GRASAS Y ACEITES 67 (1)

January–March 2016, e121

ISSN-L: 0017-3495
doi: http://dx.doi.org/10.3989/gya.0493151

Solid carbon dioxide to promote the extraction of extra-virgin olive oil

A. Zinnaia, F. Venturia,*, M.F. Quartaccia, C. Sanmartina, F. Favatib and G. Andricha

a
Department of Agriculture, Food and Environment, University of Pisa, Via del Borghetto 80, 56124 Pisa, Italy
b
Department of Biotechnology, University of Verona, Strada Le Grazie 15, 37134 Verona, Italy
*
Corresponding author: francesca.venturi@unipi.it

Submitted: 14 April 2015; Accepted: 20 July 2015

SUMMARY: The use of solid carbon dioxide (dry ice) as a cryogen is widespread in the food industry to
produce high quality wines, rich in color and perfumes. The direct addition of carbon dioxide to olives in the
solid state before milling represents a fundamental step which characterizes this innovative extraction system.
At room temperature conditions solid carbon dioxide evolves directly into the air phase (sublimation), and the
direct contact between the cryogen and the olives induces a partial solidification of the cellular water inside the
fruits. Since the volume occupied by water in the solid state is higher than that in the liquid state, the ice crystals
formed are incompatible with the cellular structure and induce the collapse of the cells, besides promoting the
diffusion of the cellular substances in the extracted oil, which is thus enriched with cellular metabolites char-
acterized by a high nutraceutical value. Furthermore, a layer of CO2 remains over the olive paste to preserve
it from oxidative degradation. The addition of solid carbon dioxide to processed olives induced a statistically
significant increase in oil yield and promoted the accumulation of tocopherols in the lipid phase, whereas a not
significant increase in the phenolic fraction of the oil occurred.
KEYWORDS: Carbonic snow; Cryomaceration; Extraction index; Extra-virgin olive oil; Product quality

RESUMEN: Dióxido de carbono sólido para promover la extracción del aceite de oliva virgen extra. El uso de dióxido
de carbono sólido (hielo seco) como criogénico está muy extendido en la industria alimentaria para producir vinos
de alta calidad, ricos en color y perfumes. La adición directa de dióxido de carbono en estado sólido a las aceitu-
nas antes de la molienda representa el paso fundamental que caracteriza este innovador sistema de extracción. En
condiciones ambientales el dióxido de carbono sólido evoluciona directamente en la fase de aire (sublimación), y
el contacto directo entre el criógeno y las aceitunas induce una solidificación parcial del agua celular dentro de los
frutos. Dado que el volumen ocupado por el agua en el estado sólido es mayor que en el estado líquido, los cristales
de hielo formados son incompatibles con la estructura celular e inducen el colapso de las células, además de pro-
mover la difusión de las sustancias celulares en el aceite extraído, que así, por lo tanto, se enriquece con metabolitos
celulares que se caracterizan por un alto valor nutracéutico. Además, una capa de CO2, permanece sobre la pasta de
aceitunas para preservar de la degradación oxidativa. La adición de dióxido de carbono sólido a las aceitunas pro-
cesadas indujo un aumento estadísticamente significativo en el rendimiento de aceite y promueve la acumulación de
tocoferoles en la fase lipídica, mientras que produce un incremento, no significativo, de la fracción fenólica del aceite.
PALABRAS CLAVE: Aceite de oliva virgen extra; Calidad del producto; Criomaceración; Índice de extracción; Nieve
carbónica

Citation/Cómo citar este artículo: Zinnai A, Venturi F, Quartacci MF, Sanmartin C, Favati F, Andrich G. 2016. Solid
carbon dioxide to promote the extraction of extra-virgin olive oil. Grasas Aceites 67 (1): e121. doi: http://dx.doi.
org/10.3989/gya.0493151.

Copyright: © 2016 CSIC. This is an open-access article distributed under the terms of the Creative Commons
Attribution-Non Commercial (by-nc) Spain 3.0 Licence.
2 • A. Zinnai, F. Venturi, M.F. Quartacci, C. Sanmartin, F. Favati and G. Andrich

