Youssef Moustakime*,Zakaria Hazzoumi,Khalid Amrani Joutei

Faculty of Science and Technology of Fez,University of Sidi Mohamed Ben Abdellah,Fez 3000,Morocco

ABSTRACT Knowing that flavored products would increase the use of olive oil by non-traditional consumers and enhance the added value of this valuable agricultural product,the virgin olive oil(VOO)was flavored with the seeds of Pimpinella anisum(Green anise)using three different methods:classic maceration,ultrasonic assisted maceration and direct addition of the essential oil(EO).These methods were compared under two main criteria:time and level of aromatization.The physico-chemical parameters and the thermal stability of flavored oils prepared by the three methods were determined by AOAC titration method and GC–MS analysis so as to compare the aromatization effect of the three methods.The trans-anethole is the major component of the EO of anise seeds as well as the indicator of the level of aromatization.GC/MS analysis results of the flavored oils showed that the diffusion of trans-anethole in the flavored oil by direct addition of EO was very important(36.3%of the total volatile fraction of the flavored oil)in comparison to the oil flavored by ultrasonic assisted maceration or classic maceration(respectively 26.59%and 23.85%).These different aromatization methods ensure an improvement in the quality of VOO with an enrichment in polyphenols estimated at 35%in the case of ultrasonic flavored oil,an increase in the content of carotenoids and chlorophylls(67%and 21%respectively)in the event of aromatization by classic maceration,and a decrease in specific absorbency at 232 nm estimated at 29%during aromatization by addition of EO as well as a decrease in the peroxide value estimated at 26%in oil flavored by classic maceration unlike in oil flavored by ultrasound which has seen an increase of around 20%.The aromatization was able to maintain the stability of the oils and its qualification as VOO with a gain in induction time in the case of treatment at 60°C estimated at 29 and 27.5 d respectively in oils flavored by addition of EO and by conventional maceration,an improvement resistance to degradation concerning K232 and K270 of all flavored oils which varied from 15 to 40 d in the case of treatment at 60°C and 3 h resistance to degradation of oils treated at 130°C for K232.Polyphenols,chlorophyll pigments and carotenoids play an important role in oxidative stability due to their antioxidant nature and their degradation during heating is very complex.All of these physico-chemical changes have increased the thermal stability of flavored oils with better resistance to oxidation of flavored oil by classic maceration in compared to oil flavored by adding EO and the oil flavored by using ultrasound.

Keywords:Aromatization Virgin olive oil Essential oil of Pimpinella anisum Classic maceration Ultrasonic assisted maceration Thermal stability Phenolic compounds

1.Introduction

The nutritional benefits of olive oil are closely related to the fatty acid composition,mainly due to the high content of oleic acid and also to the balanced proportion of saturated and polyunsaturated fatty acids.Olive oil has considerable amounts of natural antioxidants and is considered important in the prevention of many diseases[1–5].As a result,its consumption is gaining the interest of consumers mainly from northern Europe, the United States and Canada.These potential consumers are not familiar with all the applications of olive oil and may be willing to purchase ready-made olive oil preparations fortified with other ingredients related to the mediterranean diet.These flavored products would increase,on the one hand,the use of olive oil among non-traditional consumers, on the other hand,the added value of this precious agricultural product.

Aromatic plants have also been used since ancient times in food flavoring,pharmaceuticals,cosmetics and perfumery due to the presence of essential oils.Several biological activities,including antimicrobial and antioxidant properties are usually assigned to these oils or some of their components [6–8].It is well known that aromatic plants maintain the nutritional value of foods,improve the preservation qualities of food products and increase their shelf life[9–13].They are also used to enrich the taste and aroma of various foods[14].

The antioxidant activity and the aromas come mainly from the essential oil (EO) contained in these aromatic plants.The distillation of these plants leads to extracts very rich in antioxidants.Thus, different extraction methods have been developed in order to preserve the antioxidant molecules from degradation[15–19].However,flavored olive oils are usually prepared by macerating aromatic plants in oil.Using this method,the flavoring compounds are also co-extracted with others,such as natural antioxidants and pigments present in the plant,thereby altering the sensory characteristics and stability of the oil during its shelf life.One of the new methods of extracting flavors and antioxidants compounds is ultrasonic assisted extraction.

Ultrasounds are mechanical waves capable of propagating through an elastic medium.These waves temporarily displace the molecules in the medium from their original location,resulting in the formation of areas of compression and depression in the medium.When the ultrasound power is high,the areas of depression also increase.If the depression cycles are strong enough,the distance between two molecules that connects the medium could be greater than the critical molecular distance of the medium.This causes a break in the medium and the formation of what are called cavitation bubbles.The latter grow during cycles of depression and will decrease in size during cycles of compression.When the size of these bubbles reaches a critical size,they collapse during a compression cycle and this collapse generates a hot spot at very high temperature and high pressure.When this bubble collapses next to the surface of a plant matrix,for example,it is strong enough to degrade cell walls.The collapsing cavitation bubbles are capable of destroying the EO glands and therefore promoting its release into the surrounding environment.Thus,ultrasound has been used for certain flavor extractions[20–24].This extraction method provides both better yield,better quality of the extract and a gain in extraction time.The disadvantage of this extraction method is the use of chemical solvents which require purification before using the extract in the food or cosmetic industry.

