The UAE is applied in a fluid medium composed of the particulate by-product (usually in powder form) dispersed in the solvent. In these fluid systems, the mechanisms of enhanced extraction by ultrasound are mainly related to hydrodynamical phenomena, which are produced by the acoustic microstreaming and acoustic cavitation (13), whose occurrence and effects on the solid-liquid interface are important and were represented in Figure 2.

When ultrasound is applied to the fluid it causes the fluid acceleration in the same direction of the wave propagation (14). This mechanism is called acoustic streaming. When the accelerated fluid rises boundaries (container walls or sample particles), it causes turbulence that increases the agitation of the bulk flow. Therefore, acoustic streaming increases the mass and heat transfer by increasing the convection (15). As the ultrasound waves are propagated in the fluid, the alternating compression (C) and expansion or rarefaction (R) cause the vaporization of the liquid generating gas bubbles, whose size is increased during R cycles and decreased during C cycles (8, 16). When the bubbles reach a specific condition, they implode dissipating a punctual (microscopic) high quantity of energy, increasing the local temperature and pressure. This implosion is called acoustic cavitation, which results in physical shearing effects (micro jetting and pressure shockwaves) (17). According to Kumar et al. (18), acoustic cavitation is the main mechanism involved in UAE.

Considering the particulate solid-solvent system during the UAE process, due to the proximity of the solid surface of particles, the cavitation bubbles often collapse asymmetrically in the surrounding liquid forming a very fast jet (with velocities in the order of 100 m⋅s^–1) (19). This effect leads to interparticle collision, rapid heat and mass transfer at the solid surface as the boundary layer is disrupted. The accelerated interparticle collision and the implosion of cavitation bubbles on the particle surface cause particle surface erosion and fragmentation (20, 21). In addition, during acoustic cavitation, the formation of pores in the cell membranes also could occur, called “sonoporation,” which results in the release of intracellular compounds (18). In fact, pigments in plants are in organelles inside the cells, called vacuoles and plastids. In the case of anthocyanins and betalains are accumulated in the large central vacuole of most plants (5, 22), while the pigments such as chlorophyll and carotenoids are located in chloroplasts and chromoplasts plastids, respectively (23). Therefore, the rupture of these organelles and the cell wall during the UAE means a greater release of pigments.

Is important to mention, that there are other effects produced by ultrasound waves such as the microchannels formation and the “sponge effect” (24). However, these ultrasound effects have a significant impact on solid samples with a certain level of moisture (25–27). Then, considering that UAE is usually applied in a particulate fluid medium, these mechanisms are less important.

Therefore, collapsing cavitation bubbles and the propagation of the sound waves may induce different extraction mechanisms, such as fragmentation, localized erosion, pore formation, shear force, increased absorption and swelling index in the cellular matrix of the sample (18, 28). A detailed description of the extraction mechanisms induced by ultrasound is described by Chemat et al. (12). These mechanisms, in summary, result in a reduction in particle size, release of intracellular pigments, increase in the contact area of the sample with the solvent, the improvement of the solubilization of the compound, and the improvement of the transport of the solvent toward the matrix and of the compound toward the solvent (Figure 2). The increase in the extraction by UAE is attributed to the combined effect of all the mechanisms, which prevalence depends on different factors influencing the process, these factors are detailed below.

During a UAE process, several factors can be varied or controlled depending on the objective, since their levels impact the efficiency of the process and the quality of the pigment obtained. Among the main factors and their parameters are the ambient conditions (temperature, pressure), ultrasound system (actual acoustic power, frequency and amplitude), solvent (physicochemical properties, concentration), sample matrix (composition and structure), Solid/Liquid ratio (solid = by-product sample; liquid = solvent) and time of UAE process, represented in Figure 3.

