Recovery of Carotenoids via Novel Extraction Technologies for the Valorization of Tomato By-Products

Abstract

Tomato processing residues—including peels, seeds, and pomace—are rich in bioactive compounds, such as lycopene, β-carotene, cutin, pectin, and antioxidants, yet are often underutilized. This study evaluates microwave-assisted extraction (MAE) and high-pressure-assisted extraction (HPAE) for the recovery of carotenoids from TP, compared to conventional extraction (CE) using ethyl acetate. Optimal MAE conditions (150 W, 50 °C, 20 min, solid/liquid ratio of 1:10 g/mL) yielded 592.5 mg carotenoids/kg dry weight (dw), exceeding CE yields (505.3 mg/kg dw), while significantly reducing extraction time (20 min vs. 120 min). By contrast, direct HPAE (650 MPa, ambient temperature, solid/liquid ratio of 1:10 g/mL) resulted in lower carotenoid yields (ca. 84 mg/kg dw), but when used as a pre-treatment followed by stirring for 24 h, HPAE enhanced carotenoids extractability to 277.0 mg/kg dw, recovering 55% of carotenoids extracted by CE. Bioaccessibility studies showed low lycopene bioaccessibility across all methods (3.9% for HPAE, 3.4% for MAE, and 1.6% for CE). Incorporation into oils significantly improved bioaccessibility, with olive pomace oil (OPO) achieving 28.1%, compared to 8.1% in corn oil (CO). Overall, MAE and HPAE (as pre-treatment) present efficient strategies that reduce solvent usage and processing time, though they still rely on organic solvents, while strategies to enhance bioaccessibility should further be explored for effective functional ingredient development.

1. Introduction

The food industry generates substantial waste throughout its production chain, making its manipulation and valorization a key target for sustainable utilization efforts. The tomato processing industry, in particular, produces annually large quantities of waste. Globally, around 180 million tons of fresh tomatoes are produced each year, with Greece contributing approximately 400,000 tons [1]. Tomato processing by-products, including skins, seeds, pulp, spoiled tomatoes, stems, leaves, and wastewater, pose serious environmental challenges. However, they present valuable opportunities for nutrient recovery and the development of value-added products through innovative reuse strategies [2].

Tomato pomace (TP), mainly consisting of skins and seeds, constitutes about 5–10% of the tomato fruit. Dried TP comprises roughly 33% seed, 27% skin, and 40% pulp [3,4]. Producing 344 g of tomato sauce results in approximately 80.5 g of waste [5], translating to 20–40 million kg of by-product annually. TP is typically used as fertilizer or animal feed, but often lacks commercial value, remaining underutilized. Nonetheless, TP is rich in bioactive compounds (BACs), particularly carotenoids, such as lycopene and β-carotene [6]. Efficient extraction and use of lycopene from TP could add value to waste and offer a sustainable source of this beneficial compound.

Carotenoids, including lycopene, are synthesized by plants and microorganisms. constituting functional components of the human diet. These compounds enhance the immune function and reduce the risk of degenerative conditions. Lycopene (C₄₀H₅₆) is a non-cyclic, conjugated isoprenoid responsible for the red color in tomatoes, watermelon, and blood oranges [7]. It is the most effective singlet oxygen quencher among carotenoids and is known for its antioxidant properties. Fat-soluble and more bioavailable with dietary fats, lycopene consumption is linked to protection against oxidative DNA damage and cancer [8]. In food, it serves as a functional ingredient and natural colorant in compliance with EFSA regulations (Commission Regulation (EU) No 1129/2011). Lycopene is also used in cosmetics for its anti-aging benefits and in pharmaceuticals for oxidative stress-related diseases. Apart from the potential health benefits, lycopene offers an alternative solution to synthetic food colorants. From a processing perspective, extraction can be challenging due to the limited solvent choices allowed in food applications. However, isolating lycopene from tomato peels can reduce overall costs by adding value to an otherwise discarded by-product. Lycopene is insoluble in water and only sparingly soluble in organic solvents, which limits its removal from the initial plant material. Nevertheless, the efficiency of carotenoid extraction can be improved by using solvent combinations to facilitate solvent penetration and thus, bioactives diffusion. Previous research has shown that solvent systems containing hexane and ethyl acetate are most effective for extracting carotenoids from tomato seeds and peels [9]. Despite these improvements, the extraction process itself is time-consuming and carries the risk of degradation of bioactive components (carotenoids) as the samples are exposed to heat for extended periods [10].

In recent years, strategies for recovering high-value compounds from food waste have focused on technologies that preserve compound functionality and ensure economic feasibility [11]. Conventional solvent extraction, the most widely used method for lycopene recovery, often involves the use of a high amount of solvents, usually toxic ones posing health and environmental risks, as well as extended processing time, leading to insufficient yields [12,13]. While temperature-assisted extraction can be effective, thermosensitive compounds limit the applicability, leading to low yields and long processing times [14]. To overcome these issues, novel extraction technologies have emerged, offering shorter times, lower solvent use, and better selectivity [15]. Promising techniques include microwave-assisted extraction (MAE), high-pressure processing-assisted extraction (HPAE), pulsed electric fields (PEF) extraction, ultrasound-assisted extraction (UAE), enzyme-assisted extraction (EAE), pressurized liquid extraction (PLE), supercritical fluid extraction (SFE), and molecular distillation.

MAE utilizes microwave energy to heat the solvents and plant materials, enhancing the extraction of bioactive compounds. Microwaves, with frequencies between 300 MHz and 300 GHz, interact with polar molecules, producing dielectric heating. This mechanism facilitates the penetration of solvents into the plant matrix, dissolves target compounds, and enables their migration and separation [16,17,18,19]. MAE offers advantages such as reduced extraction time, lower solvent use, improved selectivity, and controlled heating [20]. Additionally, it consumes less energy and provides precise temperature control, critical for extracting thermosensitive compounds. However, it requires higher initial equipment costs and proper safety measures to manage high-power microwaves. Previous studies have applied MAE for the recovery of carotenoids from tomato by-products, mainly addressing yield optimization and compound stability [9,21,22].

