Sweeping-frequency ultrasonic preprocessing improves removal rate and stability of pigment removed from okra powders by different drying and sieving methods

Abstract Decolorization is a necessary step before the extraction of okra pectin, and the reapplication of removed pigment will benefit hierarchical utilization of resources. Based on sweeping-frequency ultrasonic preprocessing (SFUP), effects of drying methods (hot air drying (HD), freeze-drying (FD)), sieves (of 60, 80, and 120 mesh) on removal rate and stability of okra pigment (OPI) were investigated. Color deepening caused by the formation of enzymatic products within 1 h of OPI degradation was explained. The extracted OPI was finally applied to noodles to check its stability further. Results showed that compared with non-ultrasonic treatment, SFUP increased removal rate of OPI with the maximum value of 49.32%, and further enhanced the stability of OPI when applied to noodles. OPI through sieves of 80 mesh owned brighter color and more excellent stability, and colored noodles with OPI through SFUP and FD had a small total color difference and strong stability.


Introduction
Okra pigment (OPI) has a bad impact on the quality of its extracted pectin, thus decolorization is a necessary step in the extraction process of okra pectin (OP).The current decolorization methods, including activated carbon, hydrogen peroxide, resin, and alcohol-ammonia solution, are used to decolorize the pectin-extracting solution.There, the decolorization of activated carbon makes pectin mixed with impurities that can hardly be separated.Decolorization of hydrogen peroxide easily leads to poor quality of pectin based on oxidation mechanism, and the residual hydrogen peroxide threatens human health.Macroporous adsorption resin method is green and safe, but the decolorization cycle is long, and the residual pigment on the resin reduces its rate.Although alcohol-ammonia solution has an excellent decolorization effect, it is also hazardous to human health.Moreover, the pectin-extracting solution is sticky, so the pigment is wrapped easily by the pectin, and difficult to remove.Therefore, ethanol is used to decolorize okra powders at first before the pectin dissolved, and it is convenient.In addition, the ethanol is easy to volatile so it can be removed simply by rotary evaporation, which reduces damage to human health and the collected ethanol can be recycled.
Ultrasound is one of the new non-thermal pretreatment technologies.Our research group has applied sweeping-frequency ultrasonic preprocessing (SFUP) on okra treatment.It was found that SFUP effectively reduced the time of pulsing vacuum drying (PVD) by 21.74% [1] and that of freeze-drying (FD) by 24.05%, [2] respectively.Furthermore, the extraction rate of OP was improved by 65.99-66.77%,and the physicochemical properties of OP were also studied. [3]Throughout the whole OP extraction process, the main steps included drying of fresh okra !grinding of dried okra !decolorization of okra powders !OP extraction.The research, as mentioned earlier, showed that SFUP greatly shortened the drying time of fresh okra and increased the extraction yield of OP.However, the decolorization process of OPI during OP extraction received little attention, and only a preliminary investigation of SFUP on OPI was reported.On the one hand, SFUP can promote the dissolution of OPI from cells; [1] on the other hand, it can maintain the stability of OPI to a certain extent.Under the global development trend of energy conservation and emission reduction, the decolorization method with decreasing wastewater and chemical reagents is the key step to developing OP extraction technology.Additionally, if the removal rate of OPI can be enhanced and the extracted OPI can be reused, not only are environmental pressures minimized, but also resource waste in the manufacturing process is prevented.Hence, it is necessary to study OPI and effectively use it.Theoretically, the steps of pigment removal and extraction are the same.However, because the pigment might be lost during the removal or extraction process, the removal rate is slightly higher than extraction rate.Because the removal loss is not large and difficult to quantify, the removal rate is used to represent extraction rate of OPI in the following discussions.
Besides SFUP, drying procedures have an effect on OPI; two common drying methods are selected here.Hot air drying (HD), the first method, is a typical thermal drying process that uses a continuous flow of hot air to raise the temperature primarily by water evaporation. [4]FD, the second method, is a non-thermal drying process based on sublimation dehydration, and can retain the original morphology, nutrients, and active ingredients of products. [5]Moreover, the effect of sieves of okra powders on the extraction of bioactive substances is also very important for OPI.
OPI is a natural pigment from fresh okra, and its stability is affected by temperature, food additives, storage, and processing conditions.Many studies have discussed the degradation behavior of pigments in a few days or a month. [6]Furthermore, during the preliminary experiment, it was found that the color of OPI became darkened within 1 h, which was related to the formation of enzymatic products.Moreover, in order to increase appetite and attract consumers, food coloring has become essential in modern food processing nowadays, while the debate over health risks from artificial food colorant (AFC) still exists.Therefore, natural pigments are promising to replace high-risk AFC with certain biologically active substances and antioxidant capacity, such as carotenoid (Car), chlorophyll (Chl), and anthocyanins.Green tea [7] and green vegetables have been added to improve the quality of traditional noodles.Thus, green OPI was applied to noodles to demonstrate its practical feasibility.
To sum up, fresh okra was treated by SFUP (Figure 1a) first, and the preprocessing parameters of okra were optimized according to changes in monitored ultrasonic intensity and temperature (Figure 1).Two typical drying methods (HD and FD) were followed, and then the dried okra was ground into different sieves.Finally, ethanol solution was used to decolorize okra powders.The effects of different treatment methods on the removal rate and stability of OPI were studied.Finally, OPI was applied to noodles to investigate the color changes of OPI throughout the drying and cooking processes of noodles.