The UAE process is included in the green extraction methods group. The UAE compared to conventional extraction methods allows the use of a lower amount of solvent, is compatible with the use of green solvents (such as NADES and ILs), increases the extraction yield, improves the quality of the extracted compounds, and reduces the extraction time. These advantages are traduced in lower energy consumption (11, 12, 43, 45, 61–64). However, the energy consumption during a UAE process depends on several factors such as US system (probe or bath device, power, frequency and amplitude level); process application mode (continuous sonoreactor in continuous flow mode or batch reactor); the position of the ultrasound transducers; operation mode (continuous and pulsed operation) and temperature (11, 65). In addition, the energy consumption is influenced not only by the factors mentioned during the UAE processing but also by the use of other techniques associated with the UAE process, for example, the use of enzymes that allows lower energy consumption (66), also by the factors that demand energy after the extraction processing such as during the separation and purification processes. In this regard, for example, the UAE using edible oils as the solvent does not require any subsequent process saving extra power consumption as the pigment-enriched oil can be used directly (67).

Another important aspect that highly influences energy consumption during the UAE process is the plant matrix structure, plasticity and compositional differences (60, 68). For instance, the presence of sensitive compounds requires lower temperature but longer times, while a porous structure requires less time than a compact structure. Consequently, each pigment and matrix, depending on process conditions, shows different extraction behavior demanding a certain amount of energy.

Therefore, energy consumption is specific for each case being necessary for the optimization of UAE processing conditions looking for saving energy.

Carotenoids are lipophilic pigments mainly produced by plants, algae, and some animals such as arthropods and salmonids, they are responsible for the characteristic yellowish, orange, and reddish colors. Several types of carotenoids have been identified and categorized into two main groups: xanthophylls (with oxygen in its chemical formula) and carotenes (without oxygen in their chemical formula), carotenogenesis of carotenes and xanthophylls is controlled by the transcript genes, which are mainly regulated by light and temperature (69, 70). Lycopene (C₄₀H₅₆) is considered the first colored carotenoid in the biosynthesis of many other natural carotenes and xanthophylls (71). From lycopene, α and β carotenes are synthesized, then while α-carotene is converted into lutein, β-carotene is converted into cryptoxanthin, zeaxanthin, antheraxanthin, capsanthin, violaxanthin, and neoxanthin among other (69–71).

Carotenoids are important functional metabolites in the human body because of their antioxidant capacity and for being precursors of vitamin A (72), related to a significant reduction of cancer, cardiovascular diseases, and age-related macular degeneration (73).

Chlorophylls are amphiphilic pigments from photosynthetic organisms, they are distributed in several plants, algae, and cyanobacteria. Two structures of chlorophylls coexist in plants: Chlorophyll-a, which has a methyl (-CH) group at the 7-carbon position and chlorophyll-b, which has an aldehyde (-CHO) group at the 7-carbon position. Due to this structural difference, chlorophylls have a different greenish color, since chlorophyll-a has a blue-green color, while chlorophyll-b shows a blue-yellow color (4, 74, 75). Because the pigment chlorophyll is a naturally abundant compound in various fruits and vegetables and has been consumed since ancient times, its use as a food component is safe. However, the main drawback is its instability, which limits its application as an additive (71).

Anthocyanins are water-soluble pigments that belong to the wider group of polyphenols (57) and are one of the most important flavonoid compounds in plants (76). Over 500 different anthocyanins were identified from several plant sources in their flowers, leaves, and in the seed, pulp or peel of the fruits. The major plant-based anthocyanins are cyanidin, pelargonidin, delphinidin, peonidin, petunidin and malvidin. Anthocyanins (E163) colorants are capable to impart four colors: red, purple, violet, and blue, where pH is a major factor that differentiates the appearance of colors of anthocyanins (4, 5, 77).

Anthocyanins are used as natural pH-dependent colorants in the food industry, they appear to be red or pink in acidic solutions and blue to purple in alkaline conditions. Thus, these can be used for many industrial applications (67). The beneficial health effects of dietary anthocyanins have been widely demonstrated due to their antioxidant activity (67), which means that they have positive effects on the prevention of cardiovascular disease, while they also show antiproliferative, anti-inflammatory, neuroprotective activity and are considered as promising chemoprotective agents in cancer treatment (57).

Betalains are water-soluble nitrogen-containing pigments present in most plants of the order Caryophyllales, which are synthesized from the amino acid tyrosine into two structural groups: the red-violet betacyanins and the yellow-orange betaxanthins. These pigments are composed of a nitrogenous core of betalamic acid, which can either condense with amines to form betaxanthins or with imino compounds to form betacyanins (69, 78, 79). The strength of these colorants is three times higher than anthocyanins (5). Approved as a natural colorant, betalains have been used in different food products like candy, yoghurt, ice cream, salad dressing, powdered drink mix, soft drinks and dessert gelatin (68).

