Content uploaded by Jose Antonio Rocha-Uribe
Author content
All content in this area was uploaded by Jose Antonio Rocha-Uribe on Oct 29, 2015
Content may be subject to copyright.
Disponible en: http://www.redalyc.org/src/inicio/ArtPdfRed.jsp?iCve=48221175004
Redalyc
Sistema de Información Científica
Red de Revistas Científicas de América Latina, el Caribe, España y Portugal
Castillo-Ruz, María Carolina; Guillermo-Alcocer, Carlos Gerardo; Bojórquez-Gamboa,
Rubén Ricardo; Rocha-Uribe, José Antonio
Extraction of vanilla oleoresin (Vanilla planifolia Andrews) with supercritical CO2
Tecnología, Ciencia, Educación, vol. 26, núm. 2, julio-diciembre, 2011, pp. 80-84
Instituto Mexicano de Ingenieros Químicos
Distrito Federal, México
¿Cómo citar? Número completo Más información del artículo Página de la revista
Tecnología, Ciencia, Educación
ISSN (Versión impresa): 0186-6036
imiqac@sercom.com.mx
Instituto Mexicano de Ingenieros Químicos
México
www.redalyc.org
Proyecto académico sin fines de lucro, desarrollado bajo la iniciativa de acceso abierto
80 Tecnol. Ciencia Ed. (IMIQ) vol. 26 núm. 2, 2011
Extraction of vanilla oleoresin (Vanilla planifolia
Andrews) with supercritical CO2
Tecnol. Ciencia Ed. (IMIQ) vol. 14 núms.1-2,1999 80Tecnol. Ciencia Ed. (IMIQ) 26(2): 80-84, 2011
*Autor a quien debe dirigirse la correspondencia
(Recibido: Junio 3, 2011,
Aceptado: Junio 29, 2011)
María Carolina Castillo-Ruz, Carlos Gerardo Guillermo-Alcocer, Rubén Ricardo Bojórquez-
Gamboa, José Antonio Rocha-Uribe*
Universidad Autónoma de Yucatán, Facultad de Ingeniería Química, Periférico Nte. Km 33.5 Tablaje Catastral 13615, Col.
Chuburná de Hidalgo Inn, 97203 Mérida, Yucatán; México. Tel. (Phone) (+52 999) 946 0981 Ext 1177; Fax 52 (999) 946 0994,
correo-e (e-mail): antonio.rocha@uady.mx
Extracción de oleorresina de vainilla (Vanilla
planifolia Andrews) con CO2 supercrítico
Palabras clave: CO2, extracción, uido supercrítico, vainilla,
vainillina, Vanilla planifolia Andrews
Keywords: CO2, extraction, supercritical uid, vanilla,
vanillin, Vanilla planifolia Andrews
ABSTRACT
Vanilla oleoresin is a widely used natural product, mainly used as avor.
In this work, the extraction of vanilla oleoresin using supercritical CO2
is evaluated. The main extraction parameters on the oleoresin yield and
composition was determined, namely: pressure, temperature, particle
size, and contact time. Results indicate that lower particle size favors
mass transfer and oleoresin yield. High pressure and temperature also
provide higher yields of vanilla oleoresin. Extreme conditions of pressure
and temperature decrease the quality of vanillin possibly because of
the degradation of some compounds. The maximum concentration
of vanillin (97.35% w/w) is found at 408 bars and 40oC, with a mass
yield of 5.82% after 40 minutes of dynamic extraction. The extraction
of vanilla oleoresin with supercritical CO2 is possible, and optimizing
the operational conditions may provide higher oleoresin yield with
good vanillin concentration; an economic analysis would be needed to
determine the best extraction time.
RESUMEN
Actualmente la oleorresina de vainilla es uno de los productos naturales
con alto uso. Su principal aplicación es como saborizante. En este trabajo
se evalúa la extracción de oleorresina de vainilla usando CO2 supercrítico.
