Eur Food Res Technol (2006) 222: 521–526 DOI 10.1007/s00217-005-0003-4 ORIGINAL PAPER A. Parenti · P. Spugnoli · P. Masella · L. Calamai Carbon dioxide emission from olive oil pastes during the transformation process: technological spin offs Received: 11 February 2005 / Revised: 18 April 2005 / Published online: 1 December 2005 C Springer-Verlag 2005 Abstract The emission of carbon dioxide (CO 2 ) from olive paste during malaxation was investigated in a lab experi- ment using a hermetically sealed malaxation chamber. A rapid increase in the concentration of CO 2 during malaxa- tion was observed, with an average increase of 32 ml/(l min) for the initial 5 min. Then, the emission progressively de- creased to a mean rate of 1.1 ml/(l min). This was probably the result of an initial acceleration in respiration followed by the gradual onset of fermentation processes as ambient oxygen was depleted. After malaxation, small amounts of cellular fermentation products (e.g. ethanol and lactic acid) were detected in the wastewater. In order to examine this phenomenon of the inhibition of oxidation due to evolved CO 2 , malaxation experiments were conducted in both a sealed and an open-air mixing apparatus. The differences in chlorophyll concentration of the resulting oils were then measured. Large amounts of chlorophyll, about twice as much, were found in the oil produced under sealed con- ditions. This increase in the concentration of chlorophyll resulted from the limited oxidation of the sample by atmo- spheric oxygen due to the protection of the evolved CO 2 . Keywords Olive oil . Oxidation . Malaxation . CO 2 emission . Respiration . Extraction Introduction In recent years the production of olive oil has been geared toward higher quality products, especially in the traditional production areas with the introduction of protected geo- A. Parenti () · P. Spugnoli · P. Masella Dipartimento di Ingegneria Agraria e Forestale, Universit` a degli Studi di Firenze, Piazzale Cascine 15, 50144 Firenze, Italy e-mail: alessandro.parenti@unifi.it L. Calamai Dipartimento di Scienza del Suolo e Nutrizione della Pianta, Universit` a degli Studi di Firenze, Piazzale Cascine 28, 50144 Firenze, Italy graphic indication (I.G.P.) and denomination of protected origin (D.O.P.) [1–4]. To this end, specific production pro- tocols that maximize the quality of the final product were introduced. In particular, great attention is devoted to the minimization of oxidation processes during extraction be- cause they decrease oil quality [5–7]. In this respect, sev- eral studies have compared the effect of different extraction phases [8–10], olive crushers [11, 12] and malaxation times and temperatures [13–17]. Other solutions, such as the use of vertical mixers and covering the paste with inert gases (mainly with nitrogen) have also been tested with the aim of minimizing the contact of olive pastes with atmospheric oxygen [18–21]. In a series of experiments conducted in the Antinori olive milling plant at the Tignanello Estate, the relative concentration of oxygen and CO 2 was moni- tored in the gas layer above the surface of the pastes in the malaxation chamber. Because a rapid increase in the CO 2 concentration was recorded on top of the paste, further ex- periments were set up to investigate this phenomenon in detail, in a sealed environment using lab equipment that was better-suited to the measurement of small changes in CO 2 concentration. This phenomenon could potentially be used to naturally protect olive pastes from oxidation during the transformation process in order to enhance oil quality. The results of the experiment conducted on the natu- ral evolution of CO 2 from olive pastes during malaxation compared to the CO 2 produced by intact olive drupes dur- ing post-harvest respiration are presented here. In addition, the change in concentration of an oxygen-sensitive com- ponent, total chlorophyll, as a function of the malaxation methodology was monitored. Materials and methods In this experiment seven malaxation trials were conducted using two different olive cultivars, six with cv. Coratina (A, B, C, D, E, F) and one with cv. Leccino (G). In 4 out of the 6 malaxation trials with cv. Coratina and in the one done with cv. Leccino, CO 2 evolution was monitored in a hermetically sealed chamber equipped with a rubber set
Transcript
Eur Food Res Technol (2006) 222: 521–526DOI 10.1007/s00217-005-0003-4
ORIGINAL PAPER
A. Parenti · P. Spugnoli · P. Masella · L. Calamai
Abstract The emission of carbon dioxide (CO2) from olivepaste during malaxation was investigated in a lab experi-ment using a hermetically sealed malaxation chamber. Arapid increase in the concentration of CO2 during malaxa-tion was observed, with an average increase of 32 ml/(l min)for the initial 5 min. Then, the emission progressively de-creased to a mean rate of 1.1 ml/(l min). This was probablythe result of an initial acceleration in respiration followedby the gradual onset of fermentation processes as ambientoxygen was depleted. After malaxation, small amounts ofcellular fermentation products (e.g. ethanol and lactic acid)were detected in the wastewater. In order to examine thisphenomenon of the inhibition of oxidation due to evolvedCO2, malaxation experiments were conducted in both asealed and an open-air mixing apparatus. The differencesin chlorophyll concentration of the resulting oils were thenmeasured. Large amounts of chlorophyll, about twice asmuch, were found in the oil produced under sealed con-ditions. This increase in the concentration of chlorophyllresulted from the limited oxidation of the sample by atmo-spheric oxygen due to the protection of the evolved CO2.
