Journal of Food Engineering xxx (2016) 1e5

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Fractionating of green mandarin (Citrus deliciosa Tenore) essential oil

by vacuum fractional distillation
W.P. Silvestre a, c, *, F. Agostini b, L.A.R. Muniz a, G.F. Pauletti c
a
Course of Chemical Engineering, Center of Exact Sciences and Technology, University of Caxias do Sul (UCS), Brazil
b
Institute of Biotechnology, University of Caxias do Sul, Brazil
c
Course of Agronomy, Center of Biological and Health Sciences, University of Caxias do Sul, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: This work aims to evaluate the technical viability of vacuum fractional distillation to separate the
Received 14 August 2015 components of green mandarin (Citrus deliciosa Tenore) essential oil. Thermal degradation was also
Received in revised form analyzed. The obtained results demonstrate that vacuum fractional distillation is capable to separate the
29 December 2015
hydrocarbon terpenes, which were removed from the top/stages of the column, from the terpenes with
Accepted 14 January 2016
Available online xxx
other chemical functions, which remained in the bottom. Some trace compounds, such as methyl-N-
methyl anthranilate (mass fraction of 0,006 in the raw oil) and alpha sinensal (mass fraction of 0,004
in the raw oil), had their concentrations increased more than ten times after the separation, to mass
Keywords:
Fractioning column
fraction of 0,153 and 0,109, respectively. The mass fraction of the major compound of the essential oil, D-
Green mandarin essential oil limonene, was reduced from 0,707 in the raw oil, to 0,218 in the bottom. There was no evidence of
Terpenes thermal degradation in the products of the separation.
Vacuum distillation 2016 Elsevier Ltd. All rights reserved.
GC/MS

1. Introduction
The harvest of citrus fruits presents great importance world-
wide, as its production, in 2013, was approximately 111,5 millions of
tons. The main citrus fruits are orange, lemon, mandarin and lime,
which have a high commercial value as sell in nature or as a raw
material for food industry (Castro and Escobar, 2014).
The peel and the bagasse of citrus represent 40e50% of total
weight of the fruit and in general are waste of the processing of this
raw material, being both used as biomass. The bagasse is also used
as component in animal feeding. The peel has many bioactive
components, as ascorbic acid, terpenes and avonoids, which can
be extracted and after puried for various applications (Castro and
Escobar, 2014; Lohbauer, 2014).
The citrus essential oil is a mixture of terpenes, many of them
with important industrial applications, in medicine, pharmacology,
cosmetics and food industry (Bonaccorsi et al., 2009). Tough, to
have a practical application for the minority compounds (which, in
general, have more value), it is necessary to properly separate the
* Corresponding author. Laboratory of Chemistry and Soil Fertility, University of
Caxias do Sul, Road Francisco Getlio Vargas, 1135, Petropolis, 95074-270, Caxias do
Sul, RS, Brazil.
E-mail address: wpsilvestre@ucs.br (W.P. Silvestre).
mixture of substances in the raw oil (Castro and Escobar, 2014;
Embrapa, 2004; Simo ~es et al., 2004).
The genetics of the plant determines its chemotype. However,
the environmental conditions (climate, soil type, presence of nu-
trients) are capable of provoke considerable variations in the oil
and in the very chemotype, featuring an ecotype (Dellacassa,
2010). Is also important highlight that the composition of the
essential oil obtained varies considerably between the parts of the
plant, as well by the extractive process (Gamarra et al., 2006;
Simo ~es et al., 2004).
Bonaccorsi et al. (2009), Dugo et al. (2010) and Frizzo et al.
(2004) characterized the chemical composition of the Citrus deli-
ciosa Tenore essential oil obtained by cold pressing. The mandarin
essential oil is a mixture of terpenes and their oxygenated de-
rivatives. The volatile fraction of the mandarin oil obtained by cold
pressing is between 96 and 98% of the whole oil. The essential oil
has, as main compounds, D-limonene, g-terpinene, linalool, citral,
mircene, among others (Bonaccorsi et al., 2009; Favretto et al.,
1999; Teixeira et al., 2013). The major compound is the D-limo-
nene, with a percentage of 60e98% m/m of the essential oil present
in the peel. The other compounds, as g-terpinene, a-pinene, sabi-
nene, mircene, linalol, are present in variable quantities, depending
of the harvest, time of the year, location, among other factors
(Castro and Escobar, 2014; Favretto et al., 1999; Lota et al., 2014;

