Halogen chemistry in the lower troposphere - Journal de Physique IV

December 9, 2004

Time:

03:46pm

jp4121015.tex

J. Phys. IV France 121 (2004) 223–238 C EDP Sciences, Les Ulis DOI: 10.1051/jp4:2004121015

Recent applications of Differential Optical Absorption Spectroscopy: Halogen chemistry in the lower troposphere A. Saiz-Lopez and J.M.C. Plane School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK e-mail: [email protected], [email protected]

Abstract. This chapter provides a comprehensive review of the atmospheric chemistry of halogens in the lower troposphere, including a discussion of the important ways in which halogens affect this region of the atmosphere. It then describes the recent progress made in observing these species by the Differential Optical Absorption Spectroscopy (DOAS) technique. A brief description of the technique and its capabilities is also provided.

1. INTRODUCTION TO DIFFERENTIAL OPTICAL ABSORPTION SPECTROSCOPY (DOAS) DOAS is a powerful technique for measuring the concentrations at ultra-low levels of a wide variety of trace species in the atmosphere. In this chapter we will not provide a detailed discussion of the technique. The interested reader is directed to a chapter in an earlier volume of the ERCA book series [1], as well as three major reviews on DOAS that have been published in the last decade [1-4]. The DOAS technique operates by measuring the optical absorption of molecules in the ultra violet (UV) - visible - near infra red (IR) part of the electromagnetic spectrum. The light source can be artificial (e.g., a broadband Xe lamp, a halogen lamp, or a laser), or natural (direct of scattered sunlight, the moon or stars). Each molecule is identified from the unique vibrational structure of an absorption band, usually corresponding to an electronic transition of the molecule. This provides a high degree of specificity when a number of molecules absorb in the same spectral region. In addition, if a molecule has a strong absorption band, then detection limits of below 1 part per trillion (ppt, i.e. 1 part in 1012 ) are possible. Optical absorption in the atmosphere also occurs through elastic and inelastic scattering of photons by air molecules (termed Rayleigh and Raman scattering, respectively), as well as scattering by aerosol particles (Mie scattering). In addition, many atmospheric species (e.g., O3 , NO2 ) are present continuously in the atmosphere at observable concentrations. It is therefore not possible when doing atmospheric spectroscopy to measure an un-attenuated background spectrum of the light source, as one could in the laboratory. Instead, atmospheric spectra are converted into differential optical absorption spectra, and the concentrations of the individual molecular absorbers are obtained by fitting reference absorption cross sections. When working in the boundary layer with an artificial light source, atmospheric spectra are normally recorded over a pathlength of several kilometres, to enhance the optical absorption and hence the detection limits of atmospheric species. However, if the pathlength is greater than about 10 km, light losses due to beam divergence and atmospheric extinction reduce the signal quality to the point where greater pathlength no longer offers an advantage. The optimal pathlength is around 8 km. In order to avoid averaging atmospheric concentrations over too great a distance, and also for ease of installation, the light path is often folded using a retro-reflector array, so that the transmitter and receiver are co-located. When working with scattered sunlight as the light source, the slant column abundances of the molecules are actually retrieved. These are then converted into vertical column abundances using a

December 9, 2004

Time:

03:46pm

jp4121015.tex

224

JOURNAL DE PHYSIQUE IV

simple geometrical correction for low solar zenith angles (typically < 70◦ ), or a radiative transfer model for higher zenith angles. In the last few years the multiple-axis (MAX-DOAS) technique has been developed, where scattered sunlight spectra are recorded simultaneously over a range of azimuthal angles. This provides information about the location of the absorbing species: in the boundary layer, the free troposphere, or the stratosphere. The DOAS technique was developed in the 1970s by Platt and Perner [5, 6] and Noxon [7, 8]. Since then, the observation of many chemical species of importance in atmospheric chemistry has been pioneered by DOAS, including the key halogen species ClO, OClO, BrO, IO, OIO and I2 .

2. HALOGEN CHEMISTRY IN THE LOWER TROPOSPHERE Species containing halogens (i.e. Cl, Br or I) play a key role in a wide variety of atmospheric processes. For instance, catalytic cycles involving halogen radicals cause significant tropospheric O3 depletion, in certain environments leading to complete O3 removal. Halogen radicals (e.g., Cl, BrO and IO) affect the partitioning between the forms of nitrogen oxides (NO and NO2 ) and hydrogen oxides (OH and HO2 ), and can also be reactive towards organic species, thus influencing the oxidizing capacity of the lower atmosphere. Additionally, elevated levels of iodine oxides form new atmospheric particles in coastal environments.

2.1. Sources of halogens in the lower atmosphere There are two main sources of halogens in the troposphere: first, emissions of halogenated organic molecules (halocarbons) from a variety of terrestrial and marine sources; and second, the release of inorganic halogen compounds from aqueous sea-salt aerosols and, to a lesser extent and more sporadically, from volcanoes. Some halocarbons have a relatively short lifetime with respect to solar photolysis in the lower atmosphere, providing a source of reactive halogen atoms in the troposphere. By contrast, longerlived halocarbons (e.g. CH3 Cl) can be transported, principally by cloud-pumping through the tropical tropopause, into the stratosphere where they impact on stratospheric O3 .

2.1.1. Halocarbon emissions Halocarbons are emitted from a range of natural sources, as well as from anthropogenic sources such as biomass burning and industrial emissions [9]. Table 1 compares the annual emission rates, atmospheric mixing ratios and lifetimes of a number of important halocarbons. The main natural sources of halogenated hydrocarbons are macroalgae and phytoplankton in the oceans. In the last two decades, measurements of a wide range of alkyl chlorides and bromides such as CHBr3 , CH3 Br, CH3 Cl, CHCl3 , CHBr2 Cl, CHBrCl2 , CH2 BrCl and CH2 Br2 have been made at a number of oceanic locations [10-21]. Emission of CHBr3 , which is the most common form of natural reactive organic bromine [22], has even been observed in the springtime Arctic boundary layer [23]. CHCl3 and CH3 Cl, which are the most abundant natural chlorinated compounds, have also been detected in polar regions [24]. Thus, the emission of alkyl bromides and chlorides is a widespread natural phenomenon, making a substantial contribution (5 - 8 Tg y−1 ) to the global atmospheric budget, similar in magnitude to emissions from industrial sources [9]. Volatile biogenic iodocarbons of marine origin are the major source of atmospheric iodine. Evidence for the biological production of alkyl iodides from macro- and microalgae was suggested in the 1970s [25]. Since then, observations of iodocarbons such as CH3 I, CH2 I2 , CH2 ICl, CH2 IBr, C2 H5 I, C3 H7 I, CH3 CHICH3 have been made in the atmosphere [19, 20, 25-27]. These species have been found to correlate with tidal height and solar irradiance in coastal environments [19, 28].