Materials
Okra was purchased from Fujian Green Ginseng food shop and stored at 4 � C before the test.Okra with uniform length and size was washed with deionized water to remove contaminants.Before drying, okra pods were cut into 1 cm thick, and dried until moisture content was less than 10%.Wheat flour was purchased from Hebei Jinsha River Flour Co., Ltd., China, and stored at 25 � C. Moisture and protein contents were determined by the American Association for Clinical Chemistry (AACC) method 44-15A and AACC 46-11A (AACC, 2000), with values of 13.16% and 12.20%, respectively.

SFUP of fresh okra and ultrasonic field monitoring
A self-developed ultrasonic device was used to preprocess okra (Figure 1a).First, 500 g of fresh okra was put into a plexiglass reaction vessel (20 � 20 � 50 cm) containing 9.6 L water.The okra was pressed into the water with a hollow iron frame (15 � 15 � 5 cm), the ultrasonic plate transducer was fixed on the iron frame and also immersed in water so that the distance between it and the bottom of the reaction vessel was 5-6 cm.The center frequency of SFU equipment was 60 kHz, and the sweeping mode was as follows: the amplitude was ±1 kHz, the period was 100 ms, the power density was 25 W�L −1 .SFU waves could fluctuate up and down around the central frequency in a narrow range with a certain sweeping period, which produced the peak of output energy to the acting system and caused the object to vibrate strongly. [2]eal-time online monitoring of the ultrasonic intensity in time-frequency domains was carried out during SFUP of okra.Polyvinylidene fluoride (PVDF) sensor was introduced to collect acoustic (noise) signals.The effective area of the PVDF sensor was 10 � 10 mm, its thickness was 30 mm, the sensitivity was 2 � 10 −8 V/Pa, and a 50 X resistor was connected in parallel to improve the stability of acoustic signal collection and avoid distortion.The collected instantaneous pressure of the acoustic field was converted into electrical signals through the PVDF sensor, then recorded and stored in the form of voltage signals by the MDO3024 oscilloscope (MDO3024, Tektronix, USA) to study the SFU field in the time domain (Figure 1b).At the same time, 8103 hydrophone (8103, Br€ uel & Kjaer Sound & Vibration Measurement, Denmark) was also used to monitor signals of the SFU field in real-time, and that in frequency domain was studied by RSA306B real-time spectroanalyzer (RSA306B, Tektronix, USA) (Figure 1c).PVDF and hydrophone were located at the same position, directly below the middle of the ultrasonic plate transducer.Combined with time-frequency domain research of the SFU physical field, changes in SFUP itself over time were excavated from the acoustic perspective to predict energy-saving and efficient parameters.Equation (1) is the energy density (ED): where, P is the power (watts), t is the time (seconds), and v is the volume (mL).

Microstructure of okra tissue
Optical microscope (OM) (NIKON CORPORATION, Japan) equipped with a Nikon DS-Qi2 camera (NIKON CORPORATION, Japan) was used to observe pigment distribution of okra with magnification of 40, 100 and 400�.

