Sage Journals: Discover world-class research

Abstract

Objective

To refine the lycopene extraction from tomato powder with ionic liquid-microwave assistance, seeking a more efficient and cost-effective process that optimizes resource usage, lowers extraction costs, and boosts the economic value of lycopene.

Methods

The study investigated the effects of various factors on the extraction yield of lycopene. The researchers screened ionic liquids and constructed a BP (Back Propagation) neural network model to optimize the experimental data. This model was further refined using a genetic algorithm-neural network approach to determine the optimal extraction process parameters.

Results

After conducting the optimization process and subsequent verification experiments, the study identified the optimal conditions for the ionic liquid-microwave-assisted extraction of lycopene from tomato powder. These conditions were found to be a microwave power of 800 w, an extraction time of 150 s, an ionic liquid concentration of 0.17 mol/L, and a liquid-to-material ratio of 15:1 (mL/g). Under these conditions, the lycopene content achieved was 0.177 mg/g, which corresponded to the highest extraction efficiency.

Conclusions

The study optimized lycopene extraction with ionic liquid-microwave assistance, improving efficiency, reducing waste, and lowering costs to enhance its economic value from tomatoes.

Keywords

artificial neural networks tomato lycopene ionic liquids microwave extraction

Introduction

Lycopene, a fat-soluble natural pigment prevalent in tomatoes, watermelons, red grapefruits, and other vegetables, boasts a range of health benefits, including antioxidant, anti-apoptotic, antiatherosclerotic, anti-inflammatory, antihypertensive, and protective endothelial properties.^1–3 However, lycopene's stability is compromised by its sensitivity to light, oxygen, and metal cations, leading to rapid degradation under suboptimal storage conditions.⁴ As the human body lacks the ability to synthesize lycopene, it must be ingested through dietary sources. Lycopene's applications span from enhancing the value of various foods such as beverages, sauces, and baked goods as a natural pigment and nutritional supplement, to its potential use in the healthcare industry for the prevention and auxiliary treatment of certain diseases.⁵ Additionally, it finds a niche in the cosmetics sector for developing products with anti-oxidation and anti-aging effects.

Methods like mechanical crushing, heat treatment, ultrasound treatment, high-pressure homogenization, and extrusion processing can disrupt plant cells, thereby enhancing lycopene release. The bioavailability of lycopene can be influenced by its interactions with other compounds in food.^6,7 Extraction techniques for lycopene from tomatoes encompass solvent extraction,⁸ enzymatic extraction,⁹ and microwave-assisted extraction,¹⁰ etc.

Ionic liquids, composed solely of cations and anions, are molten salts at room temperature, also referred to as low-temperature molten salts. They possess environmentally friendly, stable properties, and are easily reusable, with high extraction efficiency and strong designability. These characteristics make ionic liquids a promising option for reducing environmental pollution during the extraction and separation process while significantly improving efficiency.¹¹ They excel in fields such as reaction solvents, catalysts, and adsorption and separation. As the cost of producing ionic liquids decreases, they show potential in the extraction of natural active substances. Microwaves, electromagnetic waves ranging from 300 MHz to 300 GHz, are utilized in microwave extraction, a novel technology that harnesses microwave energy to boost solvent extraction efficiency. During the microwave extraction process, microwave radiation causes rapid rotation and vibration of polar molecules, generating substantial heat energy to break down cell walls, allowing cell substances to be rapidly transferred to the extraction solvent.¹² Microwave-assisted extraction technology, known for its high efficiency, low solvent consumption, short extraction time, and absence of alcohol-insoluble residue, has been widely adopted.¹³

