Assignment of the Fundamental Modes of Hydroxyacetone Using Gas

Article pubs.acs.org/JPCA

Assignment of the Fundamental Modes of Hydroxyacetone Using Gas-Phase Infrared, Far-Infrared, Raman, and ab Initio Methods: Band Strengths for Atmospheric Measurements Rodica Lindenmaier,† Nicole Tipton,‡ Robert L. Sams,† Carolyn S. Brauer,† Thomas A. Blake,† Stephen D. Williams,‡ and Timothy J. Johnson*,† †

Pacific Northwest National Laboratory, Richland, Washington 99354, United States A. R. Smith Department of Chemistry, Appalachian State University, Boone, North Carolina 28618, United States

‡

S Supporting Information *

ABSTRACT: Hydroxyacetone (acetol) is a simple organic molecule of interest in both the astrophysical and atmospheric communities. It has recently been observed in biomass burning events and is a known degradation product of isoprene oxidation. However, its vibrational assignment has never been fully completed, and few quantitative data are available for its detection via infrared spectroscopy. Our recent acquisition of both the pressure-broadened gas-phase data and the far-IR spectra now allow for unambiguous assignment of several (new) bands. In particular, the observed C-type bands of several fundamentals (particularly in the far-infrared) and a few combination bands demonstrate that the monomer is in a planar (Cs) conformation, at least a majority of the time. As suggested by other researchers, the monomer is a cis−cis conformer stabilized by an intramolecular OH···OC hydrogen bond forming a five-membered planar ring structure. Band assignments in the Cs point group are justified (at least for a good fraction of the molecules in the ensemble) by the presence of the C-type bands. The results and band assignments are well confirmed by both ab initio MP2-ccpvtz calculations and GAMESS (B3LYP) theoretical calculations. In addition, using vetted methods for quantitative measurements, we report the first IR absorption band strengths of acetol (also in electronic format) that can be used for atmospheric monitoring and other applications. (H3CC(O)CH(O)), or formaldehyde (CH2O).11,13−15 The lower rate constant for the MACR reaction with OH (which is the dominant loss process) results in a longer atmospheric lifetime16 for methacrolein [6−10 h], as compared to the lifetime of isoprene17 [1−2 h], allowing for greater accumulation of MACR in the atmosphere. Orlando et al.15 studied the OH-initiated MACR oxidation in the presence of NO and quantified hydroxyacetone as the primary product (47 ± 5%), agreeing well with the earlier 43 ± 5% value of Tuazon and Atkinson.14 They determined the temporal profiles of methacrolein and its oxidation products by using infrared absorption spectroscopy and also confirmed the 4-day tropospheric lifetime of hydroxyacetone for the OH oxidation path.18 Existing mechanisms consider glyoxal, methylglyoxal, glycolaldehyde, and hydroxyacetone as higher-generation oxidation products of isoprene, but there are limited studies in which these are measured directly from isoprene oxidation. Moreover, some recent studies have pointed to potential gaps in our understanding of isoprene oxidation.17,19,20 Theoretical calcu-

1. INTRODUCTION Hydroxyacetone [H3CC(O)CH2OH, 1-hydroxy-2-propanone or HA] is one of the simplest molecules containing both an alcohol and a carbonyl group. It can be considered as either the methyl derivative of glycolaldehyde (the simplest sugar component) or alternatively the hydroxyl derivative of acetone, thus lending itself to its more common name of acetol. It is a volatile organic compound (VOC) that, together with other carbonyl compounds, contributes to the production of free radicals, photo-oxidants, and secondary organic aerosols (SOA) in the atmosphere. Its natural sources include tree emissions and biomass burning: Emissions from biomass burning have recently been studied under both laboratory conditions1 and field studies using both infrared and GC−MS techniques.2−5 In terms of global fluxes the biogenic emission source is the more significant, as hydroxyacetone is an important secondary oxidation product of isoprene (2-methyl-1,3-butadiene), which is produced by living organisms (trees) at a rate of ∼550 Tg/ year.6−8 The first-yield products of isoprene photooxidation in the presence of NO are methyl vinyl ketone (MVK) and methacrolein (MACR).9−12 These primary products are subsequently transformed via reactions with either OH radicals or O3 to yield second-generation products such as glycolaldehyde (HC(O)CH2OH), hydroxyacetone, methylglyoxal © 2016 American Chemical Society