Los principales parámetros variados para ver el efecto sobre el rendimiento
y la concentración de vainillina fueron: presión, temperatura, tamaño de
partícula, y tiempo de contacto. Obviamente los tamaños de particula
menores favorecen la transferencia de masa y el mayor rendimiento de
la oleorresina. Los valores mayores de presión y temperatura también
dan mejores rendimientos de la oleorresina de vainilla. Las condiciones
extremas de presión y temperatura reducen la calidad de la vainillina
posiblemente debido a la degradación de algunos compuestos. La
concentración más alta de vainillina (97.35% masa) se encuentra a una
presión de 408 bar y a una temperatura de 40oC, con un rendimiento en
masa de 5.82% con 40 minutos de extracción dinámica. La extracción de la
vainilla con CO2 supercrítico es posible y la optimización de las condiciones
de operación puede mejorar el rendimiento de la oleorresina con una buena
concentración de vainillina. Para obtener el mejor tiempo de extracción es
necesario realizar un análisis económico de factibilidad.
INTRODUCTION
Vanilla (Vanilla planifolia Andrews) is native from
southeastern Mexico and is considered the world’s main
avor (Brandt, 1996, Havkin-Frenkel y Belanger, 2011).
The rst component of vanilla aroma is vanillin, which
is obtained from a natural fermentation process known
as curing. According to Ranadive (1992), since most
of the components that provide the avor are volatile
and thermolabile, extraction and purication methods
of vanilla extracts determine the overall quality of the
nal product. Presently, the main extraction method for
vanilla oleoresin is multistage lixiviation, or leaching at
80oC employing ethyl alcohol, but if the process does
not have a good control of temperature, degradation of
vanilla may occur and a loss of extract exists.
Supercritical extraction with CO2 seems to be a good
alternative, because it operates at a lower temperature,
mass transfer is improved, and no residual solvent
Tecnol. Ciencia Ed. (IMIQ) vol. 26 núm. 2, 2011 81
is present in the nal product. These advantages are
important and seem promising for the industrial use
of this technology in the extraction of thermolabile
compounds (Moyler and Heath, 1988).
Moyler (1987) reviews the sub- and supercritical
CO2 extraction of essential oils and compare results
with steam distillation for ginger, hop, clove bud,
black pepper, juniper berry oil, and vanilla bean. Rice
and Singh obtained a patent in 1990 for the continuous
extraction of solid material using solvents such as CO2
by circulating a mixture of the solids in CO2 at elevated
pressure through a closed-loop pipeline. Carbonell
(1991) reviews extraction with CO2 of ginger, black,
green, and white pepper, and vanilla. Fu et al. (2002a,b),
report the extraction of vanilla with supercritical CO2
fluid. Kvasenkov and Kvasenkov (2010) patented
the preparation of coffee-substitute beverages using
supercritical extraction of a mixture of cocoa husk,
cinnamon and vanilla spices with liquid nitrogen.
The objective of this research is to evaluate the yield
and nal concentration of vanilla oleoresin extracted
by supercritical CO2. A study of the effect of pressure,
temperature, particle size, and contact time is also
included.
MATERIALS AND METHODS
Materials
The raw materials were vanilla beans from Papantla,
Veracruz, Mexico (from Casa Larios). The vanilla
beans were classied as superior quality based on the
parameters accepted in Mexico (Mexican Act NMX-
FF-074-1996). The content of vanillin in the vanilla
beans was determined according to the same norm. The
traditional extraction procedure consists of maceration
of vanilla beans cut to about 0.65 cm on ethanol.
Total vanillin content was 3.6% w/w after 3 days of
maceration time. The extraction yield is dened as
grams of oleoresin extracted and recovered divided by
the grams of solid material in the extraction recipient.
The concentration of vanillin was characterized by
HPLC as recommended by Thomson and Hoffmann
(1988). Yield and composition obtained in this research
are compared against the values reported by Nguyen et
al. (1991).
Solvents
The solvent was industrial-grade CO2 at 99.98% w/
w, supplied by Praxair-Mexico. Other reactants and
solvents were purchased from Merck (Mexico). Vanillin
standard with 99% purity was supplied by Sigma-
Aldrich (Mexico).
SUPERCRITICAL EXTRACTION
Equipment
The supercritical extraction of vanilla oleoresin was
performed with lab-scale equipment (model SFT-150
System Extractor from Supercritical Fluid Technologies,
Inc. Newark, DE, USA) A typical diagram is shown in
Figure 1. The capacity of the extraction cell is 0.1 L
and usually was charged with 23 g of vanilla beans.
The equipment has an air-driven reciprocating pump
to compress the CO2. The heating elements are 2000
W resistances and a temperature proportional integral
derivative (PID) controller allows xing the temperature
in the range from 0 to 120oC. The ow of CO2 is
between 1 and 330 mL/min (1-250 g/min) of liquid CO2
under normal operation.