http://dx.doi.org/10.1016/j.jfoodeng.2016.01.011
0260-8774/ 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Silvestre, W.P., et al., Fractionating of green mandarin (Citrus deliciosa Tenore) essential oil by vacuum
fractional distillation, Journal of Food Engineering (2016), http://dx.doi.org/10.1016/j.jfoodeng.2016.01.011
2 W.P. Silvestre et al. / Journal of Food Engineering xxx (2016) 1e5
Simo ~es et al., 2004).
The aldehydes, alcohols and ketones that generally appear as
minority compounds or as traces are important in the composition
of the oil's avor, being products of high value when compared to
the other terpenes that form the mixture. Its applications go from
food industry to cosmetics and pharmaceutical industry
(Bonaccorsi et al., 2009; Stcker et al., 2011; Teixeira et al., 2013).
Between these compounds, there is methyl-N-methyl anthra-
nilate, a compound used as a avoring agent and construction block
in organic chemistry (Opdyke, 1970). Another compound of interest
by the industry is the aldehyde a-sinensal, used in cosmetics,
perfume and in food industry (Smith et al., 2005).
Most terpenes are thermally unstable, decomposing or oxidizing
in high temperatures or in presence of light or oxygen (Simo ~ es et al.,
2004).
The fractional distillation is a unit operation that aims the sep-
aration of two or more substances by the volatility difference be-
tween them. This process depends of the pressure and temperature
of the system, as well the physical and chemical characteristics of
the components to be separated (Foust et al., 1982).
The literature is scarce about using the vacuum fractional
distillation for to separate essential oil components. Farah et al.
(2006) reports the use of the fractional distillation to separate
components of Myrtus communis L., with interesting results.
Although, vacuum was not applied (Farah et al., 2006). Castilo-
Herrera et al. (2007) reported the use of fractional distillation
during the hydrodistillation process to obtain Lippia graveolens
H.B.K essential oil and to improve the phenolic fraction. The
distillation column was placed between the distillation ask and
the Clevenger apparatus. The author also reports the phenolic
fraction improved due to the fractional distillation (Castilo-Herrera
et al., 2007).
One of the objectives of the vacuum fractional distillation is to
verify a possible reduction of the degradation of the components of
the essential oil by the high temperature, problem also found in the
extraction with water vapor (drag with vapor) and in the hydro-
distillation (Soto et al., 2014; Favretto et al., 1999).
The objective of this work is to evaluate the use of the vacuum
fractional distillation to separate some compounds of the green
mandarin (C. deliciosa Tenore) essential oil, in addition to verify the
effect of the pressure reduction in the boiling temperature of the
mixture (raw essential oil).
2. Material and methods
2.1. Obtainment of essential oil
The green mandarin (C. deliciosa Tenore) essential oil was ob-
tained by cold pressing of the peels, provided by EcoCitrus, located
in the city of Montenegro, in the state of Rio Grande do Sul, Brazil.
The oil was produced in October of 2013, remaining stored in amber
bottle, closed, in cold chamber (temperature between 1 and 6
C).
2.2. Description of the distillation system
The fractionating system consisted of a glass column packed
with Raschig rings with 8 mm of diameter (packing height of 8 cm),
in three stages. The column and extract gathering system were
coupled to a vacuum pump. The pump had a relief valve to adjust
the vacuum of the system. An analogic vacuum gauge, localized in
the top of the column, measured the internal pressure thereof.
Between each stage, electronic valves (solenoids) and PT-100 per-
formed the gathering of the samples and measured the vapor
