Obtaining an Extract Rich in Phenolic Compounds from Olive Pomace by Pressurized Liquid Extraction

The olive oil industry produces large volumes of wastes, which are also potential sources of bioactive compounds by developing healthy and/or functional foods. Extraction of phenolic compounds from the residues of the olive oil is mainly carried out with solvents. However, there is currently a growing public awareness about the use of organic solvents in food processing, which has pointed out the need for the application of clean technologies such as pressurized liquid extraction (PLE). Therefore, the aim of this research was to optimize the phenolic compound extraction from olive pomace by PLE, establishing the qualitative and quantitative phenolic profile by HPLC-ESI-TOF/MS. The extraction design to recover phenolics from olive pomace demonstrates a great compositional variability of PLE extracts obtained under different experimental conditions. Indeed, quantitative results have pointed out the selectivity of PLE extraction when this technique is applied to the treatment of olive pomace. PLE-optimized conditions showed higher total phenolic compound content than conventional extraction (1659 mg/kg d.w. and 281.7 mg/kg d.w., respectively). Among these phenolics, the quantity of secoiridoids and flavonoids in the optimized PLE extract was three and four times higher than in conventional extracts. Furthermore, optimal PLE conditions allowed to obtain an enriched hydroxytyrosol extract which was not detected in the conventional one.


Introduction
The food industry produces large volumes of both solid and liquid wastes, which represent a disposal and potentially environmental pollution problem. However, they are also potential sources of bioactive compounds that can be recovered and used as valuable substances by developing healthy and/or functional foods [1,2]. In this sense, the production of extra virgin olive oil (EVOO) is associated with both the generation of large quantities of wastes and the loss of phenolic compounds during the process by the partitioning between oil and by-products (olive mill waste water, olive pomace, storage by-products, and filter cake) [3,4]. D-oleuropein aglycone, demethyl oleuropein, and an aldehydic form of decarboxymethyl elenolic acid. In addition, two flavonoids (luteolin and luteolin-7-glucoside) and two lignans (pinoresinol and acetoxypinoresinol) were also detected. A non-phenolic polar compound was also identified as quinic acid. Most of these compounds have been previously described in olive oil and olive by-products [4,8,19,23,27,28]. Table 2 shows the quantitation of individual phenolic compounds for each treatment of PLE system. The total phenolic content (PC) of OP-PLE extracts ranged from 241.1 to 1141.3 mg/kg d.w OP. Figure 2 includes the total phenolic compounds and the total content in secoiridoids, phenolic alcohols, flavonoids, and lignans. Secoiridoids reached the major concentration, ranging from 103.4 to 517 mg/kg d.w OP (42.9-57.6%, respectively). The highest amounts of secoiridoids were obtained for secoiridoid derivative (m/z 407), hydroxy oleuropein (m/z 555), and oleuropein (m/z 539). With regard to phenolic alcohols, hydroxytyrosol (m/z 153) and oxidized hydroxytyrosol (m/z 151) were identified and quantified. Their concentrations ranged from 0 to 675.6 mg/kg d.w OP. The highest phenolic alcohol contents were obtained in conditions T4, T5, and T12, where the higher temperature and water content, the higher the phenolic alcohol content. This result could be explained because the high extraction temperature and water content in solvent extraction could generate the hydrolysis of secoiridoids into phenolic alcohols and acidic moieties. These results agree with those reported by [29].  Internal standard (Naringenin). Table 2 shows the quantitation of individual phenolic compounds for each treatment of PLE system. The total phenolic content (PC) of OP-PLE extracts ranged from 241.1 to 1141.3 mg/kg d.w OP. Figure 2 includes the total phenolic compounds and the total content in secoiridoids, phenolic alcohols, flavonoids, and lignans. Secoiridoids reached the major concentration, ranging from 103.4 to 517 mg/kg d.w OP (42.9-57.6%, respectively). The highest amounts of secoiridoids were obtained for secoiridoid derivative (m/z 407), hydroxy oleuropein (m/z 555), and oleuropein (m/z 539). With regard to phenolic alcohols, hydroxytyrosol (m/z 153) and oxidized hydroxytyrosol (m/z 151) were identified and quantified. Their concentrations ranged from 0 to 675.6 mg/kg d.w OP. The highest phenolic alcohol contents were obtained in conditions T4, T5, and T12, where the higher temperature and water content, the higher the phenolic alcohol content. This result could be explained because the high extraction temperature and water content in solvent extraction could generate the hydrolysis of secoiridoids into phenolic alcohols and acidic moieties. These results agree with those reported by [29].

