Supporting Information Effects of copper ions on removal of nutrients from swine wastewater and on release of dissolved organic matter in duckweed systems Qi Zhou a,b,c,* , Xiang Li b,* , Yan Lin b,* , Chunping Yang a,b,c,d,** , Wenchang Tang b , Shaohua Wu b , Dehao Li a,c,** , Wei Lou d a School of Environmental Science and Engineering, Guangdong University of Petrochemical Technology, Maoming, Guangdong 525000, China b College of Environmental Science and Engineering, Hunan University and Key Laboratory of Environmental Biology and Pollution Control (Hunan University), Ministry of Education, Changsha, Hunan 410082, China c Guangdong Provincial Key Laboratory of Petrochemcial Pollution Processes and Control, Guangdong University of Petrochemical Technology, Maoming, Guangdong 525000, China d Hunan Province Environmental Protection Engineering Center for Organic Pollution Control of Urban Water and Wastewater, Changsha, Hunan 410001, China * These authors contribute equally to this paper.
53
Embed
ars.els-cdn.com · Web viewSupporting Information Effect s of copper ions on removal of nutrients from swine wastewater and on release of dissolved organic matter in duckweed systems
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Supporting Information
Effects of copper ions on removal of nutrients from swine wastewater and on
release of dissolved organic matter in duckweed systems
This supporting information contains 37-page document, including the detailed
description for the First-order kinetic, Michaelis-Menten kinetic, Modified logistic
kinetic and modified Stern–Volmer model cited literatures, procedure of minimizing
the inner filter effect, the PARAFAC and Two-dimensional correlation analysis, the
specific procedure of additional experiment, the specific procedure of spiking
experiment with the water samples dosed with NADH, the pH and metal-spiking
experiments, discussion on the peaks of C4 , 3 tables, 15 figures and this cover page.
The three Degradation kinetics Models
First-order kinetic model:
dCdt
=-rC (1)
Integrated by the boundary condition t = 0, C = A; t = t, C= C, the above equation
1 was transformed to:
C =A e-rt (2)
Michaelis-Menten kinetic model:
dCdt
=-k1Ck2+C
(3)
Integrated by the boundary condition t = 0, C = A; t = t, C= C , the above equation
3 was transformed to:
k2 ln( C)+ C =- k1 t+A (4)
Modified logistic kinetic model:
dCdt
=-r C(1 - aCC0
) (5)
Q =1 - aCC0
(6)
Integrated by the boundary condition t = 0, C = C0; t = t, C= C, the above equation
5 was transformed to:
C=C0
a +(1- a ) ert (7)
where C is nutrient concentration (mg/L), t is time (day), A is the initial amount
of nutrient (mg/L), r is the removal rate constants(day-1), k1 is a maximum rate
approached with increasing concentration (day-1), and k2 is a pseudo-equilibrium
constant (pseudo, because as the reaction occurs, it is constantly unbalancing the
equilibrium represented by the constant). C0 is the initial amount of nutrient
(mg/L), r is the removal rate constants (day-1), a is the parameter to be estimated
(a<1), Q is the external environmental stress.
Modified Stern-Volmer model
In general, two fluorescence quenching titration (FQT) models, Ryan–Weber
non-linear model and the modified Stern–Volmer model, are widely used for
characterizing the metal binding behavior of DOM although the models still have the
limitation that they neglect the potential existence of multidentate metal binding
multiple complexes (Esteves da Silva et al., 1998). The major assumptions of the
Ryan–Weber model are that only 1:1 metal/ligand complexes are formed and that a
linear relationship exists between the metal-bound ligand concentration and the
quenched DOM fluorescence intensity (Ryan and Weber, 1982; Plaza et al., 2006).
However, the linear relationship may not be valid if stable non-fluorescent complexes
are formed and/or if fluorescence quenching data at a particular wavelength pair
represents more than two different binding sites (Smith and Kramer, 2000; Hays et al.,
2004). The modified Stern–Volmer may partially solve the problems associated with
the non-linear fluorescence response for metal–DOM interactions by estimating the
portion of unquenched fluorescence contribution. Therefore, the major assumptions of
the Ryan–Weber model are that stable non-fluorescent complexes are formed and/or
fluorescence quenching data at a particular wavelength pair represents more than two
different binding sites. Based on that, in this study, the modified Stern–Volmer model
was applied to estimate the parameters related to copper binding.
