Eshani Hettiarachchi1, Vicki H Grassian1. 1. Department of Chemistry and Biochemistry, University of California San Diego, 9500 Gilman Drive, La Jolla, California 92093, United States.
Abstract
Organonitrates (ON) are important components of secondary organic aerosols (SOAs). α-Pinene (C10H16), the most abundant monoterpene in the troposphere, is a precursor for the formation of several of these compounds. ON from α-pinene can be produced in the gas phase via photochemical processes and/or following reactions with oxidizers including hydroxyl radical and ozone. Gas-phase nitrogen oxides (NO2, NO3) are N sources for ON formation. Although gas-phase reactions of α-pinene that yield ON are fairly well understood, little is known about their formation through heterogeneous and multiphase pathways. In the current study, surface reactions of α-pinene with nitrogen oxides on hematite (α-Fe2O3) and kaolinite (SiO2Al2O3(OH)4) surfaces, common components of mineral dust, have been investigated. α-Pinene oxidizes upon adsorption on kaolinite, forming pinonaldehyde, which then dimerizes on the surface. Furthermore, α-pinene is shown to react with adsorbed nitrate species on these mineral surfaces producing multiple ON and other oxidation products. Additionally, gas-phase oxidation products of α-pinene on mineral surfaces are shown to more strongly adsorb on the surface compared to α-pinene. Overall, this study reveals the complexity of reactions of prevalent organic compounds such as α-pinene with adsorbed nitrate and nitrogen dioxide, revealing new heterogeneous reaction pathways for SOA formation that is mineralogy specific.
Organonitrates (ON) are important components of secondary organic aerosols (SOAs). α-Pinene (C10H16), the most abundant monoterpene in the troposphere, is a precursor for the formation of several of these compounds. ON from α-pinene can be produced in the gas phase via photochemical processes and/or following reactions with oxidizers including hydroxyl radical and ozone. Gas-phase nitrogen oxides (NO2, NO3) are N sources for ON formation. Although gas-phase reactions of α-pinene that yield ON are fairly well understood, little is known about their formation through heterogeneous and multiphase pathways. In the current study, surface reactions of α-pinene with nitrogen oxides on hematite (α-Fe2O3) and kaolinite (SiO2Al2O3(OH)4) surfaces, common components of mineral dust, have been investigated. α-Pinene oxidizes upon adsorption on kaolinite, forming pinonaldehyde, which then dimerizes on the surface. Furthermore, α-pinene is shown to react with adsorbed nitrate species on these mineral surfaces producing multiple ON and other oxidation products. Additionally, gas-phase oxidation products of α-pinene on mineral surfaces are shown to more strongly adsorb on the surface compared to α-pinene. Overall, this study reveals the complexity of reactions of prevalent organic compounds such as α-pinene with adsorbed nitrate and nitrogen dioxide, revealing new heterogeneous reaction pathways for SOA formation that is mineralogy specific.
Aerosols
are known to play a key role in climate and air quality,
with high concentrations leading to adverse health effects (e.g.,
cardiovascular disease, lung cancers).[1−4] Aerosols also cause reduced visibility and
haze conditions associated with urban air pollution.[5] α-Pinene, C10H16, is the most
abundant atmospheric monoterpene with an average estimated emission
of 66 Tg yr–1. It is an important precursor to several
compounds that lead to the formation of secondary organic aerosols
(SOAs).[6−9] These compounds include oxygenated organic compounds found in polluted
atmospheric conditions as well as organosulfates, ROSO3H (OS), and organonitrates, RONO2 (ON), all contributing
to SOA formation.[10−14] These α-pinene-derived compounds have been detected in ambient
air.[15−18]Atmospheric ON formation occurs by either OH radical-initiated
photochemical reactions or NO3 radical-initiated nocturnal
reactions of anthropogenic and biogenic volatile organic compounds.