The BP (Back Propagation) neural network represents a quintessential multi-layer feedforward neural network architecture, which is proficiently trained via the error back propagation algorithm. It stands as one of the most prevalent neural network models in contemporary usage, with extensive applications across diverse domains including pattern recognition, function approximation, and data prediction.¹⁴ The BP neural network is configured by specifying the number of nodes in both the input and output layers, as well as the number of hidden layers. It undergoes iterative training on pertinent input data and their corresponding desired output values through the employment of the error BP algorithm. During the training regimen, input data are funneled into the network to generate output values. These outputs are then juxtaposed with the desired outcomes to compute the error. Subsequently, the network's weights and biases are fine-tuned in the opposite direction, utilizing gradient descent and related techniques, with the intent of progressively diminishing this error. This BP process entails the calculation of gradients in a retrograde manner, from the output layer back to the input layer.¹⁵ By leveraging the gradient information gleaned from this process, the network's weights are incrementally optimized through successive training epochs until the stipulated termination criteria for training are fulfilled. This refinement enhances the network's fidelity in mapping input data, underpinning the robust application of neural network technology in addressing intricate nonlinear challenges.¹⁶

With the escalating demand for lycopene, the scientific community has become increasingly captivated by this carotenoid. Investigations into its biological functions and utility are advancing steadily, accompanied by the innovation and deployment of novel extraction technologies and methodologies designed to enhance both the efficiency and quality of lycopene extraction processes. This study outlines the construction of a BP neural network model and employs a genetic algorithm in conjunction with neural network optimization to refine the ionic liquid-microwave assisted extraction protocol for lycopene from tomato powder. The objective is to delineate the optimal process parameters for the ionic liquid-microwave assisted extraction of lycopene, thereby augmenting the extraction yield and establishing a procedural cornerstone for the further development and utilization of this valuable resource.

Results

Selection of Ionic Liquids

The data presented in Table 1 revealed that the maximum extraction yield of lycopene from tomato powder was obtained utilizing 1-butyl-3-methylimidazolium chloride for the preparation of the ionic liquid. Consequently, following a thorough experimental screening process, it had been determined that 1-butyl-3-methylimidazolium chloride will be the ionic liquid of choice for all subsequent experimental procedures.

Table 1.

Screening Results of Ionic Liquids.

Ionic Liquid	Absorbance (470 nm)	Extraction Yield (mg/g)
1-Ethyl-3-methylimidazolium Bromide	0.342	0.1156 ± 0.01
1-Ethyl-3-methylimidazolium Chloride	0.402	0.1459 ± 0.012
1-Butyl-3-methylimidazolium Bromide	0.505	0.1979 ± 0.017
1-Butyl-3-methylimidazolium Chloride	0.536	0.2136 ± 0.015
1-Butyl-3-methylimidazolium Hexafluorophosphate	0.476	0.1833 ± 0.012
1-Butyl-3-methylimidazolium Tetrafluoroborate	0.527	0.2090 ± 0.018

Single-factor Experimental Results

In the course of the single-factor experiments, the optimal ranges for microwave power, microwave time, ionic liquid concentration, and liquid-to-solid ratio were ascertained based on varying extraction efficiencies for lycopene, as depicted in Figure 1. Figure 1A revealed that the extraction yield of lycopene from tomato powder increased progressively with the enhancement of microwave power from 80 W to 640 W, with a peak observed at 800 W. Figure 1B demonstrated that the extraction yield of lycopene increased continuously with the extension of extraction time, peaking at 2.5 min. The correlation between ionic liquid concentration and lycopene extraction yield, as presented in Figure 1C, indicated that the maximum yield was achieved at a liquid-to-solid ratio of 15: 1 mL/g. Figure 1D illustrated a trend of initial increase followed by a decrease in yield, with the optimal point occurring at a concentration of 0.25 mol/L.

Figure 1.

Effects of Different Factors on Lycopene from Tomato powder. (A) Microwave Power, (B) Microwave Time, (C) Liquid–Solid Ratio, and (D) Ionic Liquid Concentration.

