Abstract
Introduction
Citrus peel, one of the main byproducts of the food industry, is an important source of phenolic compounds with preventive and protective effects. The analysis of these compounds has been widely described, however, information on the purification methods of the extracts is limited.
Objective
The objective of the present study is to determine the purification method that favors the obtaining of phenolic compounds from citrus peel extracts.
Methods
Oranges (C. sinensis) and mandarins (C. reticulata) were purchased from a local supermarket (Valencia, Spain). The peels were separated manually and cut into pieces of 25 mm2. An ultrasound-assisted extraction was performed (30 min, 400 W, < 40 °C). Purification by Solid Phase Extraction (SPE) was carried out using cartridges with 200 mg of C18. The QuEChERS procedure was performed using 2 ml DisQUE™ Tubes. The profile of phenolic compounds was analyzed by HPLC-UV.
Results
The major compounds in the samples were narirutin and hesperidin. Differences were determined according to both purification methods (p < 0.05). Hesperidin was higher in orange peel samples (2229 µg/g FW), while the highest amount of narirutin was obtained in tangerine peel (440 µg/g FW).
Conclusion
The sample purification methods are selective. The QuEChERS method showed a greater recovery of phenolic acids and quercetin. The content of phenolic acids was higher in mandarin peel samples.
Introduction
The recovery of phenolic compounds from citrus waste is one of the main activities of food science, because they have preventive and protective effects against cardiovascular and degenerative diseases, and some types of cancer. 1 Citrus waste is the main by-product generated by the food industry, accounting for approximately between 15 and 25 thousand tons. 2 Likewise, these wastes, mainly the peel, are an important source of phenolic compounds. There are several studies interested in knowing the profile of phenolic compounds from different plant samples, and the analysis of phenolic compounds from citrus residues is widely described in previous research.3–6 However, publications on clean-up methods for the purification of extracts that can favor the selectivity of polyphenol analytical processes are limited.
Solid-phase extraction has also been one of the most studied methods for the extraction of plant extracts, but little research has been done to adapt this method for the purification of extracts to obtain concentrated samples of phenolic compounds from plant residues. This method was used to eliminate pigments, waxes, glycosides, fats, and sugars, among other analytes 7 that could interfere with the results of interest, 8 and also, the process is carried out under conditions considered environmentally friendly.9,10 In addition, octadecylsilane (C18), one of the most used solid phases, can eliminate fractions with high lipid content, and the cartridge commonly used in food matrices contains between 4 and 200 mg of C18. 7
It has recently been observed that the Quick, Easy, Cheap, Effective, Rugged, and Safe (QuEChERS) is not only useful for separating contaminants in plant samples but is also postulated as a method for obtaining phenolic extracts free of other analytes. This method has been frequently used for the separation of pesticides from fruit and vegetable samples. 11 QuEChERS is based on the dispersion of salts to extract (salting effect) and the isolation of a wide range of analytes from very complex matrices. 12 This method has been previously used as a clean-up method for the analysis of limonene in citrus juices; 13 however, studies with plant peel samples describe the use of QuEChERS for the analysis of fungicide residues, 14 bactericides, 15 among other pesticides.16–18 On the other hand, recent studies describe the use of this clean-up method for the separation and analysis of phenolic compounds from fruit and vegetable extracts.19–21
Therefore, given the effectiveness of the QuEChERS method as a clean-up procedure for the purification of extracts from contaminant analytes, the study aims to test whether SPE and QuEChERS can provide purer phenolic extracts. For this purpose, the phenolic compound profile of orange and mandarin peel extracts, used as source of these analytes, obtained by ultrasounds and employing both cleaning methods is determined, to compare the content of compounds by each method using a previously validated chromatographic analysis (HPLC-UV).
Methods
The present study is a quantitative, descriptive-comparative analysis of the phenolic content in two citrus peel extracts, purified by SPE and QuEChERS, respectively.
Samples
Citrus fruits were originated from the province of Valencia. The peel was obtained from oranges (C. sinensis) and mandarins (C. reticulata) purchased from a local supermarket (Valencia, Spain), and collected from October to November 2022. After cleaning, the peel was separated manually from the fruit and cut into pieces of 25 mm2.