Received: May 19, 2016 Revised: July 9, 2016 Published: July 9, 2016 5993

DOI: 10.1021/acs.jpca.6b05045 J. Phys. Chem. A 2016, 120, 5993−6003

Article

The Journal of Physical Chemistry A

Figure 1. Three-dimensional plots of potential conformers of hydroxyacetone, as suggested from Sharma et al.30

lations by Dibble19,20 suggest that glyoxal, methylglyoxal, glycolaldehyde, and hydroxyacetone can even be formed as first-generation products from isoprene under high NOx conditions. Paulot et al.17 observed first-generation yields of glycolaldehyde and hydroxyacetone from isoprene oxidation via the δ-hydroxy channel. In addition to being an oxidation product, acetol is also a precursor of other atmospherically relevant species: Its photooxidation in the presence of NO results in methylglyoxal11 and its oxidation by OH yields both formic acid and acetic acid.21 Indeed, hydroxyacetone is removed from the atmosphere primarily via its reaction with OH radicals (τ = 4 days), but a competing mechanism for atmospheric loss is via photolysis (τ > 14 days).18 In many applications quantification of hydroxyacetone can be achieved via the use of gas-phase infrared spectroscopy. Such quantitation, however, requires the existence of high-quality reference data. Many of the results reported here were originally generated as a component of the Northwest Infrared quantitative gas-phase database22−24 [nwir.pnl.gov], which was created to detect and quantify gas-phase molecules either in the laboratory or in field measurements near the earth’s surface. The database contains the spectra of ca. 500 species, each spectrum derived from multiple burdens, and each burden having the analyte pressure-broadened to 760 Torr. Such spectra can be used for quantitation: a reference spectrum of acetol was used to determine HA emission factors (EF) in biomass burning.1 In their similar earlier studies of Indonesian and African fuels, Christian et al.5 reported acetol as a major emission product from burning rice straw with EF in the 21−34 g/kg range. In a later study Akagi et al. found1 that the EF for acetol ranged between 0.45 and 6.18 g/kg for different fuel types during biomass burning: tropical forest, savanna, crop, and pasture. More recently, we have used the PNNL data to provide quantitative reference data and also to complete the vibrational analyses of similar species relevant to biomass burning and isoprene oxidation including glyoxal/methylglyoxal,25 glycolaldehyde,26 and alkanes,27 along with isoprene itself.8 During the course of our studies, however, it has become clear that the vibrational analyses for hydroxyacetone (HA) are incomplete for various reasons, notably a lack of far-IR data and incomplete mid-IR or Raman data, but especially due to the dearth of gasphase spectra. For example, Mohaček-Grošev28 has reported a very complete analysis of HA to date but does not report any far-IR data, and though modes are assigned using Cs symmetry, her assignments are based only on condensed-phase spectra and theoretical mode descriptions. This is where the band profiles associated with gas-phase data can be of great utility. Sharma et al.29,30 have invoked more powerful experimental methods, studying HA at reduced temperatures using both jet techniques (IR, Raman) and low-temperature crystallography

along with Gaussian quantum calculations. Their results again suggest that at room temperature gas-phase HA is most stable in the Cc cis conformer with an intramolecular hydrogen bond stabilizing the molecule in a five-membered ring as seen in the first plot of Figure 1. Their results also showed that as clusters and oligomers form at low temperatures, this hydrogen bond breaks, with the O−H bond rotating nearly 180° in the dihedral angle so as to form dual intermolecular hydrogen bonds to both the carbonyl and hydroxyl oxygen atoms of the neighboring molecule. Also, in the 2009 work of Sharma et al. two conformers of HA were identified: A low-energy conformer Cc (Figure 1) is approximately 1000 cm−1 lower in energy than the Tt conformer. Because our measurements are of the room temperature gas-phase sample, the faction of Tt conformer in the present study should be negligible. The potential conformers as suggested by Sharma et al. are seen in Figure 1. In light of these considerations we have chosen to complement the seminal works of Mohaček-Grošev28 and Sharma et al.,29−31 aiming to improve the nearly complete infrared vibrational frequency assignments for this molecule, using additional methods at our disposal, e.g., far-IR spectroscopy. Equally important, we report for the first time quantitative gas-phase IR absorption band strengths that allow for quantitation of this species. We have chosen to use the available PNNL data along with ab initio and density functional theory to arrive at a more complete vibrational analysis of this atmospherically important molecule. We have complemented the quantitative PNNL mid-IR gas-phase data with additional experimental data including (1) gas-phase farinfrared data and (2) liquid-phase Raman data, along with (3) anharmonic calculations for the theoretical vibrational frequencies, allowing us to extend the very thorough studies of Mohaček-Grošev28 and Sharma et al.30 In all cases, for the experimental data we too had to labor to obtain the spectra of “dry” hydroxyacetone. Other researchers have noted the challenges associated of removing residual H2O and also that hydrolysis occurs under certain conditions.13−18,28−31 For example, we were initially misled in our assignments in the C−H stretching region due to trace liquid water bands evidencing themselves in both the IR and Raman liquidphase spectra.