1. INTRODUCTION It is possible to adopt different growing practices

during olive production or to employ different work-
Only the oil extracted from olive fruits by a mechan- ing conditions during oil extraction to enhance the
ical process at reduced temperatures, characterized content of volatile and/or phenolic components and
by a low acidity and acceptable organoleptic proper- also to determine the quality of the oil. In fact, a
ties, can be named extra-virgin olive oil, a product sufficient irrigation of the grove, an early harvest of
largely used in the Mediterranean diet (Serra-Majem the fruits, a non-stressful olive pressing, a reduced
et al., 2003; Dag et al., 2011; Fregapane and Salvador, temperature and a more prolonged kneading phase
2013). The world consumption of this product is sig- should be adopted to produce an oil characterized
nificantly increasing in recent years and this positive by a high content of volatile components and by a
trend is related not only to its nutritional and health- more pronounced aroma as well (Dag et al., 2008).
promoting characteristics (Kiritsakis, 1998; Manna To increase the phenolic fraction and then to pro-
et  al., 1997; Monteleone et  al., 1998; Ranalli et  al., duce an oil characterized by greater bitterness and
1997; Visoli and Galli, 1998; Mili, 2006; Waterman stability, it is necessary to ensure a reduced irrigation
and Lockwood, 2007; Dag et al., 2011) but also to its of the olive trees, the utilization of less ripe fruits,
sensory properties. In fact, the right balance among the use of more efficient crushers able to break also
saturated, monounsaturated and polyunsaturated the pit and a kneading phase carried on at a higher
fatty acids, along with the significant presence of temperature for a reduced time. In fact, an increase
“minor” components such as aromatic phenolics and in temperature also determines an increase in the
tocopherolic compounds (Lazzez et  al., 2008; Dag constant related to the partition equilibrium of phe-
et al., 2011) largely justify its commercial success nols between oil and water, and promotes phenol
(Fregapane and Salador, 2013). solubility in the oil phase. A prolonged kneading
The chemical and organoleptic quality enjoyed by time could promote their enzymatic (phenoloxi-
extra-virgin olive oil is a function of several factors dase and peroxidase) and/or chemical oxidation,
such as the geographical location of the olive grove, an undesirable transformation which can be sen-
the chemical and microbiological composition of the sibly reduced by a strong reduction in the oxygen
soil, the evolution of the climatic conditions during level in the surrounding atmosphere (Lercker et al.,
fruit ripening, the extraction process adopted, etc. 2007). To succeed in obtaining the desired amounts
(García et al., 1996; Kiritsakis, 1998; Zamora et al., of volatile and phenolic compounds which greatly
2001; Rotondi et  al., 2004; Abaza et  al., 2005; Ben affect the quality of an extra-virgin olive oil, it is
Temime et al., 2006; Baccouri et al., 2007; Dag et al., necessary to effectively control the working vari-
2009). Thus, the application of improved working ables involved in the malaxation phase (time, tem-
conditions could potentially offer the real possibility perature and gas composition of the surrounding
to predict the concentration of phenolic and volatile atmosphere) (Reboredo-Rodriguez et  al., 2014).
components in this product and also to modulate its However, the whole process related to oil extraction
nutraceutical properties as well as the sensorial per- cannot be reduced only to a simple physical stage
ception produced. because complex bioprocesses can potentially take
The cultivar and the ripening degree reached place in all the phases of the extraction process, and
by the olive fruits greatly affect the quality of the several enzymes able to promote many chemical
extracted oil so that only healthy fruits showing the transformations, including those related to phenolic
suitable ripening degree should be processed (Uceda and volatile compounds, could become catalytically
and Frías, 1975; Uceda et  al., 1992; Monteleone active (Del Caro et al., 2006).
et al., 1995; Mincione et al., 1996; Mincione, 2007). This complex control of all the possible chemi-
As for all biological products, the olive character- cal transformations could only partially justify the
istics are strongly related not only to the cultivar utilization in the last twenty years of the same tra-
(genetic variability) but also to the growing tech- ditional procedure to promote the extraction of oil
niques and the climatic conditions occurring in the from olive fruits. In fact, only recently has the appli-
year. Some cultivars are characterized by high phe- cation of new technologies based on ultrasound
nolic contents (Coratina), while others show more and microwave been suggested (Clodoveo, 2013;
reduced concentrations (Taggiasca). The ripening Clodoveo and Hbaieb, 2013).
degree reached by the olive fruits belonging to the Although many experimental results which evi-
same cultivar and grown in the same grove also dence that the addition of a cryogen (i.e. solid car-
determines the characteristics of the oil (Dag et al., bon dioxide) to grapes promote the extraction of
2010). In particular, the green or turning color olives aromatic and phenolic substances from the fruit
give a product marked by bitter notes which can be skin and produce structured wines of rich fragrance
related to a higher presence of phenolic components have been already largely reported, this promising
(oleocanthal). These more acute and pungent notes technology (cryomaceration) has not yet been tested
are due to tyrosol and its derivatives, particularly in olive fruits, where similar results could be poten-
deacetoxy-ligstroside. tially obtained. In fact, similar to grapes, the olive

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151
 Solid carbon dioxide to promote the extraction of extra-virgin olive oil • 3