Removal method
During OP extraction, OPI needs to be removed, and the removal step is equivalent to its extraction.10 g okra powders were immersed with 75% ethanol (1:20, w�v −1 ) at 40 � C for 1 h, and the ethanol solution was stirred continuously at 500 rpm�min −1 to achieve good mass transfer.The Buchner funnel with filter paper separated the green solution from the powder residues.In order to fully extract OPI, residues were treated repeatedly in fresh ethanol under the same conditions as above.The obtained filtrate was the crude extract of OPI, and it was concentrated at 40 � C by a rotary evaporator (R-100, BUCHI, India) and dried at −20 � C for 48 h.The OPI removal rate (mg�g −1 ), that is, its concentrations (mg�L −1 ), was quantified by a UV-Vis spectrophotometer.The concentrations of Chl a, Chl b and total Car were located at 665, 649 and 470 nm by the formula of Lichtenthaler, respectively; and calculated by Equations ( 2)-( 6): where, Yi is the extraction rate (removal rate) of various pigments (mg�g −1 ), Ci is the concentration of corresponding pigments (mg�L −1 ) calculated by Equations ( 2)-( 5), V is the volume of extraction solution (L), n is the dilution ratio, and m is the dry weight of powder (g).

Removed OPI color
The color evaluation was performed by a colorimeter in transmitting mode (CR-400, Konica Minolta, lnc., Japan).The colorimeter was calibrated by a white uniformity board and a board with camera calibration chart successively.All measurements were performed using a white board as background at the same lighting condition.Color values were expressed by CIE laboratory parameters, where L � represents brightness (0-100), a � represents green (-60-0) and red (0-60), and b � represents blue (-60-0) and yellow (0-60).The total color difference (TCD) was calculated as follows: TCD ¼ ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi where L 0 , a 0 and b 0 are initial colorimetric values of OPI extracts.

Noodle preparation and its color detection
10 g salt (2%, W�W −1 ) and 5 g dissolved OPI were mixed in water.Wheat flour (500 g) was then added under vacuum (−0.06 MPa) for 10 min (total moisture content of 35%) by a dough mixer (M50, Jiangsu natural love food Co., LTD, China).Then dough was proofed at 35 � C with 80% relative humidity for 15 min.Prepared dough crumbs were passed through the roller of the noodle machine (SK-1240, Solatek, Chengdu, China) to obtain a shaped dough sheet.Next, the sheet was folded in the pressing direction.By gradually decreasing the roll distance of the noodle machine, dough sheets with a thickness of 2.2 mm was obtained after being compressed by 6 times.Finally, dough sheets were cut into 2.5 mm wet noodles (WNs) and tested immediately.Meanwhile, other fresh WNs were dried by variable temperature drying for 3 h in the intelligent noodle drying device (SYT-030, China Academy of Agricultural Mechanization Sciences, Beijing, China).The maximum drying temperature was 50 � C, and dried noodles (DNs) were obtained.DNs were cooked in the water at 100 � C, and cooked noodles (CNs) were obtained.Specifically, 20 g DNs with 10 cm in length were cooked in 400 mL water for 5 min to gelatinize all starch.CNs were soaked in a beaker of 200 mL cold distilled water for 5 s.After draining the water through a sieve, the surface water of CNs was removed with filter paper and CNs were covered with plastic wrap to prevent evaporation.After the noodles cooked, the residual noodle soup was collected to a consistent volume in a 500 mL volumetric flask and utilized for color assessment.
The color of the aforementioned WNs, DNs, and CNs were detected, and the color of the residual soup was measured after noodle cooked at room temperature.The color change rate was calculated as follows:

Data analysis
All the experiments were conducted in three replicates and the mean ± standard deviation (SD) was used in the analysis by using Excel Version 2010 (Microsoft Corp., USA) and OriginPro Software Version 8.0 (OriginLab Corp., MA, USA).The significance was analyzed by IBM SPSS Statistics 26 software, P < 0.05 indicated statistically significant differences.