These reactions can produce ON compounds with vapor pressures low
enough to condense and form SOA.[19−23] In one such study, the formation of a series of ON
compounds from reactions of α-pinene with NO3 radicals
was observed.[22] In another study, the photochemistry
of the hydroxyl radical oxidation of α-pinene under high NO conditions was studied and the formation
of a range of ON compounds from α-pinene in the presence of
nitric oxide (NO) was reported.[24] In addition,
reactions of α-pinene with atmospheric oxidizers, such as O3 and OH radicals lead to the formation of a range of α-pinene
oxidation products that are known as first generation oxidation products.[4,6,25,26] Several of these products have been identified as pinonaldehyde,
pinonic acid, and α-pinene oxide, as well as a structurally
diverse set of organic peroxides. These oxidation products can also
act as a reactant in generating atmospheric ON compounds.[27] These oxidation reactions are generally formed
in the daytime, suggesting the formation of ON compounds from α-pinene
primarily occurs in the presence of atmospheric oxidizers and/or photochemistry.[19,28] All of these studies are primarily focused on understanding SOA
formation from gas-phase condensation.In this study, heterogeneous
reactions of α-pinene nitrogen
oxides on iron oxide (α-Fe2O3) and kaolinite
(SiO2Al2O3(OH)4) surfaces
are investigated to understand heterogeneous SOA formation on mineral
dust aerosol. Both iron oxides and kaolinite are reactive components
of mineral dust aerosol.[29] Heterogeneous
chemistry of gas-phase species, both organic and inorganic components
with atmospheric mineral dust aerosols, have been widely studied.
For instance, adsorption of α-pinene on different types of silica
surfaces in the absence of photochemistry showed reversible adsorption.[30,31] Additionally, the adsorption of NO2 and HNO3 on mineral dust has been extensively investigated.[32−35] NO2 is emitted to the atmosphere primarily via fossil
fuel combustion and vehicle exhausts.[36−39] Because of the higher correlation
of atmospheric NO2 concentrations to traffic, it is used
as a traffic-related air pollution marker.[38] Furthermore, NO2 concentrations vary significantly over
shorter distances in shorter timeframes.[38,39] Previous studies have shown that mineral surfaces exposed to NO2 yield two major surface-adsorbed species (nitrate and nitrite)
thereby producing a surface for organic compounds such as α-pinene
to interact with and facilitate the formation of ON. Similarly, gas-phase
HNO3 acid reacts on mineral dust particles to yield adsorbed
nitrates and in some cases molecular nitric acid. Despite this knowledge,
little is known about the ability of mineral dust surfaces to mediate
and catalyze ON formation. Therefore, the role of surface chemistry
in reactions of α-pinene with atmospheric NO2 and
HNO3 on iron oxide (hematite, α-Fe2O3) and kaolinite (SiO2Al2O3(OH)4) surfaces is investigated here. The experiments
in the present study were conducted under dark conditions to better
understand the underlying role of the mineral surface. For these studies,
both Fourier transform infrared (FTIR) spectroscopy and high-resolution
mass spectrometry (HRMS) were used to better understand the chemistry
of α-pinene on these atmospherically relevant surfaces.
Materials and Methods
Transmission FTIR Experiments
and Methods
Transmission FTIR spectroscopy was used to study
the adsorption
of α-pinene on hematite at 296 ± 1 K. Additional details
of this system have been previously described.[40−44] The mineral particles (hematite, α-Fe2O3, 99+%, Fisher Scientific or kaolinite, SiO2Al2O3(OH)4, Sigma-Aldrich) with
a BET surface area of 80 ± 10 and 8.4 ± 0.5 m2/g, respectively, were heated in an oven at 200 °C overnight
to remove organic contaminants and then pressed onto one-half of a
tungsten grid (ca. 5 mg). The grid was then placed in the sample IR
cell compartment, held by two stainless steel jaws. Following the
preparation of the mineral sample and placement in the IR cell, the
system was evacuated for 4 h using a turbomolecular pump. The mineral
sample was subsequently exposed to 50% RH water vapor for 2 h to hydroxylate
the surface. Once hydroxylated, the system was evacuated for another
6 h to remove water vapor in the chamber. After evacuation, the sample
was exposed to the desired pressure of α-pinene (99+%, Sigma-Aldrich)
for 20 min under dry conditions (RH < 1%). The α-pinene sample
was degassed at least three times with consecutive freeze–pump–thaw
cycles prior to use.Reactions of NO2 (26.5 ppm in
N2, Airgas) with α-pinene on mineral surfaces were
studied. First, α-pinene was allowed to adsorb on the mineral
surface for >20 min. The desired pressure of NO2 (7
mTorr)
was then introduced into the IR cell. FTIR spectra were collected
over a 4 h period through both halves of the tungsten grid to monitor
gas-phase and surface-adsorbed α-pinene. Following adsorption,
the system was evacuated overnight. The reactions of surface-adsorbed
nitrates with α-pinene were then studied. First, to obtain a
nitrated mineral surface, the mineral surface was introduced to nitric
acid vapor taken from a concentrated mixture of H2SO4 (∼96 wt %)/HNO3(∼70 wt %) in a 3:1
ratio.[45] The mineral surface was exposed
to nitric acid vapor for 4 h and evacuated overnight. Following nitration,
α-pinene was introduced to these modified mineral surfaces and
the reactions were monitored for over 4 h. The experiments with HNO3 were conducted in a different experimental but with a very
similar setup (instead of stainless steel a Teflon coated cell and
a glass system was used).Prior to and following the exposure
to α-pinene, single-beam
spectra (250 scans) of the surface and gas phase were acquired using
a resolution of 4 cm–1 and covering the spectral
range 600–4000 cm–1. For kinetic studies,
spectra were acquired every 15 s for the first 15 min of the adsorption.