The classic extraction of EO can be carried out in two different ways: the first consists in carrying out a solvent extraction before adding the extract to the food matrix of interest.This first technique poses various problems such as the use of chemical solvents, the long extraction time(2−3 h),the extraction temperature,the evaporation of the solvent and the purification of the extract.The other classic technique is steam distillation.This process also poses problems such as using large amounts of water,the plant matrix is boiled so there could be thermal damage and it is a long procedure(2–4 h).

Hamed [25] proposed the use of edible oil as an alternative solvent for the extraction of antioxidant components from natural herbs.Combining this latest extraction method and the ultrasonic technology experimented by Luque de Castro and his collaborators[26],the green anise seeds are placed directly into virgin olive oil and the mixture undergoes ultrasonic treatment.This results in a greener procedure using olive oil as a solvent(no chemical solvent), carried out at room temperature(to avoid thermal degradation) and in a considerably reduced time and cost.

The objective of this work consists,on the one hand,to produce olive oils flavored by the seeds of Pimpinella anisum(Green anise)using three different methods(classic maceration,ultrasonic assisted maceration and direct addition of EO) and to compare them under two essential criteria,time and level of aromatization.On the other hand,to evaluate the effect of these aromatization methods on the physico-chemical parameters of flavored olive oils.The thermal stability to oxidation of these flavored oils was also studied.

2.Methods

2.1.Vegetable matter

The virgin olive oil (VOO) used in this study was extracted from the olives from the Moroccan Picholine variety of Oued Amlil common urban area of the province of Taza,Morocco.

The seeds of green anise used for flavoring are of the species Pimpinella anisum cultivated in the botanical garden of the Faculty of Science and Technology of Fez, Morocco.The extraction of the EO from the seeds of green anise(200 g)is carried out by hydrodistillation in a Clevenger type apparatus.The distillation lasts three hours after the appearance of the first drop of distillate at the exit of the vapor condensation tube.The EO is dried with anhydrous sodium sulfate and stored at 4 °C in the dark before GC/MS analysis.The yield is expressed in mL per 100 g of green anise seeds.

2.2.Aromatization by classic maceration

Green anise seeds are put in VOO at a ratio of 15%(W/W)and the mixture is placed under gentle stirring at room temperature (RT).A quantity of flavored oil was taken every 24 h to evaluate the level of aromatization and to establish aromatization kinetics over time.

2.3.Aromatization by ultrasonic assisted maceration

An ultrasonic extraction reactor(Elma Transsonic TI-H,Tagent Scientific Ltd.,Germany)is used to flavor VOO by green anise seeds.The ultrasound intensity is approximately 1 W/cm2with a frequency of 25 kHz.The operation takes place at a set temperature similar to RT.The green anise seeds are put in aliquots of VOO at 15%(W/W)and are placed in the reactor.Ultrasound is applied at different times and the final flavored mixture is filtered through a coffee filter to remove traces of seeds.

2.4.Aromatization by direct addition of EO

Under gentle magnetic stirring for 24 h, the VOO(100 g)is flavored by adding the EO extracted by hydrodistillation from green anise seeds.The volume of EO used corresponds to the amount extracted from 15 g of seeds,which is estimated at 0.33 mL.

2.5.Component analysis of flavored oils by GC/MS

The samples are analyzed by a gas chromatograph(Trace GC ULTRA,Thermo Fisher Scientific,US)coupled with a mass spectrometer (Polaris Q MS with ion trap, Thermo Fisher Scientific,US)coupled with a database NIST/EPA/NIH MASS SPECTRAL LIBRARY Version 2.0 a, build Jul 2002 At the CNRST in Rabat Morocco.

2.6.Physico-chemical analysis of flavored oils

2.6.1.Determination of free fatty acids

The free fatty acid content was determined in t riplicate by the AOAC titration method [27].Seven grams of the well-mixed oil is weighed into a 250 mL flask,then 50 mL of 1%ethylene phenolphthalein is added as an indicator.The mixture was titrated with 0.1 mol/L NaOH with vigorous stirring until a permanent pale pink color appeared which persisted for at least 1 min.The free fatty acid content was calculated as a percentage of oleic acid according to the following equation:

where m is the mass of the test portion(g),N is the concentration of NaOH(mol/L),and V is the volume of NaOH(mL).

2.6.2.Determination of peroxide index and specific absorptive power

The peroxide value (PV) was determined using the AOAC method[27].Five grams of oil was weighed into a 250 mL flask.A previously prepared solution containing acetic acid and chloroform(30 mL),saturated potassium iodide(0.5 mL)and distilled water(30 mL)is added with gentle stirring.The mixture was first titrated with 0.1 mol/L Na2S2O3, then about 0.5 mL of 1% starch solution was added.The titration is continued with vigorous stirring to release all the iodine from the CHCl3layer,until the blue color disappears.PV was calculated using the following equation:

where S is the volume(mL)of Na2S2O3,and N is the concentration of Na2S2O3(mol/L).

The specific absorbance at 232 and 270 nm was determined using a UV spectrophotometer by measuring the absorbance of the 1% solution in cyclohexane at 232 and 270 nm with 1 cm passage length.

