Green solvent extraction of oil and phenolic compounds from marolo and soursop seeds: Comparative study

Abstract

Several exotic Brazilian fruits, including marolo (Annona crassiflora Mart.) and soursop (Annona muricata L.), have been identified as promising sources for the extraction of oil and phenolic compounds. This study aimed to conduct a comparative evaluation of various solvents, including hexane, ethanol, acetone, and isopropanol, for the purpose of extracting oil and bioactive compounds from the seeds of these fruits. Extractions were carried out at three temperatures (35, 45, and 55 °C) and two solvent-to-seed ratios (5:1 and 10:1, w/w). Hexane was the solvent that demonstrated the highest oil extraction efficiency, reaching up to 99.61% for marolo seeds and 96.71% for soursop seeds; however, its efficacy in extracting phenolic compounds was found to be minimal. In contrast, ethanol and isopropanol, classified as green solvents, exhibited superior efficiency in extracting phenolic compounds, particularly at 45 °C and higher dilution ratios, thereby enhancing antioxidant potential, although with lower oil yields (56.28% to 94.32% for marolo and 54.66% 82.84% for soursop when ethanol was used). Extraction with ethanol at 45 °C was found to yield a favorable balance between oil yield and phenolic content, with oil yields of 71.25% (5:1) and 91.89% (10:1) for marolo, 61.95% (5:1) and 80.82% (10:1) for soursop. Additionally, the integration of mass balance calculations with the retention index (RI) enabled a more comprehensive evaluation of solvent performance, demonstrating that solvent type significantly affected retention behavior, with hexane showing a lower RI and green solvents showing higher retention, impacting downstream solvent recovery. These findings underscore the merits of employing green solvents in the development of more sustainable and safer extraction processes while providing a more robust basis for the valorization of marolo and soursop seed by-products.

Keywords:
Marolo oil seed; Soursop oil seed; Ethanol; Isopropanol; Hexane; Bioactive compounds; Antioxidant capacity

Highlights

Ethanol achieved the highest extraction of phenolics and antioxidants among solvents

Hexane achieved the highest oil yields but showed low extraction of bioactive compounds

At 45 °C, alternative solvents achieved higher recovery of key functional compounds

1 Introduction

Brazil is characterized by its extensive biodiversity, which is distributed across a wide array of morphological and phytogeographic regions. The Cerrado, a savanna biome, is distinguished by its abundant and diverse array of exploitable plant species, which includes a wide variety of fruit-bearing plants. The consumption of these fruits primarily occurs in their natural state; however, they also serve as a source for the production of new products. Among these fruits, soursop (Annona muricata L.) and marolo (A. crassiflora Mart.) are widely consumed fresh by the local population. The industrialization of these fruits into various products, such as frozen pulp, jellies, and beverages, has proven to be a profitable and sustainable development alternative for the region. Seeds represent a significant byproduct of the fruit processing industry. These seeds contain a high concentration of oil and bioactive compounds (Almeida et al., 2024; Arruda & Pastore, 2019; Luzia & Jorge, 2014; Villacís-Chiriboga et al., 2023). On average, the weight of soursop and marolo seeds constitutes approximately 15% to 20% of the fruit's total mass, and these seeds contain approximately 30% of crude oil on a dry basis (Menezes et al., 2019).

In addition to their high oil content, soursop and marolo seeds have been found to be rich in bioactive compounds, such as phenolic compounds exhibiting high antioxidant activity (Almeida et al., 2024; Menezes et al., 2019; Mesquita et al., 2021; Roesler et al., 2006). O-coumaric acid and rutin are the predominant phenolic compounds present in these seeds (Menezes et al., 2019). In terms of fatty acid composition, unsaturated fatty acids predominate, with oleic acid as the major fatty acid in both seeds (48.54% to 49.75% for marolo and 40.35% to 47.07% for soursop seeds) (Luzia & Jorge, 2013; Menezes et al., 2019; Villacís-Chiriboga et al., 2023).

The extraction of oil from plant seeds is a key step in the valorization of fruit by-products, especially those that are rich in lipids and bioactive compounds. In addition to functioning as a valuable source of energy and essential fatty acids, vegetable oils also contain minor components such as phenolic compounds, which enhance their nutritional and functional properties. Efficient extraction methods are essential to optimize oil yield while preserving bioactive compounds. The selection of extraction conditions, including the type of solvent, temperature, and solvent-to-solid ratio, is critical in determining the quantity and quality of the extracted oil.

Hexane is the most commonly used solvent in the oil extraction industry due to its low cost and high solubility for lipids, resulting in high extraction yields. Therefore, hexane also presents several disadvantages: it is derived from non-renewable fossil fuels, is highly flammable, and exhibits neurotoxic properties (Sawada et al., 2014; Tir et al., 2012; Zhuang et al., 2018). Consequently, replacing hexane with alternative solvents has attracted increasing interest, particularly when alternative solvents are widely available, cost-effective, and offer comparable extraction efficiency and improved safety and sustainability profiles.

Ethanol is produced through biotechnological processes that are environmentally friendly and safe for human health. From an economic perspective, ethanol exhibits several advantages, including its large-scale production in Brazil and its ease of recovery and reuse (Oliveira et al., 2012). According to the findings of previous studies, ethanol extracts may also contain sugars, phosphatides, phenolic compounds, pigments, and waxes, in addition to oil (Beckel et al., 1948; Rittner, 1992). Consequently, ethanol emerges as a promising alternative solvent due to its cost-effectiveness and accessibility from various biological materials through straightforward technologies. Although ethanol is flammable, it is non-toxic and poses a lower risk of handling hazards when compared to hexane (Terigar et al., 2011). The use of ethanol has been demonstrated to reduce concerns regarding the potential toxicity of animal feed derived from residual meals (Ferreira-Dias et al., 2003).

