The extraction of bioactive compounds from agri-food wastes using green extraction procedures (e.g., supercritical fluid extraction, pulsed electric fields, ultrasound-assisted extraction, microwave-assisted extraction, enzyme-assisted extraction, pressurized liquid extraction) (Figure 2) has gained special attention due to their exceptional practices focused on economic, environmental, and safety concerns [9]. Moreover, green extraction procedures comprise six principles of green chemistry, namely: (i) the use of renewable and sustainable bio-resources, (ii) use of water or green solvents, (iii) lower energy input, (iv) co-products production from waste, (v) a minimal number of unit operations, and (iv) resulting non-denatured and biodegradable extract [9]. The following sub-sections present the most common green extraction procedures used for the recovery of bioactive compounds from agri-food wastes.

Supercritical fluid extraction (SFE) using carbon dioxide (CO₂) has been proposed as a green extraction procedure, since it requires low volumes of organic solvent to recover the value-added bioactive compounds (e.g., carotenoids, phenolic compounds) from agri-food wastes [10,11,12,13]. CO₂ is the most used supercritical fluid due to its mild critical temperature (31.2 °C) and pressure (73.8 bar), which allows for operation at moderate conditions, generally ranging from 40 to 60 °C and 200–400 bar pressure [9]. Additionally, CO₂ is non-carcinogenic, non-toxic, non-mutagenic, non-flammable, thermodynamically stable, and generally identified as safe [14]. The main benefit of this green extraction technique is that the solvent physicochemical properties can be changed by adjusting the pressure and temperature conditions within the system, consequently improving the extraction selectivity and extraction yields due to the fast diffusion of fluid through the solids [12,14]. However, the low polarity of supercritical CO₂ represents the major drawback of this procedure. This problem can be minimized by adding small percentages of co-solvents (e.g., ethanol, methanol, water) or modifiers that change the polarity of the solvent. Consequently, this results in an enhancement of the extraction yield by improving the solubility of the solute or the swelling of the solid matrix that facilitates the solute–solvent contact [15]. This versatility makes SFEs very appealing for several applications in different fields (e.g., industry, pharmaceutical).

Table 1 shows the potential of SFE in the extraction of important value-added compounds from agri-food wastes [10,11,12,13]. The bioactive compounds extracted by SFE include a wide diversity, namely, phenolic compounds from onion peels [16], antioxidants and saponins from Agave salmiana bagasse [12], and carotenoids from carrot peels [10], among others. The effect of pressure, temperature, and the addition of a co-solvent in the extraction of bioactive compounds by SFE are evaluated in some studies in Table 1. Generally, the extraction was performed by applying pressures, temperatures and co-solvents ranging from 30–400 bar, 33–230 °C and 5–15 % v/v, respectively, while the extraction time and flux ranged from 30–180 min and 1.7–133 g/min, respectively. Some studies compared the extraction efficiency of SFE with other conventional extraction procedures (e.g., Soxhlet extraction). Soldan et al. [13] compared the extraction efficiency of SFE and Soxhlet on the recovery of the bioactive compounds phenolics, flavonoids, fatty acids, and carotenoids, from Capsicum annuum waste. The results showed that the total mass yields obtained by SFE ranged from 9.38 to 10.08%, while for Soxhlet the yields ranged from 8.45 to 15.5% (w/w). Despite revealing bioactive compounds, the extracts did not show significant antioxidant activity. Natolino and collaborators [17] also performed a comparison between SFE and Soxhlet on the recovery efficiency of seed oils from pomegranate, which showed no significant difference between these two extraction procedures, as the extraction yield from SFE (0.18 ± 0.01 g/g) was similar to Soxhlet (0.19 ± 0.01 g/g). Nevertheless, SFE was faster than Soxhlet (8 h vs. 2 h of SFE) to achieve the asymptotic extraction yield and presented more oxidation stability than Soxhlet. Santos–Zea et al. [12] evaluated the effect of ultrasound on SFE for the recovery of antioxidants and saponins from agave bagasse. The data obtained showed that the use of ultrasound-assisted SFE improved the extraction yield of antioxidants and saponins from agave bagasse when a low mass load (0.043 g/cm³) was applied. Since the CO₂-SFE demonstrated low extraction efficiency of more polar compounds in some studies (e.g., phenolic compounds), a few authors proposed the use of co-solvents to enhance the extraction yields of polar and medium polar bioactive compounds [13]. Soldan et al. [13] verified that the temperature variation and the addition of a co-solvent (ethanol) were significant in increasing the total extracted mass of oleoresin, although the pressure did not have a significant effect. In sum, this green extraction technique can be easily transferred to an industrial scale to extract large quantities of matrix and obtain a great amount of extract in a single step. However, despite the exceptional extraction properties and outstanding versatility, the high processing costs and the complex industrial equipment are restricting factors [14].

