Three-Dimensional Excitation and Emission Fluorescence-Based Method for Evaluation of Maillard Reaction Products in Food Waste Treatment

Hydrothermal treatment (HT) of food waste (FW) can formMaillard reaction products (MRPs), the biorefractory organic matter due to the occurrence of Maillard reaction. However, the integrating qualitative and quantitative approach to assess MRPs is scarce. -e goal of this study was to develop a method to characterize and quantify MRPs created by HT of FW. MRPs were identified by molecular weight fractionation, indirect spectrometric indicators, and three-dimensional excitation-emission fluorescence (3DEEM) analysis. -e 3DEEM method combined with fluorescence regional integration (FRI) and parallel factor (PARAFAC) analyses was able to differentiate clearly between MRPs and other dissolved organic compounds compared to other approaches.-e volume of fluorescenceΦ from FRI andmaximum fluorescence intensity Fmax from PARAFACwere found to be suitable quantitative parameters for determination of MRPs in the hydrothermal FW system. -ese two parameters were validated with samples from hydrothermal FW under various operating temperatures and pH.


Introduction
In China, the stacking of FW has become a major issue to cause environmental problems.Recently, anaerobic digestion (AD) as an attractive waste treatment practice has been used to decrease the amount of biowaste and recover energy [1].Due to the high biodegradability and water content of FW, it becomes a good candidate for AD [2].During the process of the AD, the hydrolysis step is generally considered as the rate-limiting step for complex organic substrates degradation.erefore, hydrothermal treatment (HT) was ordinarily used as a pretreatment to promote the solubilization of complicated macromolecular solid organic matters, thus improving the AD process [3].
Despite the acceleration of dissolved properties of FW, it has been documented that HT is responsible for the formation of Maillard reaction products (MRPs) [4].On the one hand, the formation of MRPs can lead to the substrate loss during the HT of FW.On the other hand, the influence of MRPs themselves on AD merits further investigation [5].
erefore, to optimize the HT process and enhance the efficiency of AD, it is essential to provide an integrating quantitative and qualitative approach to assess the MRPs production.
Numerous methods have been developed to characterize the occurrence of MRPs by using the precision analysis instrument [6,7].However, the quantitative determination could not be achieved because there is no pure standard for the measurement of MRPs [6].In addition, these devices are time-consuming and labor-intensive, limiting their application and spreading.us, some easy-to-use and convenient characterization techniques became most commonly used methods, including the UVA 254 and color intensity [8,9].
Nowadays, three-dimensional excitation and emission fluorescence (3DEEM) is regarded as a promising tool to offer characteristic information for signature chemical structures in a complex mixture of chromophores [10].And, the qualitative characterization of MRPs has been achieved by the traditional 3DEEM method [11].However, a quantitative determination could not be realized because only one excitation/emission intensity value can be used for analysis.Recent studies have demonstrated that fluorescence regional integration (FRI) method and parallel factor analysis (PARAFAC) method were developed to integrate the area beneath EEM spectra and semiquantitatively assess the specific components in a complex system [12][13][14].However, the application of 3DEEM to the semiquantitative characterization of MRPs in the complicated hydrothermal FW system is scarce.
erefore, the utilization of 3DEEM to distinguish between MRPs and other dissolved organic matter under various hydrothermal conditions is supposed to be further explored.
is study aimed at developing a method to characterize and quantify MRPs created by HT of FW.Firstly, MRPs were characterized and evaluated with different methods.en, MRPs production was further assessed by the applicability of 3DEEM combined with FRI and PARAFAC, hence exploring the suitable fluorescence parameters for semiquantifying the MRPs in the hydrothermal FW system.

Food Waste Sample Preparations.
e FW, containing rice (44%), noodles (16%), vegetables (23%), meat (6%), and tofu (11%), was compounded based on the characteristics similar to FW collected from a canteen of Zhejiang Gongshang University (Hangzhou, China) in our previous study [15].e five components came from the same vendor at Cui Yuan farmers' market (Hangzhou, China).e FW was cut into small pieces first by hand-breaking and then crushed using a mangler.e untreated FW sample was stored at −18 °C before the HT. e main characteristics of the FW are listed in Table S1.