2.6.3.Measurement of chlorophyll and carotenoid contents

By following the procedure described by Mosquera et al.[28],a sample of olive oil(7.5 g)was placed in a 25 mL volumetric flask and filled to the graduated mark with cyclohexane.The chlorophyll fraction was measured by reading the optical density at 670 nm and the carotenoid fraction at 470 nm.The pigment concentration(C)was expressed using the following equations:

2.6.4.Extraction and determination of phenolic compounds(PC)

The extraction of oil polyphenols is performed with methanol according to the method of Vázquez-Roncero and his collaborators[29].

The assay of total polyphenols is based on the reduction of phosphomolybdic acid of Folin–Ciocalteu agent by polyphenols in alkaline medium[30].

2.7.Thermal stability test

The thermal stability of the flavored and control olive oil was tested at 60°C and 130°C.Heating at 60°C for a period of 60 d mimicking the procedure of accelerating storage at RT and heating at 130°C for 7 h resembling baking and frying conditions.Oil samples(70 g)were kept in equal parts in open vials in the dark in an oven.Samples are taken every 15 d for oils heated at 60 °C and after 1 h for oils heated at 130°C.The heat stability was evaluated by measuring the peroxide value (PV), specific absorption at 232 nm and 270 nm, chlorophylls, carotenoids and total phenol content.

2.8.Statistical analyses

An analysis of variance was performed for each parameter studied.The multiple comparison test averages Tukey post hoc is used to test for significant differences between treatments (at 5%).Univariate analysis was used to test for significant differences in treatment and their interaction for a single parameter.All statistical analyses were performed with IBM SPSS statistics,version 19.The results of every experiment are obtained from triplicates.

Fig.1.GC/MS chromatogram of the volatile compounds of the essential oil of green anise seeds.(For interpretation of the references to color in this figure legend,the reader is referred to the web version of this article.)

3.Results

3.1.EO composition of green anise seeds

The yield of essential oil obtained by hydrodistillation from the seeds of green anise is 2.2 mL/100 g.The chromatogram of the volatile compounds corresponding to this EO,obtained by GC/MS analysis,was presented in Fig.1 and the relative percentages of the different volatile compounds were described in Table 1.

The constituent volatile compounds of this EO have been identified and are divided into 13 compounds.These compounds represent 98.73%of the total volatile fraction.The main compounds of this EO are: anethole in both forms trans-anethole(76.16%)and cis-anethole(0.32%)and its isomer methyl chavicol (5.91%), the fenchone (6.58%),exobornyl acetate (5.42%) and camphene (2.34%).The other compounds are present in very low quantities and even in trace amounts(relative percentage<0.6%).

As described in Table 1,anethole was the main component of the volatile fraction characterizing the EO of green anise seeds.This compound was represented by these two forms cis and trans and its isomer methyl chavicol.Transanethole was the most predominant compound and constitutes 76.16%of the total volatile fraction.Therefore,this compound is used as an indicator of the level of aromatization of VOO.

Table 1 The volatile compounds of the essential oil of green anise seeds identified by GC/MS.

3.2.Aromatization by direct addition of EO

The aromatization of VOO by adding EO is a theoretically ideal enrichment method which allows the transmission of almost all of the volatile and aromatic fraction extracted from the green anise seeds to olive oil.Fig.2 showed the GC/MS chromatogram of the volatile compounds of olive oil flavored by this method at a ratio of 0.33 mL EO/100 g olive oil.Analysis of this oil identified the presence of anethole in its trans form.The report relating to this chromatogram showed that trans-anethole represents 36.3%of the total volatile fraction of flavored olive oil.

Fig.2.GC/MS chromatogram of volatile compounds of virgin olive oil flavored by direct addition of EO from green anise seeds(0.33 mL EO/100 g virgin olive oil).(For interpretation of the references to color in this figure legend,the reader is referred to the web version of this article.)

3.3.Aromatization by classic maceration

Monitoring the aromatization using the classic maceration of green anise seeds in VOO over time makes it possible to establish the kinetics of the diffusion of trans-anethole in the oily matrix(Fig.3).It could be seen from the analysis of the chromatograms of the volatile compounds of olive oils flavored by this method over time (Fig.4) that transanethole began to diffuse in the oily matrix from the 1st day of maceration by very low percentages,then increased regularly over time until reaching a maximum after 9 d of maceration,then decreased from the 10th day.On the 9th day,trans-anethole represented 23.85%of the total volatile fraction.This level of aromatization remained less important than that observed in the case of aromatization by direct addition of EO in which trans-anethole represented more than 36%of the total volatile fraction.

Fig.3.Evolution of the diffusion of trans-anethole during the classic maceration time of green anise seeds in virgin olive oil.

3.4.Aromatization by ultrasonic assisted maceration

Samples of VOO containing 15% (W/W) green anise seeds were sonicated for 5,10,15 and 20 min.The chromatograms of their volatile fractions(Fig.5)showed that after only 5 min of ultrasound treatment, trans-anethole began to diffuse in the oily matrix and in very large quantities.Furthermore, the GC/MS profile of the flavored oil using ultrasonic assisted maceration for 20 min did not show the presence of the volatile compound characteristic of green anise seeds.