Ultrasonic field intensity monitoring of SFUP as okra treated
Some studies have found that the higher the ultrasonic intensity or the longer the ultrasonic time, the higher the removal rate of pigment; [8] however, longterm ultrasonic treatment will consume energy and produce thermal effect, which is not conducive to food processing.Therefore, it is very important for heat-sensitive pigment to avoid overheating effect while obtaining the suitable ultrasonic intensity.Here, the intensity and temperature of SFUP were monitored to optimize the preprocessing time and improve the removal rate of OPI.The ultrasonic intensity during SFUP of okra in time (Figure 2a, b) and frequency domains (Figure 2c-d) was shown which were respectively characterized by the amplitude of ultrasonic voltage and the power spectrum of ultrasonic signals collected by aforementioned ultrasonic field monitoring system.The voltage peak is a characteristic value of ultrasonic field in time domain monitored by PVDF.The stronger the ultrasonic field, the more obvious the cavitation and mechanical effects, and the greater the pressure of the microjet generated by the collapse of cavitation bubbles.Acting on the PVDF, the larger the amount of deformation, the greater the electrical energy converted from mechanical energy, so the greater the peak voltage. [9]The power spectrum is the intensity of ultrasonic signals in the frequency domain monitored by the spectrometer.Ultrasonic waves propagate in the medium with energy propagation, the stronger the ultrasonic field is, the intense the radiated signals are, so the medium will obtain the corresponding acoustic energy and the morphology is changed due to the ultrasonic disturbance.It is beneficial to the heat and mass transfer process.It can be seen from Equation (1) that the ED applied to the system gradually increases with time when the power volume is constant (Figure 2e).However, the ultrasonic intensity in the system does not increase gradually, but instead reaches its maximum at 30 min, which is mainly related to the pulsation process of the vacuoles in the liquid (Figure 2a-d).
Figure 2a is the voltage signal waveform of SFUP field monitored in time domain.Figure 2b is the statistical analysis of voltage peaks from Figure 2a.It was found that the voltage peak reached the maximum of 0.1288 V at preprocessing time of 30 min, so the corresponding ultrasonic intensity was the strongest.Then, with the prolongation of preprocessing time, a large number of bubble nuclei have been contributed to the oscillation process from growth to collapse in the soaking solution (water).Thus, the number of nuclei decreased with the decrement of ultrasonic field intensity, and the collected voltage peak value decreased correspondingly.The voltage peak at 50 min of SFUP field was only about 65% of that at 30 min.At the same time, the temperature of the soaking solution increased at a rate of 2 � C every 10 min, and the thermal effect was difficult to ignore with the acceleration of enzymatic reaction, which reduced the freshness of okra and brought adverse effects on the later extraction of OPI and OP.Therefore, the preprocessing time of 30 min was selected for SFUP.On this basis, intensity signals of SFUP in frequency domain at 30 min were further revealed.Figure 2c is a three-dimensional (3D) waterfall spectrum.X coordinate represented the sweeping-frequency scope by the spectrometer, and the area was wide and fluctuated around the center frequency.Y coordinate represented the monitoring time and Z coordinate represented the power spectrum distribution of SFUP.The power spectrum in frequency domain had a symmetrical distribution; the farther from the center frequency, the power was reduced in a fluctuating manner.Figure 2d is the top view of Figure 2c, which visually shows the intensity distribution of SFUP at 30 min through colors.There were many blue areas, which proved once more that intensity at this time was very strong.On the other hand, the bandwidth characteristics of SFUP field were demonstrated, [1] which can cause the resonance of complex okra structure.