This model can be used to calculate the conditional stability constant KM of Cu
ions to DOM released from the duckweed system (Hur and Lee, 2011):
F0
F0-F= 1
f KMCM+1
f (8)
Here F and F0 are the measured fluorescence component scores (arbitrary units per
unit DOC, A.U./mg/L) at the total Cu2+ concentration CM (mmol/L) and the no Cu ions
addition (0 mmol/L), respectively. The parameter KM was the conditional stability
constant at the certain experimental condition, and f represents the fraction of the
initial fluorescence, which is accessible to quencher. The parameters of f and KM were
solved by plotting F0 /( F0–F) against1/CM. The excitation/emission wavelength pairs
were selected based on the PARAFAC results.
References
Esteves da Silva, J.C., Machado, A.A., Oliveira, C.J.,Pinto, M.S. 1998. Fluorescence quenching of anthropogenic fulvic acids by Cu(II), Fe(III) and UO(2)(2+). Talanta 45(6), 1155-1165.
Hays, M.D., Ryan, D.K.,Stephen, P. 2004. A modified multisite Stern-Volmer equation for the determination of conditional stability constants and ligand concentrations of soil fulvic acid with metal ions. Anal. Chem. 76(3), 848-854.
Plaza, C., Brunetti, G., Senesi, N.,Polo, A. 2006. Molecular and quantitative analysis of metal ion binding to humic acids from sewage sludge and sludge-amended soils by fluorescence spectroscopy. Environ. Sci. Technol. 40(3), 917-923.
Ryan, D.K.,Weber, J.H. 1982. Fluorescence quenching titration for determination of complexing capacities and stability constants of fulvic acid. Anal. Chem. 54(6), 12-22.
Smith, D.S.,Kramer, J.R. 2000. Multisite metal binding to fulvic acid determined using multiresponse fluorescence. Anal. Chim. Acta 416(2), 211-220.
Hur, J.; Lee, B.-M. 2011. Characterization of binding site heterogeneity for copper within dissolved organic matter fractions using two-dimensional correlation fluorescence spectroscopy. Chemosphere. 83 (11), 1603-1611.
Procedure of minimizing the inner filter effect
During the analysis of DOM fluorescence, the dissolved organic carbon (DOC) is
proportional to fluorescence intensities of several specific fluorescent fractions.
However, with the increase of DOC concentration, the fluorescence intensities of
components potentially deviated from a linear relationship. This kind of optical
phenomenon involved in a significant decrease in emission quantum yield as a result
of the absorption of excited and emitted radiation by a sample matrix, which is known
as the “inner filter effect”. Therefore, the deviation of the fluorescence intensity
during the analysis of NOM fluorescence components cannot be neglected. The inner
filter effect often has no significant influence for samples with low absorbance or
concentration, so there are many methods reported for reducing the influence of inner
filter effect (Zhu et al., 2017). And the major method was to dilute the solution in
order to decrease the absorbance or concentration of a given sample. Absorbance can
be used as an indicator for determination of the need for dilution. For example, Poulin
et al. (2014) diluted a solution containing iron (III) to a degree at absorbance of <0.2
at 254 nm to minimize inner filter effects during fluorescence analysis. In this study,
we adopted this method founded by Poulin et al. (2014) due to the similarity of
solution containing metal ions.