2. EXPERIMENTAL SECTION The spectroscopic data were recorded with different instruments as described in the following sections. 2.a. Mid-Infrared Measurements. The mid-infrared (MIR) data presented here belong to the Pacific Northwest National Laboratory (PNNL) gas-phase database and the procedures have been described in detail by Sharpe et al.22 and Johnson et al.23 In brief, the measurements were performed with an evacuated Bruker 66v/S FTIR spectrometer. A glow 5994


Article


interferometer (i.e., the light recording system) and the Raman excitation laser require frequency calibration: The Fourier interferometer frequencies are referenced to those of a HeNe laser (Connes advantage), but to better calibrate the interferometer HeNe laser frequency a series of five known Hg lamp atomic lines were used, bleeding in light from room fluorescent lamps. The frequencies were selected to cover the absolute range (∼6000 to 9000 cm−1) of the FT-Raman system and were validated using known NIST frequencies such that the mean deviation for all five lines is ≤0.2 cm−1. The Raman excitation laser is calibrated in a second step as all Raman (shift) frequencies are measured relative to it. We used a protocol that employs the frequency equivalence of the Stokes and anti-Stokes frequency shifts; the Nd:YAG 1064 nm laser wavenumber is set using the Stokes/anti-Stokes bands from a standard sample such as sulfur with Raman bands at ±479 cm−1 or even better the analyte itself (but the anti-Stokes bands are very weak beyond ca. −800 cm−1). The laser excitation laser line is nominally at 9397.58 cm−1 (air) absolute frequency but can have daily variations of ±0.50 cm−1 or more.36 The laser frequency was adjusted until the agreement was within ±0.20 cm−1, i.e., less than the inherent line widths of room temperature solids. 2.d. Sample Preparation. Acetol is notorious for adsorbing water vapor from the atmosphere and is typically delivered with ca. 10% H2O impurity. Although there are still very trace H2O vapors in our CaSO4-dried sample measurements, we believe the mid-IR spectra also represent those of the largely H2O-free monomer for the following reasons: First, our spectra show none of the features ascribed to acetol dimers or trimers in the molecular beam studies of Sharma et al. as seen in Figure 3 of their work.30 Second, the residual water content was very small (0.54% for the 25 °C data, 0.76% for the 50 °C spectra) representing only a small mole fraction. Third, the 25 and 50 °C spectra are both composite spectra, each derived from 12 or more individual measurements. When scaled for number density, the two spectra are nearly identical: Were there either water complex or oligomer formation, the formation kinetics would likely show temperature dependencies with the corresponding disappearance or formation of certain bands. Fourth, on the static system the acetol (Alfa Aesar, 95%) was chemically dried; a first unsuccessful attempt used B2O3 that resulted in purplish then red-brown gelatinous material. An alternative method dried the liquid extensively over CaSO4 (Drierite), with subsequent liquid nitrogen freeze−pump−thaw cycles to remove H2O and CO2 vapors. Fifth, the results are in reasonable agreement with preliminary (undried) acetol measurements reported earlier on the NWIR database.23 With significant water, those measurements used a disseminator apparatus and a long-path cell set to either 25 or 50 °C. The samples were introduced to the cell via an N2 gas stream, samples flow from a syringe tip atop a disseminator block that is heated37 to ca. 100 °C for the acetol measurements, and the elevated temperature would likely dissociate any oligomers upon N2 dilution. Sixth, the number of observed IR bands is consistent with both the GAMESS and MP2-predicted spectra. Water complexes or dimer/trimer formation would give rise to additional bands as both predicted and observed30 in the work of Sharma et al. Finally, our gas-phase monomer measurements match well those of Sharma et al.29 and Tuazon and Atkinson14 as well as the liquid-phase data of Mohaček-Grošev,28 most of whom removed water from the sample by pumping.

bar source, an extended range KBr beamsplitter, and a photoresistive mercury cadmium telluride (MCT) detector were used to acquire the raw interferograms. Multiple burdens of HA were measured and each single channel spectrum (600− 6500 cm−1) was derived from 256 averaged interferograms, each acquired at 0.112 cm−1 resolution, and was obtained using a Fourier transform with Mertz phase correction, boxcar apodization, and a zerofill factor of 2. Measurements were made using a 2 in. diameter gold-plated stainless manifold which, along with the sample cell, was evacuated to CH2 rock and CCC 5999


Article


Figure 4. Experimental liquid-phase infrared (top, black) gas-phase mid-IR (middle, red) and SQM theory predicted (bottom traces, A′ mode in blue, A″ modes in lilac) for the IR active bands. The liquid-phase spectrum was scaled by a factor of 50. Theoretical spectra are scaled by a factor of 40000 and vertically offset for clarity. The vapor-phase data are quantitative on the y-axis.

