Can the water quality influence the chemical composition,sensory properties,and oxidative stability of traditionally extracted argan oil?

Abstract

Argan oil (AO) is an appreciated vegetable oil thanks to its high nutritional and cosmetic values. AO extraction technology has evolved to meet the market demand. However artisanal production is still widely practiced. The present study aimed at highlighting the influence of water quality on the physicochemical and sensory properties of artisanally extracted AO. To meet this objective, AO was prepared using various water types namely: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). The obtained AOs were evaluated in terms of routinely measured quality indices: iodine, peroxide, acidity, and anisidine values, UV specific extinction coefficients, refraction index, and moisture content. Chemical composition (fatty acids, sterols content, and tocopherols content) was investigated together with oxidative stability (OS) and sensory properties. As revealed by the statistical test used, water quality impacted significantly mainly on AO chemical composition, OS, and sensory properties. Obtained results of almost studied quality attributes were consistent with the Official Moroccan Norm. The greatest values of saturated and monounsaturated fatty acids were recorded in AOMW and OAWW, respectively, while AOUW together with AOTW displayed the best record of polyunsaturated fatty acids. Moreover, the highest values of tocopherols were found in AOTW and AOUW. AODW and AOUW presented greatest values of sterols content, OS, and shelf life. Likewise, sensory analysis was satisfactory in almost obtained AOs. Principal component analysis confirmed these results and allowed a good separation among AOs especially with sterols and tocopherols. Based on these outcomes, it could be concluded that water quality is an important parameter to consider by AO producers, ultra-pure and distilled water seemed to exert an ameliorative effect on quality, stability, and shelf life of AOs.

Keywords

Argan oil artisanal extraction water quality bioactive compounds oxidative stability

1 Introduction

Argan oil (AO) is considered a relatively new global product, which is exported only from Morocco. However different European and North American companies distribute AO worldwide. AO is a non-refined vegetable oil, the more well known “virgin oil” type, produced from the argan tree [Argania spinosa (L.) Skeels], which is endemic to Southwestern Morocco, where it plays important ecological and socio-economic roles [1 –4]. AO is highly appreciated for its high cosmetic, pharmaceutical, nutritional values as well as its unique organoleptic properties owing to its high percentage of unsaturated fatty acids and insaponifiable matter composition [5 –8].

AO chemical composition is well documented [9 –13]. It contains monounsaturated (43–49%of oleic acid), polyunsaturated (29.3–36%of linoleic acid), and saturated (over 20%) fatty acids. AO is also rich in phytosterols, with schottenol and spinasterol as predominant compounds. AO is a good natural antioxidant since it contains 60 to 90 mg/100 g of tocopherols, the major compound is γ-tocopherol (81 to 92%), while both α- and δ-tocopherols are present in small amounts [12, 14]. Along with tocopherols, AO includes other antioxidative active molecules such as polyphenols [15], coenzyme Q10, and melatonin [3].

AO has been used in Morocco for centuries, either as a food, or applied to the skin for cosmetic and medicinal purposes [16, 17]. Pressing roasted kernels affords edible AO, while the unroasted ones are used to make cosmetic AO, which has a very pale gold color and a slightly bitter taste. AO consumed as a food is copper-colored and has a hazelnut-like taste, mostly resulting from the presence of volatile compounds formed during kernel roasting [18, 19]. Following these authors, AO volatile compounds encompass various chemical classes including pyrazines, aldehydes, alcohol, acids, esters, ketones, lactones, terpene, N-heterocycle, and furans. Some volatile compounds such as phenolics are known to possess a potential antioxidant power and therefore improve AO stability [15]. AO aroma volatile profile varies widely depending upon several factors such as fruit processing, extraction method, storage, among others. Such volatile compounds, produced via oxidation, and/or Maillard reaction, are present in higher amounts and actively participate in the preservation of food oil [12]. From an organoleptic point of view, AO has different tastes and smells depending on the initial preparation process and change considerably during storage following Matthäus et al. (2010) [13]. These authors reported a nutty teste in AO samples from unroasted kernels, nutty as well as roasty attributes in AO from roasted kernels. However nutty attribute disappears, during storage, within about 16 weeks and Roquefort cheese taste takes place.

Currently, AO is mainly produced in cooperatives via a semi-industrial process using mechanical extraction, allowing the production of high-quality AO [8 , 20]. Previously, the same oil was prepared by women exclusively following a laborious ancestral process which was transmitted from mother to daughter [13, 21]. The traditional process of extraction includes seven-steps:1) fruit harvesting, 2) fruit dehulling, 3) nut cracking, 4) kernel roasting, 5) kernel grinding, 6) malaxing into an oily dough, to which small amounts of water are added), and 7) the final step is traditional oil extraction [10]. This process is very laborious. Indeed, for one person, about 58 hours are needed to extract around 2 to 2.5 L of oil from 100 kg of dried fruits. It is frequently achieved in unsatisfactory sanitary conditions, and the oil obtained by the traditional process frequently result in a low oxidative stability (sometimes just a few months). In addition, the conditions of extraction and the use of poor quality water may lead to microbiological concerns [8]. Indeed, poor water quality is linked to the hydrolysis properties, leading to a decrease in nutritional and organoleptic values, and therefore depreciating the quality of the product, which, in turn, results in a low market value [8, 22].

Despite these limitations, AO exports boomed recently [10], and international demand for traditional AO also increases. This demand justifies the need for a certified oil of high quality. The aim of the present work was to investigate the influence of water quality on artisanal AO quality in order to select some types of water that could be used to produce a certified and higher quality AO.

2 Material and methods

2.1 Chemicals

All solvents used were of HPLC or analytical grade. Both iso-octane and iso-propanol were used for chromatography and cyclohexane used for UV spectrophotometer determinations at λ= 232 nm and λ= 270 nm. These chemicals were acquired from a Professional Lab (Casablanca, Morocco). For each evaluated elements (K, Na, Cu, Zn, Fe, B, Mn, Ca, Cr, Cd, Hg, and Pb), a standard solution (1000μg/mL dissolved in 2%wt HNO3) was used as a stock solution for calibration. This was supplied by Merck Millipore (Certi PUR, Darmstadt, Germany). Stock solutions were kept at 4 °C. Ultra-pure water with a maximum resistivity of 18.2 MΩ/cm was acquired from Milli-Q Millipore system (Darmstadt, Germany) and Argon alpha-gas (purity over 99.99%) was supplied by Air Liquid.

2.2 Water quality

For AO extraction, five water types were used as follows: well water (WW), tap water (TW), mineral water (MW), distilled water (DW), and ultra-pure water (UW).

2.3 Samples preparation

AO was obtained from the Tiout cooperative (Taroudant county, Morocco) using the traditional millstone method. The fruits were harvested in summer 2018 and air-dried for 3 weeks. The nuts were cracked manually to release kernels, which were roasted for 15 min to 20 min by heating gently in clay pans until reaching a brown color. The roasted kernels were then grinded, via a millstone, into an oily dough. This was then divided into 5 sub-samples, which were used for the preparation of five kinds of AO depending on the water type added to achieve the oil extraction (WW, TW, MW, DW, and UW). In fact, to the obtained oily dough, small quantities of each water type (water to kernels ratio: 250 mL/1 kg) were added regularly until reaching a smooth paste. This was constantly kneaded to separate the oil. Following water kind used for extraction, AO obtained were the following: AO extracted with WW (AOWW), with TW (AOTW), MW (AOMW), DW (AODW), and UW (AODW). The obtained oils were then filtered and kept in brown glass bottles at 4 °C.

2.4 Analytical methods

2.4.1 Determination of water physicochemical parameters

The five kinds of water used for extraction were subjected to physicochemical characterization to explain the differences observed in AO quality.

pH was measured using a pH meter Seven Excellence supplied by METTLER TOLEDO (Columbus, USA). Total hardness was determined by titration with EDTA solution after the addition of black Eriochrome indicator. Alkalimetric title was determined by titration using sulfuric acid. Chlorides were measured in a neutral medium using a solution of silver nitrate in the presence of potassium chromate. The appearance of the red color, characteristic of the silver chromate, indicates the end of this reaction.

Nine minerals (iron, copper, magnesium, manganese, calcium, potassium, sodium, boron, and zinc) were assessed following NF EN ISO 11885 using an inductively coupled plasma optical emission spectrometer ICP OES, Perkin Elmer Model Optima 8000 DV spectrometer (Waltham, USA).

2.4.2 Moisture content and quality indices of oils

Moisture content (MC), Acidity or free fatty acids (FFA), peroxide value (PV), p-anisidine (p-AV), and UV specific coefficients of extinction (K₂₃₂ and K₂₇₀) were evaluated according to the official analytical methods [23 –27]. Briefly, to determine MC, 10 g of oil were heated in a ventilated oven at 103°C, at least for one hour, until reaching a constant weight and then weighted after cooling. MC was expressed as percentage of weight loss.