In this study, although the FDOM samples were too high in concentration, all of
the samples were diluted with Milli-Q water until the concentration of DOC was less
than 10 mg/L (Abs254<0.2) to minimize the inner filter effect before the measurement
of the EEM spectra. The detailed procedure of the fluorescence measurement was as
follows: 1) DOC concentrations of all samples were measured by using a Shimadzu
TOC-VCPH analyzer (Shimadzu, Japan); 2) building the relation between the DOC
concentrations and UV254 absorbance was established to determine the dilution
multiple of each sample for measurement of 3D fluorescence. The relation between
the DOC concentrations and UV254 absorbance was shown in Figure 5. It was
obvious that there was a good linear relationship between DOC concentrations and
UV254 absorbance (R2>0.998). Based on the calculation results, when UV254
absorbance was less than 0.2, the DOC concentration reached less than 10 mg/L
through calculating. It meant that all the samples need be diluted by Milli-Q water
until the concentration of DOC was less than 10 mg/L according to initial DOC
concentrations of samples, and the aim to minimize the inner filter effect before
measuring EEM spectra. 3) DOM fluorescence spectra were measured by using an F-
4600 fluorescence spectrophotometer equipped with a 150W xenon lamp (Hitachi,
Japan). In a word, high concentrations of FDOM had been diluted to minimize the
inner filter effect before measuring EEM spectra in this study.
PARAFAC analysis
PARAFAC analysis was conducted by using the DOMFluor tool-box
(http://www.models.life.ku.dk/) in MATLAB R2012b (Math-works, Natick, MA)
(Stedmon and Bro, 2008). The toolbox allows one to identify the possible outliers and
an appropriate number of components. Prior to analysis, the Raman and Rayleigh
scatters were removed according to a homemade Matlab program. In addition, the
fluorescence at excitation and emission wavelengths both below 250 nm were not
used for further analysis to avoid the interference of noise signals (Stedmon and
Markager, 2005). PARAFAC was computed using two to seven component models
with non-negativity constraints, and the residual analysis, split half analysis, visual
inspection and model validation were applied to determine the number of
fluorescence components. After that, the obtained maximum fluorescence intensity
(Fmax) (R.U.) was used to estimate the relative levels of individual components. All
the obtained data set were further plotted using Origin 8.5 software. More detailed
information of PARAFAC analysis has been described elsewhere (Stedmon and Bro,
2008).
References
Stedmon, C.A.,Bro, R. 2008. Characterizing dissolved organic matter fluorescence with parallel
factor analysis: a tutorial. Limnol. Oceanogr-Meth. 6, 572-579.
Stedmon, C.A.,Markager, S. 2005. Resolving the variability in dissolved organic matter fluorescence in a aemperate estuary and its catchment using PARAFAC analysis. Limnol. Oceanogr. 50(2), 686-697.
Two-dimensional correlation analysis
Two-dimensional correlation analysis was conducted by using the“2D shige”
software (Kwansei-Gakuin University, Japan) to generate 2D-FTIR-CoS maps
(synchronous and asynchronous maps) and PCMW2D maps by using Cu2+
concentration as the external perturbation. Prior to analysis, the data of FTIR spectra
was corrected by baseline and smoothed, then the corrected data set were analyzed by
the“2D shige” software. After that, the synchronous and asynchronous 2DCoS maps
were generated to analysis binding sites and sequence of DOM according the Noda’s
rule (Noda, 2014). Besides, the obtained synchronous fluorescence MW2D
correlation spectra were applied to analyze conformation changes of DOM such as the
transition point and ranges with Cu2+ perturbation. The window size was selected as 5
to produce clear PCMW2D spectra (Peng et al., 2014). All the obtained data were
further plotted using Origin 8.5 software. More detailed information about the 2D-
FTIR-CoS analysis and the PCMW2D analysis was described elsewhere (Peng et al.,
2014).
References
Noda, I. 2014. Frontiers of two-dimensional correlation spectroscopy. Part 1. new concepts and noteworthy developments. J. Mol. Struct. 1069(1), 3-22.
Peng, L.L., Zhou, T., Huang, Y., Jiang, L.,Dan, Y. 2014. Microdynamics mechanism of thermal-induced hydrogel network destruction of poly(vinyl alcohol) in D2O studied by two-dimensional infrared correlation spectroscopy. J. Phys. Chem. B. 118(31), 9496-9506.
