- Email: [email protected]
H2O2 in synthetic urine
Accepted Manuscript Removal of sulfonamide antibiotics and human metabolite by biochar and biochar/ H2O2 in synthetic urine Peizhe Sun, Yaxiu Li, Tan Meng, Ruochun Zhang, Min Song, Jing Ren PII:
S0043-1354(18)30766-8
DOI:
10.1016/j.watres.2018.09.051
Reference:
WR 14101
To appear in:
Water Research
Received Date: 11 June 2018 Revised Date:
18 September 2018
Accepted Date: 20 September 2018
Please cite this article as: Sun, P., Li, Y., Meng, T., Zhang, R., Song, M., Ren, J., Removal of sulfonamide antibiotics and human metabolite by biochar and biochar/H2O2 in synthetic urine, Water Research (2018), doi: https://doi.org/10.1016/j.watres.2018.09.051. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
1
Removal of sulfonamide antibiotics and human metabolite by biochar and biochar/H2O2 in synthetic urine
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
Peizhe Sun1, Yaxiu Li1, Tan Meng1, Ruochun Zhang2*, Min Song3, Jing Ren4 1 School of Environmental Science and Engineering, Tianjin University, Tianjin 300072 China 2 Institute of Surface-Earth System Science, Tianjin University, Tianjin 300072 China 3 Ministry of Education of Key Laboratory of Energy Thermal Conversion and Control, School of Energy and Environment, Southeast University, Nanjing, Jiangsu 210096, China 4 School of Environmental Sciences, Liaoning University, Chongshan Road No.66, Shenyang 110036 China
SC
RI PT
2
Corresponding author: Ruochun Zhang; Email: [email protected]
AC C
EP
TE D
M AN U
Abstract Source-separated urine has been increasingly regarded as a promising alternative wastestream for effectively removing pharmaceuticals and human metabolites. This study investigated the removal of sulfonamide antibiotics, one category among the most frequently detected antibiotics in the environment, by biochar and biochar/H2O2 in synthetic urine matrix. The adsorption and degradation of four parent sulfonamide antibiotics, including sulfamethoxazole, sulfadiazine, sulfamethazine, sulfadimethoxine, and one human metabolite, N4-acetylsulfamethoxazole (together referred as SAs) were investigated. Biochar derived from cotton straw was applied as adsorbent for SAs and catalyst for H2O2. Results showed that the adsorption of SAs was inhibited in urine compared with that in phosphate buffer solution. Bicarbonate in urine placed major influence. Langmuir isotherm model well described the adsorption process in both buffer and urine matrices. Adsorption and desorption rates were estimated by a kinetic model, which well fitted the removal of SAs from aqueous phase at various biochar doses. The adsorption of SAs on biochar was due to multiple forces, in which van der Waals forces and hydrophobicity played major roles in distinguishing the sorption behavior of different SAs. To destruct the SAs, H2O2 was added with biochar. Except for N4-acetyl-sulfamethoxazole, all the parent SAs can be degraded in urine matrix. Carbonate radical, produced from the activation of peroxymonocarbonate by biochar, was proposed to be the major contributing reactive species in biochar/H2O2 system in urine matrix. Keywords: Source-separated urine, biochar, adsorption, sulfonamide antibiotics, H2O2 activation, carbonate radical
38
1
ACCEPTED MANUSCRIPT
39
1 Introduction Pharmaceutical residue in the environment places a major threat on the ecosystem as well as
41
human health. Wastewater treatment plants (WWTPs) has been regarded as an important source
42
for pharmaceuticals in the natural aquatic environments (Liu and Wong, 2013). Among all the
43
municipal wastewater streams entering into WWTPs, human urine accounts for less than 1% by
44
volume but contains pharmaceuticals at 2−3 orders of magnitude higher concentrations (Eawag
45
Swiss Federal Institute of Aquatic Science and Technology, 2007, Winker et al., 2008a, Winker
46
et al., 2008b). Separation of urine from other wastewater streams, namely source-separated urine,
47
has gained increasing attentions, because it is expected to be more effective for pharmaceutical
48
removal in source-separated urine than other conventional domestic wastewater. Moreover,
49
because urine contributes more than 75% total nitrogen and 50% total phosphorus in wastewater,
50
recovering nutrients from urine, as a new resource recovery strategy (Winker et al., 2008a), also
51
requires elimination of pharmaceuticals from urine. Multiple processes, including nano-filtration
52
membranes (Pronk et al., 2006), anion-exchange resin (Landry and Boyer, 2017), electrodialysis
53
(Pronk et al., 2007), struvite precipitation (Kemacheevakul et al., 2014), and advanced oxidation
54
processes (Zhang et al., 2015) have been investigated for removing pharmaceuticals from urine.
55
However, most of the investigated methods require extensive energy, material input or upgrading
56
infrastructures, and thus hinder their application. A more cost-effective process will better
57
facilitate the application of removing pharmaceuticals from their source.
AC C
EP
TE D
M AN U
SC
RI PT
40
58
Biochar is a carbonaceous material which can be produced from pyrolysis of different
59
sources of biomass, including plant materials and wastes, sludges, manures, litters, sediment, etc.
60
(Cha et al., 2016, Qambrani et al., 2017, Rosales et al., 2017). Considering cost and energy
61
saving, carbon sequestration, as well as solid waste reuse, biochar is regarded as a promising
2
ACCEPTED MANUSCRIPT
alternative adsorbent for wastewater treatment. Biochar has been proved to be effective for
63
removing varieties of pharmaceuticals including antibiotics (Wu et al., 2013, Yao et al., 2012a),
64
antiepileptic drugs (Chen et al., 2017), anti-inflammatory drugs (Lin et al., 2017), antidepressants,
65
anxiolytic and antihistaminic drugs (Calisto et al., 2015). In addition to conventional wastewater
66
treatment system, the application of biochar has also been proposed to unconventional systems,
67
such as stormwater infiltration system (Ulrich et al., 2015), reclaimed water desalination
68
concentrate (Lin et al., 2017), and other environmental compartment such as soil (Vithanage et
69
al., 2014).
SC
RI PT
62
Although it is expected that human urine is a promising matrix for effective pharmaceutical
71
removal, few studies have been carried out applying biochar. Solanki and Boyer have
72
investigated the removal of several pharmaceuticals (excluding antibiotics) in synthetic human
73
urine using biochar, but without including adsorption kinetics and isotherm studies (Solanki and
74
Boyer, 2017). The impact of urine components on pharmaceutical removal has not been
75
investigated. In addition to adsorption capacity, biochar has been reported to activate H2O2 or
76
persulfate to produce highly reactive species, which can degrade organic pollutants (Fang et al.,
77
2014, Fang et al., 2015). However, so far, few studies have been conducted for removal of
78
pharmaceuticals by radical/nonradical reactive species generated from biochar activation. And
79
the technique has not been investigated in source-separated urine matrix.
AC C
EP
TE D
M AN U
70
80
Sulfonamide antibiotics (SAs) are among the most frequently detected pharmaceuticals in
81
water systems (Bu et al., 2013, Li et al., 2017, Zhang and Li, 2011). Adsorption of
82
sulfamethoxazole (SMX) (Calisto et al., 2015, Jung et al., 2013, Wu et al., 2013, Zheng et al.,
83
2013), sulfamethazine (SMT) (Peng et al., 2016, Vithanage et al., 2014), sulfadiazine (SDZ)
84
(Peng et al., 2016), and sulfapyridine (Inyang et al., 2015, Xie et al., 2014) using biochar or
3
ACCEPTED MANUSCRIPT
modified/activated biochar has been reported. However, urine matrix is very different from
86
conventional wastewater in terms of pH and major components, thus may pose different impact
87
on adsorption process. Moreover, in urine treatment, metabolites of the pharmaceuticals should
88
also be considered. For example, only 9.5% of sulfamethoxazole was excreted in unchanged
89
form while most was excreted as metabolites (Vree et al., 1978).
RI PT
85
The objective of this study, therefore, was to investigate the kinetics and mechanisms of the
91
removal of antibiotics and human metabolite by biochar (-catalyzed) system in urine matrix.
92
Specifically, the adsorption and degradation of four parent SAs and the major human metabolite
93
of SMX were investigated. Considering the relatively long operating period, hydrolysis of urea
94
in fresh urine (referred to the urine that just left human body) will highly likely occur. Therefore,
95
hydrolyzed urine (referred to the urine that the original composition has changed due to storage)
96
was chosen in this study. This work presents SAs adsorption isotherm, kinetics and mechanisms
97
on biochar in buffer and hydrolyzed urine matrices. Chemical destruction of SAs by biochar-
98
activated H2O2 was also investigated.
99 2 Materials and methods
101
2.1 Materials
EP
100
TE D
M AN U
SC
90
Sulfamethoxazole (SMX), sulfadiazine (SDZ), sulfamethazine (SMT), sulfadimethoxine
103
(SDM), and N4-acetyl-sulfamethoxazole (NSMX) were purchased from Sigma Aldrich Inc. (St.
104
Louis, MO). All the five compounds are referred as SAs, whereas the parent SAs represents
105
SMX, SDZ, SMT and SDM. Structures and chemical properties of antibiotics are shown in Table
106
1 and Table S1. H2O2 (30%, w/w) was purchased from Aladdine (Shanghai, China). HPLC grade
107
methanol was purchased from CNW (Düsseldorf, Germany). All other chemicals and reagents
AC C
102
4
ACCEPTED MANUSCRIPT
used were in analytical grade or higher. Reagent grade deionized (DI) water (resistivity > 18 mΩ
109
cm) was prepared from a nanopure water purification system (Veolia, Paris, France). Biochar
110
was produced from pyrolysis of shredded cotton stalks, which were air-dried under 80 oC and
111
milled to pass through a 40 mesh screen. The pyrolysis condition was 350 °C with the absence of
112
oxygen.
RI PT
108
113 2.2 Adsorption experiment
SC
114
The batch adsorption experiments were carried out in 10 mL glass vials with screw caps.
116
Reaction solutions were prepared with 10 µM SA in 10 mL of synthetic urine or phosphate
117
buffer solution (PBS) at pH 9. The synthetic urine composition has been described in previous
118
literature (Zhang et al., 2015) and shown in Table 2. Due to hydrolysis of urea, high
119
concentrations of ammonia nitrogen (0.5 M) and bicarbonate (0.25 M) were present in
120
hydrolyzed urine, which also resulted in elevated pH (pH 9). PBS (with 5 mM phosphate) at pH
121
9 was employed as a baseline matrix. The vials were kept in dark and shaken at 220 rpm, 25 °C.
122
For kinetic study, biochar was applied at 0.1, 0.5, 1, 2, 5 g L−1, respectively, and aliquots were
123
taken at 1, 5, 10, 24, 120 h. For isotherm study, two more biochar doses, 0.7 g L−1 and 1.5 g L−1,
124
were added to provide more accurate results. Samples were shaken for 120 h to ensure
125
equilibrium. Before analysis, the reaction solutions were filtered with 0.45 µm polyethersulfone
126
syringe filters. Individual SAs were incubated without biochar under each corresponding
127
condition as the control samples. NaCl was applied to study the impact of ionic strength on
128
adsorption. To investigate the impact of major urine components on adsorption, bicarbonate and
129
ammonia were added separately with 1 g L−1 biochar and 10 µM SAs. Concentrations of the
130
components were the same as those in synthetic urine solution. Solution pH was adjusted by
AC C
EP
TE D
M AN U
115
5
ACCEPTED MANUSCRIPT
131
concentrated NaOH and H3PO4 solutions to pH 9. Negligible change of pH was observed over
132
the entire contact period, for both adsorption and degradation experiments. Each experiment was
133
conducted by at least duplicates.
data.
136
=
137
=
⁄
SC
135
RI PT
Freundlich (Eq. (1)) and Langmuir (Eq. (2)) isotherm model were used to fit the adsorption
134
(1) (2)
where qe is equilibrium adsorbent-phase concentration of SA, in µmol g−1; Kf is Freundlich
139
adsorption capacity parameter, in µmol g−1 (µM−1)1/n, 1/n is Freundlich adsorption intensity
140
parameter; Ce is equilibrium aqueous-phase concentration of SA, in µM; QM is Langmuir
141
maximum adsorption capacity, in µmol g−1; and bSA is the Langmuir adsorption constant of SA,
142
in µM−1. To better compare the parameters for different SAs, the amount of SA was quantified
143
using mole-based unit (µmol) instead of mass-based unit (mg).
TE D
M AN U
138
On the basis of Langmuir isotherm assumption that the adsorbates and adsorbent surface
145
sites interact as a reversible chemical equilibrium, the overall adsorption process of SAs onto
146
biochar in PBS and urine matrices was expressed using the following equation.
EP
144
AC C
k SA / kSA−1 [ SA]aq + [ BC ]vs ← →[ SA − BC ]
147 148
(3)
149
where, [SA]aq is the aqueous phase concentration of SAs, in M; [BC]vs is the vacant sites
150
concentration of biochar in the solution, in M; [SA-BC] is the concentration of SAs adsorbed
151
onto biochar, in M; kSA, in M−1 h−1, and kSA-1, in h−1, are the adsorption and desorption rates,
152
respectively. Therefore, at t = 0, [BC]vs = (biochar dose) × QM and [SA]aq = 10 µM.
153 6
ACCEPTED MANUSCRIPT
154
2.3 Degradation experiment To chemically destruct SAs, 1 mM H2O2 was added to both PBS and urine solution with 1 g
156
L−1 biochar. The samples were shaken for 1 h and then prepared for analysis. Samples after
157
reactions were treated in two ways. Extracting agent was added to recover both the residual SAs
158
that adsorbed on biochar and remained in aqueous phase. Solution containing methanol:
159
NaHCO3 9:1 (reason for adding NaHCO3 is elucidated in 3.4) was applied as extracting agent
160
and was added to samples with equal volume. After 30 min extraction, samples were filtered and
161
analyzed by HPLC. Recovery were 95%, 80%, 89%, 82%, and 90% for 10 µM SMX, SDZ,
162
SDM, SMT, and NSMX respectively.
164
2.4 Quantification of SAs
M AN U
163
SC
RI PT
155
A Waters e2695 HPLC system (Waters, Milford, MA, USA) equipped with a UV-vis
166
detector and a Zorbax-SB C18 column (4.6×150 mm, 5 µm) was used to monitor the loss of the
167
SAs. Gradient elution was used with 0.1% phosphoric acid and methanol as mobile phase. The
168
portion of methanol stayed at 10% for the first minute, then ramped to 50% at 6 min, and
169
decreased to 10% from 10 min to 15 min. The peak area of SAs was quantified at 254 nm.
EP
171
2.5 Characterization of biochar
AC C
170
TE D
165
172
Information of surface area and pore size of biochar was obtained from a surface area and
173
pore distribution analyzer (V-Sorb 2800TP, Gold APP Instruments Corporation, China). Fourier
174
transform infrared (FTIR) spectra were performed to qualitatively analyze the functional groups
175
of adsorbent (Vector 22 + TGA Bruker Optics, Germany) with the wavenumber range of 400–
7
ACCEPTED MANUSCRIPT
176
4000 cm−1. The point of zero charge (pzc) was measured by a pH meter (S220 SevenCompact,
177
Mettler Toledo, Shanghai).