In addition, betalains have been associated with several biological activities such as anti-inflammatory, antiproliferative, antimicrobial activities, and free radical scavenging potential, among others. This was demonstrated through in vivo studies, which indicate that betalain supplementations could play a beneficial role in diseases such as hypertension, cancer and dyslipidemia (69, 79, 80).

By considering the results reported by different investigations carried out in the last 5 years, below are described the types of by-products used, the identified pigments, the effect of different factors and processing parameters, among other important results during UAE of natural pigments.

UAE has been used to extract carotenoids from different fruits and vegetables by-products such as acerola, guava, genipap and umbu seeds and peels (72), carrot pomace (73, 81), cantaloupe peel (82), mandarin peel (54, 83, 84), Gac fruit peel (63), mango peel and paste (85), peel of different varieties of orange (38, 43, 48), orange pomace (86), pepper leaves (87), pomegranate peel (55), pumpkin peel (88, 89), purple passion fruit peel (90), sea buckthorn pomace (49, 91, 92), tomato peel and seeds (31, 93–95), and arthropods by-product such as shrimp shell and cephalothorax (40, 96–98). According to these studies, carotenoids can be used as a natural pigment directly in the preparation of various food products such as butter, salad dressing, beverages, ice cream, desserts, candy, bakery, edible oils, meet products, cosmetics, pharmaceutical products, also as an ingredient in feed for aquatic animals and pets.

Supplementary Table 1 shows the experimental and optimum UAE processing conditions of the studies listed above, as well as the types of carotenoids identified in each by-product. The main carotenoids pigments extracted and identified in these by-products were β-carotene, canthaxanthin (trans), lycopene, lutein, zeaxanthin, violaxanthin, and astaxanthin, being β-carotene the most frequent, which was identified in ∼ 45% of by-products. In addition, some by-products are sources of specific carotenoids, for example, in tomato residues, the main carotenoid is lycopene, while in shrimp residues, it is astaxanthin. Among UAE processing conditions, the power level, amplitude level, temperature, extraction time, type and concentration of solvent and solid/liquid ratio were analyzed. The evaluation of different experimental conditions allows identifying the best levels to obtain the maximum yield of carotenoids extraction.

Regarding the types of solvents used to extract carotenoids by UAE, solvents can be classified into conventional organic solvents (COS) and “green solvents”. On the one hand, the COS used were ethyl acetate (63), ethanol, methanol, acetonitrile (73), tetrahydrofuran, butylated hydroxytoluene (85), petroleum ether (95), among others. During carotenoid extraction, considering the polarity of the solvents used is very important. For instance, according to Benmeziane et al. (82), during the UAE of cantaloupe peel, the combination of polar (acetone) with non-polar (hexane) solvents for lipid-soluble compounds extraction seems to enhance the solubilization of the non-polar carotenoids (β-carotene), whereas individual polar solvents (ethanol and acetone) are thought to enhance the solubilization of the polar ones like lutein. Umair et al. (73) mentioned that variation in yield extraction could be because the dissimilar solvent polarities at different concentrations that lead to the increased cell wall permeability, while the use of non-polar solvent (hexane) can help to prevent degradation of heat-sensitive components. Szabo et al. (95) confirmed that the mixture of hexane/acetone lead to obtaining the maximum extraction of carotenoids recovered from tomato by-products. In addition, saponification processes were used to separate the carotenoids from ether bounds, methanolic solution of KOH (20%) was used in these cases (84), also enzymatic pre-treatment with cellulase and pectinase or polygalacturonase has been shown useful to accelerate extraction processes (81, 86).

