Introduction

ASX (Astaxanthin) with the molecular formula C40H52O4 and an orange-red color is classified as a xanthophyll and has the highest concentration among the oxygenated carotenoid derivatives1,2. ASX, with its molecular structure containing carbon-carbon chains with conjugated bonds3, possesses remarkable properties, which include the ability to remove reactive oxygen compounds, inhibit free radicals, and perform specific biological and physiological functions, such as anti-aging, antioxidant, antidiabetic, anti-inflammatory, and cognitive enhancement4,5. However, since humans and animals lack the biochemical pathways to produce carotenoids, including ASX, they must be obtained from external sources6. Due to its beneficial properties, ASX is used in numerous industries. It is widely used in the food, pharmaceutical, cosmetic and healthcare industries, where it is found in various forms such as capsules, gels, tablets, powders, creams, energy drinks, oils and extracts7,8. In addition, ASX is a widely used food additive in aquaculture, enhancing the attractive body color of aquatic animals such as salmon, shrimp and crustaceans while promoting their growth and survival9. ASX can be obtained from two main sources: natural sources such as algae, fungi, salmon, krill, crustaceans, shrimp and lobster, and commercial sources such as Fafia yeast, Haematococcus and chemical synthesis1,10.

There is a growing trend towards the use of natural ingredients in the food and nutrition market as consumers are increasingly concerned about the safety and potential negative effects of chemical products11. During shrimp processing, significant amounts of raw material are discarded as waste, resulting in economic losses12. However, this waste still contains valuable compounds such as ASX and other nutrients10. Fenneropenaeus merguiensis (F. merguiensis), a species of shrimp, is widely distributed in tropical and subtropical waters along the Indo-West Pacific coast13 and is an ideal source for the extraction of ASX as it contains high levels of this carotenoid14. Similarly, P. Maeoticus, a Gammarus species abundant in the Caspian Sea, serves as a valuable carotenoid source in the region15,16. In recent years, various methods have been used to extract ASX, including MAE (Microwave-assisted extraction), UAE (Ultrasound-assisted extraction), supercritical fluids, ionic liquids, Soxhlet, microbial fermentation, and high-pressure methods1,2,3,17,18,19. Each of these methods offers unique advantages and plays a crucial role in the extraction of ASX from various sources. Emerging technologies such as MAE and UAE are considered green techniques due to their ability to reduce Etime (Extraction time), minimize solvent consumption, and increase EE (Extraction efficiency)1,19,20.

ASX can be effectively extracted with green solvents to solve problems related to solvent separation, complete recycling, and environmental threat21. Ionic liquids are a class of environmentally friendly solvents known for their unique physical and chemical properties, such as low vapor pressure, non-flammability, high thermal stability, and renewability22. Recently, a new study focused on the extraction of ASX using ionic liquid- based microemulsions23. Microemulsions, characterized by their stability, low viscosity and high ability to solubilize hydrophilic and lipophilic compounds, offer selective extraction opportunities for biomolecules in the food and chemical industries21.

The aim of this research is the extraction of ASX from F. merguiensis and P. maeoticus using MAE and UAE as pretreatment methods in combination with organic solvents, green solvents and vegetable oil. To optimize the extraction process, Response Surface Methodology (RSM) was used under different MAE and UAE conditions. In addition, the antioxidant activity (AxAc) and other properties of the extracted ASX were analyzed. This study is the first to investigate the use of P. maeoticus Gammarus (PMG) crust, which contains an acceptable number of carotenoids and other valuable substances such as proteins and essential fatty acids, as a source for the extraction of ASX and compared it with the less studied F.merguiensis shrimp (FMS). The combination of new extraction methods with suitable solvents has shown promising potential to improve EE. In addition, two solvents (sunflower oil and ionic liquid) are discussed that are non-toxic, non-volatile and environmentally friendly, and the innovative application of ionic liquids as a green solvent in the form of microemulsions during the extraction process represents a significant advance in the field.

Results

Physical properties of the microemulsion

The density value of the microemulsion was about 0.9715 g/cm3, which was quite similar to that of water. This similarity in density suggests that the microemulsion is stable and well dispersed, making it a promising candidate for various applications. The conductivity of the microemulsion was 312 µSiemens and the particle size was 15.8 nm (Fig. 1).

Fig. 1
figure 1

The profile size of the ionic microemulsion.

Full size image

Determination of total carotenoids (total astaxanthin) by ultrasonic-assisted extraction

The results of UAE for TCC (TAC) from the samples of F.merguiensis shrimp (FMS) and PMG using different solvents with RSM are summarized in Tables 1 and 2. TCC for organic solvents were obtained between 28.476–64.142 and 9.745–33.501, for vegetable oil between 8.682–45.364 and 1.459–7.11 and by ionic microemulsion between 44.904–90.454 and 27.364–76.269 μg/g, respectively. According to the ANOVA in Tables 3 and 4, the linear model for TCC extracted with organic, oily and ionic solvents from FMS and PMG became significant (p < 0.05). Of the factors affecting UAE in the FMS and PMG samples, only the solvent-sample ratio factor was found to be significant at a significant level of p < 0.05 (Tables 3 and 4). In addition, according to the regression coefficients obtained, the effect of the solvent-to-sample ratio was greater than that of other parameters (Tables 5 and 6). The results showed (Fig. 2) that the TCC extracted from FMS and PMG decreased with increasing solvent to sample ratio.