Operational conditions
Table 1 shows the operational conditions studied in
the experimental runs (data obtained by triplicate) and
reported by Castillo-Ruz (2007). The temperature and
pressure range is wider than that reported by Nguyen
et al. (1991). The extractions were performed at 40,
45, and 50ºC, and at four different pressures 272,
Figure 1. Extraction system using supercritical
CO2: A. Air cylinder (T-1) and CO2
cylinder (T-2); B. Manometer for air
(M-1 y M-2); C. Cooler for liquid CO2
(IC-1); D. Reciprocant pump (B-1); E.
Regulatory valve for air fed to pump
(V-5); F. Extractor (E-1); G. Valve
on/off (V-6); H. Needle valve (V-7); I.
Containers: Extractor (E-2) solvent
trap (E-3); J. Flow meter (R-1)
M-1
M-2P
P
T-1 T-2
A la atmósfera
R-1
E-3
E-1
E-2
G H
V-6
V-8
V-2
V-1
V-4
V-3 V-5
V-7
I
J
F
D
C
E
B
IC-1
B-1
A
82 Tecnol. Ciencia Ed. (IMIQ) vol. 26 núm. 2, 2011
340, 408, and 476 bar (4000, 5000, 6000, and 7000
psi, respectively). The particle size was between
16 and 30 μm for most of the runs. One of the runs
included vanilla beans pieces of 5 mm in length as a
reference. Particle size was detemined with standard
sieves(Newark, US). For most of the runs, contact
time was 40 minutes, but for each pair of pressure and
temperature contact times of 30 and 60 minutes were
also tested to see the effect of these variables.
Table 1
Summary of the operational conditions used
in the experimental data to be obtained from
the temperature dependent design for vanilla
supercritical extraction
Temperature, °C
40 45 50
Pressure (bar)
272.11 272.11 272.11
340.14 340.14 340.14
408.16 408.16 408.16
476.19 476.19 476.19
For each supercritical extraction run with CO2 the
yield was calculated as follows:
yield = –––––––––––––––––––––––––––––––––
grams of vanilla oleresin extracted
grams of vanilla charged to extraction cell (1)
Extraction procedure
Referring to Figure 1, at the beginning of the extraction
the on/off valve (V-6) is closed and valves V-2 and V-4
for liquid CO2 are opened. The extraction cell (E-1)
is lled with 23 g of vanilla beans, closed, and xed.
The desired temperature set point is set on the PID
controller. The pressure is gradually increased by using
the regulatory valve V-5 until the desired pressure is
obtained. The soaking time starts once the operational
conditions for temperature and pressure are established;
this takes 15 minutes, and during this lapse there is no
ow of CO2 to the extraction cell. After the soaking
time, CO2 is allowed to ow by opening the on/off
valve (V-6); then the supercritical CO2 ows upwards
through the solid matrix of vanilla beans, extracting
the oleoresin both from the pores and the outer part of
the bean particles, diffusing out along with the CO2.
The extract is collected in the container (E-2), where
the CO2 de-pressurization occurs down to atmospheric
conditions. The continuous extraction process takes
40 minutes. The decompressed gaseous CO2 leaves
the container (E-2) and enters a second container (E-3)
which is immersed in ice. The purpose of this second
container is recovering the extract that may remain
dissolved in the gaseous CO2. The ow used for the
dynamic extraction period is measured with a ow meter
(R-1) before CO2 is discarded to the atmosphere. The
CO2 ow is regulated with a needle valve (V-7). When
the dynamic extraction period ends, valves V-1, V-2,
V-3, and V-4 are closed, and valve V-6 is opened. Then
the extraction cell (E-1) is de-pressurized, the remaining
solid matrix and the collected extract are weighed, a
mass balance is performed, and the oleoresin yield is
calculated with Equation 1.
HPLC analysis
Identication and quantication of vanilla in the extracts
was carried out by liquid chromatography (Perkin
Elmer 250) with a binary pump and a UV absorbance
detector at 275 nm. For the chromatographic separation,
an analytic column Spherisorb ODS2 was used. For
identication and quantication of vanillin, a standard
with 99% purity was obtained from Sigma-Aldrich
(Mexico).
