Emerging applications of EEM-PARAFAC for water treatment: a concise review

Emerging applications of fluorescence excitation-emission matrix with machine learning for water quality monitoring: A systematic review

Fluorescence excitation-emission matrix (FEEM) spectroscopy is increasingly utilized in water quality monitoring due to its rapid, sensitive, and non-destructive measurement capabilities. The integration of machine learning (ML) techniques with FEEM offers a powerful approach to enhance data interpretation and improve monitoring efficiency. This review systematically examines the application of ML-FEEM in urban water systems across three primary tasks of ML: classification, regression, and pattern recognition. Contributed by the effectiveness of ML in nonlinear and high dimensional data analysis, ML-FEEM achieved superior accuracy and efficiency in pollutant qualification and quantification. The fluorescence features extracted through ML are more representative and hold potential for generating new FEEM samples. Additionally, the rich visualization capabilities of ML-FEEM facilitate the exploration of the migration and transformation of dissolved organic matter in water. This review underscores the importance of leveraging the latest ML advancements to uncover hidden information within FEEM data, and advocates for the use of pattern recognition methods, represented by self-organizing map, to further elucidate the behavior of pollutants in aquatic environments. Despite notable advancements, several issues require careful consideration, including the portable or online setups for FEEM collection, the standardized pretreatment processes for FEEM analysis, and the smart feedback of long-term FEEM governance.

Zirconium-embedded ceramic membrane catalyzed moderate ozonation: dual-function synergy for simultaneous control of algal odorants and membrane fouling in water treatment

The integration of ozone with ceramic membrane offers a promising approach for treating algae-laden water but presents challenges in balancing oxidation efficacy with the preservation of cell integrity. In this study, catalytic ceramic membrane system embedded with zirconium (Zr) developed achieved 58.4 % removal of 2-methylisoborneol and 68.2 % removal of geosmin through dehydration and ring opening via hydroxyl radical-mediated degradation pathways generated in situ on the membrane under optimal ozone dosage. It is worth noting that the mild ozone concentration increased the hydrophobic interaction energy between algal cells and the filter cake layer from -13.7 mJ/m2 to -0.3 mJ/m2, thereby effectively reducing the deposition of pollutants on the membrane. By controlling oxidative intensity, the ceramic membrane's reversible and irreversible resistances were decreased by 86.7 % and 80.1 %, respectively, while maintaining >95 % algal cell integrity. This study establishes a mild catalytic oxidation paradigm for ceramic membrane-based algae-laden water treatment, achieving simultaneous degradation of odorants and membrane fouling control.

Real-time wastewater quality monitoring by fluorescence sensors: Validation for COD and CEC monitoring and implication for carbon footprint reduction

This study investigated the applicability of a protein-like fluorescence sensor for wastewater quality monitoring. Several wastewater matrices, including raw, primary, secondary and tertiary effluents from three different wastewater treatment plants were used. Furthermore, the sensor was tested for the monitoring of quaternary effluent in a pilot scale plant installed downstream of a water reuse facility. The pilot plant involved advanced oxidation processes (AOPs) and granular activated carbon (GAC) adsorption. Corrections on excitation/emission matrices (EEMs), including Inner Filter Effect (IFE) and scattering, showed no effect on linear correlation (R2=0.99) between sensor measurement and either raw or corrected benchtop protein-like fluorescence data, suggesting that for this application the signal from the sensor might be interpreted without the need for further adjustments. Furthermore, the use of quenched, diluted and filtered samples did not affect such correlations. Overall, the fluorescence sensor showed a very high capability to monitor a wide range of wastewater matrices, including raw, primary, secondary, tertiary, and quaternary effluents, providing fast information on the efficiency of the processes. The protein-like fluorescence monitoring by the real-time sensor was validated online through 9 days of 24-hour continuous monitoring of tertiary wastewater effluents. The employed fluorescence sensor was validated for monitoring the removal of contaminants of emerging concern (CEC), including a wide range of pharmaceuticals, in different AOP systems (ozone and UV based). In view of the results reported in this study, possible environmental implications for the reduction of the carbon footprint have emerged: the use of fluorescence sensors may contribute to the optimization of processes and the reduction of secondary pollution.