Hydrothermal
Treatment.Hydrothermal treatment of FW was performed in an 80 mL airtight pressure digestion vessel as described by our previous study [16] at separate batch operations at each temperature.During HT, about 30 g crushed FW was placed in the vessel.Each batch was processed for 30 min.In the first experiment, the temperature manipulations were made at 110, 120, 130, 140, 150, and 160 °C in an oil bath to explore the effect of temperature on MRPs production.In the second experiment, the different initial pH values (3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, and 10.0) were tested at 130 °C for 30 min.e time was measured from when the oil bath reached the set temperature.e vessels were cooled to ambient temperature after HT.Each treatment was performed in triple vessels.

Extraction of WEOM.
WEOM was obtained with deionized water (solid-to-water ratio of 1 : 10 w/v), and the mixture was shaken for 1 h in a horizontal shaker at 35 ± 2 °C.
e extracts were separated from the mixture by centrifugation at 10,000 rpm for 5 min and filtered using the microfiltration membrane (0.45 µm).

Synthetic MRPs Solution.
e synthetic MRPs solution was made by a concentrated solution of melanoidins, which are defined as brown substances formed during the final stage of the Maillard reaction.e formula of the concentrate was made with a 1 : 1 molar ratio of glucose and glycine with a buffer of 0.5 M Na 2 CO 3 according to the previous research [17].e solution was heated at 120 °C for 3 hours as the record [17]. is synthetic MRPs solution has been used as model MRPs to analyze the properties of MRPs [9], and it was employed to examine the availability of MRPs characterization method in the present study.

Analytical Methods of MRPs
2.5.1.Spectrometric Indicators.(1) UVA 254 .UVA was a measure of absorbance at 254 nm, measured in a 1 cm path length quartz cell.It can measure unsaturated bonds or aromaticity within dissolved organic matters [18].erefore, this spectrometric index was useful for this study as MRPs were linked to the presence of unsaturated double bonds and aromatic compounds and expressed as cm −1 •mL/g dry weight.
(2) Color Intensity.A spectrophotometer at a wavelength of 475 nm was used to determine color intensity in a 1 cm path length cell.e absorbance at this wavelength was characteristic of brown color.Characteristic color intensity was recorded in a platinum-cobalt (PtCo) unit as previously described [9].
(3) Browning Index.e Browning index of the FW solid was measured by an enzymatic digestion method which releases the brown pigments.Samples were dried for 24 h and grounded to a smaller size before use.e proposed method was modified based on pronase proteolysis created by Palombo et al. [19].e procedure was as follows: 0.3 g of the dried sample was added into a test tube which contains 5 mL deionized distilled water at 45 °C and mixed thoroughly.
en, another 0.4 mL of pronase solution was added into the mixture.After that, the test tubes were placed in a water bath, incubated for 120 min at 45 °C, and then cooled in ice water, and 1 mL trichloroacetic acid (80% TCA) was added to each tube.Finally, centrifugation (20 min at 7000 rpm) and filtration were used before the spectrometric determination.
e optical density of the filtrates was determined on a spectrophotometer.Samples were measured in a 1 mL cuvette with 1 cm pass length.
e OD of the brown index was calculated as OD � OD 420nm − OD 550nm and expressed as OD/g dry weight.
2.5.2.Molecular Weight Fractionation.Molecular weight fractionation was applied for a better separation and characterization of dissolved organic matters.Fractionation of samples was performed using an ultrafiltration centrifuge tube with different molecular weight cutoffs: 3 kDa, 10 kDa, and 30 kDa. e samples were filtered in series from 30 kDa to 3 kDa.