High-pressure-assisted extraction (HPAE) is another efficient method, operating under high hydrostatic pressure to disrupt cell walls and to release intracellular compounds. It is particularly useful for heat-sensitive substances, as it can be conducted at room temperature. HPAE offers rapid processing times, reduced energy consumption, and environmental sustainability [23]. Initially introduced by Shouqin et al. [24], HPAE has since been widely studied for extracting proteins, polysaccharides, and BACs from various sources, including tomato by-products. The process involves mixing raw material with a solvent, applying high pressure, and recovering the target compounds. This increases cell permeability, enhancing metabolite diffusion based on phase theory, which states that solubility increases with pressure. The large pressure differential between cell membranes facilitates immediate compound release. HPAE typically requires only a few minutes and low temperatures, making it suitable for thermosensitive extractions. Applications include the extraction of polyphenols, caffeine, carotenoids, pectins, flavonoids, and lycopene from various plant matrices [24,25,26,27,28]. HPAE techniques have shown strong potential for enhancing carotenoid and lycopene recovery from tomatoes and tomato by-products. Briones-Labarca et al. demonstrated that high pressure extraction (up to 450 MPa) in combination with optimized solvent mixtures significantly increased extraction yield and lycopene content from tomato pulp, improving bioactive compound release and bioaccessibility compared to conventional methods [28]. Similarly, Jurić et al. reported that the high-pressure homogenization of tomato peels (100 MPa, up to 10 passes) effectively disrupted plant cell walls, releasing up to 56% of initial lycopene content into the aqueous phase without organic solvents, along with enhanced recovery of polyphenols and proteins [29]. Complementarily, Strati et al. showed that the high-pressure-assisted extraction (100–800 MPa) of tomato waste improved carotenoid and lycopene yields by 2–64% compared to ambient pressure solvent extraction, even at lower solvent-to-solid ratios (6:1 and 4:1 mL/g) and shorter extraction times (up to 10 min) [25]. Together, these studies have confirmed that HPAE not only improves recovery efficiency and reduces reliance on harsh solvents, but supports environmentally sustainable valorization of tomato pulp and pomace.

Bioactive compounds, like carotenoids, may degrade in the gastrointestinal tract, reducing their potential health benefits. Thus, evaluating the extraction methods must include assessing compound bioaccessibility—the fraction released from the food matrix and available for intestinal absorption [30]. Since in vivo studies are costly and ethically complex, in vitro models, like the standardized INFOGEST protocol, offer a reliable alternative to simulate digestion and to assess the release and potential absorption of these compounds [31,32]. Carotenoids’ hydrophobicity limits their bioavailability and makes them vulnerable to degradation during digestion. Nonpolar carotenoids like lycopene, located within lipid droplet cores, exhibit lower bioaccessibility than polar counterparts like lutein, which are positioned at the surface. Their absorption depends on mixed micelle formation in the small intestine, involving bile salts, phospholipids, and lipid digestion products. Additionally, carotenoid type, physicochemical properties, concentration, interactions within the food matrix, and processing methods significantly influence their absorption [33,34]. Most studies in the literature concerning novel technologies focus on their extraction yield efficiency, compared to conventional extraction methods [11,25,28], but there is limited research on how these methods affect the bioaccessibility of the extracted BACs, particularly of tomato pomace. For example, Lara-Abia et al. examined the effect of HPAE extraction in the bioaccessibility of carotenoids from papaya [35].

Beyond single-compound recovery, tomato pomace is increasingly considered within a cascade biorefinery framework, where multiple products, such as pectin, polyphenols, and carotenoids, can be sequentially recovered to maximize resource efficiency. For example, Ninčević Grassino et al. demonstrated the feasibility of such an approach by extracting pectin and polyphenols from TP [27]. Almeida et al. emphasized that a biorefinery cascade approach enables the sequential recovery of these components while closing carbon and nutrient loops; examples include the extraction of bioactive compounds followed by anaerobic digestion or composting, coupled with economic and environmental assessments to guide management route selection [36]. Similarly, Casa et al. illustrated how tomato by-products—constituting up to ~10% of fresh tomato weight—can undergo fractionated valorization: extracting lycopene, cutin (useful for biopolymer synthesis), and pectin, and subsequently converting residual lignocellulosic biomass into energy via fuels, all within a cascade biorefinery model complete with mass-flow mapping [37]. Positioning carotenoid extraction within this broader context supports the circular economy objectives and highlights the potential of MAE and HPAE as complementary steps in multi-product valorization.

The present study aims to evaluate the carotenoid content of the samples derived from TP and alternatively extracted, using MAE and HPAE, compared to conventional extraction methodology that was used as the bench-mark technology. In contrast to previous studies, the present study provides a direct comparison of MAE and HPAE with conventional extraction under the same food-grade solvent system, while also assessing bioaccessibility through the standardized INFOGEST protocol. Furthermore, the combined use of spectrophotometric and HPLC-DAD analyses allowed for the demonstration of differential stability of lycopene and β-carotene under microwave and high pressure conditions. These findings highlight not only improved recovery efficiency but the nutritional functionality of carotenoids, advancing the valorization of tomato pomace within a circular economy framework.

2. Materials and Methods

2.1. Materials and Processing Procedures

The fresh tomato pomace used in this study is a solid residue from the juicing of the industrial tomato hybrid Heinz 3402, supplied from a tomato manufacturing industry in Central Greece. The material consists of approximately 89% w/w peels and flesh and 11% w/w seeds, on a dry basis, with an initial moisture of 88% w/w (on wet basis). After receiving it, the pomace was air dried for 24 h at 40 °C (HotmixPro air dryer, Modena, Italy), vacuum packaged (BOSS vacuum sealer, Hamburg, Germany) in PET12/Alu8/PE80 films and stored at −18 °C until use, to prevent degradation of the material and its contained ingredients.