Effects of algal organic matters on microporous ceramic emitters clogging in agricultural water distribution systems: Experiment and molecular simulation investigations

The mechanism by which algal organic matter (AOM) affects the clogging of ceramic emitters remains unclear, which partially reduces the operational life of agricultural water distribution systems. This paper systematically investigated the clogging phenomenon of ceramic emitters under three different AOM concentrations. The results of irrigation tests revealed that the AOM significantly affects the degree of clogging of ceramic emitters, with higher AOM concentrations leading to faster flow reduction. By analyzing the original irrigation water and effluent and characterizing the clogged emitter surface, it was demonstrated that AOM was intercepted by the ceramic emitter, forming a dense biofilm. Infrared spectroscopy analysis revealed that polysaccharides and humic substances were the main clogging components. The clogging kinetics showed that as the AOM concentration increased, the clogging of the filter cake layer gradually become dominant. Further, the mechanism of interaction between AOM and silica ceramic emitters was explored from a microscopic perspective using molecular dynamics (MD) simulation with bovine serum albumin (BSA), sodium alginate (SA), and humic acid (HA) as model clogging substances in AOM. The simulation results indicated a strong interaction between AOM molecules and silica molecules dominated by electrostatic attraction, with the strength of the interaction as SA > HA > BSA. It was hypothesized that early clogging was mainly formed by polysaccharides and humic substances combining with silica molecules, while BSA was retained later by combining with organics on the clogging layer or through size exclusion. This study provides insights into bio-clogging in microporous ceramic emitters and may offer a theoretical basis for developing measures to control emitter clogging.

Differentiating Nylon Samples with Visually Indistinguishable Fluorescence Using Principal Component Analysis and Common Dimension Partial Least Squares Linear Discriminant Analysis with Synchronous Fluorescence Spectroscopy

Fluorescence spectroscopy is an attractive candidate for analyzing samples of nylon. Impurities within the polymers formed during the synthesis and processing of nylons give rise to the observed fluorescence, allowing for nylons to be analyzed based on their impurities. Nylons from the same source are expected to display similar fluorescence profiles, and nylons with different fluorescence are expected to be from different sources. This paper investigates an important case where different nylons displayed similar fluorescence, preventing easy discrimination. Samples of Nylon 6 and Nylon 6/12 had visually indistinguishable excitation-emission matrices (EEM), excitation spectra, fluorescence spectra, and synchronous fluorescence spectra at larger Δλ. By collecting synchronous fluorescence spectra at smaller Δλ, additional features in the fluorescence profiles were identified that allowed for some discrimination between the two nylons. Combining the EEM and synchronous fluorescence data with chemometric algorithms provided a clearer differentiation between the two nylons. parallel factor analysis, principal component analysis, and common dimension partial least squares (ComDim-PLS) showed two distinct clusters in the data, with ComDim-PLS providing the greatest distinction between the clusters. The loadings revealed the variables of interest to the ComDim-PLS were the 400 nm and 335 nm bands for all synchronous fluorescence spectra, the 460 nm and 310 nm bands for the Δλ = 20 nm and Δλ = 30 nm synchronous fluorescence spectra, and the 440 nm band for the Δλ = 20 nm synchronous fluorescence spectra. The linear discriminant analysis performed with the PLS data yielded a classification accuracy of 95% with the EEM data and 100% with the synchronous fluorescence data, displaying the power of this technique to differentiate two different nylons with visually indistinguishable fluorescence spectra.