(C−O stretch) at 1107.53 cm−1 or the strong carbonyl stretch CO stretch ν5 at 1741.49 cm−1, respectively. Unlike the farinfrared, the C-type A″ modes are all quite weak in the fingerprint region (both per prediction and per experiment), but the SQM predicted frequencies are still of use. For example, the weak band near 1073 cm−1 (on the red shoulder of the stronger 1107.53 cm−1 band) has a weak A/C-type profile and is assigned as the A″ mode ν22. Similarly, the very weak band at 1231.93 is easily assigned as the A″ mode ν21, also based on SQM-predicted frequencies. The A′ modes ν7 and ν8 are predicted to be quite close (SQM theory 1440 and 1429 cm−1) but experimentally are observed with a greater separation at 1446.53 and 1411.35 cm−1. The band at 974.04 cm−1 is approximately 10 times weaker than ν12 and is distorted. We assigned it as ν13, corresponding to CC stretch combined with the in-plane CCH bend. The qualitative comparison of the region we show in Figure 3 with the spectrum shown by Jetzki et al.,46 Figure 5, is very good. The ν10 band at 1288.78 cm−1 has an A-type profile, matches perfectly the band pointed out by Jetzki et al. in their paper, and is an A′ mode band. It is also seen in the liquid phase at 1286 cm−1, matching the band seen by Mohaček-Grošev28 at 1287 cm−1. Good agreement is also seen when the assignments in this region are compared with those of Sharma et al.29 for the Ar matrix. We also observe a band at 1361.84 cm−1, assigned as a difference band ν5 − ν25. Mohaček-Grošev28 did not observe ν6 to ν8; she assigned but did not observe a CH2 deformation as an A″ band type at 1362 cm−1 (calculated) and the CH3 + CC deformation as an A′ mode band at 768 cm−1, which is not in agreement with the assignment in Table 1. The assignment of the C−H stretching modes also presents some challenges: Figure 4 presents the liquid IR (top, black),

gas-phase IR (middle, red), and also the SQM theory predicted (bottom traces, A′ mode in blue, A″ modes in lilac) for the infrared active bands. The assignment of ν2, the highest frequency C−H stretch is obvious as it is the highest frequency gas-phase band at 3022 cm−1, shifting to a clearly resolved band at 2989 cm−1 in the liquid IR spectrum. This leaves four remaining C−H stretches to assign: With the gas-phase infrared as the more reliable indicator data, the gas-phase mode at 2885 cm−1 is clearly a C-type A″ band and assigned as ν19, the lower frequency of the two A″ modes, at nearly the same frequency in liquid IR as in the gas-phase IR. This would mean the higher frequency A″ mode ν18 manifests itself as the very weak band at 2973 cm−1 in the vapor-phase IR spectrum, but in the liquid phase its intensity increases and is observed as a shoulder near 2939 cm−1 on the high-energy side of the liquid-phase unresolved triplet. The two remaining C−H fundamentals ν3 and ν4 are then ascribed to the vapor-phase shoulder at 2917 cm−1 and B-type band at 2862 cm−1, respectively. In the liquidphase infrared spectrum these are shifted to the center of the triplet at 2921 and the broad shoulder seen at ∼2842 cm−1, respectively. The other B-type band near 2803 cm−1 is assigned as the overtone 2ν8. The B-type profiles of ν4 and 2ν8 are consistent with Cs symmetry as also seen for the band shape of, e.g., the ν5 CO stretch at 1741.49 cm−1. Our assignments are also consistent with MP2-ccpvtz level theory as ν3 and ν4 are predicted to be the two strongest of the C−H fundamentals. Comparing with previous studies, we have found not only good agreement with some of the assignments of Sharma et al.29 in this region but also some differences. For ν3, for instance, we assigned it to be the CH3 symmetric stretching at 2917 cm−1, whereas Sharma et al. observed it at 2914 cm−1 (good agreement) and assigned it to be the CH2 asymmetric 6000


Article

The Journal of Physical Chemistry A stretching. They assigned the CH3 symmetric stretching to the DFT calculated value of 2987 cm−1 but failed to observe it. Our CH2 asymmetric stretching is assigned to the C-type band ν19 that we observed at 2885 cm−1. Mohaček-Grošev28 did not observe ν3, ν18, and ν19 but assigned the vibrations based on the theoretical calculations. Using these values, we believe the values in Table 1 represent the first complete vibrational assignment of hydroxyacetone. The work of Mohaček−Grošev28 is also quite complete but does not contain any far-infrared data and is limited to liquidphase studies; still, her assignments are similar to those in the present study and her mode descriptions denote the symmetric and antisymmetric descriptions of the Cs point group. In our assignments we also have used primarily the Cs point group as we considered the different acetol conformers: The methyl rotation barrier is 1 kT at room temperature so all conformers 3 will be sampled more or less equally, meaning either conformer used for assigning the spectrum should be as valid as any other. The choice of a C1 (no symmetry) conformer is neither right nor wrong, as is the choice of assignment based on a Cs planar conformer, which is consistent with some mid-IR features and several far-IR features.

Table 2. Integrated Band Intensities of Hydroxyacetone integrated band intensity (cm/molecule) × 1018 band integration limits (cm−1)

∫ 298K

∫ 323K

% diff = (∫ 298K − ∫ 323K)/∫ 298K

3620−3400 3100−2700 1850−1650 1650−1335 1335−1245 1245−1160 1160−1030 1030−910 880−760 760−665 665−570