On the other hand, the explored “green solvents” were d-limonene (43), sunflower, soy, olive and corn oils (55, 88, 92), and ionic liquids (ILs), which have remarkable characteristics and received more attention in the last years. Regarding edible oils, they have been used as extraction solvents to produce enriched oils with carotenoids from sea buckthorn pomace (49). However, it is important to highlight that oils contribute with an initial percentage of carotenoids that may affect the interpretation of the extraction yield obtained, carotenoids in oils are naturally present in greater quantity in the extra virgin olive oil, compared to extra virgin sunflower oil and refined sunflower oil (49), or refined corn oil (92). Particularly, ionic liquids (ILs) are salts with melting temperatures below 100°C, usually composed of a large organic cation and an organic or inorganic anion, that can be described as “designer solvents,” they present chemical and thermal stability, non-flammability, high ionic conductivity, a wide electrochemical potential, high density and low vapor pressure, which facilitate isolation of organic compounds. The used ILs during UAE of carotenoids were, for example, 1-n-butyl-3-methylimidazoliumtetrafluoroborate ([BMIM][BF4]), 1-hexyl-3-methylimidazoliumchloride ([HMIM][Cl]), 1-butyl-3-methylimidazolium chloride ([BMIM][Cl]) and 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM][PF6]) diluted in alcohol were used to extract carotenoids from orange peel (48). In addition, ILs-in-water microemulsions were used to extract carotenoids from shrimp waste (40).

On the other hand, the factors of temperature, power and processing time are also important during the extraction of carotenoids. In this regard, an increase in temperature from 20 to 40°C resulted in time reduction and an increase in lycopene content obtained by UAE of tomato peels (31). The increasing extraction of total carotenoids at stronger power or ultrasonic intensity was observed by Montero-Calderon et al. (38) where the high ultrasonic power allowed to obtain the maximum of carotenoid extracted from orange peel, and by Boukroufa et al. (43) who reported an increase in carotenoid concentration from citrus fruit waste with the increase of ultrasound intensity until 195 W/cm²; however, when intensity is higher than this value, the carotenoid content decreased about 40%. In the same way, Rahimi and Mikani (94) reported that the highest lycopene content (91.49 μg/100 g) was reached at 10 min and high ultrasonic power, while it was reduced until 85.90 μg/100 g, at 18.41 min. In addition, Gao et al. (40) reported that astaxanthin extracted from shrimp waste increased with used power from 30 W to 50 W, but when power was increased to 90 W, declined. Finally, regarding the UAE time, as observed in Figure 4B, the extraction yield of carotenoids from pomegranate peels along the UAE depends not only on time but also on the used solvent, the amplitude, temperature, and solid-liquid ratio (55). The same was observed by Chutia and Mahanta (90), where to extract carotenoids from passion fruit peel, a different optimum time was needed depending on if it was used olive oil or sunflower oil as solvents. Therefore, as explained in sections “Temperature,” “Ultrasound System,” and “Time of Processing,” the effects are not proportional, and high temperature, ultrasonic intensity and long times can lead to the pigment degradation.

For the S/L ratio, it has been shown that the carotenoid extraction yield increases, decreases or does not change. For example, Ordóñez-Santos et al. (54) evaluated the S/L ratios of 0.4–1.2 g/L to extract carotenoids from mandarin epicarp, where after optimization, the ratio of 0.4 g/L allowed to obtain a maximum extraction of β-carotene. In addition, carotenoid extraction yield increased at S/L of 20 g/L but decreased at a higher S/L ratio. In contrast, the increase of the S/L ratio from 100 to 200 g/L does not have a particular impact on the extraction yield of carotenoids in citrus fruits waste (43). The main reason for this effect is that a lower S/L ratio could cause greater concentration differences between phases which accelerated the carotenoid diffusion into the solvent, but also the ultrasound effects could be improved at lower S/L ratios (84, 90).

In addition to providing color, the carotenoid-rich extracts obtained through UAE exhibited antioxidant activity (49, 82) and peroxyl radical scavenging activity (48). Therefore, carotenoids can be highlighted as a promising natural colorant for application mainly in the food industry (97), also as an ingredient owning healthy components for the development of functional foods (92, 98). Extracted carotenoids can be used in the form of enriched edible oils (91), or can be integrated into high water content food products as a spray-dried powder (95). However, the carotenoids use can be limited by some interactions with other compounds in food products. For instance, the flavor and composition of some edible oils can be affected; while metals, acid value and the percentage of conjugated dienes in edible oils can be responsible for the oxyradical species (55). Also, the thermal stability of extracted carotenoids can limit the industrial use, Murador et al. (48) noted losses of 91.1% (at 60°C) and 100% (at 90°C) of all-trans-violaxanthin and losses of 89.1–98.3% for 9-cis-violaxanthin presented in ILs extract, all-trans-lutein was the most stable with losses of 35.2–48.8% at the same temperatures respectively. These results confirm violaxanthin to be the most unstable carotenoid to heating, while lutein was more stable.