Fig.6 showed the kinetics of aromatization of VOO by green anise seeds using ultrasonic assisted maceration.It could be seen that this method accelerated the aromatization very quickly so that after 15 min the level of aromatization reached its maximum with a trans-anethole content of 26.59% of the total volatile fraction.This level of aromatization was greater than that observed in the case of aromatization by classic maceration (23.85%)and less important when adding the EO of green anise seeds directly into the VOO.In addition, this compound became undetectable in the oily matrix beyond 20 min of ultrasound treatment.

3.5.Influence of the aromatization method on the physicochemical parameters of flavored oils

The study of the influence of the aromatization method on the physico-chemical parameters was carried out on flavored oils which have shown an optimal level of aromatization.Therefore, oil flavored by direct addition of EO, oil flavored by classic maceration for 9 d and flavored oil using ultrasonic assisted maceration for 15 min were the subject of this comparative study.The control was represented by VOO without aromatization.

Table 2 showed the physico-chemical changes in olive oils after aromatization by the different methods compared to VOO without aromatization.The values in Table 2 showed that whatever the aromatization process, there was a slight increase in free acidity in the oil, mainly when it came to classic maceration and maceration assisted by ultrasound.It was also observed that the three aromatization processes caused changes in the peroxide value of flavored oils;this index increased in oil flavored by ultrasonic assisted maceration,while classic maceration resulted in a significant decrease in peroxide values.The method of direct addition of EO did not show a significant influence on this index.

Fig.4.The main chromatograms by GC/MS of the volatile compounds of olive oils flavored by classic maceration of green anise seeds.(For interpretation of the references to color in this figure legend,the reader is referred to the web version of this article.)

Fig.5.GC/MS chromatograms of olive oils aromatized with green anise seeds(15%W/W)using ultrasonic assisted maceration(intensity 1 W/cm2;frequency 25 kHz).(For interpretation of the references to color in this figure legend,the reader is referred to the web version of this article.)

Furthermore,the specific coefficients K232and K270are mainly indicative,respectively,of trienes conjugation and the presence of carbonyl compounds.These two parameters are generally influenced by the aromatization of olive oil;all three methods result in a decrease in specific absorptivity at 232 nm and an increase in absorptivity at 270 nm.However,the decrease in K232observed in flavored oil by ultrasonic assisted maceration remained less significant than the other two methods, whereas the addition of EO in olive oil only resulted in slight non-significant increase in K270.However,the chlorophyll and carotenoid contents of flavored oils increase significantly using maceration whether classic or assisted by ultrasound.However,the addition of EO directly in olive oil caused only a slight modification remains without significance of these two parameters.

Fig.6.Kinetics of aromatization of VOO by green anise seeds using ultrasonic assisted maceration for 20 min.

The levels of PC naturally present in VOO generally determine the quality and stability of natural oils.The aromatization of olive oil by anise seeds improves its phenolic quality,especially when the oily matrix was treated by ultrasonic assisted maceration.In this case,the total phenol contents showed a value significantly higher than that of the control.Classic maceration leads to a significant increase in the contents of these compounds, but less than that in the previous case.The aromatization of olive oil by EO causes a slight increase,but which is even significant for this parameter.

3.6.Study of the thermal stability of flavored oils

The thermal stability of flavored oils which have shown an optimal level of aromatization of each method is a correlation between on the one hand the heating at 60°C and oil oxidation in the environmental conditions, on the other hand,between heating to 130°C and oxidation under the cooking conditions.

3.6.1.Variation of the peroxide index of flavored olive oils during thermal oxidation

Fig.7 showed the change in the peroxide number during thermal oxidation at 60°C(Fig.7a)and 130°C(Fig.7b).It was found that PV at time zero are all less than 10 meq/kg.During heating at 60°C,the oxidation rate remained stable for the first 15 d and then increased considerably afterwards to reach its maximum towards the end of the experiment (60 d), and this regardless of the aromatization method applied.

Nevertheless,we noted that the oxidation rate of the control was faster than all the other methods.In addition,the oxidation rate of oils flavored by direct addition of EO and that resulting from classic maceration exhibited much slower oxidation rates than that of the control,whereas oil flavored by ultrasonic assisted maceration has an intermediate oxidation rate.On the other hand,heating at 130°C(Fig.4b)caused a rapid increase in PV from the 1st minutes of heating,and these values reached their maximums after 7 h.Furthermore,it was observed that the oxidation rate of oils flavored by adding EO and that of oils prepared by classic maceration have greater resistance to oxidation than the other two oils.

From Fig.7,we can see that all oils have very low starting PV.The rate of oxidation then increased considerably.This first period of time is called the induction period(IP)or the induction time(IT)(the time required for a sample to reach a peroxide value of 20 meq/kg,beyond this value,the oil loses the classification of the category of VOO).Table 3 showed the induction time for the oils.During heating at 60°C,the induction periods of flavored oils by EO and by classic maceration were similar,but much longer(respectively 52 d and 50 d)compared to the induction periods of the control and the ultrasonic flavored oil.It should be noted in this case that the latter(ultrasonic flavored oil)required a longer induction period(29 d)than the control oil(23 d).This phenomenon changes during heating at 130°C since,on the contrary,it is observed that the ultrasonic aromatization has a shorter induction period than that of the control oil.Here too,the oil flavored by EO and the oil obtained by classic maceration have identical induction periods, respectively 4.25 h and 4.5 h,but longer than the other two oils.