temperature, respectively. The solenoid valves, controlled by the
PLC, directed the vapor of each stage to the condenser, which was
Please cite this article in press as: Silvestre, W.P., et al., Fractionating of green mandarin (Citrus deliciosa Tenore) essential oil by vacuum
fractional distillation, Journal of Food Engineering (2016), http://dx.doi.org/10.1016/j.jfoodeng.2016.01.011
kept at 2e5
C, using ice packs. The system and its instrumentation
was controlled by a PLC Altus FBs-24MA (14 digital entrances and
10 digital exits) and by a computer through the software Winpro-
ladder, aiming to automatize the system.
2.3. Fractionating tests
Initially, it was tried to check the separation level of the mixture
compounds in function of the distillation time and the boiling
temperature in the bottom of the column (ask), keeping the vac-
uum in a constant value of 10 kPa. For this, distillation tests were
performed, gathering only vapor of the top of the column, with
collection time of 10 min, collecting part of the distillate for analysis
and discarding the remaining. It was utilized 120 mL of raw
essential oil in each batch. The distillation was performed until the
volume of liquid in the ask reduced to near 10 mL. The collected
samples were sent to GC/MS analysis.
After, it searched to verify the distillate composition in function
of the quantity of stages the vapor passed. The column operated
under vacuum (10 kPa) until the vapor reached the top of it. The
boiling temperature of the oil at this pressure was between 80 and
90
C. Then, the solenoid valves were opened and the vapor in each
stage was gathered until the volume in the distillation ask reduced
to the minimum possible, aiming to obtain circa 25 mL of extract to
each stage (and the minimum possible volume to the bottom
product), independently of the time necessary. The collected vapor
was condensed in cooled bottles (2e5
C), and the vapor temper-
ature was also monitored. The obtained samples were sent to GC/
MS analysis.
2.4. Chromatographic analysis
The equipment used for the analysis as a gas chromatograph
Hewlett Packard, model 6890 Series, equipped with a data pro-
cessor HP-Chemstation and using a HP Innowax column
(30 m  320 mm i.d.) with lm thickness of 0,5 mm and a mass
spectrometer Hewlett Packard, model MSD5973, equipped with a
software HP Chemstation and Wiley 275 spectral library, using a
capillary column of fused silica HP-Innowax (30 m  250 mm) with
lm thickness of 0,5 mm.
The GC analysis was performed with column temperature of
40
C (8 min) to 180
C at 3
C/min, 180e230/C at 20
C/min and at
230
C for 20 min. Injector temperature was 250
C; split ratio of
1:50, ionization detector by ame with temperature of 250
C;
using hydrogen as carrier gas with ux of 1,0 mL/min, at 34 kPa. For
the quantication of the chemical compounds, it was utilized an
internal standard of 1-octanol at 30,22 g/mL (25 mL), mixed with
hexane (75 mL) and with the sample to be analyzed (10 mL). The
used volume for injection in the GC/MS was 1,0 mL.
The MS analysis was performed in a gas chromatograph coupled
to a selective mass detector, as specied before. The used temper-
ature program was the same as of the GC; interface of 280
C; split
ratio of 1:100, using helium as carrier gas at 56 kPa. The gas ux
was kept at 1,0 mL/min; ionization energy of 70 eV; the injected
volume was 1 mL of sample, diluted in hexane (1:10).
3. Results and discussion
3.1. Results of the fractionating tests
The raw oil was analyzed to characterize it. Limonene and g-
terpinene were the major compounds. This is a common chemo-
type in the C. deliciosa Tenore essential oil (Dugo et al., 2010)
(Table 1).
Through the distillation tests with removing only the top
 W.P. Silvestre et al. / Journal of Food Engineering xxx (2016) 1e5 3