Evaluation of the BP Neural Network Model's Performance

The correlation coefficient of the fitting results served as a critical metric for assessing the alignment between the BP neural network model and the empirical data. The model was developed and trained using both experimental data and synthetic samples, with the outcomes presented in Figure 2.

Figure 2.

Regression Plot Showing the Regression Coefficient of the Experimental Data and BP Neural Network Model.

Evaluation of the BP Neural Network Model's Reliability

Figure 3 displayed the comparison between the actual and predicted values of the lycopene extraction yield. A close examination of the figure revealed a strong alignment between the observed and predicted values for the test dataset, demonstrating the model's predictive accuracy. While there are instances where the predicted values in Figure 3 deviate from the actual values, either overestimating or underestimating them, these instances were relatively few. For the majority of the data points, the predicted values exhibited minimal error and a high degree of correspondence with the actual values, suggesting that the BP neural network model was reliable and stable. Consequently, this model is deemed suitable for predicting the lycopene extraction yield under various extraction scenarios.

Figure 3.

Reliability Verification Test Results of BP Neural Network Model.

Optimization for the Lycopene Extraction Process

Execution of the pertinent program yields a performance change curve for the network training, as depicted in the accompanying Figure 4 (noting that the middle curve signifies the error trend). The figure clearly illustrates that with the progression of iterations, the fitness curve exhibits a steady ascent, eventually leveling off to form a straight line. Throughout the training phase, the error diminishes, culminating in optimal fitness at the 60th generation. Subsequent to this point, the fitness curve flattens out, signifying that the extracted lycopene content has reached its peak value and remains relatively stable thereafter.

Figure 4.

The Performance of the BP Neural Network Model.

Validation of the BP Neural Network Model

Upon conducting a comprehensive global optimization of the extraction protocol, the predicted peak concentration of lycopene in tomato powder was determined to be 0.1771 mg/g. The optimized extraction parameters included a microwave irradiation duration of 151.94 s, an ionic liquid concentration of 0.1689 mol/L, and a liquid-to-solid ratio of 15.12: 1 mL/g. To facilitate practical application, these parameters were slightly refined to a microwave time of 150 s, an ionic liquid concentration of 0.17 mol/L, and a liquid-to-solid ratio of 15: 1 mL/g. Utilizing these optimized conditions, the actual lycopene content in the tomato powder was found to be 0.157 mg/g, which was slightly lower than the concentration forecasted by the BP neural network optimization.

Discussion

The selection of 1-butyl-3-methylimidazolium chloride as the ionic liquid for lycopene extraction was based on its superior performance, which yielded the highest extraction rates following a comprehensive screening process. This choice was pivotal in establishing consistent and optimal conditions for the subsequent experiments.

The effect of microwave power on lycopene yield was significant, with a notable increase observed as the power was elevated from 640 W to 800 W, peaking at this level. The enhanced energy transfer at higher microwave power levels is credited with facilitating the breakdown of tomato cell walls, thereby accelerating the release of lycopene.¹⁷ However, beyond this threshold, the yield began to decline, suggesting that prolonged or excessive microwave exposure may lead to elevated system temperatures and induce degradation reactions in lycopene.

The volume of the ionic liquid played a crucial role in the extraction process. An increase in ionic liquid volume enhanced the concentration gradient between the cells and the solvent, favoring lycopene dissolution.¹⁸ Ionic liquids are effective at absorbing microwave energy, thereby enhancing the efficiency of the microwave-assisted extraction process. However, an excessive amount of ionic liquid can absorb an undue proportion of the microwave energy, reducing the energy available for cell disruption and solubilization, and consequently leading to a decrease in extraction yield.¹⁹

The BP neural network model was developed and trained using both experimental data and synthetic samples. The model's correlation coefficient value nearing 1 signified a robust fit to the data. The BP neural network's nonlinear mapping capabilities allowed for the refinement of the extraction process parameters, constructing a sophisticated relationship model that correlates lycopene content with microwave duration, ionic liquid concentration, and liquid-to-solid ratio.

