Production of and Dissociative Electron Attachment ... - ACS Publications

Jan 13, 2015 - a flowing afterglow using a novel technique. CH2I is produced by dissociative electron attachment to CH2I2, leading to the established ...
0 downloads 0 Views 571KB Size
Subscriber access provided by CARNEGIE MELLON UNIV

Letter

Production of and Dissociative Electron Attachment to the Simplest Criegee Intermediate in an Afterglow Justin P. Wiens, Nicholas S. Shuman, and Albert A Viggiano J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/jz502569w • Publication Date (Web): 13 Jan 2015 Downloaded from http://pubs.acs.org on January 15, 2015

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 free 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 accessible to all readers and 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.

The Journal of Physical Chemistry Letters 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 11

The Journal of Physical Chemistry Letters

1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Production of and Dissociative Electron Attachment to the Simplest Criegee Intermediate In an Afterglow Justin P. Wiens, Nicholas S. Shuman, and Albert A. Viggiano* Air Force Research Laboratory, Space Vehicles Directorate, Kirtland AFB, NM 87117 *E-mail: [email protected] Abstract The simplest Criegee intermediate, CH2OO, has been produced in a flowing afterglow using a novel technique. CH2I is produced by dissociative electron attachment to CH2I2, leading to the established reaction CH2I + O2 → CH2OO + I. Presence of CH2OO is established by observation of dissociative electron attachment to yield O– using the variable electron and neutral density attachment mass spectrometry (VENDAMS) technique. The measurements establish the electron attachment rate coefficient of thermal electrons at 300 K to CH2OO as 1.2 ± 0.3 x 10-8 cm3 s-1. Thermal electron attachment is solely dissociative, and is not a promising route to producing stable CH2OO–. The results open the possibility of measuring ion–molecule chemistry involving Criegee intermediates, as well as the reactivity of other unstable radicals produced in an analogous manner. TOF Graphic

Keywords: plasma, mass spectrometry, kinetics, ion chemistry, reactions Carbonyl oxides, or Criegee intermediates (CIs), are produced in the lower atmosphere by alkene ozonolysis. These elusive and reactive species are difficult to isolate, and only recently have they been directly identified in experiments.1-5 An excellent review by Taatjes et al. has been published on CIs and their atmospheric relevance.6 Criegee decomposition plays a key role in the production of OH,6-8 which has as a substantial effect on air quality especially in urban areas. In the laboratory, CIs have been produced by several methods including ozonolysis,9 a complicated mechanism starting with a chlorine atom reacting with DMSO,1 and reaction of O2 with iodoalkyl radicals via CH2I + O2 → CH2OO + I where CH2I is created via photolysis of CH2I2.10 The McCarthy group recently generated the CH2OO by passing CH4 and O2 over a high-voltage DC discharge in vacuum;2 calculations indicated a likely mechanism is O2 abstraction of H from vibrationally excited CH3O2*, forming the CI.11 Nakajima and Endo3 created CH2OO by pulsing an electric discharge in a supersonic jet of a dilute CH2Br2/O2 mixture.