FFA was assessed using a solution of AO in ethanol with NaOH (0.1N). It was expressed as percentage of oleic acid. PV was determined via iodine titration of a solution of AO in iso-octane/acetic acid (2 : 1) and was expressed as milliequivalents of active oxygen per kilogram of oil (meq O₂/kg oil). Specific extinction coefficients (K₂₃₂ and K₂₇₀) were determined in a 1%(w/v) solution of AO in cyclohexane, the absorbance was measured at 232 and 270 nm using a SCILOGEX SP- UV1100 spectrophotometer. Refraction (RI) and iodine (IV) indices were determined according to Gharby et al. (2011) [8].

2.4.3 Fatty acids determination

Fatty acids composition was evaluated following the official analytical method (ISO 12966-2) [28]. Firstly, fatty acids (FAs) were esterified into their fatty acid methyl esters (FAMEs). Fatty acids composition was determined based on their corresponding FAMEs by gas chromatography on a CPWAX 52CB column (30 m x 0.25 mm i.d., 0.25μm film thickness) using helium (He) as a carrier gas (flow rate was 1 mL/min). Temperatures of oven, injector, and detector were 170, 200, and 230 °C, respectively. The injection volume was 2μL in a split mode with a split ratio of 1 : 50. The results were then expressed as the relative percentage of area of each individual fatty acid peak [29].

2.4.4 Phytosterols determination

Phytosterols composition was determined following the method [30]. Briefly, 5 g of AO was saponified, by boiling under reflux for 1h, using a solution of ethanolic potassium hydroxide (2 N). Then, 100 mL of water was added and extraction of unsaponifiable matter was performed using 200 mL of ethanol (EtOH). After that, the organic solution was collected, evaporated, and 20 mg of unsaponifiable matter was dissolved into 0.5 mL of chloroform then chromatographed on a silica gel plate eluted with a mixture of n-hexane and diethylether (65 : 35 v/v). The plate was then sprayed with a solution of 2,7- ichlorofluorescein (0.2%in EtOH), and the band of sterols was carefully removed. The silica gel recovered from the plate was suspended in chloroform and filtered through a paper filter. The solvent was evaporated under N₂ and the sterol composition determined after trimethylsilylation of the crude sterol fraction. Trimethylsilylated derivatives were analyzed by GC using a Varian 3800 instrument equipped with a VF-1, column (30 m and 0.25 mm i.d.) and using helium (He) as a carrier gas (flow rate of 1.6 mL/min). The column temperature was isothermal at 270°C, injector and detector temperature was set at 300°C. The injection volume was 1μL. The results were expressed as the relative percentage of the area of each individual phytosterol peak [29].

2.4.5 Tocopherols determination

Total tocopherols content was determined according to the official analytical method (ISO 9936) [31]. Content of individual tocopherols (α, β, γ and δ) was determined by HPLC using Shimadzu instruments equipped with a C18-Varian column (25 cm×4 mm; Varian Inc., Middelburg, Netherlands). A fluorescence detector was used, its wavelengths of excitation and detection were set at 290 nm and 330 nm, respectively. Eluent used was a 99 : 1 isooctane/ isopropanol (V/V) mixture and the flow rate consisted of 1.2 mL/min.

2.4.6 Oxidative stability

The oxidative stability (OS) was evaluated following the ISO 6886 official analytical method [32] using a Rancimat 743 (Metrohm Co, Basel) instrument. Briefly, OS was expressed as induction period (IP, hours) determined at 110 °C using 3 g as an AO sample with an air flow of 20 L/h. Volatile compounds released during the degradation process were collected in a flask filled with 60 mL of distilled water. Then, the conductivity of this solution was measured.

2.5 Sensory analysis

Sensory evaluation of AO samples was performed as described by Matthäus et al. (2010) [13]. To perceive AO aroma, the AO samples were tasted in special glasses of blue color used for olive oil sensory evaluations. The glasses containing AO samples were covered using a watch glass in order to maintain the volatile aroma compounds for the sensory evaluation time. Every single glass was filled with 15 mL of AO to ensure that the oil sample develops an intense aroma on one hand and the volatile compounds spread into the covered glass on the other hand. Sensory evaluation of all AO samples was carried out at room temperature. AO flavor and taste were characterized and described following a previously conceived sensory description form, established by the sensory panel. A system of scoring with a scale ranging from 0 (as not detectable) to 10 (as strongly detectable). This scoring system was used to characterize typical (nutty and roasty) and atypical (bitter, burnt, fusty, musty, rancid, wood-like, yeast-like, among others) attributes.

2.6 Statistical analyses

All determinations as well as measurements were carried out, at least, in triplicate and then averaged values were calculated. Quantitative differences were evaluated by the general linear model followed by Tukey’s test. Data statistical analyses were carried out using the SPSS package version 23 (IBM, Armonk, NY, USA). Mean values reported in tables are means followed by standard deviations (SD). Differences among mean values were considered as significant a probability level of 5%as. Correlations matrix among AO physicochemical properties was established based on mean values. Principal component analysis (PCA) was carried out on mean values of selected variables to investigate the possibility of discrimination among artisanally extracted AO (AOWW, AOTW, AOMW, AODW, and AOUW) using of STATGRAPHICS package version XVII (Statpoint Technologies, Inc., Virginia, USA).

3 Results and discussion

3.1 Water physicochemical properties

Physicochemical parameters for the five water kinds are summarized in Table 1. As evidenced in these outcomes, all water kinds displayed low hardiness values (well, tap, and mineral water) or even null in the case of distilled and ultra-pure water. Similar trends were observed for alkalimetric title, which was found to be low well, tap and mineral water) and null for both distilled and ultra-pure water. Chloride content was similar in well, tap and mineral water, a very low amount was detected in distilled water and no trace elements were detected in ultrapure water. Obtained values of water hardness, alkalinity, and chloride content were below cut-off points of drinking water (Table 1).

Table 1
Physicochemical properties and minerals content of water used for extraction namely well water (WW), tap water (TW), mineral water (MW), distilled water (DW), and ultra-pure water (UW). COF = cut-off points. nd: not detected

WW MW TW DW UW COF

Total hardness (ppm) 33.3±0.9b 9.0±0.1c 37.8±1.1a 0.2±0.0d nd 160

Alkalimetric title (ppm) 32±1a 10.0±0.7c 27.8±0.4b 1.0±0.0d 1.0±0.0d 100

Chloride (ppm) 15.0±0.0a 15.0±0.1a 15.0±0.1a 5.0±0.1b nd 250

pH 7.8±0.1b 8.3±0.1a 7.7±0.1b 7.1±0.1c 7.3±0.1 bc

Zn (mg/L) nd nd nd nd nd

Fe (mg/L) nd nd nd nd nd

B (mg/L) nd 0.01±0.01a 0.01±0.01a 0.014±0.01a nd

Mn (mg/L) nd nd nd nd nd

Mg (mg/L) 57.2±4a 40.6±2.8b 10.9±0.8c 0.1±0.0d nd

Ca (mg/L) 87.3±6a 73.8±5.2b 17.9±1.2c nd nd

Cu (mg/L) nd nd nd nd nd

Na (mg/L) 19.9±1.4c 150.3±15.5a 144.4±10.1b 4.9±0.3d nd

K (mg/L) 3.2±0.2bc 4.0±0.3ab 2.2±0.2cd nd nd

	WW	MW	TW	DW	UW	COF
Total hardness (ppm)	33.3±0.9b	9.0±0.1c	37.8±1.1a	0.2±0.0d	nd	160
Alkalimetric title (ppm)	32±1a	10.0±0.7c	27.8±0.4b	1.0±0.0d	1.0±0.0d	100
Chloride (ppm)	15.0±0.0a	15.0±0.1a	15.0±0.1a	5.0±0.1b	nd	250
pH	7.8±0.1b	8.3±0.1a	7.7±0.1b	7.1±0.1c	7.3±0.1 bc
Zn (mg/L)	nd	nd	nd	nd	nd
Fe (mg/L)	nd	nd	nd	nd	nd
B (mg/L)	nd	0.01±0.01a	0.01±0.01a	0.014±0.01a	nd
Mn (mg/L)	nd	nd	nd	nd	nd
Mg (mg/L)	57.2±4a	40.6±2.8b	10.9±0.8c	0.1±0.0d	nd
Ca (mg/L)	87.3±6a	73.8±5.2b	17.9±1.2c	nd	nd
Cu (mg/L)	nd	nd	nd	nd	nd
Na (mg/L)	19.9±1.4c	150.3±15.5a	144.4±10.1b	4.9±0.3d	nd
K (mg/L)	3.2±0.2bc	4.0±0.3ab	2.2±0.2cd	nd	nd

Values are presented as means±SD of 3 replicates. Within the same line, values followed by the same letter are not significantly different at 5%as a probability level.