The specific procedure of additional experiment
The initial media was prepared as synthetic swine wastewater and its ionic
concentrations were listed in Table S1. Different volumes CuSO4•5H2O solution was
added into the initial media to simulate different Cu2+ concentrations (0.1, 0.5, 1.0,
2.0, and 5.0 mg/L). The pH value of media was adjusted at 6.30±0.40 by using 0.10
mol/L HCl solution. Then, the initial media was sterilized in an autoclave for 30 min
at 121°C before duckweed was added into the media. 18 beakers (500 mL, 6.4
cm×6.4 cm×12.5 cm) including 2×6 duplicate sample beakers with extra copper ions
were used in six batch tests. Each beaker contained 140 mL (3.4 cm deep in the
beaker) of media, and it was initially seeded with the same amount (0.4±0.1 g fresh
weight) of duckweed to cover the 80% surface area of 6.4 cm×6.4 cm. All beakers
were placed into a light growth chamber with 16 h photoperiod at 25 °C and a
photosynthetic photon flux density of 70±10 μmol m-2 s-1 provided by wide spectrum
fluorescent tubes. Due to the decrease of pH as the duckweed grew, 0.10 mol/L NaOH
solution was added to maintain the pH around 6.30±0.40 every day. And 1 mL water
sample was taken out from each beaker by using 1 mL injector for the DOM analysis
every 24 h for 15 days.
The specific procedure of spiking experiment with the water samples dosed with
NADH
1) 10 mg/L NADH stock solution was prepared by dissolving 0.0107 g NADH-Na2
into 1 L Milli-Q water. Then 5 mg/L NADH solution was prepared by diluting the
stock solution with synthetic wastewater, and thereby its EEM fluorescence spectra
were measured and presented in S15. As shown in Figure S15, the pure NADH peak
was located at Ex/Em=260, 350/450;
2) 5 mL NADH stock solution was added into 5 m L dilution water samples of 0 Cu to
adjust the NADH concentration to 5 mg/L;
3) Then the mixture solution was shook up for the measurement of fluorescence
spectra by using an F-4600 fluorescence spectrophotometer equipped with a 150W
xenon lamp (Hitachi, Japan).
4) Finally, the EEM fluorescence spectra of the water samples dosed with NADH
were also conducted on PARAFAC analysis according the same procedure in
MATLAB R2012b.
The fluorescence maps of component C3 in the water samples dosed with NADH
at 0 Cu2+ concentrations were shown in Figure 9b. The fluorescence peak locations
were similar to those of Figure 9a, while the corresponding fluorescence peak
intensity was strengthened significantly, indicating the dosed analytes were in
accordance with the fluorophores in the water samples from duckweed system. It
convincingly demonstrated that the component C3 was NADH.
In addition, the fluorescence map of four components in the water samples dosed
with NADH at 0 Cu2+ concentrations was shown in Figure 13b. Compared with the
Figure 13a, the Figure 13b appeared some variations on the shape and position of
components. For example, the shape of component C1 has changed, and there
appeared a new peak at Ex/Em=250/420 in component C4. It was indicated that the
increase of fluorescence intensity of component C3 could influence the shape and
position of component C1and C4, which demonstrated that the presence of other
components in DOM could shift the positions of component due to the interaction
between each component in DOM (Murphy et al., 2011). It also provided another
evidence to explain why our peaks are in unusual positions and to help demonstrate
our cause.
The pH and metal-spiking experiments1. In the pH-spiking experiment, 5 mg/L NADH in synthetic wastewater was adjusted at
different pH values (pH=3.15, 4.20, 5.18, 6.07, 7.06 and 8.01) by using 0.10 mol/L
HCl and NaOH solution. Then the adjusted solution was shook up for the
measurement of fluorescence spectra by using an F-4600 fluorescence
spectrophotometer equipped with a 150W xenon lamp (Hitachi, Japan). Their
fluorescence spectra were presented in Figure S11. As shown in Figure S11, the peaks
of NADH were significantly affected at lower pH value (pH≤4.20). For example,
when the pH value of NADH solution was 4.20, there only was one peak of NADH at
Ex/Em=350/450, and another peak at Ex/Em=260/450 was disappeared. It indicated
that peak of NADH at Ex/Em=350/450 was stable, while another peak at
Ex/Em=260/450 was instable. Thus, the peak of NADH at Ex/Em=350/450 was
usually observed in many studies rather than peak at Ex/Em=260/450. Furtherly,
when the pH was lower (pH=3.15), the peak position of NADH have switched into
the Ex/Em=270/390, suggesting the fluorophore of NADH could appear significant
blue-shifted under highly acidic conditions, and the high concentrations of proton ions
could damage the structure of NADH such as the reduction in the extent of the r-
electron system and reduction of conjugated bonds in a chain structure (Senesi, 1990).
