L #2-. _ _ v‘v A "'1'!‘ r v---—... , 11 ‘2‘2'22 2 ,2722, 2,, . 2,2222“ "2222‘ 222‘ . 2222.222 ~22 "22“222222‘2 “2 “““‘“““22‘2 ,2'2‘22,2 111112112111 2,222,” 122 22,, 222‘, 222 2 MW. .':-_ “flu-“9 v-Q‘i. 4—.n,,___ u..- %:§*‘ '— -————-—l—S .- «an; .-o.-. . . -;o.o:;..u-;'-=n Cum-u? w—wm .4’ mm... -7 ' .m'; .o 34...... u... v: 2 u y . .___._, _-~;"7.. fiat}; wr:::‘.:?_“ 5'27? '53.: = 2‘““ 2 2 ~‘ 2 2 - o-~& ' u no... .0... $22:- . ' I I. ' ..,——-— n-.— 2 o." ..-, .z‘ r r a.» u... 2,, , 22, ‘ m "“4 .4- n... w... 2,,2‘22' Ala “a '3‘2' 2222‘, 22, 22222, {2221223 E 231‘ 0' '- Wt” '—— ”:63: M ---"'—":=.__..__ __, ~‘-“-.':_'-: “on; = M;.:’. " it"; u -‘ u~_ ' . " mm 2 2 n 2 2 é2““' “22“ 2222“”, ,22,2 12.2 22,. “ ‘,:‘222 22,, 2 2‘21‘21, 22 222222222222 2213;221:2222 2‘2‘222222'2’2222 22 12“,:J 22, 2' 2‘, :2'22 ,2 :22 2,, 2‘2, 2,2, ,21'222 22 2,2‘22‘2“' 2‘2222‘ ““““‘221 212,, ‘ 22" 222“‘22 2.222, 22 “ [2 ‘22 2‘,‘ ,122,1,2“ 2 ,.,.. 22 2,2,... ,. 222,22,2,,,11,,, i222,” '''' 131‘“, 22,2, 222 222222 222 ,2“ ",12, {22, 2 2 222122222” ‘ "33"” 2,22. 222 2 ‘2‘2 222 2222, 222,22 ‘2‘2 ,22‘, 522 223,2 2122222,, 2,, ,,, E12: 221" 1:2," 2,22,,1'22222,22 ‘1, 22‘ "‘2, ‘f‘ “"53 ‘22‘ 22222 2‘22 .222 2 2 2,‘,2 2 2222,32 ‘22“ '12,? 222,22 ‘222' 2,,‘2 222 222 ‘2‘2 2"‘222 22‘: '12 2,2 2,22‘, 2 ““ ‘ 22‘, '2" 2‘2,‘ '2'2‘,‘; “‘2‘2£,2 ‘ “ ‘ ‘2‘ ‘ “:22‘2 2‘, 2 2 22 2,2222, 22‘, 222222. 22 ,222, ,1 ,1, ,1,,2 ,, '22 222,,,2,,, 22,2 ‘2‘ “'2’222 222‘ 221,,', ,,222 2,2 1'2'2' “22 2‘22; '2 ‘,2 2,22,‘,1, 222 2,212,”, 21 '2222, , 2 ,2 22:2 ‘2‘ ‘2‘2‘ “‘222‘2‘2 2222,, 2‘“ 223‘- 222 22,22 2,11 ‘,, 2‘2222“, “1,111 ,221: ‘ 22222 2, 2““ ,,H, 2, ,2 22 22,1 ‘22,,212'2 52,222 221,,,,2,,,,1112,22 2,112,“, 222‘, . ., ,12,2.2,,‘,.‘,;:,2;,‘222 2 2‘ 2222 2,,‘211‘. 22122222 2222“ 22,,2, “‘ 2220‘ 22, 2 ,2 22"1 , ‘22‘1‘2‘,,§‘ 2,222 3222,,“ 222,2,22, . ‘,,““ .2 2,,21; 222 121, 31,11, 2;“:{2' W 2' “““ “ "‘2“ “"‘2 22221, 22,, “'2 22, ,, 22.22 ‘,"_2', ,l‘,‘ 22,,, 2 ' 2.2.22 222 L221,,‘. 2,, ,~, 2‘ 2,22,,22 2,222 ,2222‘ ,1 22‘2 “‘ 2" 22‘222 1'2' “,222‘L2 22 ‘2'“ 2222‘1 I221; 212,, 1,11, .' 3.6. 3. yr‘. ~ A:\ --_ ~ . . ._" “A, .- ' - . 2 . .- A u...::, ‘7': -"‘- - a .JF'gwfi $5,..— ...-. - J“. ‘M .. u, _ . N H ~vc— . 1;...” :vz" 1G" W - ,_., ,— .- .. -... ”TEL n. . - s...»- w'%: M» 3" “:me... ”‘-"-':“.J .31:- m- 28:“ '5‘" on Q...“ ‘ c 2v ' ’n“; . . . 0 5“- ~ .ai’. .. .... _ . - a:— 0.2:. o . w ' .0» an— 3.. .72.. WW '4‘- . d - .. ha 4“ . u. - (h..- H...- w... 3.. c ‘ - C. ’ g. M - "‘ ~ - . 4 _ . . ‘ -_ - - _ L. . - ' O» O ’ ' -- - - - .4 ‘h . r- . ~ ' . .' c l - ‘ V A ' . on . ,- , . ~ . : ’._ ‘ - . - . I ’ ‘ -"' '-‘~ ‘ .‘ -.- _, - . . .. 1 a I"'r . .. N V . ' . ‘ . ‘. ...~- . ~ . v t " o .. -- u v . ‘ - . -0 ‘- .I ' - v- . ,. ~ . .- . . . - .. .. I - A , - _ . ~ 1' . w u. I: .— M m .. .n Wavt'f Lin...» - 4 .. .. . - . - ... ’3: o...- .13: - .21? WI...“ ‘ . . ¢ , . -...o1a ”-12 ”euro:- 1-0;.- . f O ‘ ”m mW-‘2'm - -« . O o I?“ m. .. k .- .2“?! no . u:— c- o c. o - u .4 "3" . '3' co. “—9- .m - .- , I. m. . o . fl... ' < . o .0. ”"1337: ' u -. .- o - ....—-. o. ‘I I ; '2”- nova. 'vv—o “,1..." 2' h ’1?“ . :8... - . . -. . . . .I' ‘—.- 3.3-“ .. . . “ .. «:4: ‘- r.‘ C .- w l‘ “2 222 2‘ {‘2‘ ‘2222, 2 ““2 1,1212, 22', ,,‘,,1 222, 2“ o ' .— r” — a. .5. ' ~— '12: .‘u ‘ ' hflm—‘f'; ' 3‘35, "I .--..-—. "v‘or Em?“ -. ......""‘.~..:- ~ .. M Ma” ‘ 2-: ._..:.:. M m“- m. ‘ ~w3";v:fl mil; ' v ' u ‘ 1". . . - ~- - ' In . . IIIIIIIIIIIIIIIIIIIIIIIII Y ”W“ WWII“WWIHIIH‘IHWWWI ' \/ 3 1293 10616 3474 This is to certify that the dissertation entitled QUALITY STUDIES ON OLIVE OIL presented by Apostolos K. Kiritsakis has been accepted towards fulfillment of the requirements for Ph. D. degreein Food Science Major me Date 13 [UN I782, MSU is an Affirmative Action/54"“, “mm WWW 0.12771 1““ s _ x ,y ’34:; .- m“ ' R4133: guafiifl SWEIE University r fin ‘ ‘\_J MSU LIBRARIES n \ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. “is I EE6*EF9 ?§9l Wfigijwao CH powwwudmpa mucmcoafioo oaaumHo> cfimuumo H oHDmH 11 K + K _ 262 274 AK ‘ K268 ' 2 Usually the specific absorbance (Elim) is determined in each of the cited wavelengths. The International Olive Oil Council (1966) used the criteria of FFA, peroxide value, ultraviolet absorption and or organoleptic properties to characterize the qualities of virgin olive oil as shown in Table 2. Changes in Quality Characteristics It has been noticed (Martinez, 1975) that, during the time that oil remains in the fruit, the values of the above criteria are changed. Some become higher and some lower in organoleptic score. The free fatty acid content is increased due to hydrolysis of glycerides. Hydrolysis is caused by micro- organisms, enzymes or water present in the fruits (Martinez, 1975). The activity of some microorganisms is such that the time between the milling of the olives and the separation of the oil from the vegetable water is not sufficiently short to exclude the possibility of a certain amount of hydrolytic action on the components of the oil (Martinez, 1975). Enzyme lipases are present in the fruit and become active with the process of maturation (Martinez , 1975). Infestation of fruits by dacus oleae fly or fungi or any other damage to the fruit resultsinrniincreasein.the 12 uommumm eaausaomn< oHo.ow mN.ow omw m.mw sumceeuo uommuma mHmusHomn< oHo.ow nN.ow ONW Nm.Hw mane uummuma samuaaomn< OHo.ow om.ow omw No.Hw muuxm AHHo wx\No case Ho>mam >3 GH mSHm> prom oamao maflo coauocwuxm owwflumam mpflxonmm mmv HHo wauw> .Hfio m>flao mo maneufluo huwamao N maan 13 FFA content of the oil (Martinez, 1975; Newenshuanter and Michelakis, 1978). Studies on the aromatic constituents of the oil and on the changes that these undergo have been made by Montedoro 33_31., 1978). He noted that even during stor- age of fruits in the milL there is a loss of olive oil con- stituents due to the hydrolytic enzymatic mechanism of the cell wall. He believes that this process begins during ripening of the fruits. Storage of olive fruits for ten days caused a decrease in aldehyde content from 26.62% to 13.58% and in phenols from 104 mg/Kg oil to 89 mg/Kg oil (Montedoro 33 .31., 1978). Tocopherols, Phenols and Sterols as Natural Antioxidants Tocopherols are natural antioxidants which are pri- marily responsible for the stability of vegetable oils. There are four commonly occuring tocopherols designated a-(alpha), B-(beta), y- (gamma), and 6-(delta) tocopherols. The relative effectiveness of these four tocopherols as antioxidants is 5>y>8>a (Daubert, 1950; Dugan, 1976). A similar order (6>y>a) was reported by Sherwin (1976). He noted,however,that this order of antioxidant potency in vegetable oils may be influenced significantly by temper- ature and light conditions. If the residual tocopherols are stripped completely from vegetable oil by distillation or some other efficient method, it will be observed that 14 the oxidative stability of the oil will be reduced to an extremely low level (Sherwin, 1976). Lea and Ward (1959) observed that tocopherols were much less effective as antioxidants in light than in dark. Like other antioxidants, the tocopherols are them- selves readily oxidizable. Oxidation of tocopherols lead to the formation of tocoquinone (Tappel, 1962; Sonntag, 1979), which is not an antioxidant (Sonntag, 1979). Accord- ing to Gracian and Arevalo (1965), y-tocopherol present in olive oil is a product of a-tocopherol oxidation. Yoon and Kim (1974) in their studies in dark and sunlight-irradiated conditions observed that a-tocopherols in soybean oil showed some retarding effect on oxidation, but the effect decreased rapidly as storage time increased. Gutfinger and Letan (1974) found that lipids extracted from the olive seed kernel were higher in toco- pherols than the lipids from the fleshy part of olives (291 ug/g 011 vs 186 ug/g). In the former, the relative content of y-tocopherol was higher (25% vs 7%). Neither olive seed kernel nor olive flesh contained G-tocopherol. Fedeli (1977) noted that the concentration of dif- ferent tocopherols in olive oil was as follows: 88.5% d-tocopherol, 9.9 % B+y tocopherols and 1.6 % é-tocopherol. Using dimensional paper chromatography, Gracian and Arevalo (1965) identified only a-tocopherol in olive oil. Vitagliano (1960) and Boatella (1975) both agreed that 15 olive oil contains a-tocopherol but they reported differ- ent quantities; the first 12-102 ppm and the other 70-150 PPm- Gracian and Arevalo (1965) reported that the varia- tions in the concentration of different tocopherols in olive oil may be explained by the progressive destruction of tocopherols. Tocopherol content of olive oil can be used for detecting the adulteration of the oil with other oils (Ninnis 3E 31., 1969; Gutfinger and Letan, 1974). Accord- ing to Gutfinger and Letan (1974), addition of soybean oil to the relatively expensive olive oil can be recognized by the presence of excessive amounts of yor 5-tocopherol. On the other hand, addition of cottonseed oil to olive oil can be determined by the presence of excessive amounts of y- tocopherol. Simple as well as complex phenol structures are found in olives (Fedeli, 1977). Most of these constituents go into the aqueous phase as the oil is processed in the mill. However, a fraction remains in the oil and favors its sta- bility to oxidation (Cantanelli and Montedoro, 1969, Notte and Romito, 1971; Vazquez 3£_31,, 1976; Fedeli, 1977). Vazquez 33 31. (1976) demonstrated that the poly- phenol content of olive oil varies according to cultiva- tion procedures and environmental factors. Montedoro 33 31. (1978), on the other hand, noted that the factors which can affect the phenolic constituents of olive oil 16 are the harvesting period, the condition of fruit preser- vation and even the extraction system. Montedoro 33 31. (1978) noted that the decrease of phenolic constituents during the extraction process may be explained by the solubilization effect of the vegetation water and particularly by the dissolving of the colloidal substances (proteins and polysaccharides) which bind these components. Vazquez 33 31. (1976) found that the main poly- phenols present in virgin olive oil were tyrosol and 3- hydroxytyrosol and observed some antioxidant effect in 3-hydroxytyrosol. Phenolic acids like caffeic, vanillic, p-coumaric, p-hydroxybenzoic , and protocatechuic have been found‘ in olive oil (Vazquez 33 31., 1976). Cortesi 33 31. (1981) deter- mined (with HPLC) the phenolic acids present in virgin olive oil, refined.and solvent extracted oil and found that virgin oil had higher levels of protocatechuic and cinnamic than the other oils. Montedoro 3E_31. (1978) declared that the most inter- esting constituents from the organoleptic point of view are the polyphenols: B-3,4 dihydroxyphenylethanol (hydroxytyrosol) found only in very good quality oils, the 3-hydroxy-phenylethanol (tyrosol) and some phenolic acids, found in oils of poor quality. It is believed that hydroxytyrosol and tyrosol are derived from the hydrolysis of oleuropein while others 17 (benzoic and cinnamic acids) are derived from the hydrol- ysis of flavonoids (anthocyanins, flavones) (Montedoro and Cantarelli, 1969; Vazquez 3E 31., 1976) which are found in considerable amounts especially at the ripening stage. Notte and Romito (1971) demonstrated that polyphenols extracted from olive leaves acted as antioxidants in olive oil. These polyphenols were effective in a concentration of more than 20 mg/100 g of oil. Cantarelli and Montedoro (1969), on the other hand, observed that phenols extracted from olive oil with 80% MeOH acted as antioxidants in other oils while the extraction of phenols from olive oil caused its rapid oxidation. Gutfinger (1981) also reported that removal of polyphenols from olive oil markedly increased the oxidation rate. Fedeli 33 31. (1966) used GLC to identify stigmase terolanuicampesterol in olive oil. Fedeli (1977) reported that Italian olive oil contains 95.6% B-sitosterol, 1.3% stigmasterolanui3.1% campesterol. Boskou and Morton (1975) on the other hand, stated that Greek olive 011 contains traces of cholesterol, 2.0% campesterol, 0.5% sigmasterol, 89.5% B-sitosterol and 8% anemasterol. Gutfinger and Letan (1974) found that the sterol compositions of olive, cottonseed and avocado oils were similar. In those oils, B-sitosterol was the dominant sterol (Ca 90%), and it was accompanied by only limited amounts of stigmasterol (0-2.3% . Different amounts of sterols were present in olive oils extracted from either 18 the flesh or the seed kernel of olives but the composition of the oils' sterol fractions were almost the same (Gutfinger and Letan, 1974). According to Leone 33 31. (1978), the B-sitosterol content of olive oil depends on the cultivar and oil acidity. The determination of the sterol content of olive oil can be used for the detection of adulteration. Excessive amounts of stigmasterol indicate the presence of soybean oil in olive oil (Gutfinger and Letan, 1974). The antioxidant properties of some sterols present in olive oil were studied by Boskou and Morton (1976). They found that AS-anemasterol added to heated cottonseed oil eliminated the oxidative deterioration of the oil, while a sterol mixture consisting of 94% B-sitosterol showed some prooxidant effect. Leone 3E 31. (1976) demonstrated that any increase in the peroxide value of olive oil during storage is asso- ciated with a decrease in sterols. Effect of Harvesting on Olive Oil Quality A variety of factors influence the quality of oil present in olive fruits such as: cultivar of olives, vari- ability of soils (physical and nutritive elements), and pests (which either facilitate or impede the development of micro-organisms) (Martinez, 1975). Methods used for 19 harvesting olives influence the characteristics of the oil to a degree. There are two important harvesting factors to con- sider: the season and the method of harvesting. As to season, according to Martinez (1975), olives should be harvested at the moment of optimum maturity which is when the fruits have the maximum quantity of the oil with the best characteristics. This moment can be established visually when the fruit begins to darken. According to Martinez (1975), the fruits yield a different type of oil during each stage of maturity. The time of maturity, however, is not always the same even when; the conditions are identical and the trees are within the same olive grove. Therefore it is difficult for all of the olives to be harvested at the optimum stage because maturity is not the same from zone to zone, even with olives on the same tree. Martinez (1975) stated that fruits harvested early give a lower yield of oil, with low free fatty acid con- tent, greener in color and with a fruity savor, whereas, when the harvest is delayed, it produces a greater yield, with slightly higher acidity, a yellow color and oils that are generally less aromatic. Montedoro 3E 31., (1978) demonstrated that the con- centration of different constituents of olive fruits increases with the degree of pigmentation until a stage is 20 reached beyond which an inversion of this relationship is observed. They (1978) noticed that the highest concentra— tions of both (total volatile and total phenol) constit— uents are found during the phenological period between the semiblack and completely black olives . This period corres— ponds to the one during which the oil concentration nearly reaches its maximum and the aromatic characteristics of ”fruity" are the highest. According to Martinez (1975), collection methods which harm the olive fruits result in losses in oil quantity and quality. He reported that olive fruits col- lected from the ground produce oil of higher acidity and of different organoleptic characteristics than those gathered directly from the trees. New Systems for Olive Oil Extraction The main systems used today for olive oil extraction are the traditional, the centrifugal and the mixed. The traditional systems, used for thousands of years in the olive oil industry, are based on the application of hydrau- lic pressure. Centrifugal Systems Centrifugal systems in industrial base, were first made at the beginning of the decade of 1960 despite the fact that centrifugal extractors were known since the end of the 19th centruy. 21 Experiments recorded were not only those carried out by Boulier in 1903 and the Agricultural Experimental Sta- tion of California in 1904, but also those of Brani and Merer-Revoil in 1908, Degli Alti in 1906, of the California Packing Corporation in 1927, of Ortiz and Rouseau in 1929, of Ferraris, Pantanelli and Brandonisic in 1937, of Perogio in 1941 and of Tostorelli in 1950 (Petruccioli, 1975). Systems based on the centrifugation of olive paste have progressed considerably in the last few years. Dif- ferent companies such as ALFA-LAVAL, PIERALISI, HILLER, THEOCHARIS and others make centrifugal systems. According to Petruccioli (1975), these systems will replace traditional (presses) in the future. Advantages claimed over the traditional systems include, among others, improvement of the quality of the oil obtained. The oper- ation of the centrifugal systems is based on the different specific gravity of the constituents of olive paste (oil, water, cake). The separation of the three phases is accom- plished in a decanter (centrifugal), the main component of the processing system. The velocity of separation during the centrifugation is given by the formula: 22 2R2(Sz'sl)°g V: ge where: V = Velocity of separation R = Radius of heavier phase S1 = Specific gravity of the light phase 82 = Specific gravity oftfluaheavier phase g = Gravity e = Viscocity coefficient of light phase The processing steps in a centrifugal system are shown in Appendix 1. Mixed Systems The operation of the mixed system is based on selective filtration and on centrifugation or pressing. At the beginning of the present century,the Spaniard Acapulco discovered that if a thin layer of the paste of de-stoned olives was placed on a filter medium (this could be cotton cloth) the filtered liquid collected is oil and is practic- ally free of the vegetable water which accompanies the fruit. This observation was the basis for making a system of extraction, which, although often modified and varied, is employed today to a certain degree in combination with other systems (principally with centrifugal). The first complete installation using the new method was constructed with the aid of the agronomist Quintanilla 23 and thus the system is known throughout the world as the Acapulco-Quintanilla system. According to Petruccioli (1975), the greatest improvement on the original Acapulco system was achieved by Buendia in 1951 with his ALFIN extractor, better known in Italy and Greece under the name SINOLEA. Since then, the ALFIN extractor has received a variety of modifica- tions. In 1972 a complete system based on selective filtration and on centrifugation was constructed by the company RAPANELLI. This is known as a mixed system since it gives about 80% of the oil of paste with selective filtration through the unit of SINOLEA, while the rest is accomplished with centrifugation through a DECANTER unit. The basic steps of processing in the RAPANELLI system, giving SINOLEA and DECANTER oils, are shown in detail in Appendix 2. Effect of Extraction System on Olive Oil Quality Olive 011 quality may undergo deterioration during the processing of olives for oil extraction. Mendoza (1975) noted that the milling operation plays a very impor- tant role in the extraction of olive oil because the way it is carried out and the type of machine used have a direct influence on the processes of later operation (malaxation, decanting, centrifuging) and particularly on 24 the yield and quality of olive oil. It mainly affects its organoleptic characteristics. According to Mendoza (1975), the exposure of olive past to the air during the milling operation results in loss of aroma. He noted that the type of materials used affect considerably the quality of the oil. Traces of metals carried away from the metallic surfaces of mills cause changes in the organoleptic characteristics of the oil particularly