A comparison of the outcomes from single-factor experiments and neural network optimization revealed disparities in the optimal extraction conditions. The neural network model's enhanced expressive power likely accounts for these differences, as it can identify and fit more intricate patterns beyond the scope of single-factor experiments. However, the model's requirement for a substantial number of samples for learning and optimization is a critical consideration, and the constraint of a limited number of experimental samples may have led to increased randomness and variability in the outcomes.

The discrepancy between the predicted and measured lycopene concentrations was investigated through a critical review of the literature and detailed analysis. The variance is likely due to the inherent instability of lycopene, which is highly prone to reactions with ambient oxygen and light.²⁰ Suboptimal storage conditions, such as inadequate sealing and insufficient protection from light exposure during the experimental process, may have facilitated the interaction of lycopene with oxygen and light, triggering oxidation and degradation processes that resulted in a lower measured concentration of lycopene. These findings underscore the importance of rigorous control over storage and handling conditions to maintain lycopene integrity during the experimental process.

Limitations

The current study provides valuable insights into the optimization of lycopene extraction using ionic liquid-microwave-assisted extraction. However, it does have a few limitations that could be addressed in future research:

The study focuses solely on tomato powder from a single source and location, potentially overlooking the impact of variety, maturity, and geographical origin on extraction efficiency. This limits the generalizability of the findings and the applicability of the optimized protocol to diverse tomato sources.

While verification experiments demonstrate the validity of the optimized protocol, a larger-scale validation using diverse tomato powder batches and instruments is crucial to ensure its robustness and generalizability across different settings.

The study lacks an economic feasibility analysis, neglecting the cost-effectiveness of the process. Factors such as the cost of ionic liquids and microwave irradiation energy consumption are essential considerations for practical implementation and commercialization of the extraction method.

Conclusion

Utilizing the BP neural network to refine the experimental data, the optimized process conditions and parameters were established as follows: a microwave power of 800 W, an extraction duration of 150 s, an ionic liquid concentration of 0.17 mol/L, and a liquid-to-solid ratio of 15: 1 (mL/g), yielding a lycopene concentration of 0.177 mg/g. This study offers an innovative technical methodology and insights for the extraction of lycopene from tomato substrates.

Experimental Section

Materials and Chemicals

Tomatoes (2.0 kg) were purchased from a local marketplace in Harbin, China, in April 2023. The lycopene standard (CAS 502-65-8) was supplied by the Shanghai Aladdin Biochemical Technology Co., Ltd (Shanghai, China). Analytical-grade 1-ethyl-3-methylimidazole bromide, 1-ethyl 3-methylimidazole chloride, 1-butyl-3-methylimidazole bromide, 1-butyl-3-methylimidazole chloride, 1-butyl-3-methylimidazole hexafluorophosphate, and 1-butyl-3-methylimidazole tetrafluoroborate were purchased from Jinan Xinda Chemical Co., Ltd (Jinan, China).

Lycopene Extraction and Content Determination

Prepare the tomato samples by first thoroughly washing the freshly purchased tomatoes. Using a pair of scissors, carefully trim the tomato skin into small fragments, discarding the pulp. Transfer the skin pieces to a mortar and pestle them to a fine consistency. Evenly distribute the crushed tomato skin on a drying tray and place it in a vacuum drying oven to dehydrate. Once dried, remove the tomato skin and transfer it to a grinder to be pulverized into a powder. Collect the powder and store it in a suitable container for future experimental use.