is a fruit, namely a substrate particularly rich in which determines the point of oil picking and then
water and therefore potentially sensible to cryomac- also the degree of its contamination by vegetable
eration. The direct contact between the cryogen and water.
the olives induces a partial solidification of the cel- At the end of the extraction process, the decanter
lular water and an increase in the volume occupied was washed with an amount of water measured by
by water. The consequent laceration of the cellular a flow-meter to ensure the total spillage of the oil
membranes (cellular crash) makes the diffusion of coming out of the olive fruits. Thus, it was possible
oil and other cellular compounds (phenols, aro- to verify the mass balance of oil by equation 1:
matic substances, etc.) in the liquid phase easier and
thus facilitates their extraction from cryomacerated Mfruits×OFfruits=Mpomace×OFpomace+MOE Eq. 1
fruits.
Moreover, the lower values of temperature where Mfruits=mass of fruits (kg); OFfruits=oil frac-
reached by adding carbonic snow and the relevant tion of the olive fruits (OFfruits=weight of oil/weight
amount of gas evolved because of the interaction of fruits); Mpomace=mass of pomace; OFpomace=oil
between solid CO2 (−78 °C at atmospheric pressure) fraction in the pomace (OFpomace=weight of oil/
and the olives/olive paste (initially at room tempera- weight of pomace); MOE=mass of oil (kg), and to
ture) induce the formation of a sort of protective determine the production yield (PY) by equation 2:
inert layer over the processed olives. This gaseous
layer would be able to prevent oxidation and then PY=MOE/(Mfruits×OFfruits) Eq. 2
reduce the loss of antioxidants like phenols (Di
Giovacchino et al., 2002) without compromising the The main characteristics of the fruits (carpo-
accumulation of some components responsible for logical properties, maturity index, chemical com-
flavor (Zinnai et al., 2007). position) and of the pomace were determined by a
Although the addition of a cryogen to promote solvent extractor (SER 148-3 Velp Scientifica) (oil
oil extraction could slightly increase the production content) and by the analytical procedures reported
cost, this innovative technology is able to produce in the literature (Uceda and Frias, 1975; Ryan and
a high quality oil characterized by a strong link Robards, 1998; Mincione, 2007).
with the olives used as well as their production area. About 30 kg of olive fruits belonging to different
Thus, the product would be easily identifiable by the cultivars (Leccino, Frantoio, Correggiolo, Itrana
modern consumer who is very health-conscious and and Coratina), hand-picked and accurately washed,
critically evaluates the nutritional and organoleptic were initially charged inside the hopper to be mixed
quality of food. with carbonic snow or not.
To obtain a significant comparison among data
2. MATERIALS AND METHODS on the oil extracted after adding or not solid CO2,
the processed olive fruits were previously accurately
The extraction was carried out using a micro oil mixed to ensure in both cases (with and without
mill (Oliomio Baby®, Toscana Enologica Mori) able addition of cryogen) a comparable feed.
to mill 20–30 kg of olives. This apparatus was suit- Table 1 reports the working condition adopted
ably modified to allow the direct addition of solid for a generic experimental run carried out on with
carbon dioxide (carbonic snow) to the olive fruits and without the addition of solid carbon dioxide
and/or to the paste coming from their milling. The during the crushing phase.
main process steps followed by this micro oil mill can The analytical characterization of the oils (acid-
be summarized as follows: olives, properly cleaned ity, peroxide number, spectrophotometer indexes,
and washed, were poured into the receiving hopper total phenols, carotenoids, tocopherols, chloro-
where a screw fed a crusher equipped with a hollow phyll) was carried out according to the conventional
knife impeller. The produced paste fell into a lower
mixer, where a helicoid shaped stirrer promoted its
malaxation. The temperature reached by the paste
was maintained in the desired range by a thermal reg- TABLE 1. Working parameters adopted for a generic
experimental run carried out on adding cryogen or not
ulation system (a temperature sensor put inside the
olive paste, connected with a heat exchanger). The Cryo Traditional
suitable flow of olive paste was then sent to a bipha- kg of added CO2·kg−1 of olives (%) 10.0÷20.0 0
sic decanter by a pump equipped with a speed change
gear. The decanter (4200 rpm) favored the separation Olive temperature (°C) 11.5÷2.1 10.9
of oil from the olive pomace mixed with water (veg- Paste temperature (°C) 23.9 24.0
etable water plus the water potentially added through Malaxation time (s) 2400 2400
a flow-meter in order to induce a more efficient sepa- Extraction time (s) 4900 4300
ration). The separation efficiency of the decanter can kg of added water·kg−1 of olives (%) 8.5 9.3
be modulated by a suitable regulation of the nozzles,

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151
4 • A. Zinnai, F. Venturi, M.F. Quartacci, C. Sanmartin, F. Favati and G. Andrich