in color and taste. Moreover, these metals can act as catalysts in the oxidation of olive oil (Mendoza, 1975). He further demonstrated that heating (>25°C) of the oils during processing (milling, malaxation etc) leads to changes which significantly alter olive oil quality, because the volatile components which contribute to the aroma of good oils are lost rapidly. High temperatures can also cause changes of color to a reddish hue, increase the FFA content, and increase consuption of energy. Montedoro 3£_31, (1978) studied the effect of paste mashing (during the extraction) on the phenolic compounds in the oil. They found that, by direct pressing, oil obtained finally had 616 mg/Kg total phenols, by milling and pressing 450 mg/Kg, and by milling, mashing and pres- sing 260 mg/Kg. 25 Martinez (1975) also noted that the different sys- tems used for processing of olive oil may affect its quality. Felice 3E 31. (1979), in their studies with centri- fugal and traditional (with pressure) systems, found that oil obtained from centrifugal systems had lower total polyphenol content (expressed as gallic acid) and iron con- tent than oil from traditional systems. According to them centrifugation oil had a reduced shelf-life in spite of the lower iron content. Cakmak (1978) studied the factors influencing yield and quality of olive oil obtained by pressingixiindustrial- scale trials. He found that oil yield increased with increasing grinding time, the peroxide value increased after grinding for 15 min but the FFA content remained unaffected. He further noticed that the FFA content, the peroxide value and the spectrophotometric absorption values (at 232 and 270 nm) of the oil increased with increasing pressing time. Organoleptic properties of the oil also deteriorated with increasing pressing time. Leone 33 31. (1979) compared the qualitative charac- teristics of oil obtained by single and double pressure extraction with an automatic plant (Belgioni system) that squeezes olive pastes (at 600 Kg/cmz) without using vege- table or synthetic fibre filters. He found that the FFA 26 content of oil from the Belgioni system wasnot different from traditionally single pressure extracted oil, the spectro- photometric indices and the peroxide values were lower, the polyphenol and tocopherol content higher, the sterol and dialcohol terpene contents similar and the resistance to autoxidation higher than that of traditionalLyextracted 011. He demonstrated that the organoleptic quality of this oil corresponded to extra quality oil. Lipid Oxidation Autoxidation Several studies on autoxidation of fatty acids have been reported previously (Ross 3£_31., 1949; Boland and Ten Have, 1947; Dugan, 1961). A mechanism which is now generally accepted is that the autoxidation of lipid involves a free radical mechanism. The oxidation is initiated by H2 abstraction followed by addition of Ozto the carbon radical generated. In the entire mechanism, which involves the steps of initiation,propagation,and termination,the following reac- tions are involved (Boland and Ten Have, 1947). 1. Initiation V RH + 02 R° + °OOH 27 2. Propagation R- + 02 : ROO' ROO° + RH > ROOH + R- 3. Termination ROO° + ROO° A: ROOR + 02 ROO' + R- > ROOR R: + R- > RR RH refers to any unsaturated fatty acid in which the H is labile by reason of being on a carbon atom adjacent to a double bond. R- refers to a free radical formed by remov- al of a labile hydrogen. According to Logani and Davies (1979) the final distribution of the products of oxidation depends on the secondary reactions such as rearrangements of the inter- mediate radicals or of the final products and of further oxidation. Despite any secondary reactions involvedzhroxidation of lipids in the dark, this can be inhibited by free radi- cal quenchers such as antioxidants (Carlsson 33 31., 1976; Logani and Davies, 1979). Various extraneous influences may be present that affect the rate of oxidation. These factors are tempera- ture, light, ionizing radiation, enzymes, prooxidant metals 28 and metallic compounds, presence of oxygen, and use of antioxidants (Lea, 1962). Mechanism of PhotoOxidation Satter and DeMan (1975) reviewed the fact that chemical compounds absorb light energy in different wave- length bands depending on molecular structure. The absorp- tion of radiant energy is known as the primary photo- chemical process and results in an activated molecule: A + hv > A* (l) They pointed out that this is followed by secondary reac- tions in which the excited molecule can use the activation energy to emit light (Eq. 2) or heat (Eq. 3), form a new activated species (Eq. 4), form a bond (Eq. 5), or dis- sociate into ions (Eq. 6) or free radicals (Eq. 7). A* > A + hv (2) A* + A > A,+ A + heat (3) A* + B > A + B* (4) A* + A *> ArA. (5) A* > A+,+ e‘ (6) AB* > A‘ + B' (7) The two most important types of reactions concern- ing light induced lipid oxidation are 1) a photosensitized 29 reaction in which the light absorbing species does not undergo permanent chemical change: S + hv 4> 8* (8) S* + A A: A* + S (9) A* + B : AB (10) and 2) a photoinduced reaction in which the reactive spe— cies produced by the radiant energy initiates another reaction (such as a free radical reaction): I a: 1* (11) 1* e 1' (12) I- + A > A- + I (13) Schenck (1963) has categorized photosensitized oxi- dation reactions according to the intermediates formed. A "type I" sensitized reaction, which can proceed in the absence of oxygen, is one in which free radicals and elec- tronically excited molecules are involved. Compounds which are readily oxidized or reduced favor this type of reaction. Substances such as olefins, dienes, or aromatic compounds, that are not easily oxidized or reduced by a sensitizer favor a "type II" reaction in which oxygen par- ticipates and which occurs only through electronically excited molecules as intermediates. Oxygen may occur in a singlet or triplet state depending on the arrangement of electrons in the outer orbitals. Ground state triplet oxygen has two electrons in separate orbitals with opposed angular momentum and parallel spins. The first excited singlet state (1A) has 30 both electrons in the same orbital with the same angular momentum and opposed spins. A second excited singlet state (12) has a very short lifetime and decays to the A state after 1 X 10.11 sec. 1 The relationship between the oxygen states is shown in Table 3. Table 3 Relationship between the three states of oxygen.* State of 0 Relative Occupancy of Molecule 2 Symbol Energy Highest Orbitals Second excited 12 +37 kcal ——+f— -—:t—- First excited 1A +22 kcal :l—f A A Ground 32 0 kcal ——+—— ——+—— *Adapted from Foote (1968). The l A state has a lifetime of several psec. and is highly electrophilic in nature since it seeks electrons to fill its vacant molecular orbital (Korycka-Dahl and Richardson, 1978). Therefore 102 reacts readily with moieties containing high electron densities, such as dou- ble bonds. There are several reactions of 102 which may be of importance in biological systems. Singlet 02 can be generated in several different ways as is shown in Figure 1 . Probably the most important vmay by which singlet oxygen can be generated in fats and (3115 is through the presence of sensitizers (Rawls and Van Santen, 1970; Chan, 1977; Terao and Matsushita, 1977). ENZYMES RCOO-+RCOO- 02+38ENSITIZER Rcio sensitizer ENDOPEROXIDES 00H H202+oc1' products H20+C1 products ‘1‘; ‘ OZONIDES 2:: H 0 +0’ gfi“ on‘+onfr 2 2 2 H20+0H \\\\\\OH:§\\\\\\ ' H O . ' H202+HO2 : 2 OH 02 2H ' y‘\ e 0 +0' - 2 2 _ + 02 02+Y Fig. 1 Production of 10? by photochemical, chemical and biological systems (adapted from Krinsky, 1977). 32 Photooxidation 6f Lipids Photooxidation of lipids caused by chromophore impur- ities present such as chlorophyll, porphyrins, myoglobins, pheophytins etc., has been studied by several workers (Coe, 1937; French and Lundberg, 1944; Rawls and Van Santen, 1970; Koch, 1973; Carlsson 33 31., 1976; Chan, 1977; Krinsky, 1977; Terao and Matsushita, 1977; Malgorzata and Richardson, 1978; and Vianni, 1980). Rawls and Van Santen (1970) have identified 102 as a likely source of hydroperoxides which initiate autoxidation in oils originally free of all hydroperoxides. They (1970) proposed the following mechanism by which singlet oxygen is produced during photooxidation. 1S + hv > 18* ———————> 38* 3 * 3 g 1 * 1 S + 02 - 02 + S 103 + RH = ROOH ROOH > Free radical products (S is sensitizer, superscripts refer to the spin multi- plicity and the asterisk indicates electronic excitation). According to Rawls and Van Santen (1970), singlet oxygen (102), generated externally, was shown to react directly with extremely pure samples of methyl linolate at a rate at least 1450 times that of oxygen in the ground state. 33 Carlsson 33 31. (1976) noted that during exposure of chlorophyll in light the following reaction is involved: hv 302 l Chl > [Ch1]* = Chl + 02 The singlet oxygen (102) formed then reacts rapidly with C-C unsaturation to give hydroperoxides (Kaplan, 1971). H \ OX / I 'c=c 7 \‘ ’ I I These hydroperoxides can then cleave thermally even at room temperature and so initiate a conventional free radical autoxidation to produce more hydroperoxides (Carlsson 33 31., 1976). Although it is generally believed today (Rawls and Van Santen, 1970; Carlsson 33 31., 1976; Terao and Matsushita, 1977; Frankel 33 31., 1979) that singlet O2 is involved during photooxidation,some photosensitized oxida- tion of lipids may involve the triplet oxygen stage (Chan, 1977). In this case the sensitizer (riboflavin) reacts after light absorption with the substrate (A) to form intermediates which in turn react with ground state (trip- let) oxygen to yield the oxidation products. This is known as a Type I pathway for photo-sensitized oxidation. 34 Sens + A + hv > (Intermediates-I) (Intermediates-I) + 02 > products + Sens Besides Type I, there is aType 2 pathway for photo— sensitized oxidation (Chan, 1977). In this case,molecular oxygen rather than the substrate,is the species which reacts with the sensitizer after light absorption. Sens + 02 + hv > (Intermediates-II) (Intermediates-II) + A > products + Sens In the Type 2 photo-sensitized oxidation, singlet molecular oxygen is generally regarded as the reactive species responsible for oxygenation of the substrate. This excited molecular species is formed by reaction of ground state oxygen with an excited triplet state of the sensitizer. Unlike autoxidation, photooxidation involving trip- let oxygen doesn't involve chain reactions. ’In addition photooxidation doesn't require an induction period where- as the "dark" oxidation involves a long induction period (Logani and Davies, 1979). During autoxidation of poly-unsaturated fatty acids via a free radical mechanism, all hydroperoxides formed are conjugated (Khan 33 31., 1954). However, when methyl oleate and methyl linoleate were illuminated, in the pre- sence of chlorophyll, conjugated and nonconjugated hydro- peroxides were formed (Khan 33 31., 1954). Similar 35 observations were made by Rawls and Van Santen (1970). Nonconjugated hydroperoxides which were absent in autox- idation reactions were formed from methyl linoleate during photosensitized oxidation and on exposure to externally generated singlet oxygen. Role of Chlorophyll in Photogxidation of Oils Taufel 33 31. (1959) observed that chlorophyll in the presence of light acts as a prooxidant for methyl oleate. This pigment, however, has no prooxidant effect in the dark. (hithe contrary it acts synergistically with phenolic antioxidants. In order to have oxidation in the presence of chlorophyll in the dark, Tollen and Green (1960) showed that very strong prooxidants are necessary. Coe (1941) noted that the lower the chlorophyll value for a given oil normally treated,that is without excessive heat, the longer is the induction period of that oil. The chlorophyll value was believed to be indicative of the keeping quality of a normal oil. Clements 33 31. (1973) used sodium chlorophillin to study the participation of singlet oxygen in photo- sensitized oxidation of soybean oil. They reported that chlorophyll-like sensitizers are probably unimportant in well-refined soybean oil. Rawls and Van Santen (1970) in their studies of a possible role of singlet oxygen using chlorophyll, ob- served that singlet oxygen accounts for approximately 80% 36 of the chlorophyll induced photooxidation. They noted that the proton abstraction by the photoactivated carbonyl group of chlorophyll could account for the remaining 20% of the observed photooxidation. Vazquez 33 31. (1960) found that olive oils, because of their chlorophyll content are very sensitive to radi- ation between 320 and 720 nm whether in the presence or the absence of antioxidants. According to Francesco (1961) virgin olive oils have a chlorophyll absorption peak at 665 nm. Due to its chlorophyll content, virgin olive oil is easily oxidized and is very sensitive to light (Vitagliano, 1960). Interesse 33 31. (1971) demonstrated that under the action of light, the four pigments (chlorophyll a and b and pheophytin a and b) present in olive oil develop an oxidizing effect, while in the dark they act as anti- oxidants. Singlet Oxygen Quenchers Photosensitized oxidation is not prevented by the antioxidants commonly used to inhibit dark oxidation (Carlsson 33 31., 1976; Logani and Davies, 1979). Photo- oxidation can be prevented by singlet oxygen quenchers (Carlsson 33 31., 1976) which can have the opposite effect of sensitizers and decrease the rate of photosensitized oxidation either by physically or chemically reacting with 1 the 02. Thus they deactivate it to the ground state. 37 K 102 + quencher q > 302 + quencher B-Carotene'has been shown to quench lO2 efficiently and thereby inhibit oxidation by 102 (Foote and Denny, 1968; Carlsson 33 31., 1976; Matsushita and Terao, 1980). This quenching is believed to be the major mode of protec- tion against photodymamic action in living organisms (Foote 33 31” 1970a). In methylene blue photosensitized oxygenations, one molecule of B-carotene was shown to quench up to 1000 molecules of 102 before being consumed (Foote 33 31., 1970b). Naturally occuring a-tocopherols also quench singlet oxygen efficiently, but are themselves oxidized in the process (Carlsson 33 31., 1976). Addition of 6-tocopherol was found (Matsushita and Terao, 1980) to increase the life of B-carotene in photosensitized systems. To assess 1O2 participation in oxidation reactions, 1 O2 quencher inhibition and presence of nonconjugated hydroperoxides are now commonly used. Effect of Storage Conditions on the Oxidation of Olive Oil Pretzch (1970) reported that olive oil exposed to light and air undergoes greater oxidation than that stored in the dark. Interesse 33 31. (1971) also reported that exposure of olive oil to light accelerates oxidation. Valentinis and Romani (1960), however, noted that in the absence of air, direct sunlight causes a decrease of per- oxide and Kreis value during the storage of olive oil. 38 Gutierrez and Jimenez (1970), in their studies with virgin olive oil in polyethylene bottles, found that oil exposed to light undergoes greater oxidation than that stored in the dark. In experiments carried out in open and closed tins, the increase in peroxide value of oil in Open containers (exposed to unlimited air) was high. On the other hand the formation of peroxides in closed tins, due to the limited amount of oxygen present in the headspace, was generally insufficient to produce the typical rancid odor (Cucurachi, 1975). According to Cucurachi (1975), samples of oils with the same peroxide value have dis- similar organoleptical scores if they are packaged in different ways. In storage studies with different oils (olive, pea- nut, almond), Wiegand (1978) realized greater resistance to oxidation by olive oil than other oils. Cerezal 33 31. (1975) found that olive oil stored up to 325 days in a tank coated with epoxy resin was much more stable than olive oil stored in an iron tank with respect to FFA content, peroxide value, metal contents and organoleptic evaluation and was similar in these character- istics to olive oil stored at 4°C in glass bottles. In a study of olive oil with different types of con- tLainers, the following peroxide values were recorded after 13 months of storage under light conditions: Samples in iglass and PVC containers had PV lower than 20; samples in 39 polypropylene had PV 20-40; samples in high-density poly- ethylene had PV 30-80; and finally samples in low-density polyethylene had PV 15-45 (Gutierrez, 1975). Gutierrez (1975) reported that storage of olive oil in containers impermeable to atmospheric oxygen and light conditions results in protection of the oil from oxidation. The elimination of these factors is not always sufficient to conserve unaffected the organoleptical characteristics of the oil. Sanelli (1981a) studied the quality changes of dif- ferent oils (olive, soybean, linseed and mixture of olive with others) in relation to the presence or absence of chlorophyll during their storage in closed bottles in darkness or in open bottles in light or darkness. He found that chlorophyll decreased the formation of free fatty acid and changed the unsaturated/saturated fatty acid ratios during storage. In other studies, Sanelli (1981b) tried to determine the oxidative deterioration of virgin olive oil and other oils (refined olive, soybean oil and linseed) as pure and as mixtures in relation to the chlorophyll content and to degree of unsaturation. Samples were stored in closed glass bottles in light or darkness and in open bottles in light. Results demonstrated the great sensitivity of virgin olive oil to oxidation and the improved stability conferred by refining. 40 Unal (1978), in his storage studies with virgin olive oil in cans, colorless glass bottles or PVC bottles and stored in light (ambient temperature) or in dark (at 2811°C), found that free fatty acid content of the oil increased during the storage period of 24 months. He reported that peroxide values of samples in cans or glass bottles decreased during storage, whereas those of samples in PVC bottles increased, probably because of the 02 per- meability of the PVC. He further demonstrated that the decreases in tocopherol, B-carotene and chlorophyll were greater in illuminated samples than in those stored in darkness. Samples stored in glass or PVC bottles in the light underwent greater changes in organoleptic properties than those stored in darkness. Gutfinger 33 31. (1975) studied the changes in per- oxide value, TBA and acid values in samples of olive oil stored at 30°C in the dark in glass and PVC bottles and tin-coated cans. They found no significant differences between peroxide value of glass or PVC stored oil but lower peroxide values for canned oil in21160 day storage period without shaking. Continous shaking accelerated oxidation of glass and PVC packaged oils but slightly reduced peroxide value of some canned samples. According to them, only minor changes occurred in TBA and acid values. Cucurachi (1975) noted that the main factors con— cerning deterioration of olive oil quality during storage are: light, air, temperature and presence of metals. 