2.2. Chemicals and Reagents

The solvent used for the extractions was ethyl acetate (EtOAc), which was sourced from the commercial supplier Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). For the high-performance liquid chromatography (HPLC) analysis, the solvents used were water, acetonitrile, and methanol, HPLC-grade, from the commercial supplier Merck (Darmstadt, Germany). The pure solvents were stored under ambient conditions until their use. Enzymes and reagents required for the INFOGEST method were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). Standard lycopene was a product of Extrasynthese (Genay, France) with purity >96%. The compound was stored at −16 °C. In order to determine the calibration curves in both spectrophotometer and HPLC-DAD, an initial concentration of 100 mg/L was prepared followed by a series of dilutions (75, 50, 25, 10, 1 mg/L). β-Carotene (purity > 97%) was a product of Fluka (Buchs, Switzerland), also stored at −16 °C. The standard was used for the calibration curve of β-carotene in HPLC-DAD. In the same manner, a stock solution of 90 mg/L was prepared followed by a series of dilutions (67.5, 45, 22.5, 9 mg/L).

2.3. Semi-Batch Extraction of BACs

The semi-batch extraction has been described thoroughly in previous publications [20,38]. In this type of extraction, the material is packed in a column without any shaking or stirring. The system is fed with pure solvent, through the inlet of the extractor, which passes through the pores of the material, extracts the components, and exits the system from the outlet, as an extract. For the needs of the current experiments, the extractor was filled each time with 20 g of dried tomato pomace powder, and the solvent passed through the material at a flow of 3 mL/min using a peristaltic pump (Millipore, MA, USA), at room temperature. The duration of extractions ranged between 90 and 120 min until obtaining a practically discolorized extract from the outlet of the extractor. The solvent used for the extractions was ethyl acetate (EtOAc) in a solid-to-liquid ratio of 1:46 g/mL. The extractions were performed in duplicate experiments.

2.4. Microwave-Assisted Extraction (MAE) of Carotenoids

MAE was carried out using a laboratory scale batch apparatus (Nanjing Xianou Instruments Manufacture Co., Ltd., Nanjing, China), and EtOAc as solvent, at 150–300–500 W for 5, 10, 20, and 30 min at different solid/liquid ratios (1:10 and 1:30 g/mL). Tomato pomace samples used for MAE were air dried at 40 °C for 24 h and then ground. Supernatants were analyzed for total carotenoids, antioxidant activity, and extract dry weight, using the protocols described below.

Table 1. Experimental MAE conditions for the extraction of carotenoids from tomato pomace.

Microwave Power (Watt)	Temperature (°C)	Process Time (min)	Solid/Liquid Ratio (g/mL)
100 300 500	30 40 50	5 10 20 30	1:10 1:30

displays the experimental extraction conditions performed. Temperature and microwave power were continuously monitored throughout the procedure. The extractions were run in duplicate experiments.

2.5. High-Pressure-Assisted Extraction (HPAE) of Carotenoids

HPAE of carotenoids was applied in a pilot-scale high pressure (HP) unit (Food Pressure Unit FPU 1.01, Resato International BV, Assen, The Netherlands), with operation capability in the pressure range of 100–1000 MPa and a temperature range of 0–90 °C. The pressure transmitting fluid is deionized water. Using specified software, there is a continuous monitoring of both pressure (FPU 1.01 software, Resato International BV, Assen, The Netherlands), and temperature (PC408 software, Campbell Scientific, Inc., Logan, UT, USA). All experiments were performed in triplicate.

Ground, air-dried (40 °C for 24 h) TP was placed into multilayer pouches and mixed with ethyl acetate at different solid/liquid ratios (1:5, 1:10, and 1:30 g/mL), prior to HPAE. The efficacy of HP on the extraction of carotenoids from tomato pomace was studied in the pressure range of 250–650 MPa, at an ambient temperature, and for processing time ranging between the time required to achieve the desirable pressure and up to 20 min. Subsequent to HPAE, the extracts were separated from the pomace using vacuum filtration, and further microfiltration using 0.45 μm syringe filter, and were then collected. The total carotenoid yields of the HP-treated extracts were investigated using appropriate protocols, as described below. Furthermore, dried TP powder samples were mixed with ethyl acetate using the same procedure as above, remaining in calm for time equal to a full HP-cycle [20] to serve as control samples (HPAE-untreated samples). All experiments were performed in triplicate.

The effectiveness of HP on the extractability of carotenoids from tomato pomace was also investigated as a pre-treatment step. Experiments were conducted at 650 MPa and an ambient temperature for the time required to desirable pressure achievement, but in this case, the extracts were further extracted under stirring at 350 rpm for 24 h. These experiments were performed in duplicate.

2.6. Total Carotenoid and Lycopene Determination

The total carotenoid yields (CY) of the extracts were measured spectrophotometrically (Hitachi, Tokyo, Japan, U-2900 UV/Vis, 200 V), according to the methodology described in Strati and Oreopoulou [9], and the results were expressed as mg of lycopene equivalents per kg of dry material weight (mg/kg dw).

2.7. Determination of the Selectivity of Extraction in Tomato Pomace Extracts

In order to determine the selectivity of the extraction, the solid residue was calculated by removing the solvent from the extracts. Initially, 3 mL of tomato pomace extract were transferred to pre-weighed vials. The vials were placed in an oven (Vacutherm, Heraeus Electronics, Hanau, Germany) with a vacuum pump (V100, Büchi Labortechnik AG, Flawil, Switzerland) for solvent evaporation. The removal of the solvent was carried out at room temperature under a pressure of 100 mbar. After the solvent removal, the vials were re-weighed, and the difference in mass yielded the pure mass of the extract. The dry extract was then calculated as the ratio of the pure mass of the extract after the solvent removal to the initial volume of 3 mL placed in the oven (g/mL). Using the concentration of carotenoids in the extracts, the selectivity of extraction was calculated as the amount of mass (in mg) of total carotenoids expressed as lycopene contained in 1 g of dry extract and symbolized as “mg lycopene/g dry extract”.