No metals were detected in ultrapure water. Zn, Fe, Mn and Cu were not detected in any type of water, while Mg and Ca were in higher concentrations in well and mineral water and in low amount in tap water. Na was found in higher amount in mineral and tap water and in a low amount in well water. These results let conclude that well, mineral and tap water had a similar mineral composition, while distilled water had very poor mineral composition. A comparison with the Moroccan standard for the quality of human feed water NM 03.7.001 was also carried out and shows that the values obtained are all lower than the maximum permissible limits levels and therefore comply with that standard [33].

3.2 Moisture content and quality indices of AO

AO quality can be assessed by edible oil international standards such as Codex Alimentarius. The official Moroccan norm for AO was established in 2003 aiming at defining the specifications of “virgin” AO quality through a set of physicochemical properties [34]. It provides, among others, the qualitative classification and chemical composition of AO.

AO is classified according to basic analytical parameters specified by the official Moroccan norm [34], such as MC, FFA, PV, and K₂₇₀. These chemical parameters are considered as essential and imperative quality indicators for the virgin AO classification. The whole investigated AO indices along with MC are listed in Table 2.

Table 2
Mean values of moisture content (MC) and quality indices: free fatty acids (FFA), peroxide value (PV), UV specific coefficients of extinction (K₂₃₂ and K₂₇₀), anisidine value (p-AV), iodine value (IV), and refraction index (RI) for artisanally extracted argan oil (AO) using various types of water: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). OM Norm = official Moroccan norm

Indices AOWW AOTW AOMW AODW AOUW OM Norm

FFA (g/100g) 0.49±0.09 a 0.62±0.01 a 0.75±0.07 a 0.71±0.09 a 0.67±0.01 a ≤0.8

MC (g/100g) 0.06±0.01 b 0.06±0.00 b 0.08±0.02 b 0.19±0.06 a 0.05±0.01 b ≤0.1

IV (mg I₂/100g) 92.05±0.80 bc 96.97±0.21 ab 95.07±1.17 ab 93.43±4.94 bc 91.33±1.31 bc 91–110

RI (20°C) 1.47±0.01 a 1.47±0.01 a 1.47±0.01 a 1.47±0.01 a 1.47±0.01 a 1.463–1.472

PV(MeqO₂/kg) 0.76±0.03 a 0.57±0.11 bc 0.65±0.50 ab 0.41±0.02 cd 0.27±0.03 d ≤15

K₂₃₂ 1.50±0.01 a 1.26±0.01 b 1.49±0.05 a 1.11±0.05 c 1.16±0.02 c –

K₂₇₀ 0.14±0.05 b 0.23±0.01 a 0.18±0.01 ab 0.23±0.01 a 0.2±0.05 ab ≤0.35

p-AV 4.69±0.10 a 1.30±0.19 d 2.78±0.05 b 1.84±0.12 c 1.24±0.27 d –

Indices	AOWW	AOTW	AOMW	AODW	AOUW	OM Norm
FFA (g/100g)	0.49±0.09 a	0.62±0.01 a	0.75±0.07 a	0.71±0.09 a	0.67±0.01 a	≤0.8
MC (g/100g)	0.06±0.01 b	0.06±0.00 b	0.08±0.02 b	0.19±0.06 a	0.05±0.01 b	≤0.1
IV (mg I₂/100g)	92.05±0.80 bc	96.97±0.21 ab	95.07±1.17 ab	93.43±4.94 bc	91.33±1.31 bc	91–110
RI (20°C)	1.47±0.01 a	1.47±0.01 a	1.47±0.01 a	1.47±0.01 a	1.47±0.01 a	1.463–1.472
PV(MeqO₂/kg)	0.76±0.03 a	0.57±0.11 bc	0.65±0.50 ab	0.41±0.02 cd	0.27±0.03 d	≤15
K₂₃₂	1.50±0.01 a	1.26±0.01 b	1.49±0.05 a	1.11±0.05 c	1.16±0.02 c	–
K₂₇₀	0.14±0.05 b	0.23±0.01 a	0.18±0.01 ab	0.23±0.01 a	0.2±0.05 ab	≤0.35
p-AV	4.69±0.10 a	1.30±0.19 d	2.78±0.05 b	1.84±0.12 c	1.24±0.27 d	–

Values are presented as means±SD of 3 replicates. Within the same line, values followed by the same letter are not significantly different at 5%as a probability level.

Acidity is defined as free fatty acids in a given oil sample. It is usually expressed in g/100g of oleic acid [8, 35]. 0.8 g/100 g was established as the maximum FFA for extra virgin AO (EVAO) in the SNIMA regulation [34]. According to our results, water quality had no significant effect on FFA (Table 2). Therefore, all AO extracted with different types of water were classified into the category of EVAO as their FFA was below 0.8 g/100 g [11]. The highest level of FFA (0.75 g/100 g) was recorded in AOMW. This higher FFA was likely due to hydrolysis of triacylglycerols during oil extraction [36].

The second indicator for the quality of AO is the PV [11]. This is an indicator of primary oxidation products (hydroperoxides) [37]. As shown in Table 2, different kinds of AO showed low PV, the AOWW had the highest value (0.9 mEq O₂/kg), which was below the maximum value defined for any EVAO (15 mEq O2/kg). The lowest PV was found for AOUW (0.27 mEq O₂/kg). Another quality index evaluating the presence of primary oxidation products is K₂₃₂ [38]. In this study, PV and K₂₃₂ seemed to follow the same trend of variation owing to their correlation. Indeed, the highest K₂₃₂ value was found in AOWW (1.5) possibly resulting from oxidation processes during oil extraction.

K₂₇₀ is an important indicator used to classify the category of AO quality following to the official Moroccan norm [34]. K₂₇₀ indicates the presence of products from secondary oxidation [39]. The absorbance at λ= 270 nm showed low values for all AO samples ranging from 0.14 to 0.23. These values were below the limit (0.35) defined by the official Moroccan norm [34]. To get a complete picture on the formation of the secondary oxidation products, we determined the p-AV as a function of the water quality [14]. The p-AV ranged from 1.24 (for AOUW) to 4.69 (for AOWW), which was in full agreement with the results of K₂₇₀. Indeed, other authors have suggested that the absorbance at λ= 270 nm can replace the p-AV in the control of the oil quality.

To determine the effect of water quality used in AO artisanally extraction on the degree of fatty acids unsaturation, both RI and IV were evaluated. The water type used for AO extraction (Table 2) induced no significant modification of the degree of the unsaturation of fatty acids. Indeed, AO samples had similar RI (1.47±0.01). In addition, IV ranged from 91.33±1.31 to 96.97±0.21 mgI₂/100 g of oil. These results agree satisfactory with the official Moroccan norm [34]. Such differences could be ascribed mainly to ionic strength. For instance, deionized water (ultrapure water), which is not a model for natural water, with an extremely low ionic strength that may result in oil oxidation during extraction and create a differential repartition among molecules with different charge characteristics.

3.3 Fatty acids composition

Fatty acids profile is an important feature of vegetable oils, which is closely related to stability, cosmetic, and nutritional quality [40]. In literature, a lot of research works were devoted to AO fatty acids composition. In this regard, unsaturated fatty acids (about 80%) are the main constituents, while saturated ones constitute around 20%of the total fatty acids [10 , 41]. The results of fatty acids AO extracted with different water types are listed in Table 2. Overall, no significant differences (p < 0.05) were found with respect to the fatty acids of AOs prepared with different water qualities. The fatty acids composition was similar among AO samples in agreement with published literature and framed within the normal limits of AO set by the official Moroccan norm [34]. Furthermore, the main fatty acids identified in each AO were oleic and linoleic (Table 3). Saturated fatty acids (mainly stearic and palmitic) were found in smaller amounts. Linolenic and others fatty acids such as gadoleic and arachidic were present only in traces.