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

225

Alkyl halides photodissociate in the lower troposphere to generate halogen atoms, e.g.: CH3 Br + hν → CH3 + Br

(1)

CH2 I2 + hν →→ CH2 + 2I

(2)

CH2 IBr + hν →→ CH2 + I + Br

(3)

The absorption spectra of iodocarbons tend to be shifted to longer wavelengths than their chlorineand bromine-containing analogues, and hence the photodissociation of iodocarbons is much faster, as shown in Table 1 [29]. The main contributor to the flux of atomic I in the atmosphere appears to be the photodissociation of CH2 I2 , which has a photolytic lifetime of about 5 min [19]. Table 1. Comparison of the typical mixing ratios, release rates and atmospheric lifetimes of some important halocarbons. Halocarbon

Mixing ratios / ppt

Emission rate / 106 kg yr−1

Atmospheric lifetime

CH3 Cl

520 – 650a

∼ 3700a

∼ 1 yrb

CH3 Br

10 – 20c

∼ 206d

∼ 0.7 yrd

CH3 I

0.5 – 5c,e

1000 – 2000f

3 - 5 dayc

CH2 Cl2

20 – 40g

160-500g

0.2 - 0.4 yrg

CH2 Br2

0.5 – 4h

12 – 20h

0.2 – 0.5 yrh

CH2 I2

< 2c,e

1000 – 2000i

∼ 5 minc

CHCl3

10 – 40j

350 – 600j

∼ 0.5 yrj

CHBr3

0.5 – 40c,k

∼ 220h

∼ 0.06 yrh

CH2 ICl

< 1c

∼ 100l

Several hrc

CH2 IBr

< 1c

∼ 100l

< 1 hrc

a

Khalil and Rasmunssen [17]; bYvon and Butler [30]; c Carpenter et al. [19] and references therein; d Wayne [9]; Yokouchi et al. [27]; f Davis et al. [31]; g Khalil et al. [32]; h Carpenter and Liss [22]; i Klick and Abrahamsson [33]; j Khalil and Rasmussen [18]; k Reinfenhäuser and Heumann [34]; l estimated from Southern Ocean surface seawater concentration measurements, Carpenter et al. [35]. e

Recently, the detection of I2 in the coastal marine boundary layer (MBL) indicates that I2 is a major source of reactive iodine, at least along coastlines rich in macroalgae (seaweed and kelp) [36]. A mechanism has been proposed for the release of I2 from macroalgae [37], involving the equilibrium between HOI, I− and I2 in aqueous solution [38]: HOI + I− + H+ ↔ I2 + H2 O

(4)

The evaporation of I2 from the ocean surface by means of oxidation of I− ions in seawater under solar irradiation has also been suggested. Based on experimental results, the total emission rate of I2 from the earth’s ocean surface was estimated to be 4 × 1011 g yr−1 [39]. This may be compared with a model calculation of the sea-air flux of I2 , which yielded 2.4 × 1011 g yr−1 [40]. Another study proposed the liberation of I2 from seawater by the interaction of I− with atmospheric O3 [41].

December 9, 2004

226

Time:

03:46pm

jp4121015.tex


Figure 1. Schematic drawing of the known halogen chemistry in the marine boundary layer, showing how gas-phase species are coupled through heterogeneous reactions to halide ions in sea-salt aerosol.

2.1.2. Halogen release mechanisms from sea salt Another source of halogens in marine environments is the release of inorganic halogen species from sea-salt aerosols. This occurs through heterogeneous reactions involving nitrogen oxides, hypohalous acids HOX (X= Cl, Br or I) and strong acids. In addition, a recent laboratory study indicates that the uptake of O3 on sea salt may contribute to the release of Br and Cl atoms [42]. Figure 1 is a schematic diagram of tropospheric halogen chemistry, illustrating the central role of cycling of halogen species through sea-salt aerosol. Reactions of NO2 and N2 O5 with NaX (X= Cl or Br) release halogenated nitrogen oxides that photolyse readily or react further with sea-salt aerosol [43-48]: N2 O5 + (NaX)aerosol → XNO2 + (NaNO3 )aerosol

(5)

2NO2 + (NaX)aerosol → XNO + (NaNO3 )aerosol

(6)

XNO2 + hν → X + NO2

(7)

XNO2 + (NaY)aerosol → XY + (NaNO2 )aerosol

(8)

XY + hν → X + Y

(9)

The halogen activation channel involving N2 O5 , which is formed at night by the recombination of NO3 and NO2 , is important in the remote marine environment because reaction 5 is relatively efficient [49, 50]. In contrast, the uptake of nitrogen oxides onto aerosol surfaces is favoured in semi-polluted environments.

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

227

Another night-time source of halogenated nitrogen oxides is the uptake of NO3 [51]: NO3 + (NaX)aerosol → X + (NaNO3 )aerosol

(10)

The action of strong acids, such as H2 SO4 and HNO3 , on aerosol surfaces has also been proposed to release hydrogen halides (HX), e.g. [52]: H2 SO4 + 2(NaX)aerosol → 2HX + (NaSO4 )aerosol

(11)

HNO3 + (NaX)aerosol → HX + (NaNO3 )aerosol

(12)

Another source of halogen activation, of particular importance in remote marine locations, is the conversion of Cl− and Br− ions to halogen and inter-halogen diatomics (e.g. Br2 , IBr, BrCl and ICl) via reactions with species such as HOX and XNO3 (X = Cl, Br or I) [47, 53-59]. The reactions take place by the uptake of gaseous HOX onto acidic salt surfaces, followed by the release of the corresponding dihalogen, e.g.: HOX + Y− + H+ → XY + H2 O

(13)