2.8. HPLC-DAD Analyses of the Extracts

The determination of the carotenoids recovered from tomato pomace was carried out using the HPLC method of Szabo et al. [39]. The HPLC-DAD system was equipped with an autosampler (Agilent Infinity 1260, Agilent, Santa Clara, CA, USA), a gradient quaternary pump (HP 1100, Waldbronn, Germany), a diode array detector (Hewlett-Packard, Waldbronn, Germany), and a reversed-phase column Hypersil™ C18 (ODS 5 μm, 250 × 4.6 mm, MZ-Analysentechnik GmbH, Mainz, Germany). The mobile phase consisted of three solvents, each one containing 0.25% triethylamine, as follows: solvent A, water; solvent B, acetonitrile; solvent C, ethyl acetate. The flow rate of the mobile phase was 1 mL/min. The initial composition of the mobile phase was 9% A, 81% B, and 10% C, which changed with linear gradients to 5% A, 45% B, and 50% C, at 10 min, and finally to 1% A, 9% B, and 90% C, at 20 min. Injection volumes were 20 μL, while the compounds were detected at 474 nm. Data processing was performed using ChemStation for LC 3D software (Rev. B.04.03, Agilent Technologies, Waldbronn, Germany). Standard lycopene and β-carotene were used for the development of the respective calibration curves.

2.9. Bioaccessibility of Carotenoids Extracted from Tomato Pomace

Bioaccessibility determination was performed using the static INFOGEST 2.0 protocol [32]. The digestion process involved three phases: salivary, gastric, and intestinal. Upon completion of the digestion protocol, the samples were centrifuged at 12,000× g for 10 min at 4 °C (Heraeus Megafuge 16R, Thermo Fisher Scientific, Waltham, MA, USA) to separate the micellar phase, which contained the bioaccessible fraction of carotenoids. Carotenoids were then extracted from the micellar phase using a petroleum ether/ethanol solution in a 3:2 ratio, until the organic phase became colorless, indicating that all carotenoids were extracted from the digesta. The concentration of total carotenoids was determined spectrophotometrically and expressed as lycopene equivalents (as described in Section 2.6), and bioaccessibility (%) was then calculated using the following equation:

$$\mathrm{B}\mathrm{i}\mathrm{o}\mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y} \left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{C}\mathrm{a}\mathrm{r}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{o}\mathrm{i}\mathrm{d} \mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{i}\mathrm{n} \mathrm{m}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{a}\mathrm{r} \mathrm{p}\mathrm{h}\mathrm{a}\mathrm{s}\mathrm{e}}{\mathrm{C}\mathrm{a}\mathrm{r}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{o}\mathrm{i}\mathrm{d} \mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{i}\mathrm{n} \mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{f}\mathrm{o}\mathrm{o}\mathrm{d} \mathrm{m}\mathrm{a}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{x}}}·100$$

In addition to the evaluation of the pure extracts, the conventional extract (CE) was incorporated into different edible oils (olive pomace oil (OPO), fish oil (FO), and corn oil (CO)) at a concentration of 1% w/w. The bioaccessibility assessment of these samples was performed using the same procedure as described above. All oils were purchased from a local market.

2.10. Statistical Analysis

The experimental data were analyzed using an analysis of variance (ANOVA) (STATISTICA 7, StatSoft, Inc., Tulsa, OK, USA), with significant differences in mean values estimated at the probability threshold p < 0.05. Duncan’s multiple range test was performed to separate the data’s means when significant differences were observed.

3. Results and Discussion

3.1. Conventional Extraction of Bioactives from Dried TP Powder

Fresh tomato pomace with an initial moisture content of 88% w/w was subjected to air drying for 24 h at 40 °C. Subsequently, exhaustive extraction was performed using a percolator tightly packed with the dried and powdered TP material (25 g). EtOAc was pumped into the extractor, with an average flow of 9 mL/min. The extractor was filled up with 63 mL of solvent and then the extract started flowing from the upper outlet valve. In Figure 1 the overall yield of carotenoids is presented throughout the extraction time. At 120 min, EtOAc appeared to have depleted the TP material, and the recovered carotenoids were determined to be equal to 505 mg /kg dw, from the analysis of the total extract. Taking into consideration both the volumes of the produced extract as well as the solvent trapped in the TP, the overall solid/liquid ratio amounted up to 1:46 g/mL.

According to Figure 1, it can be observed that the yield increased rapidly for the first 5 min recovering 315 mg carotenoids/kg dw, a value that corresponds to 62.4% of the maximum yield. The solid-to-liquid ratio was determined approximately to be equal to 1:5 g/mL (including the absorbed solvent by the TP). The extraction proceeded at a lower rate in the period between 5 and 120 min. The remaining 37.6% of the total yield was acquired with the use of 1030 mL of ethyl acetate, which corresponds to a solid-to-liquid ratio of 1:41 g/mL. The observation is reasonable, since the extraction process is not a uniform phenomenon, but is characterized by two distinct phases. In the initial rapid phase, the solvent solubilizes all the readily available components distributed on the surface of the particles of the solid material. Τhe second phase is kinetically slower than the initial, governed by diffusion phenomena, including the penetration of solvent in the micro-structure of the particles, the formation of a solvation layer surrounding the extractable components, and the mass transfer of the dissolved compounds from the particle to the continuous phase [21].

The total extract was analyzed by HPLC-DAD to characterize the extracted carotenoids. The respective chromatogram, monitored at 450 nm is presented in Figure 2. Two basic components and three additional trace peaks were detected, each one bearing the spectral characteristics of carotenoids. Peaks with a retention time (r.t.) of 13.21 min, and 15.05 min matched both the spectra, as well as the retention times of lycopene and β-carotene, respectively. For confirmation, solutions of the standard compounds were spiked in the extract, and further HPLC analysis revealed the increase of the two peaks.

According to the calibration curves, lycopene and β-carotene were quantified at 123.8 and 386.2 mg/kg dw, while the trace carotenoids were determined at 23 mg lycopene equivalents/kg dw (total carotenoids: 533 mg/kg dw). The conventional photometric determination of total carotenoids (505 mg lycopene equivalents/kg dw) slightly underestimates the quantification of the compounds.