41 During the storage time, the presence of metals such as iron and copper coming from the metallic surfaces of processing systems or storage media promote oxidation. The metals act as catalysts of oxidationlnzchanging their valence. RH + M+r1 , R. + .H + M+(n-1) or ROOH + M+n > ROO- + H+ + M+(n'1) ROOH + M+(n’1) a R0- + 'OH- + M+n (where Rqunsaturated fatty acid, M:metals) Underground cisterns or tanks are recommended for storage of oil in large quantities. Thus the oil is pro- tected from excessive changes of temperatures permitting the maintenance of an average temperature of 15°C which is ideal for good storage (Cucurachi, 1975). Material used for construction of tanks, according to Cucurachi (1975), should be physically and chemically inert to the oil in order to avoid absorption of defective odors and flavor and mixing of the oil with the substances which can cause deterioration. He noted that glass is the material with the greatest inertness and is practically unattracted by the oil. This is used widely for packaging and particularly in the commercialization of small quan- tities of oil. It has been pointed out (Cucurachi, 1975) that high acidity olive oils become transparent very quickly and deposit their impurities at the bottom of the tanks but 42 good quality oils retain the vegetable water and impurities over a long period of time and those are deposited slowly. It is important therefore that the oil be free from vege- table water (free or emulsified with the oil). Due to con- tent of lipolytic enzymes or of fermentable substances, the water becomes dangerous to quality during olive oil storage. It may cause an increase in free fatty acid con- tent and deterioration of the organoleptic characteristics of the oil due to fermentation of substances (Cucurachi, 1975). Gutierrez (1975) summarized the requirements of material used for olive oil packaging as follows: 1) Be impermeable to oil. 2) Neither toxic nor foreign materials should con- taminate the oil. 3) Guarantee quality and avoid fraud which implies the use of an inviolable closure or seal. 4) To have a protective action against oxidation changes, avoiding as far as possible the action of atmospheric oxygen, light, heat and metals, particularly those that are the most active catalysts. 5) Make its commercialization and employment easy. In other words, be resistant to impact and pre- sure, easily manipulated (opened, closed) and 6) Be economical. 43 Measurement of Lipid Oxidation Many methods have been developed for measuring lipid oxidation. The most commonly used methods are: peroxide value (PV), thiobarbituric acid (TBA) test, diene conjugation method and carbonyl test. A review of methods measuring lipid oxidation was presented by Gray (1978). Peroxide Value (PV) The peroxide value involves measurement of the prim- ary oxidation products (hydroperoxides). The results are expressed as moles of peroxides or milliequivalents of oxygen per 1000 gr of fat. For peroxide determination, the iodometric methods of Lea (1931), Wheeler (1932), AOCS (1973) are widely used. The International Olive Oil Council (COI) proposed (1966) a method for measuring peroxide value in olive oil which is very similar to the Wheeler's method (1932). All the above methods are based on the measurements of the iodine liberated from potassium iodide by the peroxides present in the oil. According to Mehlenbacher (1960), the two principal sources of error in the iodometric methods (Lea, 1931; Wheeler, 1932) are: a) the absorption of iodine at unsaturated bonds of the fatty material and b) the liberation of iodine from potassium iodide by oxygen present in the solution to be titrated. The lat— ter is often referred to as the oxygen error and leads to high results in the peroxide determination. Lea 44 (1931) attempted to eliminate this error by filling the sample tube with nitrogen at the beginning of the test and assuming that the evolution of chloroform thereafter would prevent the reentry of oxygen into the tube. Wheeler (1932) used a homogeneous solution in an attempt to eliminate the need for shaking, thereby mini- mizing the effect of oxygen. It has also been established that other possible sources in the iodometric methods include variation in weight of sample, the type and grade of solvent used, the reducing agent employed, variation in the reaction conditions such as time and temperature and the constitution and reactivity of the peroxides being titrated (Gray, 1978). Terao and Matsushita (1977) proposed a method for measuring hydroperoxides in photooxidized oil. Cadmium acetate was used and the absorbance of the color formed was measured at 350 nm. Hydroperoxide concentrations were calibrated using a standard solution of benzoyl peroxide in ethanol. Eskin and Frenkel (1976) developed a colorimetric method based on complex formation between titanium and hydroperoxides resulting in a colored complex that can be measured in a spectrophotometer at 415 nm. Another spectrophotometric method for determining hydroperoxides has been developed by Takagi 33_31. (1978). After oxida- tion of iodide to iodine with the sample for 5 minutes under an inert atmosphere, the excess of iodide ion 45 is immediately converted to a cadmium complex for protec- tion from atmospheric oxygen. The iodine is then measured at 358 or 410 nm and the peroxide value is calculated from the absorbance. Tabasago 33 31. (1979) proposed a new IR spectro- scopic method for evaluating the oxidative stability of fats and oils. IR spectra were determined and were cal- culated as ratio of a varying and a constant absorbance, ie ratio of absorbance at 3450 to those at 2850, 1740 or 1465 cm"1 and ratio of absorbance at 2850, 1740 or 1465 to that 3030 cm-1. They demonstrated that data with this method were identical with those obtained by a weighing method. Maurikos 33 31. (1972) used a new polarographic method for determining the peroxide value in virgin olive oil. The supporting electrolyte was LiCl in methanol- benzene with a dropping mercury electrode. Conjugated Diene Absorption Method Mehlenbacher (1960) indicated that the oxidation of polyunsaturated fatty acids produces peroxides and the position of the double bond shifts to a conjugated form. Conjugated linkages give rise to characteristic and intense absorption bonds within the spectral range of 200-400 nm, while the absorption of isolated double bonds within the same region is very weak. 46 He noted that this characteristic is the basis for the ultraviolet absorption method for determining the state of oxidation. Farmer and Sutton (1943) observed that the ultra- violet absorption in oxidized samples, increased propor— tionally to the uptake of oxygen and to the formation of peroxides in the early stages of oxidation. Swern (1964) reported that the absorption is so weak in case of nonconjugated and saturated materials that it cannot be used for analytical purposes. When ultraviolet absorption methods indicate the presence of small quantities of conjugated compounds in natural fats, the results must be carefully interpreted because nonconjugated polyunsaturated components may have undergone conjugation as a result of autoxidation or other mishandling (Swern, 1964). In a conjugated system, dienes absorb at 233 nm while trienes absorb at 208 nm. Thus, oxidation of poly- unsaturated fatty acids is accompanied by increased ultra- violet absorption. The magnitude of change is not readily related, however, to the degree of oxidation because the effects upon the various unsaturated fatty acids vary in quality and magnitude. Therefore the changes in the ultra- violet spectrum of a given substance can be used as rela- tive measurement of oxidation, rather than its measurement per 33. (Swern, 1961; Gray, 1978). 47 In a study of the shelf-life stability of peanut butter during long and short-term storage, Angelo 33 31. (1975) found good correlation between the spectrophoto- metric determination of conjugated diene hydroperoxides and the peroxide value determinations over four and twelve week periods of storage. They (Angelo 33 31., 1975) stated that the conjugated diene hydroperoxides (CDHP) can be used as an index of progressive staling in place of, or in addition to, the peroxide value. Accord— ing to them, the CDHP method is faster than the peroxide value method, is much simpler, requires no chemical reagent, does not depend upon chemical reaction or color development and can be conducted on smaller samples. This method has been used fairly extensively in studying the autoxidation of drying oils since the conju- gation of polyunsaturated components parallels oxygen absorption (Swern, 1964). It is applicable to the analysis of peroxides in vegetable oil products containing poly- unsaturated fatty acids (Gray, 1978). Golumbic 33_31. (1946) determined absorption curves of refined and deodorized soybean oil after exposure to visible radiation in air and in nitrogen. Samples of oil exposed to visible radiation in an atmosphere of nitrogen did not develop the maximum at 234 nm characteristics of the samples exposed to air. In a study with olive oil, Bartolomeo and Sergio (1969) observed that the primary oxidation products of this 48 oil show an absorption peak at 232 nm. For the secondary products - aldehydes, ketones, etc. - they found an absorp- tion peak at 270 nm. The ultraviolet ratio A232/A270 remains almost constant in virgin olive oils stored in the dark and decreases rapidly in virgin oils exposed to sun- light due to the rapid increase of oxidation (Jiminez and Gutierrez, 1970). Montefredine and Luciano (1968) found a quasilinear relation between the absorbance at 232 nm and the peroxide value. According to Ninis and Ninni (1968), olive oil, like all oils free of conjugated double bonds, shows a slow increase in absorbtivity at 232 and 268 nm during the induction period which is followed by a sharp, sudden increase. Ultraviolet spectrophotometric analysis can also be used to predict the thermal stability (Ninnis and Ninni, 1968 and Ninnis 33 31., 1968) and even to detect the adult- eration of olive’oil (Ninnis and Ninni, 1966) . The better region of ultraviolet spectrum for the detection of olive oil adulterations is found at 208-210 nm where the vege- table oils show a specific absorption (A210 = 56-781 which is three times higher than that of olive oils (A = 13.8-21.6). 210 Thiobarbituric Acid Test (TBA) The thiobarbituric acid (TBA) test is one of the more commonly used methods for the detection of lipid oxidation. 49 However the popularity of a method is not in itself ample proof that the method fulfills all the requirements of a reproducible technique (Gray, 1978). Early investigation by Sinnhuber 33 31. (1958) helped to clarify the nature of the colorimetric reaction that occurs during the TBA test. They proposed that the chro- magen was formed through the condensation of two molecules of TBA with one molecule of malonaldehyde. However no evidence was presented that malonaldehyde could be found in all oxidizing systems. Dahle 33 31. (1962) postulated a mechanism for the formation of malonaldehyde as a secondary product in the oxidation of polyunsaturated fatty acids. This mechanism was based on investigations which showed that no color developed for linoleate even at peroxide values of 2000 or greater, but that for fatty acids with three or more double bonds the molar yield of the TBA color increased with the degree of unsaturation. They noted that only peroxides which possessed unsaturation B, y to the peroxide group were capable of undergoing cyclization with the ultimate formation of malonaldehyde. Such peroxides could only be produced from fatty acids containing three or more double bonds. Pryor 33 31. (1976) proposed a mechanism in which malonaldehyde arises at least in part from the acid cata- lyzed, or thermal decomposition of endoperoxides (2,3- dioxanorbornane compounds). They applied the theory of 50 Dahle e_t_31. (1962) to explain the formation of the thio- barbituric acid-reactive material in a diene system and demonstrated that endoperoxides can be produced in a diene system but in a lower ratio than in a triene system. Evidence that TBA can react with compounds other than those found in oxidizing systems to produce the character- istic red pigment has been presented in the literature. Dugan (1955) reported that sucrose and some compounds in woodsmoke react with TBA to give a red color in the outer layers so that cured and smoked meats require corrections for the sugar and smoke. Baumgartner 33 31. (1975) also found that a mixture of acetaldehyde and sucrose, when subjected to the TBA test,produced a 532 nm absorbing pigment identical to that produced by malonaldehyde and TBA. Tarladgis and his co-workers (1962) considered the effect of acid, heat, and oxidizing agents on the TBA reagent. They suggested steam distillation of the product to remove the volatile constituents that were assumed to be responsible for sensorial rancidity. These workerscon- cluded that the structure of TBA was altered by acid and heat treatment as well as by the presence of peroxides and recommended that blank determinations be carried out in conjunction with the test. The TBA test may be performed in two ways, either (directly on a food product followed by extraction of the cxalored pigment, or on a portion of a steam distillate of 51 the food. Both methods have in common the use of acid and heat. Dekoning and Silk (1963) reported that they were unable to successfully apply the TBA test, in either of its forms, to determine rancidity in fish oils. The TBA test has been used by some workers (Gutierrez and Romero, 1960; Casillo, 1968) to measure oxidative rancidity in olive oil. Casillo (1968) reported that this test detects the rancidity of olive oil at a lower level than other tests (peroxide value and Kreis test). MATERIALS AND METHODS The materials and methods used will be described separately for each major component of the study. I. Effect of Harvesting Processes on Quality in Olive Oil Collection of Samples Olive fruits of the cultivar "Tsounati" one of the main cultivars of Greece were used. Commercial plastic nets were used for collection of fruits from 12 olive trees. The olive trees were numbered and divided into three groups. Trees numbered from one to four formed the first group, from five to eight the second and from nine to twelve the third group. The area under the trees was covered with a layer of plastic nets and the trees were shaken so a considerable amount of olives fell on them. A second layer of nets was placed on the fallen olives to separate them from others falling later naturally. Every two weeks, a quantity of about 0.5 Kg of olives was taken from the upper net layer, a similar quan- tity from.the lower net layer, and a third one from the tree. The fruits were placed in plastic bags which were not tied for a better aeration. Samples upon arriving at 52 53 Institute of Subtropical and Olive Trees , at Chanea, where they were stored in the refrigerator (5°C) to avoid hydrolysis . Every four samples coming from a certain group (3 groups) of trees and being of a certain type (ie directly from the tree, from the upper layer of nets, or from the lower one) were mixed thoroughly. Thus nine samples were gathered for analysis; three from each group of trees. Extraction of the Oil The olives were processed for oil extraction in a small experimental "mill” of the centrifugal type. The olive fruits were ground inamill and the resulting mixture was allowed to undergo malaxation for about 15 min. Then a quantity of warm water (35-40°C) was added for easier release. After malaxation, the mixture was centrifuged for about 5 min for the solid and liquid substances to be separated. The liquids were then put in a separatory funnel, where most of the water was removed. The remain- ing oil—water mixture was centrifuged in a LABOFUGE type centrifuge, at 5000 rpm, for 10 min. Finally, clear sam- ples of olive oil were obtained. The samples were kept in the refrigerator at 5°C until they were analyzed (3-5 days after extraction). Analysis of the Fruit The oil content of olive fruit was determined by using a oxhlet extractor and the method described by Mehlenbacher (1960) for cottonseed oil with some 54 modifications. 50 g of fruits were used in a solvent extraction which lasted 12 hrs. The solvent was driven off from the oil by heating the mixture in an oven at 85°C until constant weight was obtained. The moisture content was determined by drying the samples in an oven at 100°C to a constant weight. Analysis of the Oil Free fatty acids (FFA) were determined by the offi- cial AOCS (1974) method. Twenty eight g of oil were weighed in an Erlenmeyer flask. Fifty m1 of neutralized alcohol, and 2 mlcfifphenol- phthalein indicator (1% in 95% alcohol) were added. The sample was titrated with 0.1 NaOH until the appearance of pink color persisted for 30 seconds. The percentage of free fatty acid was calculated as oleic acid. The degree of oxidation of the oil was followed by determining the peroxide value (PV) according to the method recommended by the International Olive Oil Council (1966). Two g of oil were taken by weight from the stored samples and dissolved in 25 m1 glacial acetic acid- chloroform (3:2) solution and 1 ml saturated KI solution was added. The mixture was shaken and allowed to stay for one minute in the dark. It was then diluted with 75 ml distilled water and titrated with sodium thiosulfate (0.005 or 0.01N) using 2 ml of starch indicator. Results were reported as milliequivalents Oz/Kg oil. 55 II. Comparison of Extraction Systems The following extraction systems were compared: a) Pieralisi, b) Hillerznuic) Rapanelli (Sinolea-Decanter). The processing steps for oil extraction for the systems Pieralisi and Hiller (centrifugal) are shown in Appendix 1. Appendix 2 shows steps of the Rapanelli system. As it has been pointed out in the literature review the Rapanelli system gives two types of oil, the ”Sinolea” and the "Decanter". These two types of oil were studied separately. Fruit Collection-Oil Extraction Olive fruits from the cultivar "Koroneiki" were used for this study. The fruits were collected from the same area, put in cloth sacks and transferred by truck to the institute at Chanea. As soon as the fruits were brought in, they were mixed well and the entire mass, about 4 tons, was divided into three parts, not equal, since each system requires a different quantity of fruits for normal opera- tion. Soon after the separation, fruits were processed for oil extraction in the three systems. The experiment was repeated six times during the harvesting season. Oil-Analysis The oil was first analyzed immediately after the extraction and then periodically during storage in glass con- tainers. Three different conditions of storage were explored: 56 a) dark, b) diffused natural daylight and c) direct sunlight irradiating the samples for 4 hrs a day with the remainder of the time being stored as samples (b). In another experiment, the oil extracted by each system was stored in tin cans and subjected to analysis before and during storage. The initial analysis included: a) moisture determin- ation,b) foreign materials,c) free fatty acids (FFA) and peroxide value (PV). After storage only PV was determined. Moisture Determination The moisture determination was made by the AOCS method as described by Mehlenbacher (1960). In a 500 ml distillation flask, 175 g oil were weighed and an equal volume of toluene was added. After connecting the flask with the apparatus, specified by the AOCS, the receiver was filled with toluene by pouring it through the conden- ser until it began to overflow into the distillation flask. A piece of cotton was inserted loosely into the top of the condenser. The distillation lasted until the level of the water in the receiver remained unchanged for 30 min. Then the condenser was washed with 5 ml of toluene and the receiver was immersed in water of 25°C for about 15 min. The moisture content was determined by using the formula: Volume of water x 99.7 wt of sample (g)' % moisture = 57 Foreign Materials The foreign materials of olive oil were determined by centrifugation in a Labofuge centrifuge, at 5000 rpm, for 30 min. Chlorophyll Determination The chlorophyll content of Rapanelli oil (Sinolea- Decanter) was determined at the end of the experiment by the official AOCS (1978) method. The method is described in experimental part IV. Peroxide Value (PV) The COI (1966) method described previously (Part I) was employed for the measurement of PV (initially and during storage). Free Fatty Acid (FFA) The free fatty acid Content of the oil was determined using the official AOCS (1974) method as described pre— viously (Part I). III. Effect of Packaging and Storage Conditions on Olive Oil Quality PackaginggMaterial Plastic and glass bottles were used for this study. Colorless glass bottles were obtained from a Sprite com4 pany while polyethylene plastic bottles were obtained from an olive oil company. 