Absorption spectra of α-pinene on mineral particles are reported
as the difference in the mineral spectra before and after exposure
to α-pinene. Absorption bands because of gas-phase α-pinene,
measured through the blank half of the tungsten grid, were subtracted
to obtain FTIR spectra of adsorbed α-pinene only.
HRMS Experiments and Methods
Organic
products formed on mineral surfaces following reactions of α-pinene
with NO2 and adsorbed nitrates in the dark were analyzed
using a direct-injection linear ion trap (ThermoFisher Orbitrap) high-resolution
mass spectrometer (HRMS). Adsorbed products were extracted from the
hematite or kaolinite solid substrate using methanol (CH3OH, Fisher Scientific, HPLC grade) as the solvent. The sample vial,
syringe, and all other glassware used in the transfer process were
cleaned prior to use with methanol, and Milli-Q water (Millipore Sigma,
18.2 MΩ), and baked in an oven at 500 °C to further remove
trace organics. Plastic vials used in sample preparation were sonicated
in methanol for 60 min, and washed thoroughly prior to use. All of
the samples were stored at −20 °C and analyzed within
24 h of collection. Separate experiments with approximately 100 mg
of hematite were conducted for α-pinene adsorption on hematite,
and NO2 + α-pinene reaction on hematite for extraction
and HRMS detection.HRMS analysis in both positive electrospray
ionization (ESI) and negative ESI modes was used, although the detected
ions for the reactions were observed in positive ESI mode ([M + H]+). The heated electrospray ionization (HESI) source was operated
at 100 °C. The ESI capillary was set to a voltage of 3.5 kV at
350 °C. The HESI-Orbitrap MS was calibrated prior to use. Mass
spectra were acquired with a mass range of 50–2000 Da. Peaks
with mass tolerance of >2 ppm were rejected. Compositions were
calculated
with the following element ranges: 12C, 0–60; 1H, 0–100;
16O, 0–10; 14N, 0–5; 23Na, 0–5; 39K, 0–5;
56Fe, 0–5. Tandem mass spectrometry (MS/MS) with collision
energy of 40 eV was used for structure determination.
Results and Discussion
α-Pinene Adsorption
on Hematite and Kaolinite
Surfaces at 296 K
Mineral particle surfaces exposed to gas-phase
α-pinene at various pressures under dry conditions show new
spectral features because of surface adsorption, as seen in Figure . Most notable are
the infrared absorption bands in the regions extending from 2800 to
3800 cm–1 and from 1250 to 1500 cm–1. The bands between 2800 and 3100 cm–1 were assigned
to various C—H stretching vibrations and the bands between
1300 and 1500 cm–1 were assigned to various C—H
bending modes. (Table ).[30,46,47]
Figure 1
FTIR spectra
of α-pinene adsorbed on (A) hematite and (B)
kaolinite as a function of α-pinene pressure in the spectral
regions from 2800 to 3800 cm–1 and from 1250 to
1500 cm–1. The absorbance scale is shown in the
upper left for each surface.
Table 1
FTIR Peak Assignments of Adsorbed
α-Pinene on Hematite and Kaolinite[30,46,47,49]
Peak Frequencies
Hematite
Kaolinitea
Peak Assignments
1329, 1336, 1367, 1383,
1439, 1449
1329,
1366, 1449
aliphatic
CH2 and CH3 bending, twisting and wagging modes
1471
1472
vinyl C—H bending
mode
1650
n.o.b
C=C stretching
mode
2837, 2881, 2919, 2975,
2988
2880, 2887,
2952, 2965,
2972, 2984
aliphatic
CH2, and CH3 stretching modes, and Fermi resonances
3025
3021
vinyl C—H stretching
mode
3605, 3632, 3640
n.o.
surface O—H hydrogen
bonded stretching mode
3670, 3684
n.o.
isolated surface O—H
stretching mode
Peak frequencies from the initial
FTIR spectrum shown in Figure 2A.
n.o. = not observed.