In recent years the production of olive oil has been gearedtoward higher quality products, especially in the traditionalproduction areas with the introduction of protected geo-
A. Parenti (�) · P. Spugnoli · P. MasellaDipartimento di Ingegneria Agraria e Forestale, Universita degliStudi di Firenze,Piazzale Cascine 15,50144 Firenze, Italye-mail: [email protected]
L. CalamaiDipartimento di Scienza del Suolo e Nutrizione della Pianta,Universita degli Studi di Firenze,Piazzale Cascine 28,50144 Firenze, Italy
graphic indication (I.G.P.) and denomination of protectedorigin (D.O.P.) [1–4]. To this end, specific production pro-tocols that maximize the quality of the final product wereintroduced. In particular, great attention is devoted to theminimization of oxidation processes during extraction be-cause they decrease oil quality [5–7]. In this respect, sev-eral studies have compared the effect of different extractionphases [8–10], olive crushers [11, 12] and malaxation timesand temperatures [13–17]. Other solutions, such as the useof vertical mixers and covering the paste with inert gases(mainly with nitrogen) have also been tested with the aimof minimizing the contact of olive pastes with atmosphericoxygen [18–21]. In a series of experiments conducted inthe Antinori olive milling plant at the Tignanello Estate,the relative concentration of oxygen and CO2 was moni-tored in the gas layer above the surface of the pastes in themalaxation chamber. Because a rapid increase in the CO2concentration was recorded on top of the paste, further ex-periments were set up to investigate this phenomenon indetail, in a sealed environment using lab equipment thatwas better-suited to the measurement of small changes inCO2 concentration. This phenomenon could potentially beused to naturally protect olive pastes from oxidation duringthe transformation process in order to enhance oil quality.
The results of the experiment conducted on the natu-ral evolution of CO2 from olive pastes during malaxationcompared to the CO2 produced by intact olive drupes dur-ing post-harvest respiration are presented here. In addition,the change in concentration of an oxygen-sensitive com-ponent, total chlorophyll, as a function of the malaxationmethodology was monitored.
Materials and methods
In this experiment seven malaxation trials were conductedusing two different olive cultivars, six with cv. Coratina (A,B, C, D, E, F) and one with cv. Leccino (G). In 4 out of the6 malaxation trials with cv. Coratina and in the one donewith cv. Leccino, CO2 evolution was monitored in ahermetically sealed chamber equipped with a rubber set
A Coratina paste sealed on 4.19 1.33 110B Coratina paste sealed on 4.17 1.35 78C Coratina paste sealed on 3.89 1.61 65D Coratina paste sealed on 4.19 1.33 190E Coratina paste open on 3.75 – 195F Coratina paste open on 4.10 – 120G Leccino paste sealed on 2.99 2.42 140H Coratina olives sealed off 2.47 2.90 160I Leccino olives sealed off 2.42 2.94 195
for air sampling by means of a graduated syringe. Theremaining two cv. Coratina olive samples were malaxatedin the same mixer, but it remained open to the air in orderto allow the free exchange of oxygen. These two wereused as controls.