178 2.6 Estimation of SA properties
RI PT
179
The pKa values of the secondary amine groups on SAs were obtained from previous
181
literatures (Babić et al., 2007, Bonvin et al., 2013). Solubility of the neutral form of SAs was
182
estimated by ECOSAR (v1.11) program. The apparent solubility of SAs was calculated and
183
detailed in section 3.3. The octanol-water distribution coefficient (Kow) was determined
184
experimentally by mixing equal amount of octanol with water solution containing 20 µM
185
individual SAs. The water phase was buffered with formic acid at pH 3.5. Molar refractivity and
186
electronegativity were estimated by ChemBio3D Ultra (v12.0).
M AN U
SC
180
187 3 Result and discussion
189
3.1 Biochar characterization
TE D
188
Biochar is proposed as a promising low-cost adsorbent because of production at relatively
191
low temperature compared with activated carbon. In order to demonstrate the performance of
192
low-cost biochar on the removal of pharmaceuticals in urine, biochar derived from cotton straw,
193
produced at 350 oC, was used in this study. It is also worthwhile noting that biochar produced at
194
low temperature retains more nutrient, such as total nitrogen, than those produced at higher
195
temperatures. thus potentially benefits future application as soil amendment (Shinogi and Kanri,
196
2003, Yu et al., 2013). The properties of the biochar derived from cotton straw are shown in
197
Table 3.
AC C
EP
190
8
ACCEPTED MANUSCRIPT
Generally, adsorption capacity is considered positively related to the specific surface area
199
and pore volumes. Therefore, comparing with the biochar materials through pyrolysis at
200
comparable temperatures, the properties of the biochar derived from cotton straw (shown in
201
Table 3) was better than peanut shell (Chen et al., 2017), pepper wood, sugarcane bagasse, and
202
hickory wood (Yao et al., 2012b), sediment (Wu et al., 2013), tea waste (Rajapaksha et al.,
203
2014), rice straw, wheat straw (Sun et al., 2016) and coconut shell (Solanki and Boyer, 2017),
204
but slightly worse than biochar produced from pine wood related materials (Jung et al., 2013, Xie
205
et al., 2014). The FTIR result (Fig. S1) indicated that hydroxyl, carbonyl and aromatic moieties
206
dominated on the surface of biochar, indicating interactions with pharmaceuticals through H-
207
bond and π-π interaction may likely occur. The point of zero charge was measure to be around
208
7.4 (Fig. S2), which suggested that the surface of biochar was negatively charged in PBS (pH 9)
209
and urine matrix.
M AN U
SC
RI PT
198
TE D
210 211
3.2 Adsorption of SAs in PBS and urine
212
3.2.1 Kinetics
The concentrations of SAs in aqueous phase versus time are shown in Fig. 1. Control
214
groups showed that SA concentrations remained unchanged without biochar (data not shown),
215
confirming that the decreasing concentration of SAs in aqueous phase was due to interaction
216
with biochar. SAs in samples with biochar for 120 h were extracted by methanolic solution, and
217
showed that degradation did not occur. SA adsorption occurred rapidly within the first 24 h with
218
0.5 – 5 g L−1 biochar. From 24 h to 120 h, the adsorption rates significantly decreased and
219
equilibrium was reached. For solutions with 0.1 g L−1 biochar, equilibrium was attained within 1
220
h, while the removal was only around 10%. With the increase of biochar doses, the removal of
AC C
EP
213
9
ACCEPTED MANUSCRIPT
SAs from aqueous phase significantly increased (Fig. 1). Indeed, the overall removal is higher
222
with higher adsorbent dose due to increasing adsorption sites. For the adsorption of
223
pharmaceuticals onto biochar, time to reach adsorption equilibrium may vary from less than 30
224
min (e.g., (Calisto et al., 2015) ) to several days (e.g., (Wu et al., 2013)) due to different
225
properties of adsorbent and adsorbate. In our study, 120 h was deemed sufficient to reach
226
apparent equilibrium. Therefore, 120 h was chosen for following isotherm study.
RI PT
221
Independent with biochar doses, overall removal of SAs in PBS ranked as SDM ≈ SMT ≈
228
NSMX > SMX ≈ SDZ. For example, the removal ratios at 1 g L−1 biochar were 71%, 74%, 68%,
229
49%, and 39%, for SDM, SMT, NSMX, SMX, and SDZ, respectively. In urine, similar ranking
230
was observed, whereas the removal was inhibited to different extent compared with in PBS (Fig.
231
S3). Inhibition was negligible when biochar doses were low because only small amount of SA
232
can be adsorbed even when there was no competitor. When biochar doses were relatively high,
233
inhibition was also limited because adsorption sites were sufficient for both SAs and urine
234
components that will adsorbed onto biochar. Strongest inhibition (urine vs. PBS) was observed
235
with 1 g L−1 or 2 g L−1 biochar for each SA.
236 237
3.2.2 Isotherm
EP
TE D
M AN U
SC
227
The adsorption isotherms for five SAs in both PBS and urine are shown in Fig. 2. Generally,
239
the sorption data were well described by both Freundlich and Langmuir models (shown in Fig. 2,
240
Fig. S4). For most conditions, Langmuir model presented relatively higher regression
241
coefficients (R2) (Table S2), indicating the overall adsorption process followed reversible
242
monolayer sorption mechanism. Therefore, Langmuir isotherm model was selected for
243
subsequent analysis and the fitting parameters are shown in Table 1. QM values are almost the
AC C
238
10
ACCEPTED MANUSCRIPT
same for all SAs, indicating sorption sites for different SAs were likely very similar. The subtle
245
differences of QM values for PBS and urine matrix indicated urine components did not
246
irreversibly adsorb onto biochar (except for SDZ). The abnormal QM value for SDZ in urine may
247
indicate a different mechanism from other SAs and further study is needed. QM value of this
248
study was comparable with the values reported in previous literatures for biochar produced under
249
similar condition from different materials, e.g., 2.79 mg g−1 (10.04 µmol g−1), 1.39 mg g−1 (5.0
250
µmol g−1), and 5.75 mg g−1 (20.68 µmol g−1) for SMT with tea waste biochar (Rajapaksha et al.,
251
2014), wheat biochar, and maize straw biochar (Jia et al., 2017); 4.21 mg g−1 (16.64 µmol g−1)
252
and 6.75mg/g (26.68 µmol g−1) for SMX with rice biochar and wheat biochar (Sun et al., 2016).
M AN U
SC
RI PT
244
In PBS, bSA values ranked as SDM > NSMX > SMT > SDZ > SMX; in urine, the ranking
254
was SDM > SMT > SDZ > NSMX > SMX. In addition, urine matrix significantly decreased bSA
255
for all the SAs from 1.7 times (SMT) to 7.1 times (NSMX) (Table 1). The decrease indicates that
256
urine components may compete with SAs adsorbing onto biochar. Same as QM values, bSA for
257
SDZ in urine was abnormal and even higher than in PBS. The difference among SAs in
258
adsorption kinetic and equilibrium was likely resulted from different chemical properties of SAs,
259
which was further discussed in section 3.3.
EP
261
3.2.3 Modeling of adsorption process
AC C
260
TE D
253
262
Applying Matlab Simbiology application in Matlab program, kSA and kSA-1 (in eqn. 3) were
263
obtained by fitting the data at 1 g L−1 biochar in PBS or in urine. All sorption rates are shown in
264
Table 1. The adsorption rates (kSA) of SAs in PBS matrix ranks as SDM > SDZ > SMT > NSMX >
265
SMX, whereas SMX has the highest desorption rate (kSA-1) followed by SDZ, SMT, NSMX and
266
SDM. In urine matrix, SDM and NSMX adsorbed faster than other SAs, whereas the desorption
11
ACCEPTED MANUSCRIPT
rates ranked as SMX > NSMX > SDZ > SMT > SDM. Generally, both the adsorption and
268
desorption rates of most SAs increased in urine matrix, indicating less time was required to reach
269
adsorption equilibrium. Applying the estimated sorption rates, the decrease of aqueous
270
concentrations of SAs with the presence of various doses of biochar were then simulated and
271
plotted in Fig. 1. Overall, the model has good agreement with the experimental data at various
272
biochar doses.
RI PT
267
To further evaluate this model, adsorption experiment in PBS was conducted by mixing five
274
SAs at 10 µM with 1 g L−1 biochar. The aqueous concentrations of SAs were measured after 3-
275
day adsorption. As shown in Fig. S5, the model successfully predicted the removal of all five
276
SAs from aqueous phase, which also confirmed the hypothesis that all five SAs adsorbed on
277
similar sites on biochar, thus competing with each other for the same adsorption sites when co-
278
existing in solution.
M AN U
SC
273
The adsorption rates of SAs represented the energy barrier which was required to overcome
280
for SAs to adsorb onto biochar surface. Higher adsorption rates suggested lower energy barrier
281
and higher attraction force. As reported in previous studies, the adsorption of SAs onto biochar
282
can be resulted from π+-π interactions, Coulombic forces, hydrogen bonds, van der Waals forces
283
and hydrophobic effects (Teixidó et al., 2011). In PBS matrix, the difference between individual
284
SAs may likely resulted from different molecular properties. Urine components altered the
285
surface properties of biochar, leading to the change of sorption rates. Therefore, in order to
286
elucidate the mechanisms of overall sorption process, the effects of SA properties and individual
287
urine components were evaluated and discussed in the following sections.
AC C
EP
TE D
279
288 289
3.3 Effect of pharmaceutical properties
12
ACCEPTED MANUSCRIPT
The SAs investigated in this work possessed benzene ring and negative charges on the
291
secondary amine nitrogen at pH 9, which are likely important interaction moieties with biochar
292
surface. Indeed, SAs may interact with biochar surface through Coulombic forces of amine
293
groups and π+-π interactions of benzene ring (Teixidó et al., 2011). In addition, hydrophobicity
294
of SAs may also play important role due to favorable partitioning of hydrophobic compounds to
295
the aromatic carbon-riched biochar surface from aqueous phase. Therefore, key chemical
296
properties, including octanol-water partitioning coefficient (Kow), solubility, acidity constant
297
(pKa), electronegativity and molar refractivity, of all five SAs were estimated or recited from
298
previous literatures (shown in Table 1) in order to correlate chemical properties to the different
299
sorption behavior of SAs.
M AN U
SC
RI PT
290
The ranking of pKa values of the secondary amine on SAs can be roughly considered as the
301
electron donating tendency of the side structures. Indeed, generally, SAs with lower pKa have
302
more negative charge on the secondary amine nitrogen (Table 1), suggesting lower tendency to
303
interact with the negatively charged biochar surface. However, neither pKa nor electronegativity
304
correlated with any kinetic parameters (i.e. bSA, QM, kSA, kSA-1). Therefore, interactions due to
305
electron interactions were not major forces distinguishing the sorption behavior of SAs.
EP
TE D
300
The van der Waals forces between adsorbate and adsorbent can roughly correlated with
307
molar refractivity (Ghose and Crippen, 1987). The molar refractivity of five SAs was estimated
308
by the ChemBio3D Ultra (v12.0) and shown in Table 1. SDM has the highest molar refractivity,
309
followed by SMT, NSMX, SMX, and SDZ. When plotting molar refractivity with bSA values in
310
PBS (Fig. S6), a good linear correlation was obtained, indicating that van der Waals forces play
311
an important role in distinguishing the adsorption equilibrium of different SAs onto biochar.
AC C
306
13
ACCEPTED MANUSCRIPT
312
However, van der Walls forces cannot explain the kinetic differences of SAs. Indeed, poor
313
correlations were obtained when plotting molar refractivity with kSA or kSA-1. The adsorption of hydrophobic organic compounds onto carboneous materials, such as
315
activated carbon, has been linked with their partitioning behavior between water and water-
316
immiscible organic solvent (Delgado et al., 2012). Kow has been widely applied to explain
317
sorption process. As for ionizable organic compounds, the pH-dependent octanol–water
318
distribution coefficient (D) was applied in order to take account speciation of compounds (Jung
319
et al., 2013). For SAs, D can be expressed as:
321
[ SA0 ]oct + [ SA− ]oct [SA0 ]water + [ SA− ]water
(4)
M AN U
D=
320
SC
RI PT
314
where [SA0]oct is the concentration of the neutral species present in the octanol phase, [SA-
323
]oct is the concentration of the ionized species present in the octanol phase, [SA0]water is the
324
concentration of the neutral species in the water phase, and [SA-]water is the concentration of the
325
ionized species in the water phase.
TE D
322
The relation between D and Kow can be expressed as:
327
1 1 1 ' D = K ow + K ow ≈ K ow 1 − pH − pKa pH − pKa pH − pKa 1 + 10 1 + 10 1 + 10
328
where (Kow) is the octanol–water partitioning coefficient of the neutral species, K′ow is the
329
octanol–water partitioning coefficient of the fully ionized species. For hydrophobic species, the
330
ionized species is much more soluble in water, thus K′ow is much lower than Kow. Therefore, D is
331
approximately a function of Kow and pH. Applying the Kow values measured in this study, the D
332
values of SAs were calculated and shown in Table 1. The same assumptions were also applied to
(5)
AC C
EP
326
14
ACCEPTED MANUSCRIPT
333
the estimation of solubility. Therefore, the apparent solubility at pH 9 was governed by the
334
solubility limit of neutral species (Eq. (6)).
336
RI PT
[H + ] + Ka S app ≈ S 0 + [H ]
335
(6)
where Sapp is the apparent solubility, in M; S0 is the solubility limit of neutral form, in M.
338
Attempts were made to correlated hydrophobicity related properties (i.e. Kow, D, S0, Sapp) with
339
sorption constants (i.e. bSA, QM, kSA, kSA-1). By plotting each property with sorption constants (Fig.
340
S7-S11), we found that (1) D values had linear relation with bSA in PBS with the exception of
341
NSMX; (2) the desorption rates of SAs in both PBS and urine increased with the increase of
342
apparent solubility, with the exception of NSMX. Therefore, hydrophobicity was crucial in
343
distinguishing different SAs.
M AN U
SC
337
The distribution coefficient, D, represents the difference between free energy in aqueous
345
phase and organic phase. Higher D values suggests lower energy state in aqueous solution. bSA
346
represents the difference between free energy in aqueous phase and on biochar surface. As
347
shown in Fig. S7, good linear relation was obtained between D and bSA for SMX, SDZ, SDM,
348
and SMT, expect for NSMX. This results suggested that free energy values of SAs on biochar
349
surface likely have the same trend as those in organic phase. On the other hand, water solubility
350
also represents the difference between free energy in aqueous phase and solid state. However,
351
poor correlation was obtained between water solubility and bSA. Indeed, the free energy of solid
352
state represents bond interactions other than solvation effects. Therefore, water solubility was
353
more suitable to describe sorption rates which were governed by bond interactions. As shown in
354
Fig. S8-S11, there is good correlation between water solubility and kSA and kSA-1 for all SAs,
AC C
EP
TE D
344
15
ACCEPTED MANUSCRIPT
355
expect NSMX. The unusual behavior of NSMX suggested the importance of aniline moiety on
356
the sorption process.