Quantitative Characterization of Olive Pomace Phenolic Compounds Obtained by PLE
Flavonoids ranged from 3.1 to 199.9 mg/kg d.w OP (1.2-22.6%), luteolin being the flavonoid recovered in major amount in all the treatments. Concerning lignans, different amounts of pinoresinol and acetoxypinoresinol were quantitated in the extracts. Acetoxypinoresinol was the main phenolic, whose concentration ranged from 17.6 to 48.7 mg/kg d.w OP (1.7-6%).
The results showed that although the phenolic compound profile was similar for all treatments of the OP-PLE design, the phenolic compound content was different among runs. This behavior is characteristic of the PLE method, which presents differences in the selectivity of extraction [15]. The extraction factors such as temperature and water-ethanol ratio have been associated with the dielectric constant. Thus, the lower the dielectric constant, the higher the flavonoids content (r 2 = 0.7336). However, no correlation was found between the dielectric constant and each phenolic group. It is important to consider that oxidized hydroxytyrosol (m/z 151), hydroxytyrosol (m/z 153), hydroxylated product of decarboxymethyl elenolic acid (m/z 199), hydroxy oleuropein (m/z 555), and luteolin-7-O-rutinoside (m/z 593) were identified in OP-PLE. These phenolic compounds have been reported to form by the oxidation reactions and/or hydrolysis of complex polyphenols [29].  extraction factors such as temperature and water-ethanol ratio have been associated with the dielectric constant. Thus, the lower the dielectric constant, the higher the flavonoids content (r 2 = 0.7336). However, no correlation was found between the dielectric constant and each phenolic group. It is important to consider that oxidized hydroxytyrosol (m/z 151), hydroxytyrosol (m/z 153), hydroxylated product of decarboxymethyl elenolic acid (m/z 199), hydroxy oleuropein (m/z 555), and luteolin-7-O-rutinoside (m/z 593) were identified in OP-PLE. These phenolic compounds have been reported to form by the oxidation reactions and/or hydrolysis of complex polyphenols [29].

Flavonoids; and
Molecules 2019, 24, x 5 of 17 extraction factors such as temperature and water-ethanol ratio have been associated with the dielectric constant. Thus, the lower the dielectric constant, the higher the flavonoids content (r 2 = 0.7336). However, no correlation was found between the dielectric constant and each phenolic group. It is important to consider that oxidized hydroxytyrosol (m/z 151), hydroxytyrosol (m/z 153), hydroxylated product of decarboxymethyl elenolic acid (m/z 199), hydroxy oleuropein (m/z 555), and luteolin-7-O-rutinoside (m/z 593) were identified in OP-PLE. These phenolic compounds have been reported to form by the oxidation reactions and/or hydrolysis of complex polyphenols [29]. Lignans. Oxidized hydroxytyrosol  Flavonoids ranged from 3.1 to 199.9 mg/kg d.w OP (1.2-22.6%), luteolin being the flavonoid recovered in major amount in all the treatments. Concerning lignans, different amounts of pinoresinol and acetoxypinoresinol were quantitated in the extracts. Acetoxypinoresinol was the main phenolic, whose concentration ranged from 17.6 to 48.7 mg/kg d.w OP (1.7-6%).
The results showed that although the phenolic compound profile was similar for all treatments of the OP-PLE design, the phenolic compound content was different among runs. This behavior is characteristic of the PLE method, which presents differences in the selectivity of extraction [15]. The extraction factors such as temperature and water-ethanol ratio have been associated with the dielectric constant. Thus, the lower the dielectric constant, the higher the flavonoids content (r 2 = 0.7336). However, no correlation was found between the dielectric constant and each phenolic group. It is important to consider that oxidized hydroxytyrosol (m/z 151), hydroxytyrosol (m/z 153), hydroxylated product of decarboxymethyl elenolic acid (m/z 199), hydroxy oleuropein (m/z 555), and luteolin-7-O-rutinoside (m/z 593) were identified in OP-PLE. These phenolic compounds have been reported to form by the oxidation reactions and/or hydrolysis of complex polyphenols [29].