Regarding the extraction of chlorophyll from by-products with UAE, there are very few studies carried out, in the last 5 years only 4 works were found, of which one is a conference abstract (99) and other used UAE as part of the analytical method (100). All studies evaluated chlorophyll extraction from leaves (agricultural waste produced during harvest). Chlorophyll was extracted by UAE from the powder of the dry leaves of carrot (99), papaya (101), cassava (100), and olive (102). Among the processing conditions performed in the mentioned studies, ultrasound probe and bath were used, a maximum temperature of 50°C, extraction times from 5 min to 90 min, S/L ratio from 0.033 to 0.05 g/mL using as solvents water, hexane, acetone, or a mixture of water with ethanol. According to Molina et al. (99) the UAE proved to be more efficient than maceration extraction and compared to water and hexane, ethanol allowed the extraction of greater amounts of chlorophylls. For instance, Zulqarnain et al. (101) determined that the optimal conditions for chlorophyll extraction from papaya leaf powder were ethanol 80%, 5 min, 35°C and 1 g of raw material, obtaining a chlorophyll yield of 40% (0.4 g/g) in the extract, which in the purified form was composed of chlorophyll-a (14.125 mg/g) and chlorophyll-b (19.845 mg/g). In addition, it was obtained chlorophyll-a (1.3 mg/g of freeze-dried extract) and chlorophyll-b (0.54 mg/g of freeze-dried extract) from olive leaves through UAE application (102), while in cassava leaves, it was obtained that the total chlorophyll (sum of chlorophyll-a and chlorophyll-b) content of the cassava leaves ranged from 326.27 to 747.86 mg/100 g dry matter (100).

In summary, as an alternative to conventional extraction, in recent years researchers have been working on the development of alternative UAE processes using “greener solvents” such as edible oils or ILs with more sustainable credentials. In addition, it has been observed that the carotenoid and chlorophyll extraction conditions using UAE must be studied and optimized for each type of by-product. Therefore, other types of by-products and conditions should be better explored, especially for chlorophyll extraction since there are very few studies to date.

UAE have been used to extract anthocyanins from different fruits and vegetables by-products such as bilberry press cake (76), blackberry bagasse and pulp precipitate (39, 67), blueberry peel (57, 67), grape pomace (32, 42, 45, 103, 104), eggplant peel (60, 62, 105), fig peel (106), jabuticaba eel (44, 107, 108), mulberry wine residues (41), peach waste (Plazzotta) (52), pomegranate peel (109, 110), purple corn bran (111), raspberry wine residues (112), red araçá (56), sweet cherries peel (113), wine lees (61), and huajiao peel (114). According to the reviewed authors, extracted anthocyanins can be used in the food industry such as bakery products, functional foods, and pharmaceutical and cosmetic fields, as an alternative to synthetic food colorants, also as a microencapsulated powder in yoghurt, and marshmallows.