Table 1
Composition (in mass fraction) of the mandarin essential oil obtained by cold pressing. The results obtained are compared to the ones presented by Bonaccorsi et al. (2009);
Frizzo et al. (2004) and Dugo et al. (2010).
a b c
Compound This work Bonaccorsi et al. (2009) Frizzo et al. (2004) Dugo et al. (2010)

a-Pinene 0,019 0,024 0,007 0,019

a-Tujene 0,007 0,009 0,003 0,006
b-Pinene 0,014 0,017 0,009 0,013
Sabinene 0,002 0,003 0,001 0,002
Mircene 0,016 0,017 0,015 0,014
a-Terpinene 0,002 0,004 0,008 0,003
Limonene 0,707 0,705 0,728 0,727
b-Phellandrene 0,002 0,001 e 0,001
g-Terpinene 0,194 0,195 0,180 0,184
p-Cymene 0,009 0,003 e 0,004
D-Terpinene 0,007 0,008 e 0,006
Linalool 0,002 0,002 0,002 0,001
b-Caryophylene 0,002 0,001 0,001 0,001
a-Terpineol 0,003 0,002 0,004 0,002
a-Farnesene 0,003 0,001 0,005 0,001
Methyl-N-methyl anthranilate 0,006 0,004 0,007 0,006
Thymol 0,001 0,001 0,002 0,001
a-Sinensal 0,004 0,004 0,006 0,003
a
The average value was considered.
b
The value for cold pressed Cai mandarin were considered, in the month of March (the nearest month to October, the month the oil for this work was obtained).
c
The corrected average of the peak areas was considered.

product (in intervals of 10 min between each sample collecting)
and the bottoms (at the end of the batch), it was capable to make a
qualitative analysis of the more volatile compounds and of the
products that stayed in the distillation ask. This was done to
determine which compounds left the mixture rst and what of
them were xed. The quantitative and qualitative evaluation aimed
to verify the degree of separation in each distillate and its relation
with the raw essential oil (Table 2).
In relation to distillates gathered by column stage, the degree of
separation is similar to the separation obtained by collecting only
the top product. It is observed that the qualitative separation of the
compounds by stages is similar to the distillation tests removing
the top product, with exception of the mircene, who stayed in the
ask (although the mass fraction was very low).
However, the distillation time, as well as the batch regime can
cause differences between the results (Nadais and Bernardo-Gil,
1993). It is observed again the separation in terpenes and other
chemical classes, due to the physical and chemical properties of the
compounds in the same class be similar. In qualitative level, it was
not observed great deviations in relation to the prior tests, occur-
ring a large presence of limonene and g-terpinene in the bottoms
products (Table 3).
Beyond individual analysis, it was also aimed to observe the
composition of the fractions in terms of chemical functions to both

Table 2
Average mass fractionsa of the compounds in the raw oil, distillates and in the ask obtained in the distillation tests collecting only the top of the column and analyzed by GC/
MS. The grouping of the compounds in chemical classes is also presented.

Compound Raw oil Mass in raw oil (g) Distillate Bottom Mass in bottom (g) Recovery (%)b