For the experimental procedure, precisely measure 0.5 g of the tomato powder and transfer it to a round-bottomed flask. Prepare a series of ionic liquids by dissolving various salts at a concentration of 0.25 mol/L, as indicated in Table 2. Pipette 10 mL of each ionic liquid into the flask containing the tomato powder and mix thoroughly. Set the microwave to a power of 400 W (KY-WB-201 Microwave Chemical Reactor, Shanghai Qiuzuo Scientific Instrument Co., Ltd, Shanghai, China) and irradiate for a duration of 1 min. After microwaving, carefully decant the supernatant and centrifuge it at 5000 rpm for 5 min. Collect the centrifuged supernatant and transfer it into cuvettes. Use a 721 UV-visible spectrophotometer (Tianjin Taisite Analytical Instrument Co., Ltd, Tianjin, China) to measure the absorbance at a wavelength of 470 nm. The standard curve for lycopene was constructed with solution concentration on the x-axis and absorbance on the y-axis. To determine the lycopene content, use the following formula:

X = \frac{C \times V \times N}{M}

where: X represents the lycopene content in tomatoes, expressed in mg/g; C is the concentration derived from the lycopene standard curve equation, in mg/mL; V is the volume of the lycopene extract from tomatoes, in mL; N is the dilution factor; and M is the mass of the tomato powder used, in g.

Table 2.

Configuration of Ionic Liquids.

Ionic Liquid	Molecular Weight (g/moL)	Weight (g)	Volume of Distilled Water (mL)
1-Ethyl-3-methylimidazolium Bromide	191.07	0.48	10
1-Ethyl-3-methylimidazolium Chloride	146.62	0.37	10
1-Butyl-3-methylimidazolium Bromide	219.12	0.55	10
1-Butyl-3-methylimidazolium Chloride	174.67	0.44	10
1-Butyl-3-methylimidazolium Hexafluorophosphate	284.18	0.71	10
1-Butyl-3-methylimidazolium Tetrafluoroborate	226.02	0.57	10

Single-factor Experiments

The efficiency of lycopene extraction is influenced by several key parameters, including the microwave power, microwave time, ionic liquid concentration, and liquid-to-solid ratio. To assess the impact of these variables, a systematic single-factor experimental design was employed to screen each of these four factors. The factors and their levels were set as follows: microwave power (80 W, 240 W, 400 W, 640 W, 800 W), microwave time (1 min, 1.5 min, 2 min, 2.5 min, 3 min), ionic liquid concentration (0.15 mol/L, 0.25 mol/L, 0.35 mol/L, 0.45 mol/L, 0.55 mol/L), and liquid-to-solid ratio (10: 1, 15: 1, 20: 1, 25: 1, 30: 1).

BP Neural Network Model with Genetic Algorithm Optimization for Lycopene Extraction

In the developed neural network model, the microwave processing time, ionic liquid concentration, and liquid-to-solid ratio were designated as the three neurons in the input layer. The lycopene content was set as the node in the output layer. Recognizing the necessity for a substantial dataset to train the BP Neural Network, yet acknowledging the limited number of original experimental samples available, this study incorporated the use of virtual samples to enhance the model's accuracy.

The methodology for generating virtual samples involves introducing a±Δi value to each variable of the actual samples, where the error margin Δi was fixed at 2%. Following the L₈(2⁷) orthogonal design, each actual sample gave rise to eight virtual samples.²¹ This approach resulted in a total of 120 virtual samples, with the first set of virtual samples presented in Table 3.

Table 3.

The First Set of Virtual Samples.

Serial no.	Microwave Time (s)	Ionic Liquid Concentration (mol/L)	Liquid-to-Solid Ratio (mL/g)
1	60.12	0.2505	20.04
2	60.12	0.2505	19.96
3	60.12	0.2495	20.04
4	60.12	0.2495	19.96
5	59.88	0.2505	20.04
6	59.88	0.2505	19.96
7	59.88	0.2495	20.04
8	59.88	0.2495	19.96

Utilizing a BP neural network model, which was optimized with genetic algorithms for target enhancement, both virtual samples and empirical data were employed as initial input. The execution of the program yielded experimental outcomes and predicted lycopene concentrations, with the optimized process parameters corresponding to each individual factor condition. All computational analyses were conducted using the Neural Network Toolbox within MATLAB version 9.10.0 (R2021a).