Isopropanol is another potential alternative solvent, as it can be produced from renewable sources. This solvent extraction produces oils characterized by high quality and the presence of tocopherols and phenolic compounds. Compared to hexane, isopropanol extraction is safer and less toxic (Baker & Sullivan, 1983; Gandhi et al., 2003). Acetone has also been considered a substitute for isopropanol due to its lower latent heat of vaporization and similar solubility properties to those of alcohols. However, acetone presents a fire hazard, and its risk level is comparable to that of hexane (Tir et al., 2012). Despite the availability of these alternatives, the non-selective nature of these solvents may result in the co-extraction of proteins, carbohydrates, and other compounds in conjunction with lipids.

Despite the recognized nutritional and functional potential of marolo and soursop seeds, studies on their valorization through oil and phenolic compound extraction are still limited. There is a lack of comparative analyses concerning the influence of different solvents on the efficiency and selectivity of the extraction process. This study aims to address this knowledge gap by evaluating the performance of two green solvents (ethanol and isopropanol) in comparison to acetone and hexane under varying extraction conditions. To this end, the yield of oil, total phenolic content, antioxidant activity, and retention index will be assessed.

2 Materials and methods

2.1 Raw materials and chemicals

The soursop and marolo fruits were procured from a local market in Lavras, Brazil. The fruits were washed, sanitized (hypochlorite solution, 200 g/kg), and manually pulped. The seeds were stored in plastic bags (polyethylene) at -18 °C. For the extraction process, the seeds were subjected to a drying procedure in a vacuum oven (absolute pressure = 16.8 kPa) (Tecnal, model TE-395, Piracicaba, Brazil) for a duration of 48 hours at a temperature of 45 °C. The dry seeds were subsequently ground and stored in desiccators. The composition of soursop and marolo seeds is outlined below: soursop contained 29.5% of lipids on a dry basis, 820 mg GAE/100 g of seeds of total phenolic compounds, and 77.3% of antioxidant activity measured by the DPPH method; marolo exhibited 31.1% of lipids on a dry basis, 3204 mg GAE/100 g of seeds of total phenolic compounds, and 91.3% of antioxidant activity using the same methodology. The determination of total lipids followed the methodology proposed by the Association of Official Analytical Chemists (2016).

Acetone, ethanol, isopropanol and hexane were used as solvents for the extraction of seed oil (Dinâmica Química Contemporânea Ltd., Diadema, Brazil).

2.2 Solid-liquid extraction using pure solvents

The methodology used for the solid-liquid extraction of marolo and soursop seeds was described by Araújo et al. (2019). Seed extractions were performed by adding 6 g of dried and ground seeds to a specific mass of solvent (ethanol, hexane, isopropanol, or acetone) to achieve solvent-to-solid mass ratios (SSR) of 5:1 and 10:1 (w:w). The mixtures were prepared in 125 mL Erlenmeyer flasks, which were sealed to prevent any loss of solvent. The mixture (solvent + seed) was incubated at a constant temperature (35 °C, 45 °C, and 55 °C) with stirring at 120 rpm for 16 hours, which was the time required to reach equilibrium as determined by preliminary tests. All treatments were conducted in triplicate.

A schematic representation of the extraction process and the resulting fractions is presented in Figure 1. After extraction, the system was separated into two main phases: the extract phase (EP), consisting of the solvent fraction containing the extracted oil and co-extracted compounds, and the raffinate phase (RP), composed of the solid residue with retained solvent. The extract phase was used to obtain the oil, which was subsequently used for the determination of oil yield, total phenolic content, antioxidant activity, and fatty acid composition. The raffinate phase was used to evaluate the retention index (RI), representing the amount of solution retained in the solid matrix. This approach enabled a comprehensive assessment of both extraction efficiency and solvent retention behavior.

Figure 1
Schematic diagram of the extraction process and phase separation (extract phase and raffinate phase), indicating the analytical determinations performed for each fraction.

At the end of the extraction period, a sample (approximately 10 mL) of the EP was collected using a micro syringe connected to a septum, with careful for maintaining the system’s temperature. The solvent in the samples was evaporated to constant weight at 60 °C in a vacuum oven (absolute pressure = 16.8 kPa) (Tecnal, TE-395, Piracicaba, Brazil). This procedure enabled the determination of the mass fraction of the solvent (2) in the extract phase ($w_{\mathrm{2,}EP}$), as well as the mass fraction of oil in the same phase ($w_{oil,EP}$). The inert (3) mass fraction in the extract phase ($w_{\mathrm{3,}EP}$) was considered to be zero.

After sampling from the EP, the remaining mixture was transferred to Falcon tubes and centrifuged (FANEM, Excelsa II/206-BL, Piracicaba, Brazil) at 5,000 rpm for one minute. The tubes were then placed in a water bath at a constant temperature according to the treatment. The EP was then separated from the RP, and a sample of the RP was placed in a Petri dish and dried under the same conditions as the EP sample. This procedure allowed for the determination of the mass fraction of the solvent (2) in the RP ($w_{\mathrm{2,}RP}$). The separate extract phase was used for the determination of phenolic compounds and antioxidant activity. The mass fractions of total phenolics (TP) ($w_{TP,EP}$) in the extract phase were also determined.

The variables mass of extract phase (), mass of raffinate phase (), mass fraction of the oil in the RP ($w_{oil,RP}$), and the inert mass fraction in the RP ($w_{\mathrm{3,}RP}$) were determined by global mass balance and the system’s components of interest in the system (Equations 1-4) using the Solver add-in in Excel.

The overall mass balance is given by Equation 1.

$M^M=M^{seed}+M^{solvent}=M^{RP}+M^{EP}$ (1)

The mass balance for the components analyzed is given by Equation 2.