1 2 3 4 5 6

a-Pinene 0,019 1,938 0,108 0,028 0,007 0,004 e e e e e

a-Tujene 0,007 0,714 0,046 0,009 0,002 e e e e e e
b-Pinene 0,014 1,428 0,036 0,031 0,013 0,007 0,003 0,004 e e e
Sabinene 0,002 0,204 0,019 0,005 0,003 e e e e e e
Mircene 0,016 1,632 0,037 0,032 0,020 0,013 0,008 0,005 e e e
a-Terpinene 0,002 0,204 0,200 0,005 0,005 0,004 0,003 0,002 e e e
Limonene 0,707 72,114 0,418 0,760 0,785 0,759 0,748 0,684 0,201 0,718 1,00
b-Phellandrene 0,002 0,204 0,041 0,002 0,002 0,002 0,002 0,002 e e e
g-Terpinene 0,194 19,788 0,089 0,119 0,152 0,195 0,218 0,279 0,217 0,775 3,91
p-Cymene 0,009 0,918 0,004 0,006 0,007 0,007 0,008 0,009 0,003 0,011 1,17
D-Terpineno 0,007 0,714 0,002 0,003 0,004 0,007 0,008 0,012 0,027 0,096 13,50
Linalool 0,002 0,204 e e e 0,002 0,002 0,003 0,013 0,046 22,75
b-Caryophylene 0,002 0,204 e e e e e e 0,059 0,214 99,75
a-Terpineol 0,003 0,306 e e e e e e 0,085 0,303 99,17
a-Farnesene 0,003 0,306 e e e e e e 0,085 0,303 99,17
Methyl-N-methyl anthranilate 0,006 0,612 e e e e e e 0,170 0,607 99,17
Thymol 0,001 0,102 e e e e e e 0,028 0,100 98,00
a-Sinensal 0,004 0,408 e e e e e e 0,114 0,407 99,75
Chemical classes
Hydrocarbon terpene 0,984 100,368 1,000 1,000 1,000 0,998 0,998 0,997 0,59 2,106 2,10
Alcohol 0,005 0,510 e e e 0,002 0,002 0,003 0,098 0,350 68,60
Phenol 0,001 0,102 e e e e e e 0,028 0,100 98,00
Aldehyde 0,004 0,408 e e e e e e 0,114 0,407 99,75
Amine/ether 0,006 0,612 e e e e e e 0,170 0,607 99,17
a
The marker e indicates absence of the substance in the fraction.
b
Of the bottom relative to the raw oil.

Please cite this article in press as: Silvestre, W.P., et al., Fractionating of green mandarin (Citrus deliciosa Tenore) essential oil by vacuum
fractional distillation, Journal of Food Engineering (2016), http://dx.doi.org/10.1016/j.jfoodeng.2016.01.011
4 W.P. Silvestre et al. / Journal of Food Engineering xxx (2016) 1e5

Table 3
Average mass fractionsa obtained in each of the stages of the column.

Average mass fraction

Compound Raw oil Mass in raw oil (g) Top Second stage First stage Bottom Mass in bottom (g) Recovery (%)

a-Pinene 0,019 1,938 0,037 0,005 e e e e

a-Tujene 0,007 0,714 0,013 0,003 e e e e
b-Pinene 0,014 1,428 0,033 0,011 0,005 e e e
Sabinene 0,002 0,204 0,006 0,002 e e e e
Mircene 0,016 1,632 0,032 0,018 0,008 0,003 0,019 1,17
a-Terpinene 0,002 0,204 0,005 0,004 0,003 e e e
Limonene 0,707 72,114 0,753 0,771 0,706 0,380 2,384 3,31
b-Phellandrene 0,002 0,204 0,002 0,002 0,002 0,001 0,006 3,13
g-Terpinene 0,194 19,788 0,110 0,172 0,256 0,284 1,811 9,15
p-Cymene 0,009 0,918 0,006 0,006 0,005 0,004 0,026 2,78
D-Terpineno 0,007 0,714 0,003 0,006 0,012 0,023 0,147 20,54
Linalol 0,002 0,204 e e 0,003 0,011 0,070 34,38
b-Caryophylene 0,002 0,204 e e e 0,031 0,198 96,88
a-Terpineol 0,003 0,306 e e e 0,031 0,306 100,00
a-Farnesene 0,003 0,306 e e e 0,052 0,306 100,00
Methyl-N-methyl anthranilate 0,006 0,612 e e e 0,101 0,606 98,96
Thymol 0,001 0,102 e e e 0,017 0,102 100,00
a-Sinensal 0,004 0,408 e e e 0,062 0,395 96,88
Chemical classes
Hydrocarbon terpene 0,984 100,368 1,000 1,000 0,997 0,778 4,896 4,88
Alcohol 0,005 0,510 e e 0,003 0,042 0,376 73,75
Phenol 0,001 0,102 e e e 0,017 0,102 100,00
Aldehyde 0,004 0,408 e e e 0,062 0,395 96,88
Amine/ether 0,006 0,612 e e e 0,101 0,606 98,96
a
The marker e indicates absence of the substance in the fraction.