OPI removal rate of okra powders
The removal rate of OPI is shown in Figure 3.In Figure 3a, After SFUP, the Chla þ b removal rate by U-FD was also higher than U-HD with growth ratio of 15.91%, 16.29%, and 39.93%.Compared with HD, U-HD had a maximum removal rate of 5.75% for Chla þ b.Meanwhile, the removal rate of Chla þ b by U-FD increased by 17.35% than that by FD.Researchers [10] confirmed the positive effect of ultrasound on Chl extraction.The removal rate trend of Car is similar to that of Chla þ b in Figure 3b.The addition of SFUP treatment increased the removal rate of Car by 1.66-49.32%.Researchers [11] observed Car extraction (removal) was promoted due to ultrasonic cavitation effect.However, the cavitation effect can also lead to the decomposition of Chl.Researchers [12] observed that when the ultrasonic power increased from 300 to 600 W, the Chl extraction from mulberry leaf paper decreased.Our research group [2] observed that ultrasonic pretreatment degraded okra Chl by 5.06%.In conclusion, on the one hand, SFUP effectively promoted the removal (extraction) of Chl and Car.On the other hand, it was possible to degrade Chl.From both aspects, the removal efficiency of OPI in the decolorization step before OP extraction was improved.
In terms of drying methods, compared with HD, the removal rate of Chla þ b of M60, M80, M120 OPI by FD was higher with the ratio of 11.83%, 4.80%, and 33.65%, respectively; and a higher removal rate of Car was also obtained by FD related to higher retention rate, and similar finding previously reported by Researchers [13] Chl may be degraded by light, temperature and oxygen.Here, the high temperature of HD was 50 � C, while FD was carried out at a lower temperature of 25 � C, and oxygen was isolated by a vacuum environment, resulting in a higher retention rate of OPI than HD.
In terms of okra powders with different sieve sizes, the larger the mesh, the more the removed OPI.It is found that the distribution of chloroplasts in the outer epidermis of fresh okra is denser than that in the inner sarcocarp (Figure 4).In fact, the number of chloroplasts was positively correlated with pigment content.Furthermore, the OPI distribution of okra itself was uneven and mostly in the outer epidermis.Moreover, okra outer epidermis became dense and hard due to shrinkage during the drying process, and a large portion of the outer epidermis with many chloroplasts was in M60 group correlated to more removed OPI, which can be confirmed by color of okra powders Figure 1.

Observation of okra tissue
The epidermis is greener than the inner part from the cross-section of fresh okra in Figure 4a; after SFUP, the green part had a larger distribution area.Figure 4b (40�) shows that SFUP promotes the dense green OPI to disperse widely in the tissue, making the distribution of OPI more uniform and easy to cause a high removal rate of OPI.Chloroplast aggregation and Chl loss are found in Figure 4b (100�).Figure 4b (400�) further shows that chloroplasts of fresh okra are in the middle of cells.After SFUP, chloroplasts were separated from the center to the edge of cells, some chloroplasts were destroyed, and OPI was leaked from the cell edge (red circle).Figure 4c explains that OPI generally occurs in the center of cells and is difficult to be detached; however, after SFUP, chloroplasts are transported to the cell edge and finally shattered.Thus, OPI is disjoined from chloroplasts into cells and diffused out of cells.

Color change during OPI degradation
The change of color intensity is significantly related to the real-time content or proportion of each component and the loss and formation of each substance during pigment degradation, such as Chl a (blue-green), Chl b (yellow-green) and Car mainly with b-carotene (orange) and lutein (yellow).It also includes metabolites of Chl and Car, such as pheophytin (Phein) (olive green), chlorophyllide (Chlide) (green), pheophorbide (Pheide) (olive green) and zeaxanthin (yellow).Under different oxygen, light, temperature, and pH, Chl has more than one degradation pathway, and decomposition and destruction of pigment is induced.Therefore, the influence of temperature on pigment can significantly affect the activity of enzymes related to pigment degradation. [14]hl accounted for 88.11-91.88% of total pigment for removed OPI, while Car was less (8.12-11.89).In Figure 5a, chlorophyllase (Chlase) directly takes Chl as a substrate and catalyzes the hydrolysis of phytol ester bonds to produce Chlide and phytol.In addition, Pheophytinase (PPH) catalyzes the removal of the center Mg 2þ ion of Chl/Chlide to become Phein/Pheide. [15]At the same time, there is also a possibility of deplantation on the Mg 2þ -free Chl.Chlase is also responsible for converting Phein to Pheide, leading to another Chl degradation pathway.Both proposed pathways show that Chl is cleaved into several intermediate catabolites in chloroplasts.Researchers [16] indicated the existence of two Chl degradation pathways in green plant tissues, namely (1) Phein to Pheide, and (2) Chlide to Pheide.Enzymes are also involved in the bleaching of Chl; Pheide, the direct intermediate resulting from Mg-dechelation, has its porphyrin ring oxygenolytically opened by Pheideoxygenase; hence, the loss of the green color of molecules occurred.Afterward, red Chl catabolite (RCC) is reduced by RCC reductase (RCCR).Subsequently, downstream products are degraded by non-enzymatic tautomerization. [15]At this point, Chl is degraded from green to colorless compounds, and the degradation is finished.
Car includes carotene and lutein.b-carotene 3hydroxylase (HYD) catalyzes b-carotene to produce two important intermediate and final products, namely b-cryptoxanthin and zeaxanthin.b-carotene metabolic pathway was also related to important catalytic enzymes, such as zeaxanthin epoxidase (ZEP) and violaxanthinde epoxidase (VDE).In addition, violanthin, anthoxanthin, and zeaxanthin are components of lutein metabolism. [14]The color of Car is due to the presence of chromophores that are mainly or entirely composed of conjugated double bond chains in molecules, [17] and there is a strong correlation between yellow fading and Car degradation.Due to its highly conjugated structure, lutein is prone to various reactions such as oxidation and isomerization.Thus, it could be degraded into colorless products, and its biological activity will be lost.Similar to lutein, b-carotene undergoes (Z)-isomerization after heat treatment and is eventually degraded into colorless products, and biological properties are changed. [17]he biocatalysis efficiency of enzymes is affected by their activity and stability.Enzymes have the strongest activity and can catalyze reactions more efficiently when in the optimum temperature range.Chlase has thermal stability at 20 � C; the higher the temperature, the weaker the thermal stability.Within the optimal temperature range, the catalytic reaction efficiency of other enzymes increases with the increment in temperature.Moreover, since most enzymes are proteins, food additives such as salt can also affect their activity.

