Modeling and validation of purification of pharmaceutical compounds via hybrid processing of vacuum membrane distillation.

This study provides an in-depth examination of forecasting the concentration of pharmaceutical compounds utilizing the input features (coordinates) r and z through a range of machine learning models. Purification of pharmaceuticals via vacuum membrane distillation process was carried out and the model was developed for prediction of separation efficiency based on hybrid approach. Dataset was collected from mass transfer analysis of process to obtain concentration distribution in the feed side of membrane distillation and used it for machine learning models. The dataset has undergone preprocessing, which includes outlier detection using the Isolation Forest algorithm. Three regression models were used including polynomial regression (PR), k-nearest neighbors (KNN), and Tweedie regression (TWR). These models were further enhanced using the Bagging ensemble technique to improve prediction accuracy and reduce variance. Hyper-parameter optimization was conducted using the Multi-Verse Optimizer algorithm, which draws inspiration from cosmological concepts. The Bagging-KNN model had the highest predictive accuracy (R2 = 0.99923) on the test set, indicating exceptional precision. The Bagging-PR model displayed satisfactory performance, with a slightly reduced level of accuracy. In contrast, the Bagging-TWR model showcased the least accuracy among the three models. This research illustrates the effectiveness of incorporating bagging and advanced optimization methods for precise and dependable predictive modeling in complex datasets.

Lightweight CNN combined with knowledge distillation for the accurate determination of black tea fermentation degree.

Black tea is the second most common type of tea in China. Fermentation is one of the most critical processes in its production, and it affects the quality of the finished product, whether it is insufficient or excessive. At present, the determination of black tea fermentation degree completely relies on artificial experience. It leads to inconsistent quality of black tea. To solve this problem, we use machine vision technology to distinguish the degree of fermentation of black tea based on images, this paper proposes a lightweight convolutional neural network (CNN) combined with knowledge distillation to discriminate the degree of fermentation of black tea. After comparing 12 kinds of CNN models, taking into account the size of the model and the performance of discrimination, as well as the selection principle of teacher models, Shufflenet_v2_x1.0 is selected as the student model, and Efficientnet_v2 is selected as the teacher model. Then, CrossEntropy Loss is replaced by Focal Loss. Finally, for Distillation Loss ratios of 0.6, 0.7, 0.8, 0.9, Soft Target Knowledge Distillation (ST), Masked Generative Distillation (MGD), Similarity-Preserving Knowledge Distillation (SPKD), and Attention Transfer (AT) four knowledge distillation methods are tested for their performance in distilling knowledge from the Shufflenet_v2_x1.0 model. The results show that the model discrimination performance after distillation is the best when the Distillation Loss ratio is 0.8 and the MGD method is used. This setup effectively improves the discrimination performance without increasing the number of parameters and computation volume. The model's P, R and F1 values reach 0.9208, 0.9190 and 0.9192, respectively. It achieves precise discrimination of the fermentation degree of black tea. This meets the requirements of objective black tea fermentation judgment and provides technical support for the intelligent processing of black tea.

Evaluation of lavender essential oils and by-products using microwave hydrodistillation and conventional hydrodistillation.

This study investigated the impact of two extraction methods, traditional hydrodistillation (TDH) and microwave-assisted hydrodistillation (MAH), on the essential oil yield and chemical profile of Lavandula angustifolia L., as well as the bioactive potential of the resulting wastewater. Essential oil composition was analyzed via GC-MS, revealing similar qualitative and quantitative profiles for both methods, with α-terpinolene and (-)borneol as major constituents. Wastewater analysis via LC-MS/MS and spectrophotometric assays demonstrated the presence of significant total phenolic content (3.29-1.78 mg GAE/g) and 32 individual phenolics (463.1 µg/kg for TDH; 479.33 µg/kg for MAH). These findings suggest that both essential oil and wastewater obtained by either method possess considerable bioactive potential, with the MAH method potentially offering advantages over TDH for essential oil extraction. Further exploration of wastewater applications in various industrial sectors is warranted.

Surface energy-induced anti-wetting and anti-fouling enhancement of Janus membrane for membrane distillation.