58 Storage Conditions The samples of olive oil were put in the glass or in the polyethylene plastic bottles, 100 g and 900 g, res- pectively, and stored in diffused light and in direct sun- light. One-half of the samples were covered with aluminum foil to avoid passage of light through the transparent bottles. The degree of oxidation of the samples was followed by measuring peroxide value (COI, 1960). IV. Photooxidation of Olive Oil Olive Oil Virgin* olive oil obtained from the island of Crete was used for this study. The oil was extracted from fruits of the cultivar "Koroneiki", with the centrifugal Pieralisi system, and brought to the Food Science laboratory at Michigan State University in tin cans. Chlorophyll, Carotene, Tocopherol and Pheophytin Standards Chlorophyll a, a and B-carotene and d-a-tocopherol were obtained from the Sigma Chemical Company. A mixture of pheophytin a and b was purchased from ICN Pharmaceut- icals Inc. (Plainview, N.Y.). Other Chemicals Nickel-Dibutyl di Thiocarbamate was obtained from Pealtzs' Bauer Inc. (Stanford, Conn). Caffeic acid (assay 63% by titration) was purchased from Mann Research *Extraction from healthy, mature olives by mechanical means without any chemical treatment. 59 Laboratories Inc. N.Y. Boron fluoride methanol and Folin Ciocalteau reagent were acquired from the Sigma Chemical Company. All other reagents and chemicals used were re- agent grade. Bleachinnggents Activated Charcoal Darco G-60 (MCB Manufacturing Chemists Inc., Cincinati, Ohio). Tonsil optimum Extra (L.A. Salomon & Bro. Inc., Post Washington N.Y.). Florisil (60-100 Mesh)(Fisher Scientific Company). Absorptive Magnesia Sea Sorb 43 (Fisher Scientific Company). Hyflo -superce1 and infusorial earth (Fisher Scientific Company). Light Source Two Sears, 20 watt (each), cool white fluorescent tubes were used in the photooxidation studies. Analytical Techniques Total Fatty Acids Methyl esters of olive oil were prepared according to the method of Morrisson and Smith (1964) . Boron fluoride methanol was used as reagent. The oil was put in centri- fuge tubes and the reagent was added under nitrogen in a proportion of 1 ml reagent/4-16 g of oil. Tubes were closed with screw caps and heated in a heating source (PIERCE Reacti—Therm.HEATING MODULE) for 30 minutes,cooled and opened. The esters were extracted by adding 2 volumes 60 of pentane then 1 volume of water, shaking briefly and cen- trifuging until two layers were formed. The upper layer contained the esters. The methyl esters obtained as mentioned above were injected into a Hewlett Packard Gas Cromatograph 5840 A equipped with a hydrogen flame detector. A coiled stain- less steel column, 180 cm long and 2 mm i.d. packed with 15% (W/W) DEGS on 80/100 mesh chromosorb-W was used for methyl ester separation. The column oven temperature was 100°C, the injector temperature was maintained at 210°C and the detector temperature at 350°C. The nitrogen car- rier gas was adjusted to 31.3 ml/min. The flow rate of hydrogen was 18.5 ml/min. The emerging components were identified by comparing the retention time of each to those of standard mixtures of known fatty acid methyl esters. Peak areas were calculated by an electronic integrator (5840 GC Terminal Hewlett-Packard). Chlorophyll The 1eVel of chlorophyll was determined by the official AOCS method'(1973) in a Beckman DU Spectrophotometer using a 1 cm cell. Redistilled carbon tetrachloride, was used as blank. The absorbance values werenmasuredeu:630,670 and 710 nm and the calculation was made using the formula: A _A630 + A 670 *2 0.1016 L 710 Chlorophyll (ppm)== Where: A=Absorbance, L=Cell length in cm. 61 The official AOCS method (1973) was utilized for color determination in the unbleached* and bleached olive oil. Samples without any treatment were filtered through a No. 4 Genuine Whatman filter paper and the transmittance was meaSured at 460, 550, 620 and 670nm in a Beckman DU spectrophotometer using a 1 cm cell and redistilled CCl4 as a blank. B-Carotene Two g samples of oil were weighed into 70 ml test tubes with screw caps. Twenty ml of 0.7 N alcoholic KOH solution were added, and the oil was allowed to saponify for 5 min- utes in a beaker containing boiling water. The test tube was cooled at room temperature. Twenty ml of distilled water were added and the tube was shaken for 1 minute and fifteen m1 of hexane were added and the tube was shaken vigorously for two minutes. The layers were allowed to separate. The upper layer was removed with a 20 m1 pipette bulb. This extraction was repeated four times with the same amount of solvent. The hexane layers were col- lected in a round distilling flask and the solvent was evaporated in a rotary evaporator, under vacuum, at a tem- perature of 40°C. *The term unbleached olive oil will refer to initial olive oil (virgin) in the rest of this dissertation. 62 A glass column, 40x2 cm, was prepared according to the procedure described in the A.O.A.C. (1975). The pack- ing material consisted of 2 cm of glass wool at the bottom and 12 cm of a pre-prepared mixture of 1:1 Seasorb 43 (MgO) and diatomaceous earth. About 2 cm of sodium hydroxide pellets were added and finally a 1 cm layer of anhydrous sodium sulfate placed on the top, completed the packing material of the column. The packing was carried out under reduced pressure produced by a water aSpirator. Thirty ml of 9:1 hexane-acetone mixture was added to wet the column. The extract was transferred to the column with 20 ml of hexane-acetone mixture. Fifty ml of the same mixture of solvents were used to elute the beta-carotene fraction con- tained in the sample. The solvent mixture contained in the eluate was evaporated, as described before, in order to re- duce the volume to 15-20 m1. This was transferred to a 25 ml volumetric flask and brought to volume with the same mixture of solvents. To prepare the beta-carotene standard curve, 5.0 mg of beta-carotene were dissolved in 9:1 hexane-acetone in a 250 m1 volumetric flask. From this solution, 0.2, 0.4,0.6, 0.8, 1.0 and 1.2 ml were taken and placed separately into 50 ml volumetric flasks and brought to volume using the same mixture of solvents. The absorbance of these solutions was read at 450 nm in a Beckman DU Spectrophotometer in 1 cm cell against a hexane-acetone (9:1) mixture as blank. 63 a-Tocopherol The method described by Carpenter (1978) was used to determine a-tocopherol present in olive oil. Five g of unbleached olive oil was weighed into a 50 ml volumetric flask. The sample was brought to volume with a mixture of hexane-acetone 85:15 (HPLC grade). Smaller samples were used for the oxidized oil. The solution was filtered just before injection. The injection volumes were 50 pl. The chromatographic separation was performed by a Waters Associates Liquid chromatograph on a u-Porasil column. The UV detector was set at 280 nm. d-Toc0pherol solution (10.37 ppm) was used as a stan- dard for peak identification and quantitation (by peak height). Separate curves were prepared for the non- oxidized and for oxidized samples. A straight line relationship between peak height and concentration was obtained. Phenols The total phenols were determined according to the Vazquez 33 31. (1976) method as described by T. Gutfinger (1981). Ten g of oil was put in 250 m1 Erlenmeyer flask and dissolved with 50 ml hexane. The solution was extract- ed successively with three 20 ml portions of 60% aqueous methanol. The mixture was shaken each time for 2 minutes in a BURRELL WRIST ACTION SHAKER. The combined extracts were 64 dried in a vacuum rotary evaporator at 70°C. The residue was dissolved in 1 m1 methanol. The concentration of total polyphenols in the methanolic extract was estimated with Folin Ciocalteau reagent. The procedure consisted of dilu- tion of 0.1 m1 methanolic extract with 5 ml of distilled water in a 10 ml volumetric flask and addition of 0.25 ml of Folin-Ciocalteau reagent (2N). After 3 minutes, 1 ml of saturated (Ca 35%) Na2C03 was added. The content was diluted to volume with water. The absorbance was measured after 1 hour at 725 nm in a Beckman DU Spectrophotometer using reagent solution as blank. Caffeic acid served as a standard for preparing the calibration curve ranging from 0-100 ug/lO m1 assay solution. Trace Metal About 15 g of sample was charred under 250°C for 4-5 hours and then ashed at 600°C overnight. The ash was dissolved in about 1 ml 6N HCl after adding a few drops of distilled deionized water. The measurements for Fe and Mg were taken in an IL951 Atomic Absorption Spectrophotometer according to the manufacturer's guide. Ultraviolet Absorption One g of unbleached oil was accurately weighed into a 100 ml volumetric flask and brought to volume with Spectra Grade iso-octane. Measurements were taken at 232, 262, 268, 270 and 274 nm in a Beckman DU Spectrophotometer using iso-octane as a blank. 65 Peroxide Value (PV) Peroxide values were determined by the official AOCS method (1973) and were reported as milliequivalents Oz/Kg oil. One to two g of oil, accurately weighed, were taken from the beakers, after thoroughly mixing the contents, at regular hourly intervals during irradiation. The samples were dissolved with 30 ml of glacial acetic acid-chloroform 3:2 (V/V). One ml 50% (W/V) KI solution was added. The mixture in the flask was shaken and allowed to stand for exactly 1 minute. Fifty m1 of distilled water were added and titration followed with sodium thiosulfate solution (0.01 or 0.1 N). During titration the solution was mixed constantly with a Sargent magnetic stirrer. One ml of starch solution (Fisher Scientific Company) was used as an indicator. Conjugated Dienoic Acid (CDA) The official AOCS method (1973) was used for this determination. Ninety to 130 mg of oil were weighed into a 100 ml volumetric flask. About 75 ml of Spectra Grade iso-octane was added. The flasks were allowed to stand for a few minutes at room temperature and then iso-octane was added to make up to volume. The absorbance of the solutions was measured at 233 nm in a Beckman DU Spectophotometer against an iso-octane blank. To calculate the percentage conjugated dienoic acid the following formula was used: 66 Conjugated dienoic acid % = 0.84(%§ - KO) Where: Ko=0.07 (absorptivity for esters) As= observed absorbancy at 233 nm b= cell lengh in centimeters concentration in g/liter 0 ll Thiobarbituric Acid (TBA) Test The method of Sidwell 33 31. (1974) was used for this test. The TBA solution was prepared by dissolving 0.67 g of thiobarbituric acid in distilled water with the aid of heat from a steam bath. The solution was transferred to a 100 ml volumetric flask cooled and brought to volume. The TBA reagent consisted of an equal volume of TBA solution and glacial acetic acid. Three g of oil were accurately weighed into a 125 m1 Erlenmeyer flask and dis— solved with 10 m1 of CC14. Tentmlof reagent was added. The flasks were stoppered and shaken for 4 minutes in a BURRELL WRIST ACTION SHAKER. The contents of the flask were transferred to a separatory funnel and the aqueous layer was withdrawn into a test tube. The tubes were immersed in a boiling water bath for 30 minutes, then cooled and centrifuged for 4 min. in a centrifuge (E.H. SARGENT & Co). Finally the clear portion of the contents was transfered to a 1 cm cuvette. The absorbance was measured at 530 nm in a Beckman DU Spectrophotometer against distilled water as a blank. 67 Bleaching Technique Glass columns with 4.5 cm diameter were packed in the following way: About 2 cm of glass wool was placed on the bottom and then a prepared mixture consisting of 35 g of carbon, 50 g of extra Tonsil, 15 g of florisil and 25 g of hyflo supercell was added. A layer of about one cm of infusorial earth was placed on the top. The column was packed under reduced pressure, pro— duced by a water aspirator and pressed with a glass rod. About 150 m1 of hexane was percolated through the column before the oil addition. After all hexane had passed through, a mixture of 100 g oil and 150 ml hexane was added to the column. The column was washed with 200 ml of hexane to elute all the remaining oil. The eluate was collected in 500 ml filtering flasks. This was transferred into 500 ml round flasks and the sol- vent evaporated in a rotary evaporator under vacuum at 40°C. The oil yield was more than 90%. During bleaching, the columns and the collecting flasks were covered with aluminum foil and a stream of nitrogen was applied to the top. The bleached oil obtained by this technique was kept under a nitrogen atmosphere, in the refrigerator, for later use. Sample Preparation for Illumination An appropriate quantity of sample (25-50 g) of un- bleached and bleached olive oil was put in 250 m1 beakers. 68 The additives chlorophyll a, pheophytin (a + b), d- tocopherol, and Ni chelate (Nickel-Dibutyl Di Thio- carbonate), were dissolved in a mixture of 9 m1 hexane and 1 ml acetone. Stock solutions were prepared for all addi- tives except Ni chelate. Appropriate quantities of these solutions were pipetted for each usage. Illumination of Samples with Fluorescent Light Beakers containing the different samples were placed in a stainless steel (50x29x5 cm) water bath apparatus. The interior surface of the apparatus was lined with alu- minum foil. Two 20 watt fluorescent daylight tubes were suspended approximately 6 cm above the samples. The re- maining open part of the top was covered with aluminum foil to avoid any effect of outsource light. The amount (7500 Lux) of fluorescent radiation was measured with a light meter (Luna Pro by Gossen W. Germany). The apparatus was kept in a room where the tempera- ture stayed almost constant (2412°C). The temperature inside the apparatus due to the action of light was higher (3012°C). In all the experiments the radiation source was turned off for sampling at fixed time intervals. In each sampling, the position of samples was changed to obtain illumination as uniform as possible for all samples. RESULTS AND DISCUSSION 1. Effect of Harvesting Processes on Quality in Olive Oil Results concerning the hydrolytic and oxidative deterioration of olive oil before processing, mainly during the time that olive fruits remain on the collection nets, are discussed here. Statistical analysis was made for some of the data and the results are presented either in Tables or Figures . Hydrolytic Deterioration of Olive Oil Figure 2 shows the degree of hydrolysis (increase in free fatty acid content) that olive oil undergoes during the time the fruit remained either on the tree or on the nets. In this figure, line 1 represents the change in free fatty acid content of oil extracted from fruit from the lower net layer (fruit had fallen by shaking the trees). Line 2, from the same figure, shows the change in free fatty acid content of olive oil extracted from fruit col- lected from the upper net layer. Line 3 demonstrates changes in free fatty acid content in the oil from fruits collected from the tree. Results of this figure clearly demonstrate that the increase in free fatty acid content was very high for the samples on the lower net layer, less for the samples ofthe 69 70 7 6— 1 5‘ o A4P 35 l— 12 LU .— 5 U3” 0 5 LI- 2 O 2— 1- O O A A J_ 3 Oh 4 1 J 0 15 3O 45 60 lWN“E(DANS) Figure 2 Free fatty acid content of olive oil as affected by time and method of fruit collec- tion (0:10wer net layer, o:upper net layer, A:tree). 71 upper layer and even less for the samples collected directly from the trees. The increase in the free fatty acid content of the samples collected from the tree is probably due to the pre- sence of the enzyme lipase which is activated with the pro- cess of maturation (Martinez, 1975). According to Newnschuander and Michelakis (1978), even the dacus fly infestation can cause an increase in free fatty acid con- tent. The holes formed by the exit of larvae provide the starting points for fungus development during improper storage of the fruit. Fungi in turn liberate the enzyme lipase which causes hydrolytic deterioration of the oil. Martinez (1975) reported that olives collected from the tree are of better quality than the ones gathered from the ground and should be processed separately. The high degree of hydrolytic deterioration of oil extracted from samples from the lower net layer was related to "bad" fruit conditions. The starting of fungus develop- ment in these fruits was apparent by visual inspection. Fruits collected from spots where rain water remained on the surface of the ground were in worse condition. Thus olive fruits on the lower net layer, forced to fall by shaking the trees, were the ones that showed higher free fatty acid content since they remained on the nets longer, therfore underwent greater hydrolytic deterior- ation. 72 Data in Figure 2 dealing with fruits obtained from the upper layer of nets show that the free fatty acid con- tent of the oil was less than one when they remained on the nets up to a month. From that time on, free fatty acids increased and reached a value of 3.03% when this study was terminated (two months after initiation). While this oil was still considered as "virgin” (ordinary), according to quality critieria of C01 (1966), it was not the type of oil that consumers prefer. They prefer oils with free fatty acid content <1. Such oils are known as ”extra virgin." Data in Table 4 do not show any significant statis- tical difference in the change in the free fatty acid con- tent of oil, as affected by collection time, for fruits collected from the trees. Despite the fact that we did not have a significant difference in the free fatty acid content of the oil extracted from fruits collected from the tree (Table 4), we may have had some differences in the aroma constituents of the oil. According to Martinez (1975), the best aromatic characteristics of the oil exist at the optimum stage of fruit maturation which can be established visually when the fruits begin to darken. Many of the fruits collected from the trees had passed this stage of maturity. Results dealing mainly with the hydrolytic deterior- ation of olive oil, obtained from another experiment in 73 two consecutive years appear in Table 5. The same trees were used for both years. The results shown in Table 5 indicate that the rate of the increase in the free fatty acid content of the oil was not the same for the two years. The free fatty acid content of the 011, during the time that fruits re— mained on the tree, increased from 0.8% to 1.7% for the first year, while for the second it was increased from 0.2% to 1.2% during one month. Oxidative Deterioration of Olive Oil Table 6 presents data on the peroxide value of oil from the various samples. These data demonstrate that the oxidative deterioration of olive oil like the hydrolytic (Figure l) was also higher in the samples from the lower net layer than from the upper one. The tree samples had the lower peroxide value. Table 7 shows that in one month the peroxide value of oil samples from another experiment changed from 0 to 10. This value was lower than the upper limit ( 20) pro- posed by the COI (1966) for virgin olive oil. Data of Table 6 for the oil from the upper and lower net layer had passed this limit after a month. On the con- trary,in the samples collected from the tree the peroxide value was relatively low when this study was terminated (Table 6). Table 4 Free fatty acid content of olive oil as affected by time and method of fruit collection. Origin Free Fatty Acid (Z) of Collection time in days Differences Olive fruit 0 15 30 45 60 Tree 0.14 0.27 0.33 0.44 0.53 NS Upper net layer - 0.62 0.78 2.24 3.03 * Lower net layer 0.19 0.47 1.87 5.02 6.96 ** NS: P=0.05. No significant difference with collection time at *: Significant difference with collection time at P=0.05. **: P=0.05. Table 5 Highly significant difference with collection time at Free fatty acid content of olive oil as affected by collection time of fruit from the upper net layer. Collection Free Fatty Acid (Z) Time Production year (Days) 1979-80 1980-81 0 0.0 0.0 3 0.8 0.2 7 0.9 0.5 15 1.1 1.0 22 1.4 1.1 29 1.7 1.2 Table 6 Peroxide value of olive oil as affected by time and method of fruit collection. Peroxide Value (meq Oz/Kg oil) Origin of Collection time in days Differencesa olive fruit 0 15 30 45 60 Tree 5.7 5.7 .0 10.0 7.0 * Upper net layer - 10.8 .7 32.7 22.5 * Lowernet layer 6.7 11.6 .7 47.7 43.8 ** *: Significant difference with collection time at P=0.05. **: Highly significant difference with collection time at P=0.05. Table 7 Peroxide value of olive oil, as affected by col- lection time of fruit from the upper net layer. Collection time Peroxide Value (Days) (meq OZ/Kg oil) 0 0.0 3 5.0 7 8.0 15 8.5 22 9.0 29 10.0 76 These data indicate that the oxidative deterioration of olive oil during the time that fruits remain on the tree does not increase much. Probably the epidermis of olive fruits protects the oil from coming in contact with the air. During commercial harvesting, however, the skin is damaged and the oxygen of the air may interact with the oil. Oil and Moisture Changes in the Olive Fruits As is shown in Table 8,samples collected from the nets (lower-upper) contained more oil expressed (in raw weight) than the ones obtained from the tree. This is pro- bably due to loss of water from the fruits. In fact,as Table 9 shows,the moisture content of fruits regardless of the origin was almost the same initially (0 months). Two months later, however, samples collected from the nets had about one-half the moisture content of the initial. The moisture content of fruits collected from the tree did not change appreciably. Figure 3 shows that the oil content of fruits col- lected from the tree, 15 days later from the time that this experiment was initiated, started decreasing but not significantly. Martinez (1975) reported that as maturity advances the fruit increases in weight until it is fully developed. This state is maintained for a certain time and then the fruit begins to lose weight and its oil yield is affected. 