$w_{oil,M}{M}^M=w_{oil,seed}{M}^{seed}+w_{oil,solvent}{M}^{solvent}=w_{oil,RP}{M}^{RP}+w_{oil,EP}{M}^{EP}$ (2)

The mass balance for the solvent (2) is given by Equation 3.

$w_{\mathrm{2,}M}{M}^M=w_{\mathrm{2,}solvent}{M}^{solvent}=w_{\mathrm{2,}RP}{M}^{RP}+w_{\mathrm{2,}EP}{M}^{EP}$ (3)

The mass balance for the insoluble (3) is given by Equation 4.

$w_{\mathrm{3,}M}{M}^M=w_{\mathrm{3,}seed}{M}^{seed}=w_{\mathrm{3,}RP}{M}^{RP}$ (4)

where;

, , : mass of the mixture (seed + solvent), RP and EP, respectively;

The total phenolic concentrations were determined during the extraction phase, thus allowing the extraction yield of these compounds to be calculated. In addition, the antioxidant capacity of the extracts was assayed using the DPPH radical scavenging method.

The transfer of oil ($T_{oil}\left(\%\right)$), and total phenolic compounds ($T_{TPC}\left(\%\right)$) in the extraction process was calculated using Equation 5 (Costa Rodrigues & Oliveira, 2010).

$T_i\left(\%\right)=100\left(\frac{w_{i,EP}{M}^{EP}}{w_{i,seed}{M}^{seed}}\right)$ (5)

where:

M^seed: mass of the seed used in extraction.

$w_{i,seed}$: mass fraction of component i in the seed before the extraction process. Where i represents oil or total phenolic compounds.

$w_{i,EP}$: mass fraction of component i in the seed before the extraction process.

The defatted cake obtained after the extraction process was dried and subjected to residual oil determination using a Soxhlet apparatus. This analysis aimed to validate the efficiency of the extraction process and confirm the oil yield calculations.

The retention index (RI), defined as the mass of solution retained in the solid matrix in relation to the mass of the inert solid present (mass of adhered solution/mass of inert solid or fibers), was calculated using the following Equation 6 proposed for Araújo et al. (2019).

$RI=\frac{M_{oil,RP}+M_{solvent,RP}}{M_{inert,RP}}$ (6)

2.3 Analyzes in the extract phase

2.3.1 Total phenolic compounds (TPC)

TPC were determined using the Folin-Ciocalteu reagent method originally proposed by Waterhouse (2002), with adaptations. The extracts obtained from seed characterization (Section 2.1) and from the solid-liquid extraction process (Section 2.2) were diluted according to their phenolic concentration and mixed with 2.5 mL of Folin-Ciocalteu reagent (10%) and 2.0 mL of sodium carbonate solution (20%, w/v). The resulting mixture was vortexed and kept in the dark at room temperature for 2 h. A calibration curve was prepared using aqueous gallic acid solutions, and the absorbance was measured at 750 nm using an Ultraviolet-Visible (UV-Vis) spectrophotometer. Results for seeds were expressed as grams of gallic acid equivalents (GAE) per 100 g of dry seed, while for the solid-liquid extraction experiments they were expressed as extraction yield (%) of total phenolics (%TPC).

2.3.2 DPPH• scavenging capacity

The DPPH• (2,2-diphenyl-1-picrylhydrazyl) scavenging capacity was determined following the method initially proposed by Blois (1958) with modifications by Petropoulos et al. (2018). For the determination of antioxidant capacity, 0.1 mL of each extract (obtained in Sections 2.1.3 and 2.2) was added to 3.9 mL of DPPH solution. For the control, 0.1 mL of methanol was added to the DPPH solution. Absorbance was measured after 30 min at 515 nm using an UV-Vis spectrophotometer, and the results were expressed as the percentage of free radical scavenging (%AA).

2.4 Statistical analysis

A randomized complete block design (RCB) was used with three factors (solvent, SSR, and temperature), and all treatments were performed in triplicate. Seeds were treated as blocks. Analysis of Variance (ANOVA) was conducted (F-test, p < 0.05) to evaluate factor effects and interactions. Significant differences between means were identified using Tukey’s test at a 5% significance level. Analyzes were performed using RStudio (v. 4.4.1).

3 Results and discussion

3.1 Solid-liquid extraction using different pure solvents

The results of oil extraction yield ($Toil\left(\%\right)$) from marolo and soursop seeds using pure solvents are presented in Table 1.

Table 1
Oil extraction yields ($Toil\left(\%\right)$) in marolo and soursop seeds.

	SSR	T (ºC)	$Toil\left(\%\right)$
			Ethanol	Hexane	Isopropanol
Marolo	5:1	35	56.28 ± 3.50	94.15 ± 2.02	72.56 ± 2.43
		45	71.25 ± 1.19	99.01 ± 1.35	84.24 ± 2.06
		55	75.80 ± 1.70	99.55 ± 0.95	90.97 ± 2.24
	10:1	35	79.49 ± 1.06	97.25 ± 2.45	90.29 ± 2.95
		45	91.89 ± 2.15	99.61 ± 0.35	92.57 ± 2.11
		55	94.32 ± 2.98	99.23 ± 0.71	96.06 ± 2.45
Soursop	5:1	35	54.66 ± 0.85	85.02 ± 2.42	55.44 ± 1.56
		45	61.95 ± 2.57	89.69 ± 1.25	57.51 ± 2.20
		55	64.18 ± 2.78	90.98 ± 2.88	56.73 ± 2.44
	10:1	35	75.11 ± 1.01	92.66 ± 2.32	65.35 ± 0.99
		45	80.82 ± 1.36	95.35 ± 1.99	65.88 ± 1.75
		55	82.84 ± 1.09	96.71 ± 2.52	66.17 ± 2.83

• SSR: Solvent-to-solid mass ratio.

