*2.4. Sensory Analysis*

Sensory analysis was performed by the panel of Agency for Agrofood Sector Services of Marche region (ASSAM), a fully-trained analytical taste panel recognized by the International Olive Oil Council (IOC) of Madrid, Spain, and by the Italian Ministry for Agriculture, Food, and Forestry Policy. The panel was composed of 8 assessors, 50% male and 50% female. The method applied was QDA (Quantitative Descriptive Analysis). A profile sheet IOC method T20 n. 15 modified by IBIMET-CNR and ASSAM was used, allowing to obtain a QDA of the oils' sensory profile and more complete description of the organoleptic properties of the sampled oils: the sensory assessors evaluated direct or retronasal aromatic olfactory sensations (aroma of olive fruity and green notes), gustatory sensations (olive fruity and bitterness) and tactile/kinesthetic sensation (pungency), organoleptic defects (Supplementary Table S1) as well as overall judgment. The sensory assessors had to rate the intensity of the different descriptors on a continuous 0–10 cm scale. Values of median of sensory data were calculated.

#### *2.5. Statistical Analysis*

The significance of differences at a 5% level was determined by one-way ANOVA using Tukey's test with Microsoft® Excel 2007/XLSTAT© (Version 2009.3.02, Addinsoft, Inc., Brooklyn, NY, USA). Sensory data were also processed for Principal Component Analysis (PCA) to explore data distribution patterns and to visualize the "distance" between oils produced following the differing storage times.

#### **3. Results and Discussion**

After interviewing olive growers, it was recorded that the olive and the correspondent Extra virgin olive oil (EVOO) samples collected (*n* = 228) were mainly composed of mixed varieties (blends) (45%), while the remaining samples were monovarietal from cv. Nostrana di Brisighella (25%), cv. Correggiolo (16%), cv. Leccino (8%) and other minor cultivars (6%). Furthermore, it was assessed that the olives' storage method was the same for all the analyzed samples: fruits were stored in small plastic bins with holes to allow for ventilation, and never in stacks nor in plastic or jute bags.

Olive storage duration before technological transformation ranged from 0 to over 7 days: chemical and sensory data were thus processed dividing them into four classes of storage times: ≤24 h, 2–3 days, 4–6 days, ≥7 days. Only 39% of olive samples were processed within 24 h, while 23% and 20% of olive samples were stored between 2 and 3 days and between 4 and 6 days, respectively; finally, 18% of samples were processed after 7 days of storage.

The content of free acids is an important quality factor, extensively used as the major criterion for the classification of olive oil at various commercial grades [16]. The values of free acidity, peroxide number, and K232 increased significantly along with the increase in olive storage duration (Table 1). There was a free acidity increase from 0.30% to 0.56% during the olive storage period studied; peroxide number from 6.96 to 9.56 mEqO2kg−1, K232 from 1.48 to 1.66 while K270 was not affected by time of storage probably because indicates secondary oxidation. It is interesting to note that all these values fall within the legal limit of the classification of extra virgin olive oil. This indicates that, although the oxidation process starts to take place during the olives' storage time, the phytosanitary state and the integrity of the raw material affect the speed of this process. The total phenol content of oil samples suffered a progressive reduction as olive storage duration proceeded: as observed in Table 1, oils produced within 24 h from the olive collection had a phenolic content of 243 mg kg−<sup>1</sup> of gallic acid while oils produced from olives stored for over 7 days presented 143.6 mg kg−<sup>1</sup> of gallic acid, a decrease of 41%. This phenol loss could be attributed both to bacteria and fungi proliferation on cellular fluid exuding from damaged fruits [17] and to endogenous oxidoreductases [18] as suggested by Clodoveo et al. [5], results of which were consistent with the data presented here. This impoverishment in the phenol fraction also affected the oils' stability, with a reduction from 28 h in oils produced within a day to 19 h in oils obtained from olives stored for more than a week, with a decrease of 30% (Table 1). These results agree with the studies of Vichi et al. [19] and Youssef et al. [20]. Our results partially agree with Pereira et al. [21], which found a significant decrease in oil stability during the first period of storage (0–7 days), while for peroxide number, free acidity and K232 and K270 the values were not significantly affected by storage duration. As explained by Pereira himself [21], the verified decrease in stability was due to the consumption of minor compounds such as phenols and tocopherols, that hindered the formation of peroxides.

**Table 1.** Quality index of oil samples extracted after different olive storage duration. Values are mean ± standard deviation. Values followed by different letters in the same column (a, b, c) were significantly different according to Tukey's test (*p* < 0.05).