For the case of the uptake of HOBr onto sea-salt aerosols, laboratory experiments show that the reaction is relatively fast and that Br2 is the predominant species liberated into the gas phase, rather than BrCl [60, 61]. The uptake of HOI and HOBr onto frozen and dry NaCl/NaBr surfaces has also been investigated in the laboratory [62, 63]. The relative percentage of dihalogens released was found to depend strongly on the Cl− /Br− ratio in the solution. In agreement with previous work, Br2 was the major product of the HOBr uptake on salt surfaces of similar composition to sea salt [63]. On the other hand, ICl is released from the uptake of HOI on dry frozen salt surfaces [62], representing a source of reactive I and Cl atoms in the troposphere [47]. Dry sea-salt aerosol contains by weight 2 × 10−5 % I− , compared with 0.19% Br− and 55.5% Cl− [9]. Therefore, the uptake of iodine-containing species such as HOI, formed initially from iodocarbon photolysis (Section 2.1.1), is crucial to the activation of chlorine and bromine from sea-salt aerosol in remote marine environments [47, 64]. Indeed, severe depletions of Br− ions in sea-salt aerosols are observed in locations such as the Southern Ocean [65, 66]. Br− deficits are typically greater than 40% (occasionally over 90%), compared with only about 5% for Cl− . The bromine release exhibits a strong seasonal correlation, peaking in summer when the aerosol is most acidic as a result of the gas-phase oxidation of dimethyl sulphide (DMS), leading to sulphuric acid which is absorbed into the aerosol [65]. A remarkable process involving heterogeneous chemistry occurs in the polar regions, and has been termed the “bromine explosion” [67, 68]. It occurs via the aqueous-phase reaction 13 (X and Y = Br), on sea ice that has been covered by brine. The Br2 that is formed is emitted to the gas phase, where it photolyses to produce 2 Br atoms. These react first with O3 to form 2 BrO, and then with HO2 to form 2 HOBr. The 2 HOBr then cycle through the brine-covered ice to form 2 Br2 , leading to 4 Br atoms and hence an exponential growth in the BrO concentration. Finally, HX species that form in the gas phase (see Section 2.2) are readily taken up on sea-salt aerosol [47, 60, 64, 69-71]. The resulting X− ions will then likely be recycled back to the gas-phase, so this does not represent overall an activation process of the halogen. 2.2. Impacts of halogen species on tropospheric chemistry Figure 1 shows that a large variety of halogen species, such as X, X2 , XY, XO, OXO, HOX, XNO2 and XONO2 (where X and Y = Cl, Br, I), occur in the troposphere. In this Section we consider the variety of ways in which halogens impact on the chemistry of the troposphere.

December 9, 2004

Time:

03:46pm

228

jp4121015.tex


2.2.1. Ozone depletion The role of halogen radicals in stratospheric ozone depletion is well documented [72-76], and will not be discussed further here. The halogen-catalysed depletion of O3 also occurs in the lower troposphere, often by reactions between halogen oxide radicals: X + O3 → XO + O2

(14)

Y + O3 → YO + O2

(15)

XO + YO → X + Y + O2

(16a)

net: 2O3 → 3O2 However, reaction 16 can have other product channels: XO + YO → OXO + Y → XY + O2 (+M) → XOYO

(16b) (16c) (16d)

Hence, the efficiency of O3 destruction by reaction 16a will vary depending on the halogen species X and Y involved. Note that channel 16c is really equivalent to 16a, because the XY product will photolyse very rapidly. In the ClO self-reaction (i.e., X = Y = Cl), the Cl2 O2 dimer is obtained. However, this is not stable in the troposphere and thermally decomposes back to ClO + ClO [67]. For the case of the BrO self-reaction the production of 2 Br atoms is favoured over formation of Br2 , whereas the BrO + IO cross-reaction produces OIO and Br (80 %) and I + Br (20%) [29]. The IO self-reaction is fast [77], and produces mainly OIO and I2 O2 at atmospheric pressure. The IO dimer is probably either photolysed, undergoes uptake onto existing aerosol surfaces, or is involved in the nucleation of new particles (see below) [59, 78, 79]. The efficiency of the O3 depletion cycle can be reduced by photolysis of the halogen dioxide formed in reaction 16b: OXO + hν → XO + O

(17)

O(3 P) + O2 + M → O3 + M

(18)

However, in the case of OIO, there is evidence that photodissociation also yields I + O2 [80], thereby increasing the overall O3 -depleting efficiency of the IO self-reaction. Note that OXO can also react with NO, with rate coefficients that increase down the group (I > Br > Cl): OXO + NO → XO + NO2

(19)

Reaction 16 will dominate O3 destruction when relatively high mixing ratios of halogen oxides are present (typically, more than 2 ppt of IO and BrO, for instance [81]). At lower halogen oxide concentrations, reaction with HO2 radicals becomes important [82]: XO + HO2 → HOX + O2

(20)

HOX + hν → X + OH

(21)

OH + O3 → HO2 + O2

(22)

X + O3 → XO + O2

(14)

net: 2O3 → 3O2 This cycle has been proposed to account for substantial O3 depletion events that are observed in remote marine environments at sunrise, as a result of the photolysis of halogen species that build up overnight from halogen activation on sea-salt aerosol [83, 84].

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

229

The “bromine explosion” that occurs in the polar boundary layer during spring (Section 2.1.2) produces BrO mixing ratios of 10 – 30 ppt. This quickly leads to the almost total removal of O3 over the ice, for episodes lasting up to several days [55, 85-92]. In semi-polluted environments, halogen nitrate formation can also lead to O3 depletion: XO + NO2 (+M) → XNO3

(23)

XNO3 + hν → X + NO3

(24)

NO3 + hν → NO + O2

(25)

X + O3 → XO + O2

(14)

NO + O3 → NO2 + O2

(26)

net: 2O3 → 3O2 However, the efficiency of this cycle is reduced because XNO3 can also photolyse to XO + NO2 [29], leading to a null cycle (although recent experiments in our laboratory show this does not happen in the case of INO3 ). Furthermore, the major photolysis pathway of NO3 actually produces NO2 + O [29], again leading to no overall O3 depletion. Note that the XNO3 formed in reaction 23 can also be removed from the gas phase by recycling through sea-salt aerosol, which provides an efficient route for converting NOx to NO− 3 ions in the aerosol. 2.2.2. Oxidation of organic compounds During the daytime in the lower troposphere, a photochemical steady-state is established between X and XO, where X = Br and I: X + O3 → XO + O2

(14)

XO + hν → X + O

(27)