RESULTS AND DISCUSSION
Effect of particle size on the yield of vanilla
oleoresin
Preliminary extractions at 272 bar (4000 psi) and
50oC using the 5 mm and 16-30 µm particle sizes,
were performed. As expected, the yield (4.95%) for
the smaller matrix (16-30 µm) is higher than the yield
(1.5%) for vanilla beans cut at 5 mm, confirming
that smaller size favors mass transfer of solute to the
supercritical CO2, because of a larger transfer surface-
area, according to Fick’s law. Therefore, all the other
runs were performed using the smaller particle size
(16-30 µm).
Effect of temperature and pressure on the
oleoresin yield
Figure 2 show the oleoresin yields obtained at different
pressure and temperature with a dynamic time of 40
minutes.
A general tendency of higher yields for higher
temperature and pressure conditions is observed. It seems
that pressure increments provide higher oleoresin yields.
A maximum yield of 7.77% is found for 50oC and 476.19
bar for an extraction time of 40 minutes, for grinded
vanilla beans (16-30 µm). Nguyen et al (1991) reported
Tecnol. Ciencia Ed. (IMIQ) vol. 26 núm. 2, 2011 83
yields from 0.3 to 8.0%, using a lower temperature range
(33-36oC) and lower operational pressure (110 bar), as
well as an average particle thickness of 1.8 mm, average
width of 4 mm and average length of 19 mm, and higher
extraction times (5-1540 h). These authors also report
experimental data for 110 bars, 36oC with cryoground
vanilla with a yield of 10.6% for a time of 50 hours.
Yields obtained in this research are higher than those
reported by Nguyen et al (1991), with a lower extraction
time. The reasons in order of importance from low to
high are:
a. Higher temperature (40-50 vs. 33-36oC)
b. Higher pressure (272-476 vs. 110 bar)
c. Lower particle size (0.023 vs. 19 mm)
Identication of vanillin on oleoresin
Once the yield of oleoresin was measured, the oleoresin
was analyzed for vanillin and other compounds by
liquid chromatography as recommended by Thompson
and Hoffmann (1988). Retention time for vanillin
was 7 minutes. Depending upon temperature and
pressure conditions used on the extractions, other
compounds that contributed for the vanilla oleoresin
bouquet were found. For example, on the performed
extract chromatograms (272 bars and 50oC) a peak at
2.59 minutes of retention time was observed, which
corresponds to p-hydroxybenzoic acid.
Effect of temperature and pressure on vanillin
concentration
Figure 3 shows the percentages of vanillin, measured
on the extracts by using HPLC analyses. Considering
the pressure and temperature that provided the
maximum percentage of vanillin, it was found that
intermediate pressure values between 340 and 408 bar,
and temperatures around 40 and 45oC favor vanillin
concentration. The maximum mass concentration of
vanillin (97.35%) is found at a pressure of 408 bars and
a temperature of 40oC.
Figure 2. Yield of vanilla oleoresin expressed on
mass percentage at 40, 45, 50ºC
Effect of contact time
Figure 4 shows the effect of dynamic extraction time.
As expected, there is a higher extraction yield for longer
times. It is observed that, mainly for higher pressures,
from 30 to 40 minutes the increment in the yield is about
double than the increment from 40 to 60 minutes.
Using an extraction time of 40 minutes leaves solute
in the solid matrix. Working at extraction times of 60
minutes the recovery of solute is a little higher. For
industrial extraction, an economical analysis would be
needed to determine the optimum extraction time.
Figure 4. Effect of extraction time over oleoresin
yield (30, 40, 60 min)
8
7
6
5
4
3
2
1
0
272.11 340.14 408.16 476.19
Pressure (bar)
Yield
40
40 40 40
45 45
45
45
50
50 50
50
Figure 3. Mass percentage of vanillin at different
operation conditions (40, 45, 50 ºC)
100
90
80
70
60
50
40
30
20
10
0
272.11 340.14 408.16 476.19
Pressure (bar)
40 40
40
40
45
45
45
45
50
50
50 50
Oleoresin yield
8
7
6
5
4
3
2
1
0
P=272 bar,
T=50ºC
P=340 bar,
T=45ºC
P=408 bar,
T=45ºC
P=476 bar,
T=50ºC
30 30 30
30
40 40 40
40
60
60
60
60
84 Tecnol. Ciencia Ed. (IMIQ) vol. 26 núm. 2, 2011
CONCLUSIONS
From this research the following conclusion may be
drawn:
Lower particle size favors mass transfer and oleoresin
yield, higher pressure and temperature also provide
higher yields of vanilla oleoresin.