Table 1 The independent factors and combination of independent factors for optimization the ultrasound-assisted extraction of astaxanthin with different solvents from F. merguiensis shrimp wastes
Full size table
Table 2 The independent factors and combination of independent factors for optimization the ultrasound-assisted extraction of astaxanthin with different solvents from P. maeoticus Gammarus
Full size table
Table 3 Analysis of variance for the ultrasound-assisted extraction of astaxanthin with different solvents from F. merguiensis shrimp wastes
Full size table
Table 4 Analysis of variance for the ultrasound-assisted extraction of astaxanthin with different solvents from P. maeoticus Gammarus
Full size table
Table 5 The results of constant, linear regression coefficients of variables for the ultrasound-assisted extraction of astaxanthin with different solvents from F. merguiensis shrimp wastes
Full size table
Table 6 The results of constant, linear regression coefficients of variables for the microwave-assisted extraction of astaxanthin with different solvents from F. merguiensis shrimp wastes
Full size table
Fig. 2: The three-dimensional figure of the effect of different factors of UAE on TCC extracted from the samples of FMS and PMG using different solvents.
figure 2

a Effect of solvent to sample ratio and power on carotenoid extracted from shrimp with ionic microemulsion, b Effect of solvent to sample ratio and power on carotenoid extracted from shrimp with organic solvent), c Effect of solvent to sample ratio and power on carotenoid extracted from shrimp with oil solvent, d Effect of solvent to sample ratio and power on carotenoid extracted from gammaros with ionic microemulsion, e Effect of solvent to sample ratio and power on carotenoid extracted from gammaros with organic solvent, f Effect of solvent to sample ratio and power on carotenoid extracted from gammaros with oil solvent. Red color indicates the highest and blue color indicates the lowest amount of carotenoids.

Full size image

Determination of astaxanthin by ultrasonic-assisted extraction

The ASX concentration for FMS and PMG samples in organic solvents ranged between 21.555–53.444 and 7.777–28.587, for vegetable oil between 7.364–43.628 and 1.269–6.618, and for ionic microemulsions between 41.237–85.475 and 26.555–68.142 mg/ml, respectively (Tables 1 and 2). According to the ANOVA in Tables 3 and 4, the linear model for TCC extracted with organic, oily and ionic solvents from FMS and PMG became significant (p < 0.05). Among the factors affecting the UAE of ASX from FMS and PMG by organic solvents and ionic microemulsion, only the solvent-to-sample ratio factor was significant (p < 0.05), but for the oily solvent, in addition to the solvent-to-sample ratio, the time factor was also significant (p < 0.05) (Tables 3 and 4). Figure 3 shows that the amount of ASX has a decreasing trend with increasing solvent to sample ratio.

Fig. 3: The three-dimensional figure of the effect of different factors of UAE on the ASX content extracted from the samples of FMS and PMG using different solvents.
figure 3

a Effect of solvent to sample ratio and power on astaxanthin content extracted from shrimp with ionic microemulsion), b Effect of solvent to sample ratio and power on astaxanthin content extracted from shrimp with organic solvent), c Effect of solvent to sample ratio and time on astaxanthin content extracted from shrimp with oil solvent),d Effect of solvent to sample ratio and power on astaxanthin content extracted from gammaros with ionic microemulsion), e Effect of solvent to sample ratio and power on astaxanthin content extracted from gammaros with organic solvent), f Effect of solvent to sample ratio and power on astaxanthincontent extracted from gammaros with oil solvent). Red color indicates the highest and blue color indicates the lowest amount of astaxanthin.

Full size image

Determination of astaxanthin recovery by ultrasonic-assisted extraction

The results of ARY with UAE from the samples of FMS and PMG using different solvents with RSM are summarized in Tables 1 and 2. The ARY for organic solvents ranged between 71–85.66 and 71.33–93.66, for vegetable oil between 79–97 and 47.33–92.66, and for ionic microemulsion between 91.33–98.33 and 84.66–96.66%. According to ANOVA, the difference between the solvents used to measure ARY by UAE was significant (p < 0.05) for FMS for the ionic microemulsion and for PMG for the organic model solvent. Among the factors affecting ARY, the effect of solvent to sample ratio and time was significant.

As can be seen in Fig. 4, ARY increased with increasing solvent to sample ratio and time. According to the regression coefficients, the influence of the solvent-to-sample ratio is greater than that of the time factor (Tables 5 and 6).

Fig. 4: The three-dimensional figure of the effect of different factors of UAE on the ASX recovery yield extracted from the samples of FMS and PMG using different solvents.
figure 4

a Effect of solvent to sample ratio and power on astaxanthin recovery yield extracted from gammaros with organic solvent, b Effect of solvent to sample ratio and power on astaxanthin recovery yield extracted from shrimp with ionic microemulsion. Red color indicates the highest and blue color indicates the lowest amount of astaxanthin recovery yield.