Subcritical water extraction (SCWE) is an environmentally friendly and techno-economically feasible alternative to conventional extraction procedures, such as solvent extraction. This green extraction technique uses water as an extractant, which is economic, non-flammable, and renewable. Compared to conventional extraction procedures, such as solid–liquid extraction (Soxhlet) using organic solvents, maceration, and hydrodistillation, SCWE shows higher yield and purity while applying lower extraction time [16,24]. This green extraction procedure is also distinguished by the demand for a downstream solid–liquid separation step which increases the energetic requirement of the process [70]. However, and despite large research efforts, corrosion problems have not been completely solved for the application of SCWE at an industrial scale [14].

During SCWE, the feedstock is heated in the aqueous phase at a sub-critical temperature (∼150–320 °C) and pressure (∼20–150 bar). Under these conditions, the dielectric constant (ε), surface tension, and viscosity, among other properties of the water change, enhance mass transfer and the extractability of barely water-soluble bioactive compounds, since subcritical water promotes the hydrolysis of the bonds between phenolic compounds and agri-food waste matrix [25]. Moreover, the mass transfer ratio also rises due to the reduced viscosity and surface tension of the water as well as its diffusivity [24]. A diversity of SCWE applications on the extraction of bioactive compounds from agri-food wastes has been performed, namely, stilbenes from vine co-products (e.g., cane, wood, and root) [29], phenolic compounds from onion peel [16], and protein hydrolysates from shellfish waste [27], among others. Nevertheless, the main drawback of SCWE is the risk of hydrolysis and other degradation reactions during extraction [14]. In this sense, Rodrigues et al. [27] used SCWE to recover antioxidant protein hydrolysates from shellfish waste streams. In this study, the authors assessed the impact of operating temperature (150, 200, and 250 °C), solid/liquid ratio (1:5, 1:10, and 1:15 g/mL), and heating rate (3 and 6 °C/min) on SCWE performance. It was verified that higher temperatures enabled the production of extracts with a higher antioxidant potential, possibly due to an increase of smaller peptides/free amino acids and Maillard reaction products. On the other hand, Hwang and collaborators [22] recovered hesperidin and narirutin from Citrus unshiu peel waste using SCWE combined with pulsed electric field (PEF) treatment. The data obtained demonstrated that the concentrations of hesperidin and narirutin increased with PEF treatment time, with increased yields of hesperidin and narirutin by 22.1% and 33.6%, respectively, in PEF pretreatment combined with SCWE.

Pulsed electric field (PEF) is a nonthermal agri-food processing method that applies high-intensity electric field pulses to agri-food passing through electrodes. This extraction process causes the electroporation of membranes (permeabilization) that enables the release of intracellular bioactive compounds from the matrix investigated [22,71]. The extraction efficiency of PEF treatment depends on numerous factors, involving electric field strength, total specific energy input, treatment time, and temperature. Previous studies have demonstrated that the PEF pre-treatment of moderate electric field intensity (0.5–10 kV/cm) and relatively low energy input (1–10 kJ/kg) has advantageous effects on the permeabilization of membranes of plant cells, enabling high recovery yields of intracellular compounds of interest from a wide range of food processing wastes and by-products [38]. Furthermore, PFE treatments have shown several advantages, including low solvent consumption, shorter treatment time, energy efficiency, continuous operability, ease of scale-up, non-destructive nature, and high selectivity. However, its dependence on medium composition (conductivity) and the high cost of the equipment represents the main disadvantages of PFE treatments [35,72].