The ratio of XO to X ranges from about 6 for X = I to 20 for X = Br in the mid-latitude MBL at midday. In the case of X = Cl, the photolysis rate of ClO is too slow for a steady-state to be established, and the typical daytime concentration of Cl is probably much less than 103 cm−3 . The most abundant chlorine reservoirs are HCl and ClNO3 , with concentrations around 108 cm−3 (5ppt) [47]. However, the halogen diatomics such as ICl and IBr absorb light throughout the visible, and thus photolyse very rapidly even at high solar zenith angles. If these inter-halogen compounds build up during the night as a result of heterogeneous chemistry on sea-salt aerosol (Section 2.1.2), then their photolysis will provide a significant source of Cl (and Br) at sunrise. Atomic Cl has a similar reactivity towards many non-methane hydrocarbons (RH) as the OH radical, leading to hydrogen abstraction: Cl + RH → HCl + R

(28)

The hydrogen halides also form through reaction with HO2 radicals: X + HO2 → HX + O2

(29)

The hydrogen halide formed is highly soluble in water and therefore will be irreversibly removed from the gas phase by wet/dry deposition. The only significant gas-phase reaction that HX undergo is with OH, and this increases going down the group (Cl < Br < I): HX + OH → X + H2 O

(30)

The halogen oxide radicals BrO and IO can significantly increase the daytime oxidation of DMS in the MBL [81, 93-96]: CH3 SCH3 + XO → CH3 S(O)CH3 + X

(31)

December 9, 2004

230

Time:

03:46pm

jp4121015.tex


At the levels of XO that have been observed in the MBL, the rate of oxidation of DMS by IO and BrO is about an order of magnitude faster than by OH (although there is uncertainty about the rate coefficient for the IO + DMS reaction) [81]. It is thought that CH3 S(O)CH3 (dimethyl sulphoxide) is then oxidised to methane sulphinic acid in the MBL [97]. This product is not a condensable vapour, unlike the sulphuric acid formed when DMS is oxidised by OH. Hence, oxidation of DMS by XO reduces the effectiveness of DMS emissions for producing cloud condensation nuclei and hence influencing the radiation budget and climate [98]. 2.2.3. Changes to the NOx and HOx balance The ratio of NO2 to NO is controlled principally by the reactions: NO2 + hν( < 400 nm) → NO + O(3 P)

(19)

NO + O3 → NO2 + O2

(26)

In the presence of significant halogen concentrations, the balance will be shifted towards NO2 [59]: XO + NO → X + NO2

(32)

This reaction therefore plays a similar role to peroxy radicals such as HO2 and CH3 O2 . Note, however, that reaction 32 would be followed by reaction 14, thereby consuming an O3 , and the NO2 would photolyse to produce an O3 , so the overall effect is a null cycle in terms of O3 change. In the remote MBL, the HO2 radical is formed from OH through the sequence: OH + CO → H + CO2 (+O2 ) → HO2

(33)

In the presence of BrO and IO, the HO2 /OH ratio will be reduced [19, 31, 59, 99]: XO + HO2 → HOX + O2

(20)

HOX + hν → OH + X

(21)

This cycle is particularly important in the case of iodine, because the IO + HO2 reaction is very fast, and HOI photolyses more readily than BrO [29]. 2.2.4. Removal of elemental mercury from the atmosphere Elemental mercury (Hg0 ) is emitted to the atmosphere from a variety of sources, principally coal combustion. The element has an atmospheric lifetime of about 1 year, except during the polar spring when Hg0 is converted to reactive gaseous mercury (RGM) on a timescale of only a few hours. The Hg in RGM is in the +2 oxidation state, probably as the mercuric halides (HgBr2 /HgCl2 ) and possibly as mercuric oxide (HgO). The nature of this conversion process is currently an important research area, because the soluble RGM quickly enters the foodchain in the Arctic ecosystem [100-102]. While the mechanism for Hg0 oxidation is not completely understood, it has been observed recently that Hg0 depletion events correlate with the polar O3 depletion events described in Section 2.2.1, and almost certainly involve the elevated levels of bromine generated during the “bromine explosion” (Section 2.1.2). Possible oxidation pathways are [100-102]: Br + O3 → BrO + O2

(14)

BrO + Hg → HgO + Br

(34)

Hg0 + Br(+M) ↔ HgBr

(35)

HgBr + Br(+M) → HgBr 2

(36)

0

and

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

231

2.2.5. New particle formation It has recently been proposed that the bursts of large concentrations of ultra-fine particles (diameter ≈3 nm), that are observed during daytime around low tide in the coastal MBL, are caused by the polymerization of iodine oxides produced from photolabile compounds such as CH2 I2 and I2 [36, 37, 78, 79, 103]. At the levels of I2 encountered in the coastal MBL at low tide [36], I2 would be the major contributor to the I atom flux and thus the major source of condensable iodine oxide vapour in coastal environments [37]. Although the formation of iodine oxide particles is easy to reproduce in the laboratory [36, 37, 78, 79, 103], the mechanisms whereby species such as IO, OIO and I2 O2 are oxidised to the higher oxides, such as I2 O3 , I2 O4 and I2 O5 which are likely to be highly condensable because of their large dipole moments, are not well understood. 3. DOAS OBSERVATIONS OF HALOGEN SPECIES IN THE LOWER TROPOSPHERE In the last decade, a number of reactive halogens species have been observed for the first time in the lower troposphere, all of them by DOAS. The molecules ClO, BrO, IO, OIO and I2 have been measured in a number of different locations, making use of their characteristic optical absorption features in the UV/visible spectral range. 3.1. Observations of BrO and ClO BrO has been observed by DOAS boundary layer instruments in the polar boundary layer [88, 90-92, 104], in the mid-latitude MBL [81, 105], and over salt lakes where the high salt concentrations can generate a “bromine explosion” similar to that over sea-ice in the polar spring (Section 1.2) [106-108]. DOAS measurements of BrO made over the Dead Sea have revealed mixing ratios up to 176 ppt, which produce O3 destruction rates of 10 - 20 ppb h−1 . This would quickly reduce the O3 concentration to zero if there was no mixing to dilute the bromine-rich air and entrain fresh O3 [106, 109]. Another mid-latitude location where BrO has been observed is over the Great Salt Lake, Utah. Figure 2 illustrates

Figure 2. Chlorine and bromine oxide diurnal profiles at the Great Salt Lake, Utah. O3 and NO2 are also shown; note the anti-correlation with the halogen oxides. The thin solid lines represent the detection limits of the instrument for the different molecules. Reproduced with permission of the American Geophysical Union from Stutz et al. [107].