Full size image

Determination of total carotenoids (total astaxanthin) by microwave-assisted extraction

The TCC from the samples of FMS and PMG using different solvents are as follows. TCC for organic solvents ranged between 25.634–69.253 and 26.094–12.174, for vegetable oil between 17.602–50.698 and 2.968–10.158 and by ionic microemulsion ranged between 46.872–87.967 and 28.475–89.396 μg/g, respectively (Tables 7 and 8). According to the results, the difference between TCC extracted with MAE by three organic solvents, oily and ionic microemulsion is significant (p < 0.05). Of FMS for oily solvent and ionic microemulsion and of PMG only for the organic model solvent (Tables 9 and 10). In addition, according to the results, only the effect of solvent to sample ratio on TCC by MAE of FMS and PMG is significant (p < 0.05). The results show (Fig. 5) that the TCC extracted from FMS and PMG decreases with the increase of solvent to sample ratio.

Table 7 The independent factors and combination of independent factors for optimization the microwave-assisted extraction of astaxanthin with different solvents from F. merguiensis shrimp wastes
Full size table
Table 8 The independent factors and combination of independent factors for optimization the microwave-assisted extraction of astaxanthin with different solvents from P. maeoticus Gammarus
Full size table
Table 9 Analysis of variance for the microwave-assisted extraction of astaxanthin with different solvents from F. merguiensis shrimp wastes
Full size table
Table 10 Analysis of variance for the microwave-assisted extraction of astaxanthin with different solvents from P. maeoticus Gammarus
Full size table
Fig. 5: The three-dimensional figure of the effect of different factors of MAE on TCC extracted from the samples of FMS and PMG using different solvents.
figure 5

a Effect of solvent to sample ratio and power on carotenoid extracted from shrimp with ionic microemulsion, b Effect of solvent to sample ratio and power on carotenoid extracted from shrimp with organic solvent, c Effect of solvent to sample ratio and power on carotenoid extracted from shrimp with oil solvent. Red color indicates the highest and blue color indicates the lowest amount of carotenoids.

Full size image

Determination of astaxanthin by microwave-assisted extraction

The ASX concentration for FMS and PMG wastes extracted with organic solvents ranged from 20.968–62.802 to 8.66–20.19, for vegetable oil from 16.285–41.968 to 2.44–8.663, and for ionic microemulsions from 44.951–84.348 to 27.856–71.38 mg/ml, respectively (Tables 7 and 8). According to ANOVA, the difference between the solvents used for the MAE measurement of ASX was significant (p < 0.05) for FMS for the ionic microemulsion and vegetable oil and for PMG for the organic model solvent. Of the factors that affected ASX concentration, only the effect of solvent to sample ratio was significant (p < 0.05). The results showed that the concentration of ASX extracted with MAE from FMS and PMG decreased as the solvent-to-sample ratio increased (Fig. 6). In addition, the coefficients of the model showed that the solvent to sample ratio caused the concentration of ASX to decrease steeply (Tables 11 and 12).

Fig. 6: The three-dimensional figure of the effect of different factors of MAE on the ASX content extracted from the samples of FMS and PMG using different solvents.
figure 6

a Effect of solvent to sample ratio and power on astaxanthin content extracted from gammarus with organic solvent, b Effect of solvent to sample ratio and power on astaxanthin content extracted from shrimp with ionic microemulsion, c Effect of solvent to sample ratio and time on astaxanthin content extracted from shrimp with oil solvent). Red color indicates the highest and blue color indicates the lowest amount of astaxanthin.

Full size image
Table 11 The results of constant, linear regression coefficients of variables for the ultrasound-assisted extraction of astaxanthin with different solvents from P. maeoticus Gammarus
Full size table
Table 12 The results of constant, linear regression coefficients of variables for the microwave-assisted extraction of astaxanthin with different solvents from P. maeoticus Gammarus
Full size table

Determination of astaxanthin recovery by microwave-assisted extraction

The results of ARY with MAE from the samples of FMS and PMG using different solvents with RSM are summarized in Tables 7 and 8. The ARY for organic solvents ranged from 62–98.33 to 65–77, for vegetable oil from 73–91.66 to 75–95.33, and for ionic microemulsion from 76.66–96.66 to 66.66–98%. According to ANOVA, the difference between the solvents used to measure ARY by MAE was significant (p < 0.05) for FMS for the ionic microemulsion and for PMG for the organic model solvent. Among the factors affecting ARY, the effect of solvent to sample ratio and time was significant.

As can be seen in Fig. 7, the ARY of PMG decreased with the increase of solvent to sample ratio and time. However, for the FMS samples, ARY decreased with the increase of solvent to sample ratio and ARY increased with the increase of time. According to the regression coefficients, the influence slope of the solvent-to-sample ratio is greater than the time factor (Tables 11 and 12).

