Subscriber access provided by UNIV OF SCIENCES PHILADELPHIA
Environmental Processes
Nitrogen-containing, light-absorbing oligomers produced in aerosol particles exposed to methylglyoxal, photolysis, and cloud cycling David O De Haan, Enrico Tapavicza, Matthieu Riva, Tianqu Cui, Jason Douglas Surratt, Adam C. Smith, Mary-Caitlin Jordan, Shiva Nilakantan, Marisol Almodovar, Tiffany N. Stewart, Alexia de Loera, Audrey C. De Haan, Mathieu Cazaunau, Aline Gratien, Edouard Pangui, and Jean-Francois Doussin Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b06105 • Publication Date (Web): 06 Mar 2018 Downloaded from http://pubs.acs.org on March 7, 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 31
Environmental Science & Technology
1
Nitrogen-containing, light-absorbing oligomers
2
produced in aerosol particles exposed to
3
methylglyoxal, photolysis, and cloud cycling.
4
David O. De Haan,*1 Enrico Tapavicza,*2 Matthieu Riva,3, # Tianqu Cui,3 Jason D. Surratt,3
5
Adam C. Smith,2 Mary-Caitlin Jordan,2 Shiva Nilakantan,2 Marisol Almodovar,2 Tiffany N.
6
Stewart,1 Alexia de Loera,1 Audrey C. De Haan,1 Mathieu Cazaunau,4 Aline Gratien,4
7
Edouard Pangui,4 Jean-François Doussin4
8 9 10 11 12 13 14
1
Department of Chemistry and Biochemistry, University of San Diego, 5998 Alcala Park, San
Diego CA 92110 USA 2
Department of Chemistry and Biochemistry, California State University Long Beach, 1250
Bellflower Boulevard, Long Beach, CA, 90840, USA 3
Department of Environmental Sciences and Engineering, Gillings School of Global Public
Health, The University of North Carolina at Chapel Hill, Chapel Hill, NC, 27599 USA. 4
Laboratoire Interuniversitaire des Systèmes Atmosphériques (LISA), UMR7583, CNRS,
15
Université Paris-Est-Créteil (UPEC) et Université Paris Diderot (UPD), Institut Pierre Simon
16
Laplace (IPSL), 94010 Créteil, France
17
#
Now at Université Claude Bernard Lyon 1, CNRS, IRCELYON, F-69626, Villeurbanne,
18
France
19
*Corresponding_Authors,
[email protected], 1 619 260-6882,
[email protected],
20
1 562 985-7830
ACS Paragon Plus Environment 1
Environmental Science & Technology
Page 2 of 31
21
KEYWORDS: brown carbon, aqueous secondary organic aerosol formation, alpha-dicarbonyl,
22
imine, imidazole, photolytic cloud processing, computational absorption spectra.
23
ABSTRACT. Aqueous methylglyoxal chemistry has often been implicated as an important
24
source of oligomers in atmospheric aerosol. Here we report on chemical analysis of brown
25
carbon aerosol particles collected from cloud cycling / photolysis chamber experiments, where
26
gaseous
27
methylammonium sulfate seed particles. Eighteen N-containing oligomers were identified in the
28
particulate phase by liquid chromatography / diode array detection / electrospray ionization high-
29
resolution quadrupole time-of-flight mass spectrometry. Chemical formulas were determined
30
and, for 6 major oligomer products, MS2 fragmentation spectra were used to propose tentative
31
structures and mechanisms.
32
product structures by an ab initio second order algebraic-diagrammatic-construction / density
33
functional theory approach. For five structures, matching calculated and measured absorption
34
spectra suggest that they are dominant light-absorbing species at their chromatographic retention
35
times. Detected oligomers incorporated methylglyoxal and amines, as expected, but also pyruvic
36
acid, hydroxyacetone, and significant quantities of acetaldehyde. The finding that ~80% (by
37
mass) of detected oligomers contained acetaldehyde, a methylglyoxal photolysis product,
38
suggests that daytime methylglyoxal oligomer formation is dominated by radical addition
39
mechanisms involving CH3CO*. These mechanisms are evidently responsible for enhanced
40
browning observed during photolytic cloud events.