In order to evaluate the natural respiration of whole olivedrupes that have been processed for storage, the CO2 emis-sion of the olives was monitored for about 200 min in trialsH and I in the same container without mixing. The completeexperimental design is summarized in Table 1.
The olives cv. Coratina, from the region of Apulia insouthern Italy, were harvested in early February at amedium-ripeness and were in good sanitary condition.The olives cv. Leccino, harvested in the same period inTuscany, near Florence, were at late-ripeness and in excel-lent sanitary condition.
The olives were processed into paste with a lab extru-sion mill. The consistency and granulometry of the pastesobtained were comparable to those commonly produced incommercial olive milling plants.
The olive pastes were then malaxated in a hermetically-sealed lab mixer specially designed for this experiment.
The mixer (Fig. 1) consisted of:
– A vertically-positioned cylindrical polycarbonate malax-ation chamber with a height of 300 mm and an innerdiameter of 140 mm,
– A teflon coated base equipped with a thermocouple forcontinuous temperature monitoring,
– A teflon coated cover equipped with a cone for a rubberstopper used for gas sampling and a threaded hole for amanometer,
– A vertically-rotating iron shaft with a series of 4 bendedpalettes for effective malaxation of the olive paste duringmixing,
– A 12 V DC electric motor to rotate the shaft,– And an analog manometer (full-scale = 2 bar).
The total volume of the malaxation chamber was5143 cm3. Aliquots of about 4 kg of either olives or olivepaste were used for the malaxation trials with a solid tohead-space volume ratio of about 4:1. This proportion re-constructs, on lab-scale, the volume ratio commonly usedin commercial malaxation mixers.
In addition to the measurements of CO2 concentrationtaken during the trials, samples of olive paste were taken at
Fig. 1 Mixing apparatus for olive paste malaxation
the beginning and at the end of each trail in order to evaluatethe quality of the olive oil resulting from malaxation.
For that purpose, the paste samples were centrifuged ina Beckman JA21 lab centrifuge at 10000×g in 250 mlpolypropylene bottles. The resulting oil and wastewaterwere separated and stored at −20 ◦C in 15 ml plastic falconsfor chemical analyses.
The determination of CO2 concentration was done byFT-IR method as described by Calamai et al. [22] usinga gas cell and a calibrated gas mixture containing CO2 at10 ml/l. An extensive purging of the gas cell and immediatebackground correction of the instrument was done beforeeach measurement. A calibration line in the range 0.35–15 ml/l was prepared by injecting different volumes ofthe standard gas mixture into the cell. As a preliminary run
523
showed a large range of CO2 concentration to be measured,different volumes of gas were sampled from the head-spaceof the sealed malaxation chamber over time to have peakareas falling within the range of the calibration line.
Because the set of trials was conducted in about a week, adifferent calibration was constructed each time. A homog-enization of the air in the head-space of the chamber wasobtained by a quick hand aspiration and inflation of 50 mlof the same head-space gas by means of plastic syringe (3cycles).
The total chlorophyll concentration of the olive oil wasmeasured with a spectrophotometer by directly readingthe absorbance (A) of the oil at 630, 670 and 710 nmafter centrifugation in disposable plastic cuvettes using theequation:
Total chlorophyll (mg/kg as pheophytin α)= 345.3 [A670 − (A630 + A710)/2]/L
where L is the cell thickness in mm [23, 24].The totalsugar concentration of the wastewater was measuredbefore and after malaxation by phenol reagent method[25], after precipitation of any possible interference bythe Carrez procedure. Suitable dilutions of wastewaterswere necessary to fit within a calibration curve constructedwith glucose solutions of known concentrations. Theresults were expressed as glucose equivalent. Ethanoland lactic acid concentrations in the wastewater weredetermined by using enzyme kits supplied by R-Biopharm,Germany. These kits employ the enzymatic oxidation ofthe substrate ethanol and lactic acid with the correspondingstoichiometric reduction of NAD+ to NADH + H+ whichwas determined with a spectrophotometer at 340 nm.