Statistical Analysis

All experiments were conducted in triplicate, and the data are presented as mean values ± standard deviation (SD) (n = 3). Statistical significance was determined at p < 0.05. The effects of different key parameters on lycopene extraction yield were analyzed using one-way analysis of variance (ANOVA), followed by Tukey's test for multiple comparisons to identify significant differences between groups. Data analysis and graph generation were performed using Graph Pad Prism® 7 software (San Diego, CA, USA).

Supplemental Material

sj-docx-1-npx-10.1177_1934578X251330953 - Supplemental material for Optimization of Ionic Liquid Microwave-Assisted Extraction Protocol of Lycopene from Tomato

Supplemental material, sj-docx-1-npx-10.1177_1934578X251330953 for Optimization of Ionic Liquid Microwave-Assisted Extraction Protocol of Lycopene from Tomato by Junkai Wu, Jianping Hu, Yingqian Jiang, Feiyan Zhang and Youxia Huang in Natural Product Communications

Footnotes

Acknowledgements

Special thanks to the reviewers for their valuable comments. In addition, the authors gratefully acknowledge every teacher, classmate, and friend who helped the authors with their experiment and writing.

Statement of Human and Animal Rights

This article does not contain any studies with human or animal subjects.

Statement of Informed Consent

There are no human subjects in this article and informed consent is not applicable.

ORCID iD

Junkai Wu

Ethical Considerations

Ethical Approval is not applicable for this article.

Author contributions/CRediT

Junkai Wu: Conceptualization, Methodology, Formal analysis, Investigation, Writing – original draft. Jianping Hu: Methodology, Investigation, Visualization. Yingqian Jiang: Investigation, Validation. Feiyan Zhang: Investigation, Resources. Youxia Huang: Conceptualization, Data curation, Formal analysis, Validation, Visualization, Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Scientific Research Funds of Quanzhou Medical College under grant number of XJY2412.

Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author

Supplemental Material

Supplemental material for this article is available online.

References

Mukundh Chaithanya

Uma Maheswari

Rajesh Kumar

. Anti-inflammatory and cytotoxic effect of lycopene and raspberry in-situ gel. Bioinformation. 2022;18:645–650. doi:https://doi.org/10.6026/97320630018645

Aboubakr

Farag

Elfadadny

, et al. Antioxidant and anti-apoptotic potency of allicin and lycopene against methotrexate-induced cardiac injury in rats. Environ Sci Pollut Res Int. 2023;30:88724–88733. doi:https://doi.org/10.1007/s11356-023-28686-4

Mozos

Stoian

Caraba

, et al. Lycopene and vascular health. Front Pharmacol. 2018;9:521. doi:https://doi.org/10.3389/fphar.2018.00521

Cui

. Recent technological strategies for enhancing the stability of lycopene in processing and production. Food Chem. 2023;405(Pt A):134799. doi:https://doi.org/10.1016/j.foodchem.2022.134799

Taskiran

Tastemur

. The role of nitric oxide in anticonvulsant effects of lycopene supplementation on pentylenetetrazole-induced epileptic seizures in rats. Exp Brain Res. 2021;239:591–599. doi:https://doi.org/10.1007/s00221-020-06012-5

Vitucci

Amoresano

Nunziato

, et al. Nutritional controlled preparation and administration of different tomato purees indicate increase of beta-carotene and lycopene isoforms, and of antioxidant potential in human blood bioavailability: a pilot study. Nutrients. 2021;13:1336. doi:https://doi.org/10.3390/nu13041336

Luo

Bao

Zhu

. Nutritional and functional insight into novel probiotic lycopene-soy milk by genome edited Bacillus subtilis. Food Chem. 2023;429:136973. doi:https://doi.org/10.1016/j.foodchem.2023.136973

Eslami

Carpentieri

Pataro

Ferrari

. A comprehensive overview of tomato processing by-product valorization by conventional methods versus emerging technologies. Foods. 2022;12:166. doi:https://doi.org/10.3390/foods12010166

Kuvendziev

Lisichkov

Marinkovski

, et al. Valorization of tomato processing by-products: predictive modeling and optimization for ultrasound-assisted lycopene extraction. Ultrason Sonochem. 2024;110:107055. doi:https://doi.org/10.1016/j.ultsonch.2024

10.