41
Introduction
42 43
methylglyoxal
and
methylamine
interacted
with
glycine,
ammonium
or
Electronic absorption spectra were calculated for six tentative
Methylglyoxal is believed to contribute significantly to the atmospheric formation of oligomerized1 and light-absorbing secondary organic aerosol (SOA) species2,
3
via aqueous-
ACS Paragon Plus Environment 2
Page 3 of 31
Environmental Science & Technology
44
phase reactions.4 Methylglyoxal uptake to clouds, however, is partly reversible: lab simulations
45
of cloud droplet evaporation have shown that 65 to 80% of the methylglyoxal present in a droplet
46
evaporates when the droplet dries.5, 6 Smaller drying-induced losses of aqueous organic species
47
have been observed in field studies, especially at night.7 In aqueous aerosol particles, dissolved
48
methylglyoxal forms oligomerized SOA through rapid self-reactions6, 8 or reactions with SO2,9
49
ammonium salts,10,
50
form products with low enough vapor pressures that they may remain in the aerosol phase.18
51
Alternatively, evaporation of volatile compounds may be slowed19,
52
formation causes the organic phase to solidify.21-24.
11
other aldehydes,12,
13
amine species,14 or oxidants.15-17 These reactions
20
if extensive oligomer
53
Prior studies of light-absorbing brown carbon (BrC) formed by aldehyde – ammonia reactions
54
have found that it is quickly bleached when photolyzed in the aqueous-phase,25, 26 suggesting that
55
atmospheric secondary BrC can form only at night. However, photolysis of BrC may lead to
56
greater SOA formation due to photosensitization, a process where light-absorbing species absorb
57
light and then trigger the oxidation of other species.27 In our recent study, ammonium sulfate
58
(AS) and amine-containing seed particles exposed to methylglyoxal and methylamine gas and
59
cloud processing became more light-absorbing in experiments with photolysis than in dark
60
experiments.28 This effect was attributed to autocatalytic photosensitization, where the aqueous-
61
phase photolysis of BrC produces radical species, which oligomerize with other species to form
62
even greater amounts of BrC. Here, we report high-resolution mass spectrometric measurements
63
of both the gas and aerosol phases in the autocatalytic photosensitization experiments, which
64
demonstrate the formation of oligomerized aerosol-phase products and the gas/particle exchange
65
of reactant and product species through multiple cloud cycles. We find that particle-phase
66
oligomers in most experiments are dominated by methylglyoxal photolysis products, suggesting
ACS Paragon Plus Environment 3
Environmental Science & Technology
Page 4 of 31
67
that aqueous radical addition reactions make large contributions to the formation of not only
68
oligomers15 but also aerosol BrC.
69 70
Methods
71
CESAM chamber.
Methylglyoxal addition experiments in the pressure and temperature-
72
controlled, stirred 4.2 m3 CESAM chamber29 have been described earlier.28 Briefly, different
73
types of ammonium and amine-containing seed aerosol particles were generated from 1, 2, and
74
10 mM aqueous solutions of AS, methylammonium sulfate (MAS, adjusted to pH 4 with
75
H2SO4), or glycine, respectively, and then exposed to 1 to 7 ppm methylglyoxal gas. Gas-phase
76
methylamine was present at 0.3 – 1.0 ppm in every experiment, added directly to AS seed
77
experiments or equilibrating from aerosols or walls in MAS or glycine seed experiments,
78
respectively. Cloud events lasting 3-10 min each were triggered by further additions of
79
expansion-cooled water vapor, sometimes combined with lowering pressure in the humid
80
chamber from 1.01 to 0.93 atm.
81
controllable, cloud events were repeated to achieve a total cloud processing time of 15–18 min.
82
per experiment Three xenon arc lamps illuminated the chamber for ~2 h during and after the last
83
cloud event in each experiment.