December 9, 2004

232

Time:

03:46pm

jp4121015.tex


Figure 3. Observations of BrO at the mid-latitude coastal location of Mace Head (Ireland). Empty squares and black dots correspond to BrO measurements under wind directions from mainland and the Atlantic Ocean, respectively. Daytime and night-time periods are represented by white and hatched background, respectively. Reproduced with permission of the American Geophysical Union from Saiz-Lopez et al. [81].

the measurements of BrO and ClO at this location, with values of BrO and ClO up to 6 and 15 ppt, respectively, that correlated with moderated low O3 episodes. These were the first direct observations of ClO in the mid-latitude boundary layer [107]. In the mid-latitude MBL, Multi-Axis DOAS measurements of BrO, made by observing scattered sunlight, indicate mixing ratios up to about 2 ppt [105]. These values could be indicative of background levels of the radical in the oceanic mid-latitude and tropical boundary layer. BrO has also been observed directly in the coastal MBL at Mace Head, Ireland, with an average mixing ratio of about 2 ppt and a maximum of 6 ppt [81]. Figure 3 shows the time series of BrO mixing ratios during a 6 day period of observations. It can be seen that levels above the BrO detection limit (≈ 1 ppt) are only observed when the wind is from a westerly direction (i.e., from the Atlantic ocean). The profiles are characterised by a rapid pulse after sunrise, which is likely to be caused by the buildup during the night of photolabile precursor that rapidly photodissociate in the first hour after dawn (Section 1.2.). Satellite-borne DOAS spectrometers, such as GOME, show that the tropospheric background column abundance of BrO is about 1 - 3 x 1013 molecules cm−2 . Over the polar regions during springtime, the “bromine explosion” produces large areas of elevated BrO (column abundance = 5 – 8 x 1013 molecule cm−2 ), presumably in the boundary layer or lower free troposphere [110-112].

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

233

3.2. Observations of IO, OIO and I2 Atmospheric observations of iodine species are very recent - within the last 5 years. IO has been detected in remote coastal locations by boundary-layer DOAS observations [36, 113, 114], and in the free troposphere by ground-based zenith sky DOAS measurements [115, 116] and by balloon-borne instruments [117, 118]. Observations of OIO [36, 119] and I2 have been reported in the coastal MBL [36]. Daytime measurements of IO at Mace Head, Ireland, correlate with low tide and high solar irradiance [36, 113]. At low tide, the emission rates of I2 [36] and alkyl iodides [19] are highest, and these are the photolytic precursors of atomic I and hence IO. Mixing ratios of OIO up to 3 ppt have been reported after sunset, when the photolysis of OIO ceases and any remaining IO generates OIO by the self-reaction (reaction 16b) [36, 119]. During the daytime OIO is mainly removed by photolyis, while at night uptake onto aerosol surfaces could account for the disappearance of the radical [119]. Figure 4 illustrates some recent measurements at Mace Head, Ireland, of the three iodine species, together with the NO3 radical. Panels a and b show daytime and night-time observations of I2 and OIO,

Figure 4. Diurnal concentration profiles of I2 , OIO, IO and NO3 observed at Mace Head, Ireland. The instrumental detection limit and the tidal variation are represented by thin black and thick broken lines, respectively. Daytime and night-time measurements periods are plotted as white and grey backgrounds, respectively. Reproduced with permission of the American Geophysical Union from Saiz-Lopez and Plane [36].

December 9, 2004

Time:

03:46pm

234

jp4121015.tex


made simultaneously by DOAS. Figure 4a shows that the concentrations of I2 during the day are about an order of magnitude smaller than at night. This is due to the very rapid photolysis of I2 , which has a lifetime of only about 10 s [120]. Note in Figure 4b the absence of OIO during the day, because of its rapid photolysis [80]. The nocturnal I2 concentration profile is characterised by a rapid pulse around low tide, whereas the OIO maximum occurs 1 - 2 hours later. The source of nocturnal IO and OIO appears to be the relatively fast reaction [121]: I2 + NO3 → I + INO3

(37)

I + O3 → IO + O2

(14)

IO + IO → OIO + I

(16b)

This is followed by

The NO3 mixing ratios illustrated in Figure 4d appear to be sufficient to produce the observed IO and OIO concentrations [36]. Finally, it should be noted that most field observations of iodine species have been made in mid-latitude coastal locations with productive oceans rich in macroalgae. A major question remains: how important is iodine chemistry in the open ocean marine boundary layer? One clue comes from a study of IO made off the north coast of Tenerife (Canary Islands), which is a volcanic peak with little surrounding coastal shelf [114]. Figure 5 shows measurements of the IO radical (solid points), which are compared with a model prediction of IO from a constrained model that includes a full treatment of the processing of iodine-containing species in sea-salt aerosol (solid line). The substantial levels of IO observed, which exhibit a simple diurnal solar dependence with no tidal correlation, indicate that there may indeed be significant sources of atmospheric iodine over the open ocean [59].

Figure 5. Observations and model predictions of IO at Tenerife (Canary Islands, Spain). A fully constrained photochemical box model was employed to simulate the IO measurements. Black dots and lines indicate measured and modelled IO, respectively. Reproduced with permission of the American Geophysical Union from McFiggans et al. [59].

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

235

4. SUMMARY In this chapter we have reviewed the chemistry of reactive halogen compounds in the lower troposphere. The main pathways for halogen release into the atmosphere are: natural and anthropogenic emissions of halocarbons; the emission of I2 from coastal zones rich in macroalgae and possibly from other sources over the open ocean; and the generation of halogen species from heterogeneous reactions on sea-salt aerosols. Reactive halogens play a number of very important roles in the chemistry of the lower troposphere. These include: the halogen-catalysed destruction of O3 in both the MBL and springtime polar boundary layer; increasing substantially the oxidation rate of dimethyl sulphide and yielding a different endproduct from the reactions of DMS with OH and NO3 ; increasing the NO2 /NO and OH/HO2 ratios in the boundary layer; converting elemental mercury into soluble mercury compounds during polar spring-time; and the formation in coastal regions of new particles which may grow to become cloud condensation nuclei and hence affect the global radiation budget. During the last decade, the DOAS technique has proved invaluable for the study of halogen chemistry, being used to make the first detection of ClO, BrO, IO, OIO and I2 and thus confirming the importance of halogen chemistry in the lower troposphere. Acknowledgements The work on halogen chemistry at the University of East Anglia is supported by the Natural Environment Research Council and the European Union.