Extreme conditions of pressure and temperature
decrease the quality of vanillin, possibly due to
degradation of some compounds. The maximum
concentration of vanillin (97.35% w/w) is found at a
pressure of 408 bars and a temperature of 40oC, with
a mass yield of 5.82% after 40 minutes of dynamic
extraction.
The extraction of vanilla with supercritical CO2 is
possible and optimizing the operational conditions may
provide higher oleoresin yields with a good vanillin
concentration.
An economical analysis is needed to determine the
optimum extraction time.
ACKNOWLEDGMENTS
The authors would like to acknowledge the valuable
support of:
i. Dr. Juan-Daniel Pacho-Carrillo and Dra. Rosa-
María Domínguez-Espinoza
ii. Dr. Julio C. Sacramento Rivero
iii. PRIORI Program (Program of Impulse and
Orientation to Research, Spanish acronym),
Autonomous University of Yucatan, Mexico
iv. Multimarcas y Servicios de México, S.A.
v. The experimental work was performed at the
Faculty of Chemical Engineering, Autonomous
University of Yucatan, Mexico. It was reported as
the professional thesis of María Carolina Castillo-
Ruz (2007).
NOMENCLATURE
HPLC High performance liquid chromatography
PID Proportional integral derivative controller
BIBLIOGRAPHY
Brandt, L. 1996. The creation and use of vanilla. Food Product Design
Magazine. March 01. Available from http://www.foodproductdesign.com
Carbonell, E. S. 1991. Extraction of avors with supercritical carbon
dioxide. Cereal Foods World. 36(11):935-937.
Castillo-Ruz, M. C. 2007. Extracción de oleorresina de vainilla (Vanilla
planifolia Andrews) con CO2 supercrítico . Tesis profesional.
Industrial Chemical Engineering, Universidad Autónoma de Yucatán.
Mérida, Yucatán, México.
Fu, S., Huang, M., Zhou, J., Li, S. 2002a. Determination of chemical
constituents of vanilla by supercritical CO2 uid extraction. Huaxue
Yanjiu Yu Yingyong. 14:208-210.
Fu, S., Huang, M., Zhou, J., Li, S. 2002b. Study of components of vanilla
extract by different extracting technology. Shipin Kexue. 23:109-112.
Havkin-Frenkel, D., Belanger, F. 2011. Handbook of Vanilla Science and
Technology. Wiley-Blackwell. Nueva York, NY. EEUU.
Kvasenkov, O. I., Kvasenkov, I. I. 2010. Method for production of coffee
drink “Rossiiskii”. Russian Patent RU 2402914 C1. Moscú, Rusia.
Mexican Act NMF-FF-074-1996: Food products not industrialized for human
consumption – species and condiments –entire and dry vanilla (Vanilla
fragrant Salisbury Ames o Vanilla planifolia Andrews). Diario Ocial de
la Federación. Estados Unidos Mexicanos. Mexito City, Mexico.
Moyler, D. A. 1987. Liquid carbon dioxide extraction of avor materials. En
Dev. Food Flavours [Industrial Co-op Symp.], pp. 119-29. Conference.
General review 1986. Elsevier Appl. Sci. Londres, Reino Unido.
Moyler, D. A., Heath, B. 1988. Liquid carbon dioxide extraction of essential
oils. Flavors and fragrances: A world perspective. En Proceedings of
the 10th International Congress of Essential Oils, Fragrances and
Flavors. Pp. 41-63. Elsevier. Londres, Reino Unido.
Nguyen, K., Barton, P., Spencer, J. 1991. Supercritical carbon dioxide
extraction of vanilla. The Journal of Supercritical Fluids. 4(1):40-46.
Ranadive, A. S. 1992. Vanillin and related avor compounds in vanilla
extracts made from beans of various global origins. J. Agric. Food
Chem. 40:1922-1924.
Rice, W.K., Singh, L. 1990. Dynamic supercritical uid extraction system,
US Patent No. 4898673 A. Washington, DC. EEUU.
Thompson, R. D., Hoffmann, T. J. 1988. Determination of coumarin as an
adulterant in vanilla avoring products by high performance liquid
chromatography. Journal of Chromatography. 438:369-382.