Fig. 7: The three-dimensional figure of the effect of different factors of MAE on the ASX recovery yield extracted extracted from the samples of FMS and PMG using different solvents.
figure 7

a Effect of solvent to sample ratio and power on astaxanthin recovery yield extracted from gammaros with organic solvent, b Effect of solvent to sample ratio and power on astaxanthin recovery yield extracted from shrimp with organic solvent). Red color indicates the highest and blue color indicates the lowest amount of astaxanthin recovery yield.

Full size image

Optimization

Considering the great importance of ASX pigment, the aim of this research is to optimize the extraction conditions for ASX from two sources of FMS and PMG with UAE and MAE methods using three organic solvents, oil and ionic microemulsion to maximize the extraction of ASX, TCC and EE. The optimization was performed using the numerical optimization method and Design Expert version 13 software. According to the results of Design Expert software, optimal conditions for extraction of ASX from PMG by MAE using three organic solvents, oil and ionic microemulsion were not found, but optimal conditions for MAE of FMS and UAE from shrimp were defined for FMS and PMG and the results are shown in Figs. 8–10.

Fig. 8
figure 8

The results related to the optimization of FMS extraction conditions using microwave method and organic, oily and ionic microemulsion solvents.

Full size image
Fig. 9
figure 9

The results related to the optimization of UAE of PMG by organic solvents, oil and ionic microemulsion.

Full size image
Fig. 10
figure 10

The results related to the optimization of UAE of FMS by organic, oily and ionic microemulsion solvents.

Full size image

According to the results, the conditions in which the power was 60 W, the time was 30 min and the solvent to sample ratio was five times were selected as optimal extraction conditions for UAE with a desirability of 79.6% for FMS and the conditions in which the power was 180.32 W, the time was 29.99 min and the solvent to sample ratio was five times with a desirability of 65.6% for PMG. For the optimal conditions for MAE of FMS, the power was 100 W, the time was 110 s, and the solvent-to-sample ratio was 5-fold, corresponding to a desirability of 92.4%.

DPPH scavenging results

Figure 11A illustrates that UAE achieves better results. Although BHT has the highest inhibition percentage at the same concentration (94%), UAE outperforms the oil solvent with an inhibition of 83%, surpassing the ionic solvent (71%) and the organic solvent (67%) in efficiency. The results for MAE shown in Fig. 11B are comparatively weaker than for UAE. Nevertheless, the oil solvent shows a remarkable 79% inhibition in this method, outperforming the ionic solvent (65%) and the organic solvent (58%) in inhibitory efficacy.

Fig. 11: DPPH scavenging results.
figure 11

A UAE of ASX with different solvents, B MAE of ASX with different solvents. Blue color indicates butyl hydroxytoluene, red color indicates oily solvent, green color indicates ionic microemulsion, purple color indicates organic solvent.

Full size image

Discussion

Previous research on microemulsions based on ionic liquids was limited, which prompted the current study to investigate their physical properties. In particular, the study focused on investigating the density, conductivity and particle size in these microemulsions. Gao et al. investigated the recovery of natural ASX from crab waste by UAE and the use of microemulsions. They evaluated the efficiency of microemulsions containing tributyloctylphosphonium bromide ([P4448] Br), tributyloctylphosphoniumtrifluoroacetate ([P4448] CF3COO) or tetrabutylphosphatylphosphonium ([P4448] CF3COO) in the extraction of ASX compared to organic solvents (ethanol, acetone and dimethyl sulfoxide)23. The results obtained were similar to our results.

The density of the pure ionic liquids and their microemulsions was almost as high as that of water in the temperature range of 303–323 K. The viscosity of the studied ionic liquids was 20–400 times higher than that of the organic solvents (ethanol, acetone and DMSO) and the conductivity of the studied microemulsions was similar to the density and viscosity, but in the opposite direction. A higher conductivity corresponds to a larger number of dissolved ionic species freely present and moving in the solution. The small particle size indicates that the microemulsion has excellent stability and uniformity. The high electrical conductivity of the microemulsion-ionic liquid in water results in better interaction of the solvent with the sample and higher extraction of the pigment23. To reduce the effects of volatile and toxic solvents, hydrophilic ionic liquids have been used as dispersion solvents and hydrophobic ionic liquids as extraction solvents. An ionic liquid is a salt in a liquid state which is considered one of the green solvents due to its adjustable physicochemical properties, high chemical and thermal stability, and low vapor pressure at room temperature, and has a molecular structure of cations and anions. It is different24. However, the viscosity of most ionic liquids is higher than that of organic solvents, which leads to a decrease in the mass transfer rate. Microemulsion is a promising method that enables selective extraction of biomolecules in the food and chemical industries25.