Results and discussion
Carbon dioxide emission
The levels of CO2 emissions from the pastes during malax-ation are reported in Fig. 2 as changes of concentration inthe head-space of the sealed chamber (A, B, C, D, G).
In these trials, an initial rapid increase in concentrationwas observed followed by a gradual decrease in emissionrate. The data fits a saturation equation with R2 values from0.86 to 0.97 for the different experiments as reported inTable 2. The general phenomenon is illustrated by the trendline derived from the entire set of data from the sealed trialsin Fig. 2. The initial increase in CO2 concentration was sur-prisingly large (a mean of 32 ml/(l min)) and led us to sus-pect that it was caused by a possible contribution of chemi-cal reactions (such as dissolution of carbonates in an acidicenvironment) of compounds present in the olives. This hy-pothesis was ruled out by the fact that the pH measurementsperformed on the olive mill wastewater, separated imme-diately after milling, evidenced neutral pH levels. Further-more, measurements of gas evolution performed on olivepaste aliquots of 5 g by means of a De Astis calcimetershowed no gas evolution upon treatment with 6 M HCl.
Fig. 2 CO2 emission of olive pastes during malaxation (A, B, C,D, G). The trend line is the line of best fit of all data from the sealedtrials
According to the general respiration equation:
C6H12O6 + 6O2 ⇒ 6CO2 + 12 H2O
The molar ratio of CO2/O2 is 1 and no increase in the over-all gas pressure was theoretically to be expected if CO2was derived only through aerobic respiration [26–28 ]. Inall the sealed malaxation trials a negligible initial pressureincrease was observed. This suggests that the initial por-tion of the curve was due to the onset of accelerated cellularrespiration after milling, as the result of the extensive con-tact with oxygen and the breakdown of cellular structurescaused by crushing and successive malaxation.
The final portion of the curve, that above 280 ml/l, wascharacterized by CO2 emission rates that were about 30times lower (a mean of 1,1 ml/(l min)). Assuming thatmost of the CO2 was derived from respiration, the residualoxygen concentration in this portion of the curve shouldbe low and the small amounts of evolved CO2, which wereclearly detectable, were probably the result of differentbiochemical processes as will be discussed later.
Different olive paste amounts were used in the experi-ment resulting in different head-space volumes (Table 1),
Table 2 Parameters describing CO2 emission in malaxation trials
and therefore different CO2 concentrations were expected.However, all five malaxation trials (A, B, C, D, G) showedessentially the same CO2 concentration in the head-space(Fig. 2), demonstrating that, in these experimental con-ditions, CO2 evolution is primarily controlled by oxygenconcentration rather than the amounts of olive paste in thechamber. This suggests that a greater potential respiratorycapacity of the olive pastes remained unexpressed in theseexperimental conditions.
Some slight increases in mixer pressure were recorded afew minutes after beginning the trials. However, becausegas samples were taken from the head-space to determinethe levels of CO2, exact measurements of the pressure in-crease in the chamber were not possible in this experiment.
This increase in pressure led us to believe that fermen-tation processes had taken place along with respiration, aspointed out by other authors [29–31].
Furthermore, the CO2 evolution in the final part of thecurve shown in Fig. 2 reached concentrations far above220 ml/l, indicating that fermentation processes may beoccurring in addition to respiration in order to account forsuch high concentrations. In fact, if all the atmosphericoxygen were converted to CO2 only by means of respira-tion, for an average O2 concentration in the air of 22%, thecorresponding CO2 concentration would be 220 ml/l.
If this were the case, an increase in gas pressure shouldoccur, along with the presence of fermentation products inthe olive mill wastewater. It is known that vegetal cells atlow oxygen concentration may perform alcoholic and lacticintracellular fermentation [30, 32–36].
The amounts of ethanol and lactic acid found in thewastewaters are reported in Table 3. The presence of thesetwo compounds indicates that fermentation did indeed oc-cur during the malaxation trials. A statistically significantincrease in the concentration of these two by products wasrecorded in all the trials after malaxation (Table 4). Presum-ably, these fermentation processes are related to the initialsanitary conditions of the olives and/or with malaxationtime. Further investigation is needed to clarify the ultimateconsequences of these processes on oil quality.