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

Page 2 of 11

2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In these experiments, many energetic species may also have been created in the discharge, although microwave spectroscopy was used to directly identify the CI. This work presents a novel scheme to generate the simplest Criegee intermediate, CH2OO, in a flowing afterglow-Langmuir probe (FALP) apparatus. Instead of photolysis, the precursor iodoalkyl radical CH2I is created in situ via dissociative electron attachment to CH2I2, generating I– as a co-product. Our experiment ensures rapid thermalization by He collisions of any intermediates including the CI. The method provides a route for studying the reactivity of CIs with charged species, as well as being a proof of concept for extending the use of a FALP apparatus to study reactivity of radical species. The current scheme relies on O– production via dissociative electron attachment to CH2OO as a signature of the CI via CH2OO + e– → O– + CH2O. This attachment reaction has not been previously reported, although detailed calculations of the anion and neutral structures have been undertaken in the aim of showing the feasibility of photoelectron studies.12,13 As a result, much of the current work focuses on ensuring and showing that the observed O– signal is indeed a result of attachment to CH2OO, thereby both establishing the electron attachment rate coefficient and that the simplest CI has been produced in the afterglow. The chemistry in FALP experiments can be complex because there are competing processes in the flow tube. The present experiment adds an additional complication in that the comparatively slow, neutral–neutral reaction between CH2I and O2 precedes electron attachment to the CI. Chemical kinetic modeling is necessary to provide a handle on all possible chemical reactions occurring in the weakly ionized plasma. Here we employ the variable electron and neutral density attachment mass spectrometry (VENDAMS) method developed in our laboratory, which has previously yielded kinetic data on dissociative recombination, mutual neutralization, and electron attachment for a wide variety of cations, anions, and neutrals.14 Concerning the ionic products of interest, the reaction pathways for previous measurements have not included neutral–neutral reactions. The present experiment on Criegee Intermediate production extends the utility of VENDAMS to more complex systems involving neutral– neutral chemical reactions. In Figure 1, we present a typical anion mass spectrum taken with both CH2I2 and O2 added through the reactant inlet. I– dominates the spectrum because it is created via primary attachment to CH2I2, whereas O– (inset) is a minor anion created two steps later in step 3 of Scheme 1. Scheme 1. Important Chemical Reactions in the Criegee Experiment e– + CH2I2 → CH2I + I–

(1)

CH2I + O2 → CH2OO + I

(2)

CH2OO + e– → O– + CH2O

(3)

Ar+ + O2 → O2+ + Ar

(4)

Ar+ + I– → Ar + I

(5)

O2+ + I– → neutrals

(6)

ACS Paragon Plus Environment

Page 3 of 11

The Journal of Physical Chemistry Letters

3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

O2+ + e– → O + O

(7)

O2+ + O– → neutrals

(8)

Ar+ is initially the dominant ion in the plasma and causes dissociation of CH2I2 to fragment cations and neutrals (Supporting Information), some of which are CH2I radicals. Additional sources of I– might then be dissociative electron attachment to CH2I (Supporting Information) or CH2I + O2– → CH2OO + I–, where the O2– is generated in the slow, 3-body associative attachment to O2. Including any of these channels in the model had minimal effects on the results, however; attachment to CH2I2 via Reaction (1) dominates I– production. The rate coefficients for (4)–(7) have been previously determined.15-18 Reaction (8) has not been measured and was allowed to vary in our analysis up to a maximum of 6 x 10-8 cm3 s-1 based on previously measured MN processes.16,17 For completeness in our model, we also considered the following neutral–neutral reactions: CH2I + O2 → IO + CH2O CH2O2 + I → IO + CH2O

(9) (10)

Although we cannot directly measure species in these neutral channels, previous studies have provided upper bounds on their rate coefficients.19-21 A table of all relevant rate coefficients is provided in the Supporting Information.

Figure 1. Anion spectrum of the CH2I2 + O2 system at a flow tube pressure of 1.0 Torr, [CH2I2] = 1.7 x 1010 cm-3, [O2] = 8.6 x 1013 cm-3, and [e]0 = 1.3 x 1010 cm-3. Electron attachment to CH2I2 dominates the spectrum, yielding I–. Inset shows the O– peak at m/z = 16 from electron attachment to CH2O2. Peaks at m/z = 19, 79, 81, and 149 are due to contamination from prior experiments involving F–, Br–, and triflate-