Table 2 Influence of the aromatization method on the physico-chemical parameters of olive oils.

Table 3 Induction time for flavored olive oils of different methods.

3.6.2.Evolution of the specific extinction values (K232and K270)of flavored oils during heating

Fig.7.Variation of the peroxide value(PV)of flavored olive oils obtained by different methods during thermal oxidation.

Fig.8 showed the evolution of the specific K232and K270coefficients of oils as a function of the heating time.Before heating, all the oils (control and flavored) had values of K232which varied from 1.5 to 2.1 and values of K270which varied from 0.05 to 0.11(Table 2).During heating at 60°C,the flavored oil by classic maceration showed a slight increase in the values of the specific coefficient(K232)in the initial stages of heating up to the 45th day of treatment where we observed an accelerated increase in this coefficient.In the case of the treatment of flavored oil by EO,we saw a more or less rapid increase in the initial stages of heating(15 d),followed by stabilization during the second half of the treatment.Beyond 30 d of heating,we witnessed an accelerated increase in K232to obtain,towards the end of the treatment,a coefficient significantly higher than that of oil flavored by classic maceration.However,the oil flavored by ultrasonic maceration followed the same pattern as that of the control:there is an accelerated increase in the specific coefficient K232from the initial stages of heating followed by an accentuation after the 45th day of treatment thus recording values significantly higher both for oils flavored by classic maceration and those flavored by addition of EO.

Fig.8.Variation of the values of conjugated diene(K232 and K270)of flavored olive oils during heating.

At the same time,all the oils heated at 60°C exhibited a slight increase in the values of the K270coefficient during the first 45 d,which varied between 0.24 and 0.31.From the 50th day,there was a rapid and intense increase in the values of this coefficient in all treated oils thereby recording the maximum values at the end of heating;oils flavored by EO and those by classic maceration had values close to 1 and these values were significantly higher than the values reached in oils flavored by ultrasound.However,it was observed that at this stage,the control oils showed the lowest values of K270.

Heating at 130 °C caused a significant increase in the K232coefficient from the first hour of treatment.During heat treatment,oils flavored by adding EO and by classic maceration had lower values than those of the control and oils flavored by ultrasonic maceration.At the end of the heating(7 h),the control showed a value of the specific coefficient K232significantly higher than those of all flavored oils,equivalent values in oils flavored by classic maceration and by ultrasound,while the flavored oil by the EO had the lowest value.

Furthermore,heating the oils tested at 130°C caused a rapid and intense increase in the specific coefficient K270to values between 0.22 and 0.31, and this after only one hour of heat treatment.Then, this coefficient increased slightly for all the oils to register their maxima at the end of the treatment with higher values in the control oil and the ultrasonic flavored oil.Oil flavored by EO had the lowest values throughout the heat treatment.

3.6.3.Evolution of the chlorophyll content of flavored oils during thermal oxidation

Chlorophylls are generally present in olive oil and are responsible for their greenish color.These pigments are also important for the stability of olive oil.Fig.9 illustrated the variations in chlorophyll content in flavored olive oils during thermal oxidation.

Fig.9.Evolution of the chlorophyll contents of flavored olive oils during thermal oxidation.

The chlorophyll contents in the oils varied between 2.5 mg/kg in the control and in the oil flavored by EO and 3 mg/kg in the oil flavored by classic maceration.Ultrasonic flavored oil had a chlorophyll content of 2.9 mg/kg.These contents decreased significantly during thermal oxidation for all olive oil samples heated at 60°C and 130 °C.In the case of the treatment at 60 °C, there is a more marked decrease beyond 30 d of heating.In contrast, heating at 130 °C caused a faster and more intense decrease almost an hour after the onset of thermal oxidation.At the end of the heating, it was observed that the chlorophylls of the ultrasonic flavored oils were more resistant to thermal oxidation and have higher contents than the other oil samples.

3.6.4.Evolution of the carotenoid content of flavored oils during thermal oxidation

Fig.10 showed the behavior of the carotenoids contained in flavored oils with regard to the heating process at 60°C(Fig.10a)and 130°C(Fig.10b).The level of carotenoids in the unheated oils(at zero time)is 3.2,3.4,5.4 and 5.0 mg/kg respectively for the control,the oil flavored by EO,the oil flavored by classic maceration and that flavored by ultrasound.During heating at 60°C,the carotenoid contents decreased considerably during the first 15 d to reach between 2.0 and 2.5 mg/kg for all flavored oils.Beyond this period,the carotenoid contents stabilized more or less in all the oils recording at the end values between 1.8 and 2.0 mg/kg.

Fig.10.Evolution of the carotenoid contents of flavored olive oils during thermal oxidation.

Heating the oils at 130°C caused a very rapid decrease in the carotenoid content in oils flavored by classic maceration and those flavored by ultrasound to values close to 3.5 mg/kg,and this from the first hour of heat treatment.Unlike oils flavored by EO and control, in which the decrease in carotenoid content was relatively moderate throughout the heating period.After 7 h of heat treatment at 130°C,all the flavored oils had almost equivalent carotenoid contents estimated at 2.2 mg/kg.