FTIR spectra
of α-pinene adsorbed on (A) hematite and (B)
kaolinite as a function of α-pinene pressure in the spectral
regions from 2800 to 3800 cm–1 and from 1250 to
1500 cm–1. The absorbance scale is shown in the
upper left for each surface.Peak frequencies from the initial
FTIR spectrum shown in Figure 2A.n.o. = not observed.For hematite, the frequencies observed for the absorption bands
found for adsorbed α-pinene corresponded closely to its gas-phase
vibrational frequencies suggesting that α-pinene is molecularly
adsorbed onto the hematite surface (Figure A).[30,48] With increasing α-pinene
pressure, there is an increase in intensity of bands at 3605, 3632,
and 3640 cm–1 with a concomitant loss of surface
hydroxyl groups on hematite at 3670 and 3684 cm–1. These can be attributed to the formation of π hydrogen bonds
between α-pinene and the surface hydroxyl groups.[30] The spectral feature around 1650 cm–1 was assigned to the C=C bond stretching mode of adsorbed
α-pinene.[47] Upon evacuation overnight,
almost all of the adsorbed α-pinene is removed, thus suggesting
reversible adsorption of α-pinene on hematite surfaces. Furthermore,
HRMS analysis of samples extracted from α-pinene-treated surface
detected only a minor quantity of residual α-pinene (C10H17m/z 137.13) in the
positive ESI mode. No α-pinene derivatives were confirmed during
the HRMS analysis. This indicates that adsorbed α-pinene does
not transform into other compounds on the hematite surface under dark
and dry conditions.In contrast to the hematite surface, adsorption
of α-pinene
on kaolinite shows significant changes in the vibrational frequencies
and intensities of absorption bands in the spectrum compared to the
gas phase (Figure B). Most notable is the disappearance of the peak at 3021 cm–1 corresponding to the olefinic C—H stretch,
indicating reactions of α-pinene (Figures A and 2B) with new
peaks appearing at 3074, unsaturated C—H stretch and 1771 and
1662 cm–1 suggesting surface oxidation and the formation
of C=C and C=O bonds.[30,46,49]
Figure 2
Kinetics
of the surface reaction of α-pinene on kaolinite.
(A) FTIR spectra at time t = 0 and at 225 s after
adsorption. (B) Changes in the normalized peak area for the olefinic
C—H stretch at 3021 cm–1 as a function of
time. (C) First-order kinetic fit (R2 =
0.98) yields a first-order rate constant of 0.021 s–1.
Kinetics
of the surface reaction of α-pinene on kaolinite.
(A) FTIR spectra at time t = 0 and at 225 s after
adsorption. (B) Changes in the normalized peak area for the olefinic
C—H stretch at 3021 cm–1 as a function of
time. (C) First-order kinetic fit (R2 =
0.98) yields a first-order rate constant of 0.021 s–1.
Kinetics of α-Pinene Oxidation on Mineral
Surfaces
To better understand the kinetics of reaction of
α-pinene on
kaolinite, the FTIR spectra collected within the initial 225 s were
analyzed. The initial spectrum of α-pinene adsorbed on kaolinite
contains peaks corresponding to the gas-phase α-pinene, indicating
initial molecular adsorption (Figure A). At t = 225 s, new peaks at 3074
and 1662 cm–1 appear whereas the peak at 3021 cm–1 corresponding to the olefinic C—H stretch
disappears. Furthermore, the peak area under the curve for the peak
at 3021 cm–1 (for kaolinite) was calculated and
normalized. The peak area was plotted against time. (Figure B). The kinetics of reaction
of α-pinene on kaolinite for the time, t, between
30 and 120 s was calculated according to a pseudo first-order rate
equation (eq ) are shown
in Figure C.The rate of loss of
adsorbed α-pinene, as shown by the disappearance of the peak
at 3021 cm–1, was monitored and gave a rate constant, k, of 0.021 s–1.
Formation of Surface-Bound
Products from the Adsorption of α-Pinene
on Kaolinite
HRMS analysis of surface-bound products extracted
from the kaolinite surfaces exposed to α-pinene shows two major
products (Figure ).