The results showed that the TCC extracted from FMS and PMG decreased with increasing solvent to sample ratio. The reason for this may be due to the limited solvent capacity of the solvent. These results suggest that using an ionic microemulsion as a solvent in UAE may be a good choice to extract more TCC (TAC) from the samples of FMS and PMG. The ability of the solvent to extract the desired ingredient is primarily influenced by its electrostatic and hydrophilic/hydrophobic interactions with the target molecules23. Therefore, the hydrophobic and electrostatic interactions between the cationic head groups of the microemulsion and ASX and, on the other hand, the interaction of the microemulsion anions with the two hydrophilic ends of ASX through hydrogen bonding can be considered as the reason for the better performance of the microemulsions in the extraction of ASX. Gao et al. (2020) extracted natural ASX from crab waste with UAE and using microemulsions. The effect of sonication power and time and microemulsion composition on the EE (EEAst) and extraction yield (EAst) of ASX was investigated. Compared to organic solvents (ethanol, acetone and dimethyl sulfoxide), the microemulsion containing tributyloctylphosphonium bromide, tributyloctylphosphoniumtrifluoroacetate or tetrabutylphosphatiylphosphonium resulted in a significant increase in ASX extraction due to stronger electrostatic interactions and more hydrogen bonds23.

The findings showed that the amount of ASX has a decreasing trend with increasing solvent to sample ratio. In this regard, Hamdi et al.26, in a study on the extraction of ASX from shrimp waste with a mixture of acetone and hexane in 1:1, 2:1 and 3:1 ratios, found that increasing the amount of acetone decreased the extraction of ASX. They reported that the reason for the decrease in ASX extraction with increasing amount of solvent could be due to the increase in polar solvent, which helps the extraction of other components and thus hinders the extraction of ASX26. On the other hand, with increasing time, the amount of ASX extracted by the oil solvent from both samples increased significantly. In this context, Sharayei et al.27 also reported that increasing the Et time from 5 to 15 min increased the extraction content of ASX27. Şahin and Şamlı28 reported that the amount of effective compounds by UAE occurs in two main steps depending on the time. The first step, called washing, takes place in the first 10–20 min by dissolving the soluble components on the surface of the composition. In this phase, up to 90% of the active ingredients can be extracted. The second step, called slow extraction, is the mass transfer from the surface of the compound to the solvent by the diffusion phenomenon28. Dahmoune et al. also reported that the extraction of bioactive compounds from the leaves of P. Lentiscus is intensified after 10 min29. Using eutectic solvent, Zhang et al.20 indicated that the ASX extracted from the peel was 146 μg/g waste, while from the head it was 218 μg/g waste. Extraction with the organic solvent ethanol showed that the amount of ASX extracted from the shell was 102 μg/g waste and 158 μg/g from the head. The amount of extracted ASX increased with increasing sonication time20. Bi et al. successfully extracted ASX from shrimp waste using ionic liquids and ultrasound. In this study, seven types of uniimidazolium liquids with different cations and anions were investigated. The results showed that an ionic liquid in ethanol solution with a concentration of 0.50 mol/L was the most suitable solvent. By optimizing the UAE conditions, the amount of ASX extracted increased by 98% (92.7 μg/g) compared to the conventional method (46.7 μg/g)22. These results are similar to the findings obtained in the present study.

Fig. 12
figure 12

The standard curve of ASX.

Full size image

The results showed that the ARY increased with increasing solvent to sample ratio and time. The high solvent-to-sample ratio can activate the outward diffusion movement of the intracellular matrix, which ultimately improves the efficiency of pigment extraction. These results are due to the fact that the sample can be dissolved better and more effectively in a higher volume of solvent and the effect of the cavitation phenomenon is more effective. However, a decreasing trend was observed at high solvent ratios due to the limited dissolving capacity. It was determined that the ideal ratio of sample to solvent should be 1:5. This decision was made to avoid unnecessary waste of the solvent. The reason for the increase in ARY with increasing Et time can also be attributed to the fact that the sample is exposed to the solvent for a longer time with increasing Et time and the cavitation phenomenon increases the EE.

Zou et al.30 investigated UAE of ASX from Haematococcuspluvialis. They showed that as the liquid-to-solid ratio increased, ARY also increased. However, when the liquid-to-solid ratio was increased from 20:1 to 30:1, ARY remained almost unchanged. They attributed this to the more effective solubility of the components, which can lead to an increase in EE. However, increasing the solvent to sample ratio led to a waste of solvent. As the ASX efficiency did not change after 20:1, the liquid to solid ratio of 20:1 was used in the next experiments. They also investigated the effect of etime on ARY, which was consistent with our results. Thus, the efficiency of ASX increased from 5 to 15 min and then decreased from 15 to 30 min. In fact, EE increases with time until a balance is established between the target components inside and outside the plant cells. But after equilibrium is established, increasing the time can no longer increase EE30. Hu et al.10 investigated the extraction of ASX from shrimp shells and the effects of different treatments on its concentration. They reported that in the extraction of ASX, the sample to solvent ratio should be 1:7 and a 30 min UAE gives the best results. They also found that decreasing the solid to liquid ratio does not significantly increase the output24.

The results showed that the TCC extracted from FMS and PMG decreases with the increase of solvent to sample ratio. The reason for this may be due to the limited solvent capacity of the solvent. These results are consistent with the results obtained with the amount of ASX extraction.