Changes in sugar concentration of about 1–2 g/l werefound in the olive wastewater before and after malaxation(data not reported). These variations were apparently not re-lated to malaxation time or to CO2 emission, probably dueto the variation of the samples. However, average initialsugar concentrations in the range of 35–50 g/l were mea-
Table 3 Concentration of fermentation products in wastewaters
Fig. 3 CO2 emission of intact olives during storage (trials H, I)
sured, which is comparable with what has been reported byother authors [37, 38].
The CO2 emission from intact olive drupes (control trialsH and I) in the same malaxation chamber showed valuesabout 40 times lower with respect to the CO2 emissionfrom olive pastes, specifically 0.4 and 0.8 ml/(l min) forH and I experiments respectively (Fig. 3). The data fits alinear equation with R2 of 0.95 and 0.98. These emissions,although largely variable due to the differences in variety,origin, storage time and ripeness stage of the olives, arein good agreement with those reported elsewhere for olivedrupes [39–42].
Effects on olive oil quality
The technological application of this naturally evolvingCO2 should result in a considerable increase in the qual-ity of resulting oil. It is well known that oxidation thatoccurs during transformation results in the loss of com-pounds such as polyphenols, tocopherols and chlorophyllswhich are important to human health and olive oil conser-vation [5–7, 15]. Other research has shown that by limitingthe contact of olive paste with atmospheric oxygen duringmalaxation (e.g. by means of inert gas blanketing) olive oilswith a higher concentration of anti-oxidant compounds,and a consequent quality increase, are obtained [18–21].A similar increase in quality is presumably obtained whenusing the CO2 naturally produced by the olive pastes during
525
Table 5 Total chlorophyll concentration of the oils
malaxation with a large reduction in the costs imposed bythe use of inert gases.
Furthermore, as CO2 is the heaviest component of air(Mw: 44), it is feasible that it will stratify over the pastesurface during malaxation, resulting in better protectionagainst oxidation than in the present experiment. The CO2concentration data in these trials were recorded after ho-mogenizing the gas in the head-space of the chamber. Actu-ally, the occurrence of this stratification of CO2 was verifiedin a separate experiment were two subsequent CO2 mea-surements were performed with and without mixing the airin the head-space (data not reported).
For the purposes of our research, total chlorophyll con-centration was selected as an easy-to-determine parameterthat is representative of oil quality. Further, it representsan anti-oxidant component that improves the stability ofthe oil in the dark [43–50], it is related to the nutritionaland commercial value of olive oil [51–53] and, finally, itis a sensitive parameter towards the extraction techniques[54–60].
Total chlorophyll concentrations measured in the oils pro-duced are reported in Table 4 and Fig. 4. A general increasewas observed in all trials when comparing the oils obtainedbefore and after malaxation, confirming the findings re-ported in previous experiments [61]. However, there is abig difference between the trials conducted in the sealedmalaxation chamber (A, B, C, D, G) and those that wereopen to the air (E, F).
The increase in total chlorophyll concentrations mea-sured before and after malaxation are illustrated in Fig. 4,where the high level of extraction due to the use of thesealed chamber is clearly highlighted. The increment ofthe sealed trials (A, B, C, D, G) is about six times greaterthan the others (E, F).
Other experimental variables being equal, i.e. assumingan equal dynamic of extraction, these differences proba-bly resulted from oxidation of chlorophyll by atmosphericoxygen in the trials conducted with traditional open mix-ers [54, 57–59]. In the case of the sealed mixer, the rapidCO2 emission resulting from both respiration and/or fer-mentation processes of the olive paste afforded effectiveprotection from oxidation resulting in negligible oxidationof chlorophyll and a higher final concentration as indicatedby the intense of green color of the oil.
Fig. 4 Comparison of total chlorophyll concentration incrementsafter malaxation trials
The concentrations of total chlorophyll were particu-larly high in comparison with the data reported in pre-vious literature [62–65], so possible hyper chromic ef-fects due to the release of metal ions from the malaxa-tion apparatus should be investigated [66–69]. However,as these trials were conducted in the same conditions,the observed effects must be ascribed to the tested hy-potheses. Further investigation on extraction dynamicsand evolution of anti-oxidant compounds resulting fromthe use of this technique will be the subject of a futurepaper.