These compounds were identified as 1, C20H32O4Na (compound 1), corresponding
to a pinonaldehyde dimer (m/z =
359.22) and C10H16O2Na (compound 2) corresponding to pinonaldehyde (m/z = 191.10). Therefore, it is proposed that α-pinene
oxidizes to pinonaldehyde on the kaolinite surface and readily undergoes
dimerization. Dimerization of pinonaldehyde yields two isomers, one
via aldol condensation and the other by gem diol formation followed
by subsequent dehydration.[50] Because of
the subsequent oxygen loss observed in the HRMS pattern with C20H32O4Na, C20H32O3Na, C20H32O2Na, and
C20H32ONa, the aldol condensation product was
assigned the structure shown for compound 1 (Table ).
Figure 3
HRMS pattern of the surface-bound
products formed on kaolinite
upon exposure to α-pinene. The products were identified in positive
ESI mode (M + H)+.
Table 2
Identified Surface-Bound Products
from HRMS Analysis Coupled with MS/MSa,b
The * identifies
proposed possible
structures.
All products
were identified
in positive ESI mode [M + H]+ with mass tolerance <2
ppm.
HRMS pattern of the surface-bound
products formed on kaolinite
upon exposure to α-pinene. The products were identified in positive
ESI mode (M + H)+.The * identifies
proposed possible
structures.All products
were identified
in positive ESI mode [M + H]+ with mass tolerance <2
ppm.The formation of pinonaldehyde
from α-pinene on kaolinite
surfaces likely involves surface redox sites and surface hydroxyl
groups. A possible mechanism involves dihydroxylation of the C=C
bond from the surface redox sites and surface hydroxyl groups followed
by a glycol cleavage. A similar dihydroxylation was observed previously
for limonene, an isomer of α-pinene, on kaolinite.[51] Furthermore, in our studies, a small HRMS peak
at 193.12 corresponding to dihydroxylated α-pinene with the
formula C10H18O2Na was observed supporting
a possible surface hydroxyl groups driven oxidation pathway (Scheme ). Furthermore, as
calculated using Avogadro molecular building platform, α-pinene
molecules have an average length perpendicular to C=C of 5.974
Å. The average basal distance of kaolinite between two layers
is 7.2 Å,[52] indicating that α-pinene
molecules are small enough to interact with these interlayers, thereby
possibly increasing the reactivity whereas the crystalline surface
of hematite may facilitate only the physisorption of α-pinene.
Therefore, these findings underscore the importance of the mineral
surface specificity in forming oxidation products and warrant further
investigation.
Scheme 1
Proposed Mechanism for Oxidation of α-Pinene
to Pinonaldehyde
upon Exposure to Kaolinite Surface
This proposed mechanism
was
adapted in part from Lederer et al. (2016).[51]
Proposed Mechanism for Oxidation of α-Pinene
to Pinonaldehyde
upon Exposure to Kaolinite Surface
This proposed mechanism
was
adapted in part from Lederer et al. (2016).[51]
Reactions of Adsorbed Nitrates
and Gas-Phase
NO2 with α-Pinene
In FTIR experiments, the
mineral surfaces were first nitrated by exposure to nitric acid. For
nitrated hematite and kaolinite surfaces, several coordination modes
of adsorbed HNO3 and NO3– were
observed, as has been previously discussed in detail (see Figure S1 in the Supporting Information).[33,44,53−56] It can be seen that the nitrate
coordination differs on these surfaces. Briefly, multiple coordination
modes (monodentate, bidentate, and bridging) were observed on nitrated
hematite surface.[44,57] In contrast, however, adsorbed
nitrate on kaolinite had spectral features more closely associated
with water solvated nitrate.[44,57] Several new surface
products formed following the adsorption of α-pinene with the
nitric acid exposed hematite and kaolinite surface (Figures S2 and S3). New spectral features produced from the
reaction of α-pinene with nitrated hematite and kaolinite primarily
are observed in the spectral regions extending from 2800 to 3050 cm–1 and 1450 to 1475 cm–1.[30,49,53,58] Furthermore, strong absorptions at ∼3500 cm–1 (3507, 3472, 3355 cm–1) corresponding to the presence
of different oxygenated organic compounds and hydrogen bonding were
observed on the hematite surface. Therefore, the reactions of α-pinene
with nitrated hematite and kaolinite surfaces suggest the formation
of new surface-bound products through heterogeneous reactions in the
dark and under dry conditions.The spectral features of surfaces
after the reaction of NO2(g) and α-pinene were different
from those of from nitrated surfaces with α-pinene. In these
experiments, α-pinene and NO2(g) were introduced
to the IR cell with the mineral surface present. Then the reaction
was allowed to occur over period of 4 h. After this time period, the
infrared cell was evacuated. Unlike that seen for α-pinene alone,
there remained adsorbed products on the surface following the evacuation.