analytical procedures reported in the literature the sampling date increased. This could explain the
(Gutiérrez Rosales et al., 1992; Pokorny et al., 1995; experimental values assumed by the maturity indexes
Capella et al., 1997; Tateo and Bononi, 2004). and the moisture percentages of the other runs of
The experimental data were evaluated according group A. Concerning the samples constituted by
to the statistical methods reported in the literature fruits belonging to the same cultivar (compare runs
(Snedecor and Cochran, 1979; Ryan and Robards, D1, D2 and D3), these apparent contradictions do
1998). not seem to occur. In fact, to the increasing values
assumed by the maturity index corresponds to the
3. RESULTS AND DISCUSSION expected decrease in water content and a corre-
sponding increase in oil percentage.
Table 2 reports some characteristics of olives used The olives belonging to the late cultivar Itrana
to produce extra-virgin oil by adding carbonic snow were collected and milled during the spring of the
or not. Depending on the cultivar used, it is pos- following year, but were less intact and healthy than
sible to identify five different groups of fruits. While those of the other cultivars collected the previous
group A represents a mixture of olives belonging autumn.
to two different cultivars (Frantoio and Leccino), To ensure the experimental utilization of compa-
the others are formed by fruits belonging to only rable batches of olives, samples of fruits collected at
one cultivar. Although the olives of groups C and the same date and in the same orchard were accu-
D belong to the same cultivar (Coratina) they were rately mixed to obtain a homogeneous distribu-
produced in two different Italian regions (Toscana tion of fruits before dividing them in two or more
and Basilicata, respectively). Moreover, the fruits equivalent portions. These comparable groups of
belonging to the same group but picked up at differ- olives were milled adopting the same working condi-
ent ripening stages (five sampling dates for group A tions, but adding carbonic snow or not to determine
and three for D) are characterized by chemical com- the effect induced by the addition of the cryogen.
positions which are statistically different as shown Table 3 reports the experimental data related to the
in Table 2. extraction yields (weight of oil/weight of fruits)
The experimental data related to the macro- obtained by adding (Y+CO2,s) or not (Y-CO2,s)
composition of the olives were affected by a high carbonic snow. Moreover, to highlight the possible
variability as a function of the sampling date (com- effect induced by the cryogen addition, the values of
pare data related to the five runs of group A and the percentage difference (ΔY%) occurring between
the three of group D), particularly those of group the two yields obtained by adding carbonic snow or
A, in which the fruits belong to two different culti- not are also reported in Table 3.
vars (Leccino and Frantoio) and reach their ripen- Although in two cases (A3 and C1) the amount of
ing stages at different dates. Thus, the first run of oil extracted after adding carbonic snow was lower
group A (A1) was constituted by a relevant number than that obtained without the addition, only posi-
of olives belonging to the cultivar Leccino which tive values of ΔY% were obtained in all the other
reaches its ripeness before the other (Frantoio), experimental determinations, and in some cases the
while the fraction of these olives decreased when difference occurring between the extraction yields

TABLE 2. Ripening date, maturity index and macro-compositions (mean values of percentages
related to oil, moisture and defatted dry residue together with the related confidence interval;
p=0.05) of the groups of olive fruits used in the experimental determinations

Ripening Maturity Moisture Oil content Defatted dry

Olive group date Run Index (0÷7) Cultivars % % residue %
A 08/11/2010 A1 3.0 Frantoio, 57.1±0.7 14.2±0.7 28.7±1.4
11/11/2010 A2 3.8 Leccino 52.1±0.2 16.3±0.4 31.6±0.6
04/12/2010 A3 3.6 53.5±0.6 19.0±0.3 27.4±0.9
07/12/2010 A4 4.0 52.9±0.4 20.3±0.5 26.8±0.9
09/12/2010 A5 5.1 48.2±0.3 24.2±0.4 27.6±0.7
B 12/11/2010 B1 1.0 Frantoio 50.3±0.6 18.4±1.3 31.2±1.9
C 18/11/2010 C1 2.3 Coratina 49.2±0.6 15.9±0.7 34.8±1.3
D 15/11/2010 D1 2.6 Coratina 61.4±0.3 18.3±0.2 22.2±0.3
29/11/2010 D2 2.8 55.8±0.9 19.4±0.2 27.7±1.1
13/12/2010 D3 4.2 53.2±0.1 23.5±0.1 23.2±0.2
E 16/04/2011 E 4.6 Itrana 59.5±0.4 16.2±0.2 24.3±0.6

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151
 Solid carbon dioxide to promote the extraction of extra-virgin olive oil • 5