The results showed that the concentration of ASX extracted with MAE from FMS and PMG decreased as the solvent-to-sample ratio increased. In addition, the coefficients of the model showed that the solvent to sample ratio caused the concentration of ASX to decrease steeply. the ARY of PMG decreased with the increase of solvent to sample ratio and time. However, for the FMS samples, ARY decreased with the increase of solvent to sample ratio and ARY increased with the increase of time. Studies have shown that polar solvents are more suitable for the extraction of ASX. This phenomenon is probably due to the fact that it is more difficult for non-polar solvents to diffuse through the hydrophilic layer surrounding the pigment31. Ruen-ngam et al. also found in the MAE of ASX that the EE increased with increasing time. It was reported that in materials containing some moisture, once the microwave energy is absorbed and subsequently converted to thermal energy, the moisture begins to evaporate. The vaporization causes the pressure inside the cell wall and leads to the disintegration of the cells, which ultimately leads to the release of active substances into the surrounding solvent and increases the EE32. Tomaz et al.33 attempted to optimize the extraction of phenolic compounds from Laurencia obtuse and among the factors, solvent to sample ratio was reported as the most important factor influencing extraction33.

The results showed that for both samples, the ionic microemulsion solvent showed superior performance in ASX extraction and FMS produced more TCC compared to the other sample. For both methods, ARY increased with increasing amount of solvent. In general, according to the results, MAE was more effective than UAE in extracting ASX.

When optimizing the MAE of ASX from Haematococcuspluvialis, Zhao et al.34 showed that microwave power and Et time had a similar quadratic effect on the extraction content of ASX, and the content of ASX extraction first increased with the increase of these two factors and then decreased to a certain extent. Therefore, higher power and longer Et time were not suitable for the extraction of ASX, which was attributed to the increase in temperature at high power, which affects the stability of ASX and may disrupt the structure of ASX34. Zou et al.30 extracted ASX from Haematococcusplevialis with UAE. Extraction solvent, liquid to solid ratio, extraction temperature and extraction time were optimized in this study using RSM. The optimal extraction conditions included ethanol 48.0% in ethyl acetate, a liquid to solid ratio of 20:1 (mL/g), and an extraction time of 16 min at 41.1 °C under 200 W UAE. Under this conditions, the ARY was 27.58 ± 0.40 mg/g30.

Zhao et al.35 optimized the MAE of ASX from Haematococcuspluvialis and investigated AxAc of the extracts. Four independent variables - microwave power (W), E time (s), solvent volume (ml) and extraction number - were optimized. In terms of EE, the optimal MAE conditions were a microwave power of 141 W, an Et time of 83 s, a solvent volume of 9.8 ml and a fourfold extraction number. Under these optimal conditions, ~594 μg ASX was extracted from Haematococcuspluvialis, which exactly matched the predicted content of 592 μg. Moreover, the extracts obtained under these optimal conditions exhibited a strong ability to inhibit the peroxidation of linoleic acid, a strong free radical inhibition against DPPH and also a strong reducing power35. Hooshmand et al. optimized the MAE and UAE of carotenoid pigments from Portunuspelagicus and Penaeussemisulcatus wastes and analyzed the amount of extracted pigments. The factors studied included extraction number, Et time, ultrasonic and microwave power and solvent to waste ratio. For the extraction of carotenoid pigments with UAE from crab waste, extraction was performed three times, with a waste to solvent ratio of 1:15, a power of 90 W and a time of 30 s. The amount of pigment extracted was 1407. Finally, the amount of pigment was 1407.12 ± 0.584 μg/g. Extraction from shrimp waste was performed once with a waste to solvent ratio of 1:10, a power of 150 W and a time of 105 s, and the amount of pigment 1257.43 ± 0.023 μg/g was obtained36.

DPPH is a hydrophobic radical that has a peak absorbance at 517 nm. In the DPPH assay, antioxidant compounds containing hydroxyl groups cause the reduction of DPPH by donating hydrogen atoms to DPPH radicals, resulting in a color shift of the reaction solution from deep purple to light yellow. Consequently, the absorbance decreases at the wavelength of 517 nm37.

The results showed that vegetable oil had the highest AxAc value among the different solvent methods, which was consistent with the results of Kang and Sim38. The explanation for this is that the ASX pigment is fat-soluble and is better preserved in an oily solvent, so it is protected from environmental oxygen and its AxAc is preserved more than in the other two solvents. However, despite extraction more ASX pigments, the ionic liquid solvent has lower AxAc values than the oily solvent due to partial oxidation or possible subsequent damage during the storage period. Kang and Sim treated H. pluvialis cells with common vegetable oils and the ASX oil yield reached >88%38. Dong et al.38 also evaluated four different methods for the extraction of ASX from H. pluvialis. The methods include: Hydrochloric acid pretreatment followed by acetone extraction (HCl-ACE), hexane/isopropanol (6:4, v/v) solvent mixture extraction (HEX-IPA), methanol extraction followed by acetone extraction (MET-ACE, 2-step extraction) and soybean oil extraction, were intensively studied. The results showed that the HCl-ACE method could obtain the highest yield of oil (33.3 ± 1.1%) and the amount of ASX (19.8 ± 1.1%). The DPPH reduction of the extract obtained by the HCl-ACE method was 73.2 ± 0.1%, which was the highest among the four methods. The reducing power of the extract obtained by the four extraction methods was also investigated. It was found that the extraction method using HCl-ACE proposed in this work provided the possibility of efficient extraction of ASX with high AxAc39.