Membrane distillation (MD) presents a promising alternative to conventional desalination systems, particularly for the treatment of hypersaline wastewater. However, the large-scale application of MD is hindered by challenges such as membrane wetting, membrane fouling, and low permeate flux. Herein, we proposed an air/liquid interface deposition method to fabricate a Janus membrane, termed the PVDF-PDA/PEI-Si membrane. The membrane featured a nanosieving, superhydrophilic polydopamine/polyethylenimine (PDA/PEI) layer decorated with silica nanoparticles, coupled with a microporous, hydrophobic polyvinylidene fluoride (PVDF) layer. The introduction of a dense PDA/PEI-Si layer featuring high surface energy significantly enhanced the wetting and fouling resistance of the membrane, with a minor effect on the permeate flux. The performance enhancement was particularly evident when hypersaline water containing sodium dodecyl sulfate (SDS) and oily contaminants was used as the feed. The interactions between the membrane and contaminants were calculated using the XDLVO theory and molecular dynamics simulations to elucidate the mechanisms underlying the enhanced anti-wetting and anti-fouling properties, respectively. According to the XDLVO theory, a large energy barrier must be overcome for the SDS to attach onto the PDA/PEI-Si surface. Meanwhile, molecular dynamics simulations confirmed the weak interaction energy between the oily foulants and the PVDF-PDA/PEI-Si membrane due to its high surface energy. This study presents a promising approach for the fabrication of high-performance MD membranes and provides new insights into the mechanisms underlying the enhanced anti-wetting and anti-fouling properties.

Comparative Analysis of Hydrosol Volatile Components of Citrus × Aurantium 'Daidai' and Citrus × Aurantium L. Dried Buds with Different Extraction Processes Using Headspace-Solid-Phase Microextraction with Gas Chromatography-Mass Spectrometry.

This work used headspace solid-phase microextraction with gas chromatography-mass spectrometry (HS-SPME-GC-MS) to analyze the volatile components of hydrosols of Citrus × aurantium 'Daidai' and Citrus × aurantium L. dried buds (CAVAs and CADBs) by immersion and ultrasound-microwave synergistic-assisted steam distillation. The results show that a total of 106 volatiles were detected in hydrosols, mainly alcohols, alkenes, and esters, and the high content components of hydrosols were linalool, α-terpineol, and trans-geraniol. In terms of variety, the total and unique components of CAVA hydrosols were much higher than those of CADB hydrosols; the relative contents of 13 components of CAVA hydrosols were greater than those of CADB hydrosols, with geranyl acetate up to 15-fold; all hydrosols had a citrus, floral, and woody aroma. From the pretreatment, more volatile components were retained in the immersion; the relative contents of linalool and α-terpineol were increased by the ultrasound-microwave procedure; and the ultrasound-microwave procedure was favorable for the stimulation of the aroma of CAVA hydrosols, but it diminished the aroma of the CADB hydrosols. This study provides theoretical support for in-depth exploration based on the medicine food homology properties of CAVA and for improving the utilization rate of waste resources.

Membrane electrolysis distillation for volatile fatty acids extraction from pH-neutral fermented wastewater.

Volatile fatty acids (VFAs) serve as building blocks for a wide range of chemicals, but it is difficult to extract VFAs from pH-neutral wastewater using evaporation methods because of the ionized form. This study presents a new membrane electrolysis distillation (MED) process that extracts VFAs from such fermentation solutions. MED uniquely integrates pH regulation and joule heating to facilitate the efficient evaporation of VFAs. This integration occurs alongside a hydrophobic membrane that ensures effective gas-liquid phase separation. Operating solely on electricity, MED achieved an acid flux rate of 12.03 g/m2/h at 6V. In contrast, the control results without the joule heating or pH swing only obtained a 0.23 g/m2/h and 0.32 g/m2/h flux, respectively. In addition, a physicochemical model was developed to assess the impacts of temperature on membrane surface pH. This system enhances resource recovery from waste streams and helps achieve a circular carbon economy.

Cross-modal knowledge distillation for continuous sign language recognition.