tests and what would be the separation pattern. As noted earlier, is
visible the separation of the oil in hydrocarbon terpenes and in the
other chemical functions which, due to intermolecular interactions,
tend to have higher boiling points (Soto et al., 2014), staying in the
distillation ask. It is remarkable that, both in the rst distillations
than in the superior stages, the distillates were composed exclu-
sively by hydrocarbon terpenes. The only chemical function that
was detected outside the bottoms product was the alcohol function
(linalool), being detected in the nal stages of the separation
process.
The main difference between the raw oil and the distillates was
the visual appearance. The raw oil extracted by pressing has a dark
green coloration, characteristic of the fruit's peel. The distillate is
colorless, presenting no visual relation to the oil of origin. The
bottom product, due to the increase in the concentration of colored
compounds and due to the temperature, has a dark color and a high
viscosity.
The visual appearance gives to the distilled oil a better appli-
cability in the industry, because it prevents an unwanted coloration
of the products who use the oil as a raw material (Gamarra et al.,
2006). In sensory context, the distillates present a considerably
different aroma compared to the raw oil, while the bottom product
has a very sharp odor (Oliveira et al., 2011; Stcker et al., 2011).
The presence of aldehydes, alcohols and phenols in the bottoms
products, compounds who are responsible to the nuance of the raw
oil aroma, ends to sharp the aroma of the concentrated product in
the ask. This concentrated product, due to its composition, is of
the interest of the industry, despite its color, which, depending on
its application, must be removed (Favretto et al., 1999; Gamarra
et al., 2006; Simo~es et al., 2004).
Among the compounds who stayed in the ask, it was observed
that the mass fractions of some of them were increased up to 25
times in relation to the mass fraction in the raw essential oil,
indicating the potentialities of the process (Table 4). Among these
compounds, it is remarkable an increase of the mass fraction of the
methyl-N-methyl anthranilate, thymol and a-sinensal, when
compared to the raw oil. Despite of the large increase in the mass
fraction, the concentration effect can provoke interpretation errors,
as we can see the recovery of linalool was poor (22,75%) when
compared to the other compounds, with recoveries greater than
95%.
3.2. Evaluation of the oil thermal stability
Both distillates and bottom products were analyzed by GC/MS.
None of the distillates presented evidences of thermal degradation
nor oxidation of its compounds when compared to the raw oil, with
no different compound appeared in these fractions.
The bottom products, when analyzed by MS, also did not pre-
sent oxidation and decomposition evidences. Oxides of limonene
and terpinene were detected in considerably low concentrations
(mass fraction lesser than 0,01) in some bottom products. In these
products, it was observed the appearing of these substances, which
were not detected by the GC/MS analysis in the raw oil, neither in
the tests of gathering top product (Table 5).
The low mass fraction of these compounds can indicate a small
degradation of the oil, or can be trace compounds, which, due to the
low concentration, could not be detected in the raw oil. Bonaccorsi
et al. (2009) and Dugo et al. (2010) report these compounds as trace
compounds in cold pressed mandarin peel oil. Librando and
Tringali (2005), in a work regarding the oxidation of a-pinene,
presents the role of ozone, hydroxyl radical and UV light in the
oxidation of terpenes, more than moderate heating. The concen-
tration of the bottom product by removing the light terpenes,
decreasing more than 90% of the initial volume of the raw oil, al-
lows that trace compounds, before undetectable by the GC/MS,
could be detected, making it difcult to analyze the results
(Favretto et al., 1999; Lota et al., 2014; Stcker et al., 2011).
It is more probable that these compounds were already in the
raw oil as trace compounds than being products of the exposition of
the raw oil to heat and oxygen during the operation of the column.
If degradation exists, it is very low, not harming the quality of the
products obtained by vacuum fractional distillation.