Therefore, it was demonstrated that high concentrations of proton ions could affect
the position of fluorescence components, and thereby cause unusual components.
2. In the metal-spiking experiment, 5 mg/L NADH in synthetic wastewater were
adjusted into various Cu2+ concentrations (0, 0.1, 0.5, 1.0, 2.0 and 5.0 mg/L). Then the
adjusted solution was shook up for the measurement of fluorescence spectra by using
an F-4600 fluorescence spectrophotometer equipped with a 150W xenon lamp
(Hitachi, Japan). Their fluorescence spectra were presented in Figure S12. As shown
in Figure S12, the fluorescence intensity of NADH was decreased as the Cu2+
concentrations increased, suggesting that there was apparent interaction between
copper ions and NADH, such as the complexation and oxidation-reduction. In
addition, the fluorescence intensity of NADH would disappear at high Cu2+
concentration, which was in accordance with the phenomenon of component C3 at 5.0
mg/L Cu2+ in our previous experiment and additional experiment. Therefore, it was
demonstrated that high concentrations of copper ions could affect the intensity and
shape fluorescence components, and thereby cause unusual components.
In summary, based on the pH-spiking experiment, metal-spiking experiment and
spiking experiment of water sample dosed with NADH, it was confirmed that the high
concentrations of proton and copper ions as well as the presence of other components
could cause the variation of fluorescence spectra. For example, high concentrations of
proton ions could affect the position of fluorescence components, high concentrations
of copper ions could affect the intensity and shape fluorescence components. And the
presence of other components in DOM could shift the positions of component due to
the interaction between each component in DOM, which was in agreement with the
studies of Murphy et al., (2011) who found that other components in DOM could shift
the positions of component. Therefore, the results of these experiments could help us
explain why their peaks were in unusual positions and demonstrate their cause. For
example, the component C2 missed the 220-240 nm excitation peaks for tyrosine-like
fluorescence, which could be explained that the shadow component (220-240 nm
excitation peaks) was caused by the interactions of pH, copper ions and other
component in DOM from duckweed system.
More discussion on the peaks of C4
Additionally, component C4 exhibited a wide range of emission wavelengths
(420-540 nm) in Figure 3 and Figure S8, suggesting that it had the characteristic of
large molecular weight, conjugated aromatic moieties, and complex structure (Huang
et al, 2018), which also was in accordance with the characteristic of inert products
such as humic acids like substances. Meanwhile, it can be observed that the
fluorescence intensity and shape of component C4 had few changes as the Cu2+
concentrations increased from 0 to 4.0 mg/L (Figure 3 and Figure S8), indicating that
component C4 was a kind of inert products. Therefore, according to the above
inference, the component C4 could be identified as humic-like substance. Moreover,
the peak of C4 was located in the forbidden regions (Em>2Ex), and it should be
excluded in traditional fluorescence analysis. However, due to the continuum of
organic molecules in the longer wavelengths (Chen et al., 2003), the exclusion of
forbidden regions might be arbitrary to some extent, and there may be also underlying
fluorescence spectra in the boundary of forbidden regions. Besides, the peaks
occurred in the boundary of forbidden regions, which could be ascribed to the
responses of secondary or tertiary excitation and emission when exposed to different
environment or the resultant extension of peaks in the form of shoulders into longer
emission scan wavelengths or EEMs. Meanwhile, Murphy et al., (2011) found that
other components in DOM could shift the positions of component. Thus, it remains to
be elucidated whether the peak in the boundary of the forbidden regions actually
existed, or its existence is due to the different constraints or criteria imposed during
modeling, or due to an interaction between each fluorescence component.
References
Chen, W., Westerhoff, P., Leenheer, J.A.,Booksh, K. 2003. Fluorescence excitation-emission
matrix regional integration to quantify spectra for dissolved organic matter. Environ. Sci. Technol. 37(24), 5701-5710.Huang, M., Li, Z.W., Huang, B., Luo, N.L., Zhang, Q., Zhai, X.Q. Zeng, G.M. 2018. Investigating
binding characteristics of cadmium and copper to DOM derived from compost and rice straw using EEM-PARAFAC combined with two-dimensional FTIR correlation analyses. J. Hazard. Mater. 344, 539-548.