Spectral fusion-based machine learning classifiers for discriminating membrane breakage in multiple scenarios

Membrane breakage can lead to filtration failure, which allows harmful substances to enter the effluent, posing potential hazards to human health and the environment. This study is an innovative combination of fluorescence and ultraviolet-visible (UV-Vis) spectroscopy to identify membrane breakage. It aims to unravel more comprehensive information, improve detection sensitivity and selectivity, and enable real-time monitoring capabilities. Fluorescence and UV-Vis data are extracted through variance partitioning analysis (VPA) and integrated through a decision tree algorithm to form a superior system with enhanced discrimination capabilities. VPA improves discrimination efficiency by extracting key information from spectral data and eliminating redundancy. The decision tree algorithm, on the other hand, can process large amounts of data simultaneously. In addition, the method has a wide range of applications and can be used in various scenarios accurately. The scenarios include domestic sewage, micropollutant water, aquaculture wastewater, and secondary treated sewage. The experimental results validate the application of machine learning classifiers in membrane breakage detection with an accuracy rate of 96.8 % to 97.4 %.

Light-Emitting-Diode-Induced Fluorescence from Organic Dyes for Application in Excitation-Emission Fluorescence Spectroscopy for Food System Analysis

An experimental study is presented on the possibility of using the fluorescence from organic dyes as a broadband light source together with a monochromator for applications in excitation-emission matrix (EEM) fluorescence spectroscopy. A high-power single-chip light-emitting diode (LED) was chosen as an excitation source with a central output wavelength at 365 nm to excite a fluorescent solution of Coumarin 1 dye dissolved in ethanol. Two excitation configurations were investigated: direct excitation from the LED and excitation through an optical-fiber-coupled LED. A Czerny-Turner monochromator with a diffraction grating was used for the spectral tuning of the fluorescence. A simple method was investigated for increasing the efficiency of the excitation as well as the fluorescence signal collection by using a diffuse reflector composed of barium sulfate (BaSO4) and polyvinyl alcohol (PVA). As research objects, extra-virgin olive oil (EVOO), Coumarin 6 dye, and Perylene, a polycyclic aromatic hydrocarbon (PAH), were used. The results showed that the light-emitting-diode-induced fluorescence was sufficient to cover the losses on the optical path to the monochromator output, where a detectable signal could be obtained. The obtained results reveal the practical possibility of applying the fluorescence from dyes as a light source for food system analysis by EEM fluorescence spectroscopy.

Deciphering DOM-metal binding using EEM-PARAFAC: Mechanisms, challenges, and perspectives

Dissolved organic matter (DOM) is a pivotal component of the biogeochemical cycles and can combine with metal ions through chelation or complexation. Understanding this process is crucial for tracing metal solubility, mobility, and bioavailability. Fluorescence excitation emission matrix (EEM) and parallel factor analysis (PARAFAC) has emerged as a popular tool in deciphering DOM-metal interactions. In this review, we primarily discuss the advantages of EEM-PARAFAC compared with other algorithms and its main limitations in studying DOM-metal binding, including restrictions in spectral considerations, mathematical assumptions, and experimental procedures, as well as how to overcome these constraints and shortcomings. We summarize the principles of EEM to uncover DOM-metal association, including why fluorescence gets quenched and some potential mechanisms that affect the accuracy of fluorescence quenching. Lastly, we review some significant and innovative research, including the application of 2D-COS in DOM-metal binding analysis, hoping to provide a fresh perspective for possible future hotspots of study. We argue the expansion of EEM applications to a broader range of areas related to natural organic matter. This extension would facilitate our exploration of the mobility and fate of metals in the environment.

Using multilinear regressions developed from excitation-emission matrices to estimate the wastewater content in urban streams impacted by sanitary sewer leaks and overflows

Failing sewer infrastructure introduces unknown quantities of raw wastewater into urban streams, raising human and ecological health concerns. To address this problem, we developed multilinear regressions that relate fluorescent dissolved organic matter to wastewater content. The models were constructed with the area-normalized regional volumes of excitation-emission matrices measured for mixtures of deionized water, surface water from a wastewater-impacted stream, wastewater from a sanitary sewer adjacent to the stream, and Suwannee River natural organic matter. The best performing multilinear regression had a standard error of 0.55 % wastewater. A matrix-matched calibration was used to internally validate the approach and confirm the wastewater content of select samples. The multilinear model was externally validated through (i) comparison to concentrations of contaminants of emerging concern in surface water and wastewater and (ii) extension to samples from previous campaigns that employed alternative wastewater indicators. Using the validated model, we estimated an average wastewater content of 2.4 ± 4.0 % in 165 samples collected from 14 locations in the Gwynns Falls watershed (USA) between April 2019 and April 2023. The maximum wastewater content was 35 % at a site where sanitary sewer leaks and overflows have been previously documented. The reported approach represents a cost-effective and scalable technique to estimate wastewater content in urban streams through analysis of fluorescent dissolved organic matter.