Conclusions
Olive pastes produce large amounts of CO2 during malax-ation, particularly at the beginning. This is primarily dueto cellular respiration and to a minor extent to fermenta-tion processes. When malaxation is conducted in a sealedchamber it leads to oxygen scavenging and CO2 productionwhich inhibits the oxidation of compounds responsible forthe quality of the olive oil.
The authors propose the technological use of these nat-ural CO2 emissions to reduce oxidation processes in oiltransformation plants, thus increasing oil quality througha proper engineering of the malaxation mixers (Italianpatent applied N◦ FI2004A000247). A further develop-ment would be the use of excess CO2 produced forthe saturation of other mechanical components used be-fore and after the malaxation chamber to afford furtherprotection from oxidation throughout the transformationprocess.
Acknowledgements The authors would like to thank the AntinoriEstate and Dr. Stefano Di Blasi in particular for his assistance.
Chem 81:547–5533. Torre RFD (2000) Rivista di Frutticoltura e di Ortofloricoltura
62(10):14–204. Capodagli A, Cicco A (1997) Rivista di Politica Agraria
Rassegna della Agricoltura Italiana 15(6):29–385. Servili M, Baldioli M, Federici E, Montedoro GF (1999) Atti del
4◦C ongresso Italiano di Scienza e Tecnologia degli Alimenti(CISETA) 4:294–304
6. Servili M, Mariotti F, Baldioli M, Montedoro GF (1999) ActaHorticoltu 474:609–613
7. Servili M, Mariotti F, Baldioli M, Montedoro GF (1996)Proceedings of the World Conference and Exhibition of Oilseedand Edible Oils Processing, Istanbul (Turkey), pp 289–294
8. Gimeno E Castellote AI, Lamuela-Raventos RM, De la TorreMC, Lopez-Sabater MC (2002) Food Chem 78:207–211
9. Di Giovacchino L, Costantini N, Serraiocco A, Surricchio G,Basti C (2001) Eur J Lipid Sci Technol 103:279–285
10. De Stefano G, Piacquadio P, Servili M, Di Giovacchino L,Sciancalepore V (1999) Fett/Lipid 101(9):328–332
16. Servili M, Baldioli M, Montedoro GF (1994) Proceedingsof the Second International Simposium on “Olive Growing”,Jerusalem (Israel) in Acta Horticolturae 356:331–336
18. Servili M, Selvaggini R, Taticchi A, Esposto S, Montedoro GF(2003) J Agric Food Chem 51(27):7980–7988
19. Servili M, Selvaggini R, Baldioli M, Begliomini AL, TaticchiA, Esposto S, Montedoro G (2002) La Rivista Italiana delleSostanze Grasse 79:439–441
20. Vierhuis E, Servili M, Baldioli M, Schols HA, Voragen AGJ,Montedoro GF (2001) J Agric Food Chem 49:1218–1223
21. Servili M, Baldioli M, Selvaggini R, Mariotti F, Federici E,Montedoro GF (1998) The Proceedings of the 13th InternationalSymposium on Plant Lipids Sevilla (Spain) in Advanced inPlant Lipid Research, pp 307–310
22. Calamai L, Lozzi I, Ristori GG, Fusi P, Stotzky G (2000) SoilBiol Biochem 32:815–823
23. Psomiadou E, Konstantinos X, Karakostas, Blekas G, TsimidouMZ, Boskou D (2003) Eur J Lipid Sci Technol 105:403–408
24. Pokorny J, Kalinova L, Dysseler P (1995) Pure Appl Chem67:1781–1787
25. Dubois M, Gilles KA, Hamilton JK, Rebers PA Smith F (1956)Anal Chem 28:350–356
26. de Wild HPJ, Peppelenbos HW (2001) Postharvest Bio Tech22:111–119
27. Forcier F, Raghavan GSV, Gariepy Y (1987) Int J Refrigeration10(6):353–356