3.2. Results from Microwave-Assisted Extraction of Carotenoids from Tomato By-Products

Figure 3 presents the effect of MAE parameters on the concentration of carotenoids in the extracts, expressed as mg carotenoids/kg dw. The maximum yield values for total carotenoids were found to be 592.5 and 540.5 mg carotenoids/kg dw, when MAE was performed at 150 W, 50 °C, with a solid-to-liquid ratio of 1:30 g/mL for 30 and 20 min, respectively. The corresponding values, when using the 1:10 g/mL ratio, were lower (497.4 and 444.7, respectively), indicating that a higher amount of solvent is required for the effective extraction of carotenoids from tomato processing by-products when MAE is applied (Figure 3). A general observation regarding the effect of extraction time on the yield of extracted carotenoids is that, as the extraction time increases, the yield of total carotenoids also increases. A statistical analysis of the results indicates that the effect of time is significant (p < 0.05). It was observed that the richest extracts in lycopene were achieved with extraction times of 20 and 30 min. The maximum yield value, 592.5 mg carotenoids/kg dw, was observed after 30 min, under conditions of 1:30 g/mL, 150 W, and 50 °C. The immediately following higher yield values, 558.5 and 540.5 mg carotenoids/kg dw, were obtained for a 20-min extraction time at 150 W, with a solid-to-liquid ratio of 1:30 g/mL at 30 and 50 °C, respectively. Given that the yield values for 20 and 30 min of extraction do not show statistically significant differences, the 20-min time can be selected as the optimal maximum duration required for carotenoid extraction using MAE for time-saving purposes, even though the maximum carotenoid value was recorded at 30 min.

Regarding the effect of microwave power, it was observed that applying 150 W resulted in higher (p < 0.05) yields of total carotenoids, indicating that the effect of microwave power is significant (p < 0.05). Specifically, values for the carotenoids from the extracts obtained with MAE at 150 W showed an average yield of 470 mg carotenoids/kg dw, followed by the yield values for the extracts with the MAE application at 300 and 500 W, with a yield of 430 mg carotenoids/kg dw, with no significant difference between them. Regarding the effect of microwave power, the results indicated that the most effective extractions were achieved with the lowest power, making the MAE process economically and energetically advantageous. Applying higher microwave power leads to channeling more energy through radiation into the sample for more rapid heating, due to the intensified rotation of dipoles and ion migration. Lasunon et al. demonstrated that increasing the microwave power levels from 300 to 450 W resulted in reduced quantities of trans-lycopene and β-carotene, attributed to the degradation of trans-lycopene and β-carotene during extraction [21]. Similar findings were reported by Chan et al., who found that extracting carotenoids at high microwave power levels caused the degradation of bioactive compounds, possibly due to the thermal degradation of lycopene at high microwave power (450 W) when the extraction temperature ranged from 47 to 75 °C [22]. Therefore, in this study, an additional benefit of applying lower power and achieving higher yields is the maintenance of sample temperatures at low levels, protecting the bioactive constituents extracted from tomato pomace using MAE.

Moreover, regarding the selection of the optimal extraction temperature, it was observed that an increase in the extraction temperature led to significantly (p < 0.05) higher yields of total carotenoids. The highest yield values were recorded when MAE was performed at 50 °C, while the lowest values were observed at 30 °C. Between 40 and 50 °C, no statistically significant differences were found, exhibiting an average yield of 456.0 mg carotenoids/kg dw and 463.2 mg carotenoids/kg dw, respectively. The abrupt increase in temperature should be avoided because it may potentially lead to the degradation of the carotenoids, which are thermosensitive compounds. Veggi et al. proposed 40 °C as the optimal extraction temperature for MAE, due to the preservation of carotenoids’ thermal sensitivity [18]. Conventional extraction often requires heat to facilitate the migration of the solvent toward the chromophoric compounds, such as carotenoids from tomato processing by-products (tomato pomace). Although increased temperatures are associated with improved solubility and membrane disruption of the above compounds, exposure to heat should be limited whenever possible, due to the thermosensitivity of carotenoids when dissolved in a solvent. Previous studies have shown that applying thermal treatment for a duration exceeding one hour favored the conversion of trans-to-cis isomers of lycopene over time in tomato products [40]. MAE can address the aforementioned issues, as this technology rapidly heats primarily polar constituents due to dipole rotation and ionic sliding [41]. Based on the above, the higher carotenoid values recorded with MAE (592.5 mg carotenoids/kg dw) compared to those of the conventional extraction (505 mg carotenoids/kg dw), which lasted 2 h (versus 20 min with MAE), can possibly be explained by the negative impact of the longtime extraction and consequently, the degradation of thermosensitive carotenoids, resulting in lower yield values. Lasunon et al. evaluated the effectiveness of MAE under different processing conditions for extracting bioactive compounds from tomato processing by-products, and their results indicated that higher microwave power and longer extraction times led to greater degradation of the extracted bioactive compounds [21]. The results of this study align with the above findings, as lower microwave power levels and shorter extraction times resulted in higher yields, while different conditions led to the degradation of the aforementioned bioactive compounds and consequently lower yields. Applying 150 W at 40 °C for 20 min with a solid-to-liquid ratio of 1:10 g/mL, the maximum extraction yield of carotenoids was recorded, comparable to that achieved with conventional extraction, but in significantly less extraction time.

The extracts obtained by MAE for 20 min and a solid-to-liquid ratio 1:10 were further analyzed by HPLC. Contrary to the underestimation of carotenoids by the photometric determination in the case of conventional extraction, in the abovementioned analyzed extracts of microwave extractions, photometry overestimated the total carotenoids. The correlation between HPLC-DAD and photometric results are presented in Figure 3, where CY values by HPLC were determined as the sum of lycopene, β-carotene, and the carotenoid peaks in traces. The overestimation of carotenoids by the photometric method could suggest partial degradation of the carotenoids due to the extraction conditions under microwave irradiation to compounds that produce a response at 473 nm, the wavelength at which the photometric determinations were carried out. Comparing the chromatograms between the conventional and the microwave extracts at 473 nm, it was observed that the latter produced higher baseline noise and multiple tracial peaks with non-carotenoid spectra. Also, according to Figure 4, it can be noticed that, in both cases of 500 W irradiation, the HPLC results were even lower compared with 150 W, and 300 W, evidence that could support the hypothesis of microwave power-induced degradation. In terms of individual compounds, the lycopene derived by the conventional extraction (151.2 mg/kg dw) is quite similar with the lycopene recovered by microwaves (101.4–143.3 mg/kg dw, 150 W, 50 °C), but this is not the case for β-carotene. The β-Carotene recovered by conventional extraction was determined to be 471.5 mg/kg dw; while, in the case of microwaves, β-carotene ranged between 164 (500 W, 40 °C) and 211.8 (150 W, 50 °C) mg/kg dw, indicating that β-carotene is more susceptible to degradation under microwave irradiation. This is a clear indication that microwaves intensely affect β-carotene through degradation. On the other hand, in addition to carotenoids, several non-carotenoid colored compounds could also absorb light, particularly if degradation or oxidation occurred during extraction or processing, including short-chain apocarotenoids and various oxidized derivatives, and early-stage Maillard reaction products, leading to higher values of absorbance at 473 nm and impacting the results in the case of photometric measurements.