77 Table 8 Oil content of fruit as affected by time and method of fruit collection. Oil Content (Z) Origin of Collection time in days Differencesa olive fruit 0 15 30 45 60 Tree 30.6 33.3 31.9 31.0 29.5 NS Upper net layer - 37.4 35.8 36.5 43.8 * Lower net layer 30.8 35.2 33.7 35.5 45.0 * NS: No significant difference with collection time at P=0.05. *: Significant difference with collection time at P=0.05. **: Highly significant difference with collection time at P=0.05. Table 9 Moisture content of olive fruit as affected by time and method of fruit collection. Origin Moisture Content (Z) of First Sampling Last Sampling Olive fruit 0 Months 2 Months Tree 41.73 43.54 Upper net layer 41.50 21.40 Lower net layer 41.65 17.33 78 pouomwwm mm menu Scum muflsuw m>flao mo unmucoo ouaumwoa pan HHO m ounwflm o (as) lNalNOD aunISIow O N O V 00 ms Ahmum3_uo .HHO "CV :23 32: d .meu coauomaaou he mp 7 iifi d CV 00 ( %)1N31NOD 'IIO 79 II. Comparison of Extraction Systems Initial Quality of Olive Oil as Affected by the Extraction System Table 10 presents results from the initial analysis of olive oil samples obtained by three different systems. cultivar "Koroneiki." during olive fruit processing The olives used were from the As soon as they were brought to the Institute at Chanea they were mixed well and were divided into three parts. One part of the fruits was processed by the Pieralisi system, another by the Hiller, and the third one by the Rapanelli (Sinolea-Decanter) system. As is shown in Table 10, the initial peroxide value in all of the oil samples obtained was less than 10. These data agree with those from previous experiments (Table 6) and show that olive oil undergoes some oxidative deterioration, but not high, before the removal of the fruits from the tree. With such low peroxide values (Table 10) we can not attribute any effect on oxidation from the extraction systems. Free fatty acid content exceed one. With such values virgin" (C01, 1966) . Moisture was lower than 0.5%. This is reported in the literature by Since Table 10 shows no the free fatty acid, peroxide of all samples did not the oil was rated as ”extra content on the other hand in agreement with the value Cucurachi (1975). significant differences in value, foreign materials and moisture content for the oil obtained by the three differ- ent systems, from six trials, we did not attribute any ef- fect of the extraction systems on the initial quality of the oil. 80 Table 10 FFA, PV, foreign materials, and moisture of olive oil obtained by 3 different extraction systems. Data of Analysis Trials Extraction System PV Foreign FFA merz/Kg materials Moisture (Z) oil (Z) (Z) Rapanelli-Sinolea 0.30 6.0 traces 0.3 A Rapanelli-Decanter 0.31 6.25 " 0.3 Pieralisi 0.31 7.0 ” 0.2 Rapanelli-Sinolea 0.19 8.0 0.2 0.4 B Rapanelli-Decanter 0.24 8.4 0.3 0.3 Pieralisi 0.15 7.8 0.2 0.3 Rapanelli-Sinolea 1.0 8.5 0.3 0.4 C Pieralisi 1.0 10.0 0.3 0.3 Pieralisi 0.29 7.0 0.2 0.3 D Hiller 0.36 7 5 0.2 0 4 Rapanelli-Sinolea 0.80 6.5 0.2 0.3 Rapanelli-Decanter 1.00 7.5 0 2 0.3 E Pieralisi 0.82 6.5 traces 0.2 Hiller 0.72 6.7 0.2 0.3 Rapanelli-Sinolea 0.60 7.0 traces 0.2 F Rapanelli-Decanter 0.70 8.0 ” 0.3 Pieralisi 0.60 7.0 " 0.2 *Olive fruits from each trial were of the same quality and were collected directly from the trees. 81 A difference was observed only in the color of the Rapanelli oils (Sinolea-Decanter). Decanter oil appeared darker than Sinolea oil due to its higher chlorophyll con— tent. Chlorophyll determination was made at the end of storage, in darkness studies. At that time the Sinolea and the Decanter oils contained 5.0 and 9.95 ppm chloro- phyll respectively. Carocci (1963) noted that the high chlorophyll content of the Decanter oil is probably the result of extended malaxation. Oxidation of Oil Obtained by the 3 Extraction Systems Some of the data obtained from this study were sta- tistically analyzed and the results are presented in Tables 11, 12, and 13. The values of the coefficients in these tables showing the increase of hydroperoxides, were higher in samples stored in light (directcn:diffused) than in darkness. The values of coefficients correspond- ing to direct light were greater than that corresponding to diffused light (Tables 12 and 13). This indicates that oxidation proceeded to a higher degree in the case of direct light. Figures 4 and 5 show the increase in peroxide value per unit time, for Pieralisi and Rapanelli (Sinolea- Decanter) oils stored in darkness and in diffused light. The slopes of the three lines for the diffused light are obviously larger than the ones corresponding to darkness. 82 Table 11 Correlation coefficient and regression equa— tion between peroxide value's and storage time in olive oil stored in darkness. Extraction Correlation Regression equation System Coefficient (r) a b* Pieralisi 0.92 8.17 0.61a Rapanelli- a Sinolea 0.76 7.13 0.91 Rapanelli- b Decanter 0.97 7.20 1.63 Pieralisi 0.87 9.74 0.738 Rapanelli- b Decanter 0.99 6.26 2.17 *Values b with different superscripts are statistically different at P=0.05. Table 12 Correlation coefficient and regression equa- tion between peroxide value's and storage time in olive oil stored in diffused light. Extraction Correlation Regression equation System Coefficient (r) a b* Pieralisi 0.87 11.10 3.39a Rapanelli- b Sinolea 0.78 16.16 5.18 Rapanelli- a Decanter 0.77 14.48 3.56 Pieralisi 0.93 18.02 2.573 Rapanelli- a Decanter 0.95 7.40 2.90 *ValueSIJWith different superscripts are statistically different at P=0.05. 83 .mo.oum um uCouowmap zaamoflumfiumum mum muawuomummsm ucmuemmwp nuwa n modam>e nm©.wm oo.m mm.o Anoucmooav Haamcmamm mqo.o mq.NH mm.o flmflamuowm cam.m «3.0m qw.o Aumuamooav Haamamnmm mwfi.m HN.A om.o Ammuocflmv HHHmcaamm mmm.e HN.H ca.c “maamumem en m ARV ucoflowmmooo Emumxm cowumavm seamwmhwmm cowumaouhou coauomuuxm .uanH uemuflp cw pououm paw mfimum%m ucouommwp %n pmcwmuno Hwo m>HHo GH oEHu mwmnoum can modam> opflxoumm coo3uon cowumsvo scammonwoh cam ucmwowmmooo coaumaouuoo ma manme 84 50 40- PV(meq 02/ Kg on) g J I- p Figure .AL 01r- m 2 3 4 TIME(MONTHS) Peroxide value of olive oil obtained by the Pieralisi and Rapanelli (Sinolea-Decanter) systems during storage in darkness. (lzPieralisi, 2:Rapanelli-Sinolea, 3:Rapanelli-Decanter) 85 fl? ()1 0 I 4O (5 m»30 E cs (3 3 \E 20 >' a. 1O l l _ mk- N J 0 1 2 4 5 3 TIME(MONTHS) Figure 5 Peroxide Value of olive oil obtained by the Pieralisi and Rapanelli (Sinolea-Decanter) systems during storage in diffused light. (1:Pieralisi, 2:Rapanelli-Sinolea, 3:Rapanelli- Decanter) 86 It is interesting that, although the Rapanelli- Sinolea oil contained a lesser amount of chlorophyll (5.0 ppm) than the Rapanelli-Decanter oil (9.9 ppm),. timalatter was oxidized to a greater degree in darkness. These results do not agree with the findings of Interesse 33 31. (1971), who reported that chlorophyll acted as an anti- oxidant in darkness. As shown in Figure 6, oil obtained by the Hiller system showed different degrees of oxidative deterioration during storage, under different light conditions. The oxidation proceeded at the lowest rate in darkness with a higher rate in diffused light and the highest rate in direct light (Figure 6). The same pattern was observed with the oils obtained by the Pieralisi system (Figure 7). However, in the case of Pieralisi oil, no great difference in the degree of oxidation between the levels direct and diffused light, was observed. Statistical analysis of the oils showed significant differences in peroxide value between samples stored in darkness and in diffused light for all three systems. Resistance to Oxidation of Olive Oil Obtained by the Rapanelli Sinclea and Rapanelli Decanter System For evaluation of the two types of oil (Sinolea- Decanter) obtained by the Rapanelli system, the resistance of the oil to oxidation during storage was studied. Only conditions of darkness and diffused light were used. 87 40 30- 6 U) i‘ o. 20-' C) U" 0 E >' E 10b 0 l 1 O 2 4 TIME(MONTHS) Figure 6 Peroxide value of olive oil obtained by the Hiller system during storage under different light conditions. (o:Dark, .:diffused light, A:direct sunlight) 88 150 1204- A 90 r- .5 U) 1‘ N o 3 6O - 3 3 > 2 a. 30 - 1 J J l 00 2 4 6 8 TIME (MONTHS) Figure 7 Peroxide value of olive oil obtained by the Pieralisi system, during storage under different light conditions. (1: Dark, 2: Diffused light, 3: Direct sunlight) 89 As shown in Table 14, significant differenceswere observed in the peroxide value between the oils. It is interesting that Rapanelli-Sinolea oil was oxidized to a lower degree than Rapanelli-Decanter 011 during storage in darkness. Under diffused light conditions, however, the reverse was observed. The different behavior of these two oils was probably due to the different composition. Rapanelli-Decanter oil contained 9.9 ppm chlorophyll, while the Rapanelli-Sinolea contained 5.0 ppm. With these chlorophyll values however, a higher degree of oxidation would be expected under light conditions, for Rapanelli- Decanter than Rapanelli-Sinolea oil. It seems that other components of the oil may play an important role. Table 14 Correlation coefficient and regression equa- tion between peroxide values and storage time in Rapanelli oils stored in darkness and diffused light. Extractionl Light Correlation Regression equation System Conditions Coefficient (r) a b* Dark 0.94 7.11 1.12a Rapanelli Sinolea Diffused b Light 0.94 10.78 4.41 Dark 0.98 6.79 1.19 Rapanelli Decanter Diffused Light 0.83 11.86 3.07 *Values b with different superscripts are statistically different at P=0.05. 90 Based on the observation that the two types of oil (Sinolea and Decanter) obtained by the Rapanelli system showed, each one, different resistance to oxidation under dark and light conditions,aanew study was initiated. In this study the photooxidation of olive oil containing added chlorophyll and carotene or certain other components was studied. Results of this study will be discussed in part IV. Oxidation of Olive Oil Obtained by Different Systems and Stored in Tin Containers This study reveals results concerning oxidation of oil samples originating from the same mass of fruits extracted with different systems. The oil was transferred into tin containers (5 Kg each) and stored at room tempera— ture. A 2 cm head space of air was left in each sample. The tins were tightly closed. Figure 8 presents data obtained from this study. After seven months of storage, only the oil of Rapanelli- Sinolea had a peroxide value less than 20 and was there- fore considered virgin (C01, 1966). Rapanelli-Decanter oil had exceeded this value after the sixth month of storage. Peroxide values of the samples when this study was terminated were found to be in the following order: Rapanelli-Decanter>Pieralisi>Hiller>Rapanelli-Sinolea. At the end of the test period only the Rapanelli-Decanter oil smelled rancid. Figure 9 shows that the oxidation rate in oils obtained by the above systems differ somehow but not much. 91 . €8.33.“ no .flmfiamuofim "4 .Houcmomnnwaaonmmmm 3 .mmaocflmnwaaocmmmm 3V .oHSumHoQEou Boon um muocwmfiuoo 8.3 nun pououm cam maoumhm cofluomuuxo ucouommwp hp poaflmuno Hfio o>HHo mo moam> opwxonom w ousmfim AmIpZOEVmE: o m v m N w o . _ _ _ _ d o A (I!06)I/zo bawma VN 92 AumHHflm "a .Hmaamgmem "m .umucmoaa-eaaocmamm "N .mmaoaam-aflamcaamm "Ho .mnsumuomaou_aoou um muoafimuaoo aw monoum pom maoumhm aowuomnuxm ucouomwwp up pocwmuno HHo o>HHo ow oDHm> mpwxouom mo mommouoaw mo momoam m ouowwm Am1w7h323952h m m v N P Q ‘ . lull - O (Izofix/zObawma VN 93 Since all the oils were obtained from the same quality of fruits but with different systems, it seems logical that the small differences in the resistance of oils to oxidation was the effect of the extraction system. Probably they affected the oil composition. In fact the procedures and the requirements for operation differ among the systems. Other systems require more, others a lesser amount of water for the malaxation of olive paste step. Among the two procedures in Rapanelli system the Rapanelli-Sinolea does not require addition of water for malaxation of paste. The Rapanelli-Decanter process, however, requires a lot of water. Montedoro 33 31. (1978) reported that the decrease of phenolic constituents during the extraction process may be explained by the solubilization effect of the vegetable water. The same worker noted that among the factors which can affect the phenolic constituents of the oil, the extraction process plays a very important role. Felici 33 31. (1979) determined lower polyphenol content in oils from centrifugal systems than from the traditional (press). Press systems require less water than centrifugal for operation. According to Fedeli (1977) most of the phenolic con- stituents present in olive fruits go into the aqueous phase as the oil is processed in the mill. A fraction 94 however remains in the oil and favors its oxidative sta- bility (Cantarelli, 1961; Notte and Romito, 1971; and Fedeli, 1977). The fact that the Rapanelli-Sinolea oil showed high- er resistance to oxidation than the Rapanelli-Decanter (Figure 8) could be attributed to the different phenolic composition of the two oils since water was added at the malaxation step only for the Decanter type. III. Effect of Different Packaging and Storage Conditions on the QualityofOlive Oil Results of the effect of different packing materials along with storage conditions (dark-diffused and direct sunlight) on quality of olive oil are discussed in this part of the present study. Six different samples of olive oil obtained during a collection season were used for storage studies. They were numbered as olive oils No. 1, No. 2, No. 3, No. 4, No. 5, and No. 6. In all the cases fruits from the cul- tivar "Koroneiki" were used. The oils No. 1-No. 4 were extracted by the Pieralisi system while the No. 5 and No.65 by the system Rapanelli-Sinolea and Rapanelli-Decanter, respectively. Figuresllland 11 present data obtained from.samples of olive oil No. 1 put in plastic bottles and stored in diffused and direct light. Five cm head space was left in each bottle. One-half of the samples in each case, were covered with aluminum foil. 95 75 60- 45. .6 m X \ N C) :g30- 5 >' O. 15" . 0 ° 0. 1 :4 11 In 0 1 2 3 4 5 TIME (MONTHS) Figure 10 Effect of diffused light on peroxide formation in olive oil stored in plastic bottles. (0: Plastic bottles, 0: Plastic bottles covered with alum. foil) 80 PV(mer2/Kg Oil) J J J 00 1 2 3 TIME (MONTHS) IbI— 01 Figure 11 Effect of direct sunlight on peroxide forma- tion in olive oil stored in plastic bottles. (0: Plastic bottles, A: Plastic bottles covered with alum. foil) 97 Under both light conditions (diffused-direct light) the samples in plastic bottles covered with aluminum foil developed lower peroxide values than those not covered. When this study was terminated, the peroxide value of samples covered with aluminum foil and stored in diffused or direct sunlight was 18 and 26, respectively. In the uncovered samples it was 65 and 75, respectively. While the difference in peroxide value between sam- ples exposed to light versus protected from light is large, the difference between samples exposed to diffused versus direct (only 4 hours/day) light is not great. The results obtained when samples were covered with aluminum foil agree with findings of others (Gutierrez and Jimenez, 1970). In similar studies with polyethylene bot- tles, they observed that samples stored in light were oxi— dized to a higher degree than those stored in darkness. The role of light in oxidation of olive oil has been studied by Pretzch (1970) and Interesse 33 31. (1971). They demonstrated that exposure of olive oil to light causes an increase in the oxidation rate. It seems that the natural pigments present in olive oil under the action of light develop an oxidizing activity (Interesse 33 31., (1977). It is interesting to note that, according to Valentinis and Romani (1960), direct sunlight in the ab- sence of air, caused a decrease in the peroxide value of olive oil during storage. 98 Figure 12 presents data from another experiment con- cerning the relative effect of plastic and glass bottles on the oxidation of olive oil No. 2 exposed to diffused light. As shown, samples in polyethylene bottles developed higher peroxide values than those in glass bottles. When this study was terminated the peroxide value of samples in plastic and glass bottles was 48 and 32, respectively. When samples were covered with aluminum foil lower peroxide value were recorded than in uncovered ones. Samples in covered glass bottles had lower peroxide values than those in covered plastic bottles after five months of storage. Our results agree with Gutierrez's (1975) findings. He reported higher peroxide values in polyethylene bottles than in glass bottles, but not much higher. It is interesting to note however that, according to Unal (1978), samples of olive oil in glass bottles showed a decrease in peroxide value during storage. On the other hand he observed an increase in peroxide value in samples stored in PVC bottles. He attributed that to the oxygen perme- ability of PVC. Table 15 contains data for olive oil stored in trans- parent polyethylene bottles and exposed to diffused light for a period of three months. Although the initial per- oxide value of the oil was relatively low (8.7), in one month this value exceeded 20. At this point thecfiJ. AHHOM .EDHm zuHB pouo>oo moauuon mmmao "I .mmauuon mmmau no .HAOM .EDHm noes pouo>oo mmauuon owummam n< .moauuon oaummam "0V .mmauuon mmmaw cw Ho owummam aw monoum Hwo o>HHo Ca coaumEuom mpfixouoo co uswfla powdmmap mo uoomwm NH muswaa AmI»ZO< Hanan. oDHm> HmHuHuH mflm%amcm owmuoum Eoumhm Hwo Hmcwm we we mo AHHO wM\No onv o5am> opfixouom puma puma Goauomuuxm .oz .oomam pew: oHuuflH fiuwz moauuon mmmaw Ga mmocxump CH munch N Mom monoum paw mEoumhm Gowuomuuxm unoumwwwp ma pmcwmuno Hwo o>HHo mo osam> opfixouom 0H manme 103 IV. Photooxidation Studies It has been noted in the literature that certain natural pigments found in oils can act as sensitizers, in the presence of light, initiating the photooxidation mechanism. This mechanism involves the formation of sing- let oxygen which then leads to peroxide development with- out the participation of free radicals (Rawls and Van Santen, 1970; Clements 33_31., 1973). According to Carlsson 3331. (1976) , chromophoric impurities such as chlorophyll are assumed to act as photo- sensitizers generating 1O2 by the transfer of excitation energy. One of the purposes of this work was to study the roles of chlorophyll, pheophytin, carotene and tocopherol, which occur naturally in olive oil in the photooxidation process of this oil. Efficiency of Bleaching Technique Because the natural pigments present in olive oil may affect the photooxidation mechanism, bleached olive oil was prepared from virgin oil and subsequently used as the basic material for the photooxidation studies. Several bleaching materials were used in various com- binations. The following mixture was found to give satis- factory results: charcoal (35 g), extra tonsil (50 g). florisil (15 g'), hyflo supercel (25 g), infusorial earth (ca 1 cm in the column). 