Table 3
Mean values of fatty acids (expressed as relative percentage) for artisanally extracted argan oil (AO) using various types of water: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). C14 : 0 = myristic acid, C16 : 0 = palmitic acid, C16 : 1 = palmitoleic acid, C18 : 0 = stearic acid, C18 : 1 = oleic acid, C18 : 2 = linoleic acid, C18 : 3 = linolenic acid, C20 : 0 = arachidic acid, C20 : 1 = eicosenoic acid. OM = Official Moroccan

Fatty acids (%) AOWW AOTW AOMW AODW AOUW OM Norm

C14 : 0 0.1±0.1 a 0.1±0.0 a 0.1±0.0 a 0.1±0.1 a 0.1±0. 1 a ≤0.2

C16 : 0 13.0±0.2 a 13.1±0.3 a 12.9±0.1 a 12.8±0.1 a 13.1±0.2 a 11.5–15.0

C16 : 1 0.1±0.0 a 0.1±0.0 a 0.1±0.0 a 0.1±0.0 a 0.1±0.0 a ≤0.2

C18 : 0 6.2±0.3 a 6.3±0.4 a 6.9±1.3 a 6.0±0.0 a 5.6±0.1 a 4.3–7.2

C18 : 1 48.0±0.7 a 47.7±0.3a 47.5±0.8 a 47.8±3.6 a 47.7±0.6 a 43.0–49.1

C18 : 2 31.7±1.2 a 32.1±1.3 a 31.9±0.5 a 31.8±1.1 a 32.1±0.8 a 29.3–36.0

C18 : 3 0.1±0.1 a 0.1±0.0 a 0.1±0.0 a 0.1±0.0 a 0.1±0.1 a ≤0.3

C20 : 0 0.4±0.1 a 0.4±0.1 a 0.4±0.1 a 0.4±0.1 a 0.4±0.1 a ≤0.5

C20 : 1 0.4±0.1 a 0.4±0.1 a 0.4±0.1 a 0.4±0.1 a 0.3±0.1 a ≤0.5

Fatty acids (%)	AOWW	AOTW	AOMW	AODW	AOUW	OM Norm
C14 : 0	0.1±0.1 a	0.1±0.0 a	0.1±0.0 a	0.1±0.1 a	0.1±0. 1 a	≤0.2
C16 : 0	13.0±0.2 a	13.1±0.3 a	12.9±0.1 a	12.8±0.1 a	13.1±0.2 a	11.5–15.0
C16 : 1	0.1±0.0 a	0.1±0.0 a	0.1±0.0 a	0.1±0.0 a	0.1±0.0 a	≤0.2
C18 : 0	6.2±0.3 a	6.3±0.4 a	6.9±1.3 a	6.0±0.0 a	5.6±0.1 a	4.3–7.2
C18 : 1	48.0±0.7 a	47.7±0.3a	47.5±0.8 a	47.8±3.6 a	47.7±0.6 a	43.0–49.1
C18 : 2	31.7±1.2 a	32.1±1.3 a	31.9±0.5 a	31.8±1.1 a	32.1±0.8 a	29.3–36.0
C18 : 3	0.1±0.1 a	0.1±0.0 a	0.1±0.0 a	0.1±0.0 a	0.1±0.1 a	≤0.3
C20 : 0	0.4±0.1 a	0.4±0.1 a	0.4±0.1 a	0.4±0.1 a	0.4±0.1 a	≤0.5
C20 : 1	0.4±0.1 a	0.4±0.1 a	0.4±0.1 a	0.4±0.1 a	0.3±0.1 a	≤0.5

Values are presented as means±SD of 3 replicates. Within the same line, values followed by the same letter are not significantly different at 5%as a probability level.

Figure 1 summarizes mean values of saturated (SFA), monounsaturated (MUFA), and polyunsaturated fatty acids (PUFA) in the studied AO samples. As it can be seen in these results, the best record of MUFA was found in AOWW (48.5%), while AOMW presented the highest value of SFA (20.3%). AOUW together with AOTW presented the highest level of PUFA (32.2%). Such differences in fatty acids levels among AOs extracted through various water types could be attributed to the physicochemical water properties such as pH and ionic strength. In this context, it has been shown that alkaline pH leads to deprotonation of fatty acids owing to higher level of hydroxyl ions in the bulk phase (argan dough with water). The ionization of the carboxyl groups leads to an increase of ionic repulsion among the ionized acids and therefore increasing their solubility. Likewise, dissolved calcium and other multivalent cations decrease solubility by means of binding of weakly soluble fatty acids.

Fig. 1

Mean values of fatty acids (expressed as relative percentage) for artisanally extracted argan oil (AO) using various types of water: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). SFA = saturated fatty acids, MUFA = monounsaturated fatty acids, and PUFA = polyunsaturated fatty acids. Values are means of 3 replicates.

3.4 Phytosterols composition

AO is rich in phytosterols (over 220 mg/100 g of phytosterols), spinasterol (34–44%) and schottenol (44–49%) are predominant compounds following the official Moroccan norm. Δ7-avenasterol (4–7%) along with stigmasta-8-22-dien-3β-ol (3.2–5.7%) are present in small amounts. Campesterol is found only in traces. Phytosterols are usually used to prove the authenticity of edible oils [42]. They are endowed with potent biological properties and therefore their determination is of a major interest due to their antioxidant activity and impact on health [43 –45]. In our outcomes, no significant differences (p < 0.05) were observed among AO samples in terms of phytosterols composition and reported values were consistent with the official Moroccan norm [34]. Total phytosterols content ranged from 105 to 130 mg/100 g. The highest quantity of phytosterols (130 mg/100 g) was observed in AODW followed by AOUW (128.5 mg/100 g), while the lowest value (105 mg/100 g) was observed in AOMW.

3.5 Tocopherols composition

It has been proven that tocopherols play an important role as natural antioxidants found in AO, therefore their composition is of a great importance regarding AO quality [5 , 47]. AO is widely reported to contain α β (minoritary), γ, and δ−tocopherol [14]. According to these authors, the tocopherol content in AO can be up to 90 mg/100 g and is never below 60 mg/100 g. γ-tocopherol is the major tocopherol since it constitutes about 81−92%of the total tocopherols present in AO. Table 5 shows the results of tocopherols content in AO. Small differences were observed for each tocopherol (γ, δ and, α-tocopherol). The highest amount of γ-tocopherol (86.9%) was displayed by AOUW followed by AODW (87.3%), whereas, the lowest one was found in AOMW (84.3%). It is important to note that higher γ-tocopherol content and the low linolenic acid content are responsible for the AO resistance to oxidation. For α-tocopherol, the greatest value (8.3%) was observed in AOMW and the lowest one (6.2%) was found in AOTW. Furthermore, the best score of δ -tocopherol (8.3%) presented by AOTW, however, the lowest value (6.1%) was found in AOWW. With respect to total tocopherol, no significant variations were found for the different kinds of AO, with the greatest value (110±2.5) in the case of AOTW as elucidated in Table 5. In all cases, total tocopherols content was remarkably higher than the maximum limit set by the Official Moroccan Norm (90 mg/100 g). High tocopherols content is undoubtedly a good characteristic feature for oils since tocopherols show great benefits for human health on one hand and contributes to increase OS on the other hand.

Table 5
Mean values of tocopherols content of argan oil (AO) prepared artisanally from argan kernels using different types of water quality: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). α-T = (α-tocopherol), γ-T = (γ-tocopherol), δ-T = (δ-tocopherol), and TT = total tocopherols. OM = Official Moroccan Norm

Tocopherol AOWW AOTW AOMW AODW AOUW OM Norm

α-T (%) 6.7±0.1 b 6.2±0.1 bc 8.3±0.5 a 6.7±0.2 b 6.3±0.3 bc 2.4–6.5

γ-T (%) 85.5±0.2 a 85.4±0.5 a 84.3±0.3 a 87.3±0.3 a 86.9±0.1 a 81–92

δ-T (%) 6.1±0.3 c 8.3±0.5 a 7.3±0.8 ab 6.7±0.1 bc 7.5±0.2 ab 6.2–12.8

TT (mg/100g) 102±2.5 c 110±2.5 a 100±2.5 c 100±2.5 c 105±2.5 bc 60–90

Tocopherol	AOWW	AOTW	AOMW	AODW	AOUW	OM Norm
α-T (%)	6.7±0.1 b	6.2±0.1 bc	8.3±0.5 a	6.7±0.2 b	6.3±0.3 bc	2.4–6.5
γ-T (%)	85.5±0.2 a	85.4±0.5 a	84.3±0.3 a	87.3±0.3 a	86.9±0.1 a	81–92
δ-T (%)	6.1±0.3 c	8.3±0.5 a	7.3±0.8 ab	6.7±0.1 bc	7.5±0.2 ab	6.2–12.8
TT (mg/100g)	102±2.5 c	110±2.5 a	100±2.5 c	100±2.5 c	105±2.5 bc	60–90

Values are presented as means±SD of 3 replicates. Within the same line, values followed by the same letter are not significantly different at 5%as a probability level.