Temperature stability of OPI
At 4 � C (Figure 5b), For a � , OPI lost its greenness slowly.For b � , yellowness increased in the first 20 min and then decreased.Decrement of greenness and increment of yellowness indicated a higher degradation rate of Chl than that of Car.Chl was reported to be easily destroyed, while Car naturally existed in all-(E) configuration was a thermodynamically stable isomer. [18]Since the decomposition and disappearance of Chl were ahead of Car, the solution showed a yellowness of Car.After 20 min, Chl and Car were losing color due to degradation.The enzyme activity was low at 4 � C, so the enzymatic reaction was difficult to occur without color change of enzyme products.
At 25, 50, and 100 � C (Figure 5c-e), the color change of OPI was similar.For a � and b � , OPI turned green and yellow firstly and then slowly losing greenness and yellowness.Discriminatively, yellowness and greenness darkened simultaneously within 90 min at 25 � C. At 50 � C, greenness darkened within 60 min, and the yellowness darkened at 90 min.At 100 � C, there was a significant inflection point of a � and b � curves, the greenness and yellowness darkened rapidly within 10 and 60 min, respectively.Later, OPI degraded and faded.At 25 � C, Chlase showed catalytic activity.For a � , Chl and Phein were decomposed into green Chlide and olive green Pheide by Chlase and PPH with a darker green color, and an unstable red metabolite was produced to neutralize greenness; finally, greenness was faded by further oxidation.For b � , yellow zeaxanthin was formed from Car by some enzymes.The reaction rate of enzymes became lower than the degradation rate of OPI, which showed a slow fading behavior.The color fluctuations were more pronounced at 50 � C and 100 � because enzymes catalyzed the reaction more efficiently at higher temperatures.The color inflection point occurred earlier, because the thermal stability of enzyme lost at higher temperatures and catalysis ended more quickly.The inflection point of a � was earlier than that of b � , and Car had relatively higher heat resistance than Chl.
At different temperatures, L � increased during the whole process due to the loss of OPI.At 4 and 25 � C, DE changed with little fluctuation (0-0.4 and 0-0.6).At 50 and 100 � C, DE fluctuated greatly (0-2.0 and 0-5.5) in the first 20 min because of the fast enzymatic reaction.Specifically, the enzymatic degradation rate after SFUP became lower, indicating that some enzymes that related to OPI degradation was inactivated to a certain extent.At 50 � C, the substrate was gradually exhausted hence DE (0-0.8) was not changed significantly.At 100 � C, degradation faded faster from 30 to 150 min, and DE gradually increased to 2.0.After 150 min, the degradation rate reached a maximum of 5.0, and OPI treated by SFUP showed weak stability.
Above all, OPI in FD and U-FD groups showed low L � , low a � and high b � , so colors were dark, green and yellow compared with HD and U-HD because of more Chl and Car.Researchers [19] observed similar a � results after ultrasonic treatment and FD.Researchers [20] showed that ultrasonic preprocessing increased b-carotene content by 4-42%.Ultrasonic-assisted b-carotene content was improved, which was mainly because within a certain time range, with the increase of ultrasonic irradiation time, cavitation effect and thermal effect were gradually stronger, and cavitation oscillated and turbulent behaviors in liquid, which led to the breakage of okra powders particles, the rupture of cell wall and the increase of mass transfer efficiency. [21,22]n addition, with increased extraction time the interaction between the target compounds and the solvent is higher, resulting in greater diffusion. [23,24]The researchers found that the longer the time of ultrasonic irradiation, the higher the degree of change in the cell structure of plant tissue, resulting in a greater degree of cell wall breakdown, and therefore increased porosity, and the higher the beta-carotene content. [25]In addition, the increase in temperature may also affect the increase in b-carotene content, which may increase the solubility and diffusion coefficient of the extracted compounds, making b-carotene easier to pass through the okra powders.In addition, the opening of the cell matrix and the removal of barriers to the exit of these compounds can also be influenced by the increase in temperature, resulting in increased membrane permeability and enhanced beta-carotene release. [26]ltrasound itself may disrupt the hydrophobic interaction of protein molecules through conformational changes; hence enzyme inactivation is induced.In addition, ultrasonic cavitation with continuous expansion and compression of tissue causes changes in cells, and promotes the formation of micro-channels, improving the mass transfer efficiency and reducing drying time.Hence, SFUP accelerated the drying process of okra, and it spent a short time reaching low water activity.It is known that water is essential for the integrity of the three-dimensional structure of enzyme molecules, and enzymes require a small hydration layer as the primary component of the enzymatic microenvironment in the organic medium.This layer acts as a buffer between the enzyme surface and the reaction medium of substrates. [27]Therefore, the reduced water activity put enzymes into a dormant state, restrained enzymatic reactions and excessive degradation of Chl during the long drying process (12-24 h).FD effectively preserved color; [28] thereinto, OPI of U-FD-M80 showed a green color with an overwhelming advantage of temperature stability.