3DEEM Analyses.
e 3DEEM of WEOM was measured in a 1 cm cuvette using a Hitachi F-4600 fluorescence spectrometer at room temperature (25 ± 2 °C).e scanning ranges were 200-500 nm for excitation and 250-500 nm for emission.Scanning was recorded at 5 nm intervals for excitation and 1 nm steps for emission, respectively, using a scanning speed of 2400 nm/min.e Milli-Q water blanks were subtracted in order to eliminate the effect of Raman scattering.In addition, exported EEMs were normalized by the Raman area and eliminated the primary and second Rayleigh scattering.
(1) Peak-Picking Method.A peak-picking method is used for the detection of the fluorescence intensity of easilyidentifiable peaks and their locations within the EEMs.And then, the observed peaks were analyzed by comparing their fluorescence properties with the change of operating condition.
Besides, as an important humification index in EEM, HIX can reflect the degree of humification of the sample.HIX was calculated by dividing the emission intensity into the 435-480 nm regions by intensity in 300-345 nm when excitation intensity is at a wavelength of 255 nm and is shown as follows: (2) FRI Analysis Method.e FRI approach was employed in this work to characterize the five excitation-emission regions of EEM spectra [12].According to previous studies, EEM spectra are usually divided into five areas (Table S2): aromatic protein-like fluorophores (regions I and II), fulvic acid-like fluorophores (region III), soluble microbial product-like fluorophores (region IV), and humic acid-like fluorophores (region V).By normalizing the cumulative excitation-emission area volumes to relative regional areas (nm 2 ), the volume of fluorescence (Φ (i) ) was calculated according to Chen et al. [12] within each region (i), applying the following equation: where MF (i) is a multiplication factor calculated by Equation (3); Δλ ex is the excitation wavelength interval (taken as 5 nm); Δλ em is the emission wavelength interval (taken as 1 nm); and I(λ ex λ em ) is the fluorescence intensity at each excitation-emission pair (Raman units): And the normalized excitation-emission area volumes referred to the value of region i.
e entire region was calculated, and the percent fluorescence response was then determined using the following equation: (3) PARAFAC Component Analysis Method.PARAFAC is a method that decomposes EEMs of complex WEOM into available components.PARAFAC modeling was performed in MATLAB followed by the procedure recommended in the DOMFluor toolbox [20].e PARAFAC models with two to seven components were computed, and core consistency was often applied to select the optimal number of components [21].e concentration scores of each component were represented by maximum fluorescence intensity (F max , R.U.).

Other Analytical Methods.
e TS, volatile solids (VSs), dissolved organic carbon (DOC), soluble carbohydrate, and soluble protein were all determined.
e diluent samples were operated as the same as mentioned in Section 2.3.SCOD was analyzed using standard methods [22].e DOC concentrations of the WEOM solutions were determined with a total organic carbon analyzer (TOC-L CPH, Shimadzu, Japan).Soluble protein was quantified by the Lowry−Folin method using bovine serum albumin as the standard, and carbohydrate was determined using the phenol-sulfuric acid method with glucose as the standard [23,24].

Variation of Food Waste after Hydrothermal Treatment.
FW was treated under high pressure for 30 min at 110, 120,130, 140, 150, and 160 °C.e SCOD concentrations of the hydrothermal-treated FW are shown in Figure S1.Increasing the temperature from 110 to 160 °C caused an increase in SCOD from 85.55 g/kg to 119.91 g/kg.Although the increment of SCOD was obtained, color generation by Maillard reaction was also observed in the HT process due to the high temperature and the presence of proteins and carbohydrates [25].Figure S2 shows that the treated solid gradually produced a darker brown color than the untreated FW with increasing treatment temperature.e significant color variation of hydrothermal FW was highly reliant on the formation of MRPs which depends on the reaction temperature.Nevertheless, owing to the complex and heterogeneous nature of the MRPs, it has not been possible to isolate or purify MRPs [26].erefore, there is a lack of methods for rapid, effective, and quantitative estimation of MRPs in the hydrothermal FW system.