The analysis of variance for the response variable ($Toil\left(\%\right)$) revealed a significant interaction effect between the solvent-solid mass ratio and the solvent (SSR × S). The linear effects for temperature (T), solvent (S), and SSR were also significant (p < 0.05).

Figure 2 shows the linear relationship between temperature and the oil yield obtained from marolo and soursop seeds. An increase in temperature is associated with higher extraction yields ($Toil\left(\%\right)$). This behavior can be attributed to the improved solubility of oil in the solvent and the reduced viscosity of the extraction medium at elevated temperatures, which collectively enhance mass transfer efficiency (Amarante et al., 2014). At lower temperatures, ethanol has lower solubility for oil, which can be advantageous for separating the oil from the solvent after extraction.

Figure 2
Effect of temperature on the extraction of soluble solids from marolo and soursop seeds. The data presented in this figure represent the mean values at each temperature level.

The impact of the SSR and the solvent type (SSR × S) on the oil extraction yield is shown in Table 2. An increase in the SSR (from 5:1 to 10:1) resulted in higher$Toil\left(\%\right)$, except in extractions using hexane, where no significant difference was observed between the two ratios. The lack of a significant effect of the SSR on yield when using hexane can be attributed to the high solubility of oil in this solvent. This property facilitates the efficient extraction of oil, even when the lower mass ratio is employed.

Table 2
Interaction effect between the solvent-solid mass ratio and the solvent (SSR × S) on oil extraction yield ($Toil\left(\%\right)$).

Solvents	SSR
	5:1	10:1
Ethanol	64.02 ± 8.28b,B	84.08 ± 7.48a,B
Hexane	93.07 ± 5.64a,A	96.80 ± 2.58a,A
Isopropanol	69.57 ± 15.44b,B	79.39 ± 14.99a,B
Acetone	72.48 ± 11.94b,B	83.29 ± 7.77a,B

• For each row, the same lowercase letters indicate that the mean differences are not significant at p < 0.05 by Tukey’s test. For each column, the same uppercase letters indicate that the mean differences are not significant at p < 0.05 by Tukey’s test.

The SSR is a critical parameter in extraction processes. As the dilution level increases (i.e., higher SSR), mass transfer is enhanced, along with a reduction in solution viscosity (Ntalikwa, 2021; Seth et al., 2007). This also results in more diluted extracts.

Hexane demonstrated superior efficiency in extracting oil from marolo and soursop seeds compared to other solvents. Its low dynamic viscosity (0.3 mPa s) and dielectric constant (1.88) allow it to penetrate the matrix effectively, extracting the oil more efficiently. In contrast, ethanol has a higher dynamic viscosity (1.074 mPa·s) and dielectric constant (24.60), which reduces its oil extraction efficiency (Ntalikwa, 2021).

Table 3 presents the influence of different solvents on oil extraction yield for each seed. The results show that extraction performance is strongly dependent on both solvent type and seed matrix.

Table 3
Influence of different solvents on the extraction yield of oil ($Toil\left(\%\right)$) for each seed.

Solvents	Seeds
	Marolo	Soursop
Ethanol	78.17 ± 14.02a,C	69.93 ± 11.33b,B
Hexane	98.13 ± 2.13a,A	91.73 ± 4.21b,A
Isopropanol	87.78 ± 8.39a,B	61.18 ± 5.11b,C
Acetone	85.47 ± 8.18a,BC	70.31 ± 8.26b,B

• For each row, the same lowercase letters indicate that the mean differences are not significant at p < 0.05 by Tukey’s test. For each column, the same uppercase letters indicate that the mean differences are not significant at p < 0.05 by Tukey’s test.

Isopropanol showed significantly higher oil yields than ethanol for marolo seeds, whereas the opposite trend was observed for soursop seeds, with ethanol resulting in higher yields (Table 3). Acetone exhibited intermediate performance and did not differ significantly from ethanol for both seed. These differences indicate that extraction performance depends on seed matrix composition, which influences solvent-solute interactions. Consequently, predicting the solubility of these solutes in various solvents is generally challenging (Danlami et al., 2015), leading to different results for each of the analyzed seeds.

In the studies conducted by Araújo et al. (2019), which evaluated the extraction of oil in pequi (Caryocar brasiliense Cambess.) and murici (Byrsonima crassifolia (L.) Kunth) seeds, it was found that higher yields were obtained when hexane was used as the solvent. For murici seeds, no significant differences were observed in the yields obtained using isopropanol, ethanol, and acetone. Nonetheless, for the case of pequi seeds, superior results were obtained when isopropanol was utilized as opposed to ethanol and acetone. Aquino et al. (2011) extracted oil from dried pequi pulp using a batch process in an incubator and constant temperature (50 °C) for 16 hours and an SSR ratio of 10:1 (w/w) with pure solvents. The extraction yields obtained for acetone, hexane, and ethanol were 61.07%, 60.17%, and 39.78%, respectively.

In their study, Oliveira et al. (2019) examined the effect of solvent type (ethanol, acetone, and ethyl acetate) on oil extraction from coffee beans and defatted coffee press cake. Experiments were conducted at temperatures ranging from 35 to 55 °C. Within this temperature range, acetone resulted in the highest yields of extracted compounds for both coffee beans and press cake, followed by ethyl acetate.

The RI is a pivotal parameter that quantifies the amount of solution adhering to the solid phase, otherwise known as the raffinate phase (inert + adhered solution). This parameter plays a key role in the design of the extraction process, particularly concerning the number of extraction stages and the solvent recovery step. As the amount of adhered solution increases, the overall operating costs also rise (Costa Rodrigues & Oliveira, 2010). The extent of solution adhesion to the inert is contingent on numerous factors, including the solution's viscosity, particle size, and the physical and chemical affinities between the solution and the oleaginous matrix (Toda et al., 2023). The RI for soluble oil extraction from both seeds is shown in Table 4.