Since the length of an individual cloud event was not
84
Gas generation and quantitation. Aqueous solutions of methylglyoxal (Alfa-Aesar) were
85
concentrated by pumping on a vacuum line until the entire solution was highly viscous. Gases
86
released by gentle heating were then collected in an evacuated glass bulb and flushed with dry N2
87
into the chamber.
88
reaction time-of-flight mass spectrometry (PTR-TOF-MS, Kore Technology). Pure methylamine
89
gas (Fluka, >99%) was added to reach 1 ppm in the chamber and quantified by PTR-TOF-MS.
Methylglyoxal was quantified by long path FTIR30 and proton-transfer
ACS Paragon Plus Environment 4
Page 5 of 31
90
Environmental Science & Technology
Aerosol chemical characterization.
After each experiment, cloud-processed aerosol was
91
collected without a denuder at constant pressure for ~14 h at 16 L min-1 onto a Teflon filter
92
(Tisch, 1.0 m pore size). Addition of dry N2 during filter sampling caused RH and gas-phase
93
organic species concentrations in the chamber to slowly drop by a factor of 10; thus, volatile,
94
semivolatile, and reversibly-formed aqueous aerosol species had ample opportunity to evaporate.
95
Filters were stored at -20 ºC under dark conditions until chemical analysis.
96
characterization of the particles was performed by ultra-performance liquid chromatography
97
interfaced to high-resolution quadrupole time-of-flight mass spectrometry equipped with
98
electrospray ionization (UPLC/ESI-HR-Q-TOFMS; 6520 Series, Agilent) operated in negative
99
and positive ion modes. Filters were extracted with 22 mL of methanol (LC-MS Chromasolv-
100
grade, Sigma-Aldrich) by sonication for 45 min. Methanol extracts were blown dry under a
101
gentle N2 stream at ambient temperature. The dried extracts were reconstituted with 150 µL of a
102
50:50 (v/v) solvent mixture of methanol and Milli-Q water (>18.2 MΩ). Five and 10 µL aliquots
103
were injected onto the column (Waters ACQUITY UPLC HSS T3 column, 2.1 × 100 mm, 1.8
104
µm particle size) and eluted at a 0.3 mL min-1 flow rate with a solvent mixture of methanol and
105
water, each containing 0.1% acetic acid or 0.1% ammonium acetate (LC-MS Chromasolv-grade,
106
Sigma-Aldrich) for negative (–) and positive (+) ion modes, respectively. Unretained compounds
107
were detected by diode array at 0.94 min and ESI-HR-Q-TOFMS at 1.10 min. Due to the lack of
108
standards for detected N-containing oligomer compounds, caffeine (> 99 %, Sigma-Aldrich) was
109
used as a surrogate quantification standard for positive ion mode signals, resulting in order of
110
magnitude uncertainties in estimated concentrations.
Chemical
111
Computational estimates of absorbance. Electronic structure and ab initio molecular dynamics
112
simulations (AIMD) were carried out with the TURBOMOLE quantum chemistry package.31
ACS Paragon Plus Environment 5
Environmental Science & Technology
Page 6 of 31
113
Ground state equilibrium structures of the proposed compounds were optimized using density
114
functional theory (DFT) with the approximate PBE32 and PBE033 exchange correlation
115
functionals. AIMD simulations were carried out using the PBE functional and employ the
116
resolution of identity approximation34 and the def2-SVP basis set.35
117
energies and excited state properties were calculated using second-order approximate coupled
118
cluster singles and doubles theory (CC2) and the closely related36 algebraic-diagrammatic
119
construction through second order (ADC(2))37 in combination with the resolution of identity
120
approximation38-40 and the def2-TZVP basis set.35 To take into account solvent effects we
121
applied ADC(2) with the conductor-like screening model (COSMO).41 All calculations have
122