Characterization and Evaluation of MRPs in Food Waste
after Hydrothermal Treatment

Distribution of Soluble Organic Compounds in WEOM.
Figure 1 presents the molecular weight distribution of WEOM before and after HT.DOC of the original sample was evenly in the MW analysis range and remained unchanged in each fraction after HT at 120 °C.However, molecular weight fractionation distribution was modified at 140 °C compared with 120 °C.e total DOC content had a large promotion, but significant increase of DOC from 12 to 27 g/L only occurred in the MW > 30 kDa fraction.e significant increment highlighted that dissolved organic compounds, residing in the MW > 30 kDa fraction, were expected as MRPs which have a molecular weight between 40 and 70 kDa [27].Moreover, it is well known that the formation of MRPs has a positive relationship with the carbohydrates and proteins [28].In 140 °C group, the most prevalent fraction of soluble carbohydrates and soluble proteins also resided in the MW > 30 kDa fraction, which accounts for 78% and 83%, respectively.In addition, the characteristic absorbance UVA 254 of WEOM in the MW > 30 kDa fraction was 8, 7.5, and 34 cm −1 mL/g for untreated FW, 120 °C, and 140 °C groups, respectively.e rise of this characteristic absorbance indicator showed that HT could result in the generation of MRPs from organic matter mixtures at 140 °C.Similar conclusions were obtained by Barber [29] that MRP production occurred in the typical reaction range of thermal hydrolysis of 140-165 °C.

Spectrometric Characterization of the MRPs from Food
Waste.Nowadays, various easy-to-use quantitative spectrometric characterization methods have been applied to determine the MRP production in complicated systems [19,30].ree representative indicators (ΔUVA 254 , Δbrowning index, and Δcolor intensity) were employed to demonstrate the modification of MRPs content of hydrothermal FW (Figure 2), and raw material was taken as the control.At first, ΔUVA 254 did not vary significantly (p > 0.05) after HT at 110 °C and 120 °C.e ΔUVA 254 increased significantly (p < 0.05) at 130 °C, which enhanced by 4 Journal of Chemistry 1.5-fold compared with 120 °C.en, this surrogate index continued ascending with the increase of treatment temperature, indicating a significant number of compounds accumulated when the hydrothermal temperature above 120 °C.In addition, a similar tendency was found in the result of the pronase method, and the Δbrowning index was utilized for determination of nonenzymatic browning.e changing of the Δbrowning index showed that the brown pigment production was slim to zero in 110 °C and 120 °C treatments and exhibited a regular increase beyond the critical temperature (120 °C).e tendencies of these two parameters were consistent with the transformation of the surface color of hydrothermal-treated FW.However, the accuracy of these two indicators was still in doubt because of the complexity of hydrothermal FW systems.Moreover, the Δcolor intensity of WEOM was also recorded.Nevertheless, the relative standard deviation of Δcolor intensity was higher than other methods, and this color indicator did not demonstrate the significant difference (p > 0.05) between the 130 °C and 140 °C groups, which was different from the appearance changes of hydrothermal-treated FW.Although spectrometric characterization could quickly verify the accumulation of MRPs in an indirect way, the influence of other organic matters was still not solved.