3.6 Rancimat study and shelf life of AO samples

Oil oxidation is the main factor behind the oil quality deterioration [13, 48]. It is a series of chemical alterations. These can be perceived in the initial stage through a deterioration of both taste and smell also known as rancidity. As previously reported in our works, higher content of γ-tocopherol combined with lower content of linolenic acid are involved in the resistance to AO oxidation. However, the use of water in the artisanal argan process has a detrimental effect on oxidative stability [8]. For more details regarding the influence of water during extraction on the oil OS, we determined IP by Rancimat test at 383K and 393K. Shelf lives were then estimated for AO prepared with different kinds of water. The results obtained are presented in Table 6. For both studied temperatures (383K and 393K), there were significant differences among AOs in terms of IP and shelf life. Moreover, longer IP was observed in AOUW In comparison with the other samples for both temperatures (Table 6). For instance, at 383K, the best record of the IP (15.52h) was found in AOUW followed by AODW (14.94h), AOTW (13.97h), AOMW (13.32h), and AOWW (12.69 h). IP for the oils extracted by the five water types could be ranked, in decreasing order, as follows: AOUW > AODW>AOTW>AOMW>AOWW. Nevertheless, AOUW behaved more satisfactorily under Rancimat conditions at 383K and 393K. This higher OS seemed to be linked to its notorious higher γ-tocopherol level. Shelf life is also an essential indicator to get consumer acceptance and loyalty. An equation model for the Rancimat test was used in order to predict the stability at room temperature (25°C) of AO artisanally pressed with different water types. AOWW was more sensitive to oxidation than all other AOs. Its shelf life was estimated to be lower than a half year at room temperature, followed by AOMW (0.76 year), AOTW (0.81 year), and AOTW (1.19 years). Whereas, when AO is prepared with UW, its shelf life raised to about one year and a half. This could be explained by the lack of metals, which promote oxidation, in AOUW [8, 49].

Table 6
Mean values of induction period (in hours) at two different temperatures (393 and 383 K) and shelf life (years) at 293 K of argan oil (AO) prepared artisanally from argan kernels by use of different water quality: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). K: Kelvin degree

Induction period (hours) Shelf life (years)

393 K (120°C) 383 K (110°C) 298K (25°C)

AOWW 5.45±0.5 c 12.69±0.5 c 0.70

AOMW 6.68±0.5 b 13.32±0.5 b 0.76

AOTW 6.97±0.5 b 13.97±0.5 b 0.81

AODW 7.22±0.5 ab 14.94±0.5 ab 1.19

AOUW 7.53±0.5 a 15.52±0.5 a 1.53

	Induction period (hours)	Shelf life (years)
AOWW	5.45±0.5 c	12.69±0.5 c	0.70
AOMW	6.68±0.5 b	13.32±0.5 b	0.76
AOTW	6.97±0.5 b	13.97±0.5 b	0.81
AODW	7.22±0.5 ab	14.94±0.5 ab	1.19
AOUW	7.53±0.5 a	15.52±0.5 a	1.53

Values are presented as means±SD of triplicates. Within the same column, values followed by the same letter are not significantly different at 5%as a probability level.

3.7 Sensory analysis

It is well known that AO is an interesting food with an important nutritional value owing to its higher content in terms of unsaturated fatty acids along with a high level of bioactive compounds. It is also noteworthy to highlight the importance of AO in terms of taste and particular aromas, which make it highly appreciated by consumers [7, 10]. Routinely measure quality parameters are is not sufficient to determine the quality of the oil. In fact, volatile molecules, accumulated during kernels roasting and the use the water during oil extraction, are susceptible to modify color, odor, and flavor of AO. To obtain a detailed picture of the possible impact of different water types during extraction on the sensory quality, a supplementary sensory analysis was performed. In this study, all studied samples gave satisfactory results (data not shown). Indeed, for all samples, the median of the fruity attributes is superior to zero, while the the defect median is equal to zero. Regarding the coloring aspect, we noticed that AOUW looked a bit darker in comparison to the other oil samples. In conclusion, all oil samples were satisfactory in terms of the sensory criteria of certified AO.

3.8 Correlations study

Correlation coefficients among AO physicochemical properties are illustrated in Table 7. As it can be seen in this outcomes, important negative and positive correlations were highlighted. Oil stability (OS) and shelf life (SL) were strongly and positively (p < 0.001) correlated to each other but negatively associated to PV. This means that the increase of OS result in high SL values and low values of PV. Likewise, Both OS and SL were positively correlated to tocopherols confirming their contribution to enhance OS and therefore extending oil SL. A significant (p < 0.05) positive association between total tocopherols and γ-tocopherol observed in our result, which means that an increase of γ-tocopherol (γ-T) leads to high values of total tocopherols since γ-T was the major tocopherol (more than 84%of the total tocopherols content). Total tocopherols were correlated (p < 0.001) positively to schottenol and negatively to spinasterol and campesterol. Regarding routine quality indices, K₂₃₂ and K₂₇₀ were positively linked to each other on one hand and linked positively to FFA and PV on the other hand. Similar correlations were found by other authors [49]. Furthermore, oxidation products such as hydroperoxides as well as their derivatives (measured by FFA and PV) are considered as conjugated diene (CD) and conjugated triene (CT) [50]. CD and CT absorb the light at 232 and 270 nm, respectively. Hydroperoxides (primary oxidation products) absorb at 232nm and are unstable and are quickly converted into secondary oxidation products (mostly diketones and unsaturated ketones), which absorb at 270 nm. This explains the positive correlation among K₂₃₂, K₂₇₀, FFA, and PV as discussed in Sakar et al. (2017)[51].

Table 7
Coefficients of correlation among physicochemical properties of AO. FFA = free fatty acids, PV = peroxide value, K232 and K270 = UV specific coefficients of extinction, p-AV = anisidine value, and IV = iodine value. SFA = saturated fatty acids, MUFA = monounsaturated fatty acids, and PUFA = polyunsaturated fatty acids. SC = schottenol, SS = spinasterol, Δ-7-A =Δ7-avenasterol, SDO = stigmasta-8-22-dien-3β-ol, CA = Campesterol, CH = Cholesterol, and TS = total sterols. α-T =α-tocopherol, γ-T =γ-tocopherol, δ-T =δ-tocopherol, TT = total tocopherols, OS = oil stability, and SL = shelf life. ^, ^, and ^ indicate significance at 0.05, 0.01, and 0.001 levels of probability, respectively

FFA MC IV RI PV K232 K270 p- AV SFA MUFA PUFA SC SS Δ –7 –A SDO CA CH TS α-T γ-T δ -T TT OS SL

FFA 0.60^* –0.50 0.10 –0.30 –0.40 0.30 –0.10 0.10 –0.70^* 0.10 –0.70^* 0.70^* –0.60^* –0.56 0.70^* 0.20 0.00 0.56 –0.10 0.10 –0.70^* 0.30 0.30

MC 0.20 0.30 0.10 –0.30 0.50 0.30 0.00 0.00 –0.20 –0.60 0.60^* –0.80^** –0.10 0.60^* –0.40 0.10 0.46 0.00 –0.20 –0.60^* –0.10 –0.10

IV 0.70^* –0.10 –0.30 0.60^* –0.20 –0.30 0.60^* 0.20 0.60^* –0.60^* 0.30 0.41 –0.60^* –0.10 0.50 –0.67^* 0.30 0.20 0.60^* 0.10 0.10

RI –0.60^* 0.70^* 0.90^** –0.70^* –0.20 –0.10 0.70 0.50 –0.50 0.30 0.05 –0.50 0.50 0.60^* –0.67^* 0.20 0.70^* 0.50 0.60^* 0.60^*

PV 0.90^** 0.70^* 0.90^** 0.60^* 0.30 –0.50 –0.30 0.30 –0.50 0.67^* 0.30 –0.70^* –0.80^** 0.46 –0.60^* –0.50 –0.30 –0.99^* –0.99^*

K232 0.90^** 0.70^* 0.70^* 0.10 –0.30 –0.10 0.10 –0.20 0.67^* 0.10 –0.40 –0.90^** 0.31 –0.70^* –0.30 –0.10 –0.90^* –0.90^*

K270 –0.60^* –0.50 0.00 0.40 0.20 –0.20 0.10 –0.31 –0.20 0.30 0.80^** –0.41 0.50 0.40 0.20 0.70^* 0.70^*

p- AV 0.30 0.40 –0.80^** –0.60^* 0.60^* –0.70^* 0.31 0.60^* –0.90^** –0.60^* 0.72^* –0.30 –0.80^** –0.60^* –0.90^ –0.90^