104 The efficiency of the bleaching process was monitored by measuring the color of the oil by the AOCS (1973) spectrophotometric method. The results of bleaching exper- iments are shown in Table 17. As was expected, unbleached* olive oil had high absorbance at 460 nm, which is attributed to the presence of B-carotene and other carotenoids. Separate colorimetric determination of B-carotene, showed that unbleached olive oil contained 4.2 ppm. This value is within the range of values reported by others (Gracian, 1968). The absorbance of the oil at 670 nm was high due to the presence of chlorophyll which absorbs the maximum at that wavelength. The absorbance at 620 nm was also attri- buted to the chlorophyll. Generally the color of unbleached olive oil depends on the stage of maturity of the fruits and therefore of the kind of pigments carried with the oil during the extraction process. The zero absorbance values obtained for bleached olive oil at the four wavelengths (Table 17) indicated that the bleaching technique was very efficient in remov- ing chromophoric pigments. The bleached oil appeared tofu: completely colorless by visual inspection. *For the rest of this text the term unbleached will be used for virgin olive oil. 105 Table 17 Absorbance of unbleached and bleached olive oil at different wavelengths. Absorbance (nm) Unbleached Olive Bleached Olive Oil Oil 460 0.918 0.0 550 0.062 0.0 620 0.083 0.0 670 0.582 0.0 (Bleaching agent: charcoal + tonsil + florisil + hyflosupercel + infusorial earth). Chlorophyll Content The chlorophyll content of unbleached olive oil was found to be 5.45 ppm. This value is within the range of 0.0-9.7 ppm reported by others (Vitagliano, 1960). Felice 33 31. (1970) in their studies with tradi- tional (press) and centrifugal systems found somewhat higher chlorophyll values in the oil obtained from centri- fugal systems. The chlorophyll content of olive oil depends gener- ally on the degree of maturation. Unripe fruits contain more chlorophyll than ripe ones. The olive oil used for this study was extracted from semiripe fruits. a-Tocopherol Content The a-tocopherol content of unbleached olive oil was determined by HPLC (Carpenter, 1978) and was found to be 20.69 ppm. 106 Gracian and Arevalo (1965) identified only a- tocopherol in olive oil. They proposed that y-tocopherol, sometimes present, must be considered as a product of o- tocopherol oxidation. According to Vitagliano (1960), a- tocopherol in olive oil ranges from 12-162 ppm. Boatella (1945) however reported higher concentrations of a- tocopherol (70-150 ppm) in olive oil. Total Phenols Olive oil is considered stable to autoxidation because of the presence of phenolic constituents which vary depending on cultivation procedures and environmental fac- tors (Vazquez 33 31., 1975, 1976). Cantarelli and Montedoro (1969, 1972) observed that polyphenols extracted from olive oil inhibited rancidity in other oils. Gutfinger (1981) reported that olive oils produced commercially by mechanical processes exhibited lower sta- bility to oxidation than olive oils extracted with solvent (chloroform/methanol mixture). The higher resistance to oxidation of the solvent-extracted oils is probably due to their high polyphenol content in particular to that of ortho-diphenols which are considered as better antioxidants (Gutfinger, 1981). In this study, an effort was made to determine the total poylphenol content of the unbleached oil with the Folin-Ciocalteau reagent. Figure 13 shows the standard curve used for this determination. 107 .maoconm mo cowumcHShouop How o>udo pumpGMum ma ouswwm E: mwh ._.< moz' a. 75- O ‘ . 4 1 1 O 24 48 72 96 TIME(HOURS) Figure 14 Effect of fluorescent light on peroxide value in unbleached and bleached olive oil. (0: Un- bleached olive oil, A: Bleached olive oil, a: Bleached cover with alum. foil, A: Unbleached cover with alum. foil) 114 It is interesting to note. that,in the absence of light (samples shielded with aluminum foil), bleached olive oil was oxidized to a higher degree (PV=62) than unbleached (PV=35) although the initial peroxide value of the un- bleached sample was higher than that of bleached sample (Figure 14). These results suggest that the natural anti- oxidants (phenols and toc0pherols) present in olive oil have an effect in the absence of light but have little or no effect in the presence of light. In order to determine if chlorophyll had any anti- oxidant effect in the absence of light, an experiment was run using bleached oil with added chlorophyll. Results of this study will be discussed later. Effect of Fluorescent Light on Peroxide Formation in Bleached Olive Oil Containing Chlorophyll-a,and a and B-Carotene Samples of bleached olive oil with added chlorophyll -a or a and B-carotene in doses similar to the values determined in unbleached olive oil were illuminated with fluorescent light for 84 hrs. Results obtained from this experiment are presented in Table 20. This shows that all samples were oxidized to a high but different degree. Up to 48 hrs illumination the oxidative deterioration of samples was in the following order: Samples with 6 ppm chlorophyll > samples ‘with 4 ppm chlorophyll > samples containing no additive (control) > samples containing 4 ppm B-carotene > samples 115 Haanaouoano "nape .Hfiom answasam nua3 vmum>oo mumafimuaoo mHmEmm um. 0.000 0.0.: 0.000 0.0: 0.00_ 0.000 0.000 0.000 0.00 0.000 00 0.00 0.00 0.00 0.00 0.~0 0.0 0;: 0.00 0.00 0.000 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.05 00 0.00 0.~0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 cw. 0.00 0._N 0.0m 0.00 0.2 0.00 0.00 0.0.» 0.0 0.0~ N. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 .aoosldnv .aouandlv .aou0usdnv .uoumo: d .uouaUI0 .uoumuum ago 0000 maouucoo mu can 000 :50 000 £00 00 San 0 :25 0 :30 q 600 0 Sun 0 ~0me Houucov Ava—MW AHHo 0x\voEV msam> mvflxouom .CwEDHHH .ooN « 000 um .unwwa ucmomoposaw nuws cowumaEsHHH wcfluan .mm>0uwnnm unmummmwv wcflchucoo Hwo m>HHo vozomman aw cowumEuom mvwxouom 0~.oH0mH 116 containing 6 ppm a-carotene:>samp1es containing 6 ppm B-carotene. The same order was observed when this study was terminated at 84 hrs. As is shown in Table 20, higher peroxide values were found in bleached olive oil with addition of chlorophyll-a at the level of 6 than at 4 ppm. These results indicate that the degree of photooxidation is related to the amount of chlorophyll present. However in the case of autoxidation, according to Fedeli and Brillo (1975), there is no numerical relation between the chlorophyll concen- tration and the rate of oxidation in virgin olive oil. As in the case of chlorophyll, B-carotene showed different effects in bleached olive oil at the levels of 4 and 6 ppm. Lower peroxide values were determined when carotene was used at the higher level. Terao gt El. (1979) reported that the addition of B-carotene retarded the photooxidation of methyl linoleate efficiently. B-Carotene also efficiently inhibited the deterioration of soybean oil which was exposed to photo- irradiation in the presence of tocopherols. According to Satter and Deman (1978) the inhibitory effect of B-carotene on the photooxidation of oils may be as a filter screening out radiation of active wavelengths. B-Carotene showed a more pronounced effect in pre- venting oxidation than a-carotene at the 6 ppm level. The oxidation inhibition weakened toward the end oftfluaincuba- tion period, probably due to carotene destruction. 117 Teraoe_ta_l_. (1979) demonstrated that B-carotene disappeared rapidly during irradiation and so its inhibi- tory effect was not maintained. The effectiveness of B- carotene was prolonged in the presence of 6-tocopherol. The latter seemed to protect B-carotene from oxidation. According to Terao 33 31. (1979), tocopherols and espe- cially é-tocopherol should be added together with B- carotene, when this pigment is used as an additive for protecting the oils from photooxidation. Effect of Fluorescent Light on Peroxide Formation in Bleached Olive Oil Containing Different Levels of D-a-Tocopherol The effect of fluorescent light on the peroxide for- mation of bleached olive oil with added tocopherol is shown in Table 20. The data indicate that all samples containing D-a-tocopherol developed essentially the same perixide values as those of the control up to 48 hrs illu- mination. From then on the increase in peroxide value was lower for the samples with added tocopherol. As the addition of a-tocopherol increased (from 50 to 100 to 150 ppm) the rise in peroxide value was slightly reduced. This is at least partially consistent with the observation of Oliver 35 a1. (1944) that the upper limit of tocopherol was not reached since a prooxidant effect occurs when an excess amount is present. Vianni (1980) found that , in the absence of chlorophyll and under fluorescent light , o: and y-tocopherols both gave good protection against photooxidation of bleached soybean oil. 118 Carlsson gt 31. (1976) reported that tocopherol undergoes oxidation when exposed to 1O2 in oil solutions. According to them although a-tocopherol can quench 1O2 and prevent hydroperoxide formation it undergoes rapid peroxidation itself. Terao gt a1. (1979) found that a- tocopherol scarcely inhibited the production of mono- hydroperoxides (MHP) in photooxidation studies because it disappeared rapidly during photooxidation. Matsushita and Terao (1980) found that a-tocopherol disappeared completely after 12 hrs of irradiation. Our results showed a higher effect by a—tocopherol at the longer periodwxfilluminatiOn. Effect of Fluorescent Light on Peroxide Formation of Olive Oil Containing Chlorophyll,c1, B:Carotene, D-a- Tocopherdliand Ni-Chelate The effect of B-carotene, D, a-tocopherol and Ni- chelate, which are generally considered as singlet oxy- gen quenchers, are discussed here. Bleached olive oil containing 6 ppm added chlorophyll was used. All the quenching agents were used at the level of 100 ppm. Results of this study are presented in Table 21. Samples containing 6 ppm chlorophyll and those containing 6 ppm chlorophyll + 100 ppm D-a-tocopherol showed almost similar changes in peroxide value throughout the illumination period. Our results agree with those of Vianni (1980) . He found that bothcxand~ytocopherols were unable to protect 119 Table 21 Peroxide formation in bleached olive oil con- taining added chlorophyll and other additives during illumination with fluorescent light at 32°C 0 2°C. Peroxide Value (meq Oz/Kg oil) Treatments Illumination time (hours) 0 - 8 16 24 32 B (Control) 0 23.0 , 37.0 48.0 66.0 B + 6 ppm Chl 0 27.0 45.0 51.0 72.5 B + 6 ppm Chl covered with aluminum foil 0 4.0 7.0 8.0 8.5 B + 6 ppm Chl + 100 mg/Kg carotene B 0 5.0 16.0 26.0 57.0 B + 6 ppm Chl + 100 mg/Kg D-a- tocopherol 0 31.0 54.0 63.0 75.0 B + 6 ppm Chl + 100 mg/Kg Ni Chelate. 0 10.0 16.0 25.0 47.0 B: Bleached oil Chl: Chlorophyll 120 bleached soybean oil, to which chlorophyll was added, from oxidation in the presence of light. These tocopherols, however,were effective in preventing oxidation of bleached oil only in the absence of light. Results of this experiment do not agree with those of previous experiments where some effect of D—a—tocopherol was noted during the last hrs of illumination. Probably in the presence of chlorophyll,D-a-tocopherol was easily destroyed as it has been reported by others (Terao st 31., 1979). B-Carotene kept the peroxide value of the samples at a very low level for the first 8 hrs. Subsequently the peroxide value of these samples increased more rapidly in a way parallel to that of the samples containing Ni-chelate. The fact that B-carotene prevented peroxide forma- tion to a high degree only for the first hours of illum- ination is probably due to the destruction of this pigment. Terao gt 31. (1979) reported reduction of B-carotene after 5 hrs of irradiation. The inhibitory effect of this pigment, however, was maintained for 8 hrs in the presence of 5-tocopherol (Terao gt al., 1979) When Ni chelate was added to the system, the per- oxide value of the samples remained low. These samples had lower peroxide values than those containing 8‘ carotene or D-a-tocopherol at the end of the study period. . According to Carlsson gt a_1. (1976) nickel chelates are able to retard photooxidative 121 deterioration of unsaturated food oils by near UV and visible light while the peroxy radical scavengers (phe- nols) aren't efficient. The latter interfere with the free radical chain oxidation process (Dugan, 1976). Our results demonstrate lesser effects from nickel chelate than these observed by Carlsson 33 31. (1976), probably due to differences in systems and in the concen- trations used. With a Ni-chelate concentration of 0.24% they were able to provide some protection to olive oil from photooxidation. According to Carlsson 33 a1. (1976L the inherent absorption of these colored compounds could simply screen out the active wavelengths during irradiation and so would protect the oils solely by light absorption rather than by quenching singlet oxygen. It is interesting to note that samples containing 6 ppm chlorophyll in the absence of light (shielded with aluminum foil) developed very low peroxide values at 32 hrs illumination. The peroxide value of the samples was 8.5 when this study was terminated (Table 21). Effect of Fluorescent Light on Peroxide Formation and Conjugated Diene Formation in Olive Oil Containing Chlorophyll This study shows the effect of light on peroxide formation and formation of conjugated dienes in unbleached and bleached (with and without chlorophyll) olive oil. Fifty g samples instead of 25 g were used in this experi- ment . 122 The peroxide values obtained from this particular study are presented in Figures 15, 16 and 17. Higher peroxide values were determined in samples of bleached olive oil to which chlorophyll was added than in the con- trols (Figure 15). It should be noted however that, in unbleached olive oil containing no additives, higher peroxide value were produced finally than in bleached olive oil to which 10 ppm chlorophyll was added (Figure 15 and 16). This obser- vation led to the belief that other pigments present (like pheophytin) may play an important role in the photo- catalytic oxidation of olive oil since more chlorophyll (10 ppm) than that naturally present (5.2 ppm) was added to the bleached oil. Thus a new study was initiated in which a mixture of pheophytin a and b was added to bleached olive oil. Results of this study are discussed later. Figure 17 shows the effect of light and darkness on the oxidation of unbleached olive oil. It is apparent that unbleached olive oil is very sensitive to photo- oxidation but stable to autoxidation. These results agree with previous work (Kiritsakis §£_al., 1977) The resistance of olive oil to autoxidation is accounted for probably by the low percentage of poly- unsaturated fatty acids (Table 18) and the presence of natural antioxidants such as phenols, tocopherols and 123 200 4 160 I- O O is O) 1200- O x \ CV O o 3 I 5 > a. 80-' O O 40-' I /. O ./ 0V! 1 l 1 1 0 20 4o 60 TIME (HOURS) Figure 15 Effect of fluorescent light on peroxide value of bleached olive oil containing added chloro- phyll.f (I: Bleached olive oil + 10 ppm Chl Q: Bleached olive oil) 124 180 (70 hrs,292) I50 I20 5» U! x R 90 O U 0 3 > a 60 30 __J\_ o 1 1 1 1 1 1 0 IO 20 30 4O 50 60 70 TIME (HOURS) Figure 16 Effect of fluorescent light on the peroxide value of unbleached olive oil. (0: Unbleached olive oil, 0: Unbleached olive oil with alum. foil) 125 200 150 Pv(meqo2/Kgou) 50 O 20 4O TIME(HOURS) Figure 17 Effect of fluorescent light on peroxide value in bleached olive oil. (0: Bleached olive oil, I: Bleached olive oil + 10 ppm Chl with alum. foil, 0: Bleached olive oil with alum. foil) 126 sterols . The can-tocopherol content of the olive oil used was 20.69 ppm, while the phenol content was 120 ppm. Boskou and Morton (1976) observed some antioxidant effect by A‘s-anemasterol present in olive oil and Cantarelli and Montedoro (1979) demonstrated an antioxidant effect of the phenols present in olive oil. They observed that removal of phenols from olive oil caused its rapid oxidation. Gutfinger (1981) also reported an antioxidant effect by the phenols present in olive oil and in particular by the 0-diphenols. To study if chlorophyll exhibits antioxidant activi- ty in the absence of light, samples of bleached olive oil with and without chlorophyll added were placed in beakers covered with aluminum foil and a control sample illuminated with fluorescent light. In the absence of light, no oxidation occurred up to 70 hours in bleached olive oil, no matter whether chloro- phyll was present or not (Figure 17). Table 22 shows the increase in the percentages of conjugated dienoic acid in unbleached olive oil, bleached olive oil and bleached oil to which 10 ppm chlorophyll was added, during illumination with fluorescent light. The same Table shows the increase in the percentage of conju- gated dienoic acid in bleached olive oil containing no chlorophyll in the absence of fluorescent light. In all 127 oa.am oo.om oo.oo oo.as oo.mm oo.Hm oo.oH oo.o Hasnmopoaeo age OH + emnommam oa.NH oo.HH oo.oH om.w om.w oo.m oo.a oo.o Hm>oo 0000 .Eme 3\ 0050mmam oo.ma oo.mc oo.co oo.wm oo.w~ oo.s~ ow.m oo.o HHo w>HHo emsommam oo.os om.sm om.Hm oa.w~ oo.m~ oo.qm ow.mH om.aH ago m>flao emnummanae ON co om cs om om OH o Amusv mEHu coaumcHEDHHH Uwom UHOGmHU Uwumwfihfioo N mmaafimm .unwwa unmommuoaaw £uw3 cowumcwesaaw mafiudv Hamnmouoano wcwcflmucOo H00 m>HHo vmnomman 0cm 0m£ommancs mo vflom oaofiwww vmumwsncoo N a0 mmmmnocH NN mHLMH 128 samples an increase in the values was observed throughout the illumination period. Khan.e£ El“ (1954) observed that conjugated hydro- peroxides were formed during autoxidation of unsaturated fatty acids. They also observed that when methyl lino- leate was illuminated in the presence of chlorophyll two types of hydroperoxides (conjugated and nonconjugated) were developed. Among them only the conjugated hydroperoxides absorb at 233 nm. The observed increase in the percentage of the con- jugated diene acid in our samples was probably due to the formation of conjugated diene hydroperoxides mostly from the linoleic acid present in olive oil. The oil used as GC analysis showed contained only a small amount of linolenic acid thus we did not expect much involvement of this acid in the diene conjugated values. Since noncon- jugated linoleate hydroperoxides do not absorb at 233 nm, it seems logical that there was an accumulation of more hydroperoxides in unbleached olive oil which remained undetected. Based on findings of others (Khan St 31., 1954; Rawls and Van Santen, 1970; Frankel 35 31., 1982) that photooxidation with 1O2 involves the formation of two types of hydroperoxides, the ratio was determined: _ Peroxide Value (PV) 2 Conjugated Dienoic Acid (CDA) 129 in order to get an indication of possible involvement of singlet oxygen in our studies. It was assumed that the value of 7. CDA in our samples would be related to the presence of diene conjugated hydroperoxides. Thus in a system con- taining linoleate and chlorophyll, like olive oil, during autoxidation the ratio PV/Z CDA would remain constant since all the hydroperoxides formed would be of conjugated type (Khan EE.