357
3.4 Effect of urine components
RI PT
358
The impact of major individual urine components was tested by spiking corresponding
360
amount of components into PBS with the presence of 1 g L−1 biochar. The aqueous
361
concentrations of SAs were then measured after 120-h adsorption process. Inhibition value was
362
calculated using the following equation.
364
(%)
(7)
[ SA]aq ,i /[ SA]T − [ SA]aq , PBS /[ SA]T 1 − [ SA]aq , PBS /[ SA]T
M AN U
Inhibition
363
SC
359
=
where [SA]aq,i is the aqueous phase concentration of SA with urine component i, in µM;
366
[SA]aq,PBS is the aqueous phase concentration of SA in PBS, in µM, [SA]T is the total
367
concentration of SA spiked in the solution, in µM.
TE D
365
Due to the presence of high concentrations of ionic species, the ionic strength effect on the
369
adsorption of SAs was first evaluated by adding additional amount of NaCl into PBS (5 mM) to
370
achieve the same ionic strength as urine matrix (i.e. 0.48 M). As shown in Fig. 3, the increase of
371
ionic strength by NaCl posed little impact on the sorption equilibrium, suggesting the different
372
sorption behavior in PBS and urine was not due to ionic strength difference. Total ammonia
373
nitrogen (TAN) and carbonate species were dominant species in urine matrix. The impact of
374
them was investigated in solutions containing 0.5 M TAN and 0.25 M bicarbonate, separately.
375
The solution pH was adjusted to 9. As shown in Fig. 3, TAN showed minimal impact on the
376
adsorption equilibrium for SAs. For SDM and NSMX, the addition of TAN even slightly
AC C
EP
368
16
ACCEPTED MANUSCRIPT
377
increased the sorption amount on biochar. However, bicarbonate greatly inhibited the adsorption
378
of all five SAs. Ionic species, such as chloride, ammonium and bicarbonate, likely affect the sorption of
380
ionizable organic compounds on biochar (Kah et al., 2017). On the basis of pzc value (7.4) of the
381
biochar applied in this study, most of the biochar surface possesses negative charges at pH 9.
382
Meanwhile, all SAs were predominantly (>98%) in their anionic form because the solution pH
383
was much higher than their pKa values (Table 1). A negative charge-assisted H-bond, (-)CAHB
384
(X…H…Y)-, was likely one of the major interactions between SAs and biochar at pH 9, as
385
postulated by Teixido et al (Teixidó et al., 2011). Indeed, the formation of SO2N…H…O likely
386
occurred through the interaction between negatively charged -SO2NH- group on SAs and surface
387
carboxyl or hydroxyl group on biochar due to close pKa (pzc) values (Li et al., 2013). It was
388
suggested that carboxylate or phenolate anions tend to form (-)CAHBs (Li et al., 2015).
389
Analogously, bicarbonate likely interacted with biochar surface through (-)CAHBs, therefore
390
competed with -SO2NH- group for biochar sorption sites. Due to the incapability to form (-
391
)CAHBs, chloride and ammonia/ammonium ions posed much less impact on the sorption of SAs
392
on biochar, compared with bicarbonate. The sorption of SAs was inhibited to different extent by
393
bicarbonate approximately following the order of molar refractivity (Fig. 3 and Table 1), which
394
suggested that both van der Waals forces and H-bonds were major interactions between SAs and
395
biochar. Based on the proposed sorption mechanisms, an extraction solution, containing
396
methanol and bicarbonate, was used to accelerate desorption of SAs from biochar and promote
397
extraction recovery.
AC C
EP
TE D
M AN U
SC
RI PT
379
398 399
3.5 Degradation of SAs with biochar/H2O2
17
ACCEPTED MANUSCRIPT
Before biochar can be applied as soil amendment, the contaminants adsorbed on it requires
401
chemical destruction to avoid secondary pollution. Based on previously reported research,
402
biochar may activate H2O2 and generate hydroxyl radical (Fang et al., 2014). Hydroxyl radical is
403
capable of degrading various pharmaceuticals due to its high oxidizing power and low selectivity
404
(Buxton et al., 1988). In this study, H2O2 was added to both PBS and urine matrix to investigate
405
the degradation of SAs with the presence of biochar. Removal was quantified by total extraction
406
by methanolic solution. Data were shown in Fig. 4. After 1 h, the degradation of SAs was not
407
observed in PBS, which implied that the biochar applied in this study cannot activate H2O2 to
408
generate reactive species (e.g., hydroxyl radical) that can degrade SAs at pH 9, or the reactive
409
species were scavenged immediately after generated. However, in urine matrix, SMX, SDZ,
410
SDM, and SMT were degraded by approximately 60% within 1 h, whereas the degradation of
411
NSMX was not observed. Control experiments have shown that SAs were stable with the
412
presence of H2O2 in urine matrix without biochar (data not shown). Thus, the involvement of
413
biochar was crucial for the degradation of parent SAs.
TE D
M AN U
SC
RI PT
400
The degradation of SAs in urine by biochar/H2O2 can be explained by two possible
415
mechanisms: (a) H2O2 was activated by biochar producing intermediate(s) which cannot degrade
416
SAs, but reacted with urine component(s) yielding secondary reactive species; and (b) the
417
interaction between urine components and H2O2 generated intermediate(s) which cannot degrade
418
SAs, but be activated by biochar yielding secondary reactive species. To test these two
419
hypotheses, the decomposition of H2O2 by biochar was monitored with and without urine
420
components. As shown in Fig. 5, H2O2 was stable with biochar in PBS, suggesting the biochar
421
cannot activate H2O2 (i.e. hypothesis (a) was incorrect). Among all conditions tested, H2O2 was
422
only decomposed by biochar with the presence of bicarbonate. The reaction between H2O2 and
AC C
EP
414
18
ACCEPTED MANUSCRIPT
bicarbonate produces peroxymonocarbonate (Eq. 8) (Richardson et al., 2000). Preliminary test
424
using DPD (N,N-diethyl-p-phenylenediamine) colorimetric method (Cai et al., 2017) showed
425
significant increase of absorbance in H2O2 solution with bicarbonate, indicating the formation of
426
peroxymonocarbonate. Peroxymonocarbonate has been reported to produce carbonate radical by
427
activation by transition metals (Pi et al., 2018). Similar activation pathway may likely occur that
428
biochar decomposed peroxymonocarbonate to produce carbonate radical which degraded SAs in
429
urine (Eq. 9).
SC
RI PT
423
HCO3- + H 2O2 ↔ HCO4- + H 2 O
431
HCO-4 + e- → CO•3- + OH- (activated by biochar)
432
According to our previous studies (Zhang et al., 2015, Zhang et al., 2016a), NSMX has
433
similar reactivity toward hydroxyl radical as SMX. However, SMX reacts with carbonate radical
434
at pH 9 with second-order rate constant of 2.68×108 M−1 s−1, while NSMX shows much lower
435
reactivity (<1×106 M−1 s−1) (Table 1). Amino group substituted benzene derivative was
436
speculated to be the reason because of its high reactivity toward carbonate radical (Zhang et al.,
437
2015). Second-order rate constants of SDZ and SMT toward carbonate radical were also
438
previously determined as 2.78×108 M−1 s−1 and 4.37×108 M−1 s−1, comparable to the rate constant
439
of SMX (Zhang et al., 2016b). Although the reactivity of SDM toward carbonate radical has not
440
been measured, it is highly likely at the same level as other parent SAs due to the structural
441
similarity (e.g., amino group substituted benzene moiety). Therefore, the obvious degradation
442
efficiency difference between NSMX and other SAs suggested that carbonate radical may play
443
an important role in degrading the SAs. Because parent SAs are probably the major scavengers
444
for carbonate radical, parent SAs were degraded by similar level (about 60%, shown in Fig. 4)
445
regardless of their second-order rate constants with carbonate radical.
(8) (9)
AC C
EP
TE D
M AN U
430
19
ACCEPTED MANUSCRIPT
The decomposition of H2O2 by biochar was reported in a number of retrospective literatures
447
(Fang et al., 2014, Huang et al., 2016, Yang et al., 2017). However, some types of biochar have
448
very low capability to activate H2O2. The results in this section demonstrate that a biochar, which
449
cannot decompose H2O2, activated H2O2 yielding reactive species with the presence of
450
bicarbonate. Considering the application of biochar in urine and other high-strength wastewater,
451
this activation pathway may provide alternative biochar-based catalysis system used for
452
pharmaceutical removal.
SC
RI PT
446
453
4 Conclusion
M AN U
454
The adsorption of four parent SAs, and the major human metabolite of SMX, on biochar
456
derived from cotton straw was investigated in PBS and synthetic urine matrix. Generally, the
457
adsorption reached equilibrium within around 24 h with the presence of biochar from 0.1 g L−1 to
458
5 g L−1. The sorption isotherm followed Langmiur model. By fitting experimental data, the
459
maximum sorption sites of biochar was estimated around 10 µmol g−1 for all SAs in both PBS
460
and urine matrix. A kinetic model was developed, which successfully described the adsorption
461
behavior of all SAs presence individually or co-existing in solutions. Urine matrix increased both
462
the adsorption and desorption rates, resulting shorter time to reach sorption equilibrium. The
463
adsorption of SAs on biochar was due to multiple forces, in which van der Waals forces and
464
hydrophobicity played major roles in distinguishing the sorption behavior of different SAs,
465
expect for NSMX. In urine matrix, bicarbonate was the major competitor for SAs on the sorption.
466
In order to destruct SAs, H2O2 was added to create biochar/H2O2 catalytic system. The
467
degradation of SAs, expect for NSMX, were only observed in urine matrix, which was likely
468
attributed to the reactions involving bicarbonate. Overall, biocarbonate in urine played a critical
AC C
EP
TE D
455
20
ACCEPTED MANUSCRIPT
469
role in both adsorption process and biochar/H2O2 system. The major human metabolite of SMX,
470
NSMX, behaved differently from the parent SAs, highlighting the importance of investigating
471
both parent antibiotics and their metabolites in future studies.
RI PT
472 473
Acknowledgement
474
This work was supported by the National Scientific Foundation of China (51708401, 41703101) and Tianjin Natural Science Foundation (17JCYBJC42300).
SC
475 476
Appendix A. Supplementary data
478
Supplementary data to this article can be found online.
479 480
482
TE D
481
M AN U
477
Reference
484
Babić, S., Horvat, A.J.M., Pavlović, D.M. and Kaštelan-Macan, M., 2007. Determination of pKa
EP
483
values of active pharmaceutical ingredients. Trac Trends in Analytical Chemistry 26 (11),
486
1043-1061.
487
AC C
485
Bonvin, F., Omlin, J., Rutler, R., Schweizer, W.B., Alaimo, P.J., Strathmann, T.J., Mcneill, K.
488
and Kohn, T., 2013. Direct photolysis of human metabolites of the antibiotic
489
sulfamethoxazole: Evidence for abiotic back-transformation. Environmental Science &
490
Technology 47 (13), 6746-6755.
21
ACCEPTED MANUSCRIPT
491
Bu, Q., Wang, B., Huang, J., Deng, S. and Yu, G., 2013. Pharmaceuticals and personal care
492
products in the aquatic environment in China: A review. Journal of Hazardous Materials 262
493
(Supplement C), 189-211. Buxton, G.V., Greenstock, C.L., Helman, W.P. and Ross, A.B., 1988. Critical review of rate
RI PT
494 495
constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals. Journal
496
of Physical & Chemical Reference Data 17 (2), 513-886.
Cai, M., Sun, P., Zhang, L. and Huang, C.H., 2017. UV/Peracetic acid for degradation of
SC
497
pharmaceuticals and reactive species evaluation. Environmental Science & Technology 51
499
(24), 14217-14224.
500
M AN U
498
Calisto, V., Ferreira, C.I.A., Oliveira, J.A.B.P., Otero, M. and Esteves, V.I., 2015. Adsorptive
501
removal of pharmaceuticals from water by commercial and waste-based carbons. Journal of
502
Environmental Management 152 (Supplement C), 83-90.
Cha, J.S., Park, S.H., Jung, S.-C., Ryu, C., Jeon, J.-K., Shin, M.-C. and Park, Y.-K., 2016.
504
Production and utilization of biochar: A review. Journal of Industrial and Engineering
505
Chemistry 40 (Supplement C), 1-15.
Chen, J., Zhang, D., Zhang, H., Ghosh, S. and Pan, B., 2017. Fast and slow adsorption of
EP
506
TE D
503
carbamazepine on biochar as affected by carbon structure and mineral composition. Science
508
of The Total Environment 579 (Supplement C), 598-605.
509
AC C
507
Delgado, L.F., Charles, P., Glucina, K. and Morlay, C., 2012. The removal of endocrine
510
disrupting compounds, pharmaceutically activated compounds and cyanobacterial toxins
511
during drinking water preparation using activated carbon−A review. Science of the Total
512
Environment 435−436 (7), 509.
22
ACCEPTED MANUSCRIPT
513 514 515
Eawag Swiss Federal Institute of Aquatic Science and Technology, 2007. Mix or NoMix? A closer look at urine source separation. Fang, G., Gao, J., Liu, C., Dionysiou, D.D., Wang, Y. and Zhou, D., 2014. Key role of persistent free radicals in hydrogen peroxide activation by biochar: Implications to organic
517
contaminant degradation. Environmental Science & Technology 48 (3), 1902-1910.
RI PT
516
Fang, G., Liu, C., Gao, J., Dionysiou, D.D. and Zhou, D., 2015. Manipulation of persistent free
519
radicals in biochar to activate persulfate for contaminant degradation. Environmental
520
Science & Technology 49 (9), 5645-5653.
Ghose, A.K. and Crippen, G.M., 1987. Atomic physicochemical parameters for three-
M AN U
521
SC
518
522
dimensional-structure-directed quantitative structure-activity relationships. 2. Modeling
523
dispersive and hydrophobic interactions. Journal of Chemical Information and Computer
524
Sciences 27 (1), 21-35.
Huang, D., Wang, Y., Zhang, C., Zeng, G., Lai, C., Wan, J., Qin, L. and Zeng, Y., 2016.
526
Influence of morphological and chemical features of biochar on hydrogen peroxide
527
activation: Implications on sulfamethazine degradation. Rsc Advances 6 (77).
TE D
525
Inyang, M., Gao, B., Zimmerman, A., Zhou, Y.M. and Cao, X.D., 2015. Sorption and cosorption
529
of lead and sulfapyridine on carbon nanotube-modified biochars. Environmental Science and
530
Pollution Research 22 (3), 1868-1876.
AC C
EP
528
531
Jia, M., Wang, F., Bian, Y., Stedtfeld, R.D., Liu, G., Yu, J. and Jiang, X., 2018. Sorption of
532
sulfamethazine to biochars as affected by dissolved organic matters of different origin.
533
Bioresource Technology 248 (Part B), 36-43.