SFA –0.50 0.30 –0.10 0.10 –0.30 0.67^* 0.10 0.10 –0.90^ 0.21 –0.99^* 0.30 –0.10 –0.60^* –0.60^*

MUFA –0.60^* 0.20 –0.20 0.10 0.21 –0.20 –0.70^* 0.30 –0.15 0.50 –0.6 0.20 –0.30 –0.30

PUFA 0.60^* –0.60^* 0.50 0.21 –0.60^* 0.90^** 0.10 –0.67^* –0.30 0.99^*** 0.60^* 0.50 0.50

SC –0.99^* 0.90^ 0.41 –0.99^* 0.50 0.30 –0.98^* 0.10 0.60^* 0.99^* 0.30 0.30

SS –0.90^ –0.41 0.99^* –0.50 –0.30 0.97^* –0.10 –0.60^* –0.98^* –0.30 –0.30

Δ –7 –A 0.05 0.90^ 0.60^* 0.40 –0.87^ 0.30 0.50 0.90^ 0.50 0.50

SDO –0.41 –0.15 –0.56 –0.29 –0.67^* 0.21 0.41 –0.67^* –0.67^*

CA –0.50 –0.30 0.98^*** –0.10 –0.60^* –0.97^* –0.30 –0.30

CH 0.20 –0.56 –0.10 0.90^ 0.50 0.70^* 0.70^*

TS –0.46 0.90^ 0.10 0.30 0.80^ 0.80^**

α-T –0.21 –0.67^* –0.98^*** 0.46 0.46

γ-T –0.30 0.10 0.60^* 0.6^*

δ -T 0.60^* 0.50 0.50

TT 0.30 0.30

OS 0.99^***

SL

3.9 Principal component analysis (PCA)

PCA was performed as a multivariate statistical approach to investigate the possibility of separation among artisanally extracted AOs obtained using various kinds of water (AOWW, AOTW, AOMW, AODW, and AOUW) based on selected dependent variables (quality indices, fatty acids, phytosterols, and tocopherols). PCA was successfully used for the same purpose in previous works[52, 53]. In our results, from all extracted principal components (PCs), the first two were retained since they explained over 80%in our data variability. The points plotted on the surface determined by both axis 1 and 2 (Fig. 2A) are related to AO samples, which appear to be distributed along the first component (PC1). Toward the positive direction of PC1, AODW together with AOUW interacted with higher scores of K₂₇₀, FFA, PV, and MC. while AOTW was associated to the best record of IV. In contrast, both AOWW and AOMW were linked to higher records of PV, p-AV, and K₂₃₂ on the negative values of PC1. Similarly, points distributed on the surface delimited by axis 1 and 2 are linked to AOs as an independent variable, while dependent variables consist of fatty acids (Fig. 2B). On the first component (PC1), both AOUW and AOTW were associated to the greatest values of C16 : 0, C18 : 2, and PUFA, while AOMW was marked by higher records of C18 : 0 and SFA. On the contrary, on the negative values of PC1, AOWW together with AODW were associated with higher levels of C18 : 1 and MUFA. Figure 2C presents the distribution of AOs (as an independent variable) versus phytosterols as dependent variables on the plan formed by PC1 and PC2. From these outcomes, AO samples were distributed following PC1. On the positive side of PC1, both AOTW and AOWW were associated with the greatest values of schottenol (SCO) and Δ-7-avenasterol (Δ-7-AS), while AODW together with AOUW were characterized by the greatest value of total sterols (TS). On the other side of this component, AOMW associated with higher values of stigmasta-8-22-dien-3β-ol (SDO) and spinasterol (SS). Finally, Fig. 2D summarizes interactions between AOs (independent variable) on one hand and tocopherols content as dependent variables on the other hand. AOs seemed to be distributed following the first component (PC1). AOWW, AODW, and AOMW were plotted toward the positive direction of PC1, with the highest level of α-tocopherol (α-T) being associated with AOMW. Furthermore, on the negative side of the same component, AOTW was marked by higher δ-tocopherol (δ-T) and total tocopherols (TT), meanwhile AOUW had the greatest level of γ-tocopherol (γ-T). The results of PCA confirmed comparison of mean values presented in Tables 2 –5. PCA was used successfully in AO chemistry to discriminate among AOs. AO chemical composition is under the dependency of different factors such as clones, tree age, harvest year, and geographical origin. Aithammou et al. (2019) used PCA to separate successfully AO from old and younger trees [54].

Fig. 2

Principal component analysis (PCA) projections on PC1 and PC2. Eigenvalues are symbolized as blue segments representing parameters that most affect each principal component. Plotted points are mean values of each studied parameter of argan oil (AO) samples prepared using various types of water: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). Quality indices (A): MC = moisture content, FFA = free fatty acids, PV = peroxide value, K232 and K270 = UV specific coefficients of extinction, p-AV = anisidine value, and IV = iodine value. Fatty acids (B): C14 : 0 = myristic acid, C16 : 0 = palmitic acid, C16 : 1 = palmitoleic acid, C18 : 0 = stearic acid, C18 : 1 = oleic acid, C18 : 2 = linoleic acid, C18 : 3 = linolenic acid, C20 : 0 = arachidic acid, C20 : 1 = eicosenoic acid, SFA = saturated fatty acids, MUFA = monounsaturated fatty acids, and PUFA = polyunsaturated fatty acids. Phytosterols(C): SCO = schottenol, SS = spinasterol, Δ-7-AS =Δ7-avenasterol, SDO = stigmasta-8-22-dien-3β-ol, and TS = total sterols. Tocopherols (D): α-T =α-tocopherol, γ-T =γ-tocopherol, δ-T =δ-tocopherol), and TT = total tocopherols. Values are means of 3 replicates.

Table 4

Mean values of phytosterols content of argan oil (AO) artisanally prepared from argan kernels using different types of water quality: well water (AOWW), tap water (AOTW), mineral water (AOMW), distilled water (AODW), and ultra-pure water (AOUW). Sco = schottenol, SS = spinasterol, Δ7-AS =Δ7-avenasterol, SDO = stigmasta-8-22-dien-3β-ol, CamS = campesterol, ChoS = cholesterol, and TS = total sterols. OM = Official Moroccan Norm

Sterols	AOWW	AOTW	AOMW	AODW	AOUW	OM Norm
Sco (%)	46.0±0.1 ab	46.6±0.5 a	43.7±0.5 d	45.7±0.1 b	46.4±0.1 ab	44–49
SS [%]	39.2±0.5 c	38.6±0.4 c	41.6±0.5 a	39.9±0.6 bc	39.0±0.1 c	34–44
Δ7-AS (%)	4.7±0.2 a	4.7±0.6 a	4.1±0.6 a	4.6±0.6 a	4.8±0.6 a	4–7
SDO (%)	4.6±0.6 a	4.6±0.6 a	4.6±1.1 a	4.5±1.4 a	4.5±0.6 a	3.2–5.7
CamS (%)	0.3±0.1 a	0.2±0.2 a	0.3±0.1 a	0.3±0.1 a	0.2±0.0 a	≤0.4
ChoS (%)	0.1±0.1 a	0.1±0.1 a	0.1±0.3 a	0.1±0.2 a	0.1±0.1 a	≤0.4
TS (mg/100g)	119.8±2.5 b	120.6±2.5 b	105±2.5 c	130.5±2.5 a	128.5±2.5 a	≤220

Values are presented as means±SD of 3 replicates. Within the same line, values followed by the same letter are not significantly different at 5%as a probability level.

4 Conclusion

Based on physicochemical indices (FFA, PV, K₂₃₂, K₂₇₀), all AO samples showed an acceptable quality and not much variation was recorded among AOs extracted using various types of water. Likewise, no significant variations were found in terms of fatty acid and sterol composition. In contrast, we found a significant (p < 0.05) influence of the water type on tocopherols content. In fact, the best results were found in AO artisanally pressed with UW.

Regarding oxidative stability, all results demonstrated that the AO stability of was under the dependency of water type used for extraction. In addition, shelf life at room temperature of AOUW was found to be much higher (about one year and half) as compared to the remaining AO samples. Finally, taken all together, our results demonstrated that water quality as an important factor involved in AO oxidation. Traditionally extracted AOUW was highly appreciated by the sensory panel. Based on these outcomes, AO producers should take into account water quality used for extraction giving some impacts on the chemical and sensorial quality of the obtained AO. Thanks to its higher performance in terms of OS, chemical and sensory quality, ultra-pure water might be recommended for AO extraction.

Footnotes

Acknowledgments

We thank Association Ibn Al Baytar, HSB (Agadir), cooperative Taitmatine (Tiout), GIE TARGANINE and Afoulki cooperative for their support and assistance in this work.

Author contribution

Zaineb Boukyoud: Resources, Writing - review & editing.

Mohamed Ibourki: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Writing - original draft.

Said Gharby: Resources, Writing - review & editing.

El Hassan Sakar: Data acquisition, Writing - review & editing.

Laila Bijla: Data acquisition, Writing - review & editing.

Hajar Atifi: Data acquisition, Interpretation, Writing - review & editing.

Bertrand Matthaus: Supervision, Resources, Software, Writing - review & editing.

Abdelatif Laknifli: Supervision, Resources, Software, Writing - review & editing.

