Subscriber access provided by University of Winnipeg Library
Ecotoxicology and Human Environmental Health 2.5
Oxidative Potential by PM in the North China Plain: Generation of Hydroxyl Radical Xiaoying Li, Xiaobi Michelle Kuang, Caiqing Yan, Shexia Ma, Suzanne E Paulson, Tong Zhu, Yuanhang Zhang, and Mei Zheng Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b05253 • Publication Date (Web): 30 Nov 2018 Downloaded from http://pubs.acs.org on December 1, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 29
Environmental Science & Technology
1
Oxidative Potential by PM2.5 in the North China Plain: Generation of Hydroxyl
2
Radical
3
Xiaoying Li1#, Xiaobi M. Kuang2#, Caiqing Yan1, Shexia Ma3, Suzanne E. Paulson2,
4
Tong Zhu1, Yuanhang Zhang1, and Mei Zheng1*
5
1SKL-ESPC
6
Peking University, Beijing 100871, China
7
2Department
8
Angeles, Los Angeles, CA 90095, USA
9
3South
and BIC-ESAT, College of Environmental Sciences and Engineering,
of Atmospheric and Oceanic Sciences, University of California at Los
China Institute of Environmental Sciences, Ministry of Environmental
10
Protection, Guangzhou 510655, China
11
* Corresponding author: Mei Zheng,
[email protected] 12
# Both
13
ABSTRACT
contributed equally as the first author.
14
Adverse health effects of ambient PM2.5 (dp99%)
97
were obtained from J&K Scientific Ltd., China. Sodium chloride (guaranteed reagent
98
grade, GR) was from Xilong Chemical Co., Ltd., China. Disodium Terephthalate (TA,
99
ACS) was from Tokyo Chemical Industry Co., Ltd., Japan. Sodium citrate tribasic
100
dihydrate (99.0%), ascorbic acid sodium salt (Asc, >99%), uric acid sodium salt (UA,
101
ACS), nitric acid (Cit, >69%), 2,2,2-Trifluoroethanol, and chelex-100 (95%) were
102
purchased from Sigma Aldrich (St. Louis, MO, USA).
103
Sampling and chemical analysis. Ambient PM2.5 samples were collected at an
104
urban (Beijing) and a suburban (Wangdu) site in NCP from 9th June to 8th July, 2014
105
(Figure S1, Supporting Information) during the CAREBEIJING-NCP project. Wangdu
106
is a small town in Hebei Province, and close to Baoding (35 km) and Shijiazhuang (90
107
km), which are two major industrial cities in NCP. Consecutive 23-h PM2.5 samples
108
were collected in Beijing, typically starting from 9:00 a.m. and ending at 8:00 a.m. the
109
next day. PM2.5 samples in Wangdu were collected during the daytime (from 8:00 a.m.
110
to 5:30 p.m.) and nighttime (from 6:00 p.m. to 7:30 a.m. the next day). Teflon filters
111
were weighted before sampling and re-weighted after sampling using a microbalance
112
(Mettler Toledo) with the sensitivity of 0.00004 g to determine the mass concentration
113
of PM2.5. Chemical compositions of PM2.5 were analyzed, including OC, EC, water-
114
soluble ions (Na+, NH4+, K+, Mg2+, Ca2+, NO3-, SO42-) by ion chromatography (ICS-
115
2000 and ICS-2500, DIONEX), and trace metals (Al, Fe, Mn, Ti, Co, Cr, Ni, Cu, Pb,
116
Zn, Cd, V, As, Se, Mo, and Co) by inductively coupled plasma-mass spectrometry
117
(ICP-MS). A detailed description of sampling and chemical analyses is provided in
118
Text S1.