Table 1 presents a diversity of PFE applications for the extraction of bioactive compounds from agri-food wastes, such as lycopene from tomato peels [38], ellagic acid from pomegranate peels [35], and phenolic compounds from lemon peels [33], among others. Pollini et al. [30] compared different extraction techniques, such as ultrasound-assisted extraction (UAE), ultraturrax extraction (UTE), accelerated solvent extraction (ASE), and PEF extraction pre-treatment to identify the most efficient method to recover phenolic compounds from apple pomace. The extraction efficiency of phloridzin, the main phenolic compound in apples, increased by applying PFE at a low intensity and for a long duration (2 kV/cm and 100 kJ/kg), using EtOH:H₂O (70:30, v/v). In another study, Lal et al. [32] combined PFE with microwave-assisted extraction to recover pectin polysaccharide from jackfruit waste, but the pectin yield obtained was not significant when compared to conventional processes. Radjha et al. [35] compared the aqueous extraction efficiency and biological activities of phenolic compounds from pomegranate peels assisted by infrared (IR), ultrasound (US), PFE, and high-voltage electrical discharges (HVED). The data showed that the PFE selectively extracted and enhanced the recovery of ellagic acid (≈740 μg/g dm), whereas HVED (≈345 μg/g dm) intensified the gallic acid extraction compared to US, IR, PFE and WB. Peiró and collaborators [33] evaluated the influence of PEF of different intensities (3–9 kV/cm and 0–300 μs) on the extraction of phenolic compounds from lemon peel residues, which increased by around 300%, giving maximum values of 84 mg of hesperidin in 100 g FW and 176 mg of eriocitrin in 100 g FW.

Ultrasound-assisted extraction (UAE) is a green extraction procedure and a techno-economically feasible alternative to conventional extraction procedures. This technique has gained attention in recent years, due to its excellent advantages compared to traditional extraction procedures, such as reduced solvent volumes, shorter extraction time, and use of common laboratory equipment (e.g., ultrasonic bath), making it an environmentally sustainable and economical extraction procedure [73,74]. Yet, the solid–liquid separation and drying are certainly the main disadvantages of the UAE process. This extraction procedure is based on the cavitation process induced by compression and expansion cycles associated with the passage of ultrasounds (20 kHz–100 MHz frequency) through the sample. The acoustic waves promote the distance between molecules and consequently generate spaces among them, forming bubbles. The implosion of the cavitation bubbles causes inter-particle collisions resulting in particle disruption and enhanced diffusion of extractable bioactive compounds into the solvent [70,75]. A large amount of energy is released by bubble implosions, causing significant changes in the local temperature and pressure, liquid circulation, and turbulence, consequently increasing the mass transfer rate [47]. Moreover, the extraction efficiency of UAE can be significantly influenced by the sample properties (e.g., consistency, rheology, particle mobility) which affect ultrasound energy dispersion [75].

The UAE has been extensively applied at the lab scale in diverse food fields [70]. Ben-Othman and collaborators [39] used the response surface method (RSM) with a Box–Behnken design to select the best extraction efficiency of UAE for the recovery of phloretin and other phenolic compounds from apple tree leaves (Malus domestica Borkh.) from different cultivars from Estonia. The optimal extraction conditions were 14.4 min of extraction time, 10% sonication amplitude, and 10 g of sample per 100 mL solvent (70% ethanol, w/w). By applying the ideal conditions, the phloretin concentration ranged from 292 to 726 µg/g and the antioxidant activity from 6.06 to 11.42 mg GAE/g in the local winter cultivars “Paide taliõun” and “Tellissaare”, respectively. Martín-García et al. [42] used RSM to evaluate the effect of solvent composition, extraction time, and ultrasound power on the recovery of proanthocyanidins from brewers’ spent grains. The highest content of proanthocyanidins was obtained using 80/20 acetone/water (v/v), 55 min, and 400 W, which resulted in 1.01 mg/g dw of proanthocyanidins from brewers’ spent grains. In another study, da Rocha et al. [44] compared the extraction efficiency of microwave-assisted extraction and UAE of bioactive compounds from grape pomace. The results showed that both extraction procedures allowed the recovery of 45% of the anthocyanins when compared to the exhaustive extraction with methanol acidified solution.