Optimization of Recovering Phenolic Compounds from Olive Pomace by PLE Using Response Surface Methodology (RSM)
When characterizing the profile of phenolic compounds of the different treatments by PLE, the occurrence of hydroxytyrosol was observed in some extracts of olive pomace. Hydroxytyrosol presents comparative advantages over other polyphenols in olive residues, due to its higher antioxidant capacity and biological effects such as prevention of heart disease, tumor, and antithrombotic effect, among others [18,30]. In addition, the European Food Safety Agency (EFSA) approved health claims associated with consumption of hydroxytyrosol. In this context, the hydroxytyrosol content and its derivatives were used as response variables for the experimental design of olive pomace extracts by PLE. Table 3 summarizes the experimental conditions (independent variables values: Ethanol content and extraction temperature) and response variables (hydroxytyrosol, hydroxy D-oleuropein aglycone, hydroxy-oleuropein, demethyloleuropein, decarboxymethyl oleuropein aglycone, and oleuropein content and yield). RSM was used to optimize each response variable considering linear, quadratic, and cross-product interactions of the independent variables at the 95% confidence level. The analysis of the variance (ANOVA) for the extraction of phenolic compounds from OP by PLE is shown in Table 4. Non-significant terms were removed from the equation, but when quadratic or cross-product interactions of the independent variables were significant, the linear forms of independent variables were considered in the quadratic equation, because they are fundamental elements of the mathematical model.
The hydroxytyrosol content ranged from 0 to 258.3 mg/kg ( Table 3). The linear, quadratic, and cross-product forms of ethanol content (E) and extraction temperature (T) were significant on the hydroxytyrosol content. The model explained 78.8% of the variability in hydroxytyrosol content (R 2 adj. d.f., Table 4).
The quadratic regression equation describing the effect of the independent variables on the hydroxytyrosol content was the following (Equation (1)):  As it can be seen in surface graph (Figure 3), the hydroxytyrosol content was higher when high extraction temperature and low ethanol content where applied. The hydroxy D-oleuropein aglycone and hydroxy D-oleuropein ranged from 0 to 95 mg/kg and from 0.4 to 134.6 mg/kg, respectively ( Table 3). The effect of temperature and ethanol content quadratic forms were significant on hydroxy D-oleuropein aglycone content, while the quadratic form of temperature was only significant on hydroxy D-oleuropein content. In both response variables, the linear forms of temperature and ethanol content were considered in the quadratic equations. The model explained 75.1% and 64.7% of the variability in hydroxy D-oleuropein aglycone and hydroxy D-oleuropein content, respectively (R 2 adj. d.f., Table 4). The quadratic regression equations describing the effect of the independent variables on hydroxy D-oleuropein aglycone (Equation (2)) and hydroxy D-oleuropein (Equation (3)) content are as follows:   Figure 1 shows the HPLC-DAD-ESI-TOF/MS chromatograms and the tentative identification of phenolic compounds in the olive pomace extract obtained by conventional extraction system (OP) and olive pomace extract by optimal PLE (OP-PLE). In both olive pomace extracts, it is possible to identify different phenolic compounds which have previously been reported in olive fruits and olive derivatives [31]. Table 5 shows phenolic compound content of both OP and OP-PLE extracts. The total phenolic compound content in OP-PLE extract was higher than in OP (1659 mg/kg d.w. and 281.7 mg/kg d.w., respectively), showing a total phenolic compound value equivalent to 588.9% with respect to the extract obtained by conventional extraction. This value is much higher than those previously reported (201-256%) [8]. These results can be attributed to the high temperature and pressure of the PLE extraction method, process conditions that improve the interaction capacity between the phenolic compounds and the extraction solvent (ethanol and water mixture). On the other hand, the surface tension and the viscosity of the extraction solvent are reduced by the increase of temperature, which leads to better wetting and penetration in the matrix (olive pomace), thus increasing the mass transfer and therefore the extraction of phenolic compounds [17,19,22,23,27,28].