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

Page 4 of 11

4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

containing species using the same apparatus. A complete list of the impurity peaks and their likely identities is provided in the Supporting Information. The source of the O– signal in Fig. 1 is established using the VENDAMS method by monitoring the relative abundance of product ions as a function of the initial plasma density (i.e. the concentration of Ar+ and e– at the time when reactants are introduced to the afterglow). The primary benefit of the VENDAMS approach is that the shape of the O– curve is a strong function of the sequence of reactions producing it. Fig. 2 shows the best fit modeled curve assuming production of O– from attachment to CH2OO (i.e. Reaction 3 above), as well as modeled curves assuming production from alternate possible mechanisms, either charge exchange between O2– and O, charge exchange between I– and any oxygencontaining impurity, or via electron attachment directly to any oxygen-containing impurity. The magnitudes of each of these modeled curves are a function of the magnitudes of the rate coefficients for each reaction, which may or not be known. Importantly, the shapes of the curves are primarily a function of the particular sequence of reactions leading to the product, and only attachment to CH2OO satisfactorily fits the data.

Figure 2. VENDAMS plot of observed O– branching (solid points; point near [e]0 = 2 x 109 is an experimental outlier) relative to I– (not shown) as a function of initial electron density [e]0 after 7 ms reaction time at 300 K and 1.5 Torr with [CH2I2]0 = 1.1 x 1010 cm-3 and [O2]0 = 3.5 x 1014 cm-3. The best fit model for this data set (solid red curve) assumes O– production via electron attachment to CH2OO with k3 = 7 ±4 x 10-9 cm3 s-1, and the reported value from seven data sets of k3 = 1.2 ± 0.3 x 10-8 cm3 s-1. Fits at the uncertainty limits of this data set correspond to k3min = 4.2 x 10-9 and k3max = 1.2 x 10-8 cm3 s-1 (dashed red curves; see ref 14 for explanation) are also shown. Models of O– signal that could arise from other mechanisms (ion–molecule reaction of any oxygenated impurity XO with I–, blue curve; electron

ACS Paragon Plus Environment

Page 5 of 11

The Journal of Physical Chemistry Letters

5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

attachment to XO, green curve; O2– + O, orange curve) are shown for comparison. The green and blue curves have been arbitrarily offset for clarity. The impurity channels can, in some combination, reproduce the general magnitude of the data depending on the rate constants, but not the shape of the data. VENDAMS data was also collected with only one of the reactants, either CH2I2 or O2, present. When only O2 is added (Figure 3), the anion spectrum is very sparse, with small peaks at m/z = 32 (O2–) and halide impurity peaks at 35 and 37 (Cl–), 79 and 81 (Br–), and 127 (I–); no measurable O– peak is observed. When only CH2I2 is added (Figure 4), numerous peaks appear in the anion spectrum, but the main peaks are I– from (1), Br– from attachment to a Br-containing contaminant present in the CH2I2, and Cl– from fast attaching Cl-containing species, likely CCl4, used in previous experiments on the apparatus. Many of these peaks are absent in Figure 1 because these anions presumably react with O2 or O2+. A few percent O–, relative to the amount when O2 is also added, may also be present, presumably from trace O2 amounts present in the flow tube due to ppm level impurity in the He buffer gas. The absence of an O– signal under these two conditions supports the notion that the chemistry yielding the O– signal observed in Figure 1 requires both CH2I2 and O2 to be present in the flow tube.

Figure 3. Anion spectrum with only O2 added at [O2] = 1.8 x 1014 cm-3 and a flow tube pressure of 1.5 Torr. No signal can be seen at 16 amu (inset).

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

Page 6 of 11

6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. Anion spectrum with only CH2I2 added at [CH2I2] = 1.3 x 1010 cm-3 and [e]0 = 1.4 x 1010 cm-3 and a flow tube pressure of 1.5 Torr. Any O– that may be present (inset) is within the noise.