Please cite this article in press as: Silvestre, W.P., et al., Fractionating of green mandarin (Citrus deliciosa Tenore) essential oil by vacuum
fractional distillation, Journal of Food Engineering (2016), http://dx.doi.org/10.1016/j.jfoodeng.2016.01.011
 W.P. Silvestre et al. / Journal of Food Engineering xxx (2016) 1e5 5

Table 4
Comparison between the mass fraction of some compounds in the bottom product compared to the raw essential oil.

Substance Average mass fraction (raw Mass in raw oil Average mass fraction Mass in bottom Relative increase to the raw oil Recovery
oil) (g) (bottom) (g) (times) (%)

Methyl-N-methyl 0,006 0,612 0,170 0,607 28,33 99,17

anthranilate
a-Sinensal 0,004 0,408 0,114 0,407 28,50 99,75
Thymol 0,001 0,102 0,028 0,100 28,00 98,00
Linalool 0,002 0,204 0,013 0,046 6,50 22,75
a-Terpineol 0,003 0,306 0,085 0,303 28,33 99,17

Table 5
Summary of the compounds detected in the GC/MS analysis of the bottom products and were not detected in the raw oil.

Substance Average mass fraction (bottom) Mass (g) Chemical class Possible origin

Terpinen-4-ol 0,005 0,018 aliphatic alcohol isomerization of a-terpineol

Perillaldehyde 0,008 0,029 aldehyde oxidation of limonene
Trans-carveol 0,004 0,014 aliphatic alcohol oxidation of limonene
Cis-b-terpineol 0,003 0,011 aliphatic alcohol isomerization of a-terpineol
a-Caryophylene 0,003 0,011 sesquiterpene isomerization of b-caryophylene