Chemicals
4-hydroxybenzoic acid (99%), catechin (≥99.0%), vanillic acid (≥97.0%), chlorogenic acid (≥95.0%), narirutin (≥98%), caffeic acid (≥99.0%), naringin (≥95.0%), hesperidin (≥97.0%), p-coumaric acid (≥98.0%), ferulic acid (≥99.0%), naringenin (≥95%), kaempferol (≥90%), hesperetin (≥98.0%), and trans-cinnamic acid (≥98.0%) were purchased from Sigma-Aldrich (Steinheim, Germany). Apigenin (96.86%), quercetin dihydrate (97.02%), and rutin trihydrate (97.67%) were obtained from HWI Analytik GmbH (Ruelzheim, Germany). Analysis grade ethanol, HPLC grade methanol (MeOH), analysis grade hydrochloric acid, and HPLC grade acetonitrile were purchased from J.T. Baker Chemical Co. (Deventer, The Netherlands). Analysis grade formic acid was purchased from PanReac AppliChem (Darmstadt, Germany). Ultrapure water (18.2 MΩ cm) was used for preparing the mobile phase, and other aqueous solutions were obtained using a Milli-Q water purification system (Millipore, Molshein, France). DisQUE™ tubes (2 ml; 150 mg MgSO4/25 mg PSA/25 mg C18) were purchased from WATERS (Irlanda). C18 (200/6 mg ml−1) cartridges (55 μm, 70 Å) for SPE were purchased from Phenomenex (Steinheim, Germany). Nylon membrane filters (0.22 μm) were obtained from Millex, Millipore (Tullagreen, Carrigtwohill, Ireland).
Preparation of Extracts
Citrus peel extracts were prepared employing an ultrasonic processor QSONICA Q500 (Newtown, CT, USA). The equipment was used at a frequency of 20 kHz. Ultrasound-assisted extraction was performed following conditions: 400 W, an amplitude of 80%, a temperature of 35–40 ◦C, and a 30 min extraction period, according to the method previously described by Montero-Calderon et al 22 A 6 g of fresh peels were placed in a glass beaker using ethanol and water (50:50, v/v) as a solvent in a solid-liquid ratio of 1:10 (w/v). The extracts were centrifuged (4000 rpm, 4 °C, 5 min) and the supernatants were filtered using a Whatman no. 1 membrane filter (Whatman International Ltd, UK). Samples were brought to 45 ml and collected in a volumetric flask to subsequently be stored in dark conditions at −20 ◦C until use.
SPE Clean-up Method
The SFE clean-up procedure used to separate the polyphenols from the citrus peel extracts was based on the procedure validated by Anticona et al 23 A 5 ml of extract was diluted to 20 ml with 0.1% formic acid in Milli-Q water. C18 (200 mg) cartridges were preconditioned by loading 3 ml of MeOH twice and 3 ml of Milli-Q water twice, after which the samples were loaded. Solvents and samples were left in the cartridges for 2 min before loading. After, the cartridges were washed with 5 ml of 0.1% formic acid in Milli-Q water. Polyphenol fractions were eluted with 3 ml of 0.1% formic acid in MeOH. All methanolic fractions were dried under a light nitrogen stream and redissolved in 1 ml of HCl 0.6 mol/L, in 75% (v/v) aqueous MeOH. Samples were filtered through a 0.22 μm nylon membrane for subsequent analysis on the HPLC-UV system. Results were expressed as µg/g of fresh weight (FW). The entire process is described in Figure 1.

Extraction and clean-up methods (SPE and QuEChERS) process for the separation of phenolic compounds from citrus peel.
QuEChERS Clean-up Method
This cleaning procedure was performed using the method described by Mnyandu and Mahlambi 24 with some modifications. A 1.5 ml of the ultrasonic citrus peel extract was introduced into the DisQUE™ tubes (2 ml) containing the homogenized mixture of 150 mg MgSO4/25 mg PSA/25 mg C18. It was dissolved in vortex for 1 min and then, centrifuged at 4000 rpm for 5 min at 5 °C. The solid material should be at the bottom of the tube and a layer of supernatant (liquid) should form on top of the solid material. The extracts (supernatant) were dried under a light stream of nitrogen and re-dissolved with 0.3 ml of HCl 0.6 M (in 75% aqueous MeOH) (v/v). Samples were filtered through a 0.22 μm nylon membrane for subsequent analysis on the HPLC-UV system. Results were expressed as µg/g of FW. The entire process is described in Figure 1.