Fluorescent Dissolved Organic Matter Components as Surrogates for Disinfection Byproduct Formation in Drinking Water: A Critical Review

Disinfection byproduct (DBP) formation, prediction, and minimization are critical challenges facing the drinking water treatment industry worldwide where chemical disinfection is required to inactivate pathogenic microorganisms. Fluorescence excitation-emission matrices-parallel factor analysis (EEM-PARAFAC) is used to characterize and quantify fluorescent dissolved organic matter (FDOM) components in aquatic systems and may offer considerable promise as a low-cost optical surrogate for DBP formation in treated drinking waters. However, the global utility of this approach for quantification and prediction of specific DBP classes or species has not been widely explored to date. Hence, this critical review aims to elucidate recurring empirical relationships between common environmental fluorophores (identified by PARAFAC) and DBP concentrations produced during water disinfection. From 45 selected peer-reviewed articles, 218 statistically significant linear relationships (R2 ≥ 0.5) with one or more DBP classes or species were established. Trihalomethanes (THMs) and haloacetic acids (HAAs), as key regulated classes, were extensively investigated and exhibited strong, recurrent relationships with ubiquitous humic/fulvic-like FDOM components, highlighting their potential as surrogates for carbonaceous DBP formation. Conversely, observed relationships between nitrogenous DBP classes, such as haloacetonitriles (HANs), halonitromethanes (HNMs), and N-nitrosamines (NAs), and PARAFAC fluorophores were more ambiguous, but preferential relationships with protein-like components in the case of algal/microbial FDOM sources were noted. This review highlights the challenges of transposing site-specific or FDOM source-specific empirical relationships between PARAFAC component and DBP formation potential to a global model.

Recovering fluorescence spectra hidden by scattering signal: In search of the best smoother

Interpolation of the scattering areas in fluorescence excitation-emission matrices is a useful preprocessing method in fluorescence spectroscopy and data modelling. Commonly used row-by-row interpolation using piecewise cubic Hermite interpolating polynomials smoother (PCHIP), however, frequently leads to artifacts because it does not make any use of the information in the other dimension. We have suggested the way of constructing the penalty matrices for Whittaker smoothing that removed one of the main sources of difference between the axis of multiparametric signal - the grid step size - thus making it possible to reduce the number of parameters to optimize. We have compared Whittaker smoother with various surface interpolation methods, including LOESS, Kriging, multilevel B-spline approximation, and PCHIP for the purpose of data preprocessing before PARAFAC modelling of fluorescence signal on a model dataset. The two leaders by signal reconstruction and reconstruction of PARAFAC loadings are LOESS and Whittaker smoothing; the latter is additionally shown to have fundamentally interpretable parameters, which are easier to optimise for the whole dataset. Moreover, Whittaker keeps the shape of the signal and is resistant to variations in data structure and noise level that is very important in numerous applications. We also tested smoothers performance for Åsmund Rinnan fluorescence dataset and the high performance of Whittaker was proved. We can recommend the Whittaker smoothing as a perfect tool for interpolation of scattering areas in florescence spectra of seawaters with low signal-to-noise ratio.