8.38 10.34 17.72 12.23 7.77 2.26 18.04 1.96 1.18 1.09 1.42

8.47 10.41 18.11 12.10 7.79 2.21 18.39 1.90 1.15 1.01 1.35

−1.12 −0.70 −2.20 1.12 −0.36 2.13 −1.96 3.40 2.68 7.57 4.80

1030−910 cm−1 the integrated band intensity is slightly higher (3.4%) in the 298 K spectrum than it is in the 323 K spectrum; this range includes the ν23 fundamental with which a combination band at ∼968 cm−1 overlaps. Similarly, one of the largest differences arises for the combination bands in the 760−665 cm−1 integration range, which have a 7.6% greater intensity in the 298 K spectrum than in the 323 K spectrum. This is likely due to larger fractional effects for weaker bands due to the challenges in baseline fitting in regions with weaker signal/noise ratios, rather than any actual temperature dependencies. In general, as expected, the two spectra compare very well; there is no evidence of shifted peaks, and quantitative band integral differences are small. If, for the 298 K data, one sums the areas of all the band integrals the total is 82.39, as opposed to a sum of 82.89 for the 323 K data. This is a difference of only 0.6%, showing good agreement and a value far less than the estimated experimental uncertainties.24,32 This is also the expected result as the total integrated intensities for a molecule’s infrared spectrum should be conserved when the temperature of the sample is changed: Not only are higher rotational states within a given band more populated with increasing temperature at the cost of lower rotational states, but also the population is shifted to higher vibrational modes with increases in temperature. That is, intensity is shifted out of fundamental modes and other bands that originate in the ground state and moved into hot bands as has been shown by Chackerian et al.50 For atmospheric monitoring, several bands from the fingerprint region of the spectrum (Figure 3) could be used. The ν5 1741.36 cm−1 (CO stretch) band is obviously strong but would be obscured by the lines of the H2O bending mode in most sensing scenarios. However, the strong ν12 band at 1107.53 cm−1 is a good candidate for trace detection in the LWIR window. The differential cross section of this band is approximately 0.7 × 10−3 (ppm m)−1, meaning that in a 1 m path an optical system with a noise level less than 0.7 × 10−3 OD will detect 1 ppm of HA, and a noise level of 0.7 × 10−6 (achievable in most modern laser systems) would be able to detect 1 ppb of HA through 1 m.

4. RESULTS AND DISCUSSIONS: INTEGRATED BAND STRENGTHS In their study of the OH-initiated oxidation of methacrolein to hydroxyacetone and other products, Orlando et al.15 used quantitative IR spectroscopy to quantify the reagents and other products but did not explicitly report the reference HA spectrum, only observing that it agreed with the values of Tuazon and Atkinson.14 The Tuazon and Atkinson study does report a quantitative HA spectrum in Figure 2, from which we can estimate cross sections. The agreement between the present data and their spectrum (implicitly with that of Orlando et al.) is excellent: Using the scale indicated by Tuazon and Atkinson14 in their Figure 2, we calculated the intensities of their two strong bands corresponding to our ν5 and ν12. The agreement with our experimental results is very good, within 2%. We note that both the above papers used their internally generated reference spectra to quantify the production rate of HA in methacrolein photolysis. However, the band strength data reported here represent the first comprehensive quantitative infrared cross sections of vapor-phase hydroxyacetone. The mid-infrared quantitative spectra of HA reported here are pressure broadened to 760 Torr, and were recorded at a resolution sufficient to resolve the natural bandwidths, and are thus optimized for tropospheric monitoring. The integrated intensities shown in Table 2 were obtained using the OPUS software (v 5.5), with integration method B; this method uses the area under the peak using a simple straight baseline drawn between the two integration limits. The composite spectra are in fact recorded at 298 and 323 K but are scaled to correspond to an optical path of 1 m, with an analyte concentration of 1 ppmv scaled to a number density corresponding to a temperature of 296 K and 1 atm pressure. The integrated band intensities reported in Table 2 have been converted to the more common units of cm/molecule using Napierian (loge) absorbances. The integration domains are specified in the table and are given in cm−1. As seen in Table 2, the integrated band strengths for the two temperatures agree to within 3% for most of the measured bands with but a few exceptions. In the band integration range

5. SUMMARY We have updated the PNNL database with the spectra of “dry” hydroxyacetone and reported seminal quantitative gas-phase IR absorption band strengths that allow for quantitation of this species. We have expanded the quantitative aspect of the NWIR 6001