TABLE 3. Extraction yields (kg of oil·kg−1 of olives, %) With the exception of oils from olives belonging
obtained adopting the same experimental conditions but to this particular cultivar (Ea, Eb and Ec) which,
adding (Y+CO2,s) or not (Y-CO2,s) carbonic snow to the same
portion of fruits together with the value assumed by the as previously reported, were less intact and healthy
difference percentage [ΔY%=(Y+CO2,s–Y-CO2,s)×100/Y-CO2,s] than the other fruits, all the oils were character-
ized by better analytical values, not only of those
ΔY%=(Y+CO2,s– expected for an extra-virgin olive oil but also for a
Run Y+CO2(s) Y-CO2(s) Y-CO2,s)×100/Y-CO2,s PDO product (Table 4). Moreover, no significant
A1,1* 12.2 11.0 10.9 differences occurred among the values determined
A1,2* 13.0 11.8 10.2 for these qualitative indexes as a function of the
A2 16.3 13.2 23.5 extraction technology adopted or of the possible
A3 15.5 15.7 −1.3
addition of carbonic snow. In any case, the micro
oil mill used was able to produce high quality
A4 16.0 14.2 12.7 extra-virgin olive oil without any cryogen addition,
A5 15.2 14.6 4.1 so any possible quality improvement due to the use
B1 15.8 15.2 4.0 of carbonic snow would assume a more significant
C1 12.0 12.8 −6.2 meaning.
D1 14.8 14.5 2.2
Among all the quality indexes determined in the
oils, only those related to the contents of tocopherols
D2 16.4 15.0 9.3
and total phenols varied significantly following the
D3 17.3 16.2 6.8 addition of carbonic snow to the olive fruits. Table 5
Ea** 8.8 8.5 3.5 reports the amounts of tocopherols and total phe-
Eb** 9.2 8.5 8.2 nols co-extracted with oil adopting the same experi-
Ec** 12.6 8.5 48.2 mental conditions but adding (Toc/Phtot+CO2,s)
or not (Toc/Phtot-CO2,s) carbonic snow to olives
belonging to the same homogeneous group of fruits.
While for tocopherols (vitamin E), with the only
evaluated after adding carbonic snow or not was par- exception of the D1 sample, a clear increase in the
ticularly significant (23.5% for A2 and 48.2% for Ec). extraction yield was always obtained; the effect
Moreover, if increasing amounts of carbonic induced by cryogen on the extraction of total phe-
snow (0, 3, 4.5 and 6  kg) were added to the same nols was affected by a higher variability. In fact, for
amount (30 kg) of olives (E group) an increase in oil many samples an increase in total phenol concen-
yield was always obtained. trations (A1.1, A2, A3, A5, B1, C1, Ea, Eb and Ec)
The mean value and the related confidence was observed when cryogen was used, but in some
interval of the percentage differences between cases (A1.2, A4, D1, D2, D3) the opposite situation
extraction yields obtained after adding cryogen occurred. To evaluate the effect induced by cryogen
to the olive fruits or not were also determined addition on the co-extraction of tocopherols and
(ΔY%  mean±c.i.=9.3±1.9%; p=0.05). To ensure a total phenols with oil, the same statistical procedure
normal distribution of the experimental data, the adopted to analyze the experimental data concern-
statistical analysis was carried out with the cor- ing the oil yield was used. A positive mean value
responding values of arcsin √ΔY% (Snedecor and of percentage difference was obtained for both
Cochran, 1979). The addition of carbonic snow tocopherols (ΔToc%mean=6.0) and total phenols
to the olive fruits produced an increase in oil of (ΔPhtot%mean=5.1), but while for tocopherols a
about 9%. Thus, if without any cryogen addition confidence interval (c.i.=0.6; p=0.05) lower than
16 kg of extra-virgin olive oil can be extracted from the mean value was obtained, for total phenols
100  kg of olive fruits, from the same amount of the confidence interval (c.i.=10.6; p=0.05) resulted
the same fruits more than 17 kg of product can be higher than the corresponding mean value. Thus,
obtained following carbonic snow addition. while the addition of carbonic snow determined
With regards to the qualitative parameters of a significant increase in tocopherol content in the
the oil extracted after adding carbonic snow or oil, the increase in total phenols did not prove to
not, Table 4 reports the values for acidity, peroxide be statistically significant. As reported in the litera-
number and the spectrophotometric indexes (K232, ture (Clodoveo, 2013; Clodoveo and Hbaieb, 2013;
K270 and ΔK), together with their related confidence Reboredo-Rodriguez et al., 2014), phenol extraction
interval (p=0.05). Being that the values calculated is not promoted by working at reduced temperatures
for ΔK (K268–(K262+K274)/2) are lower than 0.01 because the constant related to the partition equi-
for almost all the oils, these data are not reported in librium of phenols between oil and water decreases
Table 4. Only two samples of oil extracted from olive substantially. On the contrary, the direct contact of
fruits belonging to the Itrana cultivar which showed carbonic snow with the olive fruits induces a par-
values greater than or comparable to this limit (ΔKEa tial solidification of fruit water and then the lac-
(ΔK=0.01±0.05 and ΔKEc=−0.08±0.15) were found. eration of the cellular membranes (cellular crash)

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151
6 • A. Zinnai, F. Venturi, M.F. Quartacci, C. Sanmartin, F. Favati and G. Andrich

TABLE 4. The values determined for total acidity, number of peroxides and spectrophotometric indexes (mean ±c.i.;
p=0.05) in the oils extracted from different samples of olive fruits tested, adding (+) or not (-) carbonic snow