Characterization of Olive Pomace Extracts Obtained Under Optimal PLE Conditions (OP-PLE)
Higher hydroxytyrosol contents than that of this study were reported in extracts of olive pomace (var. Picual) (2800 mg hydroxytyrosol/kg d.w.) using ethanol-water as solvent and static-dynamic process [7,18]. However, the initial phenolic compounds concentration in a matrix depends on several agronomic (including cultivar, ripening stage, geographic origins, and tree irrigations, among others) and technological factors (like the process temperature and water content) [3,8,32]. Future studies are warranted to compare both techniques using the same matrix.
In OP extract, the main phenolic compound class was secoiridoids, with more than half of the total phenolic content (54.8%): Secoiridoid derivated 1 (16.9%), derived from oleuropein aglycone (8.6%), oleuropein (7.7%), and ligstroside (7.0%). These results are in agreement with other studies, in which the main phenolic compounds are secoiridoids (between 50 and 70%) [8]. Flavonoids are the second chemical group in importance (23%), with luteolin being the main compound of this subclass (17.7%). In contrast, in OP-PLE extract obtained under optimum conditions, the main phenolic compound class is phenolic alcohols with 42.5% (phenolic compound class that were not found in the OP), followed by secoiridoids (26.4%) and flavonoids (16.4%), with luteolin being the main phenolic compound (13.2%). The appearance of phenolic alcohols and the increase of the elenolic acid derivatives in OP-PLE could be related to degradation reactions of secoiridoids, followed by several reactions, such as oxidation, hydration, and loss of the carboxylic and carboxymethyl groups due to the high temperatures used in the extraction method [33][34][35].
On the other hand, the OP-PLE extract obtained under optimal conditions had a processing yield of 17.2%, which is higher to that found in olive extracts obtained with the PLE method. Table 5. Phenolic compound content of olive pomace extracts by pressurized liquid extraction in optimal conditions (OP-PLE) and olive pomace extract by conventional extraction (OP). The demethyl oleuropein concentration ranged from 0 to 21.6 mg/kg ( Table 3). The quadratic form of temperature was significant; however, the linear form was considered in the quadratic equations. The linear form of ethanol content and cross-product forms of ethanol as extracting solvent and temperature were significant for demethyl oleuropein content. The model explained 80.6% of the variability (R2 adj. for d.f., Table 4).
The quadratic regression equation of demethyl oleuropein content (Equation (4)) is the following: As can be seen in the response surface graph (Figure 3), the highest demethyl oleuropein content was achieved with high temperature and high ethanol content.
The decarboxymethyl oleuropein aglycone concentration ranged from 0 to 19.1 mg/kg (Table 3). Only the quadratic form of temperature and ethanol content were significant; however, the linear form and cross-product forms of ethanol as extracting solvent and temperature were considered in the quadratic equations for decarboxymethyl oleuropein aglycone. The model explained 83.6% of the variability (R2 adj. for d.f., Table 4).
The quadratic regression equation of decarboxymethyl oleuropein aglycone content (Equation (5)) is: As can be seen in the response surface graph (Figure 3), the highest decarboxymethyl oleuropein aglycone content was achieved with intermediate values of temperature and ethanol content.
The oleuropein concentration ranged from 3.7 to 113.4 mg/kg ( Table 3). The linear, quadratic form of temperature, linear form of ethanol content and cross-product forms of ethanol as extracting solvent and temperature were significant, so were considered in the quadratic equations for oleuropein content. The model explained 88.6% of the variability (R2 adj. for d.f., Table 4).
The quadratic regression equation of oleuropein content is the following (Equation (6)): As can be seen in the response surface graph (Figure 3), the highest oleuropein content was achieved with intermediate temperature and high ethanol content. According to the information described above, it is observed that the highest hydroxytyrosol content was obtained in the treatments with high extraction temperatures (T4 and T5), while hydroxy-D-oleuropein aglycone, dimethyl oleuropein, and decarboxymethyl-oleuropein aglycone were not detected in these experiments. In contrast, hydroxytyrosol was not detected in the treatments carried out at the lowest temperatures (T6, T8, and T11). These results show that high extraction temperatures would cause rupture of secoiridoides to phenolic alcohols [14].
Regarding oleuropein, the highest value was obtained in the treatment performed at high temperature with high ethanol content (T9).
In relation to yield, this ranged from 3.7 to 25.7% (Table 4). The linear, quadratic, and cross-product forms of ethanol as extracting solvent and temperature were significant for yield (Equation (7)). The model explained 93.6% of the variability (R2 adj. for d.f.) in yield (Table 4).
As can be seen in the surface graph (Figure 3), the yield showed the highest values when using high temperature extraction and a minor proportion of ethanol in the extraction solvent (T4 and T5). Finally, multiple optimization taking into account all response variables was evaluated (Desirability Function). Figure 4 shows the surface response graphic. The optimal conditions for the extraction of phenolic compounds and yield were 136.5 • C and 52.3% of ethanol in the ethanol:water mixture. In the graph of the response surface, the optimum values are located at the yellow zone with 0.7 of desirability. Figure 1 shows the HPLC-DAD-ESI-TOF/MS chromatograms and the tentative identification of phenolic compounds in the olive pomace extract obtained by conventional extraction system (OP) and olive pomace extract by optimal PLE (OP-PLE). In both olive pomace extracts, it is possible to identify different phenolic compounds which have previously been reported in olive fruits and olive derivatives [31]. Table 5 shows phenolic compound content of both OP and OP-PLE extracts. The total phenolic compound content in OP-PLE extract was higher than in OP (1659 mg/kg d.w. and 281.7 mg/kg d.w., respectively), showing a total phenolic compound value equivalent to 588.9% with respect to the extract obtained by conventional extraction. This value is much higher than those previously reported (201-256%) [8]. These results can be attributed to the high temperature and pressure of the PLE extraction method, process conditions that improve the interaction capacity between the phenolic compounds and the extraction solvent (ethanol and water mixture). On the other hand, the surface tension and the viscosity of the extraction solvent are reduced by the increase of temperature, which leads to better wetting and penetration in the matrix (olive pomace), thus increasing the mass transfer and therefore the extraction of phenolic compounds [17,19,22,23,27,28].