3.3. Results from High-Pressure-Assisted Extraction of Carotenoids from Tomato By-Products

The effect of processing time on HPAE was studied at an applied pressure of 250 MPa, and the results are presented in Figure 5. Interestingly, increasing the extraction time under HP from the time required to achieve the desired pressure (ca. 40–60 s) to 20 min did not significantly affect the extractability of the studied BACs. The total carotenoid yields after 0, 10, and 20 min of HPAE were approximately 57.7 (±5.73), 63.4 (±3.94), and 55.2 (±8.83) mg/kg dw, respectively. Similar results on time effectiveness have been previously reported for the HPAE of phenolic compounds from olive pomace at 250–600 MPa for 1–5 min processing time [20,42].

Figure 6 presents the results of HPAE of carotenoids from tomato pomace for different applied pressures, and solid/liquid ratios of 1:30, 1:10, and 1:5 g/mL. The values represent the yield of total carotenoids (mg/kg dw) after applying HPAE for a time equal to the time required to achieve the desired pressure (40–60 s, depending on the pressure level) at an ambient temperature.

The increase in the applied pressure from atmospheric to ultra-high pressures had a significant (p < 0.05) effect on extractability (mean yields: 65.7, 66.0, and 84.2 mg/kg dw for application of 250, 450, and 650 MPa, respectively). The effect of ultra-high pressures is related to changes in the diffusion coefficient due to alterations in the cellular membrane of plant tissues under HP conditions, resulting in increased permeability of the cellular membranes, thus enhancing the penetration of the extraction solvent into the tomato cells [43]. At the same time, HP causes protein denaturation, to which the carotenoids’ molecule is bound, destruction of the tomato cell wall, and thus facilitates the efficient pigment release and extraction of carotenoids [44,45].

Regarding the effect of the solid/liquid ratio, it was observed that a decrease of the ratio (i.e., increasing the amount of solvent) leads to an increase in the yield of total carotenoids, especially for solid/liquid ratios higher than 1:10 g/mL (statistical average yield: 62.0, 67.0, and 76.7 mg/kg dw for ratios of 1:5, 1:10, and 1:30 g/mL, respectively). Similar results have been reported in the literature for carotenoid extraction from tomato by-products [25] and for the use of various solvents at ratios of 1:4, 1:6, and 1:10 g/mL for total phenols, tannins, and flavonoids from date palm using ethanol at liquid/solid ratios of 10–70 mL/g [46]. In order to determine the most efficient conditions for HPAE of carotenoids from tomato pomace, the extraction yields were calculated as percentages relative to both the HPAE-untreated sample (i.e., TP diluted in an appropriate mass of solvent and remain in calm for the time equal to a full HP-cycle) and the corresponding conventional solid/liquid extraction (

Table 2. Percentage of extraction yield (%) of HPAE carotenoids compared to untreated and conventionally treated tomato pomace.

	Yield Increase Compared to HPAE-Untreated Pomace			Recovery Compared to Conventionally Extracted Carotenoids
Solid/Liquid Ratio (g/mL)	1:30	1:10	1:5	1:30	1:10	1:5
Untreated	-	-	-	12.5	10.3	11.9
250 MPa	13.6	10.8	13.0	14.2	11.4	13.4
450 MPa	24.8	9.96	3.40	15.6	11.3	12.3
650 MPa	48.1	55.1	30.1	18.5	16.0	15.4

). The application of HPAE using EtOAc at solid/liquid ratios of 1:30 and 1:10 g/mL resulted in the majority of cases in increased yields of extracted carotenoids compared to the untreated samples across all applied pressures.

The composition of the extracts obtained by HPAE at a solid/liquid ratio of 1:10 g/mL was examined by HPLC-DAD analyses. Lycopene, β-carotene, and total carotenoids presented a slight increase with the applied pressure: lycopene, 60.4 mg/kg dw (250 MPa), 66.8 mg/kg dw (450 MPa), 71.9 mg/kg dw (650 MPa); β-carotene, 207.1 mg/kg dw (250 MPa), 231.9 mg/kg dw (450 MPa), 255.3 mg/kg dw (650 MPa); total carotenoids, 274.3 mg/kg dw (250 MPa), 306.9 mg/kg dw (450 MPa), 336.8 mg/kg dw (650 MPa) (Figure 2). The ratio of β-carotene/lycopene ranged between 3.4 and 3.5, close to the ratio derived from the conventional extraction, i.e., 3.1, but higher. The microwave extracts were clearly different from the above, with the respective ratio to vary between 1.5 and 1.6. This might suggest that the main contribution of HPAE is the more efficient recovery of β-carotene. The specific coefficient of absorption (E^1%) for lycopene in ethyl acetate at 473 nm was determined by Strati and Oreopoulou to be 2963 (L g⁻¹ cm⁻¹) [9]. The respective value for β-carotene (in ethyl acetate at 473 nm) was determined to be 1528 (L g⁻¹ cm⁻¹), almost half of lycopene. This fact means that, if the photometric determination of β-carotene is conducted as lycopene equivalents, the concentration of the compound will be underestimated. Therefore, the increased yield of β-carotene by the use of HPAE, which was verified by HPLC, did not correspond to a proportional increase by photometric determination (lycopene equivalents).