119
Surrogate Lung Fluid (SLF) solution. All extractions were performed in a cell-free
120
SLF solution, which is a surrogate of lung fluid, but it has to be recognized that it is
121
different from actual lung fluid. It consists of phosphate-buffered saline (PBS,
122
including 114 mM NaCl, 7.8 mM Na2HPO4 and 2.2 mM KH2PO4 to buffer the solution 5 / 29
ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 29
123
at pH 7.2-7.4), four antioxidants (200 µM Asc, 300 µM Cit, 100 µM GSH, and 100 µM
124
UA), and TA (10 mM).30,41-48 The PBS was treated by column chromatography with
125
Chelex-100 resin to remove metals, stored in the refrigerator, and generally used within
126
one month. Four antioxidants were freshly prepared and added to the PBS just prior to
127
extraction.
128
Particle extraction and quantification of ·OH. In this study, half of a Teflon filter
129
was first treated with 50 µL of 2,2,2-Trifluoroethanol to increase water-solubility of
130
particles on the filters.42,47 It is also added to filed blank filter extractions and the
131
resulting signals are subtracted off to correct the sample results. Vidrio et al.47 reported
132
that TFE could enhance ·OH production by 80±40% compared to the same sample
133
extracted by SLF in the absence of TFE. The average absolute mass of PM2.5 on
134
extracted filters is 238±143 μg (mean±standard deviation) and 410±228 μg in Beijing
135
and Wangdu, respectively. Before each filter was placed in a Teflon dish, 10 mL of
136
SLF solution was added. TA was added and reacted with ·OH to produce TAOH, which
137
is stable and strongly fluorescent. TAOH was detected at excitation/emission
138
wavelength (λex/λem) of 320/420 nm using a sensitive fluorometer (Cary Eclipse,
139
Varian), and then aliquots of 200 µL of solution were measured at 0, 20, 40, 60, 80, 100
140
and 120 minutes to determine the amount of TAOH formed. For each sample and blank
141
filter, TAOH was measured three times and the mean value of the results was taken as
142
the final concentration of TAOH. The production of ·OH by PM2.5 in SLF increased
143
linearly during two hours of extraction, with the formation rate of 0.49 ng/μg·PM2.5·h
144
and 0.52 ng/μg·PM2.5·h in Beijing and Wangdu, respectively (Figure S2). To better
145
compare with other studies which applied similar method with this study30,33,42,43,47, the
146
production of ·OH showed in this study was presented as the total ·OH generated in
147
SLF after 2-hour reaction, normalized to PM2.5 mass (ng/µg·PM2.5) and air volume
148
(ng/m3·air), rather than the formation rate of ·OH.
149
Calibration with TAOH standards at 50, 100, 200, 500 and 800 nM was performed
150
daily. 10 mM of TA in the SLF solution is sufficient to react with >98% of ·OH. The
151
concentration of ·OH in each solution was determined using Equation 1,
152
[·OH] = [TAOH] / yTAOH 6 / 29
ACS Paragon Plus Environment
Equation 1
Page 7 of 29
Environmental Science & Technology
153
where [TAOH] is the measured concentration of TAOH, yTAOH is the molar yield of
154
TAOH produced from the reaction of ·OH with TA in SLF, which is 0.35 at pH 7.2.42,49
155
The SLF solution blanks and field blanks were treated in the same way with PM2.5
156
samples. The amount of TAOH formed in the SLF solution blanks was less than 1% of
157
the amount of TAOH formed in samples, and thus could be ignored. The results of the
158
field blanks was ~10% of the amount of TAOH formed in samples, and were subtracted
159
off to correct the sample results.
160
RESULTS AND DISCUSSION
161
High ·OH production (ng/m3·air) by PM2.5 in heavily polluted days. Figure 1
162
shows the mass concentration of PM2.5 and ·OH production (ng/m3·air) by PM2.5 in
163
Beijing and Wangdu. The daily variation of PM2.5 in Beijing and Wangdu is similar,
164
indicating the regional nature of air pollution in NCP. However, the mass
165
concentrations of PM2.5 in Wangdu in most samples are higher than those in Beijing,
166
although it is a suburban site in NCP. The average mass concentrations of PM2.5 in
167
Beijing and Wangdu are 51.4±31.5 µg/m3 and 70.5±37.4 µg/m3, respectively (Table 1).