Zoubida Charrouf: Supervision, Resources, Software, Writing - review & editing.

Funding

The authors report no funding.

Conflict of interest

The authors have no conflict of interest to report.

References

Ourrach

, Rada

, Pérez-Camino

, et al. Detectionof argan oil adulterated with vegetable oils: New markers. Grasas yAceites 2012;63:355–364. https://doi.org/10.3989/gya.083811

Guillaume

, Pioch

, Charrouf

. Fruit Oils: Chemistry and Functionality; 2019.

Venegas

, Cabrera-Vique

, García-Corzo

, et al. Determination of coenzyme Q 10, coenzyme Q 9, and melatonin contentsin virgin argan oils: Comparison with other edible vegetable oils. Journal of Agricultural and Food Chemistry 2011;59:12102–12108..

Monfalouti El

, Guillaume

, Denhez

, Charrouf

. Therapeuticpotential of argan oil: A review. Journal of Pharmacy andPharmacology 2010;62:1669–1675. https://doi.org/10.1111/j.2042-7158.2010.01190.x

Guillaume

, Charrouf

. Argan oil and other argan products: Use indermocosmetology. European Journal of Lipid Science and Technology 2011;113:403–408. https://doi.org/10.1002/ejlt.201000417

Lucarini

, Durazzo

, Nicoli

, et al. Cold pressed argan (Argania spinose) oil. Elsevier Inc. 2020.

Sevindik

, Amanpour

, Tsouli Sarhir

, et al. Characterization ofKey Odorants in Moroccan Argan Oil by Aroma Extract DilutionAnalysis. European Journal of Lipid Science and Technology 2019;121:1–8. https://doi.org/10.1002/ejlt.201800437

Gharby

, Harhar

, Guillaume

, et al. Oxidative stability ofedible argan oil: A two-year study. LWT - Food Science andTechnology 2011;44:1–8. https://doi.org/10.1016/j.lwt.2010.07.003

Charrouf

, Harhar

, Gharby

, Guillaume

. Enhancing the value of argan oil is the best mean to sustain the argan grove economy and biodiversity, so far. OCL - Oleagineux Corps Gras Lipides 2008;15:269–271.

10.

Charrouf

, Guillaume

. The argan oil project: Going from utopia toreality in 20 years. OCL - Oilseeds and fats, Crops and Lipids 2018;25:1–5. https://doi.org/10.1051/ocl/2018006

11.

Gharby

, Harhar

, El Monfalouti

, et al. Chemical and oxidativeproperties of olive and argan oils sold on the Moroccan market. Acomparative study. Mediterranean Journal of Nutrition andMetabolism 2012;5:31–38. https://doi.org/10.1007/s12349-011-0076-5

12.

Harhar

, Gharby

, Kartah

, et al. Influence of Argan KernelRoasting-time on Virgin Argan Oil Composition and OxidativeStability. Plant Foods for Human Nutrition 2011;66:163–168. https://doi.org/10.1007/s11130-011-0220-x

13.

Matthäus

, Guillaume

, Gharby

, et al. Effect of processingon the quality of edible argan oil. Food Chemistry 2010;120:426–432. https://doi.org/10.1016/j.foodchem.2009.10.023

14.

Gharby

, Guillaume

, Nounah

, Harhar

, Hajib

, Matthaus

, Charrouf

. Shelf life of Moroccan prickly pear (Opuntia ficusindica) and argan (Argania spinosa) oils: a comparative study. Grasas Aceites. 2021;72:e397. https://doi.org/10.3989/gya.1147192

15.

El Monfalouti

, Charrouf

, Belviso

, et al. Analysis andantioxidant capacity of the phenolic compounds from argan fruit(Argania spinosa (L. ) Skeels). European Journal of LipidScience and Technology 2012;114:446–452. https://doi.org/10.1002/ejlt.201100209

16.

Mohammed

FAE

, Bchitou

, Bouhaouss

, et al. Can the dietary elementcontent of virgin argan oils really be used for adulterationdetection?. Food Chemistry 2013;136:105–108. https://doi.org/10.1016/j.foodchem.2012.07.098

17.

Zaanoun

, Gharby

, Bakass

, et al. Kinetic parameterdetermination of roasted and unroasted argan oil oxidation underRancimat test conditions. Grasas y Aceites. . 2014:65. https://doi.org/10.3989/gya.122713

18.

Monfalouti

, Charrouf

, Giordano

, Guillaume

, Kartah

, Harhar

, Gharby

, Denhez

, and Zeppa

. Volatile Compound Formation During Argan Kernel Roasting Hanae. Natural Product Communications 2012;8:33–36.

19.

Matthäus

, Bonte

, Sinning

, Charrouf

. Aroma-RelevantVolatile Compounds as Markers for the Sensory Quality of Argan Oil. European Journal of Lipid Science and Technology 2019;121. https://doi.org/10.1002/ejlt.201900279

20.

Charrouf

, Guillaume

. Secondary metabolites from Argania spinosa(L.) skeels. Phytochemistry Reviews 2002;1:345–354. https://doi.org/10.1023/A:1026030100167

21.

Charrouf

, Guillaume

. Argan oil: Occurrence, composition andimpact on human health. European Journal of Lipid Science and Technology 2008;110:632–636. https://doi.org/10.1002/ejlt.200700220

22.

Hilali

, Charrouf

, Soulhi

AEA

, et al. Influence of origin andextraction method on argan oil physico-chemical characteristics and composition. Journal of Agricultural and Food Chemistry 2005;53:2081–2087. https://doi.org/10.1021/jf040290t

23.

ISO 662. Détermination de la teneur en eau et en matiéres volatiles. Corps gras d’origines animale et végétale. 2016.

24.

ISO 660. Détermination de l’indice d’acide et de l’acidité. Corps gras d’origines animale et végétale. 2009.

25.

ISO 3960. No TitleDétermination de l’indice de peroxyde—Détermination avec point d’arrêt iodométrique. Corps gras d’origines animale et végétale. 2017.

26.

ISO 6885. Détermination de l’indice d’anisidine. Corps gras d’origines animale et végétale. 2006.

27.

ISO 3656. Détermination de l’absorbance dans l’ultraviolet, exprimeé sous la forme d’extinction spécifique en lumiére ultraviolette. Corps gras d’origines animale et végétale. 2011.

28.

ISO 12966-2. Chromatographie en phase gazeuse des esters méthyliques d’acides gras—Partie 2: Préparation des esters méthyliques d’acides gras. Corps gras d’origines animale et végétale. 2011.

29.

Gharby

, Ravi

, Guillaume

, et al. 2-methyloxolane asalternative solvent for lipid extraction and its effect on thecactus (Opuntia ficus-indica L.) seed oil fractions. OCL 2020;27:27. https://doi.org/10.1051/ocl/2020021

30.

ISO 12228-1. Determination of individual and total sterols contents—Gas chromatographic method. Animal and vegetable fats and oils. 2014.

31.

ISO 9936. Determination of tocopherol and tocotrienol contents by high-performance liquid chromatography. Animal and vegetable fats and oils. 2016.

32.

ISO 6886. Détermination de la stabilité á l’oxydation (essai d’oxydation accéléré). Corps gras d’origines animale et végétale. 2016.

33.

Snima . Qualite Des Eaux D ’ Alimentation Humaine. Norme Marocaine NM 2001;037001:1–14.

34.

SNIMA. P r o j e t d e N o r m e M a r o c a i n e. 2003.

35.

Wang

, Yu

, Zhu

, et al. Chemical Profiles of Twenty-threeMonovarietal Olive Oils Produced in Liangshan Region of China. Journal of Oleo Science 2020;2020:1–11. https://doi.org/10.5650/jos.ess19265

36.

Rigane

, Ayadi

, Boukhris

, et al. Characterisation and phenolicprofiles of two rare olive oils from southern Tunisia: Dhokar andGemri-Dhokar cultivars. Journal of the Science of Food and Agriculture 2013;93:527–534. https://doi.org/10.1002/jsfa.5815

37.

Gharby

, Harhar

, Farssi

, et al. Influence of roasting olivefruit on the chemical composition and polycyclic aromatichydrocarbon content of olive oil. OCL –Oilseeds and fats, Cropsand Lipids 2018;25. https://doi.org/10.1051/ocl/2018013

38.

Gharby

, Guillaume

, Elibrahimi

, Charrouf

. Physico-ChemicalProperties and Sensory Analysis of Deodorized Argan Oil. ACS Food Science & Technology 2021;1:275–281. https://doi.org/10.1021/acsfoodscitech.0c00107

39.

Harhar

, Gharby

, Guillaume

, Charrouf

. Effect of argan kernelstorage conditions on argan oil quality. European Journal of LipidScience and Technology 2010;112:915–920. https://doi.org/10.1002/ejlt.200900269

40.