Continuous Sign Language Recognition (CSLR) is a task which converts a sign language video into a gloss sequence. The existing deep learning based sign language recognition methods usually rely on large-scale training data and rich supervised information. However, current sign language datasets are limited, and they are only annotated at sentence-level rather than frame-level. Inadequate supervision of sign language data poses a serious challenge for sign language recognition, which may result in insufficient training of sign language recognition models. To address above problems, we propose a cross-modal knowledge distillation method for continuous sign language recognition, which contains two teacher models and one student model. One of the teacher models is the Sign2Text dialogue teacher model, which takes a sign language video and a dialogue sentence as input and outputs the sign language recognition result. The other teacher model is the Text2Gloss translation teacher model, which targets to translate a text sentence into a gloss sequence. Both teacher models can provide information-rich soft labels to assist the training of the student model, which is a general sign language recognition model. We conduct extensive experiments on multiple commonly used sign language datasets, i.e., PHOENIX 2014T, CSL-Daily and QSL, the results show that the proposed cross-modal knowledge distillation method can effectively improve the sign language recognition accuracy by transferring multi-modal information from teacher models to the student model. Code is available at https://github.com/glq-1992/cross-modal-knowledge-distillation_new.

Integrated supercritical fluid extraction of essential oils.

Supercritical fluid extraction (SFE) stands out as an incredibly efficient, environmentally conscious, and fast method for obtaining essential oils (EOs) from plants. These EOs are abundant in aromatic compounds that play a crucial role in various industries such as food, fragrances, cosmetics, perfumery, pharmaceuticals, and healthcare. While there is a wealth of existing literature on using supercritical fluids for extracting plant essential oils, there's still much to explore in terms of combining different techniques to enhance the SFE process. This comprehensive review presents a sophisticated framework that merges SFE with EO extraction methods. This inclusive categorization encompasses a range of methods, including the integration of pressurized liquid processes, ultrasound assistance, steam distillation integration, microfluidic techniques, enzyme integration, adsorbent facilitation, supercritical antisolvent treatments, molecular distillation, microwave assistance, milling process and mechanical pressing integration. Throughout this in-depth exploration, we not only elucidate these combined techniques but also engage in a thoughtful discussion about the challenges they entail and the array of opportunities they offer within the realm of SFE for EOs. By dissecting these complexities, our objective is to tackle the current challenges associated with enhancing SFE for commercial purposes. This endeavor will not only streamline the production of premium-grade essential oils with improved safety measures but also pave the way for novel applications in various fields.

Improving membrane distillation performance by Fe(II) activated sodium percarbonate oxidation during the treatment of shale gas produced water.

Membrane distillation (MD) offers promise for recycling shale gas produced water (SGPW), while membrane fouling is still a major obstacle in standalone MD. Herein, sodium percarbonate (SPC) oxidation was proposed as MD pretreatment, and the performance of the single MD, SPC-MD hybrid process and Fe(II)/SPC-MD hybrid process for SGPW treatment were systematically evaluated. Results showed that compared to raw SGPW, the application of SPC and Fe(II)/SPC led to the decrease of the fluorescent organics by 28.54 % and 54.52 %, respectively. The hydrophobic fraction decreased from 52.75 % in raw SGPW to 37.70 % and 27.20 % for SPC and Fe(II)/SPC, respectively, and the MD normalized flux increased from 0.19 in treating raw SGPW to 0.65 and 0.81, respectively. The superiority of SPC oxidation in reducing the deposited membrane foulants and restoring membrane properties was further confirmed through scanning electron microscopy observation, attenuated total reflection fourier transform infrared, water contact angle and surface tension analyses of fouled membranes. Correlation analysis revealed that hydrophobic/hydrophilic matters and fluorescent organics in SGPW took a crucial role in MD fouling. The mechanism of MD fouling mitigation by Fe(II)/SPC oxidation was attributed to the decrease in concentrations and hydrophobicity of organic by synergistic oxidation, coagulation and adsorption.

Influence of photothermal nanomaterials localization within the electrospun membrane structure on purification of saline oily wastewater based on photothermal vacuum membrane distillation.