Table 4
Retention index (kg adhered solution/kg inert) for the extraction of marolo and soursop seeds.

	SSR	T (°C)	Ethanol	Hexane	Isopropanol
Marolo	5:1	35	1.82 ± 0.01	1.17 ± 0.01	1.69 ± 0.01
		45	1.78 ± 0.01	1.14 ± 0.01	1.64 ± 0.01
		55	1.69 ± 0.02	1.15 ± 0.01	1.64 ± 0.05
	10:1	35	1.67 ± 0.04	1.09 ± 0.01	1.46 ± 0.02
		45	1.68 ± 0.01	1.07 ± 0.01	1.47 ± 0.03
		55	1.68 ± 0.02	1.06 ± 0.02	1.48 ± 0.04
Soursop	5:1	35	2.01 ± 0.01	1.10 ± 0.03	2.10 ± 0.03
		45	1.81 ± 0.01	1.06 ± 0.03	2.02 ± 0.03
		55	1.50 ± 0.01	0.99 ± 0.02	1.48 ± 0.01
	10:1	35	1.21 ± 0.03	1.02 ± 0.04	1.33 ± 0.03
		45	1.17 ± 0.03	0.92 ± 0.03	1.33 ± 0.03
		55	1.01 ± 0.03	0.89 ± 0.04	1.31 ± 0.04

In evaluating the RI, only the linear effects of the S and the SSR were found to be significant (p < 0.05). The impact of solvent selection on RI is illustrated in Figure 3.

Figure 3
Effect of solvents on the retention index of marolo and soursop seeds during the extraction.

The solvents isopropanol, ethanol, and acetone did not show significant differences in RI. The lowest values were observed when hexane was used. The low RI of hexane can be attributed to its low viscosity and polarity (Abraham et al., 1988). The elevated RI values exhibited by ethanol, isopropanol, and acetone are attributable to their pronounced polar interaction with the surface of the solid matrix (Terigar et al., 2011). In their study, Wlśniak et al. (1987) observed that hexane was associated with lower solvent retention (holdup) compared to isopropanol in the context of jojoba meal.

Costa Rodrigues & Oliveira (2010) extracted rice bran oil using an ethanol-water mixture as solvent with a solid-liquid ratio of 1:3. Their study found RI values ranging from 0.77 (0.04% water in the solvent at 50 °C) to 1.12 (20% water in the solvent at 50 °C), demonstrating that solvent polarity influences RI. Toda et al. (2023) investigated the effect of absolute and hydrated ethanol and isopropanol (6% and 12% hydration, respectively) on the retention index values within the temperature range of 60 to 90 °C. The study revealed that, except for isopropanol (12% water), no discernible temperature-related effects on the RI values were observed. However, an inverse relationship was noted, with RI decreasing as temperature increased in the RI values for isopropanol (12% water).

In this study, the RI decreased as the SSR increased. The average values dropped from 1.549 kg of adhered solution per kg of inert to 1.279 kg of adhered solution per kg of inert when the SSR shifted from 5:1 to 10:1. This reduction can be explained by the lower retention of solution at higher solvent concentrations, as increased solvent volume reduces solution viscosity and improves the efficiency of solvent recovery.

The extraction of total phenolic content yield ($T_{TPC}$) of marolo seeds varied widely, ranging from 0% (using hexane, SSR 5:1, 35 °C) to 99.25% (with ethanol, SSR 5:1, 45 °C) (Figure 4). For soursop seeds, the $T_{TPC}$ ranged from 11.37% (hexane, SSR 10:1, 35 °C) to 107.63% (ethanol, SSR 10:1, 45 °C) (Figure 4).

Figure 4
Total phenolic compounds in extracts of soursop and marolo seeds using different solvents, percentage value.

The results showed that the linear effects of T, SSR, S, and the interaction between temperature and solvent (T × S) exerted a significant influence on the extraction of total phenolic compounds (p < 0.05). As illustrated in Table 5, the interaction effects of temperature and solvent (T × S) on the extraction of total phenolic compounds ($T_{TPC}$) are demonstrated. The extraction efficiency was found to be highest for ethanol, while hexane resulted in the lowest efficiency. Furthermore, acetone extraction yielded a greater quantity of phenolic compounds in comparison with isopropanol. These trends were consistent across both seeds.

Table 5
Extraction yield of total phenolic compounds ($T_{TPC}$) for different temperature and solvent (T × S) interactions, results expressed in percentage.

Solvents	Temperature (°C)
	35 °C	45 °C	55 °C
Ethanol	90.51bA	102.13aA	90.46bA
Hexane	6.22aD	11.68aD	10.18aC
Isopropanol	53.74bC	73.49aC	75.57aB
Acetone	72.72bB	85.56aB	84.61aA

• For each row, the same lowercase letters indicate no significant difference in the means (p < 0.05, Tukey’s test). For each column, the same uppercase letters indicate no significant difference in the means (p < 0.05, Tukey’s test).

The higher $T_{TPC}$obtained with polar solvents can be attributed to the polar nature of phenolic compounds. As reported by Oliveira et al. (2019), the impact of various solvents (acetone, ethanol, ethyl acetate, hexane, isopropanol, and petroleum ether) on total phenolic content yield from green coffee beans and press cake was evaluated. The extraction process was carried out using the Soxhlet method, with two distinct extraction times (3 hours and 5 hours). The results indicated that the phenolic compound yield from green coffee bean extract was higher than that from the press cake extract for all solvents except ethyl acetate. The highest phenolic yields, in decreasing order, were obtained with ethanol, isopropanol, acetone, ethyl acetate, hexane, and petroleum ether. The prolonged extraction period did not increase yield, likely due to compound degradation.