3.6.5.Evolution of the total phenol contents of flavored oils during thermal oxidation

Polyphenols are compounds naturally present in olive oils and are the main compounds responsible for the stability of oils during storage and heating.Fig.11 showed the changes in the content of PC during the heat treatment of oils at 60°C(Fig.11a)and at 130°C(Fig.11b).Before heating,the amount of polyphenols present in EO flavored oils(233 mg/kg)was slightly higher than that observed in the control estimated at 217 mg/kg.On the other hand,oils flavored by classic maceration and those flavored by ultrasound have higher concentrations than the previous ones,respectively 250 and 292 mg/kg.

Fig.11.Evolution of the total phenol content of flavored olive oils during thermal oxidation.

However,it was found that the heat treatment,whether at 60°C or 130°C,resulted in a decrease in the polyphenol levels in all the oil samples to reach their minimums at the end of the heating.However, the behavior of the PC changed depending on whether the heat treatment was carried out at 60°C or at 130°C:at 60°C,the polyphenol content of ultrasonic flavored oil dropped sharply after 15 d of heating,then the decrease became slight compared to other oils,while at 130°C,4 h of heating was sufficient for the polyphenol content of ultrasonic flavored oils to drop from 292 to 114 mg/kg,then the drop became moderate.The concentration of polyphenols in oil flavored by classic maceration decreased moderately during heating at 60°C,while 2 h of heating at 130 °C was sufficient to cause a drop in the polyphenol contents from 250 to 162 mg/kg,then,the decrease in these compounds became moderate.

It should be noted that during thermal oxidation at 60°C,the oils flavored by classic maceration and by ultrasound showed significantly higher contents than those of the control and oil flavored by EO.

4.Discussion

The hydrodistillation of the green anise seeds, the subject of our study, allows an estimated EO yield of 2.2 mL/100 g to be obtained.This yield is lower than that declared by Acimovic et al.[31] which obtained an estimated content of 3.91 mL/100 g.On the other hand,the yield of EO that we obtained remains in the range determined by the work of Ullah and Honermeier [32] which stated that the green anise seeds contained about 1.5 to 5.0 mL/100 g of EO.

The composition of EO,as confirmed by several authors[31,33–35],depends on internal and external factors that influence the plant such as genetic characteristics and ecological conditions.Agricultural practices also have critical effects on the yield and oil composition in EO crops, although EO has major components that can vary widely depending on the period of ripening.However,the analysis of the chromatogram of the volatile compounds corresponding to our EO,obtained by GC/MS analysis showed the presence of 13 volatile compounds which represented 98.73% of the total volatile fraction and of which transanethole is the major component with 76.16%.As a result,the concentration of trans-anethole in the oils extracted from the seeds,object of our study,is low in comparison to other work carried out: the detailed description of all the components detected by GC/MS found in EOs seeds of green anise from a recent study[31]revealed the presence of 12 compounds which represent more than 99% of the total volatile fraction and of which trans-anethole is the most abundant compound with 96.8%.Likewise, Gende et al.[34]found that green anise seed oils were particularly rich in trans-anethole with 96.3%.However,we see that the cis form of anethole is present with 0.32%in our oil while it is in trace form in other oils.As for the anethole isomer,methyl chavicol, it is more abundant in our oil (5.91%)compared to the EO of the work of Acimovic et al.[31]and those of Özcan and Chalchat[36].According to these authors,estragola represents 0.19%and 2.4%respectively;methyl chavicol being produced exclusively by the isomerization of trans-anethole,and the reverse situation is also possible[37].

Trans-anethole therefore constitutes the majority of volatile compound in green anise seeds.This compound is used as an indicator of the aromatization of olive oil.As a result,it was used to determine the evolution of the level of olive oil aromatization by the three processes used in this work.We note first of all that ultrasound considerably increases the kinetics of aromatization.Olive oils were flavored by 15 g of Anise seeds per 100 g of VOO (15% W/W) and after 15 min of sonication, these oils already contained 26.59% of diffusible trans-anethole in their total volatile fractions(Fig.6).Under the same experimental conditions,with a classic maceration of green anise seeds(15%W/W),the diffusion of trans-anethole barely reached 23.85% of the total volatile fraction and this after 9 d of maceration(Fig.3).The passive diffusion process of EO from anise seeds to the surrounding medium was slow in the classic maceration and required 9 d to produce fully.However,this process remained less important than in the case of ultrasound.Cavitation bubbles produced by applying ultrasound in an elastic medium can explode on the surface of seeds and destroy plant cells.The EO glands exposed to the cavitation bubbles explode,releasing the essential oil into the surrounding environment.As the surrounding medium is VOO,the explosion of the glands makes the aromatization process much faster than the classic diffusion process.However,the aromatization by adding the EO of green anise seeds at a ratio of 0.33 mL/100 g of VOO(volume of EO extracted from 15 g of seeds)showed a level of very important aromatization estimated at a fraction of trans-anethole representing 36.3%of the total volatile fraction of this flavored oil.

