperature ow-
ing to the endothermicity of the adsorption process, and
thus, lower yields are obtained at higher temperatures.
 Supercritical Carbon Dioxide Extraction of Andrographolide from Andrographis paniculata
The extracts collected at high temperature were
yellowish in color and contained less andrographolide.
This fact agrees well with supercritical carbon dioxide+ 13
ethanol extraction of andographolide conducted by Ge
et al.[8], where he found that greenish extract con-
tained more andrographolide when compared to yel- 14
lowish extract. He also observed that at high tempera-
ture and pressure, the extraction rate was fast, which
was not investigated in this study. However, he re- 15
ported that the active ingredients’ contents were obvi-
ously low because andrographolide became unstable 16
at high temperatures and because of the possible oc-
currence of degradation.
4 CONCLUSIONS 17
In this study, the andrographolide extraction yield
of approximately 0.0174g andrographolide per gram
of andrographolide present in the A. paniculata leaves 18
using only supercritical CO2 extraction was obtained.
The extraction rate was found to be affected by the
intraparticle mass transfer of andrographolide. The
best extraction condition was found to be at 40℃,
－ 19
10MPa, and a flow rate of 2ml·min 1 for a 3g sample
of A. paniculata leaves.
20
REFERENCES
1 Wongkittipong, R., Prat, L., Damronglerd, S., Gourdon,
C., “Solid-liquid extraction of andrographolide from 21
plants-experimental study, kinetic reaction and model”,
Sep. Purif. Technol., 40, 147—154(2000).
2 Calabrese, C., Berman, S.H., Babish, J.G., “A phase I 22
trial of andrographolide in HIV positive patients and
normal volunteers”, Phytother. Res., 14, 333—338(2000).
3 Choudhury, B.R., Haque, S.J., Poddar, M.K., “In vivo
and in vitro effects of kalmegh (Andrographis paniculata)
extract and andrographolide on hepatic microsomal drug 23
metabolising enzymes”, Planta Med., 53, 135—140(1987).
4 Rajani, M., Shrivastava, N., Ravishankara, M.N., “A
rapid method for isolation of andrographolide from An- 24
drographis paniculata Nees (Kalmegh)”, Pharmaceut.
Biol., 38, 204—209(2000).
5 Tonthubthimthong, P., Chuaprasert, S., Douglas, P.,
Luewisutthichat, W., “Supercritical CO2 extraction of
nimbin from neem seeds—An experimental study”, J. 25
Food Eng., 47, 289—293(2001).
6 Jarvis, A.P., Morgan, E.D., “Isolation of plant products
by supercritical-fluid extraction”, Phytochem. Anal., 8,
217—2221997).
7 Roy, B.C., Goto, M., Kodama, A., Hirose, T., “Super-
critical CO2 extraction of essential oils and cuticular 26
waxes from peppermint leaves”, J. Chem.Technol. Bio-
technol., 67, 21—26(1996).
8 Ge, F.H., Lin, X.X., Huan, X.F., Shi, Q., Liang, B., Li, J., 27
Zhong, G., “Study on extraction of active ingredients
from Andrographis paniculata using orthogonal experi-
ment with supercritical carbon dioxide”, J. Chinese Med.
Mater., 25(2), 101—102(2002). (in Chinese) 28
9 Elkanzi, E.M., Singh, H., “ Extraction of used lubricating
oils with supercritical carbon dioxide”, Arabian J. Sci.
Eng., 26(2C), 11—23(2001).
10 Tonthubthimthong, P., Douglas, P.L., Douglas, S.,
Luewisutthichat, W., Teppaitoon, W., Pengsopa, L., “Ex- 29
traction of nimbin from neem seeds using supercritical
CO2 and a supercritical CO2-methanol mixture”, J. Su-
percrit. Fluids, 30, 287—301(2004).
11 Saldana, M.D.A., Zetzl, C., Mohamed, R.S., Brunner, G. , 30
“Decaffeination of guaraná seeds in a microextraction
column using water-saturated CO2”, J. Supercrit. Fluids,
22, 119—127(2002). 31
12 Macnaughton, S.J., Foster, N.R., “ Supercritical adsorp-
View publication stats
883
tion and desorption behaviour of DDT on activated car-
bon using carbon dioxide”, Ind. Eng. Chem. Res., 34,
275—282(1995).
Goto, M., Sato, M., Hirose, T., “Extraction of peppermint
oil by supercritical carbon dioxide”, J. Chem. Eng. Japan,
26(4), 401—407(1993).