Article


sions from Indonesian, African, and Other Fuels. J. Geophys. Res. 2003, 108 (D23), 4719−4732. (6) Guenther, A. B.; Jiang, X.; Heald, C. L.; Sakulyanontvittaya, T.; Duhl, T.; Emmons, L. K.; Wang, X. The Model of Emissions of Gases and Aerosols from Nature Version 2.1 (MEGAN2.1): an Extended and Updated Framework for Modeling Biogenic Emissions. Geosci. Model Dev. 2012, 5, 1471−1492. (7) Paulot, F.; Crounse, J. D.; Kjaergaard, H. G.; Kürten, A.; St. Clair, J. M.; Seinfeld, J. H.; Wennberg, P. O. Unexpected Epoxide Formation in the Gas-phase Photooxidation of Isoprene. Science 2009, 325, 730− 733. (8) Brauer, C. S.; Blake, T. A.; Guenther, A. B.; Sharpe, S. W.; Sams, R. L.; Johnson, T. J. Quantitative Infrared Absorption Cross Sections of Isoprene for Atmospheric Measurements. Atmos. Meas. Tech. 2014, 7, 3839−3847. (9) Tuazon, E. C.; Atkinson, R. A Product Study of the Gas-phase Reaction of Isoprene with the OH Radical in the Presence of NOx. Int. J. Chem. Kinet. 1990, 22, 1221−1236. (10) Paulson, S. E.; Flagan, R. C.; Seinfeld, J. H. Atmospheric Photooxidation of Isoprene Part 1: The Hydroxyl Radical and Ground State Atomic Oxygen Reactions. Int. J. Chem. Kinet. 1992, 24, 79−101. (11) Grosjean, D.; Williams, E. L., II; Grosjean, E. Atmospheric Chemistry of Isoprene and of its Carbonyl Products. Environ. Sci. Technol. 1993, 27, 830−840. (12) Montzka, S. A.; Trainer, M.; Goldan, P. D.; Kuster, W. C.; Fehsenfeld, F. C. Isoprene and its Oxidation Products, Methyl Vinyl Ketone and Methacrolein, in the Rural Troposphere. J. Geophys. Res. 1993, 98 (D1), 1101−1111. (13) Tuazon, E. C.; Atkinson, R. A Product Study of the Gas-phase Reaction of Methyl Vinyl Ketone with the OH Radical in the Presence of NOx. Int. J. Chem. Kinet. 1989, 21, 1141−1152. (14) Tuazon, E. C.; Atkinson, R. A Product Study of the Gas-phase Reaction of Methacrolein with the OH Radical in the Presence of NOx. Int. J. Chem. Kinet. 1990, 22, 591−602. (15) Orlando, J. J.; Tyndall, G. S.; Paulson, S. E. Mechanism of the OH-initiated Oxidation of Methacrolein. Geophys. Res. Lett. 1999, 26, 2191−2194. (16) Gierczak, T.; Burkholder, J. B.; Talukdar, R. K.; Mellouki, A.; Barone, S. B.; Ravishankara, A. R. Atmospheric Fate of Methyl Vinyl Ketone and Methacrolein. J. Photochem. Photobiol., A 1997, 110, 1−10. (17) Paulot, F.; Crounse, J. D.; Kjaergaard, H. G.; Kroll, J. H.; Seinfeld, J. H.; Wennberg, P. O. Isoprene Photooxidation: New Insights into the Production of Acids and Organic Nitrates. Atmos. Chem. Phys. 2009, 9, 1479−1501. (18) Orlando, J. J.; Tyndall, G. S.; Fracheboud, J.-M.; Estupiñan, E. G.; Haberkorn, S.; Zimmer, A. The Rate and Mechanism of the Gasphase Oxidation of Hydroxyacetone. Atmos. Environ. 1999, 33, 1621− 1629. (19) Dibble, T. S. Intramolecular Hydrogen Bonding and Double Hatom Transfer in Peroxy and Alkoxy Radicals from Isoprene. J. Phys. Chem. A 2004, 108, 2199−2207. (20) Dibble, T. S. Prompt Chemistry of Alkenoxy Radical Products of the Double H-atom Transfer of Alkoxy Radicals from Isoprene. J. Phys. Chem. A 2004, 108, 2208−2215. (21) Butkovskaya, N. I.; Pouvesle, N.; Kukui, A.; Mu, Y. J.; Le Bras, G. Mechanism of the OH-initiated Oxidation of Hydroxyacetone Over the Temperature Range 236−298 K. J. Phys. Chem. A 2006, 110, 6833−6843. (22) Sharpe, S. W.; Johnson, T. J.; Sams, R. L.; Chu, P. M.; Rhoderick, G. C.; Johnson, P. A. Gas-phase Databases for Quantitative Infrared Spectroscopy. Appl. Spectrosc. 2004, 58, 1452−1461. (23) Johnson, T. J.; Profeta, L. T. M.; Sams, R. L.; Griffith, D. W. T.; Yokelson, R. L. An Infrared Spectral Database for Detection of Gases Emitted by Biomass Burning. Vib. Spectrosc. 2010, 53, 97−102. (24) Kochanov, R. V.; Gordon, I. E.; Rothman, L. S.; Sharpe, S. W.; Johnson, T. J.; Sams, R. L. Comment on “Radiative Forcings for 28 Potential Archean Greenhouse Gases” by Byrne and Goldblatt (2014). Clim. Past 2015, 11, 1097−1105.

library to estimate acetol atmospheric detectability and detection limits; the infrared absorption band strengths reported here can be used for biogenic fluxes, biomass burning atmospheric monitoring, and other applications. In addition, we have improved the infrared vibrational frequency assignments for hydroxyacetone. To this end we have complemented the quantitative PNNL mid-IR data with additional experimental data including gas-phase far-infrared spectra and liquid-phase mid-IR and Raman data, along with anharmonic calculations for the theoretical vibrational frequencies. We have combined these results with ab initio theoretical methods, to obtain a more complete vibrational analysis of this atmospherically important species. Our results and band assignments are well confirmed by both ab initio MP2-ccpvtz calculations and GAMESS (B3LYP) theoretical calculations. The theoretical calculations indicate that the methyl group is rotating nearly freely at room temperature, but that the intramolecular hydrogen bond creates an otherwise planar structure and that it is appropriate to label the vibrational modes under the symmetry labels of the Cs point group.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.6b05045. Optimized geometries, dipoles, methyl dihedral angles, and methyl torsion harmonic frequencies (TXT)