The addition of green anise seeds to VOO results in a slight increase in the free acidity in the oil.In all cases,all the acidity values are lower than the limits set by European regulation 2568/91 for VOO(free acidity ≤2.0%).However,this method causes changes in the physico-chemical parameters of the oil.These changes vary depending on the aromatization process used.The maceration of green anise seeds in VOO,whether classic or assisted by ultrasound,results in significant changes in all the physico-chemical parameters studied.Unlike aromatization by direct addition of EO,where it is found that these parameters change very little compared to the control oil.Ayadi et al.[38]and Zouari et al.[39]attributed all the physico-chemical changes that could be caused by the aromatization of olive oil to the migration of particular compounds from aromatic plants to the oily matrix.These particular compounds can be organic acids,PC,pigments,antioxidants,essential oils,etc.

The change in these physico-chemical parameters of flavored oils could explain the differences observed in the resistance of flavored olive oils to thermal oxidation.In fact,all the flavored oil samples exhibit prior peroxide indices(before heat treatment)of less than 10 meq/kg.These values are lower than the maximum allowed by European regulations for their classification as virgin oils (20 meq/kg).During heating at 60°C,oils aromatized by adding EO and those by classic maceration resist thermal oxidation since their induction time exceeds 50 d compared to the control and the ultrasonic flavored oil which show ITs estimated at 23 and 29.5 d respectively.During the treatment at 130°C,thermal decomposition occurs in oils flavored by ultrasound before the control,while those flavored by classic maceration and those by addition of EO, thermal oxidation are late.Several authors have estimated the induction period for the oxidation of olive oil as the time required to reach a PV value of 70 meq/kg[3,40,41].In this study,the induction time for each flavored olive oil was estimated as the time required for a sample to reach a peroxide value of 20 meq/kg.When this value exceeds the maximum authorized limit,olive oil consequently loses its classification in the category“VOO”.On the other hand,we can conclude that despite the limited differences in their peroxide and acidity values,flavored oils show different deterioration patterns and oxidation kinetics.In fact,the control oil shows less thermal stability than the ultrasonic flavored oil,followed by the flavored oil by classic maceration and finally the most stable oil is the flavored olive oil by direct addition of the EO.The evolution of the peroxide index of these flavored VOOs could be explained by changes during the oxidation process,reaching this maximum value due to the formation of hydroperoxides.The results obtained are consistent with those previously reported[3,38,42–44].

The specific extinction coefficients at K232and K270are mainly indicative,respectively,of the conjugation of trienes and the presence of carbonyl compounds.They reflect the oxidized state of the oil.The higher its extinction at 232 nm,the more peroxidized it is.Similarly,the higher the extinction at 270 nm,the richer the oil in secondary oxidation products and its poor shelf life[40,45].Their maximum values authorized by the International Olive Council(2011) are 2.60 and 0.25 respectively for K232and K270for VOO.As a result,we find when heating at 60°C,oils flavored by addition of EO and those flavored by classic maceration have a higher resistance to degradation compared to ultrasonic flavored oils and control oil samples since these oils lost their name of VOO respectively after 55,45,30 and 15 d for K232.While this classification is lost for the K270after 30 d of treatment at 60 °C for the control and beyond 45 d for the flavored oils.In addition,the acceleration of the degradation of flavored and control oils after heating for one hour at 130°C indicated the sensitivity of olive oil even after aromatization.As a result,K270exceeds the standards authorized by the International Olive Council(2015)for all oils at most after 2 h of heat treatment and lose their classification as VOO.The same is true for K232in the case of oil flavored by ultrasonic assisted maceration and control oil which lose their classification as VOO after 2 h of heating at 130 °C, whereas oils flavored by classic maceration and those by adding EO resist up to 5 h.Our results are consistent with those of Ayadi et al.[38] and Caponio et al.[42].

In fact,the samples of oils flavored by green anise seeds by classic maceration and those by addition of the EO of the seeds showed significantly low values of the peroxide index, K232and K270during storage period (represented by heating at 60°C).This result indicated the effectiveness of aromatization with green anise seeds on the thermal stability of VOO.In addition,the holding capacity of olive oil is a highly desirable attribute especially if the product is to be marketed in areas with hot climatic conditions [40, 46].The effectiveness of aromatization by classic maceration or by adding EO is also confirmed following a stability of up to 5 h under cooking conditions (heating at 130 °C).This efficiency could be explained by the higher content of strong antioxidants present in the EO of each aromatic plant[5,14].Several studies of different extracts of green anise seeds have shown very strong antioxidant power,reducing power, scavenging of DPPH radicals and superoxide anions,scavenging of hydrogen peroxide and metal chelation activities in comparison with synthetic antioxidants such as butylated hydroxyanisole(BHA),butylated hydroxytoluene(BHT)and α-tocopherol[47–51].

In addition to polyphenols,chlorophylls and carotenoids play an important role in oxidative stability due to their antioxidant nature and are primarily responsible for the color of VOO varying from yellow-green to greenish gold[52].Moreover,the degradation of chlorophyll pigments and carotenoids present in olive oil during heating is very complex.Understanding the steps in the degradation of these pigments is limited by one main difficulty:this degradation gives rise to the formation of various end products,some of which are colorless.The overlap of these products with degradation products adds to this difficulty[52].The pigment content of the oil,mainly the concentration of the chlorophyll and carotenoid fractions,gradually decreases during the heating time(Figs.9,10).However,the transformations of these pigment fractions during storage are different.In fact, the chlorophyll contents decrease continuously as a function of the heating time,the carotenoid contents however decrease considerably after a short heating period to reach a value which tends towards a minimum.These differences in the degradation kinetics lead to the conclusion that the carotenoids provided by green anise seeds are more sensitive to temperature than chlorophyll.The same findings were made by Ayadi et al.[38]in their work on flavoring VOO with different Tunisian aromatic plants.