References [1] Plane J.M.C., “Spectroscopic techniques for atmospheric measurements”, Topics in atmospheric and interstellar physics and chemistry, C.F. Boutron Ed. (Les Editions de Physique, Paris, 1995) pp. 309-334. [2] Platt U., Differential optical absorption spectroscopy (DOAS), Air monitoring by spectroscopy techniques, M.W. Sigrist Ed. (John Wiley, London, 1994) pp. 27-83. [3] Plane J.M.C. and Smith N., Atmospheric monitoring by differential optical absorption spectroscopy, Spectroscopy in environmental science, R.J.H. Clark and R.E. Hester Eds. (John Wiley, London, 1995) pp. 223-262. [4] Plane J.M.C. and Saiz-Lopez A., UV-visible Differential Optical Absorption Spectroscopy (DOAS), Analytical techniques for atmospheric measurement, D.E. Heard Ed. (Blackwell Publishing, Oxford, 2004) In press. [5] Platt U., Perner D. and Patz H.W., J. Geophys. Res. 84 (1979) 6329-6335. [6] Platt U., Perner D., Winer A.M., Harris G.W. and Pitts J.N., Geophys. Res. Lett. 7 (1980) 89-92. [7] Noxon J.F., Science 189 (1975) 547-549. [8] Noxon J.F., Norton R.B. and Henderson W.R., Geophys. Res. Lett. 5 (1978) 675-678. [9] Wayne R.P. Chemistry of atmospheres (Oxford University Press, Oxford, 2000). [10] Theiler R., Cook J.C., and Hager L.P., Science 202 (1978) 1094-1096. [11] Penkett S.A., Jones B.M.R., Rycroft M.J. and Simmons D.A., Nature 318 (1985) 550-553. [12] Class T.H. and Ballschmiter K., J. Atmos. Chem. 6 (1988) 35-46. [13] Mano S. and Andreae M.O., Science 263 (1994) 1255-1257. [14] Blake N.J., Blake D.R., Sive B.C., Chen T.Y., Rowland F.S., Collins J.E., Sachse G.W. and Anderson B.E., J. Geophys. Res.-Atmos. 101 (1996) 24151-24164. [15] Pilinis C., King D.B. and Saltzman E.S., Geophys. Res. Lett. 23 (1996) 817-820. [16] Schall C., Heumann K.G. and Kirst G.O., Fresenius J. Anal. Chem. 359 (1997) 298-305. [17] Khalil M.A.K. and Rasmussen R.A., Atmos. Environ. 33 (1999) 1305-1321. [18] Khalil M.A.K. and Rasmussen R.A., Atmos. Environ. 33 (1999) 1151-1158.

December 9, 2004

236

Time:

03:46pm

jp4121015.tex


[19] Carpenter L.J., Sturges W.T., Penkett S.A., Liss P.S., Alicke B., Hebestreit K. and Platt U., J. Geophys. Res.-Atmos. 104 (1999) 1679-1689. [20] Carpenter L.J., Malin G., Liss P.S. and Kupper F.C., Glob. Biogeochem. Cycle 14 (2000) 11911204. [21] Yvon-Lewis S.A., King D.B., Tokarczyk R., Goodwin K.D., Saltzman E.S. and Butler J.H., J. Geophys. Res.-Oceans 109 (2004) art. no.-C02008. [22] Carpenter L.J. and Liss P.S., J. Geophys. Res.-Atmos. 105 (2000) 20539-20547. [23] Sturges W.T., Cota G.F. and Buckley P.T., Nature 358 (1992) 660-662. [24] Keene W.C., Khalil M.A.K., Erickson D.J., McCulloch A., Graedel T.E., Lobert J.M., Aucott M.L., Gong S.L., Harper D.B., Kleiman G., Midgley P., Moore R.M., Seuzaret C., Sturges W.T., Benkovitz C.M., Koropalov V., Barrie L.A. and Li Y.F., J. Geophys. Res.-Atmos. 104 (1999) 8429-8440. [25] Burreson B.J., Moore R.E. and Roller P., Tetrahedron Lett. (1975) 473-476. [26] Schall C. and Heumann K.G., Fresenius J. Anal. Chem. 346 (1993) 717-722. [27] YokouchiY., Mukai H.,Yamamoto H., Otsuki A., Saitoh C. and NojiriY., J. Geophys. Res.-Atmos. 102 (1997) 8805-8809. [28] Carpenter L.J., Hebestreit K., Platt U., and Liss P.S., Atmos. Chem. Phys. 1 (2001) 9-18. [29] DeMore W.B., Sander S.P., Golden D.M., Hampson R.F., Kurylo M.J., Howard C.J., Ravishankara A.R., Kolb C.E. and Molina M.J. Chemical kinetics and photochemical data for use in stratospheric modelling(http://jpldataeval.jpl.nasa.gov), (Jet Propulsion Laboratory, Pasadena, California, 2004). [30] Yvon S.A. and Butler J.H., Geophys. Res. Lett. 23 (1996) 53-56. [31] Davis D.J., Crawford J., Liu S., McKeen S., Bandy A., Thornton D., Rowland F. and Blake D., J. Geophys Res. 101 (1996) 2135-2147. [32] Khalil M.A.K., Moore R.M., Harper D.B., Lobert J.M., Erickson D.J., Koropalov V., Sturges W.T. and Keene W.C., J. Geophys. Res.-Atmos. 104 (1999) 8333-8346. [33] Klick S. and Abrahamsson K., J. Geophys. Res.-Oceans 97 (1992) 12683-12687. [34] Reifenhauser W. and Heumann K.G., Atmos. Environ. 26 (1992) 2905-2912. [35] Carpenter L.J., Wevill D.J., Hopkins J.R., McQuaid J.B. and Palmer C.J., in preparation (2004) [36] Saiz-Lopez A. and Plane J.M.C., Geophys. Res. Lett. 31 (2004) art. no.-L04112. [37] McFiggans G., Coe H., Burgess R., Allan J., Cubison M., Alfarra M.R., Saunders R., Saiz-Lopez A., Plane J.M.C., Wevill D.J., Carpenter L.J., Rickard A.R. and Monks P.S., Atmos. Chem. Phys. 4 (2004) 701-713. [38] Truesdale V.W. and Canosamas C., J. Chem. Soc.-Faraday Trans. 91 (1995) 2269-2273. [39] Miyake Y. and Tsunogai S., J. Geophys. Res. 68 (1963) 3989-3993. [40] Liss P.S. and Slater P.G., Nature 247 (1974) 181-184. [41] Garland J.A. and Curtis H., J. Geophys Res. 86 (1981) 3183-3186. [42] Mochida M., Hirokawa J. and Akimoto H., Geophys. Res. Lett. 27 (2000) 2629-2632. [43] Finlayson-Pitts B.J., Ezell M.J. and Pitts J.N., Nature 337 (1989) 241-244. [44] Behnke W., Kruger H.U., Scheer V. and Zetzsch C., J. Aerosol. Sci. 22 (1991) S609-S612. [45] Behnke W., George C., Scheer V. and Zetzsch C., J. Geophys. Res.-Atmos. 102 (1997) 3795-3804. [46] Schweitzer F., Mirabel P. and George C., J. Atmos. Chem. 34 (1999) 101-117. [47] McFiggans G., Cox R.A., Mossinger J.C., Allan B.J. and Plane J.M.C., J. Geophys. Res.-Atmos. 107 (2002) art. no.-4271. [48] Herrmann H., Majdik Z., Ervens B. and Weise D., Chemosphere 52 (2003) 485-502. [49] Zetzsch C. and Behnke W., Ber. Bunsen-Ges. Phys. Chem. 96 (1992) 488-493. [50] George C., Behnke W., Scheer V., Zetzsch C., Magi L., Ponche J.L. and Mirabel P., Geophys. Res. Lett. 22 (1995) 1505-1508. [51] Gershenzon M.Y., Il’in S., Fedotov N.G., Gershenzon Y.M., Aparina E.V. and Zelenov V.V., J. Atmos. Chem. 34 (1999) 119-135.