Characterization of Olive Pomace Extracts Obtained Under Optimal PLE Conditions (OP-PLE)
Higher hydroxytyrosol contents than that of this study were reported in extracts of olive pomace (var. Picual) (2800 mg hydroxytyrosol/kg d.w.) using ethanol-water as solvent and static-dynamic process [7,18]. However, the initial phenolic compounds concentration in a matrix depends on several agronomic (including cultivar, ripening stage, geographic origins, and tree irrigations, among others) and technological factors (like the process temperature and water content) [3,8,32]. Future studies are warranted to compare both techniques using the same matrix. In OP extract, the main phenolic compound class was secoiridoids, with more than half of the total phenolic content (54.8%): Secoiridoid derivated 1 (16.9%), derived from oleuropein aglycone (8.6%), oleuropein (7.7%), and ligstroside (7.0%). These results are in agreement with other studies, in which the main phenolic compounds are secoiridoids (between 50 and 70%) [8]. Flavonoids are the second chemical group in importance (23%), with luteolin being the main compound of this subclass (17.7%). In contrast, in OP-PLE extract obtained under optimum conditions, the main phenolic compound class is phenolic alcohols with 42.5% (phenolic compound class that were not found in the OP), followed by secoiridoids (26.4%) and flavonoids (16.4%), with luteolin being the main phenolic compound (13.2%). The appearance of phenolic alcohols and the increase of the elenolic acid derivatives in OP-PLE could be related to degradation reactions of secoiridoids, followed by several reactions, such as oxidation, hydration, and loss of the carboxylic and carboxymethyl groups due to the high temperatures used in the extraction method [33][34][35].
On the other hand, the OP-PLE extract obtained under optimal conditions had a processing yield of 17.2%, which is higher to that found in olive extracts obtained with the PLE method.

Conventional Extraction Procedure of Phenolic and Other Polar Compounds from Olive Pomace (OP)
Conventional extraction of phenolic compounds from olive pomace was performed by solid-liquid extraction. Lyophilized olive pomace (5.0 g d.w.) was soaked for 120 min with 20 mL of a mixture of methanol:water (80: 20). After this, the samples were centrifuged at 10,000 rpm for 15 min, and the supernatants were collected and filtered through a 0.45 µm filter. Each procedure was carried out in triplicate. The OP extracts were frozen at -20 • C until analysis.