Interaction between biochar-dissolved organic matter and chlorophenols during biochar adsorption

Biochar (BC) has been widely applied in the remediation of chlorophenols (CPs) from contaminated sites in which the role and mechanisms of BC dissolved organic matter (BDOM), as a crucial component of BC, with CPs are largely unknown and thus need to be investigated. In this study, DOM was derived from peanut hulls (PDOM) and corn stalks (CDOM) as BC sources, and the interactions between PDOM/CDOM and 2,4,6-trichlorophenol (TCP) were analysed using excitation-emission matrix spectroscopy (EEM) in combination with multiple models. EEM combined with fluorescence region integration (EEM-FRI) indicated that humic-like materials were the major materials of both PDOM and CDOM (percentage fluorescence response R_i,n > 60%), and CDOM contained more protein- and fulvic-like materials than PDOM. Based on EEM in combination with parallel factor analysis (EEM-PARAFAC), 4 components were obtained, and the percentage decrease in maximum fluorescence intensities (F_max) showed that the main components interacting with TCP in PDOM/CDOM were protein- and fulvic-like components (> 25%). Moreover, the modified Stern-Volmer model was used to calculate the stability constants (Log K_TCP) of PDOM/CDOM and TCP for the first time, and the mechanism of static quenching was dominant for interacting with TCP in PDOM (Log K_TCP: 4.36-4.65) and CDOM (Log K_TCP: 3.53-4.73). Furthermore, the sequential TCP binding of fluorescent components in BDOM generally followed the order of protein-like → short-wavelength fulvic-like → long-wavelength fulvic-like → humic-like components. These findings will provide a basis for screening biochar as a functional material for CP remediation applications and for understanding the environmental chemical behaviour of leached DOM during biochar application.

Effect of ferrous-activated calcium peroxide oxidation on forward osmosis treatment of algae-laden water: Membrane fouling mitigation and mechanism

Forward osmosis (FO) is a high-efficiency and low-energy consumption way for algae-laden water treatment, whereas membrane fouling is still an unavoidable problem in its practical application. In this work, a strategy of ferrous-activated calcium peroxide (Fe(II)/CaO₂) was proposed to control FO membrane fouling in the purification of algae-laden water. With the treatment of Fe(II)/CaO₂, the aggregation of algal contaminants was promoted, the cell viability and integrity were well preserved, and the fluorescent organics were efficiently removed. With respect to the fouling of FO membrane, the flux decline was generally alleviated, and the flux recovery was promoted to varying degrees under different process conditions. It could be revealed through the extended Derjaguin-Landau-Verwey-Overbeek theory that the adhesion of contaminants and membrane surfaces was reduced by Fe(II)/CaO₂ treatment. The interface morphologies and functional groups of membrane verified that Fe(II)/CaO₂ could mitigate the fouling by reducing the amount of algal contaminants adhering to the FO membrane. The co-coagulation of in-situ Fe(III) together with Ca(OH)₂, as well as the oxidation of ^•OH were the main mechanisms for fouling mitigation. In sum, the Fe(II)/CaO₂ process could effectively improve the efficiency of FO for algae-laden water treatment, and has broad application prospects.

Use of Fluorescence Spectroscopy and Chemometrics to Visualise Fluoroquinolones Photodegradation Major Trends: A Confirmation Study with Mass Spectrometry

In this work, we employed EEM-PARAFAC (fluorescence excitation-emission matrices-parallel factor analysis) as a low-cost tool to study the oxidation pathways of (fluoro)quinolones. Amounts of 12.5 μM of enrofloxacin (ENR), ciprofloxacin (CIP), ofloxacin (OFL), oxolinic acid (OA), and flumequine (FLU), as individual solutions, were irradiated under UVA light. A 5-component PARAFAC model was obtained, four of them related to the parent pollutants, named as ENR-like (including CIP), OFL-like, OA-like, and FLU-like, and an additional one related to photoproducts, called ENRox-like (with an emission red-shift with respect to the ENR-like component). Mass spectrometry was employed to correlate the five PARAFAC components with their plausible molecular structures. Results indicated that photoproducts presenting: (i) hydroxylation or alkyl cleavages exhibited fingerprints analogous to those of the parent pollutants; (ii) defluorination and hydroxylation emitted within the ENRox-like region; (iii) the aforementioned changes plus piperazine ring cleavage emitted within the OA-like region. Afterwards, the five antibiotics were mixed in a single solution (each at a concentration of 0.25 μM) in seawater, PARAFAC being also able to deconvolute the fingerprint of humic-like substances. This approach could be a potential game changer in the analysis of (fluorescent) contaminants of emerging concern removals in complex matrices, giving rapid visual insights into the degradation pathways.