Extractions performed at 45 °C using ethanol yielded the highest total phenolic content at 35 °C and 55 °C. An increase in temperature to 55 °C resulted in a decrease in $T_{TPC}$, likely due to the thermal degradation of polyphenols, particularly those with lower heat resistance (Volf et al., 2014). Zuorro (2015) reported a similar trend. The study found that the extraction of polyphenols from coffee grounds peaked at a yield of 97.8% at a temperature of 45-50 °C. However, there was a sharp decline in yield at higher temperatures.

For acetone and isopropanol, increasing the temperature from 35 °C to 45 °C significantly increased, but no differences were observed between 45 °C and 55 °C. Hexane extractions showed no temperature-dependent variation in $T_{TPC}$.

Mesquita et al. (2021) extracted compounds from soursop seeds using supercritical fluid and Soxhlet methods, with hexane and ethanol as solvents for 6 h. Soxhlet extraction yielded 15,600 and 21,900 mg GAE/kg dry matter (db) for hexane and ethanol, respectively.

Changing the SSR from 5:1 to 10:1 improved total phenolic content (from 61.51% to 64.64%). The effect of SSR on $T_{TPC}$ may vary; for instance, Anastásio & Carvalho (2013) studied the influence of temperature, extraction time, and SSR on phenolic extraction from sweet potato using water as the solvent. They found that only SSR significantly affected phenolic extraction, with no significant influence observed for temperature or extraction time, whereas Nepote et al. (2005) reported no significant effect of SSR on phenolic extraction from peanut skins with ethanol.

Figure 5 illustrates the antioxidant capacity of marolo and soursop seed extracts obtained with different solvents, as measured by free radical scavenging activity. As shown, marolo seed extracts displayed superior antioxidant activity compared to soursop extracts when using ethanol, isopropanol, and acetone. This difference is consistent with findings by Menezes et al. (2019), who reported higher antioxidant activity in marolo seeds.

Figure 5
Antioxidant activity of soursop and marolo seeds extracts using different solvents, percentage value.

The solvent had a significant linear effect on antioxidant activity. Ethanol extracts exhibited higher activity than isopropanol and hexane extracts, and similar activity to acetone extracts (Figure 6). Temperature and SSR had no significant effect on antioxidant activity.

Figure 6
Effect of solvents on the antioxidant activity of marolo and soursop seed extracts.

4 Conclusion

The present study demonstrated that solvent type, extraction temperature, and solvent-to-solid mass ratio (SSR) are critical parameters that influence the efficiency of oil and phenolic compound recovery from marolo and soursop seeds. Hexane showed the highest oil extraction yields and the lowest retention indices; however, its performance in extracting phenolic compounds was markedly inferior to that of green solvents. Among the alternative solvents evaluated, ethanol was identified as the most effective for phenolic compound recovery and antioxidant activity, particularly at 45 °C. While ethanol and isopropanol yielded lower oil yields compared to hexane, their advantages in terms of safety, environmental sustainability, and bioactive compound recovery support their potential as promising alternatives. These findings support the feasibility of using green solvents in the valorization of native fruit by-products, contributing to more environmentally responsible processes for the food, pharmaceutical, and cosmetic industries.

Acknowledgements

The authors appreciate the support from the Federal University of Lavras (UFLA), CAPES and CNPq for the scholarships. This work was supported by the FAPEMIG under grant CAG APQ-01689/13.

Data Availability Statement

All data generated or analyzed in this study are included in this published article.

• Cite as:
  Menezes, E. G. T., & Queiroz, F. (2026). Green solvent extraction of oil and phenolic compounds from marolo and soursop seeds: Comparative study. Brazilian Journal of Food Technology, 29, e2025133. https://doi.org/10.1590/1981-6723.1332025
• Funding:
  Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG) [CAG APQ 01689/13].