been carried out with a dielectricity constant of 80.1 and a refractive index of 1.3325. The
123
ADC(2)/COSMO approach has been successfully applied to incorporate solvent effects on
124
electronic excitations.42, 43
Electronic excitation
125
Conformational changes of the molecules were taken into account by generating a Boltzmann
126
ensemble of each molecule using AIMD. For cyclic compounds, Born-Oppenheimer molecular
127
dynamics (BOMD) was efficient enough to sample the conformational space. For non-cyclic,
128
flexible molecules, we used replica-exchange molecular dynamics (REMD)44-46 to obtain an
129
ensemble accounting for the different conformers. In BOMD and REMD a time step of 50 a.u
130
and a Nosé-Hoover thermostat47, 48 with a characteristic response time of 500 a.u. was used. CC2
131
and ADC(2) absorption spectra of the proposed molecules were calculated for the equilibrium
132
ground state structures and by averaging spectra of >250 snapshot structures obtained from
133
BOMD or REMD, as previously described.44,
134
excitation energies were broadened using a Gaussian line shape with full widths at half
135
maximum (FWHM) between 0.1 and 0.5 eV.49
49-51
For comparison with experimental spectra,
ACS Paragon Plus Environment 6
Page 7 of 31
Environmental Science & Technology
136
Since experimental spectra were measured at slightly acidic conditions, all structures were
137
generated as protonated, singly positively charged species. For compounds with several amino or
138
imino groups, we carried out geometry optimization and total energy calculations prior to MD
139
simulations in order to obtain the most likely protonation state. The lowest ten excited singlet
140
states were calculated for each spectrum.
141 142
Results
143
A summary of experiments involving gas-phase methylglyoxal where product species were
144
detected in filter extracts is shown in Table 1. Each experiment had at least two reduced nitrogen
145
species present:
146
methylamine. The particle-phase species detected in each experiment are listed in Table 2, with
147
possible precursor molecule combinations for each product formula listed in Table S1.
one ionized in the seed aerosol (AS, MAS, or glycine), and gas-phase
148 149
150 151 152 153 154 155
Table 1: Summary of Methylglyoxal Gas Addition Chamber Experiments Expt #
[MeGly]a (ppm)
Cloud [MeAm] Seed Aerosol processing (ppm) time Type (min)
Total illum. time (h)
Aerosol mass on filter (ug)b
Products detected on filter
A
0.80
1.0
AS
15
2.5
14
3
B
3.6a
~0.3c,d
MAS
16
1.9
50
15
C
6.9a
~0.6d
Glycine 18
1.7
70
6
Notes: a: Peak methylglyoxal (MeGly) concentrations as detected by PTR-TOF-MS. In experiments B and C, [MeGly] was 1.0 ppm prior to gas phase addition due to equilibration from the walls of the humidified chamber. b: based on SMPS size distribution and assumed density = 1 g/cm3. c: peak methylamine (MeAm) concentration detected by PTR-TOF-MS, volatilized from methylammonium sulfate (MAS) seeds or d: equilibrated from walls of humidified chamber.
156
ACS Paragon Plus Environment 7
Environmental Science & Technology
Page 8 of 31
157
Table 2: Water-Soluble Aerosol-Phase Products Quantified in Each Methylglyoxal Gas
158
Addition Experiment by UPLC/(+)ESI-HR-QTOFMS Quantity (ng m-3)a Detected Mass
Formula
Ionizing Species
Retention time tr (min) 1.6 1.1c 1.1c 1.1c 1.2 5.6 1.1c/2.2 2.5 1.1c 2.3 1.1c/2.6 6.0 1.1c/1.8 5.3 5.5 1.1c/2.8 5.9 7.0
∆ (ppm)b
Expt A (AS)
Expt B Expt C (MAS) (Gly)
159 160 161 162
102.0921 C5H11NO H+ 0.19 0.0 650 0.0 111.0901 C6H10N2 H+ 3.04 0.0 380 32 + 116.1067 C6H13NO H 4.05 0.0 25 0.0 132.1019 C6H13NO2 H+ 1.16 0.0 40 0.0 133.0999 C5H12N2O2 H+ 1.17 0.0 17 0.0 + 138.0968 C8H11NO H 1.79 0.0 54 0.0 + 141.1022 C7H12N2O H 3.23