oil acidity (g of oleic Number of peroxides

acid·100 g−1 of oil) (mEq O2·kg−1 oil) K232 K270
Run + − + − + − + −
A1,1* 0.36±0.01 0.63±0.03 4.82±0.58 6.53±0.01 1.37±0.15 1.44±0.22 0.07±0.01 0.10±0.03
A1,2* 0.67±0.01 0.75±0.03 7.86±0.52 6.52±0.34 1.26±0.05 1.27±0.07 0.10±0.01 0.12±0.03
A2 0.19±0.01 0.18±0.01 7.45±0.10 7.44±0.30 1.64±0.10 1.57±0.10 0.15±0.02 0.13±0.01
A3 0.26±0.01 0.25±0.01 9.36±0.42 7.51±0.74 1.33±0.02 1.91±0.17 0.11±0.02 0.14±0.01
A4 0.55±0.01 0.54±0.01 7.14±0.38 6.80±0.57 1.78±0.12 1.04±0.14 0.17±0.03 0.11±0.01
A5 0.68±0.06 0.61±0.03 8.19±0.30 8.22±0.50 1.60±0.15 1.62±0.12 0.16±0.02 0.14±0.01
B1 0.32±0.01 0.29±0.02 6.75±0.18 6.26±0.63 1.45±0.13 1.52±0.14 0.09±0.02 0.09±0.03
C1 0.40±0.02 0.26±0.03 5.91±0.68 7.08±0.47 1.43±0.09 1.61±0.08 0.11±0.02 0.12±0.01
D1 0.26±0.01 0.21±0.02 9.25±0.70 9.06±0.10 1.54±0.14 1.69±0.13 0.13±0.04 0.15±0.01
D2 0.21±0.04 0.22±0.02 8.90±1.22 9.13±0.58 1.59±0.09 1.66±0.23 0.13±0.04 0.15±0.03
D3 0.29±0.01 0.32±0.01 5.29±0.12 6.52±0.68 1.34±0.07 1.60±0.14 0.10±0.02 0.12±0.01
Ea** 0.53±0.01 0.37±0.01 11.08±1.74 10.93±0.64 2.52±0.10 2.46±0.25 0.04±0.01 0.14±0.02
Eb** 1.64±0.02 = 10.55±1.50 = 2.00±0.02 = 0.07±0.01 =
Ec** 1.65±0.01 = 10.17±2.41 = 1.67±0.15 = 0.12±0.02 =
° <0.80 <20.00 <2.50 <0.22
°° ≤0.50 <12.00 <2.20 <0.20

*A1,1 and A1,2 represent two replications of oil extraction carried out adding carbonic snow or not to similar samples of
olive fruits (A1).
**experimental runs carried out adding different amounts (0; 3 ; 4.5; 6 kg) of carbonic snow to 30 kg of olive fruits belonging
to the same lot of olives (E=Itrana cultivar).
=experimental value equal to that reported for Ea.
°Values established for an extra virgin olive oil (Reg 2568/91); °°Values requested by the disciplinary of a PDO extra virgin
olive oil.

TABLE 5. Amounts of tocopherols and total phenols co-extracted with oil adopting the same experimental conditions
but adding (Toc/Phtot+CO2,s) or not (Toc/Phtot-CO2,s) carbonic snow to the same lot of fruits together with the
value of the difference in percentage [ΔToc/Phtot%=(Toc/Phtot+CO2,s−Toc/Phtot−CO2,s)×100/Toc/Phtot−CO2,s]

Run Toc+CO2,s (ppm) Toc-CO2,s (ppm) ΔToc% Phtot+CO2,s (ppm) Phtot-CO2,s (ppm) ΔPhtot%
A1.1* 206 198 4.0 341 335 1.8
A1.2* 220 198 11.1 227 251 −9.6
A2 193 181 6.6 502 450 11.6
A3 197 189 4.2 248 238 4.2
A4 179 174 2.9 186 251 −25.9
A5 184 170 8.2 556 501 11.0
B1 135 125 8.0 537 428 25.5
C1 255 221 15.4 557 498 11.8
D1 294 294 0.0 443 466 −4.9
D2 261 253 3.2 445 553 −19.5
D3 249 248 0.4 333 475 −29.9
Ea** 294 257 14.4 44 36 22.2
Eb** 291 257 13.2 46 36 27.8
Ec** 286 257 11.3 54 36 50.0

*A1,1 and A1,2 represent two replications of oil extraction carried out adding carbonic snow or not to similar samples of
olive fruits (A1).
**experimental runs carried out adding different amounts (0; 3 ; 4.5; 6 kg) of carbonic snow to 30 kg of olive fruits belonging
to the same lot of olives (E=Itrana cultivar).

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151
 Solid carbon dioxide to promote the extraction of extra-virgin olive oil • 7