£l°’ 1954) and therefore all would be de- tected at 233 nm. However if this system is exposed to light, the ratio would increase, along with the presence of chlorophyll, since the nonconjugated hydroperoxides formed by singlet oxygen would not be detected at 233 nm. Our results showed that in the unbleached olive oil containing natural chlorophyll, the ratio PV/Z CDA continued increasing along with the peroxide value (Figure 18) as long as the chlorophyll color remained per- sistent. In contrast, in bleached olive oil containing added chlorophyll (10 ppm), while the peroxide value in- creased continually with incubation time, the ratio of PV/Z CDA increased up to 10 hrs and then leveled off (Figure 19). The subsequent leveling of the ratio coin- cides with the disappearance of chlorophyll as observed by visual inspection. When the light was excluded during illumination of bleached olive oil, both the peroxide value and the PV/Z CDA ratio remained rather consistent throughout the incubation. 130 9 360 PV/% CONJUG. DIENOIC ACID d N O_ 1 1 1 1 1 1 O 0 IO 20 so 40 so so 70 TIME (nouns) Figure 18 Peroxide value and PV/Z CDA ratio in unbleached olive oil illuminated with fluorescent light. (I: PV, 0: PV/Z CDA) 240 ° PV(m9002/Kg Oil) PV/%CONJUG. DIENOIC ACID 131 3 '270 A A A A A ; a 2 - ~180 ' '6 D) X \ N _ C) 0' d) 3 i '1— ‘ 9° 0 1 1 1 1 1 1 0 0 IO 20 3O 4O 50 60 70 TIME ( HOURS) Figure 19 Peroxide value and PV/Z CDA in bleached olive oil containing added chlorophyll and illumin- ated with fluorescent light. (A; PV/Z CDA, I: PV) 132 (Figure 20). This indicates that the few hydroperoxides formed during autoxidation were conjugated. Data of Figures 18 and 19 suggest that l 02 was involved in the photooxidation of bleached (with added chlorophyll) and unbleached olive oil which has naturally occurring sensitizers. Effect of Fluorescent Light on Peroxide Formation in Olive Oil Containing Chlorophyll-a and Pheophytin a and b Based on the observation that the increase in the ratio PV/Z CDA for samples with added chlorophyll happened at the early illumination period, an experiment was con- ducted using bleached olive oil to which chlorophyll or pheophytin was added. Peroxide value, diene conjugation and TBA tests (for some samples) were conducted every 2 hrs throughout the illumination period. Figure 21 demonstrates the effect of fluorescent light on peroxide formation in samples containing either chlorophyll or pheophytin and exposed to light. The same figure demonstrates the role of chlorophyll in the absence of light. Pheophytin as well as chlorophyll promoted oxidation of bleached olive oil in the presence of light. It is obvious that both molecules catalyzed olive oil photo- oxidation. Rawls and Van Santen (1970) using chlorophyll a and pheophytin (a + b) and methyl linoleate as substrate observed 133 0.8 160 0-6 t - 120 Q U < L) C) “2; ?-.‘ a O 0 o 4 ~ -so :2” 2 a g 0 U 3 32 E \ V >' :> m m 02/ iso (toieje 0—4! ‘fl‘ ‘4. ‘ilf’ 20911 Tf—————‘io 0 IO 20 30 4O 50 60 70 TIME ( HOURS) Figure 20 Peroxide value and PV/Z CDA ratio in bleached olive oil containing no additives in the absence of fluorescent light. (I: PV, g: PV/Z CDA) Figure 21 Effect of fluorescent light on the peroxide value of bleached olive oil containing added chlorophyll and pheophytin. (A: Bleached olive oil + 10 ppm pheop a + b, I: Bleached olive oil + 10 ppm Chl a, o: Bleached olive oil + 10 ppm Chl a with alum. foil, A: Bleached olive oil) 75 134 60- PV(meq O2/Kg Oil) U o I IS- l Figure 21 4 TIME(HOURS) 1 6 135 that the photooxidation products in both cases showed a very similar pattern and the results were very close. In the studies with the two pigments, pheophytin and chlorophyll, the latter exhibited a greater prooXidant effect during the first six hrs of illumination (Figure 21). From then on the samples containing pheophytin showed a greater increase in peroxide formation. The more pronounced effect of chlorophyll in the first hrs of illumination could be attributed to the Mg contribution to the energy distribution of the pigment. Both pigments have a porphyrin structure and the only important structural difference between them is that chlorophyll contains magnesium while in the case of pheophytin that element has been replaced by hydrogen as issfiunnlbelow: Chlorophyll + 2H+ > Pheophytin + Ms++ Actually chlorophyll is considered a very labile molecule. The fact that a higher oxidation effect resulted from pheophytin than from chlorophyll in the last hrs of ill- umination was probably due to chlorophyll destruction. It was obvious by visual inspection that chlorophyll was completely destroyed (disappearance of the green color) after six hrs of illumination, whereas the color of the pheophytin containing samples did not change appreciably. Rawls and Van Santen (1970) observed destructioncflfchloro- phyll during their photooxidation studies but they reported a low rate of destruction. Ramunni (1964) 136 reported that even the natural chlorophyll present in olive oil undergoes complete decomposition during storage of olive oil in the presence of light. Sastry gt 31. demonstrated that bleaching of chloro- phyll occurs during photooxidation. He proposed the fol- lowing mechanism for the bleaching of chlorophyll in the presence of visible light: Chl hV > 4: |Chl|* ———————> 3Chl 3Chl + 302 > 102 + Chl 102 + RH > ROOH ROOH hV > > ROO. + H. ROO- + Chl > ROOH + Chl' Chl° + 102 > Chl- - o2 V Chl' - 02 + H- Chl - O2 (bleached) Hydroperoxides are formed due to the action of sing— let oxygen and these may give rise to peroxy radicals on exposure to light. The peroxy radicals abstract hydrogen atoms from chlorophyll, thus disturbing its conjugated electron system. The resulting peroxy radical of chlorophyll combines with a proton and is stabilized it- self to a stable peroxide. Results concerning the role of chlorophyll in dark- ness disagree with the findings of Interesse £5 £1. (1971L who reported that chlorophylls a and b act as anti- oxidants. 137 Figure 22 shows a drop in the ratio PV/Z CDA for samples containing chlorophyll after six hrs illumination time. This was probably due, as mentioned previously, to chlorophyll destruction. Samples containing pheophytin however showed a continous increase in that ratio (PV/ Z CDA). These samples had maintained some of their ini- tial color when this study was terminated. The increase in the ratio PV/Z CDA was more pro- nounced at the first hrs of illumination (Figure 22) pro- bably due to the higher activity of the pigments. Photooxidation by 102 form isomeric hydroperoxides at each end of the double bond systems in unsaturated fatty acids. Consequently when linoleic acid, which is present in olive oil, is oxidized by 102, hydroperoxides can be formed at C9, C10, C12 or Cl with double bonds at 10- 3 11 and 12-13 for C at 8-9 and 12-13 for C 9-10 and 9’ 10’ 13-14 for C12 and at 9-10 and 11-12 for C13. It is ob- vious that the 10-OOH is B-y to the 12-13 double bond whereas the 12-OOH is B-y to the 9-10 double bond. Thus, during the photooxidation by 102 there is formation of hydro- peroxides , which can lead to the development of B-y systems . According to Dahle g£_§l. (1969) and Pryor gt a1. (1976) the hydroperoxide possessing double bonds 8, y to the peroxide group is capable of undergoing cycliza- tion and fission with ultimate formation of malonaldehyde. When TBA values were assessed along with peroxide value, the samples containing bleached olive oil and 10 HHHHOOHOHHO uEu HHO m>HHO OmsommHm uHm 138 OOO.O O.m OOH.O m.- O ONO.O m.~ OOH.O N.~N O OOO.O O.H OHO.O a.HH N OO0.0 0.0 OOO.O O.O O m «me >m HHOM .EDHN mHDOm HuHs Hno Baa OH + Hm H5O sag OH + Hm .unwwa unmommHOSHm mo monomnm mnu aw Ho moCmmmum oSu CH .HH050 Iouoano 0cflcflmucoo Hwo m>HHo vmnomman mo mmdam> coaumuomnm <0H 0cm mDHm> muwxoumm 0N manme 139 HHsnaouoHHO uH5 HHo m>HHo OmnommHm nHm OO.OO OO.OAH OO.O O0.0 O OO.Hm OO.OOH HO.O OH.m O NN.NN OO.OOH OH.O HO.H N O0.0 OO.O OO.O O0.0 O HHOH .asHm HHH3 HHom .esHm HHHz Hnu sag OH + Hm Hno sag OH + Hm HnO aOO OH + Hm . .HHO Baa OH + Hm musom <00 " Opr " >m OHHmm >m OHHmm .uswfla unwommuosam mo monomnm map c0 Ho moammmna mnu ca Hahna IouoHSU 0GHGMMuaoo 0H0 m>HHo wonomman mo <0H\>m 05m <00 N\>m mo moaumm 0N mHQMH 140 ~15 9230- U 4 2 C) Z E O - L9 3 0 Z C) U 33 \ >15*- 0. 4 O l ' . 1 1 1 O 2 4 6 TIME(HOURS) Figure 22 PV/Z CDA ratio vs time of bleached olive oil containing added chlorophyll and pheophytin. (A: Bleached, A; Bleached + 10 ppm pheop (a + b), a: Bleached + 10 ppm Chl a) 141 ppm chlorophyll showed higher peroxide values and TBA values than the samples containing the same amount of chlorophyll and covered with aluminum foil (Table 23). The ratio of PV/Z CDA and PV/TBA for the above samples are shown in Table 24. These ratios in both groups of samples were increased with the illumination time. The rate of the increase in the PV/TBA ratio as in the case of PV/Z CDA (Figure 22) was more pronounced at the first hrs of illumination (Table 24). The values of these two ratios were higher in the case of samples containing 10 ppm chlorophyll and exposed to light than the ones with the same amount of chlorophyll but protected from the light. Although the TBA values of the bleached olive oil containing 10 ppm chlorophyll and exposed to fluorescent light were higher than those of bleached olive oil con- taining the same amount of chlorophyll and protected from the light (shielded with aluminum foil), we can not say that these small differences in TBA values (Table 23) were the result of 1O2 participation. However, these results corroborate somewhat the finding from peroxide value and diene measurement which gave an indication that singlet oxygen was involved in the photooxidation of olive oil. SUMMARY AND CONCLUSIONS The relative quality of olive oil extracted from fruits collected directly from the tree and from plastic nets was studied. The effects of the extraction systems: Pieralisi, Hiller and Rapanelli (Sinolea-Decanter) and of packaging (glass-plastic) and storage conditions (dark, diffused, and direct light) were studied as well. In addition, the effect of fluorescent light on the photo- oxidation of olive oil containing either natural or added (after bleaching) substances was investigated. During the time the fruits remained on the tree, neither hydrolytic nor oxidative deterioration of the oil were noticeable. These deteriorations,however, became sig- nificant in oil extracted from fruits which remained on the collection nets longer than a month. Beyond a month, the free fatty acid content of the oil exceeded one and the peroxide value exceeded 20, thus it was no longer con- sidered as ”virgin” oil. There was no significant effect of the various pro- cessing systems on the initial quality of the oil, as evaluated by the peroxide values, free fatty acids, mois- ture and foreign materials. 142 143 Some difference in the color of Rapanelli Decanter oil was observed. It was darker than the others asixzcon- tained more chlorophyll. The peroxide values of the oil extracted by the sys- tems Pieralisi, Hiller and Rapanelli-Sinolea after seven months of storage in darkness did not differ significantly among the systems. The peroxide values of Rapanelli- Decanter oil however were higher than the others and dif- fered significantly (P=0.05). The oil obtained from the above systems was oxidized to a different degree in darkness, in diffused light and in direct sunlight. The samples stored in darkness were significantly less oxidized when compared with those stored in diffused and direct sunlight. The samples stored in diffused and in direct sunlight were not oxi- dized differently from each other. Under diffused light conditions, the oil extracted by the Rapanelli-Decanter system was oxidized to a lesser degree than that of Rapanelli-Sinolea. The reverse, however, was observed when the two oils were stored in darkness. The packaging materials influenced the olive oil sta— bility during storage in diffused light. Glass bottles provided better protection from oxidation than plastic bottles of polyethylene. Exclusion of light with aluminum foil resulted in lower peroxide values in both types of oil containers. The peroxide value of the oil in commer- cial polyethylene bottles, stored in diffused light, was 144 greater than 20 in one month. Therefore this oil was no longer considered as ”virgin" oil. Even the color of the oil was destroyed after 3 months storage probably due to chlorophyll destruction. Indeed, relatively rapid destruc- tion of the chlorophyll added to the bleached olive oil was observed in photooxidation studies. The olive oil used for photooxidation studies with fluorescent light was satisfactorily bleached with a mix- ture of different bleaching agents. Zero absorbance values were determined at 460 nm, 550 nm, 620 nm and 670 nm. The bleaching procedure reduced the peroxide value of the oil to zero. The free fatty acid content of the oil was also reduced. Unbleached olive oil was oxidized to a greater degree than the bleached olive oil in the presence of fluorescent light. In the absence of fluorescent light, however, the reverse was observed. This indicated that the natural antioxidants (tocopherols and phenols) present in olive oil had a pronounced effect in the absence of light, while they had little or no effect in the presence of light. Appar- ently photooxidation.destroys these compounds quiterapidly. At the level of 4 ppm, chlorophyll added to the bleached olive oil showed a lower prooxidative effect than at the level of 6 ppm. Therefore there is an apparent relationship between the chlorophyll concentration and the degree of photooxidaiton. Since unbleached olive oil con- tained 5.2 ppm chlorophyll and the added chlorophyll at the 145 level of 4 ppm promoted oxidation, it seems logical that the chlorophyll content of olive oil was sufficient to catalyze photooxidation. Both, chlorophyll a and pheophytin a + b, increased the oxidation rate significantly in the bleached olive oil when present at the level of 10 ppm. Higher oxidation rates were achieved with chlorophyll than with pheophytin during the first 6 hrs of incubation. The effect of pheo- phytin was more pronounced from that point on. This rever- sal in rates was probably due to chlorophyll destruction. It was obvious that after six hrs illumination time the green color of the chlorophyll containing samples faded considerably. This pigment did not seem to function as an antioxidant in darkness. B-Carotene at the concentration of 4 ppm prevented the photooxidation of bleached olive oil to a lesser degree than at the level of 6 ppm. At the same level of concentration (6 ppm» a-carotene was less effective than B-carotene in preventing photooxidation of bleached olive oil. The protective effect of both carotenes was more pronounced in the first hrs of incubation. The weakening of the effect toward the end of the incubation period was probably due to destruction of the carotene. D-a-tocopherol showed some inhibition of oxidation of bleached olive oil containing no chlorophyll after 48 hrs of incubation. As the addition of a-tocopherol in- creased (50 to 100 to 150 ppm) the rise in peroxide value 146 was slightly reduced. D-d-tocopher012hlthe presence of chlorophyll did not prevent peroxide formation in bleached olive oil illuminated with fluorescent light. This was probably due to destruction of the tocopherol by photo- oxidation. Ni-chelate provided some protection against the photooxidation of bleached olive oil containing 6 ppm added chlorophyll. At the concentration of 100 ppm, Ni— chelate afforded a greater protecting effect than B- carotene. This indicates that Ni-chelate may exhibit a greater singlet oxygen quenching effect. The values of the PV/Z CDA ratio indicated that both conjugated and nonconjugated hydroperoxides were formed when bleached olive oil containing chlorophyll or pheo- phytin was exposed to fluorescent light. Also conjugated and nonconjugated hydroperoxides probably were formed when unbleached olive oil containing natural chlorophyll was illuminated with fluorescent light. These results pro- vide additional evidence for the probable implication of singlet O2 in oxidation of olive oil when sensitizers are present. The conclusions drawn from this study are summarized as follows: 1) The oxidative and hydrolytic deteriorations of olive oil are not appreciable during the time the fruit remains on the trees. 147 2) Oil from olives which remain on collection nets longer than a month suffers oxidative and hydrolytic deter— ioration. 3) The compared extraction systems (Pieralisi, Hiller, Rapanelli-Sinolea and Rapanelli-Decanter) did not affect the initial quality of the oil appreciably. 4) The small differences in the degree of olive oil oxidation during storage in darkness indicated that all the systems, except the Rapanelli-Decanter, had similar effects on the tendency of the oil to oxidation. 5) The oxidation of olive oil proceeds slowly in darkness, more readily in diffused light and even greater in direct sunlight. 6) Glass packaging materials give better protection against oxidation than polyethylene plastic bottles during storage of olive oil in diffused light. Olive oil should not be bottled in transparent plastic bottles in order to minimize oxidative deterioration during storage. 7) The natural substances (phenols and tocopherols) present in olive oil exhibit an appreciable antioxidant effect only in the absence of light. 8) Chlorophyll did not exhibit antioxidant activity in darkness under conditions of tests carried out here. In contrast, it acted as a photosensitizer under light conditions promoting oxidation to a high degree. 148 9) Pheophytin also appears to participate in the photooxidative mechanism and to undergo less degradation than chlorophyll upon exposure to fluorescent light. This indicates that there may be a higher photooxidative effect in olive oil from pheophytin, if present, than from chlorophyll. 10) Ni-chelate exhibits a greater effect in pre- venting photooxidation of olive oil than the B-carotene does and thus may be a better quencher of singlet oxygen. 11) The presence of conjugated and nonconjugated hydroperoxides indicated that singlet (102) oxygen could have been involved in the photooxidation of olive oil, when chlorophyll or pheophytin was present. 12) It appears that the best way to avoid oxidative degradation in olive oil is protecting it from light and oxygen. APPENDICES 149 APPENDIX I STEPS IN EXTRACTION OF OLIVE OIL BY THE PIERALISI PROCESS OLIVE FRUITS I REMOVAL OF LE AVES WASHING I MILLING OF FRUITS I MALAXATION OF PASTE I ADDITION OF WATER l OLIVE CAKE CENTRIFUGATION (IN DECANTER) l VEGET. WATER OLIVE OIL WITH SOME WATER I CENTRIFUGATION —-—-> VEGET. WATER I VIRGIN OLIVE OIL 150 APPENDIX 2 STEPS IN EXTRACTION OF OLIVE OIL BY THE RAPANELLI PROCESS OLIVE FRUITS I REMOVAL OF LEAVES I WASHING I MILLING I MALAXATION OF PASTE REMOVAL OF MOST OF THE OIL PRESENT WITH “SINOLEA MACHINE ” I VIRG.OLIVE <—— CENTRIFUGATION—> VEGET. WATER OIIISINOLEA) l MALAXATION OF PASTE I ADDITION OF WATER I CENTRIFUGATION (IN DECANTER) I OLIVE WITH SOME MOISTURE I CE NTRIFUGATION oqu oII(DECANTER) VEGET. WATER LI ST OF REFERENCES LIST OF REFERENCES Amellotti, G., A. Dachetta, D. Grieco, and K. Martin. 1973. Analysis of pressed olive oils in Liguria in relation to the olive harvesting period. Riv. Ital. Sost.Grasse. 50:30 Angelo, A.S., R.L. Dry, and L.E. Brown. 1975. Compar- ison of methods for determining peroxides in pro- cessed whole peanut products. J. Am. oil Chem. Soc. 52:34. A.O.A.C. 1975. Official Methods of Analysis. 12th ed. Ass. Offic. Anal. Chem., Washington, D.C. Bartolomeo, D., and R. Sergio. 1969. Physicochemical features and acidic composition of some meridional virgin olive oils. Riv. Ital. Sost. Grasse. 