23
ACCEPTED MANUSCRIPT
534
Jung, C., Park, J., Lim, K.H., Park, S., Heo, J., Her, N., Oh, J., Yun, S. and Yoon, Y., 2013.
535
Adsorption of selected endocrine disrupting compounds and pharmaceuticals on activated
536
biochars. Journal of Hazardous Materials 263 (Part 2), 702-710. Kah, M., Sigmund, G., Xiao, F. and Hofmann, T., 2017. Sorption of ionizable and ionic organic
538
compounds to biochar, activated carbon and other carbonaceous materials. Water Research
539
124 (Supplement C), 673-692.
Kemacheevakul, P., Chuangchote, S. and Otani, S., 2014. Phosphorus recovery: Minimization of
SC
540
RI PT
537
amount of pharmaceuticals and improvement of purity in struvite recovered from
542
hydrolyzed urine. Environmental Technology 35 (23), 3011-3019.
543
M AN U
541
Landry, K.A. and Boyer, T.H., 2017. Fixed bed modeling of nonsteroidal anti-inflammatory drug
544
removal by ion-exchange in synthetic urine: Mass removal or toxicity reduction?
545
Environmental Science & Technology 51 (17), 10072-10080.
Li, S., Shi, W., Liu, W., Li, H., Zhang, W., Hu, J., Ke, Y., Sun, W. and Ni, J., 2017. A
TE D
546 547
duodecennial national synthesis of antibiotics in China's major rivers and seas (2005-2016).
548
Science of the Total Environment 615, 906-917. Li, X., Gámiz, B., Wang, Y., Pignatello, J.J. and Xing, B., 2015. Competitive sorption used to
EP
549
probe strong hydrogen bonding sites for weak organic acids on carbon nanotubes.
551
Environmental Science & Technology 49 (3), 1409-1417.
552
AC C
550
Li, X., Pignatello, J.J., Wang, Y. and Xing, B., 2013. New insight into adsorption mechanism of
553
ionizable compounds on carbon nanotubes. Environmental Science & Technology 47 (15),
554
8334-8341.
24
ACCEPTED MANUSCRIPT
555
Lin, L., Jiang, W. and Xu, P., 2017. Comparative study on pharmaceuticals adsorption in
556
reclaimed water desalination concentrate using biochar: Impact of salts and organic matter.
557
Science of The Total Environment 601 (Supplement C), 857-864. Liu, J.-L. and Wong, M.-H., 2013. Pharmaceuticals and personal care products (PPCPs): A
RI PT
558 559
review on environmental contamination in China. Environment International 59
560
(Supplement C), 208-224.
Peng, B., Chen, L., Que, C., Yang, K., Deng, F., Deng, X., Shi, G., Xu, G. and Wu, M., 2016.
SC
561
Adsorption of antibiotics on graphene and biochar in aqueous solutions induced by π-π
563
interactions. Scientific Reports 6, 31920.
564
M AN U
562
Pi, L., Yang, N., Han, W., Xiao, W., Wang, D., Xiong, Y., Zhou, M., Hou, H. and Mao, X., 2018. Heterogeneous activation of peroxymonocarbonate by Co-Mn oxides for the efficient
566
degradation of chlorophenols in the presence of a naturally occurring level of bicarbonate.
567
Chemical Engineering Journal 334, 1297-1308.
569 570
Pronk, W., Palmquist, H., Biebow, M. and Boller, M., 2006. Nanofiltration for the separation of pharmaceuticals from nutrients in source-separated urine. Water Research 40 (7), 1405-1412. Pronk, W., Zuleeg, S., Lienert, J., Escher, B., Koller, M., Berner, A., Koch, G. and Boller, M.,
EP
568
TE D
565
2007. Pilot experiments with electrodialysis and ozonation for the production of a fertiliser
572
from urine. Water Science and Technology 56 (5), 219-227.
573
AC C
571
Qambrani, N.A., Rahman, M.M., Won, S., Shim, S. and Ra, C., 2017. Biochar properties and
574
eco-friendly applications for climate change mitigation, waste management, and wastewater
575
treatment: A review. Renewable and Sustainable Energy Reviews 79 (Supplement C), 255-
576
273.
25
ACCEPTED MANUSCRIPT
577
Rajapaksha, A.U., Vithanage, M., Zhang, M., Ahmad, M., Mohan, D., Chang, S.X. and Ok, Y.S.,
578
2014. Pyrolysis condition affected sulfamethazine sorption by tea waste biochars.
579
Bioresource Technology 166 (166), 303-308. Richardson, D.E., Yao, H., Karen M. Frank, a. and Bennett, D.A., 2000. Equilibria, kinetics, and
581
mechanism in the bicarbonate activation of hydrogen peroxide: Oxidation of sulfides by
582
peroxymonocarbonate. Journal of the American Chemical Society 122 (8), 1729-1739.
583
Rosales, E., Meijide, J., Pazos, M. and Sanromán, M.A., 2017. Challenges and recent advances
SC
RI PT
580
in biochar as low-cost biosorbent: From batch assays to continuous-flow systems.
585
Bioresource Technology 246 (Supplement C), 176-192.
587 588 589 590
Shinogi, Y. and Kanri, Y., 2003. Pyrolysis of plant, animal and human waste: Physical and chemical characterization of the pyrolytic products. Bioresource Technology 90 (3), 241-247. Solanki, A. and Boyer, T.H., 2017. Pharmaceutical removal in synthetic human urine using biochar. Environmental Science-Water Research & Technology 3 (3), 553-565.
TE D
586
M AN U
584
Sun, B., Fei, L., Bao, Q., Liu, Z., Song, Z. and Zhu, L., 2016. Impact of low molecular weight organic acids (LMWOAs) on biochar micropores and sorption properties for
592
sulfamethoxazole. Environmental Pollution 214, 142-148.
593
EP
591
Teixidó, M., Pignatello, J.J., Beltrán, J.L., Granados, M. and Peccia, J., 2011. Speciation of the ionizable antibiotic sulfamethazine on black carbon (biochar). Environmental Science &
595
Technology 45 (23), 10020.
596
AC C
594
Ulrich, B.A., Im, E.A., Werner, D. and Higgins, C.P., 2015. Biochar and activated carbon for
597
enhanced trace organic contaminant retention in stormwater infiltration systems.
598
Environmental Science & Technology 49 (10), 6222-6230.
26
ACCEPTED MANUSCRIPT
599
Vithanage, M., Rajapaksha, A.U., Tang, X., Thiele-Bruhn, S., Kim, K.H., Lee, S.-E. and Ok, Y.S., 2014. Sorption and transport of sulfamethazine in agricultural soils amended with
601
invasive-plant-derived biochar. Journal of Environmental Management 141 (Supplement C),
602
95-103.
RI PT
600
Vree, T.B., Hekster, Y.A., Baars, A.M., Damsma, J.E. and Kleijn, E.V.D., 1978. Determination
604
of trimethoprim and sulfamethoxazole (co-trimoxazole) in body fluids of man by means of
605
high-performance liquid chromatography. Journal of Chromatography B Biomedical
606
Sciences & Applications 146 (1), 103-112.
Winker, M., Faika, D., Gulyas, H. and Otterpohl, R., 2008a. A comparison of human
M AN U
607
SC
603
608
pharmaceutical concentrations in raw municipal wastewater and yellowwater. Science of the
609
Total Environment 399 (1), 96-104.
610
Winker, M., Tettenborn, F., Faika, D., Gulyas, H. and Otterpohl, R., 2008b. Comparison of analytical and theoretical pharmaceutical concentrations in human urine in Germany. Water
612
Research 42 (14), 3633-3640.
613
TE D
611
Wu, M., Pan, B., Zhang, D., Xiao, D., Li, H., Wang, C. and Ning, P., 2013. The sorption of organic contaminants on biochars derived from sediments with high organic carbon content.
615
Chemosphere 90 (2), 782-788.
Xie, M., Chen, W., Xu, Z., Zheng, S. and Zhu, D., 2014. Adsorption of sulfonamides to
AC C
616
EP
614
617
demineralized pine wood biochars prepared under different thermochemical conditions.
618
Environmental Pollution 186, 187-194.
619
Yang, J., Pignatello, J.J., Pan, B. and Xing, B., 2017. Degradation of p-nitrophenol by lignin and
620
cellulose chars: H2O2-mediated reaction and direct reaction with the char. Environmental
621
Science & Technology 51 (16), 8972-8980.
27
ACCEPTED MANUSCRIPT
622
Yao, H., Lu, J., Wu, J., Lu, Z., Wilson, P.C. and Shen, Y., 2012a. Adsorption of fluoroquinolone
623
antibiotics by wastewater sludge biochar: Role of the sludge source. Water, Air, & Soil
624
Pollution 224 (1), 1370. Yao, Y., Gao, B., Chen, H., Jiang, L., Inyang, M., Zimmerman, A.R., Cao, X., Yang, L., Xue, Y.
RI PT
625 626
and Li, H., 2012b. Adsorption of sulfamethoxazole on biochar and its impact on reclaimed
627
water irrigation. Journal of Hazardous Materials 209 (Supplement C), 408-413.
Yu, W., Hu, Y., Xu, Z., Wang, S. and Xing, G., 2013. Comparisons of biochar properties from
SC
628
wood material and crop residues at different temperatures and residence times. Energy &
630
Fuels 27 (10), 5890-5899.
631
M AN U
629
Zhang, R., Sun, P., Boyer, T.H., Zhao, L. and Huang, C.-H., 2015. Degradation of
632
pharmaceuticals and metabolite in synthetic human urine by UV, UV/H2O2, and UV/PDS.
633
Environmental Science & Technology 49 (5), 3056-3066.
Zhang, R., Yang, Y., Huang, C.-H., Li, N., Liu, H., Zhao, L. and Sun, P., 2016a. UV/H2O2 and
TE D
634
UV/PDS treatment of trimethoprim and sulfamethoxazole in synthetic human urine:
636
Transformation products and toxicity. Environmental Science & Technology 50 (5), 2573-
637
2583.
638
EP
635
Zhang, R., Yang, Y., Huang, C.-H., Zhao, L. and Sun, P., 2016b. Kinetics and modeling of sulfonamide antibiotic degradation in wastewater and human urine by UV/H2O2 and
640
UV/PDS. Water Research 103 (Supplement C), 283-292.
641
AC C
639
Zhang, T. and Li, B., 2011. Occurrence, transformation, and fate of antibiotics in municipal
642
wastewater treatment plants. Critical Reviews in Environmental Science and Technology 41
643
(11), 951-998.
28
ACCEPTED MANUSCRIPT
644
Zheng, H., Wang, Z., Zhao, J., Herbert, S. and Xing, B., 2013. Sorption of antibiotic sulfamethoxazole varies with biochars produced at different temperatures. Environmental
646
Pollution 181 (Supplement C), 60-67.
AC C
EP
TE D
M AN U
SC
RI PT
645
29
ACCEPTED MANUSCRIPT Table 1. Chemical properties, Langmuir isotherm parameters and kinetics modeling
SDZ
SDM
SMT
NSMX
5.6
6.28
5.97
7.42
5.07
-0.075
0.014
0.004
0.006
-0.088
64.5
64.2
77.8
74
72.9
Sapp (M) c
13.71
3.92
0.20
0.43
35.05
Kow d
9.40
0.72
58.00
1.13
27.20
0.0023
0.0009
0.0339
0.0186
0.0020
pKa a Electonegativity b Molar Refractivity
D
PBS
2.68×108
2.78×108
N.A.
4.37×108
<1×106
bSA (µmol L−1)
0.12
0.15
1.89
0.98
1.34
QM (µmol g−1)
12.5
10.7
8.3
11.8
8
kSA (×104 M−1 h−1)
0.23
1.37
1.55
1.22
0.76
kSA-1 (h−1)
0.185
0.059
0.007
0.016
0.009
0.06
0.31
0.54
0.58
0.19
13.6
6.5
11.5
10.9
12.8
kSA (×104 M−1 h−1)
1.17
2.08
1.24
1.22
2.49
kSA-1 (h−1)
0.304
0.123
0.022
0.030
0.147
Urin bSA(µmol L−1) e
EP
QM (µmol g−1)
TE D
kSA/CO3·- e
M AN U
(cm3 mol−1) b
RI PT
SMX
SC
parameters of SAs
AC C
a pKa values of the secondary amine groups on SAs were obtained from Babić et al., 2007;
Bonvin et al., 2013
b Electronegativity and molar refractivity were estimated using ChemBio3D Ultra (v12.0)
program
c Solubility at pH 9 was calculated based on the solubility of neutral form of SAs predicted by
ECOSAR program
d Kow values were experimentally determined in this study e The second-order rate constants between SA and carbonate radical were obtained from Zhang et al., 2015 and Zhang et al., 2016
ACCEPTED MANUSCRIPT Table 2. Synthetic hydrolyzed human urine recipe. MW
Concentration
(g mol−1)
(mol L−1)
NaCl
58.44
0.06
Na2SO4
142.04
0.015
KCl
74.55
0.04
NH4OH
35.04
0.25
NaH2PO4
119.98
EP AC C
SC 0.0136
M AN U 79.06
TE D
NH4HCO3
RI PT
Species
0.25
ACCEPTED MANUSCRIPT Table 3. Characteristics of biochar. SBET (m2 g−1)
Vt (cm3 g−1)
Dp (nm)
pzc
68.4
0.074
4.3
7.4
AC C
EP
TE D
M AN U
SC
RI PT
SBET: Brunauer–Emmett–Teller surface area; Vt: total pore volume; Dp: average pore diameter; pzc: point of zero charge
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 1. Sorption kinetics (dots) and simulation data (lines) of SAs in PBS (a – e) and urine (f – j). Data points were the average of at least duplicate samples.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2. Adsorption isotherm (dots) and Langmuir isotherm model (lines) plotted as equilibrium concentration on biochar (qe) vs. aqueous-phase equilibrium
AC C
EP
TE D
concentration (Ce) for different SAs in PBS and urine.
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
L−1 biochar compared with in PBS.
M AN U
Fig. 3. Inhibition of different urine components on the adsorption of SAs on 1 g
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 4. Degradation of SAs under biochar/H2O2 conditions in PBS and urine.
AC C
EP
TE D
at 10 µM. Reaction time was 1 h.
M AN U
Biochar dose was 1 g L−1; initial H2O2 dose was 1 mM. SAs were spiked individually
ACCEPTED MANUSCRIPT
SC
RI PT
1
2
Fig. 5. Decomposition of H2O2 with biochar under different conditions. Biochar dose was 1
M AN U
3 4
g L−1; initial H2O2 dose was 1 mM; ammonia concentration was 0.5 M; chloride concentration
5
was 0.1 M; bicarbonate concentration was 0.25 M.