which makes the diffusion of all the compounds “Souri” olive orchard converted to irrigation. J. Sci. Food
easier and therefore of phenols in the liquid phase Agric. 88, 1524–1528. http://dx.doi.org/10.1002/jsfa.3243.
Dag A, Bustan A, Avni A, Zipori I, Lavee S, Riov J. 2010.
as well. These considerations could partially justify Timing of fruit removal affects concurrent vegetative
the results obtained: the cellular crash could have growth and subsequent return bloom and yield in olive
increased phenol extraction (DPhtot%mean>0), (Olea europaea L.). Sci. Hort. 123, 469–472. http://dx.doi.
org/10.1016/j.scienta.2009.11.014.
while the reduced temperature would have induced Dag A, Keremb Z, Yogevb N, Zipori I, Laveec S, Ben-Davida E.
phenol accumulation in the water phase; the rate 2011. Influence of time of harvest and maturity index on
of diffusion could be due to the amount of liquid olive oil yield and quality. Sci. Hort. 127, 358–366. http://
dx.doi.org/10.1016/j.scienta.2010.11.008.
water actually present during the phases of malaxa- Del Caro A, Vacca V, Poiana M, Fenu P, Piga A. 2006. Influence
tion and centrifugation. Moreover to a preliminary of technology. storage and exposure on components of
and a non-professional sensory evaluation, the oil extra virgin olive oil (Bosana cv) from whole and de-stoned
appeared to have a richer aroma and a more com- fruits. Food Chem. 98, 311–316. http://dx.doi.org/10.1016/j.
foodchem.2005.05.075.
plex and varietal taste. Di Giovacchino L, Sestili S, Di Vincenzo D. 2002. Influence of
olive processing on virgin olive oil quality. Eur. J. Lipid
4. CONCLUSIONS Sci. Tech. 104, 587–601.
Fregapane G, Salvador MD. 2013. Production of superior qual-
ity extra virgin olive oil modulating the content and profile
According to these preliminary results, the addi- of its minor components. Food Res. Int. 54, 1907–1914.
tion of carbonic snow to olive fruits constitutes a http://dx.doi.org/10.1016/j.foodres.2013.04.022.
promising new technology which allows to signifi- García JM, Seller S, Pérez-Camino MC. 1996. Influence of
fruit ripening on olive oil quality. J. Agric. Food Chem. 44.
cantly increase the yield of oil extracted from olive 3516–3520. http://dx.doi.org/10.1021/jf950585u.
fruits to produce a high quality product (extra-­virgin Gutiérrez-Rosales F, Perdiguero S, Gutiérrez R, Olías JM. 1992.
olive oil) characterized by an increased content Evaluation of bitter taste in virgin olive oil. J. Am. Oil Chem.
Soc. 69, 394–395. http://dx.doi.org/10.1007/BF02636076.
of vitamin E (tocopherols) and also of phenols. Kiritsakis AK. 1998. Olive Oil Handbook. AOCS Press. Champaign.
Further research must be carried on in the near IL. ISBN: 978-1-4614-7776-1 (Print) 978-1-4614-7777-8
future to evaluate the real potential of this innova- (Online); http://dx.doi.org/10.1007/978-1-4614-7777-8.
tive technology. In particular, this research will be Lazzez A, Perri E, Caravita MA, Khlif M, Cossentini M.
2008. Influence of olive maturity stage and geographical
aimed to determine the best ratio between cryogen origin on some minor components in virgin olive oil of
and olive fruits as well as the optimum values of the Chemlali Variety. J. Agric. Food Chem. 56, 982–988.
time and temperature to employ during the extrac- http://dx.doi.org/10.1021/jf0722147.
Lercker G, Bendini A, Cerretani L. 2007. Qualità, composizione
tion process, particularly during malaxation, the e tecnologia di produzione degli oli vergini di oliva. Prog.
phase which greatly affects the qualitative charac- Nutr. 9, 134–148.
teristics of the extra-virgin olive oil produced. Manna C, Galletti P, Cucciolla V, Moltedo O, Leone A, Zappia
V. 1997. The protective effect of the olive oil polyphenol
(3.4-dihydroxyphenyl)-ethanol counteracts reactive oxygen
REFERENCES metabolite-induced cytotoxicity in Caco-2 cells. J. Nutr.
127, 286–292. PMID: 9039829.
Abaza L, Taamalli W, Ben Temime S, Daoud D, Gutiérrez F, Mili S. 2006. Olive oil marketing in non-traditional markets:
Zarrouk M. 2005. Natural antioxidant composition as cor- prospects and strategies. New Medit. 5, 27–37.
related to stability of some Tunisian virgin olive oils. Riv. Mincione A, Runcio A, Valenzise R, Sorgonà L, Santacaterina
Ital. Sost. Grasse, 82, 12–18. S, Poiana M. 2007. Ricerche sull’olio di oliva monovari-
Baccouri B, Ben Temime S, Taamalli W, Daoud D, M’sallem M, etali. Nota XVI. Contributo alla caratterizzazione dell’olio
Zarrouk M. 2007. Analytical characteristics of virgin olive estratto dalle olive della cv Frantoio coltivata in due diversi
oils from two new varieties obtained by controlled crossing areali della Provincia di Reggio Calabria. Riv. Ital. Sost.
on Meski variety. J. Food Lipids, 14, 19–34. http://dx.doi. Grasse, 84, 133–156.
org/10.1111/j.1745-4522.2006.00067.x. Mincione B, Poiana M, Giuffrè AM, Modafferi V, Giuffrè F. 1996.
Ben Temime S, Taamalli W, Bacccouri B, Abaza L, Daoud D, Ricerche sugli oli monovarietali. Nota II. Caratterizzazione
Zarrouk M. 2006. Changes in olive oil quality of Chétoui dell’olio di Peranzana. Riv. Ital. Sost. Grasse. 73, 245–257.
variety according to origin of plantation. J. Food Lipids, 13, Mincione B. 2007. L’olio di Oliva vergine della provincia di Reggio
88–99. http://dx.doi.org/10.1111/j.1745-4522.2006.00036.x. Calabria prodotto da cultivar non autoctone - Caratteristiche
Capella P, Fedeli E, Bonaga G, Lercker G. 1997. Il Manuale degli Merceologiche e Qualitative. UNIRC (CEN.S.A.) Laruffa
Oli e dei Grassi. Ed. Tecniche nuove, Milano. ISBN 88 7081 Editore. 105–129. ISBN 978 88 7221 319 3.
979 5. Monteleone E, Caporale G, Carlucci A, Pagliarini E. 1998.
Clodoveo ML. 2013. New advances in the development of inno- Optimisation of extra virgin olive oil quality. J. Sci. Food
vative virgin olive oil extraction plants: Looking back to Agric. 77, 31–37.
see the future. Food Res. Int. 54, 726–729. http://dx.doi. Monteleone E, Caporale G, Lencioni L, Favati F, Bertuccioli M.
org/10.1016/j.foodres.2013.08.020. 1995. Optimization of virgin olive oil quality in relation
Clodoveo ML, Hbaieb RH. 2013. Beyond the traditional vir- to fruit ripening and storage. Dev. Food Sci. 37, 397–418.
gin olive oil extraction systems: Searching innovative and http://dx.doi.org/10.1016/S0167-4501(06)80168-8.
sustainable plant engineering solutions. Food Res. Int. 54, Pokorny J, Kalinova L, Dysseler P. 1995. Determination of chlo-
1926–1933. http://dx.doi.org/10.1016/j.foodres.2013.06.014. rophyll pigments in crude vegetable oil. Pure Appl Chem. 67,
Dag A, Ben-David EA, Fiume P, Perri E. 2009. Cultivation: 1781–1787. http://dx.doi.org/10.1351/pac199567101781.
Problems and Perspectives. Sixth Framework Programme Ranalli A, De Mattia G, Ferrante ML. 1997. Comparative
Priority 5; Food Quality and Safety Priority. Call 4-C; evaluation of the olive oil given by a new processing
MAC-oils. The Scientific Handbook. Avellino. Italy. 1–55. ­system. Int. J. Food Sci. Technol. 32, 289–297. http://dx.doi.
Dag A, Ben-Gal A, Yermiyahu U, Basheer L, Nir Y, Kerem Z. org/10.1046/j.1365-2621.1997.00116.x.
2008. The effect of irrigation level and harvest mechani- Reboredo-Rodrìguez P, Gonzàlez-Barreiro C, Cancho-Grande
zation on virgin olive oil quality in a traditional rain-fed B, Simal-Gàndara J. 2014. Improvements in the malaxation