Supplementary Table 2, for the works listed above, shows the UAE processing conditions and the optimum levels, as well as the types of anthocyanins identified in each by-product. The main anthocyanins extracted and identified were Galactoside derivates such as Delphinidin-3-galactoside, Cyanidin-3-galactoside, Petunidin-3-galactoside, Peonidin-3-galactoside, Malvidin-3-galactoside; Glucoside derivates such as Delphinidin-3-glucoside, Cyanidin-3-glucoside, Petunidin-3-glucoside, Peonidin-3-glucoside, Malvidin-3-glucoside, Malvidin-3-glucoside; Arabinoside derivates such as Delphinidin-3-arabinoside, Cyanidin-3-arabinoside, Petunidin-3-arabinoside, Peonidin-3-arabinoside, Malvidin-3-arabinoside; Rutinoside derivates such as Cyanidin-3-rutinoside, Delphinidin-3-rutinoside; Petunidin-3-rutinoside; Acetylgalactoside derivates such as Delphinidin-3-(6-acetyl)-galactoside, Cyanidin-3-(6-acetyl)-galactoside, Petunidin-3-(6-acetyl)-galactoside, Peonidin-3-(6-acetyl)-galactoside, Malvidin-3-(6-acetyl)-galactoside; Acetylglucoside derivates such as Cyanidin-3-(6-acetyl)-glucoside; Acetylhexoside derivates such as Cyanidin-3-acetyl-hexoside; Rutinoside-glucoside derivates such as Delphinidin-3-rutinoside-5-glucoside, Malvidin-3-rutinoside-5-glucoside; and Coumaroyl derivates Peonidin-3-(6-O-p-coumaroyl) and Malvidin-3-(6-O-p-coumaroyl).

The conditioning stage of the by-products has an important effect on the extraction of anthocyanins. For example, the anthocyanin extraction from grape pomace by UAE was enhanced more significatively in freeze-dried by-product samples compared to fresh ones (32). In contrast, anthocyanins were not detected in the dried peach waste sample compared to the frozen sample (52). On the other hand, among the evaluated UAE experimental conditions are the temperature, type and concentration of solvent and S/L ratio, power level and extraction time.

Regarding the temperature, a higher extraction yield was observed in anthocyanin obtained from wine lees when the temperature of extraction increased from 25 to 35°C (29). However, González et al. (32) found that increasing the temperature from 15 to 60°C did not have a significant effect on the UAE of anthocyanins from grape pomace. This reinforces the fact that the use of high temperatures, mainly above 50°C, can be detrimental. Different solvents were used to extract anthocyanins with UAE, where distilled, purified or deionized water has been commonly used in 44% of cases, either at neutral pH (26% of cases), acidic pH = 1.5–3.5 (26% of cases, by using citric acid or chloride acid) or basic pH = 12 with NaOH (4% of cases) being optimum the use of acidified water. Other commonly used solvents are hydroalcoholic solutions like ethanol (67% of cases) almost acidified, methanol (11% of cases), isopropanol (7.4% of cases), and hydrochloric acid solution (7.4% of cases). Specifically, it was recommended the use of aqueous ethanol solution (50% v/v) to extract anthocyanins from wine lees (29) and jabuticaba by-products (44), or acidified ethanol to extract anthocyanin from eggplant peels (60). In addition, Machado et al. (67), compares the yield extraction using acidified water (pH 2.0) and aqueous ethanol solutions (50, 70% v/v) obtaining that 70% v/v hydroalcoholic solution allows the best yield extraction. Dranca and Oroian (105) compare the use of ethanol, methanol and propanol, obtaining the highest recovery with methanol solution. On the other hand, NADES were also applied (11% of cases), as emerging green solvents. For instance, for the UAE of anthocyanins from blueberry peels (57), and grape pomace (45), it was applied different NADES. Paniæ et al. (45) compared the use of NADES with acidified ethanol solution (70% v/v) obtaining similar effects for ChCit (Choline Chloride:Citric acid) and ChProMa (Choline:Proline:Malic acid) at pH 0.49–3.27. Anthocyanins are highly polar compounds that are better solubilized in polar than in non-polar solvents, and their stability depends on pH value, they are stable at pH 2.0 and are degraded at pH > 7.0 (61). Acid-based NADES have polarities similar to water and at concentrations of 25% v/v decrease their viscosity improving the mass transfer during the UAE of anthocyanins from grape pomace (45).

Regarding the S/L ratio, Goula et al. (55) evaluated the S/L ratios from 100 to 0.333 g/L, being the ratio of 100 g/L the necessary for optimum conditions, while the anthocyanin extraction from wine lees slightly decreased as the S/L ratio decreased from 100 to 25 g/L (29). More and Arya (110) and Chen et al. (111) studied this factor at 20–100 g/L and 28.6–66.7 g/L respectively. All these studies demonstrated that the increase of the S/L ratio decrease the yield extraction, this agrees with the expected mass transfer phenomenon, where at a higher gradient between solid and solvent, the driving force during the mass transfer is greater. Nevertheless, a lower S/L ratio may cause more solvent consumption, so also the optimum ratio needs to be investigated for each case.