AC C
EP
TE D
6
1
ACCEPTED MANUSCRIPT Highlights: Adsorption of sulfonamide antibiotics in synthetic urine by biochar was investigated Adsorption was well described by Langmuir isotherm model and kinetic model Hydrophobicity and van der Waals forces are crucial in distinguishing different SAs
RI PT
Biochar can activate peroxymonocarbonate and degrade sulfonamide antibiotics
AC C
EP
TE D
M AN U
SC
Carbonate radical was proposed as major reactive species for antibiotic degradation
S0043-1354(18)30766-8
DOI:
10.1016/j.watres.2018.09.051
Reference:
WR 14101
To appear in:
Water Research
Received Date: 11 June 2018 Revised Date:
18 September 2018
Accepted Date: 20 September 2018
Please cite this article as: Sun, P., Li, Y., Meng, T., Zhang, R., Song, M., Ren, J., Removal of sulfonamide antibiotics and human metabolite by biochar and biochar/H2O2 in synthetic urine, Water Research (2018), doi: https://doi.org/10.1016/j.watres.2018.09.051. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
1
Removal of sulfonamide antibiotics and human metabolite by biochar and biochar/H2O2 in synthetic urine
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
Peizhe Sun1, Yaxiu Li1, Tan Meng1, Ruochun Zhang2*, Min Song3, Jing Ren4 1 School of Environmental Science and Engineering, Tianjin University, Tianjin 300072 China 2 Institute of Surface-Earth System Science, Tianjin University, Tianjin 300072 China 3 Ministry of Education of Key Laboratory of Energy Thermal Conversion and Control, School of Energy and Environment, Southeast University, Nanjing, Jiangsu 210096, China 4 School of Environmental Sciences, Liaoning University, Chongshan Road No.66, Shenyang 110036 China
SC
RI PT
2
Corresponding author: Ruochun Zhang; Email: [email protected]
AC C
EP
TE D
M AN U
Abstract Source-separated urine has been increasingly regarded as a promising alternative wastestream for effectively removing pharmaceuticals and human metabolites. This study investigated the removal of sulfonamide antibiotics, one category among the most frequently detected antibiotics in the environment, by biochar and biochar/H2O2 in synthetic urine matrix. The adsorption and degradation of four parent sulfonamide antibiotics, including sulfamethoxazole, sulfadiazine, sulfamethazine, sulfadimethoxine, and one human metabolite, N4-acetylsulfamethoxazole (together referred as SAs) were investigated. Biochar derived from cotton straw was applied as adsorbent for SAs and catalyst for H2O2. Results showed that the adsorption of SAs was inhibited in urine compared with that in phosphate buffer solution. Bicarbonate in urine placed major influence. Langmuir isotherm model well described the adsorption process in both buffer and urine matrices. Adsorption and desorption rates were estimated by a kinetic model, which well fitted the removal of SAs from aqueous phase at various biochar doses. The adsorption of SAs on biochar was due to multiple forces, in which van der Waals forces and hydrophobicity played major roles in distinguishing the sorption behavior of different SAs. To destruct the SAs, H2O2 was added with biochar. Except for N4-acetyl-sulfamethoxazole, all the parent SAs can be degraded in urine matrix. Carbonate radical, produced from the activation of peroxymonocarbonate by biochar, was proposed to be the major contributing reactive species in biochar/H2O2 system in urine matrix. Keywords: Source-separated urine, biochar, adsorption, sulfonamide antibiotics, H2O2 activation, carbonate radical
38
1
ACCEPTED MANUSCRIPT
39
1 Introduction Pharmaceutical residue in the environment places a major threat on the ecosystem as well as
41
human health. Wastewater treatment plants (WWTPs) has been regarded as an important source
42
for pharmaceuticals in the natural aquatic environments (Liu and Wong, 2013). Among all the
43
municipal wastewater streams entering into WWTPs, human urine accounts for less than 1% by
44
volume but contains pharmaceuticals at 2−3 orders of magnitude higher concentrations (Eawag
45
Swiss Federal Institute of Aquatic Science and Technology, 2007, Winker et al., 2008a, Winker
46
et al., 2008b). Separation of urine from other wastewater streams, namely source-separated urine,
47
has gained increasing attentions, because it is expected to be more effective for pharmaceutical
48
removal in source-separated urine than other conventional domestic wastewater. Moreover,
49
because urine contributes more than 75% total nitrogen and 50% total phosphorus in wastewater,
50
recovering nutrients from urine, as a new resource recovery strategy (Winker et al., 2008a), also
51
requires elimination of pharmaceuticals from urine. Multiple processes, including nano-filtration
52
membranes (Pronk et al., 2006), anion-exchange resin (Landry and Boyer, 2017), electrodialysis
53
(Pronk et al., 2007), struvite precipitation (Kemacheevakul et al., 2014), and advanced oxidation
54
processes (Zhang et al., 2015) have been investigated for removing pharmaceuticals from urine.
55
However, most of the investigated methods require extensive energy, material input or upgrading
56
infrastructures, and thus hinder their application. A more cost-effective process will better
57
facilitate the application of removing pharmaceuticals from their source.
AC C
EP
TE D
M AN U
SC
RI PT
40
58
Biochar is a carbonaceous material which can be produced from pyrolysis of different
59
sources of biomass, including plant materials and wastes, sludges, manures, litters, sediment, etc.
60
(Cha et al., 2016, Qambrani et al., 2017, Rosales et al., 2017). Considering cost and energy
61
saving, carbon sequestration, as well as solid waste reuse, biochar is regarded as a promising
2
ACCEPTED MANUSCRIPT
alternative adsorbent for wastewater treatment. Biochar has been proved to be effective for
63
removing varieties of pharmaceuticals including antibiotics (Wu et al., 2013, Yao et al., 2012a),
64
antiepileptic drugs (Chen et al., 2017), anti-inflammatory drugs (Lin et al., 2017), antidepressants,
65
anxiolytic and antihistaminic drugs (Calisto et al., 2015). In addition to conventional wastewater
66
treatment system, the application of biochar has also been proposed to unconventional systems,
67
such as stormwater infiltration system (Ulrich et al., 2015), reclaimed water desalination
68
concentrate (Lin et al., 2017), and other environmental compartment such as soil (Vithanage et
69
al., 2014).
SC
RI PT
62
Although it is expected that human urine is a promising matrix for effective pharmaceutical
71
removal, few studies have been carried out applying biochar. Solanki and Boyer have
72
investigated the removal of several pharmaceuticals (excluding antibiotics) in synthetic human
73
urine using biochar, but without including adsorption kinetics and isotherm studies (Solanki and
74
Boyer, 2017). The impact of urine components on pharmaceutical removal has not been
75
investigated. In addition to adsorption capacity, biochar has been reported to activate H2O2 or
76
persulfate to produce highly reactive species, which can degrade organic pollutants (Fang et al.,
77
2014, Fang et al., 2015). However, so far, few studies have been conducted for removal of
78
pharmaceuticals by radical/nonradical reactive species generated from biochar activation. And
79
the technique has not been investigated in source-separated urine matrix.
AC C
EP
TE D
M AN U
70
80
Sulfonamide antibiotics (SAs) are among the most frequently detected pharmaceuticals in
81
water systems (Bu et al., 2013, Li et al., 2017, Zhang and Li, 2011). Adsorption of
82
sulfamethoxazole (SMX) (Calisto et al., 2015, Jung et al., 2013, Wu et al., 2013, Zheng et al.,
83
2013), sulfamethazine (SMT) (Peng et al., 2016, Vithanage et al., 2014), sulfadiazine (SDZ)
84
(Peng et al., 2016), and sulfapyridine (Inyang et al., 2015, Xie et al., 2014) using biochar or
3
ACCEPTED MANUSCRIPT
modified/activated biochar has been reported. However, urine matrix is very different from
86
conventional wastewater in terms of pH and major components, thus may pose different impact
87
on adsorption process. Moreover, in urine treatment, metabolites of the pharmaceuticals should
88
also be considered. For example, only 9.5% of sulfamethoxazole was excreted in unchanged
89
form while most was excreted as metabolites (Vree et al., 1978).
RI PT
85
The objective of this study, therefore, was to investigate the kinetics and mechanisms of the
91
removal of antibiotics and human metabolite by biochar (-catalyzed) system in urine matrix.
92
Specifically, the adsorption and degradation of four parent SAs and the major human metabolite
93
of SMX were investigated. Considering the relatively long operating period, hydrolysis of urea
94
in fresh urine (referred to the urine that just left human body) will highly likely occur. Therefore,
95
hydrolyzed urine (referred to the urine that the original composition has changed due to storage)
96
was chosen in this study. This work presents SAs adsorption isotherm, kinetics and mechanisms
97
on biochar in buffer and hydrolyzed urine matrices. Chemical destruction of SAs by biochar-
98
activated H2O2 was also investigated.
99 2 Materials and methods
101
2.1 Materials
EP
100
TE D
M AN U
SC
90
Sulfamethoxazole (SMX), sulfadiazine (SDZ), sulfamethazine (SMT), sulfadimethoxine
103
(SDM), and N4-acetyl-sulfamethoxazole (NSMX) were purchased from Sigma Aldrich Inc. (St.
104
Louis, MO). All the five compounds are referred as SAs, whereas the parent SAs represents
105
SMX, SDZ, SMT and SDM. Structures and chemical properties of antibiotics are shown in Table
106
1 and Table S1. H2O2 (30%, w/w) was purchased from Aladdine (Shanghai, China). HPLC grade
107
methanol was purchased from CNW (Düsseldorf, Germany). All other chemicals and reagents
AC C
102
4
ACCEPTED MANUSCRIPT
used were in analytical grade or higher. Reagent grade deionized (DI) water (resistivity > 18 mΩ
109
cm) was prepared from a nanopure water purification system (Veolia, Paris, France). Biochar
110
was produced from pyrolysis of shredded cotton stalks, which were air-dried under 80 oC and
111
milled to pass through a 40 mesh screen. The pyrolysis condition was 350 °C with the absence of
112
oxygen.
RI PT
108
113 2.2 Adsorption experiment
SC
114
The batch adsorption experiments were carried out in 10 mL glass vials with screw caps.
116
Reaction solutions were prepared with 10 µM SA in 10 mL of synthetic urine or phosphate
117
buffer solution (PBS) at pH 9. The synthetic urine composition has been described in previous
118
literature (Zhang et al., 2015) and shown in Table 2. Due to hydrolysis of urea, high
119
concentrations of ammonia nitrogen (0.5 M) and bicarbonate (0.25 M) were present in
120
hydrolyzed urine, which also resulted in elevated pH (pH 9). PBS (with 5 mM phosphate) at pH
121
9 was employed as a baseline matrix. The vials were kept in dark and shaken at 220 rpm, 25 °C.
122
For kinetic study, biochar was applied at 0.1, 0.5, 1, 2, 5 g L−1, respectively, and aliquots were
123
taken at 1, 5, 10, 24, 120 h. For isotherm study, two more biochar doses, 0.7 g L−1 and 1.5 g L−1,
124
were added to provide more accurate results. Samples were shaken for 120 h to ensure
125
equilibrium. Before analysis, the reaction solutions were filtered with 0.45 µm polyethersulfone
126
syringe filters. Individual SAs were incubated without biochar under each corresponding
127
condition as the control samples. NaCl was applied to study the impact of ionic strength on
128
adsorption. To investigate the impact of major urine components on adsorption, bicarbonate and
129
ammonia were added separately with 1 g L−1 biochar and 10 µM SAs. Concentrations of the
130
components were the same as those in synthetic urine solution. Solution pH was adjusted by
AC C
EP
TE D
M AN U
115
5
ACCEPTED MANUSCRIPT
131
concentrated NaOH and H3PO4 solutions to pH 9. Negligible change of pH was observed over
132
the entire contact period, for both adsorption and degradation experiments. Each experiment was
133
conducted by at least duplicates.
data.
136
=
137
=
⁄
SC
135
RI PT
Freundlich (Eq. (1)) and Langmuir (Eq. (2)) isotherm model were used to fit the adsorption
134
(1) (2)
where qe is equilibrium adsorbent-phase concentration of SA, in µmol g−1; Kf is Freundlich
139
adsorption capacity parameter, in µmol g−1 (µM−1)1/n, 1/n is Freundlich adsorption intensity
140
parameter; Ce is equilibrium aqueous-phase concentration of SA, in µM; QM is Langmuir
141
maximum adsorption capacity, in µmol g−1; and bSA is the Langmuir adsorption constant of SA,
142
in µM−1. To better compare the parameters for different SAs, the amount of SA was quantified
143
using mole-based unit (µmol) instead of mass-based unit (mg).
TE D
M AN U
138
On the basis of Langmuir isotherm assumption that the adsorbates and adsorbent surface
145
sites interact as a reversible chemical equilibrium, the overall adsorption process of SAs onto
146
biochar in PBS and urine matrices was expressed using the following equation.
EP
144
AC C
k SA / kSA−1 [ SA]aq + [ BC ]vs ← →[ SA − BC ]
147 148
(3)
149
where, [SA]aq is the aqueous phase concentration of SAs, in M; [BC]vs is the vacant sites
150
concentration of biochar in the solution, in M; [SA-BC] is the concentration of SAs adsorbed
151
onto biochar, in M; kSA, in M−1 h−1, and kSA-1, in h−1, are the adsorption and desorption rates,
152
respectively. Therefore, at t = 0, [BC]vs = (biochar dose) × QM and [SA]aq = 10 µM.
153 6
ACCEPTED MANUSCRIPT
154
2.3 Degradation experiment To chemically destruct SAs, 1 mM H2O2 was added to both PBS and urine solution with 1 g
156
L−1 biochar. The samples were shaken for 1 h and then prepared for analysis. Samples after
157
reactions were treated in two ways. Extracting agent was added to recover both the residual SAs
158
that adsorbed on biochar and remained in aqueous phase. Solution containing methanol:
159
NaHCO3 9:1 (reason for adding NaHCO3 is elucidated in 3.4) was applied as extracting agent
160
and was added to samples with equal volume. After 30 min extraction, samples were filtered and
161
analyzed by HPLC. Recovery were 95%, 80%, 89%, 82%, and 90% for 10 µM SMX, SDZ,
162
SDM, SMT, and NSMX respectively.
164
2.4 Quantification of SAs
M AN U
163
SC
RI PT
155
A Waters e2695 HPLC system (Waters, Milford, MA, USA) equipped with a UV-vis
166
detector and a Zorbax-SB C18 column (4.6×150 mm, 5 µm) was used to monitor the loss of the
167
SAs. Gradient elution was used with 0.1% phosphoric acid and methanol as mobile phase. The
168
portion of methanol stayed at 10% for the first minute, then ramped to 50% at 6 min, and
169
decreased to 10% from 10 min to 15 min. The peak area of SAs was quantified at 254 nm.
EP
171
2.5 Characterization of biochar
AC C
170
TE D
165
172
Information of surface area and pore size of biochar was obtained from a surface area and
173
pore distribution analyzer (V-Sorb 2800TP, Gold APP Instruments Corporation, China). Fourier
174
transform infrared (FTIR) spectra were performed to qualitatively analyze the functional groups
175
of adsorbent (Vector 22 + TGA Bruker Optics, Germany) with the wavenumber range of 400–
7
ACCEPTED MANUSCRIPT
176
4000 cm−1. The point of zero charge (pzc) was measured by a pH meter (S220 SevenCompact,
177
Mettler Toledo, Shanghai).