Today, synergistic combination of special nanomaterials (NMs) and electrospinning technique has emerged as a promising strategy to address both water scarcity and energy concerns through the development of photothermal membranes for wastewater purification and desalination. This work was organized to provide a new perspective on membrane design for photothermal vacuum membrane distillation (PVMD) through optimizing membrane performance by varying the localization of photothermal NMs. Poly(vinylidene fluoride) omniphobic photothermal membranes were prepared by localizing graphene oxide nanosheets (GO NSh) (1) on the surface (0.2 wt%), (2) within the nanofibers structure (10 wt%) or (3) in both positions. Considering the case 1, after 7 min exposure to the 1 sun intensity light, the highest temperature (∼93.5 °C) was recorded, which is assigned to the accessibility of GO NSh upon light exposure. The case 3 yielded to a small reduction in surface temperature (∼90.4 °C) compared to the case 1, indicating no need to localize NMs within the nanofibers structure when they are localized on the surface. The other extreme belonged to the case 2 with the lowest temperature of ∼71.3 °C, which is consistent with the less accessibility of GO NSh during irradiation. It was demonstrated that the accessibility of photothermal NMs plays more pronounced role in the membrane surface temperature compared to the light trapping. However, benefiting from higher surface temperature during PVMD due to enhanced accessibility of photothermal NMs is balanced out by decrease in the permeate flux (case 1: 1.51 kg/m2 h and case 2: 1.83 kg/m2 h) due to blocking some membrane surface pores by the binder. A trend similar to that for flux was also followed by the efficiency. Additionally, no change in rejection was observed for different GO NSh localizations.

Innovative hyper-thermophilic aerobic submerged membrane distillation bioreactor for wastewater reclamation.

For the first time, a hyper-thermophilic aerobic (>60 °C) bioreactor has been integrated with direct submerged membrane distillation (MD), highlighting its potential as an advanced wastewater treatment solution. The hyper-thermophilic aerobic bioreactor, operating up to 65 °C, is tailored for high organic removal, while MD efficiently produces clean water. Throughout the study, high removal rates of 99.5% for organic matter, 96.4% for ammonia, and 100% for phosphorus underscored the impressive adaptability of microorganisms to challenging hyper-thermophilic conditions and a successful combination with the MD process. Despite the extreme temperatures and substantial salinity accumulation reaching up to 12,532 µS/cm, the biomass of microorganisms increased by 1.6 times over a 92-day period, representing their remarkable resilience. The distillation flux ranged from 6.15 LMH to 8.25 LMH, benefiting from the temperature gradient in the hyper-thermophilic setting and the design of the tubular submerged MD membrane module. The system also excels in pH control, utilizing fewer alkali and nutritional resources than conventional systems. Meiothermus, Firmicutes, and Bacteroidetes, the three dominant species, played a crucial role, showcasing their significance in adapting to high salinity and decomposing organic matter.

Ammonia recovery via direct contact membrane distillation: Modeling and performance optimization.

Ammonia recovery from wastewater has positive environmental benefits, avoiding eutrophication and reducing production energy consumption, which is one of the most effective ways to manage nutrients in wastewater. Specifically, ammonia recovery by membrane distillation has been gradually adopted due to its excellent separation properties for volatile substances. However, the global optimization of direct contact membrane distillation (DCMD) operating parameters to maximize ammonia recovery efficiency (ARE) has not been attempted. In this work, three key operating factors affecting ammonia recovery, i.e., feed ammonia concentration, feed pH, and DCMD running time, were identified from eight factors, by a two-level Plackett-Burman Design (PBD). Subsequently, Box-Behnken design (BBD) under the response surface methodology (RSM) was used to model and optimize the significant operating parameters affecting the recovery of ammonia though DCMD identified by PBD and statistically verified by analysis of variance (ANOVA). Results showed that the model had a high coefficient of determination value (R2 = 0.99), and the interaction between NH4Cl concentration and feed pH had a significant effect on ARE. The optimal operating parameters of DCMD as follows: NH4Cl concentration of 0.46 g/L, feed pH of 10.6, DCMD running time of 11.3 h, and the maximum value of ARE was 98.46%. Under the optimized conditions, ARE reached up to 98.72%, which matched the predicted value and verified the validity and reliability of the model for the optimization of ammonia recovery by DCMD process.