December 9, 2004

Time:

03:46pm

jp4121015.tex

ERCA 6

237

[52] Wayne R.P., Poulet G., Biggs P., Burrows J.P., Cox R.A., Crutzen P.J., Hayman G.D., Jenkin M.E., Lebras G., Moortgat G.K., Platt U. and Schindler R.N., Atmos. Environ. 29 (1995) 2677-2881. [53] Fan S.M. and Jacob D.J., Nature 359 (1992) 522-524. [54] Mozurkewich M., J. Geophys. Res.-Atmos. 100 (1995) 14199-14207. [55] Tang T. and McConnell J.C., Geophys. Res. Lett. 23 (1996) 2633-2636. [56] Sander R. and Crutzen P.J., J. Geophys. Res.-Atmos. 101 (1996) 9121-9138. [57] Vogt R., Crutzen P.J. and Sander R., Nature 383 (1996) 327-330. [58] Sander R., Vogt R., Harris G.W. and Crutzen P.J., Tellus Ser. B-Chem. Phys. Meteorol. 49 (1997) 522-532. [59] McFiggans G., Plane J.M.C., Allan B.J., Carpenter L.J., Coe H. and O’Dowd C., J. Geophys. Res.-Atmos. 105 (2000) 14371-14385. [60] Abbatt J.P.D. and Waschewsky G.C.G., J. Phys. Chem. A 102 (1998) 3719-3725. [61] Fickert S., Adams J.W. and Crowley J.N., J. Geophys. Res.-Atmos. 104 (1999) 23719-23727. [62] Holmes N.S., Adams J.W. and Crowley J.N., Phys. Chem. Chem. Phys. 3 (2001) 1679-1687. [63] Adams J.W., Holmes N.S. and Crowley J.N., Atmos. Chem. Phys. 2 (2002) 79-91. [64] Vogt R., Sander R., Von Glasow R. and Crutzen P.J., J. Atmos. Chem. 32 (1999) 375-395. [65] Ayers G.P., Gillet R.W., Cainey J.M. and Dick A.L., J. Atmos. Chem. 33 (1999) 299-319. [66] Gabriel R., von Glasow R., Sander R., Andreae M.O. and Crutzen P.J., J. Geophys. Res.-Atmos. 107(D19) doi: 10.1029/2001JD001133 (2002) 8032. [67] Platt U. and Janssen C., Faraday Discuss. (1995) 175-198. [68] Wennberg P.O., Nature 397 (1999) 299-301. [69] Abbatt J.P.D., Geophys. Res. Lett. 21 (1994) 665-668. [70] Abbatt J.P.D. and Nowak J.B., J. Phys. Chem. A 101 (1997) 2131-2137. [71] Sander R., Surv. Geophys. 20 (1999) 1-31. [72] Stolarsky R.S. and Cicerone R.J., Can. J. Chem.-Rev. Can. Chim. 52 (1974) 1610-1615. [73] Molina M.J. and Rowland F.S., Nature 249 (1974) 810-812. [74] Wofsy S.C., McElroy M.B. and Yung Y.L., Geophys. Res. Lett. 2 (1975) 215-218. [75] McElroy M.B., Salawitch R.J., Wofsy S.C. and Logan J.A., Nature 321 (1986) 759-762. [76] Solomon S., Nature 347 (1990) 347-354. [77] Cox R.A., Bloss W.J., Jones R.L. and Rowley D.M., Geophys. Res. Lett. 26 (1999) 1857-1860. [78] Jimenez J.L., Bahreini R., Cocker D.R., Zhuang H., Varutbangkul V., Flagan R.C., Seinfeld J.H., O’Dowd C.D. and Hoffmann T., J. Geophys. Res.-Atmos. 108 (2003) art. no.-4318. [79] O’Dowd C.D., Jimenez J.L., Bahreini R., Flagan R.C., Seinfeld J.H., Hameri K., Pirjola L., Kulmala M., Jennings S.G. and Hoffmann T., Nature 417 (2002) 632-636. [80] Ashworth S.H., Allan B.J. and Plane J.M.C., Geophys. Res. Lett. 29 (2002) art. no.-1456. [81] Saiz-Lopez A., Plane J.M.C. and Shillito J.A., Geophys. Res. Lett. 31 (2004) art. no.-L03111. [82] Adams J.W. and Cox R.A., J. Phys. IV 12 (2002) 105-124. [83] Nagao I., Matsumoto K. and Tanaka H., Geophys. Res. Lett. 26 (1999) 3377-3380. [84] Galbally I.E., Bentley S.T. and Meyer C.P., Geophys. Res. Lett. 27 (23) (2000) 3841-3844. [85] Barrie L.A., Bottenheim J.W., Schnell R.C., Crutzen P.J. and Rasmunssen R.A., Nature 334 (1988) 138-141. [86] Finlayson-Pitts B.J., Livingston F.E. and Berko H.N., Nature 343 (1990) 622-625. [87] McConnell J.C., Henderson G.S., Barrie L.A., Bottenheim J.W., Niki H., Langford C.H. and Templeton E.M.J.,Nature 355 (1992) 150-152. [88] Hausmann M. and Platt U., J. Geophys. Res.-Atmos. 99 (1994) 25399-25413. [89] Lebras G. and Platt U., Geophys. Res. Lett. 22 (1995) 599-602. [90] Tuckermann M., Ackermann R., Gölz C., Lorenzen-Schmidt H., Senne T., Stutz J., Trost B., Unold W. and Platt U., Tellus 49B (1997) 533-555. [91] Martinez M., Arnold T. and Perner D., Ann. Geophys.-Atmos. 17 (1999) 941-956. [92] Honninger G. and Platt U., Atmos. Environ. 36 (2002) 2481-2489.