168
PM2.5 mass concentrations in the nighttime in most samples are slightly higher
169
compared to daytime samples in Wangdu. The ·OH production (ng/m3·air) shows the
170
similar day-to-day variation trend with PM2.5, and it is higher in Wangdu compared to
171
Beijing. The average ·OH production is 47.0±16.5 ng/m3·air and 61.0±21.9 ng/m3·air
172
in Beijing and Wangdu, respectively. The air in Wangdu exhibits higher oxidative
173
potential due to its larger concentration of PM2.5.
174
Considering the National Ambient Air Quality Standard of China and the PM2.5
175
concentration range during our sampling period, samples collected on days with daily
176
average PM2.5 concentration below 35 µg/m3 are defined as clean days, 35 µg/m3 to 75
177
µg/m3 as lightly polluted days, and above 75 µg/m3 as heavily polluted days. From
178
clean days to heavily polluted days, ·OH production increases 2.0 and 1.6 times in
179
Beijing and Wangdu, respectively. As Figure 2 and Table 1 show, the ·OH production
180
(ng/m3·air) increases with PM2.5 concentration in both Beijing and Wangdu, and the
181
increasing trend goes to flat when PM2.5 concentration reaches higher level. It is 7 / 29
ACS Paragon Plus Environment
Environmental Science & Technology
182
consistent with the integrated exposure-response in epidemiologic studies, which
183
describes the relationship between health endpoints (e.g., ischemic heart disease, stroke,
184
chronic obstructive pulmonary disease) and exposure to PM2.5.50,51 It indicates that as
185
PM2.5 concentration increases, e.g., for every increased 10 µg/m3 of PM2.5, less ROS
186
are produced.
187
The ·OH production per unit mass (ng/µg·PM2.5) decreases with PM2.5
188
concentration. Figure 3 presents the production of ·OH per unit mass of PM2.5
189
decreases with increasing ambient PM2.5 concentration. Regardless of the sampling
190
location, a power function is observed between ·OH production (ng/µg·PM2.5) and
191
ambient PM2.5 mass concentration, and similar trend was also observed in Seoul,
192
Korea32. There are two possible reasons for the observed pattern. One possible reason
193
is that the increased mass in PM2.5 has little influence on ·OH generation, and the other
194
is that the possible inhibiting effect of high levels of redox active metals in the
195
extraction solution on ·OH generation, as the reaction to produce ·OH is a reversible
196
reaction.
197
For the first possibility, SO42-, NO3- and NH4+ dominate the increment of PM2.5 in
198
both Beijing (80%) and Wangdu (83%) (Table S1), while they do not contribute to the
199
formation of ROS due to their non-redox activity45,46. Many studies reported that it is
200
the redox-active components in PM2.5 inducing the formation of ROS, including trace
201
metals, quinones, and HULIS28,30,34,35, whereas SO42-, NO3- and NH4+ are non-redox
202
active.
203
The second possibility is the inhibiting effect of redox active metals to ·OH formation
204
in SLF, due to their high concentrations in heavily polluted days. Charrier and
205
Anastasio30 investigated the concentration dependence of ·OH production from water-
206
soluble Fe, Cu, and quinones at atmospherically relevant concentrations and found that
207
·OH produced by water-soluble Cu in simplified laboratory solutions is linearly related
208
to the Cu concentration at low concentrations, but flattens out above ~200 nM. In this
209
study, the average concentration of Cu (digested with strong acids) in the extraction is
210
110 ± 82 nM and 204 ± 175 nM in Beijing and Wangdu, respectively. While the 8 / 29
ACS Paragon Plus Environment
Page 8 of 29
Page 9 of 29
Environmental Science & Technology
211
concentrations of water-soluble Cu in the extraction solutions were not measured, Cu
212
is one of the most soluble metals in particles, with reported fraction soluble compared
213
to total Cu in the 60~80% range52,53. Similar concentration-dependent behavior was
214
observed for several quinones, but not for ·OH produced by water-soluble Fe.30
215
However, Charrier and Anastasio30 also confirmed that the mixtures of Fe and Cu
216
mitigate the behavior and result in a synergistic increase in ·OH production. Because
217
the concentrations of Cu in the extraction solutions are not too high in this study, a
218
similar concentration-dependence may not be occurring. However, future study is
219
needed to further investigate the inhibiting effects of redox active metals (e.g. Fe, Mn,
220
Cu) and/or quinones in PM2.5.
221
Therefore, the decrease of ·OH production (ng/µg·PM2.5) with PM2.5 concentration
222
in this study is most likely due to the fact that the increased PM2.5 in NCP is mainly
223
secondary inorganic components (sulfate, nitrate, and ammonium) and these
224
components are not the key species that contribute to the generation of ROS.
225
Correlations between chemical species in PM2.5 and ·OH production. To
226
investigate which species are more responsible for ·OH generation in Beijing and
227
Wangdu, chemical species in PM2.5 were analyzed and Pearson correlation analysis was
228
conducted between ·OH production and chemical species in PM2.5.
229
Correlations between ·OH production and major chemical components in PM2.5. The
230
average mass concentration of major chemical species in PM2.5 in Beijing and Wangdu
231
is shown in Figure 4. The average concentrations of OC and EC in Wangdu are slightly
232
higher than Beijing (1.44 times for OC and 1.25 times for EC), while the relative
233
contributions of OC and EC to PM2.5 are similar (Figure 5). Similar to OC and EC, the
234
concentrations of SO42- and NH4+ in Wangdu are 1.50 and 1.30 times higher compared
235
to those in Beijing, and this may be due to more coal combustion (sulfur dioxide, SO2)11
236
and livestocks and agricultural emissions (ammonia gas, NH3)54 in the suburban site
237
Wangdu. The concentrations of NO3- are 10.7±10.2 μg/m3 and 10.6±10.0 μg/m3 in
238
Beijing and Wangdu, respectively, but the relative contributions of NO3- to PM2.5 in
239
Beijing is higher than that in Wangdu, due to higher contributions from vehicle exhaust. 9 / 29
ACS Paragon Plus Environment
Environmental Science & Technology
240
Figure 6 presents the Pearson correlation between ·OH production and PM2.5
241
chemical components. It can be seen that OC, not water-soluble organic carbon
242
(WSOC), is strongly correlated with ·OH production in Beijing and these relationships
243
are not statistically significant in Wangdu. WSOC, HULIS and quinones, have all been
244
associated with ROS formation as shown in other studies.30,31,34,55-57 Some studies have
245
demonstrated that the aging process of organic aerosol could increase the oxidative
246
potential of PM2.5.52,58,59 Liu et al.14 also reported that WSOC has a high contribution
247
to ROS activity in Beijing and it appears to be associated with heavy traffic emissions.
248
In addition, Lin et al.60 found that the oxygenated products of PAHs in Beijing are
249
quinones, and secondary formation is a significant source of PM2.5 toxicity in non-
250
heating season due to the generation of harmful ROS by quinones.
251
Similar to OC, EC is also strongly correlated with ·OH production in Beijing, but not
252
in Wangdu. This might be related to the different sources of PM2.5 in Beijing and
253
Wangdu, which result in different atmospheric composition of PM2.5, especially for
254
carbonaceous aerosol. In Beijing, vehicle exhaust is one of the most important sources
255
of OC and EC in PM2.5 in summer, accounting for 63% of the carbonaceous components
256
in PM2.5.61,62 Wangdu is a suburban site, with industry from its surrounding area and
257
biomass burning exhibiting higher importance and impact.63 Several studies have found
258
associations between EC and adverse health effects, but the mechanism is not well
259
understood.64-67 Hu et al.58 found the high correlation between EC and ·OH, and
260
proposed this correlation is actually due to the correlation between OC and ·OH, since
261
EC and OC are normally from the same sources, such as vehicle exhaust. It has also
262
been found that fresh emitted BC has low oxidative potential, while it increases as soot
263
(or EC, BC) particles oxygenated by O366 or secondary organic carbon, PAHs and
264
quinones adsorbed on soot surfaces25-27,52,59. Therefore, the good correlation between
265
EC and ·OH production in Beijing suggests that the measured EC in ambient samples
266
is more aged EC.
267
However, the correlations between ·OH production and SO42-, NO3-, and NH4+ are
268
all negative, and such negative correlations are stronger for Beijing relative to Wangdu
269
(Figure 6). This is consistent with the findings of Hu et al.68 and Ntziachristos et al.69, 10 / 29
ACS Paragon Plus Environment
Page 10 of 29
Page 11 of 29
Environmental Science & Technology
270
indicating SO42-, NO3-, and NH4+ do not have a direct pathway to induce the formation
271
of ROS. This negative correlation provides supportive evidence to explain the findings
272
that the production of ·OH per unit mass (ng/μg·PM2.5) decreases with ambient PM2.5
273
concentration, because the majority of increased PM2.5 mass is mainly composed of
274
SO42-, NO3-, and NH4+.
275
Correlations between ·OH production and trace metals in PM2.5. It should be noticed
276
that the trace metals analyzed in this study are digested with strong acids and they are
277
not water-soluble fraction only. The difference of metals in PM2.5 at the two sites
278
(Beijing and Wangdu) is significant (Figures 4, 5). In Beijing, the average mass
279
concentration of all metals is 0.54±0.17 µg/m3, accounting for 1.1% of PM2.5, while
280
their contribution is 2.4% in Wangdu (1.67±1.09 µg/m3). Several studies have shown
281
that water-soluble metals which undergo redox cycle are strongly associated with ROS
282
formation, of which, Fe, Cu, and Mn are more active.14,30,31,34,35,43,46 The relative
283
contributions of Fe, Cu, and Mn (digested with strong acids) to all metals in Beijing are
284
almost double those in Wangdu, especially Fe, which account for 42% of all metals in
285
Beijing vs. 24% in Wangdu (Figure 5). The relative contributions of Cu and Mn to all
286
metals in all samples are quite low, but it is still higher in Beijing than Wangdu. Redox
287
inactive Al accounts for nearly half of all metals in Wangdu samples (45%), about two
288
times higher than that in Beijing (25%). The relative contributions of other metals are
289
much the same at both sites. Therefore, trace metals considered to be more redox active
290
are more enriched in the urban site Beijing than the suburban site Wangdu.
291
Quite a few studies have demonstrated that different chemical species in PM2.5 have
292
different abilities to contribute to the generation of ·OH in SLF.14,28-30,34,35 For example,
293
in simplified solutions, water-soluble Cu and Fe generate the most ·OH, with additional
294
contributions from a subset of quinones.30 A recent published paper showed that water-
295
soluble transition Cr and Zn are correlated to oxidative potential of PM2.5, and As and
296
V have a high contribution to ROS activity in Seoul, Korea.32 A lab study also found
297
that both nano- TiO2 and ZnO could induce the formation of ·OH.70 Therefore, the
298
difference in trace metals in PM2.5 could also be one of the most important reasons for
299
the difference for the oxidative potential of PM2.5 in Beijing and Wangdu, besides the 11 / 29
ACS Paragon Plus Environment
Environmental Science & Technology
300
aged OC and EC.
301
As Figure 6 shows, the metals measured with strong acid digestion are strongly
302
correlated with ·OH production both in Beijing and Wangdu, although the extent of the
303
correlation varies with species. Fe, Ba, Mn, Al, Ca2+, Mo, and Se all have strong
304
correlations (p