Extraction of Phenolic Compounds from Olive Pomace by PLE (OP-PLE)
PLE was performed using a Dionex ASE 350 Accelerated Solvent Extractor (Thermo Fisher Scientific, Leicestershire, UK). All extractions were done using 34-mL extraction cells, containing 5 g of lyophilized olive pomace mixed homogeneously with 10 g of sand. Prior to extraction of phenolic compounds, a preliminary clean-step based on the use of n-hexane as the solvent and 1500 psi at room temperature as the experimental conditions was carried out to remove the lipophilic fraction from the olive pomace. After this step, the extraction of phenolic compounds from the olive pomace was performed according to a central composite design, with a total of 12 runs (4 experimental points, 4 axial points, and 4 central points). The ethanol percentage (0 to 100%) and temperature (40 to 176 • C) were evaluated as independent variables to cover a wide range of dielectric constants (from 19 to 65.5 F/m, Table 6). The pressure and extraction time were 1500 psi and 20 min, respectively. All of the experiments were conducted randomly to avoid systematic errors. The obtained extracts were protected from light, filtered through a 0.45 µm regenerated cellulose filter and evaporated under vacuum in Speed Vac (Thermo Scientific, Leicestershire, UK). Yield (Y), hydroxytyrosol, and hydroxytyrosol-containing compounds were used as response variables for the experimental design of olive pomace extracts by PLE (hydroxityrosol (HYTY), hydroxyoleouropein aglycone, hydroxyoleuropein, demethyl oleuropein, decarboxymethyl oleuropein, and oleouropein).
Response surface methodology (RSM) was applied to determine the optimal conditions for the OP-PLE system by multiple response optimization using the desirability function (DF) where the response variables were maximized. The data were fitted to a second-order regression model according to Equation (8).
where Y is the response; subscripts i and j range from 1 to the number of variables (n = 2); b 0 is the intercept term; b i values are the linear coefficients; b ij values are the quadratic coefficients; and X i and X j are the levels of independent variables. The HPLC analyses were performed in a high-performance resolution liquid chromatography (HPLC) system (Agilent Technologies, Waldbronn, Germany) equipped with a vacuum degasser, autosampler, a binary pump, and diode-array-detector (DAD). This equipment was coupled to a time-of-flight mass spectrometry (TOF-MS) microTOF (Bruker Daltonik, Bremen, Germany). The TOF mass spectrometer was equipped with a model G1607A ESI interface (Agilent Technologies, Palo Alto, CA, USA) operating in negative ion mode. The analytical column used was a 150 mm × 4.6 mm internal diameter, 1.8-µm Zorbax Eclipse Plus C18 (Agilent Technologies, Palo Alto, CA, USA).
The compounds separated were monitored with DAD (240 and 280 nm) and MS. At this stage, the use of a splitter was required for the coupling with the MS detector, as the flow arriving to the TOF detector had to be 0.25 mL/min in order to ensure reproducible results and a stable spray. External mass spectrometer calibration was performed with sodium acetate clusters (5 mM sodium hydroxide in water/2-propanol 1/1 (v/v), with 0.2% of acetic acid) in high-precision calibration (HPC) regression mode.
The phenolic compound quantification was performed with calibration curves elaborated with commercial standards: Hydroxytyrosol was used to quantify hydroxytyrosol and oxidized hydroxytyrosol; tyrosol was used to quantify tyrosol and ligstroside; caffeic acid was used to quantify verbascoside; oleuropein was used to quantify oleuropein, its isomers and derivatives, oleosides and elenolic acid derivatives, and secoiridoids; pinoresinol was used to quantify pinoresinol and acetoxypinoresinol; luteolin was used to quantify luteolin; and luteolin 7-glucoside was used to quantify luteolin 7-glucoside and luteolin 7-rutinoside. All phenolic compound standard solutions were prepared at a concentration of 1000 mg/L by dissolving the appropriate amount of the compound in methanol and then serially diluting to working concentrations (0.5 to 50 mg/L). Naringenin was added at 25 µg/mL and used as internal standard.

Statistical Analysis
Linear regression (95% confidence limit) was used to determine the correlation between dielectric constant and each phenolic compound class for the design from olive pomace extracts by PLE. A one-way analysis of variance (ANOVA) and Duncan's multiple range test were performed to test for differences in phenolic compound content between olive pomace extracts by pressurized liquid extraction in optimal conditions (OP-PLE) and olive pomace extract by conventional extraction (OP). The statistical analyses were performed using Statgraphics Centurion XV (StatPoint Inc., Warrenton, VA, USA, 2011).

Conclusions
The extraction design of phenolic compounds from olive pomace by PLE demonstrates the great variability of extracts that can be obtained by changing the extraction conditions, presenting differences in the selectivity of extraction. The olive pomace extract obtained under optimized conditions showed a higher concentration in phenolics than that obtained by conventional extraction, besides presenting hydroxytyrosol, a compound which was not found in the extraction by maceration. The olive pomace extract obtained under optimized PLE conditions can be used in the design of bioactive food ingredients, as well as nutraceuticals.