Characterization of MRPs by 3DEEM Fluorescence
Analysis.
e objective of using 3DEEM fluorescence was to monitor the change of fluorescence properties in WEOM and to analyze the effect of temperature on MRPs generation in the hydrothermal FW system.Typical 3DEEM spectra of treated WEOM samples at different temperatures are shown in Figure 3, and detailed fluorescence properties are summarized in Table 1 by the "peak-picking" method.As expected, there were obvious differences in the 3DEEM spectra of the WEOM with the rise of temperature, and the maximal transformation was found between 130 °C and 140 °C groups.Moreover, Table 1 shows the change of peak intensity at the soluble microbial region (λ ex /λ em � 285/357-361), which decreased from 3.58 R.U. (raw material) to 0.10 R.U. (160 °C), and the sharp decrease trend started at the 140 °C group.To the best of our knowledge, soluble microbial products (SMPs) contain various complex organic materials, such as proteinaceous material, carbohydrates, humic and fulvic substances, and organic acids [31,32].us, the organic matters resided in the region of soluble microbial by-products were considered as soluble carbohydrates and proteins because other organic matters would not appear in this region.Conversely, the peak intensity at the humic acid-like region (λ ex /λ em � 320-360/419-438) increased from 0.43 to 3.34 R.U. as the temperature increased, and obvious transformation also occurred at 140 °C group.e peak ratio (V/IV) ascended with the increase of temperature, which implied that the generation and accumulation of humic acid-like organic fluorescent molecules were attributed to the polymerization of the soluble carbohydrates and proteins during the HT.In addition, the movement of humic-like peaks location within the 3DEEM (λ ex /λ em ) to longer wavelength (red-shifting), from λ ex /λ em � 320/428 to 360/438, also indicated the humic acid-like fluorophores were concentrating and becoming more refractory [33].Besides, the HIX value has been applied to evaluate the humification extent of dissolved organic matter [34].From Table 1, the increase of HIX values after 140, 150 and 160 °C treatments were indicative of more humic WEOM.It is generally known that MRPs are highly aromatic and resemble humic substances; it suggested that 3DEEM might serve as a good descriptor to prove the MRPs production during the HT.
Overall, compared with the 3DEEM analysis method, spectrometric approaches cannot provide a direct Besides, Figures 1(b) and 1(c) show that even though the molecular weight fractionation was used, the existence of organic macromolecular compounds (carbohydrates and proteins) might also affect the accuracy of UVA 254 measurement.Accordingly, 3DEEM fluorescence analysis was simple, had good selectivity, and provided a wealth of  Journal of Chemistry information about WEOM, especially helpful in characterizing WEOM and interpreting the complex information in EEM [35].However, fluorescence information was still restricted because only one peak location and fluorescence intensity value was used for analysis, and it should be noticed that the direct quantification of MRPs using 3DEEM can be of major interest for investigation of WEOM from FW.Recently, FRI analysis and PAR-AFAC component estimation were developed to provide a more complete data analysis than the traditional "peakpicking" technique [36].erefore, the FRI analytical method was used to reveal the transformation of organic compounds in WEOM by integration of the volume beneath each EEM region.According to the five regions of the FRI analysis, it can be found that the obvious characteristic peaks were presented in region IV and region V. e variation of percent fluorescence response P (i,n) of the WEOM is shown in Figure 4.
e P (i,n) values of region IV and region V remained almost constant in 110, 120, and 130 °C groups.When temperature increased from 130 to 160 °C, the P (i,n) value of region IV decreased from 49.8% to 16.9%, and the P (i,n) of region V increased from 12.1% to 66.1%.Moreover, the P (i,n) value of region I also had a little reduction above 130 °C.ese results implied that humic acid-like material regarded as MRPs in region V were efficiently accumulated when HT temperature was beyond 130 °C.It was assumed that the disappearance of soluble carbohydrates and proteinaceous products (region IV) and tryptophan-like substances (region I) were the source material for the polymerization of MRPs (region V). erefore, region IV (carbohydrates and proteinaceous products) matters and region V (MRPs) matters can be effectively differentiated based on the FRI method.
Furthermore, PARAFAC can capture the heterogeneity of WEOM samples, thereby decomposing EEM spectra into various individual fluorescent components.Two fluorescent components evaluated by PARAFAC using the COR-CONDIA procedure were the soluble microbial by-productlike component (C 1 ) and humic acid-like component (C 2 ).Individual components are shown in Figure 5 as contour plots.e excitation and emission loadings are also given in Figure 5, and excitation and emission characteristics of the components identified in this study and probable source of components are depicted in Table S3.In Table S3, two PARAFAC components are pointed out in detail: C 1 with peaks of ex/em � 280/360 nm was associated with soluble microbial products, defined as carbohydrates and proteins which could be easily biodegradable [37,38], and C 2 with specific peaks of ex/em � 360/441 nm originated from humic acid-like substances [9,28].In the current study, the humic acid-like component identified by PARAFAC was MRPs, which had higher ex/em than the other synthetic or isolated humic compounds [11,27,39].Furthermore, the observed humic-like compounds were forcefully identified as MRPs by comparing its fluorescence property (ex/em) with synthetic MRPs solution created in our study (Figure S3). e maximum fluorescence intensities (F max ) of component C 2 remained constant at a low content when the temperature was in the range of 110 °C to 130 °C and then increased from 22.37 to 155.79 R.U. by further increasing the temperature from 130 to 160 °C (Figure S4), which indicated little accumulation of nonbiodegradable MRPs at temperatures below 130 °C and mass production of MRPs when temperature beyond 130 °C.Meanwhile, the results in Figure 6 also show that the F max of C 1 had a slight reduction at 110 °C, 120 °C, and 130 °C groups but substantially dropped with the further increase of temperature.Results from PARAFAC were consistent with the FRI analysis, which suggested biodegradable organic compounds and MRPs in WEOM can be separated by PARAFAC.
As mentioned above, the distribution of P (i,n) and the 2component PARAFAC estimation provided complementary information, showing that humic acid-like matter (MRPs) and SMP materials (carbohydrates and proteins) which dominated the WEOM underwent an increase and decrease, respectively, as the operating temperature increased.ese results proved FRI and PARAFAC can differentiate fluorescence characteristic disparities between MRPs and other organic compounds.And, it also can be seen that Maillard reaction (C 2 ) was accelerated at elevated temperature by consuming organics in C 1 .
erefore, 3DEEM combined with FRI/PARAFAC could be employed to analysis of MRPs generated from hydrothermal-treated FW.Φ (V) parameter from FRI and maximum fluorescence intensity in the C 2 component F max (C 2 ) parameter from PARAFAC were used as the characterization parameters for assessing the generation of MRPs.All the data were based on dry weight, and raw material was taken as control.e corresponding results of fluorescent indictors are presented in Figure 6.It can be seen from Figure 5 that the impact of temperature on the ΔΦ (V) was minimal in 110, 120, and 130 °C groups (p > 0.05), staying at a low level from 5.1 × 10 4 to 9.0 × 10 4 R.U.•nm 2 •mL/g.But then ΔΦ (V) increased significantly from 2.9 × 10 5 (140 °C) to 1.2 × 10 6 R.U.•nm 2 •mL/g (160 °C).A similar tendency was found in ΔF max (C 2 ) (Figure 5).e ΔF max (C 2 ) remained unchanged as temperature increased from 110 to 130 °C and then went through a successive and relatively large increase from 33.5 to 141.3 R.U.•mL/g as temperature further increased.ese results were consistent with the fact that MRPs are temperature dependent [29].