Goto, M., Roy, B.C., Kodama, A., Hirose, T. , “Model-
ling supercritical fluid extraction process involving sol-
ute-solid interaction”, J. Chem. Eng. Japan, 31(2), 171—
177(1998).
Vetere, A., “A short-cut method to predict the solubilities
of solids in supercritical carbon dioxide”, Fluid Phase
Equil., 148, 83—93(1998).
Lucas, S., Calvo, M.P., Palencia, C., Cocero, M.J.,
“Mathematical model of supercritical CO2 adsorption on
activated carbon: Effect of operating conditions and ad-
sorption scale-up”, J. Supercrit. Fluids, 32, 193 —
201(2004).
Doker, O., Salgin, U., Sanal, I., Mehmetoglu, U., Calimli,
A., “Modelling of extraction of β-carotene from apricot
bagasse using supercritical CO2 in packed bed extractor”,
J. Supercrit. Fluids, 28, 11—19(2004).
Marentis, R.T., “Steps to developing a commercial su-
percritical carbon dioxide processing plant”, In: Super-
critical Fluid Extraction and Chromatography, Carpentier,
B.A., Sevenants, M.R., eds., American Chemical Society,
Washington, DC (1988).
Roy, B.C., Goto, M., Hirose, T., “Extraction of ginger oil
with supercritical carbon dioxide: experiments and mod-
eling”, Ind. Eng. Chem. Res., 35, 607—612(1996).
Span, R., Wagner, W., ”A new equation of state carbon
dioxide covering the region from triple-point temperature
to 1100K at pressure up to 800MPa.”, J. Phys. Chem. Ref.
Data, 25(6), 1509—1596(1996).
Jossi, J.A., Stiel, L.I., Thodos, G., “The viscosity of pure
substances in the dense gaseous and liquid phases”,
AIChE J., 8(1), 59—62(1962).
Liu, H., Silva, C.M., Macedo, E.A., “Unified approach to
the self-diffusion coefficients of dense fluids over wide
ranges of temperature and pressure-hard-sphere,
square-well, Lennard-Jones and real substances”, Chem.
Eng. Sci., 53(13), 2403—2422(1998).
Baysal, T., Starmans, D.A.J., “Supercritical carbon diox-
ide extraction of carvone and limonene from caraway
seed”, J. Supercrit. Fluids, 14, 225—234(1999).
de Franca, L.F., Reber, G., Meireles, M.A.A., Machado,
N.T., Brunner, G., “Supercritical extraction of carote-
noids and lipids from buriti (Mauritia flexuosa), a fruit
from the Amazon region”, J. Supercrit.Fluids, 14, 247—
256(1999).
D’Andrea, A., Ferri, D., Maccioni, O., van der Esch, S.A.,
Vitali, F., “Application of SFE technology to the extrac-
tion of active substances from Azadirachta indica (A.
juss) seeds”, Kleeburg, H., eds., In: Practice Oriented
Results on the Use and Production of Neem Ingredients
and Pheromones”, Drug und Graphic, Giessen (1994).
Brunner, G., Birtich, A., Johannsen, M., “Supercritical
fluid extraction of oil-palm components”, J. Supercrit.
Fluids, 8, 46—50(1995).
de Lucas, A., de la Ossa, E.M., Rincón, J., Blanco, M.A.,
Gracia, I., “Supercritical fluid extraction of tocopherol
concentrates from olive tree leaves”, J. Supercrit. Fluids,
22, 221—228(2002).
Yang, Y., Gharaibeh, A., Hawthorne, S.B., Miller, D.J.,
“Combined temperature & modifier effects on super-
critical CO2 extraction efficiencies of polycyclic aromatic
hydrocarbons from environmental samples”, Anal. Chem.,
67(3), 641—646(1995).
Kiriamiti, H.K., Rascol, E., Marty, A., Condoret, J.S.,
“Extraction rates of oil from high oleic sunflower seeds
with supercritical carbon dioxide”, Chem. Eng. Process.,
41, 711—718(2002).
Vetere, A., “A short-cut method to predict the solubilities
of solids in supercritical carbon dioxide”, Fluid Phase
Equilib., 148, 83—93(1998).
Skerget, M., Knez, Z., “Modelling high pressure extraction
processes”, Comp. Chem. Eng., 25, 879—886(2001).
Chin. J. Ch. E. 15(6) 877 (2007)