References

• Abraham, G., Hron-Sr, R. J., & Koltun, S. P. (1988). Modeling the solvent extraction of oilseeds. Journal of the American Oil Chemists’ Society, 65(1), 129-135. https://doi.org/10.1007/BF02542564
  » https://doi.org/10.1007/BF02542564
• Almeida, R. F., Moreno, I. F., Machado, A. P. O., Meireles, M. A. A., da Silva, L. K. F., & Batista, E. A. C. (2024). Araticum (Annona crassiflora Mart.): A critical review for the food industry. Food Research International, 184, 114241. PMid:38609221. https://doi.org/10.1016/j.foodres.2024.114241
  » https://doi.org/10.1016/j.foodres.2024.114241
• Amarante, R. C. A., Oliveira, P. M., Schwantes, F. K., & Morón-Villarreyes, J. A. (2014). Oil extraction from castor cake using ethanol: Kinetics and thermodynamics. Industrial & Engineering Chemistry Research, 53(16), 6824-6829. https://doi.org/10.1021/ie500508n
  » https://doi.org/10.1021/ie500508n
• Anastásio, A., & Carvalho, I. S. (2013). Phenolics extraction from sweet potato peels: Key factors screening through a Plackett–Burman design. Industrial Crops and Products, 43, 99-105. https://doi.org/10.1016/j.indcrop.2012.07.011
  » https://doi.org/10.1016/j.indcrop.2012.07.011
• Aquino, L. P., Borges, S. V., Queiroz, F., Antoniassi, R., & Cirillo, M. A. (2011). Extracción de aceite del fruto pequi (Carroçar brasiliense Camb.) utilizando diversos disolventes y sus mezclas. Grasas y Aceites, 62, 245-252. https://doi.org/10.3989/gya.091010
  » https://doi.org/10.3989/gya.091010
• Araújo, A., Oliveira, É., Carvalho, G., Menezes, E., & Queiroz, F. (2019). Modeling of cosolvent and temperature effects on ethanolic extraction of pequi and murici seeds. Emirates Journal of Food and Agriculture, 31, 153-163. https://doi.org/10.9755/ejfa.2019.v31.i3.1927
  » https://doi.org/10.9755/ejfa.2019.v31.i3.1927
• Arruda, H. S., & Pastore, G. M. (2019). Araticum (Annona crassiflora Mart.) as a source of nutrients and bioactive compounds for food and non-food purposes: A comprehensive review. Food Research International, 123, 450-480. PMid:31284996. https://doi.org/10.1016/j.foodres.2019.05.011
  » https://doi.org/10.1016/j.foodres.2019.05.011
• Association of Official Analytical Chemists – AOAC. (2016). Official methods of analysis of the AOAC International (20th ed.). Gaithersburg: AOAC.
• Baker, E. C., & Sullivan, D. A. (1983). Development of a pilot-plant process for the extraction of soy flakes with aqueous isopropyl alcohol. Journal of the American Oil Chemists’ Society, 60(7), 1271-1277. https://doi.org/10.1007/BF02702099
  » https://doi.org/10.1007/BF02702099
• Beckel, A. C., Belter, P. A., & Smith, A. K. (1948). Solvent effects on the products of soybean oil extraction. Journal of the American Oil Chemists’ Society, 25(1), 7-9. https://doi.org/10.1007/BF02553630
  » https://doi.org/10.1007/BF02553630
• Blois, M. S. (1958). Antioxidant determinations by the use of a stable free radical. Nature, 181(4617), 1199-1200. https://doi.org/10.1038/1811199a0
  » https://doi.org/10.1038/1811199a0
• Costa Rodrigues, C. E., & Oliveira, R. (2010). Response surface methodology applied to the analysis of rice bran oil extraction process with ethanol. International Journal of Food Science & Technology, 45(4), 813-820. https://doi.org/10.1111/j.1365-2621.2010.02202.x
  » https://doi.org/10.1111/j.1365-2621.2010.02202.x
• Danlami, J. M., Arsad, A., & Zaini, M. A. A. (2015). Characterization and process optimization of castor oil (Ricinus communis L.) extracted by the Soxhlet method using polar and non-polar solvents. Journal of the Taiwan Institute of Chemical Engineers, 47, 99-104. https://doi.org/10.1016/j.jtice.2014.10.012
  » https://doi.org/10.1016/j.jtice.2014.10.012
• Ferreira-Dias, S., Valente, D. G., & Abreu, J. M. F. (2003). Comparison between ethanol and hexane for oil extraction from Quercus suber L. fruits. Grasas y Aceites, 54(4), 378-383. https://doi.org/10.3989/gya.2003.v54.i4.225
  » https://doi.org/10.3989/gya.2003.v54.i4.225
• Gandhi, A. P., Joshi, K. C., Jha, K., Parihar, V. S., Srivastav, D. C., Raghunadh, P., Kawalkar, J., Jain, S. K., & Tripathi, R. N. (2003). Studies on alternative solvents for the extraction of oil-I soybean. International Journal of Food Science & Technology, 38(3), 369-375. https://doi.org/10.1046/j.1365-2621.2003.00683.x
  » https://doi.org/10.1046/j.1365-2621.2003.00683.x
• Luzia, D. M. M., & Jorge, N. (2013). Bioactive substance contents and antioxidant capacity of the lipid fraction of Annona crassiflora Mart. seeds. Industrial Crops and Products, 42, 231-235. https://doi.org/10.1016/j.indcrop.2012.05.027
  » https://doi.org/10.1016/j.indcrop.2012.05.027
• Luzia, D. M. M., & Jorge, N. (2014). Study of antioxidant activity of non-conventional Brazilian fruits. Journal of Food Science and Technology, 51(6), 1167-1172. PMid:24876651. https://doi.org/10.1007/s13197-011-0603-x
  » https://doi.org/10.1007/s13197-011-0603-x
• Menezes, E. G. T., Oliveira, E. R., Carvalho, G. R., Guimarães, I. C., & Queiroz, F. (2019). Assessment of chemical, nutritional and bioactive properties of Annona crassiflora and Annona muricata wastes. Food Science and Technology, 2019(1, Suppl.2), 662-672. https://doi.org/10.1590/fst.22918
  » https://doi.org/10.1590/fst.22918
• Mesquita, P. C., Rodrigues, L. G. G., Mazzutti, S., Silva, M., Vitali, L., & Lanza, M. (2021). Intensified green-based extraction process as a circular economy approach to recover bioactive compounds from soursop seeds (Annona muricata L.). Food Chemistry: X, 12, 100164. PMid:35024607. https://doi.org/10.1016/j.fochx.2021.100164