Free acidity is expressed as g/100 g of oleic acid; peroxide number as mEq O2kg−<sup>1</sup> oil; OSI, oxidative stability index, as hours; total phenol as mg kg−<sup>1</sup> of gallic acid.

> Fatty acid and sterolic profile can be used as an exceptional compositional marker for olive oil authenticity [11]. Fatty acid composition of all of the EVOO samples extracted from olives after different storage duration was characterized by the high oleic acid content (Table 2), coherent with the cold climate of the region; the relationship between fatty acid composition and climate is well known [22]. Several studies reported that fatty acid composition in oils did not show any change as the period for which olives were stored prior to crushing increased [10,23] and neither did they even when the olive storage period was very long, e.g., 45 days, as reported by Gutierrez et al. [24]. However, other studies [20,21] found differences in fatty acids content during storage, in agreement with our results. Specifically, we found differences in the content of C16:0, C18:1 and C18:2. In fact, the C16:0 content tended to increase as the olive storage period lengthened, while concentrations of C18:1 and C18:2 did not show a clear trend.


**Table 2.** Fatty acid composition of oils produced by olives after different olive storage duration. Values are mean ± standard deviation. Values followed by different letters in the same column (a, b) were significantly different according to Tukey's test (*p* < 0.05).

The content of α-tocopherol, the naturally occurring form of vitamin E assimilated by the human body, found in oils obtained from olives belonging to different conservation classes showed a statistically significant decreasing trend (Table 3). Vitamin E is an antioxidant, working as peroxyl radical scavenger that terminates chain reactions [25]. As it is well documented, oxidation phenomena are the main cause of tocopherol degradation [26]; the data here presented showed that milling olives within 24 h from the collection was the only way to protect the tocopherol fraction. In this study, the decrement in α-tocopherol content found after 7 days of olive storage was about 17%, consistent with the reduction of 22% of total tocopherol content found by Yousfi et al. [10] for cv. Arbequina and Pereira and colleagues [21] for cv. Verdeal Transmontana. An important and significant decrease in α-tocopherol after a short (48 h) olive storage period was as well found for cv. Nostrana di Brisighella oils while not for cv. Correggiolo oils [27].

**Table 3.** Tocopherol and carotenoid content in oil samples after different olive storage durations. Values are mean ± standard deviation. Compounds are expressed as mg of relative standard compound per kg of oil. Values followed by different letters in the same column (a, b) were significantly different according to Tukey's test (*p* < 0.05).


In the present study, carotenoid pigments decreased during the first three storage times analyzed, in agreement with other works [28,29]. At the last time of storage duration analyzed (>7 days), olive oils showed an increase in carotenoid content, in line with Yousfi and colleagues [10], who hypothesized a greater extractability of the pigments in olives during storage, due to the degradation of the chloroplast membranes; the degradation was found to be a consequence of the growing dehydration of the olives during storage [10].

A correlation analysis was carried out to quantify the intensity of the connection between EVOO properties and olive storage duration (Table 4). A positive correlation (Pearson) with storage duration was found for the parameter acidity, peroxide number, K232 and palmitic acid content, while palmitoleic and stearic acid, total phenol content and oil stability exhibited a negative correlation, diminishing with the increase in storage duration. In the case of fruits left for long periods before transformation, in the produced oils was observed, in addition to an increase in free acidity, even a gradual depletion in the content of oleic acid and total phenols, with the consequent reduction in stability during storage. It is also important to underline the correlation (r = 0.388, *p* ≤ 0.0001) between oleic acid and OSI found in this work (data not shown). In fact, a high concentration of oleic acid enhances the oil stability and EVOO oxidative stability is mainly linked to its fatty acid composition, therefore the induction period is the result of the fatty acid composition and the simultaneous activity of various prooxidant and antioxidant factors endogenous in the oils [30].


**Table 4.** Pearson correlation between oil chemical parameters and olive storage duration.

The sensory profile was modified according to the olive storage duration before processing, with the main positive attributes (olive fruity, green notes, bitterness and pungency) undergoing a statistically significant reduction with the increasing of olive storage days (Table 5). This result was in line with the previously recorded decrease in total phenols: the typical bitter taste and pungent note of fresh EVOO rich in total phenols decreased in intensity as olive storage duration lengthened. The intensity of the defects perceived by the sensory assessors increased with the progress of the days of olives storage (Table 5). Even in oils processed within 24 h, an intensity of the defect of 0.29 was recorded, probably attributable to the percentage of olives harvested late and thus overripe. In detail, the perceived defects were attributable to incorrect management of the raw material: in fact, they were 67% for the fusty defect, 29% for the vinegary and only 4% for the musty defect. During prolonged olive storage duration, the drupe tissues are damaged, resulting in the secretion of fluids favoring the growth of undesirable microorganisms [17]; increased temperature can also increase drupe respiratory activity, leading to undesirable metabolic processes accelerating fruit deterioration and characterized by the fusty sensory defect [31].