Food additive stability of OPI
After adding an equal proportion of 10% salt and starch solution, the stability of OPI is given in Figure 5f-g.In Figure 5f with monotonous changes of L � , a � , and b � , the solution continues to lose greenness and yellowness because of OPI degradation and there is no obvious formation of enzyme products.The enzymatic reaction of OPI was inhibited after the salt added, it was known that enzymes are mainly composed of proteins associated with salting out of protein enzymes.In the absence of OPI enzymatic products, the color continued to degrade, resulting in a steady change in DE (0-0.8).
The starch content in wheat flour is more than 70%, and the rest is protein and other contents.In order to exclude the influence of other substances and consider the wide applicability of OPI in food, the stability of OPI in starch was observed.In Figure 5g, compared with the stability under the aforementioned conditions with an increment of L � , it is abnormal with a decrement trend after starch is added.The change rate of a � and b � was slow because the starch had no significant negative effect on OPI degradation.The brightness of starch in the solution decreased due to oxidation, including polyphenol oxidase and nonpolyphenol oxidase (non-PPO).Starch, a promising coagulant and flocculant, also has physical adsorption.Starch molecules are easily associated with some polar organic compounds through hydrogen bonds, resulting in partial adsorption of pigments, the enzymatic reaction and degradation rate are weakened.The decrement of L � was also related to the gradual absorption of OPI on starch and the color got darker.According to DE, after 150 min, the stability of OPI by SFUP in starch solution (0-0.6) was stronger than that without SFUP (0-0.8).
Besides, the stability of OPI (DE) was M80 > M60 > M120 after salt and starch added, and the concentration of OPI in M80 group was higher than that of M120, which was sufficient to maintain stability in additives.