December 9, 2004

238

Time:

03:46pm

jp4121015.tex


[93] Toumi R., Geophys. Re. Lett. 21 (1994) 117-120. [94] Bedjanian Y., Poulet G. and LeBras G.,Int. J. Chem. Kinet. 28 (1996) 383-389. [95] James J.D., Harrison R.M., Savage N.H., Allen A.G., Grenfell J.L., Allan B.J., Plane J.M.C., Hewitt C.N., Davison B. and Robertson L.,J. Geophys. Res.-Atmos. 105 (D21) (2000) 2637926392. [96] Boucher O., Moulin C., Belviso S., Aumont O., Bopp L., Cosme E., von Kuhlmann R., Lawrence M.G., Pham M., Reddy M.S., Sciare J. and Venkataraman C., Atmos. Chem. Phys. 3 (2003) 49-65. [97] Sander R., Keene W.C., Pszenny A., Arimoto R., Ayers G.P., Baboukas E., Cainey J.M., Crutzen P.J., Duce R., Hoenninger G., Huebert B., Maenhaut W., Mihalopoulos N., Turekian V. and Van Dingenen R., Atmos. Chem. Phys. 3 (2003) 1301-1336. [98] Charlson R.J., Lovelock J.E., Andreae M.O. and Warren S.G., Nature 326 (1987) 655-661. [99] Jenkin M.E., Cox R.A. and Candeland D.E., J. Atmos. Chem. 2 (1985) 359-375. [100] Lindberg S.E., Brooks S., Lin J., Scott K., Goodsite M., Tilden M.S., Landis M. and Stevens R., Abstr. Pap. Am. Chem. Soc. 222 (2001) 66-ENVR. [101] Goodsite M.E., Plane J.M.C. and Skov H., Environ. Sci. Technol. 38 (2004) 1772-1776. [102] Calvert J.G. and Lindberg S.E., Atmos. Environ. 37 (2003) 4467-4481. [103] Burkholder J.B., Curtius J., Ravishankara A.R. and Lovejoy E.R., Atmos. Chem. Phys. 4 (2004) 19- 34. [104] Honninger G., Leser H., Sebastian O. and Platt U., Geophys. Res. Lett. 31 (2004) art. no.-L04111. [105] Leser H., Honninger G. and Platt U., Geophys. Res. Lett. 30 (2003) art. no.-1537. [106] Hebestreit K., Stutz J., Rosen D., Matveiv V., Peleg M., Luria M. and Platt U., Science 283 (1999) 55-57. [107] Stutz J., Ackermann R., Fast J.D. and Barrie L., Geophys. Res. Lett. 29 (2002) art. no.-1380. [108] Honninger G., Bobrowski N., Palenque E.R., Torrez R. and Platt U., Geophys. Res. Lett. 31 (2004) art. no.-L04101. [109] Matveev V., Peleg M., Rosen D., Tov-Alper D.S., Hebestreit K., Stutz J., Platt U., Blake D. and Luria M., J. Geophys. Res.-Atmos. 106 (2001) 10375-10387. [110] Wagner T. and Platt U., Nature 395 (1998) 486-490. [111] Richter A., Wittrock F., Eisinger M. and Burrows J.P., Geophys. Res. Lett. 25 (1998) 2683-2686. [112] Van Roozendael M., Wagner T., Richter A., Pundt I., Arlander D.W., Burrows J.P., Chipperfield M., Fayt C., Johnston P.V., Lambert J.C., Kreher K., Pfeilsticker K., Platt U., Pommereau J.P., Sinnhuber B.M., Tornkvist K.K. and Wittrock F., Adv. Space Res. 29 (2002) 1661-1666. [113] Alicke B., Hebestreit K., Stutz J. and Platt U., Nature 397 (1999) 572-573. [114] Allan B.J., McFiggans G., Plane J.M.C. and Coe H., J. Geophys. Res.-Atmos. 105 (2000) 1436314369. [115] Wittrock F., Muller R., Richter A., Bovensmann H. and Burrows J.P., Geophys. Res. Lett. 27 (2000) 1471-1474. [116] Friess U., Wagner T., Pundt I., Pfeilsticker K. and Platt U., Geophys. Res. Lett. 28 (2001) 19411944. [117] Pundt I., Pommereau J.P., Phillips C. and Lateltin E., J. Atmos. Chem. 30 (1998) 173-185. [118] Bosch H., Camy-Peyret C., Chipperfield M.P., Fitzenberger R., Harder H., Platt U. and Pfeilsticker K., J. Geophys. Res.-Atmos. 108 (2003) art. no.-4455. [119] Allan B.J., Plane J.M.C., and McFiggans G., Geophys. Res. Lett. 28 (2001) 1945-1948. [120] Saiz-Lopez A., Saunders R.W., Joseph M., Ashworth, S.H. and Plane J.M.C., Atmos. Chem. Phys. 4 (2004) 1443-1450. [121] Chambers R.M., Heard A.C. and Wayne R.P., J. Phys. Chem. 96 (1992) 3321-3331.