178 2.6 Estimation of SA properties
RI PT
179
The pKa values of the secondary amine groups on SAs were obtained from previous
181
literatures (Babić et al., 2007, Bonvin et al., 2013). Solubility of the neutral form of SAs was
182
estimated by ECOSAR (v1.11) program. The apparent solubility of SAs was calculated and
183
detailed in section 3.3. The octanol-water distribution coefficient (Kow) was determined
184
experimentally by mixing equal amount of octanol with water solution containing 20 µM
185
individual SAs. The water phase was buffered with formic acid at pH 3.5. Molar refractivity and
186
electronegativity were estimated by ChemBio3D Ultra (v12.0).
M AN U
SC
180
187 3 Result and discussion
189
3.1 Biochar characterization
TE D
188
Biochar is proposed as a promising low-cost adsorbent because of production at relatively
191
low temperature compared with activated carbon. In order to demonstrate the performance of
192
low-cost biochar on the removal of pharmaceuticals in urine, biochar derived from cotton straw,
193
produced at 350 oC, was used in this study. It is also worthwhile noting that biochar produced at
194
low temperature retains more nutrient, such as total nitrogen, than those produced at higher
195
temperatures. thus potentially benefits future application as soil amendment (Shinogi and Kanri,
196
2003, Yu et al., 2013). The properties of the biochar derived from cotton straw are shown in
197
Table 3.
AC C
EP
190
8
ACCEPTED MANUSCRIPT
Generally, adsorption capacity is considered positively related to the specific surface area
199
and pore volumes. Therefore, comparing with the biochar materials through pyrolysis at
200
comparable temperatures, the properties of the biochar derived from cotton straw (shown in
201
Table 3) was better than peanut shell (Chen et al., 2017), pepper wood, sugarcane bagasse, and
202
hickory wood (Yao et al., 2012b), sediment (Wu et al., 2013), tea waste (Rajapaksha et al.,
203
2014), rice straw, wheat straw (Sun et al., 2016) and coconut shell (Solanki and Boyer, 2017),
204
but slightly worse than biochar produced from pine wood related materials (Jung et al., 2013, Xie
205
et al., 2014). The FTIR result (Fig. S1) indicated that hydroxyl, carbonyl and aromatic moieties
206
dominated on the surface of biochar, indicating interactions with pharmaceuticals through H-
207
bond and π-π interaction may likely occur. The point of zero charge was measure to be around
208
7.4 (Fig. S2), which suggested that the surface of biochar was negatively charged in PBS (pH 9)
209
and urine matrix.
M AN U
SC
RI PT
198
TE D
210 211
3.2 Adsorption of SAs in PBS and urine
212
3.2.1 Kinetics
The concentrations of SAs in aqueous phase versus time are shown in Fig. 1. Control
214
groups showed that SA concentrations remained unchanged without biochar (data not shown),
215
confirming that the decreasing concentration of SAs in aqueous phase was due to interaction
216
with biochar. SAs in samples with biochar for 120 h were extracted by methanolic solution, and
217
showed that degradation did not occur. SA adsorption occurred rapidly within the first 24 h with
218
0.5 – 5 g L−1 biochar. From 24 h to 120 h, the adsorption rates significantly decreased and
219
equilibrium was reached. For solutions with 0.1 g L−1 biochar, equilibrium was attained within 1
220
h, while the removal was only around 10%. With the increase of biochar doses, the removal of
AC C
EP
213
9
ACCEPTED MANUSCRIPT
SAs from aqueous phase significantly increased (Fig. 1). Indeed, the overall removal is higher
222
with higher adsorbent dose due to increasing adsorption sites. For the adsorption of
223
pharmaceuticals onto biochar, time to reach adsorption equilibrium may vary from less than 30
224
min (e.g., (Calisto et al., 2015) ) to several days (e.g., (Wu et al., 2013)) due to different
225
properties of adsorbent and adsorbate. In our study, 120 h was deemed sufficient to reach
226
apparent equilibrium. Therefore, 120 h was chosen for following isotherm study.
RI PT
221
Independent with biochar doses, overall removal of SAs in PBS ranked as SDM ≈ SMT ≈
228
NSMX > SMX ≈ SDZ. For example, the removal ratios at 1 g L−1 biochar were 71%, 74%, 68%,
229
49%, and 39%, for SDM, SMT, NSMX, SMX, and SDZ, respectively. In urine, similar ranking
230
was observed, whereas the removal was inhibited to different extent compared with in PBS (Fig.
231
S3). Inhibition was negligible when biochar doses were low because only small amount of SA
232
can be adsorbed even when there was no competitor. When biochar doses were relatively high,
233
inhibition was also limited because adsorption sites were sufficient for both SAs and urine
234
components that will adsorbed onto biochar. Strongest inhibition (urine vs. PBS) was observed
235
with 1 g L−1 or 2 g L−1 biochar for each SA.
236 237
3.2.2 Isotherm
EP
TE D
M AN U
SC
227
The adsorption isotherms for five SAs in both PBS and urine are shown in Fig. 2. Generally,
239
the sorption data were well described by both Freundlich and Langmuir models (shown in Fig. 2,
240
Fig. S4). For most conditions, Langmuir model presented relatively higher regression
241
coefficients (R2) (Table S2), indicating the overall adsorption process followed reversible
242
monolayer sorption mechanism. Therefore, Langmuir isotherm model was selected for
243
subsequent analysis and the fitting parameters are shown in Table 1. QM values are almost the
AC C
238
10
ACCEPTED MANUSCRIPT
same for all SAs, indicating sorption sites for different SAs were likely very similar. The subtle
245
differences of QM values for PBS and urine matrix indicated urine components did not
246
irreversibly adsorb onto biochar (except for SDZ). The abnormal QM value for SDZ in urine may
247
indicate a different mechanism from other SAs and further study is needed. QM value of this
248
study was comparable with the values reported in previous literatures for biochar produced under
249
similar condition from different materials, e.g., 2.79 mg g−1 (10.04 µmol g−1), 1.39 mg g−1 (5.0
250
µmol g−1), and 5.75 mg g−1 (20.68 µmol g−1) for SMT with tea waste biochar (Rajapaksha et al.,
251
2014), wheat biochar, and maize straw biochar (Jia et al., 2017); 4.21 mg g−1 (16.64 µmol g−1)
252
and 6.75mg/g (26.68 µmol g−1) for SMX with rice biochar and wheat biochar (Sun et al., 2016).
M AN U
SC
RI PT
244
In PBS, bSA values ranked as SDM > NSMX > SMT > SDZ > SMX; in urine, the ranking
254
was SDM > SMT > SDZ > NSMX > SMX. In addition, urine matrix significantly decreased bSA
255
for all the SAs from 1.7 times (SMT) to 7.1 times (NSMX) (Table 1). The decrease indicates that
256
urine components may compete with SAs adsorbing onto biochar. Same as QM values, bSA for
257
SDZ in urine was abnormal and even higher than in PBS. The difference among SAs in
258
adsorption kinetic and equilibrium was likely resulted from different chemical properties of SAs,
259
which was further discussed in section 3.3.
EP
261
3.2.3 Modeling of adsorption process
AC C
260
TE D
253
262
Applying Matlab Simbiology application in Matlab program, kSA and kSA-1 (in eqn. 3) were
263
obtained by fitting the data at 1 g L−1 biochar in PBS or in urine. All sorption rates are shown in
264
Table 1. The adsorption rates (kSA) of SAs in PBS matrix ranks as SDM > SDZ > SMT > NSMX >
265
SMX, whereas SMX has the highest desorption rate (kSA-1) followed by SDZ, SMT, NSMX and
266
SDM. In urine matrix, SDM and NSMX adsorbed faster than other SAs, whereas the desorption
11
ACCEPTED MANUSCRIPT
rates ranked as SMX > NSMX > SDZ > SMT > SDM. Generally, both the adsorption and
268
desorption rates of most SAs increased in urine matrix, indicating less time was required to reach
269
adsorption equilibrium. Applying the estimated sorption rates, the decrease of aqueous
270
concentrations of SAs with the presence of various doses of biochar were then simulated and
271
plotted in Fig. 1. Overall, the model has good agreement with the experimental data at various
272
biochar doses.
RI PT
267
To further evaluate this model, adsorption experiment in PBS was conducted by mixing five
274
SAs at 10 µM with 1 g L−1 biochar. The aqueous concentrations of SAs were measured after 3-
275
day adsorption. As shown in Fig. S5, the model successfully predicted the removal of all five
276
SAs from aqueous phase, which also confirmed the hypothesis that all five SAs adsorbed on
277
similar sites on biochar, thus competing with each other for the same adsorption sites when co-
278
existing in solution.
M AN U
SC
273
The adsorption rates of SAs represented the energy barrier which was required to overcome
280
for SAs to adsorb onto biochar surface. Higher adsorption rates suggested lower energy barrier
281
and higher attraction force. As reported in previous studies, the adsorption of SAs onto biochar
282
can be resulted from π+-π interactions, Coulombic forces, hydrogen bonds, van der Waals forces
283
and hydrophobic effects (Teixidó et al., 2011). In PBS matrix, the difference between individual
284
SAs may likely resulted from different molecular properties. Urine components altered the
285
surface properties of biochar, leading to the change of sorption rates. Therefore, in order to
286
elucidate the mechanisms of overall sorption process, the effects of SA properties and individual
287
urine components were evaluated and discussed in the following sections.
AC C
EP
TE D
279
288 289
3.3 Effect of pharmaceutical properties
12
ACCEPTED MANUSCRIPT
The SAs investigated in this work possessed benzene ring and negative charges on the
291
secondary amine nitrogen at pH 9, which are likely important interaction moieties with biochar
292
surface. Indeed, SAs may interact with biochar surface through Coulombic forces of amine
293
groups and π+-π interactions of benzene ring (Teixidó et al., 2011). In addition, hydrophobicity
294
of SAs may also play important role due to favorable partitioning of hydrophobic compounds to
295
the aromatic carbon-riched biochar surface from aqueous phase. Therefore, key chemical
296
properties, including octanol-water partitioning coefficient (Kow), solubility, acidity constant
297
(pKa), electronegativity and molar refractivity, of all five SAs were estimated or recited from
298
previous literatures (shown in Table 1) in order to correlate chemical properties to the different
299
sorption behavior of SAs.
M AN U
SC
RI PT
290
The ranking of pKa values of the secondary amine on SAs can be roughly considered as the
301
electron donating tendency of the side structures. Indeed, generally, SAs with lower pKa have
302
more negative charge on the secondary amine nitrogen (Table 1), suggesting lower tendency to
303
interact with the negatively charged biochar surface. However, neither pKa nor electronegativity
304
correlated with any kinetic parameters (i.e. bSA, QM, kSA, kSA-1). Therefore, interactions due to
305
electron interactions were not major forces distinguishing the sorption behavior of SAs.
EP
TE D
300
The van der Waals forces between adsorbate and adsorbent can roughly correlated with
307
molar refractivity (Ghose and Crippen, 1987). The molar refractivity of five SAs was estimated
308
by the ChemBio3D Ultra (v12.0) and shown in Table 1. SDM has the highest molar refractivity,
309
followed by SMT, NSMX, SMX, and SDZ. When plotting molar refractivity with bSA values in
310
PBS (Fig. S6), a good linear correlation was obtained, indicating that van der Waals forces play
311
an important role in distinguishing the adsorption equilibrium of different SAs onto biochar.
AC C
306
13
ACCEPTED MANUSCRIPT
312
However, van der Walls forces cannot explain the kinetic differences of SAs. Indeed, poor
313
correlations were obtained when plotting molar refractivity with kSA or kSA-1. The adsorption of hydrophobic organic compounds onto carboneous materials, such as
315
activated carbon, has been linked with their partitioning behavior between water and water-
316
immiscible organic solvent (Delgado et al., 2012). Kow has been widely applied to explain
317
sorption process. As for ionizable organic compounds, the pH-dependent octanol–water
318
distribution coefficient (D) was applied in order to take account speciation of compounds (Jung
319
et al., 2013). For SAs, D can be expressed as:
321
[ SA0 ]oct + [ SA− ]oct [SA0 ]water + [ SA− ]water
(4)
M AN U
D=
320
SC
RI PT
314
where [SA0]oct is the concentration of the neutral species present in the octanol phase, [SA-
323
]oct is the concentration of the ionized species present in the octanol phase, [SA0]water is the
324
concentration of the neutral species in the water phase, and [SA-]water is the concentration of the
325
ionized species in the water phase.
TE D
322
The relation between D and Kow can be expressed as:
327
1 1 1 ' D = K ow + K ow ≈ K ow 1 − pH − pKa pH − pKa pH − pKa 1 + 10 1 + 10 1 + 10
328
where (Kow) is the octanol–water partitioning coefficient of the neutral species, K′ow is the
329
octanol–water partitioning coefficient of the fully ionized species. For hydrophobic species, the
330
ionized species is much more soluble in water, thus K′ow is much lower than Kow. Therefore, D is
331
approximately a function of Kow and pH. Applying the Kow values measured in this study, the D
332
values of SAs were calculated and shown in Table 1. The same assumptions were also applied to
(5)
AC C
EP
326
14
ACCEPTED MANUSCRIPT
333
the estimation of solubility. Therefore, the apparent solubility at pH 9 was governed by the
334
solubility limit of neutral species (Eq. (6)).
336
RI PT
[H + ] + Ka S app ≈ S 0 + [H ]
335
(6)
where Sapp is the apparent solubility, in M; S0 is the solubility limit of neutral form, in M.
338
Attempts were made to correlated hydrophobicity related properties (i.e. Kow, D, S0, Sapp) with
339
sorption constants (i.e. bSA, QM, kSA, kSA-1). By plotting each property with sorption constants (Fig.
340
S7-S11), we found that (1) D values had linear relation with bSA in PBS with the exception of
341
NSMX; (2) the desorption rates of SAs in both PBS and urine increased with the increase of
342
apparent solubility, with the exception of NSMX. Therefore, hydrophobicity was crucial in
343
distinguishing different SAs.
M AN U
SC
337
The distribution coefficient, D, represents the difference between free energy in aqueous
345
phase and organic phase. Higher D values suggests lower energy state in aqueous solution. bSA
346
represents the difference between free energy in aqueous phase and on biochar surface. As
347
shown in Fig. S7, good linear relation was obtained between D and bSA for SMX, SDZ, SDM,
348
and SMT, expect for NSMX. This results suggested that free energy values of SAs on biochar
349
surface likely have the same trend as those in organic phase. On the other hand, water solubility
350
also represents the difference between free energy in aqueous phase and solid state. However,
351
poor correlation was obtained between water solubility and bSA. Indeed, the free energy of solid
352
state represents bond interactions other than solvation effects. Therefore, water solubility was
353
more suitable to describe sorption rates which were governed by bond interactions. As shown in
354
Fig. S8-S11, there is good correlation between water solubility and kSA and kSA-1 for all SAs,
AC C
EP
TE D
344
15
ACCEPTED MANUSCRIPT
355
expect NSMX. The unusual behavior of NSMX suggested the importance of aniline moiety on
356
the sorption process.