Semiquantitative Characterization of Fluorescent
Besides, the correlations between 3DEEM fluorescence parameters and spectrometric parameters were analyzed to check whether fluorescence parameters could effectively characterize the MRPs in FW after HT.Pearson correlations for 3DEEM fluorescence parameters and spectrometric parameters are summarized in Table S4.Generally, spectrometric parameters were strongly correlated with 3DEEM fluorescence parameters (r > 0.880, p < 0.01), which indicated that these two fluorescent analytical techniques could be employed to monitor the generation of MRPs and semiquantitatively determine the content of MRPs in the hydrothermal FW system.In addition, these two indicators were expected to be tested at various situations, which help their generalization and application.Apart from temperature, pH also has an impact on MRPs formation.From previous studies, pH and alkalinity were reported to increase the degree of polymerization, thereby causing the accumulation of MRPs [40].erefore, it is meaningful to evaluate the effect of pH on MRPs production during HT in FW.In order to semiquantitatively determine their production, 3DEEM-FRI/PARAFAC methods were applied.FW with initial pH from 3.0 to 10 was treated at 130 °C for 30 min (Figure 7).As pH increased from 3.0 to 5.0, the mean ΔΦ (V) decreased from 7.05 × 10 5 to 2.29 × 10 5 R.U.• nm 2 •mL/g and reached the minimum at pH 5.0.Nevertheless, an obvious increase of MRPs was observed from a minimum of 2.29 × 10 5 to 1.39 × 10 6 R.U.•nm 2 •mL/g as the pH increased from 5.0 to 10.0.
e ΔF max (C 2 ) from PARAFAC analysis, which was associated with the contribution of MRPs, had a similar tendency to ΔΦ (V) and increased from the minimum of 14.24 R.U. at pH 5.0 to the maximum of 90.04 R.U. at pH 10.Previous studies have stated that high initial pH (pH > 5) could accelerate the Maillard reaction rate because Schiff-base matter formed easily [41].Moreover, low initial pH could also cause the formation of MRPs as it favors 1,2-enolisation reaction pathway and results in the ascent of compounds like furfural or 5-hydroxymethyl-2-furaldehyde (HMF) [42].erefore, the results above further verified that ΔΦ (V) and ΔF max (C 2 ) allow an effective evaluation for MRPs production in the hydrothermal FW system.
Besides, hydrothermal time and composition of FW also affect the occurrence of Maillard reaction.us, the effect of these hydrothermal conditions on MRPs production can be evaluated according to these two parameters.In this way, the formation of MRPs can be effectively controlled under the optimized hydrothermal condition, thus preventing the substrate loss and promoting limited hydrolysis efficiency simultaneously.
erefore, the increase of bio-converted energy can be achieved by reducing substrate consuming from MRPs formation.