However, the CYs with the use of HPAE remained lower than the conventional extraction according to either the photometric measurements or the HPLC-DAD analyses. The maximum yield of the total carotenoid recovery, relative to conventional extraction, was observed in the case where HPAE was performed using a solid/liquid ratio of 1:30 g/mL and at 650 MPa, resulting in recovery percentages of approximately 18.5% compared to conventional extraction (photometric measurements). In order to investigate whether HPAE could lead to higher recovery yields of carotenoids from tomato pomace, the latter was subjected to HPAE using ethyl acetate at solid/liquid ratios of 1:30, 1:10, and 1:5 g/mL, at 650 MPa, and an ambient temperature. Subsequently, the samples underwent further extraction under stirring at 350 rpm at atmospheric pressure and an ambient temperature for 24 h. Based on the results obtained (Figure 7), it became evident that, when HP is applied as a pre-treatment step of the raw material, prior to CE, the recovery of bioactive compounds can be enhanced.

As in the case of carotenoid extraction under HP conditions and in the case of HP pre-treatment of tomato pomace, the solid/liquid ratio plays a crucial role in improving the recovery of bioactive constituents, with higher solvent ratios leading to better carotenoid recovery. The maximum extraction rate was observed for the first 8 h of extraction, while the concentration of carotenoids reached its maximum value after 20–24 h of extraction. These maximums were 277, 258, and 182 mg/kg dw for solid/liquid ratios of 1:30, 1:10, and 1:5 g/mL, respectively, and they were 362%, 342%, and 336% higher than the corresponding zero-time concentrations, i.e., immediately after HP processing, with a recovery percentage reaching 55%, 51%, and 36%, respectively, compared to the carotenoid concentration after conventional extraction. Since slight but not statistically significant (p > 0.05) differences were observed between the solid/liquid ratios of 1:30 and 1:10 g/mL, the use of lower solvent amounts (1:10 g/mL) is suggested, aligning with the circular economy approaches. Finally, taking into consideration the effect of HP on the extractability of carotenoids, the optimal extraction conditions of HPAE of carotenoids include the pre-treatment of TP at 650 MPa and ambient temperature for the time required for pressure build-up, followed by a simple stirring process for 24 h, which leads to the efficient recovery of carotenoids using quite lower amounts of solvent compared to conventional extraction.

As expected, the HPLC analysis of the final extract at a solid/liquid ratio of 1:10 at 24 h revealed a different picture concerning the yields of the main compounds and the comparison with the conventional extraction. Lycopene was determined to be 158.1 mg/kg dw, β-carotene to be 516.5 mg/kg dw, and total carotenoids to be 697.9 mg/kg dw. The above values exceed the respective values for conventional extraction by 28% (lycopene), 34% (β-carotene) and 31% (total carotenoids). This result is also depicted in Figure 4, together with the rest of the studied extracts.

3.4. Bioaccessibility of Extracted Carotenoids

Lycopene bioaccessibility was assessed using extracts obtained under the optimal conditions for each extraction technique. The results are presented below in

Table 3. Bioaccessibility (%) of lycopene in extracts obtained through MAE, HPAE, and CE.

Sample	Lycopene Bioaccessibility (%)
MAE: 150 W, 50 °C, 1:10 g/mL, 20 min	3.4 a (±0.8)
HPAE: 650 MPa, 1:30 g/mL, 1 min	3.9 a (±0.7)
CE: 2 h	1.6 b (±0.4)

Superscript letters a and b represent significant differences (p < 0.05) between the samples. Data are described as mean value ± standard deviation (n = 3).

.

The results indicate low lycopene bioaccessibility across all extraction methods. No statistically significant difference (p > 0.05) was observed between MAE and HPAE, while CE exhibited lower bioaccessibility. These low percentages are consistent with previous findings in the literature, which report that lycopene bioaccessibility is often limited, even falling below detectable levels [47,48]. Mixed micelles, formed by bile acids in the small intestine, are essential for the absorption of lipophilic compounds, which have limited solubility in the aqueous environment of the gastrointestinal tract. Moreover, carotenoids are highly susceptible to oxidation, which can further reduce their bioavailability after digestion. Non-thermal extraction techniques, such as HPAE, can be used to recover lycopene while minimizing heat-induced degradation, thus preserving some of its quality attributes. However, the bioaccessibility percentages reported in such cases remain relatively low [45]. Lycopene is also reported to undergo trans (all-E) to cis (Z) isomerization during processing, with cis isomers being more bioaccessible [49]; however, we did not assess the isomer profiles in this study. The bioaccessibility percentages reported for non-thermal extraction remain relatively low [47], in accordance with our results.

Considering the lipophilic nature of carotenoids and the essential role of dietary fats in micelle formation during digestion, many studies have shown that carotenoid bioaccessibility significantly increases when co-consumed with lipids [50,51]. Accordingly, in this study, lycopene extract (CE) was incorporated into selected edible oils with different fatty acid composition (olive pomace oil (OPO), fish oil (FO), and corn oil (CO)) to evaluate their effect on lycopene bioaccessibility. The results are presented in Figure 8 below.

All tested oils exhibited significantly higher lycopene bioaccessibility compared to the pure extract, consistent with the previous findings that emphasize the positive effect of lipids on carotenoid absorption. Among the oils tested, OPO exhibited the highest lycopene bioaccessibility percentage (28.1 ± 0.4%), while CO showed the lowest (8.1 ± 0.1%). These findings align with the existing literature on oils with similar fatty acid composition. Liu et al. reported total carotenoid bioaccessibility in carrots ranging from 5 to 15% with the addition of 2% corn oil, depending on the homogenization method used [52]. Similarly, González-Casado et al. observed lycopene bioaccessibility percentages of approximately 25–30% when olive oil was added, while sunflower oil presented lower values (5–20%), depending on the tomato product [53]. The variation in bioaccessibility is largely influenced by the oil’s fatty acid composition. Oils rich in polyunsaturated fatty acids (PUFAs), such as FO and CO, tend to form larger mixed micelles in the small intestine. These larger micelles often face greater difficulty diffusing through the unstirred water layer, thereby limiting their uptake by intestinal epithelial cells. Additionally, PUFAs are highly prone to oxidation, which can degrade carotenoids during their passage through the gastrointestinal tract (GIT), further reducing bioaccessibility [46]. By contrast, oils rich in monounsaturated fatty acids (MUFAs), such as OPO, are known to enhance carotenoid bioavailability. In vivo studies suggest that MUFAs improve carotenoid absorption by stimulating a stronger postprandial lipemic response, whereas long-chain n-3 PUFAs result in a smaller increase in blood fat levels [54,55]. Furthermore, all tested oils contain long-chain fatty acids, which contribute to the formation of mixed micelles with high solubility for lipophilic compounds. They also promote the production of chylomicrons in epithelial cells, facilitating the absorption of carotenoids and reducing the effects of first-pass metabolism [56]. Techniques such as encapsulation may further enhance the bioaccessibility of lipophilic compounds like lycopene, as reported in the literature [47,48,56].