Microwave-assisted extraction (MAE) is a green and cost-effective extraction technique that has gained a lot of attention recently, due to its enhanced productivity, reduced extraction time, less solvent requirement, simplicity, and low set-up costs [4]. MAE involves electromagnetic radiations, transmitted as waves in the frequency range from 300 MHz to 300 GHz [4]. This technique is based on the principle that the energy absorbed during the passage of microwaves through the medium is converted into thermal energy, which facilitates the processing, due to higher extraction temperature and resultant faster mass transfer rate [4,76]. The heating effect of microwaves depends on the dielectric properties of the mixture of the solvent. When a solvent placed in contact with the sample is heated, MAE leads to the disruption of the hydrogen bonds, which results in the dipole rotation of the molecules and migration of the ions. Consequently, this process allows for the diffusion of the solvent, and thus the dissolution of the components [4]. MAE can be influenced by a wide range of parameters, namely microwave power, frequency, irradiation time, the particle size of the sample matrix, the composition of the solvent, extraction temperature, pressure, and the number of cycles. The choice of a suitable solvent for extraction is important and depends on the solubility, dielectric constant, and dissipation factors. Solvents with both high dielectric constant and dissipation factor can lead to a better extraction, which can be accomplished by mixtures of water with other solvents, such as ethanol or methanol [76].

MAE has been frequently used in the extraction of bioactive compounds, especially for plant materials [76]. Tran, Akanbi, Kirkman, Nguyen and Vuong [49] provided a method for the recovery of total phenolics, flavonoids, chlorogenic acid, and caffeine from coffee pulp using an MAE system. The results of this study showed that the sample-to-solvent ratio and ethanol concentration significantly affected the recovery yields of the bioactive compounds and the antioxidant capacity. Under the optimal conditions (Table 1), the extraction yields of total phenolic compounds, flavonoids, chlorogenic acid, and caffeine were 38.68, 27.00, 6.95, and 5.47 (mg/g dw), respectively. The extracts showed high antioxidant capacities, with values measured by ABTS, DPPH, and FRAP assays as 87.95, 9.3, and 65.31 (mg Trolox equivalents/g dw), respectively. In another study, Kurtulbaş, Sevgen, Samli and Şahin [50] extracted phenolic compounds and anthocyanins from peach peels, with the highest total phenolic content (TPC) being 19.35 mg of gallic acid equivalents/g of fresh plant matter and a total anthocyanin of 1.12 mg of cyn-3-glu/g of fresh plant matter, under the optimal MAE conditions (Table 1). After the extract was obtained, the samples were exposed to several storage media, such as −20 °C, 4 °C, and 25 °C in dark and 25 °C in light and the storage stability was monitored in terms of 4 bioactive properties (TPC and total anthocyanin contents, p-hydroxybenzoic acid and p-coumaric acid). In a general way, the degradation rate rose with storage temperature. The longest shelf life in terms of total phenols, anthocyanins, and major phenolic compounds (p-hydroxybenzoic acid and p-coumaric acid) was calculated as 111, 107, 88, and 83 days under deep freezer conditions at −20 °C. Zhang and collaborators [54] extracted flavonoid compounds from spent sweet potato leaves with natural deep eutectic solvents (NADESs) coupled with MAE. The highest extraction yield (40.21 ± 0.23 mg of rutin equivalents/g of sweet potato leaves) was obtained with NADES-2 synthesized by choline chloride and malic acid (molar ratio 1:2). The extracts were recovered by macroporous resin for the biological activity detection of flavonoid compounds, in which the AB-8 macroporous resin provided a recovery yield of 85.46% ± 2.33%. Additionally, the in vitro bioactivity experiments confirmed that the flavonoid compounds had good DPPH and O₂⁻ radical-scavenging activity, as well as inhibitory effects on E. coli, S. aureus, E. carotovora, and B. subtilis. Rodríguez García and Raghavan [55] evaluated the potential of MAE as a green technique to obtain phenolics. The authors extracted phenolic compounds (vanillic, sinapic, caffeic, chlorogenic, ferulic, gallic, neochlorogenic, and p-coumaric acids, identified by HPLC) from broccoli by-products (stems, leaves, and florets). MAE was found to increase the phenolic yield up to 45.70% for broccoli leaves, 133.57% for broccoli florets, and 65.30% for broccoli stems, in less time compared with maceration extraction. Despite the advantages of MAE over conventional extraction methods, the high dependency on the solvent nature and the extraction temperature limits the application of MAE [4,76].