In particular, the spectrum of the hematite surface showed peaks at
2957, 2928, and 2876 cm–1 as well as at 1472 and
1440 cm–1, suggesting the presence of α-pinene
and/or its derivatives on the surface. Additionally, the appearance
of peaks at 1687 and 1215 cm–1 implies the formation
of oxygenated pinene derivatives on the surface. For the kaolinite
surface, the presence of oxygenated organic compounds was shown by
the spectral features around 3745, 3779, ∼3500, and 1660 cm–1. The FTIR spectra of surface-bound products are provided
in the Supporting Information (Figures S2 and S3). Furthermore, the products formed on these surfaces were
analyzed with HRMS. In particular, the analysis of solvent extracted
surface products provided was used to identify specific products.To identify these surface-bound products formed and to understand
possible reaction pathways, these products formed on mineral surfaces
were extracted and analyzed via high-resolution mass spectrometry
(HRMS). The proposed structures were further confirmed using tandem
mass spectrometry (MS/MS). From the reaction of α-pinene with
adsorbed nitrates on hematite, HRMS analysis suggested the formation
of a series of ON compounds (Figure A, Table ). Compound 3 (C10H16NO3, m/z = 198.15) and compound 4 (C10H18NO3, m/z = 200.17) were identified as the major ON compounds
formed. The MS/MS analysis of compounds 3 and 4 confirmed the formation of fragments C10H16ON (m/z = 166.12) and C10H18ON (m/z = 168.14),
respectively, among others. Compound 5 (C10H18NO4, m/z = 216.23) was identified with low relative intensity and with fragments m/z = 184.13 and 109.10 corresponding to
C10H18O2N and C8H13 respectively. A smaller peak observed for compound 6 (C10H17O, m/z = 153.15) was assigned to α-pinene oxide. A possibility
for compound 6 is isopinocamphone. However, the absence
of a carbonyl peak ∼1740 cm–1 in the FTIR
spectrum (Figure S2) eliminated this possibility.
Furthermore, the formation of the fragment C8H13 (m/z = 109.10) from α-pinene
was confirmed through MS/MS analysis. Additionally, a smaller peak
at C10H17 (m/z = 137.13) was identified as a fragment of α-pinene oxide,
which also can be assigned to unreacted α-pinene that may remain
on the surface.
Figure 4
HRMS patterns in positive ESI mode, [M + H]+, of surface-bound
products from the reactions of α-pinene with adsorbed nitrate
(A) on hematite (B) on kaolinite.
HRMS patterns in positive ESI mode, [M + H]+, of surface-bound
products from the reactions of α-pinene with adsorbed nitrate
(A) on hematite (B) on kaolinite.Scheme illustrates
the proposed formation pathways for the identified ON compounds. The
formation of α-pinene oxide suggests that adsorbed nitrate on
hematite oxidizes α-pinene. C10H17O (m/z = 153.15) was not observed in experiments
with α-pinene adsorption on hematite, suggesting the epoxidation
of α-pinene likely involves adsorbed nitrate, and perhaps iron-nitrate-redox
chemistry.[59] This is then followed by a
nucleophilic attack from adsorbed nitrate on either carbon attached
to the O atom on α-pinene oxide (electrophile) can form the
two isomers of compound 5 (Scheme A). Compound 3 may form by dehydration
of compound 5(b). The formation of Compound 4 may occur via the interaction of π bonds in α-pinene
with the oxygen atoms in adsorbed nitrates leading to the formation
of a carbocation intermediate. The reaction of α-pinene π
bonds with acid groups leading to carbocation formation was previously
proposed.[51,60−62] The formed carbocation
then can react with adsorbed nitrate species on the hematite surface
to form the isomers of compounds 4(a) or 4(b), respectively (Scheme B). Though the
formation of both isomers is possible, a higher yield of the compound 4(b) is expected because of the higher stability
of tertiary carbocations.
Scheme 2
Proposed Mechanisms for the Formation of
(A) Compounds 3 and 5 from Adsorbed Nitrate
and α-Pinene Oxide,
Formed from the Interaction of α-Pinene with Adsorbed Nitrates
on Hematite and (B) Compound 4 from Adsorbed Nitrate
and α-Pinene
The structures 4(a) and 4(b) are structural
isomers of compound 4. Similarly, the structures 5(a) and 5(b) are structural isomers of compound 5.