Gross[53]indicated that the presence of oxygen is a crucial factor in the degradation of carotene and that even a low concentration of oxygen leads to a significant loss of pigments.Along with the presence of oxygen,the presence of free radicals can also accelerate the degradation kinetics of carotenoids[52].It is accepted that the oxidation of carotenoids depends on the simultaneous oxidation of unsaturated fats [54].These unsaturated fats are probably oxidized in the early stages of oil storage by lipoxygenase,and the oxidation product in turn oxidizes the carotenoids.However,Ayadi et al.[38]correlated the presence of both free radicals and oxygen to explain the sharp decrease in carotenoid content after a short heating period.

The PC naturally present in olive oil are a source of natural antioxidants.The main compounds of these polyphenols present in olive oil are hydroxytyrosol and its derivatives in addition to tyrosol and its derivatives[55,56].This author also indicated that the polyphenol content is different from one oil to another and that there are wide limits that have been reported (from 50 to 1,000 mg/kg), but the values are generally between 100 and 300 mg/kg.However,the polyphenol content of an olive oil is dependent on several factors such as the variety of olive,the stage of maturation,the extraction system,and the conditions of processing and storage[4,55,57].In addition,the rate of phenolic enrichment of flavored oils differs depending on the aromatization method used.In fact, ultrasonic assisted maceration facilitates,much more than classic maceration,the release of the PC contained in the seeds of anise to the oily matrix:ultrasound is widely used for the destruction of cell walls and membranes in the purpose of releasing the contents of plant cells.

This also results in an easier release of the PC into the surrounding environment,especially those which are free.However, the work of Amrani Joutei [58] on grapes has shown that the accumulation of PC in the vacuole, their condensation and binding to the tonoplast as well as the binding of condensed tannins to cell walls make their extractability difficult.In addition,he showed that vacuolar membranes exhibit resistance to ultrasound which,attributed to an elasticity of the tonoplast,prevents the release of these tannins into the surrounding medium,even at an intensity of 20 kHz.In addition,aromatization by adding EO in VOO results in less enrichment than the two maceration processes.Work on the chemical constituents of the EO of anise seeds has enabled the isolation of a number of PC characteristic of the umbelliferous family[33,59,60].Further studies have been carried out to isolate flavonoids from anise and have led to the isolation of quercetin 3-glucuronide,rutin,luteolin 7-glucoside,isoorientine and isovitexin as compounds crystalline and apigenin 7-glucoside and luteolin glycoside as non-crystalline compounds[33,61].Different behaviors of flavored oils with respect to their polyphenol contents were noticed during heat treatment and can be explained by the efficiency and stability of the compounds which could migrate from green anise seeds to the olive oil during the aromatization process.Identical behaviors have been observed by Ayadi et al.[38].Several data are available in the literature concerning the composition of EO and extracts of green anise seeds[31,33,62–65].In our study,the abundance of natural active substances present in green anise seeds can act synergistically as scavengers of free radicals and/or contributed to the protection of the initial total phenols such as tocopherol,the main antioxidant contained in the VOO, against degradation by thermal oxidation [66].Indeed, these natural components (depending on the plant material)can react with free radicals in olive oil,thus effectively inhibiting the loss of tocopherols being the most important natural antioxidant in olive oil heat induced.However,the antioxidant evaluation of green anise extracts showed very high activities in all dosed methods[62].The same properties were reported by Al-Ismail and Aburjai[67]who evaluated the antioxidant activity of anise seed extracts and positively correlated with the higher abundance of PC(expressed as total phenols).

5.Conclusion

The aromatization test of VOO with green anise seeds is carried out with the aim of improving the quality of the oil and its thermal stability.Three aromatization methods were compared in this study:aromatization by direct addition of EO resulted in a very high degree of aromatization than aromatization by ultrasonic assisted maceration.The latter method in turn exhibits a greater degree of aromatization than aromatization by classic maceration.The maceration of green anise seeds in VOO, whether classic or assisted by ultrasound,results in significant modifications of all the physico-chemical parameters studied,unlike aromatization by direct addition of the EO, where these parameters change very little.The change in these parameters of flavored oils could explain the differences observed in the resistance of flavored olive oils to thermal oxidation: oil aromatized by classic maceration shows greater thermal stability than that aromatized by addition of EO and low resistance in aromatized oil by ultrasonic assisted maceration.

Author Contributions

Youssef Moustakime:Writing,review&editing,methodology,investigation supervision,results interpretation,data curation; Zakaria Hazzoumi: Methodology, treatment of GC/MS spectra, results interpretation; Khalid Amrani Joutei: Supervision, results interpretation, resources,review.

Conflicts of Interest

The authors declare that there are no conflicts of interest.

Acknowledgement

We warmly thank Mr.Toufik Haloui who assured us of the aromatization test of VOO by green anise seeds using ultrasonic assisted maceration.