357
3.4 Effect of urine components
RI PT
358
The impact of major individual urine components was tested by spiking corresponding
360
amount of components into PBS with the presence of 1 g L−1 biochar. The aqueous
361
concentrations of SAs were then measured after 120-h adsorption process. Inhibition value was
362
calculated using the following equation.
364
(%)
(7)
[ SA]aq ,i /[ SA]T − [ SA]aq , PBS /[ SA]T 1 − [ SA]aq , PBS /[ SA]T
M AN U
Inhibition
363
SC
359
=
where [SA]aq,i is the aqueous phase concentration of SA with urine component i, in µM;
366
[SA]aq,PBS is the aqueous phase concentration of SA in PBS, in µM, [SA]T is the total
367
concentration of SA spiked in the solution, in µM.
TE D
365
Due to the presence of high concentrations of ionic species, the ionic strength effect on the
369
adsorption of SAs was first evaluated by adding additional amount of NaCl into PBS (5 mM) to
370
achieve the same ionic strength as urine matrix (i.e. 0.48 M). As shown in Fig. 3, the increase of
371
ionic strength by NaCl posed little impact on the sorption equilibrium, suggesting the different
372
sorption behavior in PBS and urine was not due to ionic strength difference. Total ammonia
373
nitrogen (TAN) and carbonate species were dominant species in urine matrix. The impact of
374
them was investigated in solutions containing 0.5 M TAN and 0.25 M bicarbonate, separately.
375
The solution pH was adjusted to 9. As shown in Fig. 3, TAN showed minimal impact on the
376
adsorption equilibrium for SAs. For SDM and NSMX, the addition of TAN even slightly
AC C
EP
368
16
ACCEPTED MANUSCRIPT
377
increased the sorption amount on biochar. However, bicarbonate greatly inhibited the adsorption
378
of all five SAs. Ionic species, such as chloride, ammonium and bicarbonate, likely affect the sorption of
380
ionizable organic compounds on biochar (Kah et al., 2017). On the basis of pzc value (7.4) of the
381
biochar applied in this study, most of the biochar surface possesses negative charges at pH 9.
382
Meanwhile, all SAs were predominantly (>98%) in their anionic form because the solution pH
383
was much higher than their pKa values (Table 1). A negative charge-assisted H-bond, (-)CAHB
384
(X…H…Y)-, was likely one of the major interactions between SAs and biochar at pH 9, as
385
postulated by Teixido et al (Teixidó et al., 2011). Indeed, the formation of SO2N…H…O likely
386
occurred through the interaction between negatively charged -SO2NH- group on SAs and surface
387
carboxyl or hydroxyl group on biochar due to close pKa (pzc) values (Li et al., 2013). It was
388
suggested that carboxylate or phenolate anions tend to form (-)CAHBs (Li et al., 2015).
389
Analogously, bicarbonate likely interacted with biochar surface through (-)CAHBs, therefore
390
competed with -SO2NH- group for biochar sorption sites. Due to the incapability to form (-
391
)CAHBs, chloride and ammonia/ammonium ions posed much less impact on the sorption of SAs
392
on biochar, compared with bicarbonate. The sorption of SAs was inhibited to different extent by
393
bicarbonate approximately following the order of molar refractivity (Fig. 3 and Table 1), which
394
suggested that both van der Waals forces and H-bonds were major interactions between SAs and
395
biochar. Based on the proposed sorption mechanisms, an extraction solution, containing
396
methanol and bicarbonate, was used to accelerate desorption of SAs from biochar and promote
397
extraction recovery.
AC C
EP
TE D
M AN U
SC
RI PT
379
398 399
3.5 Degradation of SAs with biochar/H2O2
17
ACCEPTED MANUSCRIPT
Before biochar can be applied as soil amendment, the contaminants adsorbed on it requires
401
chemical destruction to avoid secondary pollution. Based on previously reported research,
402
biochar may activate H2O2 and generate hydroxyl radical (Fang et al., 2014). Hydroxyl radical is
403
capable of degrading various pharmaceuticals due to its high oxidizing power and low selectivity
404
(Buxton et al., 1988). In this study, H2O2 was added to both PBS and urine matrix to investigate
405
the degradation of SAs with the presence of biochar. Removal was quantified by total extraction
406
by methanolic solution. Data were shown in Fig. 4. After 1 h, the degradation of SAs was not
407
observed in PBS, which implied that the biochar applied in this study cannot activate H2O2 to
408
generate reactive species (e.g., hydroxyl radical) that can degrade SAs at pH 9, or the reactive
409
species were scavenged immediately after generated. However, in urine matrix, SMX, SDZ,
410
SDM, and SMT were degraded by approximately 60% within 1 h, whereas the degradation of
411
NSMX was not observed. Control experiments have shown that SAs were stable with the
412
presence of H2O2 in urine matrix without biochar (data not shown). Thus, the involvement of
413
biochar was crucial for the degradation of parent SAs.
TE D
M AN U
SC
RI PT
400
The degradation of SAs in urine by biochar/H2O2 can be explained by two possible
415
mechanisms: (a) H2O2 was activated by biochar producing intermediate(s) which cannot degrade
416
SAs, but reacted with urine component(s) yielding secondary reactive species; and (b) the
417
interaction between urine components and H2O2 generated intermediate(s) which cannot degrade
418
SAs, but be activated by biochar yielding secondary reactive species. To test these two
419
hypotheses, the decomposition of H2O2 by biochar was monitored with and without urine
420
components. As shown in Fig. 5, H2O2 was stable with biochar in PBS, suggesting the biochar
421
cannot activate H2O2 (i.e. hypothesis (a) was incorrect). Among all conditions tested, H2O2 was
422
only decomposed by biochar with the presence of bicarbonate. The reaction between H2O2 and
AC C
EP
414
18
ACCEPTED MANUSCRIPT
bicarbonate produces peroxymonocarbonate (Eq. 8) (Richardson et al., 2000). Preliminary test
424
using DPD (N,N-diethyl-p-phenylenediamine) colorimetric method (Cai et al., 2017) showed
425
significant increase of absorbance in H2O2 solution with bicarbonate, indicating the formation of
426
peroxymonocarbonate. Peroxymonocarbonate has been reported to produce carbonate radical by
427
activation by transition metals (Pi et al., 2018). Similar activation pathway may likely occur that
428
biochar decomposed peroxymonocarbonate to produce carbonate radical which degraded SAs in
429
urine (Eq. 9).
SC
RI PT
423
HCO3- + H 2O2 ↔ HCO4- + H 2 O
431
HCO-4 + e- → CO•3- + OH- (activated by biochar)
432
According to our previous studies (Zhang et al., 2015, Zhang et al., 2016a), NSMX has
433
similar reactivity toward hydroxyl radical as SMX. However, SMX reacts with carbonate radical
434
at pH 9 with second-order rate constant of 2.68×108 M−1 s−1, while NSMX shows much lower
435
reactivity (<1×106 M−1 s−1) (Table 1). Amino group substituted benzene derivative was
436
speculated to be the reason because of its high reactivity toward carbonate radical (Zhang et al.,
437
2015). Second-order rate constants of SDZ and SMT toward carbonate radical were also
438
previously determined as 2.78×108 M−1 s−1 and 4.37×108 M−1 s−1, comparable to the rate constant
439
of SMX (Zhang et al., 2016b). Although the reactivity of SDM toward carbonate radical has not
440
been measured, it is highly likely at the same level as other parent SAs due to the structural
441
similarity (e.g., amino group substituted benzene moiety). Therefore, the obvious degradation
442
efficiency difference between NSMX and other SAs suggested that carbonate radical may play
443
an important role in degrading the SAs. Because parent SAs are probably the major scavengers
444
for carbonate radical, parent SAs were degraded by similar level (about 60%, shown in Fig. 4)
445
regardless of their second-order rate constants with carbonate radical.
(8) (9)
AC C
EP
TE D
M AN U
430
19
ACCEPTED MANUSCRIPT
The decomposition of H2O2 by biochar was reported in a number of retrospective literatures
447
(Fang et al., 2014, Huang et al., 2016, Yang et al., 2017). However, some types of biochar have
448
very low capability to activate H2O2. The results in this section demonstrate that a biochar, which
449
cannot decompose H2O2, activated H2O2 yielding reactive species with the presence of
450
bicarbonate. Considering the application of biochar in urine and other high-strength wastewater,
451
this activation pathway may provide alternative biochar-based catalysis system used for
452
pharmaceutical removal.
SC
RI PT
446
453
4 Conclusion
M AN U
454
The adsorption of four parent SAs, and the major human metabolite of SMX, on biochar
456
derived from cotton straw was investigated in PBS and synthetic urine matrix. Generally, the
457
adsorption reached equilibrium within around 24 h with the presence of biochar from 0.1 g L−1 to
458
5 g L−1. The sorption isotherm followed Langmiur model. By fitting experimental data, the
459
maximum sorption sites of biochar was estimated around 10 µmol g−1 for all SAs in both PBS
460
and urine matrix. A kinetic model was developed, which successfully described the adsorption
461
behavior of all SAs presence individually or co-existing in solutions. Urine matrix increased both
462
the adsorption and desorption rates, resulting shorter time to reach sorption equilibrium. The
463
adsorption of SAs on biochar was due to multiple forces, in which van der Waals forces and
464
hydrophobicity played major roles in distinguishing the sorption behavior of different SAs,
465
expect for NSMX. In urine matrix, bicarbonate was the major competitor for SAs on the sorption.
466
In order to destruct SAs, H2O2 was added to create biochar/H2O2 catalytic system. The
467
degradation of SAs, expect for NSMX, were only observed in urine matrix, which was likely
468
attributed to the reactions involving bicarbonate. Overall, biocarbonate in urine played a critical
AC C
EP
TE D
455
20
ACCEPTED MANUSCRIPT
469
role in both adsorption process and biochar/H2O2 system. The major human metabolite of SMX,
470
NSMX, behaved differently from the parent SAs, highlighting the importance of investigating
471
both parent antibiotics and their metabolites in future studies.
RI PT
472 473
Acknowledgement
474
This work was supported by the National Scientific Foundation of China (51708401, 41703101) and Tianjin Natural Science Foundation (17JCYBJC42300).
SC
475 476
Appendix A. Supplementary data
478
Supplementary data to this article can be found online.
479 480
482
TE D
481
M AN U
477
Reference
484
Babić, S., Horvat, A.J.M., Pavlović, D.M. and Kaštelan-Macan, M., 2007. Determination of pKa
EP
483
values of active pharmaceutical ingredients. Trac Trends in Analytical Chemistry 26 (11),
486
1043-1061.
487
AC C
485
Bonvin, F., Omlin, J., Rutler, R., Schweizer, W.B., Alaimo, P.J., Strathmann, T.J., Mcneill, K.
488
and Kohn, T., 2013. Direct photolysis of human metabolites of the antibiotic
489
sulfamethoxazole: Evidence for abiotic back-transformation. Environmental Science &
490
Technology 47 (13), 6746-6755.
21
ACCEPTED MANUSCRIPT
491
Bu, Q., Wang, B., Huang, J., Deng, S. and Yu, G., 2013. Pharmaceuticals and personal care
492
products in the aquatic environment in China: A review. Journal of Hazardous Materials 262
493
(Supplement C), 189-211. Buxton, G.V., Greenstock, C.L., Helman, W.P. and Ross, A.B., 1988. Critical review of rate
RI PT
494 495
constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals. Journal
496
of Physical & Chemical Reference Data 17 (2), 513-886.
Cai, M., Sun, P., Zhang, L. and Huang, C.H., 2017. UV/Peracetic acid for degradation of
SC
497
pharmaceuticals and reactive species evaluation. Environmental Science & Technology 51
499
(24), 14217-14224.
500
M AN U
498
Calisto, V., Ferreira, C.I.A., Oliveira, J.A.B.P., Otero, M. and Esteves, V.I., 2015. Adsorptive
501
removal of pharmaceuticals from water by commercial and waste-based carbons. Journal of
502
Environmental Management 152 (Supplement C), 83-90.
Cha, J.S., Park, S.H., Jung, S.-C., Ryu, C., Jeon, J.-K., Shin, M.-C. and Park, Y.-K., 2016.
504
Production and utilization of biochar: A review. Journal of Industrial and Engineering
505
Chemistry 40 (Supplement C), 1-15.
Chen, J., Zhang, D., Zhang, H., Ghosh, S. and Pan, B., 2017. Fast and slow adsorption of
EP
506
TE D
503
carbamazepine on biochar as affected by carbon structure and mineral composition. Science
508
of The Total Environment 579 (Supplement C), 598-605.
509
AC C
507
Delgado, L.F., Charles, P., Glucina, K. and Morlay, C., 2012. The removal of endocrine
510
disrupting compounds, pharmaceutically activated compounds and cyanobacterial toxins
511
during drinking water preparation using activated carbon−A review. Science of the Total
512
Environment 435−436 (7), 509.
22
ACCEPTED MANUSCRIPT
513 514 515
Eawag Swiss Federal Institute of Aquatic Science and Technology, 2007. Mix or NoMix? A closer look at urine source separation. Fang, G., Gao, J., Liu, C., Dionysiou, D.D., Wang, Y. and Zhou, D., 2014. Key role of persistent free radicals in hydrogen peroxide activation by biochar: Implications to organic
517
contaminant degradation. Environmental Science & Technology 48 (3), 1902-1910.
RI PT
516
Fang, G., Liu, C., Gao, J., Dionysiou, D.D. and Zhou, D., 2015. Manipulation of persistent free
519
radicals in biochar to activate persulfate for contaminant degradation. Environmental
520
Science & Technology 49 (9), 5645-5653.
Ghose, A.K. and Crippen, G.M., 1987. Atomic physicochemical parameters for three-
M AN U
521
SC
518
522
dimensional-structure-directed quantitative structure-activity relationships. 2. Modeling
523
dispersive and hydrophobic interactions. Journal of Chemical Information and Computer
524
Sciences 27 (1), 21-35.
Huang, D., Wang, Y., Zhang, C., Zeng, G., Lai, C., Wan, J., Qin, L. and Zeng, Y., 2016.
526
Influence of morphological and chemical features of biochar on hydrogen peroxide
527
activation: Implications on sulfamethazine degradation. Rsc Advances 6 (77).
TE D
525
Inyang, M., Gao, B., Zimmerman, A., Zhou, Y.M. and Cao, X.D., 2015. Sorption and cosorption
529
of lead and sulfapyridine on carbon nanotube-modified biochars. Environmental Science and
530
Pollution Research 22 (3), 1868-1876.
AC C
EP
528
531
Jia, M., Wang, F., Bian, Y., Stedtfeld, R.D., Liu, G., Yu, J. and Jiang, X., 2018. Sorption of
532
sulfamethazine to biochars as affected by dissolved organic matters of different origin.
533
Bioresource Technology 248 (Part B), 36-43.
23
ACCEPTED MANUSCRIPT
534
Jung, C., Park, J., Lim, K.H., Park, S., Heo, J., Her, N., Oh, J., Yun, S. and Yoon, Y., 2013.