Murphy, K.R., Adam, H., Sachin, S., Henderson, R.K., Andy, B., Richard, S.,Khan, S.J. 2011. Organic matter fluorescence in municipal water recycling schemes: toward a unified PARAFAC model. Environ. Sci. Technol. 45(7), 2909-2916.
Table S1Ionic concentrations of synthetic swine wastewater
ionic concentrations(mg/L) Value
COD 220±10.51
NH3-N 75.2±7.62
PO4-P 14.6±2.23
NO3-N 98.3±4.14
Ca2+ 120±5.63
Mg2+ 24.5±2.32
K+ 97.5±3.40
Na+ 175.5±2.42
Cl- 285±4.53
SO42- 123±5.72
Fe-EDTA 39.9±2.51
Minor elements 2.54±0.22
Table S2a The important parameters and statistics indexes of the three models for NH3-N removal at various Cu2+ concentrations
Table S3 Peaks of traditional DOM and their descriptionComponent plot Ex (nm)
/Em (nm)
Description Reference
350/440
The peak B at excitation–emission maxima of 350/440 nm was attributed to NADH (Lindemann et al. 1998). NADH is usually generated by the catabolism of glucose to pyruvate via glycolysis process (Nath and Das, 2004). In anaerobic hydrogen-producing process, due to the lack of electron transfer chain, NADH could be accumulated (Morel et al., 2004), and the intracellular NADH was released into the solution as the microorganisms died (Li et al., 2011).
selected componen
ts plots were
collected from Li et al., 2011
270/332
The component was considered as tyrosine-like substance belonging to the type of protein-like materials. The component was also found in other aquatic environments such as high-strength nitrogenous wastewater (Kothawala et al., 2012), lake water (Wang et al., 2009), submerged membrane bioreactor (Fellman et al., 2011).
Selected component plot was collected from Zhu et al. 2017
300/390
Components G2 was identified as microbial humic-like Generally, the generation of microbial by-products from the biodegradation within horizontal subsurface flow constructed wetland system, presumably partially from the leaves and roots of the matured plants (Wei et al., 2009). And it usually is a mixture of fluorescent compounds including humic-like fluorescence fulvi-like and protein-like fluorescence, or the presence of a complex fluorophore in which energy can be transferred internally (Coble and Brophy, 1994).
Selected component
plot was collected
from Murphy et al., 2011
250/250, 500
The component was considered as humic-like substances (Rosario-Ortiz et al., 2007).Besides, Components exhibited a wide range of emission wavelength (420-540 nm), it was suggested that they have the characteristic of large molecular weight, conjugated aromatic moieties, and complex structure (Huang et al., 2018).
Selected component plot was collected from Zhu et al. 2014
ReferenceCoble, P.G., Brophy, M.M. 1994. Investigation of the geochemistry of dissolved organic matter in
coastal waters using optical properties, pp. 377-389.Fellman, J.B., Dogramaci, S., Skrzypek, G., Dodson, W.,Grierson, P.F. 2011. Hydrologic control
of dissolved organic matter biogeochemistry in pools of a subtropical dryland river. Water Resour. Res. 47(6), 667-671.
Kothawala, D.N., Von, W.E., Koehler, B.,Tranvik, L.J. 2012. Selective loss and preservation of lake water dissolved organic matter fluorescence during long-term dark incubations. Sci. Total Environ. 433, 238-246.
Li, W.H., Sheng, G.P., Lu, R., Yu, H.Q., Li, Y.Y.,Harada, H. 2011. Fluorescence spectral characteristics of the supernatants from an anaerobic hydrogen-producing bioreactor. Appl. Microbiol. Biotechnol. 89(1), 217-224.
Lindemann, C., Marose, S., Nielsen, H.O.,Scheper, T. 1998. 2-Dimensional fluorescence spectroscopy for on-line bioprocess monitoring. Sensors & Actuators B Chemical 51(1–3), 273-277.