Proposed Mechanisms for the Formation of
(A) Compounds 3 and 5 from Adsorbed Nitrate
and α-Pinene Oxide,
Formed from the Interaction of α-Pinene with Adsorbed Nitrates
on Hematite and (B) Compound 4 from Adsorbed Nitrate
and α-Pinene
The structures 4(a) and 4(b) are structural
isomers of compound 4. Similarly, the structures 5(a) and 5(b) are structural isomers of compound 5.In contrast to lower molar mass ON compounds observed
in the reactions
of α-pinene with nitrated hematite, high molecular weight ON
compounds were observed with nitrated kaolinite surfaces (Figure B). Among these were
compound 7 (C20H31NO7, m/z = 398.34) and compound 8 (C30H48NO9, m/z = 566.43). These compounds were further confirmed
by analyzing their MS/MS fragmentation patterns. For instance, the
fragment C30H48ON (m/z = 438.37) was produced by the parent peak corresponding
to compound 8 at 566.43. Apart from these ON compounds,
the oxidation products, compound 2 (pinonaldehyde) and
compound 9 (C10H16O3, m/z = 207.22, pinnonic acid),
were observed in minor quantities.Pinonaldehyde and its dimer
were produced from α-pinene upon
exposure to kaolinite surfaces. Therefore, a similar product formation
can be expected from α-pinene reactions on nitrated kaolinite
surfaces. The detection of compound 9 (pinonic acid)
suggests at least some of the produced pinonaldehyde further oxidizes
on the nitrated kaolinite surface. Therefore, it can be speculated
that the aldehyde end of the produced pinonaldehyde, pinonaldehyde
dimer, and pinonaldehyde trimer oxidizes to form a carboxylic acid
end in the presence of nitrated kaolinite surfaces. The adsorbed nitrate
groups react with this carboxylic acid end to form PAN-analogues (peroxy
acyl nitrate) of these compounds, resulting the observed compounds 7 and 8 in this study. Furthermore, this observation
suggests more α-pinene was oxidized to form pinonaldehyde in
the presence of adsorbed nitrate on kaolinite, thereby facilitating
its trimerization. Thus, again, these results underline the important
role of the surface chemical reactions in formation of ON from α-pinene
in the dark and low RH environments.In gas-phase reactions
of α-pinene with NO2(g)
in the presence of hematite, the formation of a mixture of oxidation
products of α-pinene was observed (Figure A, Table ). Here, compounds 2 (pinonaldehyde) and 10 (C10H19O, m/z = 155.14, isopinocampheol) and m/z = 153.15 corresponding to C10H17O were identified. However, no evidence of ON formation was seen.
Therefore, the major identified SOA products from the reaction of
gas-phase NO2 with α-pinene in the presence of hematite
are oxidation products of α-pinene. In contrast to reactions
of adsorbed nitrate on hematite with α-pinene, here the peak
at m/z = 153.15 corresponding to
C10H17O is most likely several species of similar
mass including α-pinene oxide, isopinocamphone, and α-pinene
derived alcohols such as verbenol.[63−65] This is because of the
complex mass fragmentation pattern produced with a strong fragment
at C7H9O (m/z = 109.06), along with C10H17 (m/z 137.13) with many smaller peaks for fragments
below m/z = 100, all corresponding
to possible isomers of C10H17O.
Figure 5
HRMS patterns in positive
ESI mode, [M + H]+, of surface-bound
products from the reactions of α-pinene with NO2 (A)
on hematite (B) on kaolinite.
HRMS patterns in positive
ESI mode, [M + H]+, of surface-bound
products from the reactions of α-pinene with NO2 (A)
on hematite (B) on kaolinite.In contrast to the variety of surface-bound oxidation products
seen from the hematite surface, for kaolinite only compound 1 (pinonaldehyde dimer), compound 2 (pinonaldehyde),
and compound 11 (C30H48O6, m/z = 527.33, pinonaldehyde trimer)
were observed. Therefore, it is likely that the presence of NO2 in the gas phase causes further oxidation of α-pinene
to pinonaldehyde, which then leads to trimerization. Similar evidence
of oligomerization of pinonaldehyde has been observed previously.[66]The oxidation products of α-pinene
are widely seen in the
atmosphere and have shown to readily form in the presence of typical
atmospheric oxidants.[4,6,25,26] However, their formation in the presence
of gas-phase NO2 and in the absence of other oxidizers
such as O3 and HO• was not previously
observed. Therefore, it is proposed that the α-pinene oxidation
by NO2 can occur in the presence of mineral surfaces, thereby
producing the observed oxidation products of α-pinene. The formed
oxidation products adsorb onto the mineral surface, thus increasing
SOA formation. Additionally, these findings underscore the importance
of particle mineralogy and the role that different mineral dust surfaces
play in this chemistry.