Beyond carotenoid extraction, tomato pomace can be valorized more effectively within a cascade biorefinery framework, where different bioactives are recovered in a sequential manner from the same biomass. Such an approach enhances both economic feasibility and resource efficiency, since pectin, polyphenols, proteins, and carotenoids can be obtained as complementary product streams rather than isolated targets. Ninčević Grassino et al. illustrated this potential by combining high hydrostatic pressure and ultrasound-assisted techniques to recover pectin and polyphenols from tomato peel waste [27]. In this context, carotenoid extraction via MAE or HPAE could be strategically integrated as a subsequent step, enabling multi-product valorization and supporting circular economic objectives. Future research should therefore focus on the design of integrated biorefinery schemes that link these complementary recovery processes into a coherent, scalable strategy.

Nevertheless, it should be noted that both MAE and HPAE, despite their advantages in reducing extraction time and solvent consumption, remain dependent on organic solvents, such as ethyl acetate. This reliance constitutes a limitation for food-related applications, where the range of permitted solvents is narrow and concerns about solvent residues and environmental impact persist. Similar extraction challenges related to solvent dependence and scalability have been highlighted by Radić et al. [57]. Therefore, while our findings demonstrate the potential of MAE and HPAE for the valorization of tomato pomace, future research should prioritize the exploration of greener solvent systems (e.g., natural deep eutectic solvents, limonene) or solvent-free technologies (e.g., supercritical CO₂) to further enhance sustainability and industrial feasibility.

4. Conclusions

The valorization of tomato pomace through the recovery of bioactive carotenoids represents a promising strategy for the sustainable utilization of food industry by-products. In this study, MAE and HPAE were evaluated as alternative technologies to conventional solvent extraction, for the efficient recovery of carotenoids, particularly lycopene and β-carotene. The findings clearly demonstrated that both MAE and HPAE offer distinct advantages, depending on the processing objectives and operational constraints.

MAE proved to be highly effective in extracting carotenoids from tomato pomace, achieving yields comparable to or exceeding those obtained through conventional extraction, while significantly reducing processing time and solvent consumption. Optimal conditions amongst those investigated were established at 150 W, 50 °C, and a solid/liquid ratio of 1:10 g/mL, with an extraction time of 20 min. Notably, lower microwave power favored the preservation of thermosensitive carotenoids, minimizing degradation risks. However, the application of higher microwave power or extended processing times was associated with the potential degradation of β-carotene, highlighting the necessity for strict control of operational parameters to preserve compound integrity. Despite the benefits observed at the laboratory scale, the scalability of MAE remains a critical consideration due to the need for specialized equipment and energy management protocols in industrial applications.

HPAE applied at pressures up to 650 MPa demonstrated the ability to enhance cell wall disruption and solvent penetration, facilitating carotenoid release at an ambient temperature. While carotenoid yields from direct HPAE were lower than those obtained via MAE or conventional extraction, a significant finding emerged when HPAE was applied as a pre-treatment step. This approach, followed by subsequent conventional extraction under stirring, resulted in enhanced carotenoid recovery, surpassing conventional extraction yields by approximately 55% at the optimal HPAE conditions (650 MPa, 25 °C, and a solid/liquid ratio of 1:10 g/mL). This combined process leverages the structural disruption achieved under high pressure and the prolonged solvent contact during stirring to maximize bioactive compound recovery, with the added benefit of reduced solvent usage. HPAE, as a nonthermal technique, also mitigates thermal degradation risks, making it particularly suitable for sensitive compounds.

Furthermore, this study assessed the bioaccessibility of the recovered carotenoids, revealing uniformly low bioaccessibility percentages across all extraction methods. MAE and HPAE extracts exhibited slightly higher bioaccessibility compared to conventional extracts. However, the absolute values remained below 5%, consistent with the known challenges associated with the intestinal absorption of non-polar carotenoids like lycopene. Co-administration with edible oils, particularly those rich in monounsaturated fatty acids, such as olive pomace oil, significantly improved carotenoid bioaccessibility, underscoring the critical role of dietary lipids in enhancing micellar incorporation during digestion.

In conclusion, both MAE and HPAE represent viable, more efficient, and solvent-saving alternatives for the recovery of high-value carotenoids from tomato processing by-products compared to conventional extraction methodologies. Nevertheless, in line with the circular economy principles, future studies should explore greener alternatives, such as natural solvents (e.g., limonene, natural oils, etc.), or solvent-free techniques, like supercritical CO₂, to better balance efficiency with environmental impact. While MAE offers rapid, solvent-efficient extraction, HPAE’s potential is maximized when employed as a pre-treatment. Although this study focused exclusively on carotenoid recovery, comprehensive valorization of tomato pomace should ideally follow a cascade biorefinery approach. Previous work has demonstrated the sequential recovery of pectin and polyphenols from TP, highlighting the feasibility of multi-product valorization. In this context, carotenoid extraction using MAE or HPAE could be integrated as one step within a broader cascade strategy, thereby enhancing overall resource efficiency and alignment with circular economy principles. To ensure that these approaches are not only environmentally sustainable but economically viable at larger scales, future research should incorporate a life cycle assessment (LCA) and a technoeconomic analysis (TEA). These tools are essential to evaluate the overall environmental footprint, economic feasibility, and scalability of the technologies, providing a robust framework for guiding their transition from laboratory to industrial application. Additionally, research should continue to optimize industrial-scale applications and to develop techniques, such as encapsulation or nanoemulsion-based delivery systems, to enhance both the process and storage stability as well as the bioaccessibility of recovered carotenoids, thereby supporting the broader goals of the circular economy and functional food development.