Comparison of quantity, quality and antibacterial activity of essential oil Mentha longifolia (L.) L. under different traditional and modern extraction methods.

Extraction is the first and most important step in obtaining the effective ingredients of medicinal plants. Mentha longifolia (L.) L. is of considerable economic importance as a natural raw material for the food and pharmaceutical industries. Since the effect of different extraction methods (traditional and modern methods) on the quantity, quality and antimicrobial activity of the essential oil of this plant has not been done simultaneously; the present study was designed for the first time with the aim of identifying the best extraction method in terms of these features. For this purpose, extracting the essential oil of M. longifolia with the methods of hydrodistillation with Clevenger device (HDC), steam distillation with Kaiser device (SDK), simultaneous distillation with a solvent (SDE), hydrodistillation with microwave device (HDM), pretreatment of ultrasonic waves and Clevenger (U+HDC) and supercritical fluid (SF) were performed. Chemical compounds were identified by gas chromatography coupled with mass spectrometer (GC-MS). Antimicrobial activity of essential oils against various clinical microbial strains was evaluated by agar diffusion method and determination of the minimum inhibitory concentration and minimum bactericidal concentration (MIC and MBC). The results showed that the highest and lowest yields of M. longifolia leaf essential oil belonged to HDC (1.6083%) and HDM (0.3416%). The highest number of compounds belonged to SDK essential oil and was equal to 72 compounds (with a relative percentage of 87.13%) and the lowest number of compounds was related to the SF essential oil sample (7 compounds with a relative percentage of 100%). Piperitenone (25.2-41.38%), piperitenone oxide (22.02-0%), pulegone (10.81-0%) and 1,8-cineole (5-35.0%) are the dominant and main components of M. longifolia essential oil were subjected to different extraction methods. Antimicrobial activity results showed that the lowest MIC value belonged to essential oils extracted by HDM, SDK, SDE and U+HDC methods with a value of 1000 µg/mL was observed against Gram-negative bacteria Shigella dysenteriae, which was 5 times weaker than rifampin and 7 times weaker than gentamicin. Therefore, it can be concluded that in terms of efficiency of the HDC method, in terms of the percentage of compounds of the HDM method, and in terms of microbial activity, the SDK, HDM and U+HDC methods performed better.

Preparation of High-Purity Docosahexaenoic Acid Ethyl Ester from Algal Oil through Enzymatic Ethanolysis.

Docosahexaenoic acid plays a crucial role in infant brain function, and the market demand of high-purity docosahexaenoic acid is continuously increasing. The availability of docosahexaenoic acid in natural fish oil is limited, prompting the exploration of alternative sources like microalgae. For algal oil, enzymatic ethanolysis is preferred to chemical methods because the former is milder and can avoid docosahexaenoic acid oxidation. However, enzymatic methods have generally low yield due to the poor substrate-specificity of lipase to long-chain polyunsaturated fatty acids, affecting the yield and purity of docosahexaenoic acid. Therefore, we developed an efficient process to produce high-purity docosahexaenoic acid ethyl ester from algal oil, by screening lipases, optimizing enzymatic ethanolysis and applying molecular distillation. Lipase UM1 was the best lipase to produce ethyl ester from algal oil with the highest ethyl ester yield (95.41%). Meanwhile, it was a catalyst for the reaction of long-chain polyunsaturated fatty acids with ethanol. The fatty acid docosahexaenoic acid conversion rates exceeded 90%. After molecular distillation, a final product containing 96.52% ethyl ester was obtained with a docosahexaenoic acid content up to 80.11%. Our findings provide an highly effective enzymatic method for the production of high-purity docosahexaenoic acid ethyl esters, with potential commercial applications.

Membrane fouling analysis of air-gap membrane distillation (AGMD) for recovery of water and removal of antibiotics from a model wastewater containing antibiotics and humic acid.