Concerning ultrasound power intensity and processing time effects during the extraction of anthocyanins, it has been shown that when ultrasonic power increased up to 300 W, it was reached the maximum of anthocyanin content extracted from mulberry wine residues; however, a further increase in ultrasonic power decreased the yield (41). The same behavior was observed for anthocyanins extracted from jabuticaba peel increased as the ultrasound intensity increased from 1.1 to 7.3 W/cm², but at 13 W/cm² the anthocyanins decreased (44). In contrast, in some cases the power level variation also may not affect pigment extraction, this was observed by da Rocha and Noreña (42) during UAE of anthocyanins from grape pomace. Regarding time, in samples with higher geometry, a longer extraction time is usually required to cause cell disruption and increase the release of pigments, this was observed during anthocyanin extraction from eggplant peel cut into squares (60). However, in powdered samples, too large extraction times can cause a reduction in yield (39). In fact, as observed in Figure 4A, this behavior was reported in anthocyanins extracted from red araçá peel (56) and mulberry wine residues (41) were after reaching the highest anthocyanin content at 90 min, after this time their content decreased, suggesting degradation. Furthermore, any variation in anthocyanin concentration after a certain processing time also was reported. This was observed during the extraction of anthocyanin from wine lees, where after 15 min a steady anthocyanin concentration was achieved (29). In the same manner, during UAE of anthocyanins from blueberry peels, at 30 min the equilibrium concentration was reached (57). In addition, Varo et al. (76) noted that the highest extraction yield was reached at approximately between 5 and 7 min using UAE, and became constant at 15–20 min, so increasing times beyond these values would not be necessary as it could degrade the pigment. The possible reasons given for these behaviors have been described in sections “Ultrasound System” and “Time of Processing”; however, usually, the maximum power used can achieve the highest yield in a short time.

Regarding characteristics reported in anthocyanin extracts, the instrumental color was the most reported. The range of color parameters L* (lightness), a* (redness: green to red) and b* (yellowness: blue to yellow) obtained were L*: 5.09–79.7; a*: 21.97–70.9 and b*: 7.39–36.01 (42, 44, 76). Some variations in the behavior of color parameters have been reported in some cases during processing. Among them, by increasing processing time, L* decreases, while a* and b* increase. Otherwise, L* and b* decrease, and a* increases when the water/ethanol ratio increases. The decrease in L* parameter means that the extracts become darker, with the a* increase and b* decrease, red color was established. This inverse behavior between a* and b* parameters is correlated with the high content of anthocyanins. In contrast, lower values of L* but higher values of a* and b* are related to brownish color as a possible consequence of the pigment degradation. On the other hand, the decrease of all parameters indicates the loss of red color, this behavior was related to high ultrasound intensities and the decrease of bioactive compounds and their antioxidant capacity (44).

In summary, evaluation of anthocyanin extraction kinetics by UAE is important as well as optimization studies since the different types of solvents and their characteristics, as well as the different levels of power, S/L ratio, temperature and time can produce results of increase, decrease or no variation in the yield of extracted anthocyanins. Furthermore, few studies have evaluated and/or proposed options to improve the stability of anthocyanins extracted by the UAE process from by-products. In this regard, the encapsulation of anthocyanins in whey protein isolate and chitosan (113) and alginate-Ca²⁺ (103), were evaluated. Bruno Romanini et al. (103) recommended the encapsulated anthocyanins for future food applications since it protects anthocyanins against light. In fact, anthocyanin complex with biomacromolecules (such as proteins) has the potential to enhance the thermal stability, oxidative stability and photostability of anthocyanin extracts preventing anthocyanin degradation during food processing (115–117). Therefore, the stability and interaction with other compounds need to be implemented in studies of pigments extracted by the UAE process.