Stability of OPI during the noodle-making process
Colored noodles are the perfect combination of color, aroma, taste and shape.Not only are the natural nutrition of vegetables retained, but also the characteristics and strength of noodles were not affected.
Based on studies of OPI, M60 OPI had the highest removal rate, but its stability was relatively low, because pigment extracts may contain a larger proportion of other alcohol-soluble substances such as polyphenol flavonoids.Thus, the application of M80 OPI with brighter color and more excellent stability was emphatically studied with higher dyeing efficiency.In addition, the texture of removed OPI in noodles can be given in Figure S1.Photos and CIELAB color of noodles in Control (without OPI), HD-M80, U-HD-M80, FD-M80 and U-FD-M80 are shown in Figure 6a and Table 1.Noodles in Control group were off-white.With the addition of OPI, noodles turned green, namely all groups of WNs, DNs, CNs and even the remaining noodle soup.In Table 1, compared to WNs with HD OPI, the brightness of WNs with FD OPI was 2.18% higher, and the color was 28.9% greener.In addition, SFUP had no significant effect on WNs with HD OPI, but the greenness of WNs with FD OPI was reduced by 6.26%.This was related to the structure of FD OPI by SFUP, resulting in the color difference of noodles.
In Figure 6b, SFUP significantly reduces DE of noodles with HD OPI and FD OPI by 20.08% and 85.76% during the noodle drying process, respectively.Meanwhile, SFUP also significantly increased DE of noodles with HD OPI by 32.49% during noodle cooked process but had no significant effect on DE of noodles with FD OPI.This was related to the SFUP enhanced stability for the combination of OPI and noodles.Among all the samples, noodles by U-FD-M80 showed the most outstanding comprehensive performance in dried and cooked stages of noodles with the smallest DE loss and the strongest stability.
In Figure 6c and Table 1, after WNs drying, the brightness of samples with added OPI increased by 3.65% on average, while greenness and yellowness decreased by 43.63% and 15.01%, respectively, compared with WNs, which was caused by a large loss of pigments.Here, WNs were dried at 50 � C with a low loss of OPI.In Table 1, the dyed WNs without SFUP were greener than those with SFUP; surprisingly, after noodles dried, groups with SFUP turned greener than those without SFUP.Figure 6c shows that although greenness was lost due to the drying process of noodles, SFUP can retain the green and yellow colors better.And smaller DE of OPI by SFUP in starch indicated similar results in Section 3.3.3.Researchers [29] showed that ultrasonic-assisted synthesized pigments showed narrow particle size distribution, brighter color and better thermal stability.Thus, the particle size of OPI was reduced after SFUP, and OPI can be uniform and stable when mixed with food additives and starch of noodles.Researchers [30] also showed that ultrasonic treatment caused significant changes in the properties of the adsorbed layer and was a promising method for surface modification of pigments in terms of dispersion stability.
In Table 1 and Figure 6d after the noodles cooking, the brightness and yellowness decrease by 6.05% and 26.55%, respectively, but greenness increases by 15.51% compared with DNs.SFUP did not bring a significant difference, but FD retained green color better than HD.Hence, it can be said that the stronger stability of FD-OPI was shown as mingled in noodles.The color of the remaining noodles soup in Table 1 indicates that some nutrients of noodles may be lost in the soup during the cooking process.According to the small change of CIELAB values, there is very little color loss for noodles with OPI, and the color of noodles has a certain loss after drying and cooking processes.

Conclusion
Decolorization is an inevitable step in the whole extraction process of OP.It is a safe and green choice to apply the removed natural OPI to food and avoid resource waste.SFUP was used to remove OPI with a maximum value of 49.32%.Meanwhile, its stability in the application of noodles was significantly enhanced.FD OPI had a higher removal rate and a greener color than HD OPI.For different meshes of okra powders, M60 OPI had the highest removal rate, while M80 OPI had higher purity and was more stable in food additives; therefore, it can dye more efficiently in the application of noodles.
Furthermore, degradation is the most noticeable problem of natural pigments, and the degradation of pigments in a long time is inevitable.It was found that the color of OPI deepened in a short time (less than 90 min) during degradation.The yellow-green was simultaneously lightened in a long process (After 1 h of degradation).When OPI was applied to noodles, greenness was lost during the noodles drying process but appeared again during the cooking process.The stability of OPI by SFUP was significantly enhanced when applied to noodles.In view of this, the successful use of OPI in noodles can catch the interest of manufacturers in the prospect of more applications.

Figure 1 .
Figure 1.Preparation of okra powders based on SFUP, different drying methods, and meshes.

Figure 3 .
Figure 3. Pigment removal rate from okra powders with chlorophyll (a) and carotenoids (b).

Figure 6 .
Figure 6.Application of OPI in noodles.(a) Photos of OPI dyed noodles, (b) changes of DE, (c, d) changes of L � , a � and b � of WNs, DNs, CNs and remaining noodle soup.

Table 1 .
Changes of L � , a � and b � of WNs, DNs, CNs and remaining noodle soup.