Conclusions and Atmospheric
Implications
The reactions of α-pinene on hematite
and kaolinite surfaces
at 296 K yield various ON and oxidation products of α-pinene.
In particular, α-pinene oxidized to pinonaldehyde followed by
dimerization on the kaolinite surface, whereas no such product formed
on hematite. When NO2 is present or the surface is first
exposed to HNO3, multiple ON products, as well as oxidation
products, were identified with both mineral surfaces. The proposed
formation pathways of these ON compounds occur via an adsorbed nitrate
species. In the case of kaolinite, the adsorbed nitrates react with
pinonaldehyde and its oligomers to produce a carboxylic acid end which
then reacts with adsorbed nitrates to form PAN-type compounds. In
contrast to kaolinite, adsorbed nitrate reacts with α-pinene
on hematite to yield lower molar mass ON compounds. Furthermore, the
fate of gas-phase reaction products of the gas-phase α-pinene
plus NO2 reaction was also investigated. These gas-phase
reactions lead to several oxidized α-pinene derivatives that
form strong interactions with these mineral surfaces. Eleven unique
surface-bound products were identified.Overall, this study
shows how mineral dust aerosol not only has
the potential to act as seed particles for SOA formation by facilitating
adsorption of lower volatility oxidized organic compounds but also
can provide a reactive surface for reactions to occur that is mineral
specific, leading to the formation of ON and other SOA components.
Additionally, these findings suggest further investigation of the
heterogeneous and multiphase mineral oxide-mediated chemistry of these
abundant monoterpenes under various environmental conditions, such
as high RH environments, daytime with abundant sunlight, and other
mineral surfaces (e.g., other metal oxides, other clay types). These
types of studies will provide the basis for further understanding
of the role of mineral dust aerosol on air quality, climate change,
and human health.
Authors: F D Lopez-Hilfiker; C Mohr; E L D'Ambro; A Lutz; T P Riedel; C J Gaston; S Iyer; Z Zhang; A Gold; J D Surratt; B H Lee; T Kurten; W W Hu; J Jimenez; M Hallquist; J A Thornton Journal: Environ Sci Technol Date: 2016-02-09 Impact factor: 9.028
Authors: Mikael Ehn; Joel A Thornton; Einhard Kleist; Mikko Sipilä; Heikki Junninen; Iida Pullinen; Monika Springer; Florian Rubach; Ralf Tillmann; Ben Lee; Felipe Lopez-Hilfiker; Stefanie Andres; Ismail-Hakki Acir; Matti Rissanen; Tuija Jokinen; Siegfried Schobesberger; Juha Kangasluoma; Jenni Kontkanen; Tuomo Nieminen; Theo Kurtén; Lasse B Nielsen; Solvejg Jørgensen; Henrik G Kjaergaard; Manjula Canagaratna; Miikka Dal Maso; Torsten Berndt; Tuukka Petäjä; Andreas Wahner; Veli-Matti Kerminen; Markku Kulmala; Douglas R Worsnop; Jürgen Wildt; Thomas F Mentel Journal: Nature Date: 2014-02-27 Impact factor: 49.962
Authors: Véronique Perraud; Emily A Bruns; Michael J Ezell; Stanley N Johnson; John Greaves; Barbara J Finlayson-Pitts Journal: Environ Sci Technol Date: 2010-08-01 Impact factor: 9.028
Authors: Jason D Surratt; Jesse H Kroll; Tadeusz E Kleindienst; Edward O Edney; Magda Claeys; Armin Sorooshian; Nga L Ng; John H Offenberg; Michael Lewandowski; Mohammed Jaoui; Richard C Flagan; John H Seinfeld Journal: Environ Sci Technol Date: 2007-01-15 Impact factor: 9.028
Authors: G Duporté; P-M Flaud; E Geneste; S Augagneur; E Pangui; H Lamkaddam; A Gratien; J-F Doussin; H Budzinski; E Villenave; E Perraudin Journal: J Phys Chem A Date: 2016-09-29 Impact factor: 2.781
Authors: Madeline R Lederer; Allison R Staniec; Zoe L Coates Fuentes; Daryl A Van Ry; Ryan Z Hinrichs Journal: J Phys Chem A Date: 2016-11-21 Impact factor: 2.781