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151
8 • A. Zinnai, F. Venturi, M.F. Quartacci, C. Sanmartin, F. Favati and G. Andrich

process to enhance the aroma quality of extra virgin olive Uceda M, Frías L. 1975b. Harvest dates. Evolution of the fruit oil
oils. Food Chem. 158, 534–545. http://dx.doi.org/10.1016/j. content. oil composition and oil quality. En: Proceedings del
foodchem.2014.02.140. Segundo Seminario Oleicola Internacional. COI. Cordoba.
Rotondi A, Bendini A, Cerretani L, Mari M, Lercker G, Toschi 125–128.
TG. 2004. Effect of olive ripening degree on the oxidative Uceda M, Frías ML, Ruano MT. 1992. Diferenciación de varie-
stability and organoleptic properties of cv. Nostrana di dades de aceituna por la composición ácidica de su aceite.
Brisighella extra virgin olive oil. J. Agric. Food Chem. 52, In: Abstracts of the First International Symposium on
3649–3654. http://dx.doi.org/10.1021/jf049845a. Olive Growing. Córdoba. Spain. pp. 35–38.
Ryan D, Robards K. 1998. Phenolic compounds in olives. Visoli F, Galli C. 1998. Olive oil phenols and their potential effects
Analyst, 123, 31–44. http://dx.doi.org/10.1039/a708920a. on human health. J. Agric. Food Chem. 46. 4292–4296. http://
Serra-Majem L, Ngo de la Cruz J, Ribas L, Tur JA. 2003. Olive oil dx.doi.org/10.1021/jf980049c.
and the Mediterranean diet: beyond the ­rhetoric. Eur. J. Clin. Waterman E, Lockwood B. 2007. Active components and clini-
Nutr. 57, 2–7. http://dx.doi.org/10.1038/sj.ejcn.1601801. cal applications of olive oil. Altern Med Rev. 12, 331–342.
Snedecor GW, Cochran WG. 1979. Statistical methods. 6th PMID:18069902.
Ed. The Iowa State University Press. Ames. Iowa. USA. Zamora R, Alaiz M, Hidalgo FJ. 2001. Influence of cultivar
Tateo F, Bononi M. 2004. Guida all’Analisi Chimica degli Alimenti and fruit ripening on olive (Olea europaea) fruit protein
vol. 3. Oli e Grassi Vegetali ARS Edizioni Informatiche, content. Composition and antioxidant activity. J. Agric.
Milano. ISBN: 9788889260067. Food Chem. 49, 4267–4270. http://dx.doi.org/10.1021/
Uceda M, Frías L. 1975a. Épocas de recolección. Evolución del jf0104634.
contenido graso del fruto y de la composición y calidad Zinnai A, Venturi F, Calamita Y, Andrich G. 2007. La cri-
del aceite. in: IOOC (Ed.). Proceedings of II Seminario omacerazione prefermentativa nella produzione di vini
Oleícola International. Córdoba. Spain. di qualità. Ind. Bev. 209, 227–232.

Grasas Aceites 67 (1), January–March 2016, e121. ISSN-L: 0017–3495 doi: http://dx.doi.org/10.3989/gya.0493151