■ ■

AUTHOR INFORMATION

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported by the Department of Defense’s Strategic Environmental Research and Development Program (SERDP), resources conservation project RC-2640 as well as the U.S. Department of Energy, National Nuclear Security Administration, Office of Defense Nuclear Nonproliferation R&D (NA-22). We gratefully thank both sponsors for their support. PNNL is operated for the U.S. Department of Energy by the Battelle Memorial Institute under contract DE-AC0676RLO 1830.

■

REFERENCES

(1) Akagi, S. K.; Yokelson, R. J.; Wiedinmyer, C.; Alvarado, M. J.; Reid, J. S.; Karl, T.; Crounse, J. D.; Wennberg, P. O. Emission Factors for Open and Domestic Biomass Burning for Use in Atmospheric Models. Atmos. Chem. Phys. 2011, 11, 4039−4072. (2) Akagi, S. K.; Craven, J. S.; Taylor, J. W.; McMeeking, G. R.; Yokelson, R. J.; Burling, I. R.; Urbanski, S. P.; Wold, C. E.; Seinfeld, J. H.; Coe, H.; et al. Evolution of Trace Gases and Particles Emitted by a Chaparral Fire in California. Atmos. Chem. Phys. 2012, 12, 1397−1421. (3) Akagi, S. K.; Burling, I. R.; Mendoza, A.; Johnson, T. J.; Cameron, M.; Griffith, D. W. T.; Paton-Welsh, C.; Weise, D. R.; Reardon, J.; Yokelson, R. J. Field Measurements of Trace Gases Emitted by Prescribed Fires in Southeastern US Pine Forests Using an Open-path FTIR System. Atmos. Chem. Phys. 2014, 14, 199−215. (4) Yokelson, R. J.; Burling, I. R.; Gilman, J. B.; et al. Coupling Field and Laboratory Measurements to Estimate the Emission Factors of Identified and Unidentified Trace Gases for Prescribed Fires. Atmos. Chem. Phys. 2013, 13, 89−116. (5) Christian, T. J.; Kleiss, B.; Yokelson, R. J.; Holzinger, R.; Crutzen, P. J.; Hao, W. M.; Saharjo, B. H.; Ward, D. E. Comprehensive Laboratory Measurements of Biomass-burning Emissions: 1. Emis6002