Conclusion
Compared to traditional methods, the 3DEEM analysis is proved to be a more sensitive method to estimate the occurrence of Maillard reaction in the hydrothermal FW system, providing valuable information when the MRPs are formed.However, its utility is limited to the quantifying of the MRPs.e FRI and PARAFAC analytic methods were established to further distinguish MRPs from the other dissolved organic compounds by integration of the volume beneath each EEM region and capturing the heterogeneity of samples.And ΔΦ (V) from FRI and ΔF max (C 2 ) from PARAFAC can be considered liable parameters for semiquantifying MRPs under various temperature and pH.e minimum MRPs were validated at temperature below 130 °C and pH 5.0.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.their associated names.Table S3: comparison of the fluorescence component peak location in the current study with those reported in the previous literature and probable source description.Table S4: Pearson correlations among spectrometric parameters and EEM fluorescence parameters.Figure S1: SCOD of hydrothermal-treated food waste at different temperatures.e data were based on wet weight.Figure S2: the appearance of hydrothermal-treated food waste (untreated; 110 °C, 120 °C, 130 °C, 140 °C, 150 °C, and 160 °C treated).Figure S3: EEM spectra of the synthetic MRP solution produced by glycine and glucose after hydrothermal treatment.e peak location of synthetic MRP solution was λ ex /λ em � 355/435; the λ ex /λ em value is higher than the other humic isolates tested, which explains why this product was MRPs. Figure S4: F max values of the two components (C 1 and C 2 ) in WEOM from food waste after hydrothermal treatment.(Supplementary Materials)

Figure 2 :
Figure 2: ree spectrometric indicators of MRPs production from FW at different temperatures.

Figure 4 :
Figure 4: e distribution of FRI in WEOM from food waste after hydrothermal treatment.

Figure 5 :
Figure 5: Two-component PARAFAC model generated from samples of WEOM: (a and b) contour plots of the spectral shapes; (c and d) line plots of the loadings.

Table 1 :
Fluorescent properties of treated groups at each temperature.