535
Adsorption of selected endocrine disrupting compounds and pharmaceuticals on activated
536
biochars. Journal of Hazardous Materials 263 (Part 2), 702-710. Kah, M., Sigmund, G., Xiao, F. and Hofmann, T., 2017. Sorption of ionizable and ionic organic
538
compounds to biochar, activated carbon and other carbonaceous materials. Water Research
539
124 (Supplement C), 673-692.
Kemacheevakul, P., Chuangchote, S. and Otani, S., 2014. Phosphorus recovery: Minimization of
SC
540
RI PT
537
amount of pharmaceuticals and improvement of purity in struvite recovered from
542
hydrolyzed urine. Environmental Technology 35 (23), 3011-3019.
543
M AN U
541
Landry, K.A. and Boyer, T.H., 2017. Fixed bed modeling of nonsteroidal anti-inflammatory drug
544
removal by ion-exchange in synthetic urine: Mass removal or toxicity reduction?
545
Environmental Science & Technology 51 (17), 10072-10080.
Li, S., Shi, W., Liu, W., Li, H., Zhang, W., Hu, J., Ke, Y., Sun, W. and Ni, J., 2017. A
TE D
546 547
duodecennial national synthesis of antibiotics in China's major rivers and seas (2005-2016).
548
Science of the Total Environment 615, 906-917. Li, X., Gámiz, B., Wang, Y., Pignatello, J.J. and Xing, B., 2015. Competitive sorption used to
EP
549
probe strong hydrogen bonding sites for weak organic acids on carbon nanotubes.
551
Environmental Science & Technology 49 (3), 1409-1417.
552
AC C
550
Li, X., Pignatello, J.J., Wang, Y. and Xing, B., 2013. New insight into adsorption mechanism of
553
ionizable compounds on carbon nanotubes. Environmental Science & Technology 47 (15),
554
8334-8341.
24
ACCEPTED MANUSCRIPT
555
Lin, L., Jiang, W. and Xu, P., 2017. Comparative study on pharmaceuticals adsorption in
556
reclaimed water desalination concentrate using biochar: Impact of salts and organic matter.
557
Science of The Total Environment 601 (Supplement C), 857-864. Liu, J.-L. and Wong, M.-H., 2013. Pharmaceuticals and personal care products (PPCPs): A
RI PT
558 559
review on environmental contamination in China. Environment International 59
560
(Supplement C), 208-224.
Peng, B., Chen, L., Que, C., Yang, K., Deng, F., Deng, X., Shi, G., Xu, G. and Wu, M., 2016.
SC
561
Adsorption of antibiotics on graphene and biochar in aqueous solutions induced by π-π
563
interactions. Scientific Reports 6, 31920.
564
M AN U
562
Pi, L., Yang, N., Han, W., Xiao, W., Wang, D., Xiong, Y., Zhou, M., Hou, H. and Mao, X., 2018. Heterogeneous activation of peroxymonocarbonate by Co-Mn oxides for the efficient
566
degradation of chlorophenols in the presence of a naturally occurring level of bicarbonate.
567
Chemical Engineering Journal 334, 1297-1308.
569 570
Pronk, W., Palmquist, H., Biebow, M. and Boller, M., 2006. Nanofiltration for the separation of pharmaceuticals from nutrients in source-separated urine. Water Research 40 (7), 1405-1412. Pronk, W., Zuleeg, S., Lienert, J., Escher, B., Koller, M., Berner, A., Koch, G. and Boller, M.,
EP
568
TE D
565
2007. Pilot experiments with electrodialysis and ozonation for the production of a fertiliser
572
from urine. Water Science and Technology 56 (5), 219-227.
573
AC C
571
Qambrani, N.A., Rahman, M.M., Won, S., Shim, S. and Ra, C., 2017. Biochar properties and
574
eco-friendly applications for climate change mitigation, waste management, and wastewater
575
treatment: A review. Renewable and Sustainable Energy Reviews 79 (Supplement C), 255-
576
273.
25
ACCEPTED MANUSCRIPT
577
Rajapaksha, A.U., Vithanage, M., Zhang, M., Ahmad, M., Mohan, D., Chang, S.X. and Ok, Y.S.,
578
2014. Pyrolysis condition affected sulfamethazine sorption by tea waste biochars.
579
Bioresource Technology 166 (166), 303-308. Richardson, D.E., Yao, H., Karen M. Frank, a. and Bennett, D.A., 2000. Equilibria, kinetics, and
581
mechanism in the bicarbonate activation of hydrogen peroxide: Oxidation of sulfides by
582
peroxymonocarbonate. Journal of the American Chemical Society 122 (8), 1729-1739.
583
Rosales, E., Meijide, J., Pazos, M. and Sanromán, M.A., 2017. Challenges and recent advances
SC
RI PT
580
in biochar as low-cost biosorbent: From batch assays to continuous-flow systems.
585
Bioresource Technology 246 (Supplement C), 176-192.
587 588 589 590
Shinogi, Y. and Kanri, Y., 2003. Pyrolysis of plant, animal and human waste: Physical and chemical characterization of the pyrolytic products. Bioresource Technology 90 (3), 241-247. Solanki, A. and Boyer, T.H., 2017. Pharmaceutical removal in synthetic human urine using biochar. Environmental Science-Water Research & Technology 3 (3), 553-565.
TE D
586
M AN U
584
Sun, B., Fei, L., Bao, Q., Liu, Z., Song, Z. and Zhu, L., 2016. Impact of low molecular weight organic acids (LMWOAs) on biochar micropores and sorption properties for
592
sulfamethoxazole. Environmental Pollution 214, 142-148.
593
EP
591
Teixidó, M., Pignatello, J.J., Beltrán, J.L., Granados, M. and Peccia, J., 2011. Speciation of the ionizable antibiotic sulfamethazine on black carbon (biochar). Environmental Science &
595
Technology 45 (23), 10020.
596
AC C
594
Ulrich, B.A., Im, E.A., Werner, D. and Higgins, C.P., 2015. Biochar and activated carbon for
597
enhanced trace organic contaminant retention in stormwater infiltration systems.
598
Environmental Science & Technology 49 (10), 6222-6230.
26
ACCEPTED MANUSCRIPT
599
Vithanage, M., Rajapaksha, A.U., Tang, X., Thiele-Bruhn, S., Kim, K.H., Lee, S.-E. and Ok, Y.S., 2014. Sorption and transport of sulfamethazine in agricultural soils amended with
601
invasive-plant-derived biochar. Journal of Environmental Management 141 (Supplement C),
602
95-103.
RI PT
600
Vree, T.B., Hekster, Y.A., Baars, A.M., Damsma, J.E. and Kleijn, E.V.D., 1978. Determination
604
of trimethoprim and sulfamethoxazole (co-trimoxazole) in body fluids of man by means of
605
high-performance liquid chromatography. Journal of Chromatography B Biomedical
606
Sciences & Applications 146 (1), 103-112.
Winker, M., Faika, D., Gulyas, H. and Otterpohl, R., 2008a. A comparison of human
M AN U
607
SC
603
608
pharmaceutical concentrations in raw municipal wastewater and yellowwater. Science of the
609
Total Environment 399 (1), 96-104.
610
Winker, M., Tettenborn, F., Faika, D., Gulyas, H. and Otterpohl, R., 2008b. Comparison of analytical and theoretical pharmaceutical concentrations in human urine in Germany. Water
612
Research 42 (14), 3633-3640.
613
TE D
611
Wu, M., Pan, B., Zhang, D., Xiao, D., Li, H., Wang, C. and Ning, P., 2013. The sorption of organic contaminants on biochars derived from sediments with high organic carbon content.
615
Chemosphere 90 (2), 782-788.
Xie, M., Chen, W., Xu, Z., Zheng, S. and Zhu, D., 2014. Adsorption of sulfonamides to
AC C
616
EP
614
617
demineralized pine wood biochars prepared under different thermochemical conditions.
618
Environmental Pollution 186, 187-194.
619
Yang, J., Pignatello, J.J., Pan, B. and Xing, B., 2017. Degradation of p-nitrophenol by lignin and
620
cellulose chars: H2O2-mediated reaction and direct reaction with the char. Environmental
621
Science & Technology 51 (16), 8972-8980.
27
ACCEPTED MANUSCRIPT
622
Yao, H., Lu, J., Wu, J., Lu, Z., Wilson, P.C. and Shen, Y., 2012a. Adsorption of fluoroquinolone
623
antibiotics by wastewater sludge biochar: Role of the sludge source. Water, Air, & Soil
624
Pollution 224 (1), 1370. Yao, Y., Gao, B., Chen, H., Jiang, L., Inyang, M., Zimmerman, A.R., Cao, X., Yang, L., Xue, Y.
RI PT
625 626
and Li, H., 2012b. Adsorption of sulfamethoxazole on biochar and its impact on reclaimed
627
water irrigation. Journal of Hazardous Materials 209 (Supplement C), 408-413.
Yu, W., Hu, Y., Xu, Z., Wang, S. and Xing, G., 2013. Comparisons of biochar properties from
SC
628
wood material and crop residues at different temperatures and residence times. Energy &
630
Fuels 27 (10), 5890-5899.
631
M AN U
629
Zhang, R., Sun, P., Boyer, T.H., Zhao, L. and Huang, C.-H., 2015. Degradation of
632
pharmaceuticals and metabolite in synthetic human urine by UV, UV/H2O2, and UV/PDS.
633
Environmental Science & Technology 49 (5), 3056-3066.
Zhang, R., Yang, Y., Huang, C.-H., Li, N., Liu, H., Zhao, L. and Sun, P., 2016a. UV/H2O2 and
TE D
634
UV/PDS treatment of trimethoprim and sulfamethoxazole in synthetic human urine:
636
Transformation products and toxicity. Environmental Science & Technology 50 (5), 2573-
637
2583.
638
EP
635
Zhang, R., Yang, Y., Huang, C.-H., Zhao, L. and Sun, P., 2016b. Kinetics and modeling of sulfonamide antibiotic degradation in wastewater and human urine by UV/H2O2 and
640
UV/PDS. Water Research 103 (Supplement C), 283-292.
641
AC C
639
Zhang, T. and Li, B., 2011. Occurrence, transformation, and fate of antibiotics in municipal
642
wastewater treatment plants. Critical Reviews in Environmental Science and Technology 41
643
(11), 951-998.
28
ACCEPTED MANUSCRIPT
644
Zheng, H., Wang, Z., Zhao, J., Herbert, S. and Xing, B., 2013. Sorption of antibiotic sulfamethoxazole varies with biochars produced at different temperatures. Environmental
646
Pollution 181 (Supplement C), 60-67.
AC C
EP
TE D
M AN U
SC
RI PT
645
29
ACCEPTED MANUSCRIPT Table 1. Chemical properties, Langmuir isotherm parameters and kinetics modeling
SDZ
SDM
SMT
NSMX
5.6
6.28
5.97
7.42
5.07
-0.075
0.014
0.004
0.006
-0.088
64.5
64.2
77.8
74
72.9
Sapp (M) c
13.71
3.92
0.20
0.43
35.05
Kow d
9.40
0.72
58.00
1.13
27.20
0.0023
0.0009
0.0339
0.0186
0.0020
pKa a Electonegativity b Molar Refractivity
D
PBS
2.68×108
2.78×108
N.A.
4.37×108
<1×106
bSA (µmol L−1)
0.12
0.15
1.89
0.98
1.34
QM (µmol g−1)
12.5
10.7
8.3
11.8
8
kSA (×104 M−1 h−1)
0.23
1.37
1.55
1.22
0.76
kSA-1 (h−1)
0.185
0.059
0.007
0.016
0.009
0.06
0.31
0.54
0.58
0.19
13.6
6.5
11.5
10.9
12.8
kSA (×104 M−1 h−1)
1.17
2.08
1.24
1.22
2.49
kSA-1 (h−1)
0.304
0.123
0.022
0.030
0.147
Urin bSA(µmol L−1) e
EP
QM (µmol g−1)
TE D
kSA/CO3·- e
M AN U
(cm3 mol−1) b
RI PT
SMX
SC
parameters of SAs
AC C
a pKa values of the secondary amine groups on SAs were obtained from Babić et al., 2007;
Bonvin et al., 2013
b Electronegativity and molar refractivity were estimated using ChemBio3D Ultra (v12.0)
program
c Solubility at pH 9 was calculated based on the solubility of neutral form of SAs predicted by
ECOSAR program
d Kow values were experimentally determined in this study e The second-order rate constants between SA and carbonate radical were obtained from Zhang et al., 2015 and Zhang et al., 2016
ACCEPTED MANUSCRIPT Table 2. Synthetic hydrolyzed human urine recipe. MW
Concentration
(g mol−1)
(mol L−1)
NaCl
58.44
0.06
Na2SO4
142.04
0.015
KCl
74.55
0.04
NH4OH
35.04
0.25
NaH2PO4
119.98
EP AC C
SC 0.0136
M AN U 79.06
TE D
NH4HCO3
RI PT
Species
0.25
ACCEPTED MANUSCRIPT Table 3. Characteristics of biochar. SBET (m2 g−1)
Vt (cm3 g−1)
Dp (nm)
pzc
68.4
0.074
4.3
7.4
AC C
EP
TE D
M AN U
SC
RI PT
SBET: Brunauer–Emmett–Teller surface area; Vt: total pore volume; Dp: average pore diameter; pzc: point of zero charge
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 1. Sorption kinetics (dots) and simulation data (lines) of SAs in PBS (a – e) and urine (f – j). Data points were the average of at least duplicate samples.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2. Adsorption isotherm (dots) and Langmuir isotherm model (lines) plotted as equilibrium concentration on biochar (qe) vs. aqueous-phase equilibrium
AC C
EP
TE D
concentration (Ce) for different SAs in PBS and urine.
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
L−1 biochar compared with in PBS.
M AN U
Fig. 3. Inhibition of different urine components on the adsorption of SAs on 1 g
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 4. Degradation of SAs under biochar/H2O2 conditions in PBS and urine.
AC C
EP
TE D
at 10 µM. Reaction time was 1 h.
M AN U
Biochar dose was 1 g L−1; initial H2O2 dose was 1 mM. SAs were spiked individually
ACCEPTED MANUSCRIPT
SC
RI PT
1
2
Fig. 5. Decomposition of H2O2 with biochar under different conditions. Biochar dose was 1
M AN U
3 4
g L−1; initial H2O2 dose was 1 mM; ammonia concentration was 0.5 M; chloride concentration
5
was 0.1 M; bicarbonate concentration was 0.25 M.
AC C
EP
TE D
6
1
ACCEPTED MANUSCRIPT Highlights: Adsorption of sulfonamide antibiotics in synthetic urine by biochar was investigated Adsorption was well described by Langmuir isotherm model and kinetic model Hydrophobicity and van der Waals forces are crucial in distinguishing different SAs
RI PT
Biochar can activate peroxymonocarbonate and degrade sulfonamide antibiotics
AC C
EP
TE D
M AN U
SC
Carbonate radical was proposed as major reactive species for antibiotic degradation