Article

The Journal of Physical Chemistry A (25) Profeta, L. T. M.; Sams, R. L.; Johnson, T. J.; Williams, S. D. Quantitative Infrared Intensity Studies of Vapor-phase Glyoxal, Methylglyoxal, and 2,3-Butanedione (diacetyl) with Vibrational Assignments. J. Phys. Chem. A 2011, 115, 9886−9900. (26) Johnson, T. J.; Sams, R. L.; Profeta, L. T. M.; Akagi, S. K.; Burling, I. R.; Yokelson, R. J.; Williams, S. D. Quantitative IR Spectrum and Vibrational Assignments for Glycolaldehyde Vapor: Glycolaldehyde Measurements in Biomass Burning Plumes. J. Phys. Chem. A 2013, 117, 4096−4107. (27) Williams, S. D.; Johnson, T. J.; Sharpe, S. W.; Yavelak, V.; Oates, R. P.; Brauer, C. S. Quantitative Vapor-phase IR Intensities and DFT Computations to Predict Absolute IR Spectra Based on Molecular Structure: I. Alkanes. J. Quant. Spectrosc. Radiat. Transfer 2013, 129, 298−307. (28) Mohaček-Grošev, V. Vibrational Analysis of Hydroxyacetone. Spectrochim. Acta, Part A 2005, 61, 477−484. (29) Sharma, A.; Reva, I.; Fausto, R. Matrix-insolation Study and Ab Initio Calculations of the Structure and Spectra of Hydroxyacetone. J. Phys. Chem. A 2008, 112, 5935−5946. (30) Sharma, A.; Reva, I.; Fausto, R.; Hesse, S.; Xue, Z.; Suhm, M. A.; Nayak, S. K.; Sathishkumar, R.; Pal, R.; Guru Row, T. N. Conformation-changing Aggregation in Hydroxyacetone: A Combined Low-temperature FTIR, Jet, and Crystallographic Study. J. Am. Chem. Soc. 2011, 133, 20194−20207. (31) Sharma, A.; Reva, I.; Fausto, R. Conformational Switching Induced by Near-infrared Laser Irradiation. J. Am. Chem. Soc. 2009, 131, 8752−8753. (32) Johnson, T. J.; Sams, R. L.; Blake, T. A.; Sharpe, S. W.; Chu, P. M. Removing Aperture-induced Artifacts from Fourier Transform Infrared Intensity Values. Appl. Opt. 2002, 41 (15), 2831−2839. (33) Bonen, D.; Johnson, T. J.; Sarkar, S. Characterization of Principal Clinker Minerals by FT-Raman Microspectroscopy. Cem. Concr. Res. 1994, 24 (5), 959−965. (34) Smith, M. W.; Dallmeyer, I.; Johnson, T. J.; Brauer, C. S.; McEwen, J.-S.; Espinal, J. F.; Garcia-Perez, M. Structural Analysis of Char by Raman Spectroscopy: Improving Band Assignments Through Computational Calculations from First Principles. Carbon 2016, 100, 678−692. (35) Kunkel, B. M.; Su, Y.-F.; Tonkyn, R. G.; Stephan, E. G.; Joly, A. G.; Birnbaum, J. C.; Jarman, K. H.; Johnson, T. J. Raman Database Considerations for Near-infrared Systems. Proc. SPIE 2011, 8189, 818905−1−818905−9. (36) Williams, S. D.; Johnson, T. J.; Gibbons, T.; Kitchens, C. L. Relative Raman Intensities in C6H6, C6D6, and C6F6: A Comparison of Different Computational Methods. Theor. Chem. Acc. 2007, 117 (2), 283−290. (37) Johnson, T. J.; Sams, R. L.; Burton, S. D.; Blake, T. A. Absolute Integrated Intensities of Vapor-phase Hydrogen Peroxide (H2O2) in the Mid-infrared at Atmospheric Pressure. Anal. Bioanal. Chem. 2009, 395, 377−386. (38) Stanton, J. F.; Gauss, J.; Harding, M. E.; Szalay, P. G. with contributions from Auer, A. A.; Bartlett, R. J.; Benedikt, U.; Berger, C.; Bernholdt, D. E.; Bomble, Y. J.; et al. CFOUR, a Quantum Chemical Program Package. http//cfour.de. (39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; et al. Gaussian 09, Revision B.01; Gaussian, Inc.: Wallingford, CT, 2009. (40) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A. General Atomic and Molecular Electronic Structure System. J. Comput. Chem. 1993, 14, 1347−1363. Gordon, M. S.; Schmidt, M. W. Advances in Electronic Structure Theory: GAMESS a Decade Later. In Theory and Applications of Computational Chemistry: the First Forty Years; Dykstra, C. E., Frenking, G., Kim, K. S., Scuseria, G. E., Eds.; Elsevier: Amsterdam, 2005; pp 1167−1189. (41) PQS version 4.0; Parallel Quantum Solutions: 2013 Green Acres Road, Fayetteville, Arkansas 72703; http://www.pqs-chem.com. (42) Baker, J.; Jarzecki, A. A.; Pulay, P. Direct Scaling of Primitive Valence Force Constants: An Alternative Approach to Scaled Quantum Mechanical Force Fields. J. Phys. Chem. A 1998, 102, 1412−1424.

(43) Pulay, P.; Török, F. On the Parameter Form of the Force Constant Matrix II: Investigation of the Assignment with the Aid of the Parameter Form. Acta. Chim. Acad. Sci. Hung. 1965, 47, 273−279. (44) Matthews, D. A.; Stanton, J. F. Quantitative Analysis of Fermi Resonances by Harmonic Derivatives of Perturbation Theory Corrections. Mol. Phys. 2009, 107, 213−222. (45) Blom, C. E.; Altona, C. Application of Self-consistent-field Ab Initio Calculations to Organic Molecules 2: Scale Factor Method for Calculation of Vibrational Frequencies from Ab Initio Force Constants − Ethane, Propane, and Cyclopropane. Mol. Phys. 1976, 31, 1377− 1391. (46) Jetzki, M.; Luckhaus, D.; Signorell, R. Fermi Resonance and Conformation in Glycolaldehyde Particles. Can. J. Chem. 2004, 82, 915−924. (47) Kattija-Ari, M.; Harmony, M. D. The Microwave Spectrum and Conformation of Hydroxyacetone: The Influence of Hydrogen Bonding on the Barrier to Internal Rotation of the Methyl Group. Int. J. Quantum Chem. 1980, 18, 443−453. (48) Bond, M. R.; Johnson, T. J.; Willett, R. D. Structural, Spectroscopic, and Electron Paramagnetic Resonance Studies on Two Solid Phases of bis-dipropylammonium tetrachlorocuprate(II) (DPACC). Can. J. Chem. 1988, 66, 963−973. (49) Bell, S. Ab Initio Study of the Barriers to Methyl Torsion and Torsional Frequencies of Acetyl Molecules. Spectrochim. Acta, Part A 2005, 61, 1471−1477. (50) Chackerian, C.; Sharpe, S. W.; Blake, T. A. Anhydrous Nitric Acid Integrated Absorption Cross Sections: 820−5300 cm−1. J. Quant. Spectrosc. Radiat. Transfer 2003, 82, 429−441.

6003


Assignment of the Fundamental Modes of Hydroxyacetone Using Gas

Recommend Documents