UAE has been applied to obtain extracts rich in betalains from by-products such as beetroot peel and pomace (46, 118), prickly pear peel (47), pitahaya or dragon fruit peel (119), quinoa seed husks (68), and harvest waste such as amaranth flowers (120) and beet leaves (121). Supplementary Table 3 shows the processing conditions by UAE of the works listed above, as well as the types of betalains identified in each by-product. The amount of betalain present in the by-product can be less than, equal to or greater than that in the edible part of the raw material. For example, Fernando et al. (46) showed that the yield of betalains in beet residue is equivalent to the content in the whole beet or beet juice.

Among the experimental conditions evaluated for UAE extraction are the power level, amplitude level, extraction time, type and concentration of solvent and S/L ratio. Adequate control of these factors allows for obtaining the best extraction yield and betalain content. For instance, Melgar et al. (47) during the UAE of opuntia fruit peels obtained the highest total betacyanin content (197.51 mg/g) by applying a time of 1.5 min, S/L ratio of 5 g/L, methanol 50% at 20°C. In contrast, the lowest total betacyanin content (72.01 mg/g) was obtained at the processing conditions of 1.5 min, 25 g/L, methanol 100% and 20°C. This demonstrates that the S/L ratio must be adequate to guarantee the non-degradation of the betalains as well as the non-saturation of the solvent. On the other hand, the concentration of the ideal solvent is related to the affinity of the pigment. According to Fernando et al. (46), since betalains are hydrophilic pigments; the mixture of organic solvents with water increases the extraction yield when compared to pure organic solvents such as alcohols (ethanol, methanol, among others). However, although pure water can improve the betalain yield, it has caused difficulties during the solute separation by filtration due to coextraction of mucilaginous compounds such as pectin Considering this, solvents such as methanol 50% to extract betalains from opuntia peels (47), or ethanol 30% to extract betalains from beetroot waste (46) were recommended. On the other hand, the factors of power and processing time are also important during the extraction of betalains. It has been shown that up to a certain point there is an increase in the concentration of pigments extracted as the power or process time is increased; however, if the power and time are too high, they can lead to the deterioration of the extracted pigment. In this regard, Šeremet et al. (118), after 30 min of UAE processing, reported an amount of betaxanthin of 8.61 ± 0.08 mg vulgaxanthin I/g dry matter (dm), which after 60 min decreased to 6.98 ± 0.06 mg vulgaxanthin I/g dm. Therefore, depending on the extraction matrix, it should be evaluated whether it is suitable to use pure water or a combination of water with organic solvents in percentages not greater than 50%, also the required ultrasound power and processing time should be optimized for each by-product type.

In addition to providing color, the betalain rich extracts obtained through UAE present antioxidant and antibacterial properties. Therefore, it can be highlighted as a promising natural colorant for application mainly in the food sector as an alternative to synthetic food colorants (46, 119, 120). However, compared to synthetic colorants, their wider use could be greatly limited by their poor stability during processing and storage. Laqui-Vilca et al. (68) evaluated the thermal stability (at 90°C) of betalains extracted by UAE from husks of the quinoa seeds, reporting that the betalains from husks of quinoa “Bright red Pasankalla” showed higher stability (first-order rate constants (k) of 0.019 ± 0.008 min^–1 and half-life (t_1/2) of 37 min) than those from “Red Pasankalla” quinoa samples, but both presented thermal stability similar to betalains from beetroot. On the other hand, Fernando et al. (46) evaluated the betalain stability, reporting that during the 4-week storage, betalains quickly degraded at room temperature in contrast to -20°C. Betalain degradation is evidenced by a marked reduction of a* values and increased values of b* (i.e. reduction of red color and increase of yellow color) along with the storage. Although, studies that have evaluated the stability of betalains are still scarce and the stability of betalains from other by-products is not known.

Finally, it is important to mention that during the extraction of betalains, some processing conditions have not been clearly defined. For example, not all studies report the type of ultrasound device used as well as the set temperature of processing or the temperature reached during processing. These factors are important since the type of ultrasound equipment and its characteristics, as well as the temperature, reached during the process, strongly influence the quality of extraction using UAE (12), as was described in section 2.2. In addition, more studies are recommended to better explore the betalain purification and bioactivity, as well as their physical stability evaluation inside and outside a food matrix. In addition, new studies are required to explore the optimal processing conditions in other by-products rich in betalains.