HPLC-UV Analysis
Chromatographic analysis was performed according to the validated method described by Anticona et al 23 using an HPLC system (Agilent Technologies 1120 Compact LC), equipped with a UV detector, and a C18 column (250 × 4.6 mm, particle size 5 μm; Luna PFP2, Phenomenex). The mobile phase was a linear gradient with a combination of solvent A: water/formic acid (95/5, v/v) and solvent B: acetonitrile/solvent A (60/40, v/v). The gradient program used was as follows: 0–10 min, linear 100%–85% A; 10–20 min, linear from 85% to 82% A; 20–50 min, linear from 82% to 0% A, 50–70 min, isocratic 0% A; 70–75 min, linear from 0% to 100% A; 75–80 min, isocratic 100% A. The analyses were conducted at a flow rate of 0.8 ml/min, at 280 nm, and an injection volume of 20 μl. The quantification of the compounds was calculated from the peak area according to the calibration curves in different ranges (n = 8 calibration points) of each external standard as follows: 4-hydroxybenzoic acid, caffeic acid, vanillic acid, quercetin dihydrate, chlorogenic acid, p-coumaric acid, ferulic acid, trans-cinnamic acid, naringenin, apigenin (0 to 50 μg/ml); rutin trihydrate, naringin, hesperetin (0-100 μg/ml); narirutin (0-400 μg/ml); hesperidin (50-400 μg/ml), as was previously used. 23 Finally, the results were expressed as μg/g of FW.
Statistical Analysis
All measurements were carried out in triplicate (n = 3) with three independent samples according to each variety and clean-up method. The results were presented as mean values ± standard deviation (SD). Statistical analyses were performed using a T-test to verify whether there were significant differences (P < 0.05) in polyphenol values in relation to the clean-up method used for each sample. The analysis was performed using the R Software (version 4.3.2). 25
Results
The Phenolic composition with differences between both species was obtained in both clean-up methods. In Figure 2, the representative chromatograms of the orange peel extracts are shown, and it can be observed that the hesperidin peak is higher, especially in the extract purified by SPE-C18. Figure 3 shows the phenolic profile of the mandarin peel extracts. It can be observed that the narirutin peak is higher compared to the orange peel extracts. Also, most of the unknown peaks are different depending on the samples analyzed.

Chromatograms of orange peel extracts at 280 nm by HPLC-UV: a. Orange-C18, b. Orange-QuEChERS, Peaks: 1– 4-hydroxybenzoic acid; 2– caffeic acid; 3– vanillic acid; 6– Quercetin dihydrate; 7– p-coumaric acid; 8– chlorogenic acid; 9– rutin trihydrate; 10– ferulic acid; 11– narirutin; 12– naringin; 13– hesperidin; 17– trans-cinnamic acid; 19– naringenin; 20– apigenin; 21– hesperitin; 4–5, 14–16, 18 unknown compounds.

Chromatograms of Mandarin peel extracts at 280 nm by HPLC-UV: a. Mandarin-C18, b. Mandarin-QuEChERS, Peaks: 1– 4-hydroxybenzoic acid; 2– caffeic acid; 3– vanillic acid; 6– Quercetin dihydrate; 7– p-coumaric acid; 8– chlorogenic acid; 9– rutin trihydrate; 10– ferulic acid; 11– narirutin; 12– naringin; 13– hesperidin; 17– trans-cinnamic acid; 19– naringenin; 20– apigenin; 21– hesperitin; 4–5, 14–16, 18 unknown compounds.
The concentration of phenolic compounds in orange and mandarin samples by C18 and QuEChERS is shown in Table 1. Regarding the quantities per variety, differences can be observed depending on the purification method used.
Phenolic Compounds Content in Orange and Mandarin Peel Extracts Purified with SFE (C18) and QuEChERS.
SFE (C18): solid fase extraction; Q: QuEChERS. Values are expressed in µg/g of fresh weight (FW). Each value is the mean ± the standard deviation (SD) of triplicate determinations. Different letters within a row indicate statistically significant differences (p < 0.05), among clean-up methods in orange peel (a-b) and mandarin peel (A-B), respectively.