Chada

PSN

Santos

Rodrigues

LGG

, et al. Non-conventional techniques for the extraction of antioxidant compounds and lycopene from industrial tomato pomace (Solanum lycopersicum L.) using spouted bed drying as a pre-treatment. Food Chem. 2022;13:100237. doi:https://doi.org/10.1016/j.fochx.2022.100237

11.

Ding

Liu

. Magnetic ionic liquid-based liquid-liquid microextraction followed by ultra-performance liquid chromatography coupled with triple-quadrupole tandem mass spectrometry for simultaneous determination of neurotransmitters in human cerebrospinal fluid and plasma. Talanta. 2023;262:124690. doi:https://doi.org/10.1016/j.talanta.2023.124690

12.

Olalere

Gan

Akintomiwa

Adeyi

. Optimisation of microwave-assisted extraction and functional elucidation of bioactive compounds from Cola nitida pod. Phytochem Anal. 2021;32:850–858. doi.:https://doi.org/10.1002/pca.3030

13.

Ferrara

Beccaria

Cordero

Purcaro

. Microwave-assisted extraction in closed vessel in food analysis. J Sep Sci. 2023;46:e2300390. doi:https://doi.org/10.1002/jssc.202300390

14.

Liu

. Incomplete big data imputation mining algorithm based on BP neural network. J Intell Fuzzy Syst. 2019;37:1–10. doi:https://doi.org/10.3233/JIFS-179278

15.

Ming

Liu

Lei

Huang

Chen

. Rapid determination of the total content of oleanolic acid and ursolic acid in chaenomelis Fructus using near-infrared spectroscopy. Front Plant Sci. 2022;13:978937. doi:https://doi.org/10.3389/fpls.2022.978937

16.

Qin

Zheng

, et al. Prediction of bitterant and sweetener using structure-taste relationship models based on an artificial neural network. Food Res Int. 2022;153:110974. doi:https://doi.org/10.1016/j.foodres.2022.110974

17.

Kim

Bai

Park

. The effects of high-power microwaves on the ultrastructure of Bacillus subtilis. Lett Appl Microbiol. 2008;47:35–40. doi:https://doi.org/10.1111/j.1472-765X.2008.02384.x

18.

Wang

Yang

Peng

, et al. Optimization of ionic liquids-based microwave-assisted hydrolysis of puerarin and daidzein derivatives from Radix Puerariae Lobatae extract. Food Chem. 2018;256:149–155. doi:https://doi.org/10.1016/j.foodchem.2017.12.080

19.

Liu

Zhu

Chen

, et al. Mild acetylation and solubilization of ground whole plant cell walls in EmimAc: a method for solution-state NMR in DMSO-d(6). Anal Chem. 2020;92:13101–13109. doi:https://doi.org/10.1021/acs.analchem.0c02124

20.

Corrêa-Filho

Santos

Brito

Moldão-Martins

Alves

. Stability and in vitro bioaccessibility of microencapsulated tomato (Solanum Lycopersicum L.) pomace extract. Bioengineering(Basel). 2022;9:311. doi:https://doi.org/10.3390/bioengineering9070311

21.

Jiang

Zheng

Yuan

, et al. Study on extraction technology and antioxidant activity of total alkaloids from Hemsleya chinensis based on orthogonal design and BP neural network. Heliyon. 2023;9:e20680. doi:https://doi.org/10.1016/j.heliyon.2023.e20680

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB