S
1
Supporting Information: Hydroxymethanesulfonate (HMS)
Forma
tion during Summertime Fog in an
Arctic Oil Field
Jun Liu
1
, Matthew J. Gunsch
1
, Claire E. Moffett
2
, Lu Xu
3
, Rime El Asmar
4
, Qi
Zhang
5
,
Thomas B. Watson
6
, Hannah M. Allen
3
, John D. Crounse
7
, Jason St. Clair
8,9
, Michelle
Kim
7
, Paul O. Wennberg
3,7
, Rodney J. Weber
4
, Rebecca J. Sheesley
2
,
and Kerri A.
Pratt
1,10*
1
Department of Chemistry, University of Michigan, Ann Arbor, MI, USA
2
Department of Environmental Science, Baylor University, Waco, TX, USA
3
Division of Geological
and Planetary Sciences, California Institute of Technology,
Pasadena, CA, USA
4
Department of Environmental Toxicology, University of California, Davis, CA, USA
5
School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta,
GA, USA
6
Department of Environmental and Climate Sciences, Brookhaven National Laboratory,
Upton, NY, USA
7
Division
of Engineering
and
Applied Science, California Institu
te of Technology,
Pasadena, CA, USA
8
Atmospheric Chemistry and Dynamics Lab, NASA Goddard Space Flight Center,
Greenbelt, MD, USA
9
Joint Center for Earth Systems Technology, University of Maryland Baltimore County,
Baltimore, MD, USA
10
Department of Ear
th & Environmental Sciences, University of Michigan, Ann Arbor,
MI, USA
*Corresponding Author:
prattka@umich.edu
S
2
ToF
-ACSM
measurements and
excess sulfate
mass estimation
Non
-refractory particulate matter < 1 μm (NR
-PM
1
) mass concentrations of organics,
sulfate, nitrate, ammonium, and chloride were measured by the ToF
-ACSM
1, 2
at the
Olikt
ok Point field site
from August 18 to September 20, 2016.
These data were previously
reported and discussed by Gunsch et al.
3
ToF
-ACSM NR
-PM
1
sulfate mass concentrations
were correlated with PM
1
sulfate measured by IC
(r
2
=0.74) (Figure S
6).
Previously, Zhou
et al.
4
showed good correlation between the online aerosol mass spectrometer, ACSM, and
particle-
into
-liquid sampler (PILS)
-IC
measurements of sulfate at a mid
-latitude coastal
site.
ToF
-ACSM data have previously been used to estimate HMS mass concentrations.
5- 7
We followed equation 1 in Song et. al.
7
to estimate th
e ToF
-ACSM
excess
SO
+
and
SO
2
+
(excess sulfate) mass
, attributed to organosulfate compounds
. For inorganic sulfate, SO
+
and
SO
2
+
signals
were assumed to
contribute to a constant fraction of
H
푦푦
SO
x
+
, which is the
sum of SO
+
,
SO
2
+
,
SO
3
+
,
HSO
3
+
, and
H
2
SO
4
+
.
8
The two fractions,
SO
+
/
H
푦푦
SO
x
+
and
SO
2
+
/H
2
SO
4
+
in inorganic sulfate
, were
determined
for
time periods when they were both
low and stable
. The excess mass was then calculated by subtracting the calculated SO
+
and
SO
2
+
mass, assumed to be exclusively inorganic sulfate using these fractions, from the
measured SO
+
and
SO
2
+
mass.
ToF
-ACSM measurements suggest that excess sulfate could, on average, contribute
30% (0.07
휇휇
g/m
3
) of NR
-PM
1
“sulfate” and 6% of total NR
-PM
1
mass at Oliktok Point
(Figure S
4). For comparison, Song et al.
7
used this approach, assuming excess sulfate to
be HMS as an upper bound, and estimated the HMS contribution to be PM
1
sulfate in
Beijing winter haze to be 17%. However, the uncertainty in such a HMS estimation is
S
3
high.
9- 12
In this study, no correlation was observed between the
calculated
HMS
production
rate
and
the rate of change
in the ToF
-ACSM
excess SO
+
and
SO
2
+
mass
(R=
-0.02), thereby
inhibiting the ability to
determine
the contribution
of HMS
to the excess SO
+
and
SO
2
+
mass
(excess sulfate)
. The ToF
-ACSM excess sulfate mass
concentration
s were
~60
times higher,
on average, than the IC HMS mass concentration
s for filter sampl
e time periods
with
signals above the limit of detection
(Figure S
4).
Excess SO
+
and
SO
2
+
mass was 4% lower in fog periods compared to no-
fog periods,
supporting compounds
beyond HMS contributing to the excess SO
+
and
SO
2
+
mass.
One
of
the
other
possible
causes for different levels of excess SO
+
and
SO
2
+
signals
is the possible
variation
in the relative abundance of inorganic sulfates, such as (NH
4
)
2
SO
4
, NH
4
HSO
4
,
and H
2
SO
4
.
9, 10
Fragmentation of organosulfate
s sometimes can be indistinguishable from
inorganic sulfates,
11
which also adds to the
uncertainty of this estimation.
A recent study
also showed
that
the quantification of different sulfur organic compounds with this method
is challen
ging, as the fragmentation patterns only have subtle differences and are sensitive
to matrix effects.
12
In this study, no apparent correlation was found between RH and the ToF
-ACSM total
sulfate mass concentration (R=0.01), or between RH and the ATOFMS HMS
-containing
particle number fraction (R=0.12), in
contrast to several previous HMS studies in mid
-
latitude polluted sites.
5, 7, 13
The ToF
-ACSM sulfate mass concentration was also not
enhanced during fog periods. Nevertheless, the positive relationship between fog periods
and the prese
nce and abundance of HMS within individual particles supports the role of
fog processing in HMS formation in the Arctic oil field.
S
4
Mass transport processes of SO
2
and HCHO
at the gas
-l iquid i
nterface
, in the gas
phase, and
within fog d
roplets
The mass transport processes include diffusion of HCHO and SO
2
in the
gas and
aqueous phase
s, and their equilibrium at the gas
-liquid i
nterface.
The characteristic times
of
these processes were calculated to determine if
they
limit the rate of this aqueous
reaction. The
following
equations
(1-
5)
are from Seinfeld and Pandis
.
14
We did not
examine the characteristic times of the aqueous
-phase dissociation reactions here because
they tend to be short as compared with all timescales of interest.
14
The characteristic time
for gas
-phase diffusion to a particle
is described as
:
휏휏
푑푑푑푑
=
푅푅
푝푝
2
퐷퐷
푔푔
(1)
where R
p
is the
fog droplet size,
assumed to be 10
휇휇
m based on previous fog droplet
size
distributions measurements
on the North Slope of Alaska
.
15
퐷퐷
푑푑
is the diffusion coefficient
of SO
2
(1.03
×
10
−5
m
2
s
-1
)
16
or
HCHO (1.8
×
10
−9
m
2
s
-1
)
17
in the
gas phase.
The
characteristic time to
achieve e
quilibrium at the gas
-liquid interface (s)
is described as
:
휏휏
푠푠푠푠푠푠푠푠푠푠푠푠푠푠
=
2
휋휋휋휋휋휋휋휋 퐷퐷
푠푠푎푎
(
퐻퐻
퐴퐴
∗
)
2
(2)
where M is molecular weight, R is the universal gas constant, T is temperature,
퐷퐷
푠푠푎푎
is the
diffusion coefficient of SO
2
(1.62
×
10
−10
m
2
s
-1
)
16
or
HCHO (2
×
10
−10
m
2
s
-1
)
17
in the
aqueous
phase,
퐻퐻
퐴퐴
∗
is the Henry's law constant
(HCHO:
0.1593 mol m
-3
Pa
-1
; SO
2
: 0.0288
mol m
-3
Pa
-1
) at the
campaign average temperature (276K)
.
18
The characteristic time of
aqueous
-phase diffusion in a droplet is described as:
휏휏
푑푑푠푠
=
푅푅
푝푝
2
휋휋
2
퐷퐷
푎푎푎푎
(3)
S
5
The characteristic times
of mass transport
described
above were compared to the
characteristic time for aqueous phase chemical reactions:
휏휏
퐻퐻퐻퐻퐻퐻퐻퐻
=
1
(
k
1
α
1
+
k
2
α
2
)[
SO
2
(
aq
)
]
(4)
휏휏
푆푆퐻퐻
2
=
1
(
k
1
α
1
+
k
2
α
2
)[
HCHO
(
aq
)
]
(5)
The
measured
gas phase mole ratio
s of SO
2
and HCHO in Table S1 were examined here
.
HCHO hydration and dehydration were previously reported to be fast and generally not
rate-
limiting for HMS production.
7, 19, 20
For
a fog droplet
with pH
5, the average characteristic time for
aqueous phase
chemical
reaction
ranged from
10
to 3000 h
for
both
HCHO
and
SO
2
. In comparison, t
he
characteristic times under the same pH for gas
-phase diffusion, aqueous diffusion,
and gas
-
liquid interface equilibrium ranged
from 1
×
10
−12
to
6
×
10
−3
s. Therefore, t
he
characteristic times o
f mass transport were at least six
orders of magnitudes smaller than
the characteristic time for aqueous phase chemical reactions. Thus
, these mass transfer
processes are not rate-
limiting factors in HMS formation
in fog droplets
and do not need
to be taken into account for the HMS production rate calculation.
S
6
Single
-p article c
hemical c
haracterization
ATOFMS single
-particle chemical characterization
during this study is described
by
Gunsch et al.
3
T he 32,880 individual particle mass spectra were clustered using the neural
network algorithm ART
-2a.
21
Eight unique individual particle types were identified
based
on
the presence and intensity of ion peaks
with comparison to
previous laboratory and field
studies
.
22
The
characteristic mass spectra for each of the
single-
particle
types
(sea spray
aerosol (SSA), elemental carbon (
EC
), EC and organic carbon (ECOC
), OC
–amine–sulfate,
OC, biomass burning, mineral dust, and incineration particles
) are described in detail by
Gunsch et al
.
3
Briefly,
OC
–amine–sulfate
particles
, which accounted for 81%, by number,
of the HMS
-containing particles,
were characterized by OC (e.g.,
m/z
27, C
2
H
3
+
), sulfate
(m/z
−
97, HSO
4
-
), and alkylamines (
m/z
58, 59, 72, 86, 101, and 118).
The OC
–amine–
sulfate particles were most abundant
during oil field plumes
, with diethylamine and
trimethylamine attributed to anthropogenic sources within the oil field and trimethylamine
from biogenic sources.
3
. ECOC particles, which contributed to 15%, by number,
of the
HMS
-containing particles showed high degree of similarity to particles from diesel
combustion, and were identified by carbon cluster ions (C
n
+/-
), OC (
m/z
27, C
2
H
3
+
),
oxidized OC (
m/z
43, C
2
H
3
O
+
), sulfate (
m/z
-97, HSO
4
-
), and sulfuric acid (
m/z
-195,
H
2
SO
4
HSO
4
-
). SSA particle mass spectra featured abundant sodium and chloride peaks.
EC particle mass spectra were characterized by carbon cluster ions
and phosphate. The OC
particle type featured only positive ions, including
m/z
27 (C
2
H
3
+
), 37 (C
3
H
+
), and 43
(C
2
H
3
O
+
). The biomass burning particles also included only positive ions, including an
intense
m/z
39 (K
+
), plus organic carbon peaks. The lack of negative ions is indicative of
the accumulation of water during transport.
23
The incineration particles featured an intense
S
7
m/z
39 (K
+
), as well
m/z
113, 115, 117 (K
2
Cl
+/-
), as well as nitrate (
m/z
-46, 62) and copper
(
m/z
63, 65). Only the incineration particles contained both potassium and chloride
,
suggesting a likely interference of KCl
2
-
at
m/z
-111, as noted in the main text. Additional
details about the source attributions, number and mass concentrations, and time series of
these particle types are described by Gunsch et al.
3
S
8
PM
1
filter s
ampling and ion chromatography (IC)
analysis
The PM
1
medium volume sampler
(URG Corp., Chapel Hill, NC)
was elevated on a
platform ~10 m above ground level.
The sample flow was split into two filter lines, with
the quartz line having a
flow rate of 82 L min
-1
. The flow rate
was
checked prior to each
sample with a calibrated digital flowmeter
. The
90 mm diam
eter
quartz fiber filters
(Pall
Tissuquartz, Port Washington, NY)
were baked prior to sampling at 500°C for 12 h and
stored in aluminum foil
-lined petri dishes
(foil was also baked at 500°C for 12 h) and
storage bags in a
-10°C freezer before and after sampling.
Field blanks were taken
periodically by placing an unsampled filter in the filter holder, placing it in the sampler
momentarily, and then removing it and placing the filter in storage. Field blanks were
treated in the same manner as sampled filters. Filters were shipped from Oliktok Point to
Baylor University and then to Georgia Institute of Technology in a cooler with ice packs.
Prior to filter extraction for IC measurement, a 1.5-
4.5 cm
2
section of each filter was
removed for other analysis. T
he remaining
filter was extracted in 15 mL of deionized
water
(>18 MOhm) by sonication for 30 min in 50 mL centrifuge tubes (89039-
656, VWR
International, LLC
, Radnor, PA).
The
IC
analytical system included an autosampler (Dionex AS40, 5 mL vials with
filt
ering cap, Thermo Fisher Scientific, Waltham, MA) and a Metrohm Peak Ion
Chromatograph (Compact 761, Metrohm, Herisau, Switzerland). The IC was operated with
a 250 μL sample loop, a Metrosep A Supp 5
-150/4.0 anion column with
carbonate/bicarbonate eluent (
3.2 mM Na
2
(CO
3
)/1.0 mM NaHCO
3
), isocratic separation, a
flow rate of 0.70 mL/min, and conductivity detection. This method separates HMS
(HOCH
2
SO
3
-
) and sulfate
(SO
4
2-
) peaks
(Figure S2). The instrument was calibrated by
S
9
serial dilution of liquid standards composed of a mix of ions (IV
-STOCK
-59
-125 mL,
Inorganic Ventures, Christianburg, VA), whereas the HMS standard was run separately.
This standard was prepared from solid sodium hydroxymethanesulfonate
(Na
-HMS,
112704-
100
G, Sigma
-Aldrich) that had been desiccated and then liquid
concentrations
were based on a gravimetric analysis
of HMS (removing mass of Na
+
, i.e., measured Na
-
HMS mass times 0.83) and a known volume of pure (10 M
Ω
) water
. The HMS defined
here is the peak
in the IC chromatogram that has the same retention time as this HMS
standard. Calibrations show that a fraction of the HMS is converted to sulfate during the
IC analysis (less than 2% of measured sulfate for these data), which means that sulfate is
slightl
y over
-estimated when HMS is present, whereas the loss of HMS is accounted for in
the calibration.
The sulfate limit of detection ranged from 0.8-
3 ng/m
3
.
S
10
Figure S1.
Map of the North Slope of Alaska oil field
extent
, and the location of the three
closest airports
(Nuiqsut, Deadhorse
, and Ugnu
-Kuparuk).
The map background was
acquired
by ArcGIS 10.3.1
with the World Imagery basemap
. Oil
field extent was obtained
from http://dog.dnr.alaska.gov.
S
11
Figure S2.
A)
Mixed standard solution
and B) example ambient (
Sep. 7
-11, 2018 sample)
IC chromatogram
s showing separation of HMS
and sulfate
.
A
B
S
12
Figure S
3.
A) Temperature and B) relative humidity
(RH)
measured at Oliktok Point
, AK.
C) Observed fog periods
from meteorological measurements at Oliktok P
oint
and weather
archives (https://www.weatherforyou.com)
at
Utqiaġvik and
the three nearby airports
(Ugnu
-Kuparuk, Deadhorse
, and Nuiqsut). Periods with no data are shown as
white gaps.
S
13
Figure S
4.
Differences in average A) OC
-amine-
sulfate and B) ECOC individual particle
mass spectra for particles with and without HMS. Positive on the y
-axis means ion signals
higher in HMS
-containing particles, and negative on the y
-axis means ion signals higher
in particles lacking HMS.
S
14
Figure S
5.
Time series of the mass concentration of sulfate and excess sulfate from ToF
-
ACSM measurements at Oliktok Point, AK, as well as HMS and sulfate measured by IC.
Oliktok Point fog periods (Figure S2) are also shown.
Th
e last filter sample was collected
from Sep. 14 9:00 –
Sep. 15 7:47 and Sep. 17 8:34 –
Sep. 18 7:37, with the break in
sampling shown.
HMS mass concentrations below the limit of detection are shown as 0.
5×
the limit of detection.
S
15
Figure S
6.
PM
1
sulfate measured by ToF
-ACSM (averaged by filter sampling periods,
with 30% uncertainty
24
shown) versus IC, with 20% uncertainty shown.
S
16
Table S1.
Average SO
2
and HCHO concentrations
at less than 2000
m above sea level
measurements
from
the
ATom
-1 flight campaign
. For measurements b
elow
the
detection
limit
*, half of the measurement
uncertainty values were
used in the production rate
calculations.
Location
Longitude
Latitude
Altitude
(m)
Date
Time
(UTC)
Average
[SO
2
] (ppb)
(uncertainty)
Average
[HCHO]
(
ppb
)
Beaufort Sea 1
-
136.79 to
-137.
77
77.9
0
to
78.46
950
-
2000
8/1/16 20:40:00
to
8/1/16 20:47:00
-
0.029*
(0.022)
0.083
Beaufort Sea 2
-
142.
77 to
-142.82
78.
93 to
78.48
200
-
2000
8/1/16 21:05:00
to
8/1/16 21:17:00
0.001*
(0.014)
0.083
Deadhorse
-
148.
22 to
-148.57
70.64 to
69.97
70
-
2000
8/1/16 22:30:00
to
8/1/16 22:44:00
0.040
0.282
Fairbanks
-
147.
98 to
-148.11
65.24 to
64.7
200
-
2000
8/1/16 23:30:00
to
8/1/16 23:41:00
0.237
0.794
Anchorage
-
150.
4
0
to
-149.98
61.
43 to
61.17
50
-
2000
8/2/16 00:
31:00
to
8/2/16 00:39:00
0.194
0.631
S
17
References:
1. Fröhlich, R.; Cubison, M.; Slowik, J.; Bukowiecki, N.; Prévôt, A.; Baltensperger, U.;
Schneider, J.; Kimmel, J.; Gonin, M.; Rohner, U., The ToF
-ACSM: a portable aerosol
chemical speciation monitor with TOFMS detection.
Atmospheric Measurement
Techniques
2013,
6
, (11), 3225-
3241.
2. Ng, N. L.; Herndon, S. C.; Trimborn, A.; Canagaratna, M. R.; Croteau, P. L.; Onasch,
T. B.; Sueper, D.; Worsnop, D. R.; Zhang, Q.; Sun, Y. L.; Jayne, J. T., An Aerosol
Chemical Speciation Monitor (ACSM) for Routine Monitoring of the Composition
and Mass Concentrations of Ambient Aerosol.
Aerosol Science and Technology
2011,
45
, (7), 780-
794.
3. Gunsch, M. J.; Liu, J.; Moffett, C. E.; Sheesley, R. J.; Wang, N.; Zhang, Q.; Watson, T.
B.; Pratt, K. A., Diesel Soot and Amine-
Containing Organic Sulfate Aerosols in an
Arctic Oil Field.
Environmental Science & Technology
2019,
54
, (1), 92
-101.
4. Zhou, S.; Collier, S.; Xu, J.; Mei, F.; Wang, J.; Lee, Y.-
N.; Sedlacek, A. J.; Springston, S.
R.; Sun, Y.; Zhang, Q., Influences of upwind emission sources and atmospheric
processing on aerosol chemistry and properties at a rural location in the
northeastern U.S.
Journal of Geophysical Research
2016,
121
, 6049-
6065.
5. Gilardoni, S.; Massoli, P.; Paglione, M.; Giulianelli, L.; Carbone, C.; Rinaldi, M.;
Decesari, S.; Sandrini, S.; Costabile, F.; Gobbi, G. P., Direct observation of aqueous
secondary organic aerosol from biomass-
burning emissions.
Proceedings of the
Nationa
l Academy of Sciences
2016,
113
, (36), 10013-
10018.
6. Ge, X.; Zhang, Q.; Sun, Y.; Ruehl, C. R.; Setyan, A., Effect of aqueous
-phase processing
on aerosol chemistry and size distributions in Fresno, California, during
wintertime.
Environmental Chemistry
2012,
9
, (3), 221-
235.
7. Song, S.; Gao, M.; Xu, W.; Sun, Y.; Worsnop, D. R.; Jayne, J. T.; Zhang, Y.; Zhu, L.; Li,
M.; Zhou, Z., Possible heterogeneous chemistry of hydroxymethanesulfonate
(HMS) in northern China winter haze.
Atmospheric Chemistry and Physi
cs
2019,
19
, (2), 1357-
1371.
8. Shao, J.; Chen, Q.; Wang, Y.; Lu, X.; He, P.; Sun, Y.; Shah, V.; Martin, R. V.; Philip, S.;
Song, S.; Zhao, Y.; Xie, Z.; Zhang, L.; Alexander, B., Heterogeneous sulfate aerosol
formation mechanisms during wintertime Chinese
haze events: air quality model
assessment using observations of sulfate oxygen isotopes in Beijing.
Atmos. Chem.
Phys.
2019,
19
, 6107–6123.
9. Allan, J. D.; Delia, A. E.; Coe, H.; Bower, K. N.; Alfarra, M. R.; Jimenez, J. L.;
Middlebrook, A. M.; Drewnick, F.; Onasch, T. B.; Canagaratna, M. R., A generalised
method for the extraction of chemically resolved mass spectra from Aerodyne
aerosol mass spectrometer data.
Journal of Aerosol Science
2004,
35
, (7), 909-
922.
10.
Jayne, J. T.; Leard, D. C.; Zhang, X. F.; Davidovits, P.; Smith, K. A.; Kolb, C. E.;
Worsnop, D. R., Development of an aerosol mass spectrometer for size and
composition analysis of submicron particles.
Aerosol Sci. Technol.
2000,
33
, (1
-2),
49-
70.
11.
Farmer, D.; Matsunaga, A.; Docherty, K.; Su
rratt, J.; Seinfeld, J.; Ziemann, P.;
Jimenez, J., Response of an aerosol mass spectrometer to organonitrates and
organosulfates and implications for atmospheric chemistry.
Proceedings of the
National Academy of Sciences
2010,
107
, (15), 6670-
6675.