Extracted ASX obtained by both UAE and MAE methods was able to inhibit DPPH radicals. However, due to the sensitivity of ASX pigment to heat and the destructive effect of long-term microwave irradiation on ASX pigment, ASX extracted by the MAE method exhibited a lower AxAc value than ASX extracted by the UAE method. The shorter the Et time and the lower the power of the device, the better the pigments are retained and the higher the AxAc value with MAE. Our results were in line with the results of Roy et al.40. They extracted ASX from shrimp waste using natural, deep eutectic solvents with UAE and investigated its application in bioactive films. The extracted ASX extract clearly showed strong AxAc of ABTSþ and DPPHþ. Sowmya and Sachindra reported that ASX exhibits strong AxAc and even a small amount can show its activity in the final product41.

Methods

Materials

F. merguiensis and P. maeoticus were freshly prepared with species and obtained in winter from a reputable Iranian research institute. Commercially available ASX (>98% purity), α-diphenyl-β-picrylhydrazyl (DPPH) and butylated hydroxytoluene (BHT) were purchased from Sigma-Aldrich (USA). HPLC ethanol, propanol, ethyl acetate, tributyloctylphosphonium bromide, N-butanol and Triton X-100 with high purity were procured from Merck Chemicals Co (Germany). We also purchased high quality, antioxidant-free refined sunflower oil from a reputable Iranian company.

Preparation of the samples

The shells of F. merguiensis and P. maeoticus samples were thoroughly washed with distilled water and then freeze-dried at −50 °C for 48 h (Christ-Alpha 1–4, LD freeze dryer, Germany). After freeze-drying, the powders were sieved through a laboratory sieve with a mesh size <15 µm. The resulting powders were then stored at −18 °C for further use.

Preparation of the solvents

In the experiment, we used a mixture of ethanol and ethyl acetate in a 1:2 ratio as the organic solvent, while refined sunflower oil served as the solvent for the vegetable oil. As an environmentally friendly solvent, we prepared an ionic liquid microemulsion in water using the following method: a mixture of tributyloctylphosphonium bromide, Triton X 100 and n-butanol (in a mass ratio of 3:1:0.62:0.25:0.13) was combined and stirred thoroughly. These compounds constituted the non-polar phase, the surface-active co-surfactant and the polar phase of the microemulsion, respectively23.

Characterization of the microemulsion

The density of the microemulsion was measured with a pycnometer and its conductivity with a conductivity meter (Inolab, WTW, Germany). The particle size of the microemulsions was determined with a zeta sizer (Nano SZ-100, Horiba, Japan) based on dynamic light scattering (DLS).

Experimental design and optimization

RSM was used to determine the extraction conditions and optimize the method to obtain the highest amount of ASX. This method allows the study of the main and interacting variables to determine the optimal extraction conditions. For UAE, three independent variables were considered: the solvent to sample ratio (X1) is in the range of 5–20 times, the sonication power (X2) is in the range of 60–200 W and the sonication time (X3) is in the range of 5–30 min. Similarly, for MAE, the solvent to sample ratio (X1) in the range of 5–20 times, the microwave power (X2) in the range of 100–300 W and the E time (X3) in the range of 5–110 s were considered as independent variables. The selection of these variable ranges was based on the analysis of various studies on the extraction of ASX from different sources. The amount of ASX extracted was measured using a spectrophotometer and was considered as the dependent variable in the experimental design. A centralized composite experimental design was used for the RSM. Each independent variable was set at five levels and coded as values of 0, ±1 and ±α, which increased the precision of the experiment. Design–Expert13 software was used to calculate the extraction conditions for each solvent in the UAE and MAE methods. In addition, six replicates were performed at the central point to estimate the experimental error. Model determination and numerical optimization were performed with Design Expert software. To analyze the dependent variables, the quadratic polynomial model was used in the software.

$$\begin{array}{ll}Y={\beta}_{0}+{\beta}_{1}{\rm{X}}_{1}+{\beta}_{2}{\rm{X}}_{2}+{\beta}_{3}{\rm{X}}_{3}+{\beta}_{12}{\rm{X}}_{1}{\rm{X}}_{2}+{\beta}_{13}{\rm{X}}_{1}{\rm{X}}_{3}\\\quad+{\beta}_{23}{\rm{X}}_{2}{\rm{X}}_{3}+{{\beta}_{1}}^{2}{\rm{X}}_{{1}}^{2}+{{\beta}_{2}}^{2}{\rm{X}}_{{2}}^{2}{{\beta}_{3}}^{2}{\rm{X}}_{{3}}^{2}\end{array}$$

In this equation, Y is the response variable and it is the function or response variable X1, X2 and X3 are the solvent to sample ratio, power and Et time respectively. The values of β1, β2 and β3 are the linear effects of solvent to sample ratio, power and Et time respectively, the values of β12, β22 and β32 are the coefficients of the quadratic effects of solvent to sample ratio, power and Et time and the values of β12, β13 and β23 are the coefficients of the interaction of solvent to sample ratio and extraction power and solvent to sample ratio and Et time and extraction power and time respectively. Analysis of variance (ANOVA) was used to test the significance of the linear, quadratic and reciprocal effects of the coefficients of the regression model for each answer at the 0.05 level.