The study investigates the efficiency of air-gap membrane distillation (AGMD) in water recovery and antibiotics removal from wastewater, focusing on high-concentration scenarios. Experimental findings reveal enhanced membrane performance with increasing the feed temperature, resulting in vapor permeate fluxes of up to 5 kg/m2.h at higher temperatures. Despite experiencing flux reduction caused by fouling from humic acid (HA) in the feed antibiotics solution, the antibiotics consistently maintain near-complete rejection rates (>99%) over 48 h. The foulant on the membrane surface was illustrated by SEM imaging. To know the temperature polarization and the fouling resistance, mathematical modeling was used, and it validates experimental results, elucidating temperature polarization effects and mass transfer coefficients. An increase in feed flow rates reduced thermal boundary layers, enhancing heat flux. Higher temperatures reduced HA fouling resistance. Therefore, AGMD proves effective in water recovery and antibiotics removal, with mathematical models aiding fouling understanding for future research and detailed computational fluid dynamics (CFD) models.

Bioenzyme mediated hydrodistillation (BMHD) for extraction of mint oil from mentha leaves: improvement in yield and menthol content.

The present study optimized pre-treatment conditions for bioenzyme-mediated hydro-distillation (BMHD) for extraction of mint oil from mentha leaves and the results were compared with those of traditional hydro-distillation (HD) method using response surface methodology. The bio-enzymes produced from moringa leaves had maximum pectinase activity (287.04 µg of sugar/min/ml) followed by xylanase (87.78 µg of sugar/min/ml) while endoglucanase, exoglucanase and amylase activities were comparatively low. The optimized conditions for HD were 69.08 temperature for 173.70 min with water:sample of 10.0. The optimized conditions for enzyme pre-treatment of mentha leaves by BMHD were enzyme concentration of 8%, for a period of 120 min at an incubation period of 40 â„ƒ. The yield (%) and menthol content (%) of the oil at optimized conditions by HD were 1.55 ml/100 g of sample and 56.40% menthol content, respectively, and for BMHD the yield and menthol content (%) of the oil at optimized conditions were 3.69% and 72.80%, respectively. It was found that BMHD leads to a 130% increase in the yield of mint oil and a 10% increase in menthol content as compared to the HD method. No significant difference in physical parameters was observed in mint oil extracted via both methods. Therefore, BMHD is a cost-effective and sustainable approach having an edge over the HD method without compromising the quality and could be a viable approach for commercial purposes.

Profiling of seized Cannabis sativa L. flowering tops by means of microwave-assisted hydro distillation and gas chromatography analyses.

This research aimed to support police forces in their battle against illicit drug trafficking by means of a multi-technique approach, based on gas chromatography. In detail, this study was focused on the profiling of volatile substances in narcotic Cannabis sativa L. flowering tops. For this purpose, the Scientific Investigation Department, RIS Carabinieri of Messina, provided 25 seized samples of Cannabis sativa L. The content of Δ9-tetrahydrocannabinol (THC), useful to classify cannabis plant as hemp (≤ 0.2 %) or as marijuana (> 0.2 %), was investigated. Essential oils of illicit drug samples were extracted using a microwave-assisted hydro-distillation (MAHD) system; GC-MS and GC-FID analytical techniques were used for the characterization of the terpenes and terpenoids fingerprint. Furthermore, the enantiomeric and carbon isotopic ratios of selected chiral compounds were investigated using a heart-cutting multidimensional GC (MDGC) approach. The latter exploited a combination of an apolar column in the first dimension, and a chiral cyclodextrin-based column in the second one, prior to parallel isotope-ratio mass spectrometry (C-IRMS) and MS detection. Finally, all the data were gathered into a statistical model, to demonstrate the existence of useful parameters to be used for the classification of seized samples.

Progress in membrane distillation processes for dye wastewater treatment: A review.