As shown in Figure 2, the shape of the O– signal as a function of [e]0 precludes the signal from having arisen from electron attachment to or ion–molecule reaction with a species present as an impurity. The proper shape of that curve would be reproduced by assuming dissociative electron attachment to O3, which may be produced by dissociative recombination of O4+ (which is formed slowly22 via the threebody process M + O2 + O2+ → M + O4+). However, attachment to O3 is known to proceed slowly,23 and this mechanism according to the modeling yields four orders of magnitude less O– than does the scheme involving the Criegee intermediate, and therefore cannot explain the data. Some O– should be produced by associative attachment to O2 (as mentioned earlier) followed by O2– + O reaction; however, the primary O-atom source is dissociative recombination of O2+, which yields a curve with much too steep a slope (Fig. 2). O+ radiatively recombines negligibly slowly with electrons in the afterglow,24 and O atoms will not associatively attach thermal electrons,25 making O– formation via this channel impossible. We can also reject the possibility that the O– signal arises from an isomer of CH2OO. The neutral–neutral reaction (2) has been well characterized and proceeds with a pressure-independent rate coefficient k2 = 1.6 ± 0.2 x 10-12 cm3 s-1 (see ref 26 and references therein). This reaction is believed to proceed through an excited CH2IO2 complex,6 and branching to stabilized CH2IO2 as opposed to the CI depends on the pressure and buffer gas composition.26 Under our experimental conditions, branching to the CI in (2) should be near unity. In any case, neither formic acid nor dioxirane—if formed—is exothermic to dissociative electron attachment forming O–. Formic acid has been shown not to attach thermal (300 K) electrons at all.27,28 A less stable isomer, dioxymethane, would be exothermic to dissociative attachment; however, this species has been shown to rapidly isomerize to formic acid.29 On

ACS Paragon Plus Environment

Page 7 of 11

The Journal of Physical Chemistry Letters

7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

that note, the isomerization barriers between the CI and dioxirane or between dioxirane and dioxymethane are substantial, and isomerization between these species cannot occur at the energies of the present experiment. No anion peak at m/z = 46 (CH2O2–) was observed in the present experiments. After establishing that the observed O– signal arose from dissociative electron attachment to the CI, we took seven VENDAMS data sets with varying amounts of both CH2I2 and O2 reactants. Best fits to the data and uncertainties were obtained using Monte Carlo-sampling of the full parameter space and least squares comparison of the simulated O–:I– ratios to the experimental data, as explained in detail elsewhere.14 The seven data sets give the average electron attachment rate coefficient k3 = 1.2 ± 0.3 x 10-8 cm3 s-1 excluding reaction channels (9) and (10). This result was invariant to the reactants’ concentrations when either or both were varied over ~1 order of magnitude. Including additional channels in the reaction scheme may cause modest changes in the O–:I– ratio and thereby the reported Criegee attachment rate coefficient. IO formed via (9) may attach electrons to form I–; if the attachment is fast (3 x 10-7 cm3 s-1), the inclusion of (9) lowers k3 by 0.4 eV above the dissociation barrier to CH2O + O–, the rate coefficient for that dissociation will be on the order of 1014 s-1, orders of magnitude larger than that for autodetachment and on a much faster time scale than stabilization to the parent anion. The present results coupled with the ab initio calculations suggest that production of the Criegee Intermediate anion via thermal electron attachment is unlikely to succeed, instead leading quickly to dissociated products. 7

3

-1

In the course of these experiments, we also determined the rate coefficient for CH2I2 attachment (1) to be k1 = 3.0 ± 1.5 x 10-8 cm3 s-1 at 300 K using the established FALP technique of monitoring electron depletion as a function of reaction time34 by adding a known concentration of CH2I2 to the

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters

Page 8 of 11

8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

afterglow and translating the Langmuir probe along the flow tube. The attachment is purely dissociative, forming I–. Despite the extensive literature on electron attachment to halomethanes,35,36 we are unable to find a prior reported measurement. Although not relevant to the current study, it is interesting to note that attachment to CH2I2 runs counter to the general trends noted for halomethanes. Typically the rate coefficients for attachment increase with substitutions of species in the order of H