EEM-PARAFAC as a convenient methodology to study fluorescent emerging pollutants degradation: (fluoro)quinolones oxidation in different water matrices

Commercial (fluoro)quinolones ((F)Qs), ciprofloxacin (CIP), enrofloxacin (ENR), ofloxacin (OFL), oxolinic acid (OA) and flumequine (FLU) (3 μM each), were degraded with solar-photo-Fenton in a compound parabolic concentrator photoreactor (total volume 5 L) in ultra-pure water at pH = 5.0, salty water at pH = 5.0, and simulated wastewater at pH = 5.0 and 7.5. Iron speciation (its hydrolysis and the complexation with (F)Qs 15 μM and/or chlorides 0.5 M) was calculated at pH 5.0, observing, negligible formation of Fe(III)-chloride complexes, and that >99 % of the total (F)Qs are forming complexes stoichiometry 1:1 with Fe(III) (which also increases the percentage of Fe(OH)²⁺), being minoritarian the free antibiotic form. On the other hand, EEM-PARAFAC (fluorescence excitation-emission matrices-parallel factor analysis) was employed to simultaneously study the behaviour of: i) 4 structure-related groups corresponding to parent pollutants and slightly oxidised by-products, ENR-like (including CIP), OFL-like, OA-like, FLU-like; ii) intermediates still showing (F)Q characteristics (exhibiting analogous fluorescent fingerprint to ENR-like one, but shifted to shorter wavelengths); iii) humic-like substances. The scores from the 4 PARAFAC components corresponding to the parent pollutants were plotted vs. accumulated energy, exhibiting slower decay than their individual removals (measured with HPLC-UV/vis) due to the contribution of the aforementioned by-products to the overall fluorescence. Moreover, thiabendazole (TBZ) 3 μM was added as fluorescence interference. The presence of (F)Qs greatly enhanced TBZ degradation due to (F)Q-Fe(III) complex formation, keeping iron active at pH = 5.0 for Fenton process. The EEM-PARAFAC model was able to recognise the former six components plus an additional one attributable to TBZ-like.

Treatment of Wastewater from Thermal Desorption for Remediation of Oil-Contaminated Soil by the Combination of Multiple Processes

Thermal desorption (TD) is one of the methods commonly used to remediate contaminated soil. However, as water is the liquid adsorbent of the off-gas treatment system in the TD stage, the wastewater generated after multiple cycles in the TD stage has low biodegradability and contains complex organic pollutants. In addition to petroleum hydrocarbon, there are also a lot of ammonia, emulsified oil, phenols, aldehydes, and ketones. In this study, effective removal of contaminants was achieved using a combined process of demulsification and flocculation (DF), ammonia stripping (AS), Fenton oxidation (FO), and reverse osmosis (RO). The combined process was optimized, and the maximum chemical oxygen demand (COD), NH3-N, turbidity, and extractable petroleum hydrocarbons (EPH) removal efficiencies reached 93.3%, 79.8%, 97.6%, and 99.9%, respectively. The FO was the key process for the efficient removal of contaminants. Ultraviolet-visible (UV/Vis), excitation-emission matrix (EEM), fluorescence spectroscopy, and gas chromatography-mass spectroscopy (GC-MS) showed that refractory macromolecular organic pollutants in water were removed, especially aromatics, phenols, and conjugated aldehydes or conjugated ketones, and further ring cleavage of benzene rings and carbocycles with carbon double bonds was observed. The cost-benefit analysis of the combined process was also carried out. The operating cost was 8.73 US$/m3, indicating that the combined process involved moderate costs for recalcitrant wastewater treatment. No studies have been published on combined processes for the treatment of wastewater from TD for the remediation of oil-contaminated soils. Therefore, this study could provide fundamental information based on experimental results and guidelines for wastewater treatment in engineering applications.