Ultrasound-assisted extraction

UAE was conducted by mixing the solvents with the powder samples in the proportions mentioned in the previous section. Extraction was performed at 25 °C using an ultrasonic bath (Sono Swiss, SW3H, 280 W) operated at a frequency of 40 kHz and at 41 °C, varying the solvent to sample ratio, power and time. After UAE, the resulting extract was centrifuged at 8500 rpm for 10 min while maintaining the temperature at 25 °C. This process separated the supernatant of the sample from the settled portion. The settled part was then subjected to a new extraction using the same method. Finally, the supernatant obtained was filtered with a 0.45 µm filter to prepare it for analysis30.

Microwave-assisted Extraction

First, the powder samples were mixed with the solvents in the specified ratios at 25 °C and MAE was performed using microwaves (MW domestic multiwave oven, LG, Korea) with a frequency of 2.45 GHz and different values for radiation power and time. After MAE, the obtained extract was centrifuged at 10,000 rpm for 15 min at 25 °C and then the supernatant was separated and filtered with a 0.45 µm filter. When extracted with green solvent or vegetable oil, the resulting extract was diluted with ethanol and then filtered to prepare it for further analysis34,42.

Analysis

The absorbance of the extract samples was measured using a UV–Vis spectrophotometer (UV-1601, Shimadzu, Japan) at a wavelength of 450 nm. It is noteworthy that at this specific wavelength, most of the detected TCC consists of ASX. TCC was calculated using the following equation, which is based on the study by43:

$${\rm{TCC}}=({\rm{absorbance}}\; {\rm{at}}\,450\,{\rm{nm}}\times {\rm{dilution}}\; {\rm{factor}}\times 100)/({\rm{sample}}\; {\rm{weight}}\times {\rm{E}}1 \% )$$

This equation was used to determine the TCC content in the extract samples, which comprises the predominant ASX content among other carotenoids, measured at 450 nm. A specific extinction coefficient (E1%) of 2100 was used to calculate the total ASX content (TAC). It should be noted that TAC accounts for most of the TCC detected at this wavelength. Finally, TCC was calculated using the following equation43:

$${\rm{TCC}}({{\upmu }}{\rm{g}}/{\rm{g}}{\rm{{DW}}})=({\rm{V}}\times {\rm{A}}\times 100)/(21\times {\rm{W}})\times 100$$
(1)

where V is the volume (mL) of the solvent used, A is the absorbance and W is the dry weight (g) of the biomass in the solvent.

To create the standard curve of ASX, various dilutions of pure ASX with concentrations from 0 to 1 μg/ml were prepared. For this purpose, 2-propanol was used as a solvent for the pure ASX pigment. The ASX absorbance of the extracts was measured at 478 nm using a UV–Vis spectrophotometer. The ASX concentration (AC) of the extracts was calculated using the following equation44:

$${\rm{AC}}({\rm{mg}}/{\rm{mL}})=({\rm{absorbance}})/1.97$$
(2)

After analyzing the extract samples, the ARY was calculated according to the following equation45:

$${\rm{ARY}}( \% )=({\rm{extracted}}\; {\rm{ASX}})/({\rm{total}}\; {\rm{ASX}})\times 100$$
(3)

For the DPPH scavenging activity assay, an ethanolic DPPH solution (0.2 mM) was prepared. Then different concentrations of the extracts (from 0.05 to 1 mg/mL) were mixed with 2 mL of ethanolic DPPH solution for 30 min. After the 30-min incubation period, the absorbance of the sample mixture was measured using a UV–vis spectrophotometer at a wavelength of 517 nm. The DPPH scavenging activity was then calculated using the following formula. This test was compared to AxAc of the synthetic antioxidant BHT at similar concentrations46.

$${\rm{Free}}\,{\rm{DPPH}}\,{\rm{radical}}\,{\rm{scavenging}}\,\% = ({\rm{Abs}}\_{\rm{control}}- {\rm{Abs}}\_{\rm{sample}})/{\rm{Abs}}\,({\rm{control}})\times{100}$$
(4)

where Abs_Control and Abs_Sample represent the absorbance of the blank and sample at 517 nm, respectively.

Statistical analysis

The results of the ASX analysis obtained with the spectrophotometer were processed in the Design–Expert13 software to optimize the extraction with RSM. The significance level for the statistical analysis was set at 5%, which means that any p-value < 0.05 is considered statistically significant. Statistical analysis of data from other tests was performed using IBM SPSS Statistics 22 (IBM Corporation, Armonk, NY, USA) based on one-way ANOVA. For all tests conducted, including the ASX analysis and other experiments, each test was performed three times to ensure reliability and consistency of results. Reported results were presented as mean ± standard deviation.