Textile and cosmetic industries generate large amounts of dye effluents requiring treatment before discharge. This wastewater contains high levels of reactive dyes, low to none-biodegradable materials and chemical residues. Technically, dye wastewater is characterised by high chemical and biological oxygen demand. Biological, physical and pressure-driven membrane processes have been extensively used in textile wastewater treatment plants. However, these technologies are characterised by process complexity and are often costly. Also, process efficiency is not achieved in cost-effective biochemical and physical treatment processes. Membrane distillation (MD) emerged as a promising technology harnessing challenges faced by pressure-driven membrane processes. To ensure high cost-effectiveness, the MD can be operated by solar energy or low-grade waste heat. Herein, the MD purification of dye wastewater is comprehensively and yet concisely discussed. This involved research advancement in MD processes towards removal of dyes from industrial effluents. Also, challenges faced by this process with a specific focus on fouling are reviewed. Current literature mainly tested MD setups in the laboratory scale suggesting a deep need of further optimization of membrane and module designs in near future, especially for textile wastewater treatment. There is a need to deliver customized high-porosity hydrophobic membrane design with the appropriate thickness and module configuration to reduce concentration and temperature polarization (CP and TP). Also, energy loss should be minimized while increasing dye rejection and permeate flux. Although laboratory experiments remain pivotal in optimizing the MD process for treating dye wastewater, the nature of their time intensity poses a challenge. Given the multitude of parameters involved in MD process optimization, artificial intelligence (AI) methodologies present a promising avenue for assistance. Thus, AI-driven algorithms have the potential to enhance overall process efficiency, cutting down on time, fine-tuning parameters, and driving cost reductions. However, achieving an optimal balance between efficiency enhancements and financial outlays is a complex process. Finally, this paper suggests a research direction for the development of effective synthetic and natural dye removal from industrially discharged wastewater.

Contrasting mixed scaling patterns and mechanisms of nanofiltration and membrane distillation.

Oriented towards the pressing needs for hypersaline wastewater desalination and zero liquid discharge (ZLD), the contrasting mixed scaling of thermal-driven vacuum membrane distillation (VMD) and pressure-driven nanofiltration (NF) were investigated in this work. Bulk crystallization was the main mechanism in VMD due to the high salinity and temperature, but the time-independent resistance by the adsorption of silicate and organic matter dominated the initial scaling process. Surface crystallization and the consequent pore-blocking were the main scaling mechanisms in NF, with the high permeate drag force, hydraulic pressure, and cross-flow rate resulting in the dense scaling layer mainly composed of magnesium-silica hydrate (MSH). Silicate enhanced NF scaling with a 75% higher initial flux decline rate attributed to the MSH formation and compression, but delayed bulk crystallization in VMD. Organic matter presented an anti-scaling effect by delaying bulk crystallization in both VMD and NF, but specifically promoted CaCO3 scaling in NF. Furthermore, the incipient scaling was intensified as silicate and organic matter coexisted. The scaling mechanism shifted from surface to bulk crystallization due to the membrane concentration in both VMD and NF. This work fills the research gaps on mixed scaling mechanisms in different membrane processes, which offers insights for scaling mitigation and thereby supports the application of ZLD.

Prototype-based sample-weighted distillation unified framework adapted to missing modality sentiment analysis.

Missing modality sentiment analysis is a prevalent and challenging issue in real life. Furthermore, the heterogeneity of multimodality often leads to an imbalance in optimization when attempting to optimize the same objective across all modalities in multimodal networks. Previous works have consistently overlooked the optimization imbalance of the network in cases when modalities are absent. This paper presents a Prototype-Based Sample-Weighted Distillation Unified Framework Adapted to Missing Modality Sentiment Analysis (PSWD). Specifically, it fuses features with a more efficient transformer-based cross-modal hierarchical cyclic fusion module. Subsequently, we propose two strategies, namely sample-weighted distillation and prototype regularization network, to address the issues of missing modality and optimization imbalance. The sample-weighted distillation strategy assigns higher weights to samples that are located closer to class boundaries. This facilitates the obtaining of complete knowledge by the student network from the teacher's network. The prototype regularization network calculates a balanced metric for each modality, which adaptively adjusts the gradient based on the prototype cross-entropy loss. Unlike conventional approaches, PSWD not only connects the sentiment analysis study in the missing modality to the full modality, but the proposed prototype regularization network is not reliant on the network structure and can be expanded to more multimodal studies. Massive experiments conducted on IEMOCAP and MSP-IMPROV show that our method achieves the best results compared to the latest baseline methods, which demonstrates its value for application in sentiment analysis.