Hajib

, Nounah

, Harhar

, et al. Oil content, lipid profiling andoxidative stability of ‘sefri’ Moroccan pomegranate (Punicagranatum L.) seed oil. OCL –Oilseeds and fats, Crops and Lipids 2021;28. https://doi.org/10.1051/ocl/2020069

41.

Harhar

, Gharby

, Kartah

, et al. Effect of harvest date ofArgania spinosa fruits on Argan oil quality. Industrial Crops and Products 2014;56:156–159. https://doi.org/10.1016/j.indcrop.2014.01.046

42.

Hilali

, Charrouf

, Soulhi

AEA

, et al. Detection of argan oiladulteration using quantitative campesterol GC-analysis. JAOCS, Journal of the American Oil Chemists’ Society 2007;84:761–764. https://doi.org/10.1007/s11746-007-1084-y

43.

Gharby

, Harhar

, Matthäus

, et al. The chemical parametersand oxidative resistance to heat treatment of refined and extravirgin Moroccan Picholine olive oil. Journal of Taibah University for Science 2016;10:100–106. https://doi.org/10.1016/j.jtusci.2015.05.004

44.

Hajib

, Harhar

, Gharby

, et al. Is geographic origin a good marker for cumin seed oil (Cuminum cyminum L.)? Rivista Italiana delle Sostanze Grasse 2018;95:155–159.

45.

Taoufik

, Zine

, El

Hadek M

, et al. Oil content and mainconstituents of cactus seed oils opuntia ficus indica of differentorigin in Morocco. Mediterranean Journal of Nutrition and Metabolism 2015;8:85–92. https://doi.org/10.3233/MNM-150036

46.

Gharby

, Harhar

, Bouzoubaa

, et al. Chemical characterizationand oxidative stability of seeds and oil of sesame grown in Morocco. Journal of the Saudi Society of Agricultural Sciences 2017;16:105–111. https://doi.org/10.1016/j.jssas.2015.03.004

47.

Delgado

, Al-Hamimi

, Ramadan

, et al. Contribution of tocolsto food sensorial properties, stability, and overall quality. Journal of Food Quality 2020;2020. https://doi.org/10.1155/2020/8885865

48.

Madawalaa

SRP

, Kochharb

, Duttaa

. Lipid components andoxidative status of selected specialty oils. Grasas y Aceites 2012;63:143–151. https://doi.org/10.3989/gya.083811

49.

Choe

, Min

. Mechanisms and factors for edible oil oxidation. Comprehensive Reviews in Food Science and Food Safety 2006;5:169–186. https://doi.org/10.1111/j.1541-4337.2006.00009.x

50.

Gharby

, Hajib

, Ibourki

, et al. Induced changes in olive oilsubjected to various chemical refining steps: A comparative study ofquality indices, fatty acids, bioactive minor components, andoxidation stability kinetic parameters. Chemical Data Collections 2021;33:100702. https://doi.org/10.1016/j.cdc.2021.100702

51.

Sakar

, El Yamani

, Rharrabti

. Variability of oil content and its physico-chemical traits from five almond (Prunisdulcis) cultivars grown in Northern Morocco. Journal of Materials and Environmental Science 2017;8:2679–2686.

52.

Ibourki

, Gharby

, Guillaume

, et al. Profiling of mineralelements and heavy metals in argan leaves and fruit by-productsusing inductively coupled plasma optical emission spectrometry andatomic absorption spectrometry. Chemical Data Collections 2021;35:100772. https://doi.org/10.1016/j.cdc.2021.100772

53.

Sakar

, El Yamani

, Boussakouran

, et al. Variability of oilcontent and its physicochemical traits from the main almond [Prunusdulcis Mill. DA Webb] cultivars grown under contrasting environmentsin north-eastern Morocco. Biocatalysis and AgriculturalBiotechnology.52. 2021;32:101952. https://doi.org/10.1016/j.bcab.2021.101952

54.

Aithammou

, Hallouti

, Daoud

, et al. Effect of clones, year ofharvest and geographical origin of fruits on quality and chemicalcomposition of Argan oil. Food Chemistry 2019;297:124749. https://doi.org/10.1016/j.foodchem.2019.05.024

	Induction period (hours)		Shelf life (years)
	393 K (120°C)	383 K (110°C)	298K (25°C)
AOWW	5.45±0.5 c	12.69±0.5 c	0.70
AOMW	6.68±0.5 b	13.32±0.5 b	0.76
AOTW	6.97±0.5 b	13.97±0.5 b	0.81
AODW	7.22±0.5 ab	14.94±0.5 ab	1.19
AOUW	7.53±0.5 a	15.52±0.5 a	1.53

	FFA	MC	IV	RI	PV	K232	K270	p- AV	SFA	MUFA	PUFA	SC	SS	Δ –7 –A	SDO	CA	CH	TS	α-T	γ-T	δ -T	TT	OS	SL
FFA		0.60^*	–0.50	0.10	–0.30	–0.40	0.30	–0.10	0.10	–0.70^*	0.10	–0.70^*	0.70^*	–0.60^*	–0.56	0.70^*	0.20	0.00	0.56	–0.10	0.10	–0.70^*	0.30	0.30
MC			0.20	0.30	0.10	–0.30	0.50	0.30	0.00	0.00	–0.20	–0.60	0.60^*	–0.80^**	–0.10	0.60^*	–0.40	0.10	0.46	0.00	–0.20	–0.60^*	–0.10	–0.10
IV				0.70^*	–0.10	–0.30	0.60^*	–0.20	–0.30	0.60^*	0.20	0.60^*	–0.60^*	0.30	0.41	–0.60^*	–0.10	0.50	–0.67^*	0.30	0.20	0.60^*	0.10	0.10
RI					–0.60^*	0.70^*	0.90^**	–0.70^*	–0.20	–0.10	0.70	0.50	–0.50	0.30	0.05	–0.50	0.50	0.60^*	–0.67^*	0.20	0.70^*	0.50	0.60^*	0.60^*
PV						0.90^**	0.70^*	0.90^**	0.60^*	0.30	–0.50	–0.30	0.30	–0.50	0.67^*	0.30	–0.70^*	–0.80^**	0.46	–0.60^*	–0.50	–0.30	–0.99^***	–0.99^***
K232							0.90^**	0.70^*	0.70^*	0.10	–0.30	–0.10	0.10	–0.20	0.67^*	0.10	–0.40	–0.90^**	0.31	–0.70^*	–0.30	–0.10	–0.90^*	–0.90^*
K270								–0.60^*	–0.50	0.00	0.40	0.20	–0.20	0.10	–0.31	–0.20	0.30	0.80^**	–0.41	0.50	0.40	0.20	0.70^*	0.70^*
p- AV									0.30	0.40	–0.80^**	–0.60^*	0.60^*	–0.70^*	0.31	0.60^*	–0.90^**	–0.60^*	0.72^*	–0.30	–0.80^**	–0.60^*	–0.90^**	–0.90^**
SFA										–0.50	0.30	–0.10	0.10	–0.30	0.67^*	0.10	0.10	–0.90^**	0.21	–0.99^***	0.30	–0.10	–0.60^*	–0.60^*
MUFA											–0.60^*	0.20	–0.20	0.10	0.21	–0.20	–0.70^*	0.30	–0.15	0.50	–0.6	0.20	–0.30	–0.30
PUFA												0.60^*	–0.60^*	0.50	0.21	–0.60^*	0.90^**	0.10	–0.67^*	–0.30	0.99^***	0.60^*	0.50	0.50
SC													–0.99^***	0.90^**	0.41	–0.99^***	0.50	0.30	–0.98^***	0.10	0.60^*	0.99^***	0.30	0.30
SS														–0.90^**	–0.41	0.99^***	–0.50	–0.30	0.97^***	–0.10	–0.60^*	–0.98^***	–0.30	–0.30
Δ –7 –A															0.05	0.90^**	0.60^*	0.40	–0.87^**	0.30	0.50	0.90^**	0.50	0.50
SDO																–0.41	–0.15	–0.56	–0.29	–0.67^*	0.21	0.41	–0.67^*	–0.67^*
CA																	–0.50	–0.30	0.98^***	–0.10	–0.60^*	–0.97^***	–0.30	–0.30
CH																		0.20	–0.56	–0.10	0.90^**	0.50	0.70^*	0.70^*
TS																			–0.46	0.90^**	0.10	0.30	0.80^**	0.80^**
α-T																				–0.21	–0.67^*	–0.98^***	0.46	0.46
γ-T																					–0.30	0.10	0.60^*	0.6^*
δ -T																						0.60^*	0.50	0.50
TT																							0.30	0.30
OS																								0.99^***
SL