Discussion
The analysis of the phenolic compound profile of orange and mandarin peel extracts by SPE-C18 and QuEChERS was performed HPLC-UV. 9 The recognition of each compound was carried out by determination of retention times, as well as spectral matching, and subsequent comparison with the data of the phenolic compound standards. The major compounds in the samples from both methods were narirutin and hesperidin. These results agree with what was observed in previous research in both orange and tangerine peel treated by ultrasound and analyzed by HPLC-UV a 280 nm.26–28
Solid phase extraction is one of the main clean-up methods included in the analytical process of phenolic compounds, however in the present study, the use of QuEChERS also provided interesting results. The composition observed during the first 32 min of the chromatographic method, mostly phenolic acid and quercetin, indicates that the QuEChERS method shows a higher recovery. However, the peaks of narirutin (peak 11) and hesperidin (peak 13), as well as the following compounds, are most of the profiles extracted with the SPE-C18 method. This indicates the affinity of the phenolic compounds (more or less polar) to the solid phase of the methods. The composition of MgSO4, from the QuEChERS method, increases the attraction of polar compounds, such as phenolic acids and simple phenols. 29 In contrast, the nonpolar nature of C18 allows for a higher affinity of more complex phenolic compounds such as flavonoids. 30 Given that the QuEChERS process is not yet validated, which is a limitation of the study, it is important to note that these results could benefit from an in-depth study of the recovery of phenolic compounds using this process, which will allow robust differences to be established between the two methods used.
As can be observed in Table 1, of the phenolic acids, caffeic acid (5-61.6 µg/g FW), p-coumaric acid (3.5-11.8 µg/g FW) and ferulic acid (5.9-9.5 µg/g FW) content in orange peel, are in the range of concentrations obtained in a previous study (0.2-21.1; 0.6-20.7; 3.8-74.2 µg/g FW, respectively). 22 He et al 31 obtained similar concentrations of caffeic acid (10.9 µg/g FW) and chlorogenic acid (13.9 µg/g FW); however, they obtained a greater amount of ferulic acid (27 µg/g FW) in hybrid orange peel. For its part, Karoui and Marzouk 32 obtained similar concentrations of 4-hydroxybenzoic acid (20 µg/g FW), vanillic acid (20 µg/g FW) and trans-cinnamic acid (20 µg/g FW).
The concentration of hesperidin ranges between 1837 and 2229 µg/g FW, being the C18 orange samples, the ones with the highest concentration. Also, hesperidin was the most abundant phenolic compound in orange peel in a previous study. 33 Our results were superior to those obtained by Montero-Calderon et al 22 in orange peel (C. sinensis) treated by ultrasounds (30 min) (1130 µg/g FW). He et al 31 determine a concentration of 2658 µg/g FW in orange hybrid peel (C. sinensis L. x C. unshiu Marc). Celano et al 26 and Gayed et al 34 obtained higher values in C. sinensis (9543 µg/g FW) and C. aurantium (4691 µg/g FW) respectively. On the other hand, the rutin content in orange peel samples (12.6-43.5 µg/g FW) was lower than that observed by Karoui and Marzouk 32 C. aurantium L peel (140 µg/g FW).
In mandarin peel samples, the content of phenolic acids is higher than in orange peel samples, except for chlorogenic acid. The caffeic acid (7.4-95.9 µg/g FW), chlorogenic acid (4.4-8.6 µg/g FW), p-coumaric acid (6.3-16.8 µg/g FW) and ferulic acid content (19.8-33.6 µg/g FW) was similar to the concentrations obtained by He et al 31 (8.5-12.9; 8,8-18.7; 12.2; 14.4-80.4 µg/g FW, respectively) in hybrid mandarins peel. About the flavonoid content in mandarin peel extracts purified by C18 and QuEChERS, the naringin and hesperidin concentrations were similar to that was obtained by He et al 25 in hybrid mandarin peels (20.2; 837 - 7995 µg/g FW, respectively).
Conclusions
This study allows us to estimate the chromatographic profile of the phenolic composition of orange and mandarin peel extracts purified by SPE and QuEChERS. The results indicate that the clean-up methods are selective according to the polar or nonpolar affinity of the compounds. In the case of phenolic acids, the concentration was higher in samples purified with QuEChERS, except for trans-cinnamic acid. Hesperidin and narirutin were the major compounds in the samples analyzed, their content being higher in extracts purified with C18. Likewise, the concentration of the polyphenol compounds varies depending on the species analyzed. These results are of interest for studies on the isolation of phenolic compounds from plant matrices and represent an alternative to obtaining clean extracts. A validation of the QuEChERS method could be useful to optimize the process conditions and this would help to establish robust differences between the cleaning methods under evaluation. In this sense, future research with different samples of plant extracts would favor the knowledge of analytical methods, including a clean-up step, for the recovery of phenolic compounds.
Footnotes
Acknowledgments
Mayra Anticona thanks to President of the Republic Scholarship from the Ministry of Education of the Republic of Peru and the national program of scholarships and PRONABEC for the support.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical Statement
This article does not contain any studies with human or animal participants.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