  » https://doi.org/10.1016/j.fochx.2021.100164
• Nepote, V., Grosso, N. R., & Guzmán, C. A. (2005). Optimization of extraction of phenolic antioxidants from peanut skins. Journal of the Science of Food and Agriculture, 85(1), 33-38. https://doi.org/10.1002/jsfa.1933
  » https://doi.org/10.1002/jsfa.1933
• Ntalikwa, J. W. (2021). Solvent extraction of jatropha oil for biodiesel production: Effects of solvent-to-solid ratio, particle size, type of solvent, extraction time, and temperature on oil yield. Journal of Renewable Energy, 2021, 1-8. https://doi.org/10.1155/2021/9221168
  » https://doi.org/10.1155/2021/9221168
• Oliveira, C. M., Garavazo, B. R., & Rodrigues, C. E. C. (2012). Liquid–liquid equilibria for systems composed of rice bran oil and alcohol-rich solvents: Application to extraction and deacidification of oil. Journal of Food Engineering, 110(3), 418-427. https://doi.org/10.1016/j.jfoodeng.2011.12.027
  » https://doi.org/10.1016/j.jfoodeng.2011.12.027
• Oliveira, E. R., Carvalho, G. R., Cirillo, M. A., & Queiroz, F. (2019). Effect of ecofriendly bio-based solvents on oil extraction from green coffee bean and its industrial press cake. Brazilian Journal of Chemical Engineering, 36(4), 1307-1318. https://doi.org/10.1590/0104-6632.20190364s20190102
  » https://doi.org/10.1590/0104-6632.20190364s20190102
• Petropoulos, S. A., Fernandes, Â., Barros, L., & Ferreira, I. C. (2018). A comparison of the phenolic profile and antioxidant activity of different Cichorium spinosum L. ecotypes. Journal of the Science of Food and Agriculture, 98(1), 183-189. PMid:28573740. https://doi.org/10.1002/jsfa.8453
  » https://doi.org/10.1002/jsfa.8453
• Rittner, H. (1992). Extraction of vegetable oils with ethyl alcohol. Oléagineux, 47, 39-42.
• Roesler, R., Malta, L. G., Carrasco, L. C., & Pastore, G. (2006). Evaluation of the antioxidant properties of the Brazilian Cerrado fruit Annona crassiflora (araticum). Journal of Food Science, 71(2), C102-C107. https://doi.org/10.1111/j.1365-2621.2006.tb08882.x
  » https://doi.org/10.1111/j.1365-2621.2006.tb08882.x
• Sawada, M. M., Venâncio, L. L., Toda, T. A., & Rodrigues, C. E. C. (2014). Effects of different alcoholic extraction conditions on soybean oil yield, fatty acid composition and protein solubility of defatted meal. Food Research International, 62, 662-670. https://doi.org/10.1016/j.foodres.2014.04.039
  » https://doi.org/10.1016/j.foodres.2014.04.039
• Seth, S., Agrawal, Y. C., Ghosh, P. K., Jayas, D. S., & Singh, B. P. N. (2007). Oil extraction rates of soybean using isopropyl alcohol as solvent. Biosystems Engineering, 97(2), 209-217. https://doi.org/10.1016/j.biosystemseng.2007.03.008
  » https://doi.org/10.1016/j.biosystemseng.2007.03.008
• Terigar, B. G., Balasubramanian, S., Sabliov, C. M., Lima, M., & Boldor, D. (2011). Soybean and rice bran oil extraction in a continuous microwave system: From laboratory to pilot-scale. Journal of Food Engineering, 104(2), 208-217. https://doi.org/10.1016/j.jfoodeng.2010.12.012
  » https://doi.org/10.1016/j.jfoodeng.2010.12.012
• Tir, R., Dutta, P. C., & Badjah-Hadj-Ahmed, A. Y. (2012). Effect of the extraction solvent polarity on the sesame seeds oil composition. European Journal of Lipid Science and Technology, 114(12), 1427-1438. https://doi.org/10.1002/ejlt.201200129
  » https://doi.org/10.1002/ejlt.201200129
• Toda, T. A., Barreiro, M. M., Cunha, G. B., & Rodrigues, C. E. C. (2023). Performance of alcoholic solvents in the continuous countercurrent spent coffee grounds oil extraction. Journal of Industrial and Engineering Chemistry, 118, 268-279. https://doi.org/10.1016/j.jiec.2022.11.012
  » https://doi.org/10.1016/j.jiec.2022.11.012
• Villacís-Chiriboga, J., Prandi, B., Ruales, J., Van Camp, J., Sforza, S., & Elst, K. (2023). Valorization of soursop (Annona muricata) seeds as alternative oil and protein source using novel de-oiling and protein extraction techniques. Lebensmittel-Wissenschaft + Technologie, 182, 114777. https://doi.org/10.1016/j.lwt.2023.114777
  » https://doi.org/10.1016/j.lwt.2023.114777
• Volf, I., Ignat, I., Neamtu, M., & Popa, V. (2014). Thermal stability, antioxidant activity, and photo-oxidation of natural polyphenols. Chemical Papers, 68(1), 121-129. https://doi.org/10.2478/s11696-013-0417-6
  » https://doi.org/10.2478/s11696-013-0417-6
• Waterhouse, A. L. (2002). Wine phenolics. Annals of the New York Academy of Sciences, 957(1), 21-36. PMid:12074959. https://doi.org/10.1111/j.1749-6632.2002.tb02903.x
  » https://doi.org/10.1111/j.1749-6632.2002.tb02903.x
• Wlśniak, J., Hillel, J., & Katz, O. (1987). Holdup and extraction characteristics of jojoba meal. Journal of the American Oil Chemists’ Society, 64(9), 1352-1354. https://doi.org/10.1007/BF02540796
  » https://doi.org/10.1007/BF02540796
• Zhuang, X., Zhang, Z., Wang, Y., & Li, Y. (2018). The effect of alternative solvents to n-hexane on the green extraction of Litsea cubeba kernel oils as new oil sources. Industrial Crops and Products, 126, 340-346. https://doi.org/10.1016/j.indcrop.2018.10.004
  » https://doi.org/10.1016/j.indcrop.2018.10.004
• Zuorro, A. (2015). Optimization of polyphenol recovery from espresso coffee residues using factorial design and response surface methodology. Separation and Purification Technology, 152, 64-69. https://doi.org/10.1016/j.seppur.2015.08.016
  » https://doi.org/10.1016/j.seppur.2015.08.016

Edited by

• Associate Editor:
  Juliano Lemos Bicas.

Publication Dates

• Publication in this collection
  20 July 2026
• Date of issue
  2026

History

• Received
  13 Nov 2025
• Accepted
  24 Apr 2026