46:467. Baumgartner, W.A., N. Baker, V.A. Hill, and E.T. Wright. 1975. Novel interference in thiobarbituric acid assay for lipid peroxidation. Lipids 10:309. Boatella, R. 1975. Analysis of the tocopherols of vegetable oils by gas-phase chromatography. J. Ann. Brom. 27:287. Bolland, J.L., and P. Ten Have. 1947. Kinetic studies in the chemistry of rubber and related materials IV. The inhibitory effect of hydroquinone on the thermal oxidation of ethyl linoleate. Trans. Faraday Soc. 43:201. Boskou, D. 1978. Stability of natural terpenoids in heated olive oil. Grasas y Aceites. Vol. 29(3): 193. Boskou, D. and H. Katsikas. 1979. Effect of olive oil hydrocarbons and triterpene alcohols on the stabil- ity of heated cotton seed oil. Acta Alimen. Vol. 8(3):317. Boskou, D. and D. Morton. 1975. Changes in the sterol Composition of olive oil on heating. J. Sci. Food Agric. 26:1149. 151 152 Boskou, D. and D. Morton. 1976. Effect of plant sterols on the rate of deterioration of heated oils. J. Sci. Food Agric. 27:928. Cakmak, D. 1978. Determination of optimum numerical values for yield and quality in olive oil extraction. Ege Universitesi Ziraat Fakultesi Dergisi. 15(3):l83. Cantarelli, C and G. Montedoro. 1969. Phenolic compounds in olive oil. Riv. Ital. Sost. Grasse 46:115. Cantarelli, C. and G. Montedoro. 1972. Phenolic sub- stances present in olive oil. Riv. Ital. Sost. Grasse 46:115. Carlsson, D.J., T. Suprunchuk and H.M. Wiles. 1976. Photooxidation of unsaturated oils. Effects of singlet oxygen quenchers. J. Am. Oil Chem. Soc. 53:656. Carocci, C. 1963. Ancoro Complesso Oleario. Sinolea Per l'estrazione dell' olici della olive. Inst. Sper. Per. l'olive. E' l'oleit. Imperia. Carpenter, A.P. Jr. 1978. Determination of tocopherols in vegetable oils. J. Am. Oil Chem. Soc. 55:668. Casillo, R. 1968. Quality evaluation of virgin olive oils. Thiobarbituric acid test. Riv. Ital. Sost. Grasse 45:753. Cerezal, A., Del Bario Perez, F. Gutierrez Rosales, R. Gutierrez Consalez-Quijano. 1975. Virgin olive oil conservation in storage tanks coated with epoxy resins. Grasas y Aceites 26(5):287. Chan, H.W.S. 1977. Photo-sensitized oxidation of un- saturated fatty acid methyl esters. The Identifica- tion of different pathways. J. Am. Oil Chem. Soc. 54:100. Christakis, G., M.K. Fordyce, and C.S. Kurtz. 1980. The biological and medical aspects of olive oil. IIIrd International Congress on the Biological Value of Olive Oil. Chania, Greece. Clements, A.H., R.H. Van Den Engh, D.J. Frost, K. Hoogenhout, and J.R. Nooi. 1973. Participation of singlet oxygen in photosensitized oxidation of 1.,4- dienoic systems and photooxidation of soybean oil. J. Am. Oil Chem. Soc. 50:325. 153 Coe, M.R. 1937. Photochemical studies of rancidity rate of peroxide development under constant intensity of light. Oil and Soap 14:171. Coe, M.R. 1941. Photochemical studies of rancidity. The Chlorophyll value in relation to autoxidation. Oil and Soap 18:227. Cortesi, N., A. Ponziani and E. Fedeli. 1981. Character- ization of virgin and refined olive oils by HPLC of polar components: Riv. Ital. Sost,Grasse 58(3):108. Council International Olive Oil (C01). 1966. Trade stan- dard for virgin olive oils. Document COI No. 2/66/ 15-II. Cucurachi, A. 1975. Final operations In Olive Oil Technology. Ed. by Martinez,M. Food and Agric. Org. of the United Nations. Rome. Cucurachi, A., L. Camera, F. Angerosaandbm Solinas. 1975. Erythrodiol content of "dark green" olive oil. Riv. Ital. Sost. Grasse 52(8):266. Dahle, L.R., E.G. Hill and R.T. Holman. 1962. The thio- barbituric acid reaction and the autoxidation of polyunsaturated fatty acid methyl esters. Arch. Biochem. Biophys. 98:253. Daubert, B.F. 1950. Chemical composition of soybean oil. In Soybean and Soybean Products. Ed. by Markley, K.S. Inter. Publ. Inc. New York. Vol. I. Dekoning, A.J. and M.H. Silk. 1963. The 2-thiobarbituric acid reagent for determination of oxidative rancidity of fish oils. J. Am. Oil Chem. Soc. 40:165. Dugan, L.R. Jr. 1955. Stability and rancidity. J. Am. Oil Chem. Soc. 32:605. Dugan, L.R., Jr. 1961. Development and inhibition of oxidative rancidity of foods. Food Technol. 15:10. Dugan, L.R., Jr. 1976. Lipids 1n Principles of Food Science. Ed. by Fennema, O.R. Marcel Dekker, New YorE, Part 1. Chapter 4. Eskin, N.A.M. and C. Frenkel. 1976. A simple and rapid method for assessing rancidity of oils based on the formation of hydroperoxides. J. Am. Oil Chem. Soc. 53:746. 154 Farmer, E.H. and D.A. Sutton. 1943. The course of aut- oxidation reactions in polyisoprenes and allied com- pounds. Part IV. The isolation and constitution of photochemically-formed methyl oleate peroxide. J. Chem. Soc. 48:392. Fedeli, E. 1977. Lipids of olives. Prog. Chem. Fats Other Lipids. 15:55. Fedeli, E. and A. Brillo. 1975. Mechanism of autoxidation phenomena in fats. 11 Olive oil. Riv. Ital. Sost. Grasse 52:109. Fedeli, E., A. Brillo and G. Jacini. 1973. Metals affecting the autoxidation of vegetable oils. Riv. Ital. Sost. Grasse 50:102. Fedeli, E., E. Camurati, N. Cortesi, G. Favini, V. Cirio, and G. Vita. 1974. 12th World Congress ISF, Paper 117 Milan. Fedeli, E., A. Lanzani, P. Capella, and G. Jacini. 1966. Triterpene alcohols and sterols of vegetable oils. J. Am. Oil Chem. Soc. 43:254. Felice, M. De, T. Comes, M. Catalano. 1979. Oil extrac- tion from olives by continous industrial pressing, 3-year research results. Riv. Ital. Sost. Grasse 56:361. Flath, R.A., R.R. Forrey, and D.C. Guadagni. 1973. Aroma components of olive oil. J. Agr. Food Chem. 21:948. Foote, C.S. 1968. Mechanism of photosensitized oxidation. Science 162:963. Foote, C.S. and R.W. Denny. 1968. Chemistry of singlet oxygen. VII Quenching by B-carotene. J. Am. Oil Chem. Soc. 90:6233. Foote, C.S., V.C. Chang, and R.W. Denny. 1970a. Chemistry of Singlet Oxygen. X. Carotenoid quenching parallels biological protection. J. Am. Chem. Soc. 92:5216. Foote, C.S., R.W. Denny, L. Weaver, V. Chang, and J. Peters. 1970b. Quenching of singlet oxygen. Annals N.Y. Acad. Sci. 171:139. Francesco, F. 1961. Instrumental analysis of oils and fats. III. Visible Spectrophotometry. Olearia 15:153. ‘155 Frankel, E.N., W.E. Neff and T.R. Bessler. 1979. Analysis of autoxidized fats by gas chromatography- mass spectrometry. V. Photosensitized oxidation. Lipids 14:961. Frankel, E.N., W.E. Neff, E.Selke and D. Weisleder. 1982. Photosensitized oxidation of methyl linoleate secondary and volatile thermal decomposition pro— ducts. Lipids 17:11. French, C.S. and W.O. Lundberg. 1944. The fluorescence of chlorophyll in fats in relation to rancidity. Oil and Soap 21:23. Golumbic, G., C.J. Martins and B.F. Daubert. 1946. Flavor reversion in soybean oil. 11. The effect of atmospheres of different oxygen concentrations on the flavor reversion of soybean oil. Oil and 23:360. Gracian, J. 1968. The chemistry and analysis of olive oil. In Analysis and Characterization of Oils, Fats and fat products. Ed. hy Boekenoogen, H.E. Inter. Publ. Incor. London, New York. Gracian, J. and G. Arevalo. 1965. Los tocopherols en los aceites vegetales con especial referencia a1 aceite de oliva. Grasas y Aceites 16:278. Gray, J.I. 1978. Measurement of lipid oxidation: A review. J. Am. Oil Chem. Soc. 55:539. Gutfinger, J. 1981. Polyphenols in olive oils. J. Am. Oil Chem. Soc. 58:966. Gutfinger, J. and A. Letan. 1974. Studies of unsaponifi- ables in several vegetable oils. Lipids 9:658. Gutfinger, J., O. Vatariu, M. Alter, A. Letan. 1975. Chemical characteristics of Israeli olive oils. Grasas y Aceites 26:8. Gutierrez, R.G.Q. and A.V. Romero. 1960. Estudios sobre el enranciamiento del aceite de oliva. XI. Comparacion entre las diferences pruedas para la determinacion del grado de rancidez. Grasas y Aceites 11:67. Gutierrez, R.G.Q. 1975. Bottling and canning. In Olive Oil Technology. Ed. by Martinez, M. Food and Agric. Ofg. of the Ufiited Nations. Rome. 156 Gutierrez, R.G.Q. and J.M.O. Jimenez. 1970. PackagingIIE olive oil in commercial type containers. 11. Packaging in glass, tin, polyethylene and polyvinvl chloride containers. Peroxide number evolution. Grasas y Aceites 21:217. Interesse, F.S.P., D. Ruggiero and M. Vitagliano. 1971. Autoxidation of olive oil. Effects of chlorophyll pigments. Ind. Agr. 9:318. Iverson, J.L. and D. Firestone. 1965. Fatty acid com- position of olive oil by urea fractionation liquid chromatography. J. Assoc. Anal.(fluan.48:1l91. Jimenez, O.J.M. and R.G.Q. Gutierrez. 1970. Packaging of olive oil in commercial type containers. III. Preservation in glass, tin, polyethylene, and PVC containers. Changes in absorptivity at 232 and 270 nm. Grasas y Aceites 21:329. Kaplan, M.L. 1971. "Singlet" oxygen. Chem Technol. 1:621. Khan, N.A., W.E. Tolberg, D.H. Wheeler and W.O. Lundberg. 1954. Photochemical oxidation of fatty acid esters with and without chlorophyll. Ultraviolet and infrared studies of products. J. Am. Oil Chem. Soc. 31:460. Kiritsakis, A., C.M. Stine and L.R. Dugan, Jr. 1977. Effect of selected antioxidants on the stability of virgin olive oil. Unpublished data. Koch, E. 1973. Dynamic reaction analysis of olefin oxid- ation products generated by photosensitized oxygen- ation or thermal oxidation with triphenyl phosphite ozonide in solution. Anal. Chem. Vol. 45:2120. Korycka-Dahl, M.B. and T. Richardson. 1978. Activated oxygen species and oxidation of food constituents. Crit. Rev. in Food Sci. &Nutr. 10:209. Krinsky, N.T. 1977. Singlet oxygen in biological systems. Trends Bioch. Sci. 2:35. Lea, C.H. 1931. The effect of light on the oxidation of fats. Proc. Royal Soc. London 108B:175. Lea, C.H. and R.T. Ward. 1959. Relative antioxidant activities of the seven tocopherols. J. Sci. Food Agr. 10:537. .157 Lea, C.H. 1962. The oxidative deterioration of food lipids. In Symposium on Foods, Lipids and Their Oxidation. Ed. by SchultZTlH.W., E.A. Day and R.O. Sinnhuber. AVI Westport, Conn. Leone, A.M., F. Lamparelli and E. La Notte. 1978. Sterols and terpene dialcohols in Apulian olive oil. Riv. Ital. Sost. Grasse. 55:342. Leone, A.M., E. La Notte and F. Lamparelli. 1976. The sterolic fraction of olive oil and its analytical significance. Riv. Tech. Alim. Nutr. Um. 6(4):205. Leone, A.M., E.La Notte, V.A. Liuzzi and M. Santoro. 1979. An advanced plant for olive oil production. Yield and product characteristics._ Riv. Ital. Sost. Grasse. 56:12. Lercker, G., P. Capella and P.L. Deserti. 1973. Volatile and aromatic compounds of extra virgin olive oils. Sci. Technol. Alimenti 3(5):299. Logani, M.K. and R.E. Davies. 1979 Lipid oxidation. Biological effects of antioxidants. A Review Pre- sented at the International Congress on Oilseeds and Oils. New Delhi, India. Malgorzata, B.K.D. and T. Richardson. 1978. Activated oxygen species and oxidation of food constituents. CRC Crit. Rev. in Food Sci. & Nutri. 10:209. Martinez, M.J.M. 1966. Riccerche Sull' Estrazione dell' olio d'oliva. VIII Congresso Italia no di studi sulla Sost. Grasse Imperia Santemo 20-24 Aprille. Martinez, J.M.S. 1975. Preliminary operations. In Olive Oil Technology. Ed. by Martinez,M.J.M. Food and Agric. Organ. of the United Nations. Rome. Matsushita S. and J. Terao. 1980. Singlet oxygen initiated photooxidation of unsaturated fatty acid esters and inhibitory effects of tocopherols and B-carotene. In Autoxidation in foods and biological s stems. Ed._by Michael G. Simic and Marcus Karel. genum Press New York and London. Maurikos, P., D. Gegkou and G. Iliopoulos. 1972. Polaro- graphic study of peroxides in olive oil. Prakt. Panelleniou Chem.Syned. 4th Athens. Chem Chron. l, 118. Mehlenbacher, V.C. 1960. The analysis of fats and oils. The Gerrard Press Publ. Illinois. 158 Mendoza, J.A. 1975. Milling-Malaxatrion. In Olive Oil Technology. Ed. by Martinez, M.J.M. Food and Agric. Org. of the Unit. Nations. Rome. Montedoro, G., M. Bertuccioli, and F. Anichini. 1978. Aroma analysis of virgin olive oil by head space volatiles and extraction (polyphenoles) techniques. In Flavor of Foods and Beverages. Chemistry and TechnOlogy. Ed. by Charalampous, G.and G. Inglett. Acad. Press. New York. Montedoro, G. and C. Cantarelli. 1969. Phenolic com- pounds in olive oils. Riv. Ital. Sost. Grasse 46:115. Montefredine, A. and L. Luciano. 1968. Quality character- istics of virgin Italian olive oils. Proposal of a new classification of virgin olive oils. Boll. Lab. Chim. Prov. 19:784. Morisson, W.R. and L.M. Smith. 1964. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoride-methanol. J. Lipid Res. 5:600. Nawar, W.W. 1970 The flavor of olive oil. 29th Nat. Meeting Inst. Food Tech. Chicago Ill. Newenschuander, P. and S. Michelakis. 1978. The infest- ation of dacus oleas at harvest time and its influ- ence on yield and quality of olive oil in Crete. Sonderdrukaus 86:420. Ninnis, L.N. and M.L. Ninni. 1966. L'importance pour analyse de l'huile d'olive de l'absorption dans l'ultra-violet (190 a 220 m). Rev. Fran. des Corps Gras. 60. N. 1:11. Ninnis, L.N. and M.L. Ninni. 1968. Stabilite thermique de l'huile d'olive et sa prevision par son absorption dans l'ultra-violet A 232 at A 268 nm. Rev. Fran. des Corps. Gras 15(N.7):441. Ninnis, L.N., M.L. Ninni and A. Chatoupis. 1968. Sur quelques ralations entre la reaction A l'acide nitrique et le degre d'oxydation de 1'huile d'olive. Oleagineux. 23 amnee (N.7):461. Ninnis, L., M. Ninnis, A. Chatoupis and E. Voudouris. 1969. Tocopherol content of Greek olive oil and its significance for detection of adulteration. Pract. Acad. Athenon. 37:210. Notte, E. and N. Romito. 1971. Autoxidation of olive oil Effects of polyphenols. Ind. Agr. 9:325. 159 Official and Tentative Methods of the Am. Oil Chem. Soc. 1973. Chlorophyll. CC 13d-55. Official and Tentative Methods of the Am. Oil Chem. Soc. 1973. Color. Photometric method. CC 13c-50. Official and Tentative Methods of the Am. Oil Chem. Soc. 1973c. Peroxide Value. cd 8-53. Official and Tentative Methods of the Am. Oil Chem. Soc.1973 Spectrophotometric determination of conjugated dienoic acid. Ti la-64. Official and Tentative Methods of the Am. Oil Chem. Soc. 1974. Free fatty acid. Ca 5a-40. Olias Jimenez, J.M. and R. Gutierrez Gonzales-Quijano. 1973. Canning of olive oil in commercial type con- tainers. VI. Conservation in glass, tin, poly- ethylene and polyvinyl chloride containers. Results of the organoleptic tests. Grasas y Aceites 24:79. Olias Jimenez, J.M., J. Cabtera Martin and R, Gutierrez Gonzales-Quijano. 1974. Sensorial perception ratio of the aromas of packed olive oils. Grasas y Aceites 25:34. Oliver, G.D., W.S. Singleton and A.E. Bailey. 1944. Molec- urarly distilled peanut oil antioxidants and pure a— tocopherol as stabilizing agents for fats of poor keeping quality. Oil and Soap 21:188. Petruccioli, G. 1975. Oil extraction In Olive Oil Technology. Ed by Martinez, M.J.M. Food and Agric. Organ. of the United Nations. Rome. Pretzch, G. 1970. Peroxide number. Olive Oil Studies. Prakitan-tenbriefe 16:86. Pryor, W.A., J.P. Stanley and E. Blair. 1976. A suggested mechanism for the formation of TBA-reactive material from prostaglandin-like endoperoxides. Lipids 11:370. Ramunni, A. 1964. Storage of olive oil. Ann. Foc. Soc. 30:1. Rawls, H.R. and P.J. Van Santen. 1970. A possible role for singlet oxygen in the initiation of fatty acid autoxidation. J. Am. Oil Chem. Soc. 47:121. Ross, J., A.I. Gebhart and J.E. Gerecht. 1949. The Aut- oxidation of Methyl Oleate. J. Ann Oil Chem. Soc. 26:282. 160 Sanelli, B. 1981a. Keeping quality of vegetable oils as related to their degree of unsaturation and chloro- phyll content. Note 11. Variations of acidity iodine number, and fatty acid composition. Riv. Ital. Sost. Grasse 58:184. Sanelli, B. 1981b. The keeping quality of vegetable oils in relation to the degree of unsaturation and the chlorophyll content. I. Autoxidation. Riv. Ital. Sost. Grasse 58:125. Sastry, Y.S.R., P.V. Rao and G. Lakshminarayana. 1973. Bleaching behavior of chlorophyll substrates under vacuum on exposure to visible light. Oleagineux. 28:467. Satter, A. and J.M. DeMan. 1975. Photooxidation of milk and milk products. A review. CRC Crit. Rev. in Food Sci. & Nutr. 7:13. Satter, A. and J.M. DeMan. 1976. Stability of edible oils and fats to fluorescent light irradiation. J. Am. Oil Chem. Soc. 53:473. Schenk, G.O. 1963. Photosensitization. Ind. Eng. Chem. 55:40. Sherwin, E.R. 1976. Antioxidants for vegetable oils. J. Am. Oil Chem. Soc. 53:430. Sidwell, C.G., H. Salwin, M. Benca and J.H. Michell, Jr. 1954. The use of thiobarbituric acid as a measure of fat oxidation. J. Am. Oil Chem. Soc. 31:603. Sinnhuber, R.O., T.C. Yu and Y.T. Chang. 1958. Charact- erization of the red pigment formed in the 2- thiobarbituric acid determination of oxidative ran- cidity. Food Res. 23:626. Sonntag, N.O.V. 1979. Structure and composition of fats and oils. In Bailey's Indust. Oil and Fat Prod. Ed. by Swern, D. Inter. Publ. Inc. New York. Swern, D. 1961. Primary products of olefinic autox- idation. In Autoxidation and Antioxidants. Vol. 1. Ed: Lundberg, W.D. Inter. Publ. Inc. London, New York. Swern, D. 1964. In Bailey's Industrial and Fat Products. 3rd edit. Ed. by Swern, D. Inter. Publ. Inc. London, New York. 161 Tabasago, M., K. Morikawa, S. Masugama. 1979. A new spectral method for assessing oxidative stability of oils and fats. J. of Jap. Oil Chem. Soc. 28(4):291. Takagi, T., Y. Mitsuno and M. Masumura. 1978. Determina- tion of peroxide value by the colorimetric iodine method with protection of iodide as cadmium complex. Lipids 13:147. Tappel, A.L. 1962. Hematin compounds and lipoxidase as biocatalysts. In Lipids and Their Oxidation. Ed by Schultz H.W., Day, E.A. and Sinnhuber, R.O. Avi. Publ. Co. Inc. West. CT. Tarladgis, B.G., A.M. Pearson and L.R. Dugan, Jr. 1962. The chemistry of the 2—thiobarbituric acid test for the determination of oxidative rancidity in foods. I. Some important side reactions. J. Am. Oil Chem. Soc. 39:34. Taufel, F., C. Franzke and G. Heber. 1959. The role of chlorophyll in oxidation of olefinic fats. Fette Seifen Anst. 61:1225. Terao, J. and S. Matsushita. 1977. Products formed by photosensitized oxidation of unsaturated fatty acid esters. J. Am. Oil Chem. Soc. 54:234. Terao, J., R. Yamauchi, H. Murakami and S. Matsushita. 1979. Inhibitory effects of toc0pherols and B- carotene on singlet oxygen - initiated photo- oxidation of methyl linoleate and soybean oil. J. Food Proces. and Preser. 4:79. Tiscornia, E. and G.C. Bertini. 1972. Recent analytical data on chemical composition and structure of olive oil. Riv. Ital. Sost. Grasse 49:3. Tollin, G. and G. Green. 1960. Light induced single electron transfer reactions between chlorophyll and quinones in solution. 21. Some general features of Kinetics and mechanism. Biochem. Biophys. Acta. 60:524. Unal, K. 1978. Change in quality of virgin olive oil during storage for 24 months. Ege Univ. Ziraat Fakul. Dergisi. 15(3):l99. Valentinis, G. and B. Romani. 1960. Changes of rancidity and peroxide number in edible oil. Boll. Lab. Chim. Prov. 11:351. 162 Vazquez, R.A., J.M.R. Borbolla and Alcala. 1960. The influence of light on olive oil alterations. Grasas y Aceites 11:163. Vazquez, R.A., J.C. Del Valle and J.L.M. Del Valle. 1973. Determination of total polyphenols in olive oils. Grasas y Aceites 29:350. Vazquez, R.A., J.C. Del Valle and J.L.M. Del Valle. 1975. Polyphenol content and stability of olive oils. Grasas y Aceites 26:14. Vazquez, R.A., J.C. Del Valle and J.L.M. Del Valle. 1976. Phenolic compounds in olive fruits. Poly- phenols in olive oil. Grasas y Aceites 27:185. Vianni, R. 1980. Photooxidation studies on soybean oil. Dissertation. Michigan State University. Viola, L. 1975. Proceedings of the second International Congress on the biological value of olive oil. Torremolinos, Spain. Ed. by C01. Vitagliano, M. 1960. Minor constituents of olive oil. Min. Gras.e Sap. Col. e Vern 37:136. Wheeler, D.H. 1932. Peroxide formation as a measure of autoxidative deterioration. Oil and Soap 9:89. Wiegand, B. 1978. Studies on the storage stability of fatty oils. Fehr. Diet.Pharm. Ztg 123:2333. Yoon, S.U. and D.H. Kim. 1974. Relative effectiveness of some antioxidants in a dark and a sunlight irradiated condition. Korean J. Food Sci. Tech. 1:42.