Morel, E., Santamaria, K., Perrier, M., Guiot, S.R.,Tartakovsky, B. 2004. Application of multi-wavelength fluorometry for on-line monitoring of an anaerobic digestion process. Water Res. 38(14), 3287-3296.
Murphy, K.R., Hambly, A., Singh, S., Henderson, R.K., Baker, A., Stuetz, R.,Khan, S.J. 2011. Organic matter fluorescence in municipal water recycling schemes: toward a unified PARAFAC model. Environ. Sci. Technol. 45(7), 2909-2916.
Nath, K.,Das, D. 2004. Improvement of fermentative hydrogen production: various approaches. Appl. Microbiol. Biotechnol. 65(5), 520-529.
Wang, Z.W., Wu, Z.C.,Tang, S.J. 2009. Characterization of dissolved organic matter in a submerged membrane bioreactor by using three-dimensional excitation and emission matrix fluorescence spectroscopy. Water Res. 43(6), 1533-1540.
Wei, L.L., Zhao, Q.L., Xue, S., Jia, T., Tang, F.,You, P.Y. 2009. Behavior and characteristics of DOM during a laboratory-scale horizontal subsurface flow wetland treatment: Effect of DOM derived from leaves and roots. Ecol. Eng. 35(10), 1405-1414.
Zhu, G.C., Bian, Y.N., Hursthouse, A.S., Wan, P., Szymanska, K., Ma, J., Wang, X.F.,Zhao, Z.L. 2017. Application of 3-D Fluorescence: Characterization of Natural Organic Matter in Natural Water and Water Purification Systems. Journal of Fluorescence 27(9), 1-26.
Rosario-Ortiz, F.L., Snyder, S.A.,Suffet, I.H. 2007. Characterization of dissolved organic matter in drinking water sources impacted by multiple tributaries. Water Res. 41(18), 4115-4128.
Huang, M., Li, Z.W., Huang, B., Luo, N.L., Zhang, Q., Zhai, X.Q.,Zeng, G.M. 2018. Investigating binding characteristics of cadmium and copper to DOM derived from compost and rice straw using EEM-PARAFAC combined with two-dimensional FTIR correlation analyses. J. Hazard. Mater. 344, 539-548.
Figure S1. First-order kinetic model for NH3-N (a), TP (c) respectively; Michaelis-
Menten kinetic model for NH3-N (b), TP (d) respectively, by duckweed under various
Cu2+ concentrations within 38 days.
0 5 10 15 20 25 30 35 40 450
10
20
30
40
50
60
70
80
90 (a)
NH
3-N (m
g/L)
Cultivation (day)
0 Cu 0.1 Cu 0.5 Cu 1.0 Cu 2.0 Cu 5.0 Cu
0 5 10 15 20 25 30 35 40 45
0
3
6
9
12
15
18 (c)
TP (m
g/L)
Cultivation (day)
0
3
6
9
12
15
18
0 5 10 15 20 25 30 35 40 45
(d)
Cultivation (day)
TP (m
g/L)
10
20
30
40
50
60
70
80
90
0 5 10 15 20 25 30 35 40 45
(b)
Cultivation (day)
NH
3-N (m
g/L)
Figure S2. Growth curves for duckweed at various concentrations of Cu2+ within 38
days.
Figure S3. The assimilation of Cu2+ by duckweed at various initial concentrations of
Cu2+ within 38days.
Figure S4. Fluorescence quenching curves of each PARAFAC-derived component at
various initial concentrations of Cu2+. The curve corresponds to the decrease of C3
component.
0
10
20
30
40
50
60
70
80
38 d32 d8 d0 d 26 d21 d12 d4 d
DO
C (m
g/L)
0 Cu 0.1 Cu 0.5 Cu 1.0 Cu 2.0 Cu 5.0 Cu
Figure S5. DOC concentrations at various concentrations of Cu2+ within 38 days.
(a)
(f)(e)
(d)(c)
(b)
Figure S6. The original fluorescence images of sample in 21th day under various Cu2+
concentrations in the duckweed systems. (a) 0 mg/L Cu2+; (b) 0.1 mg/L Cu2+; (c) 0.5