4. Conclusions
We observed that the vacuum fractional distillation is effective
to separate between chemical classes. The hydrocarbon terpenes
were separated from the ketones, alcohols and aldehydes and these
remained in the distillation ask. The equipment was not able to
separate efciently the terpenes between them without the use of
more stages, but the separation between chemical classes is of in-
terest of the industry.
The column's pressure inuenced in an important way the
boiling point of the essential oil, from near 180
C at 100 kPa (room
pressure) to 80e90
C at 10 kPa (maximum vacuum of the column).
There was no evidence of degradation of the distillates. In the
bottoms, it was observed the appearing of compounds that prob-
able are traces and, because of the reduction of the volume of oil,
could be detected; or they can also be arising from oxidation pro-
cesses. However, this degradation (if exists) did not reduce the
quality of the obtained products, both the distillates as the bottoms.
It was observed a considerable increase of the concentration of
some interest compounds in the bottoms products, with relative
increases in the mass fraction (compared to the raw oil) of more
than 25 times in some of these compounds. However, the recovery
varies for each compound.
Acknowledgments
The authors thank EcoCitrus for providing the essential oil and
the Secretariat of Science and Technological Innovation of the state
of Rio Grande do Sul by the incentive and the nancing of the
equipment, both of them giving support to the achievement of this
research.
References
Bonaccorsi, I.L., et al., 2009. Characterization of mandarin (Citrus deliciosa Ten.)
essential oil. Determination of volatiles, non-volatiles, physico-chemical indices
and enantiomeric ratios. Nat. Product. Commun. 4 (11), 1595e1600.
Castilo-Herrera, G.A., et al., 2007. Extraction method that enriches phenolic content
in oregano (Lippia graveloens H.B.K.) essential oil. J. Food Process Eng. 30,
661e669.
Castro, M.D.L., Escobar, C.A.L., 2014. Towards a comprehensive exploitation of citrus.
Trends Food Sci. Technol. xx, 1e13.
Dellacassa, E., 2010. Normalizacio n de productos naturales obtenidos de especies de
la ora aroma tica latinoamericana: proyecto CYTED IV.20. EDIPUCRS, Porto
Alegre (org.).
Dugo, P., et al., 2010. Analytical characterization of mandarin (Citrus deliciosa Ten.)
essential oil. Flavour Fragr. J. 21, 34e46.
Embrapa, 2004. Citrus map in the RS. Cultiv. HF 08e09.
Farah, A., et al., 2006. Fractional distillation effect on the chemical composition of
Moroccan myrtle (Myrtus communis L.) essential oils. Flavour Fragr. J. 21,
351e354.
Favretto, L., et al., 1999. Chemometric analysis and extraction processes of mandarin
essential oils. Anal. Chim. Acta 312, 199e205.
Foust, A.S., et al., 1982. Principles of Unit Operations, second ed. Guanabara Dois, Rio
de Janeiro.
Frizzo, C.D., et al., 2004. Composition and seasonal variation of the essential oils
from two mandarin cultivars of Southern Brazil. J. Agric. Food Chem. 52,
3036e3041.
Gamarra, F.M.C., et al., 2006. Inuence on the quality of essential lemon (Citrus
aurantifolia) oil by distillation process. Braz. J. Chem. Eng. 23 (1), 147e151.
Librando, V., Tringali, G., 2005. Atmospheric fate of OH initiated oxidation of ter-
penes. Reaction mechanism of a-pinene degradation and secondary organic
aerosol formation. J. Environ. Manag. 75, 275e282.
Lohbauer, C., 2014. Citrus Panorama in World. Available in: http://www.citrusbr.
com.br. Accessed in Aug. 28, 2014.
Lota, M.L., et al., 2014. Chemical variability of peel and leaf essential oils of man-
darins from Citrus reticulata Blanco. Biochem. Syst. Ecol. 28, 61e78.
Nadais, M.H., Bernardo-Gil, M.G., 1993. Vapour-liquid equilibria of a-
pinene limonene at reduced pressures. Fluid Phase Equilib. 91, 312e330.
Oliveira, J.V., et al., 2011. Fractionation of citronella (Cymbopogon winterianus)
essential oil and concentrated orange oil phase by batch vacuum distillation.
J. Food Eng. 102, 348e354.
Opdyke, D.L.J., 1970. Metyhl-N-methyl Anthranilate. Fragrance Raw Materials
Monographs, pp. 791e792.
Simo~ es, C.M.O., et al., 2004. Farmacognosia: da planta ao medicamento, 5 ed.
Editora UFRGS/Editora da UFSC, Porto Alegre/Floriano polis (org.).
Smith, R.L., et al., 2005. A procedure for the safety evaluation of natural avor
complexes used as ingredients in food: essential oils. Food Chem. Toxicol. 43,
345e363.
Soto, A., et al., 2014. Improved concentration of citrus essential oil by solvent
extraction with acetate ionic liquids. Fluid Phase Equilib. 361, 37e44.
Stcker, C.Z., et al., 2011. Uso de destilaa ~o fracionada (DF) para obtena ~o de pro-
dutos diferenciados em perfumaria a partir de o leos essenciais de ctricos. In:
Reunia ~o da Sociedade Brasileira de Qumica, Floriano polis, SC.

Teixeira, D.M.A., et al., 2013. Oleos essenciais aplicados em alimentos: uma revisa ~o.
Rev. Bras. Pesqui. em Aliment. 4 (2), 69e81.

Please cite this article in press as: Silvestre, W.P., et al., Fractionating of green mandarin (Citrus deliciosa Tenore) essential oil by vacuum
fractional distillation, Journal of Food Engineering (2016), http://dx.doi.org/10.1016/j.jfoodeng.2016.01.011