**Table 5.** The intensity of sensory attributes (aroma of olive fruity and green notes, flavor of bitterness, pungency and defect) of oil samples related to different olive time storage. Values are median ± standard deviation. Values followed by different letters in the same column (a, b, c) were significantly different according to Tukey's test (*p* < 0.05).


The Principal Component Analysis (PCA) of the sensory data explained 91.4% of the variability and confirmed the strong influence of olive storage duration on the sensory characteristics of the oils produced (Figure 1). Most of the samples that transformed within 24 h are, in fact, positioned in the first quadrant of the PCA, showing the greatest intensities of positive attributes such as the aroma of olive fruity and green notes and bitterness. The small percentage of oil samples that, despite having been transformed within 24 h, are positioned in the fourth quadrant relative to the presence of sensory defects, is probably attributable to oils produced from overripe olives [32].

**Figure 1.** Principal component analysis (PCA) plot of sensory data.

#### **4. Conclusions**

The objective of this study, unlike other studies which investigated preprocessing storage as a way to modulate a positive reduction in the bitter taste of phenol-rich varieties with the aim of improving consumer acceptance [31], was to identify the criticality of the olive storage phase, highlighting its influence on the depletion of the EVOOs' chemical and sensory characteristics.

The authors are aware that this experimental design included many variables that affect the final quality of the oil: cultivar, ripeness, cultivation environment, seasonality and variation of technological parameters of extraction. However, thanks to the high rate of sampling, repeated for three consecutive production years, the single effect of the different variables was reduced. From this study, it emerged that some quality, nutritional and sensory parameters were affected by olive storage duration, independent of the varietal composition of the starting material. However, variety and ripeness degree influence the time window available to leave the olives on trees [33]. The knowledge of the effect of the harvest time window (early harvest or late harvest) on the olive oil final quality is important especially in years with late fly attacks, when it is recommended to harvest early rather than treating with chemicals, since a sustainable olive growing satisfies consumers who are increasingly attentive to the consumption of genuine and healthy products.

This study was carried out on purpose in a practical context characterized by all the limits listed above, thus the results provide an important photograph of the critical points of the olive storage phase from harvesting to pressing. By acting on the critical points, it is possible to improve the chain of olive oil production.

Free acidity, peroxide number, K232, total phenols, stability, α-tocopherol, lutein, βcarotene and organoleptic properties significantly decreased between the first and second storage interval, thus after 24 h of olive storage the final EVOO's quality was already substantially impoverished. Many specifications for PDO and PGI productions indicate

48 h as the maximum allowed storage duration; however, it is important to underline that by keeping storage within 24 h it is possible to maximize the potential of the olives, thus producing the oils with the highest nutritional and sensory properties expected by selecting cultivars known for the high quality of the final product. While PDO and IGP oils are produced according to strict production regulations, this study is aimed at blends productions that represent most of the processed olives in all of the Italian regions: for this kind of production the improving of the crucial phase of olive storage duration is important, and the results of this study are clearly significant for their applicability. The purpose of this work was to provide guidelines for obtaining a high-quality product at the time of processing, initial high quality being pivotal during the oil storage phase. The associations of olive producers guarantee their associates the supply of plastic aerated bins, together with guidelines aimed at reducing the olives' storage times from harvesting to processing. Ensuring the chemical and sensory oil quality during shelf life is, in fact, becoming the purpose of the most recent labeling regulations [34]: the community regulation states that what is indicated on the label should correspond with what is expected at the end of the product's shelf life.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/foods10102296/s1. Table S1: Definition of sensory descriptors.

**Author Contributions:** Conceptualization, A.R.; methodology, A.R., G.B.; formal analysis, L.M., G.B.; investigation A.R., G.B.; resources, A.R., G.B.; data curation, A.R., G.B.; writing—original draft preparation, A.R., L.M.; writing—review and editing, L.N., G.B.; supervision, A.R.; project administration, A.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors gratefully thank Matteo Mari for his technical support, Massimiliano Magli for his support in the statistical analysis, the panel of ASSAM-Marche for sensory analysis and Simon Brown for providing language help.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

