Forests emit large quantities of volatile organic compounds (VOCs) to the atmosphere. Their condensable oxidation products can form secondary organic aerosol, a significant and ubiquitous component of atmospheric aerosol, which is known to affect the Earth's radiation balance by scattering solar radiation and by acting as cloud condensation nuclei. The quantitative assessment of such climate effects remains hampered by a number of factors, including an incomplete understanding of how biogenic VOCs contribute to the formation of atmospheric secondary organic aerosol. The growth of newly formed particles from sizes of less than three nanometres up to the sizes of cloud condensation nuclei (about one hundred nanometres) in many continental ecosystems requires abundant, essentially non-volatile organic vapours, but the sources and compositions of such vapours remain unknown. Here we investigate the oxidation of VOCs, in particular the terpene α-pinene, under atmospherically relevant conditions in chamber experiments. We find that a direct pathway leads from several biogenic VOCs, such as monoterpenes, to the formation of large amounts of extremely low-volatility vapours. These vapours form at significant mass yield in the gas phase and condense irreversibly onto aerosol surfaces to produce secondary organic aerosol, helping to explain the discrepancy between the observed atmospheric burden of secondary organic aerosol and that reported by many model studies. We further demonstrate how these low-volatility vapours can enhance, or even dominate, the formation and growth of aerosol particles over forested regions, providing a missing link between biogenic VOCs and their conversion to aerosol particles. Our findings could help to improve assessments of biosphere-aerosol-climate feedback mechanisms, and the air quality and climate effects of biogenic emissions generally.

We present a hypothesis that autoxidation (inter- and intramolecular hydrogen abstraction by peroxy radicals) plays an important role in the oxidation of organic compounds in the atmosphere, particularly organic matter associated with aerosol. In the laboratory, we determine the rate of this process at room temperature for a model system, 3-pentanone. We employ ab initio calculations to investigate H-shifts within a broader group of substituted organic compounds. We show that the rate of abstraction of hydrogen by peroxy radicals is largely determined by the thermochemistry of the nascent alkyl radicals and thus is highly influenced by neighboring substituents. As a result, autoxidation rates increase rapidly as oxygen-containing functional groups (carbonyl, hydroxy, and hydroperoxy) are added to organic compounds. This mechanism is consistent with formation of the multifunctional hydroperoxides and carbonyls often found in atmospheric aerosol particles.

The prompt formation of highly oxidized organic compounds in the ozonolysis of cyclohexene (C6H10) was investigated by means of laboratory experiments together with quantum chemical calculations. The experiments were performed in borosilicate glass flow tube reactors coupled to a chemical ionization atmospheric pressure interface time-of-flight mass spectrometer with a nitrate ion (NO3(-))-based ionization scheme. Quantum chemical calculations were performed at the CCSD(T)-F12a/VDZ-F12//ωB97XD/aug-cc-pVTZ level, with kinetic modeling using multiconformer transition state theory, including Eckart tunneling corrections. The complementary investigation methods gave a consistent picture of a formation mechanism advancing by peroxy radical (RO2) isomerization through intramolecular hydrogen shift reactions, followed by sequential O2 addition steps, that is, RO2 autoxidation, on a time scale of seconds. Dimerization of the peroxy radicals by recombination and cross-combination reactions is in competition with the formation of highly oxidized monomer species and is observed to lead to peroxides, potentially diacyl peroxides. The molar yield of these highly oxidized products (having O/C > 1 in monomers and O/C > 0.55 in dimers) from cyclohexene ozonolysis was determined as (4.5 ± 3.8)%. Fully deuterated cyclohexene and cis-6-nonenal ozonolysis, as well as the influence of water addition to the system (either H2O or D2O), were also investigated in order to strengthen the arguments on the proposed mechanism. Deuterated cyclohexene ozonolysis resulted in a less oxidized product distribution with a lower yield of highly oxygenated products and cis-6-nonenal ozonolysis generated the same monomer product distribution, consistent with the proposed mechanism and in agreement with quantum chemical modeling.

Peroxy radicals formed by addition of OH and O 2 to the olefinic carbon atoms in methacrolein react with NO to form methacrolein hydroxy nitrate and hydroxyacetone. We observe that the ratio of these two compounds, however, unexpectedly decreases as the lifetime of the peroxy radical increases. We propose that this results from an isomerization involving the 1,4-H-shift of the aldehydic hydrogen atom to the peroxy group. The inferred rate (0.5 ± 0.3 s −1 at T = 296 K) is consistent with estimates obtained from the potential energy surface determined by high level quantum calculations. The product, a hydroxy hydroperoxy carbonyl radical, decomposes rapidly, producing hydroxyacetone and re-forming OH. Simulations using a global chemical transport model suggest that most of the methacrolein hydroxy peroxy radicals formed in the atmosphere undergo isomerization and decomposition.

A model for predicting the intelligibility of processed noisy speech is proposed. The speech-based envelope power spectrum model has a similar structure as the model of Ewert and Dau [(2000). J. Acoust. Soc. Am. 108, 1181-1196], developed to account for modulation detection and masking data. The model estimates the speech-to-noise envelope power ratio, SNR(env), at the output of a modulation filterbank and relates this metric to speech intelligibility using the concept of an ideal observer. Predictions were compared to data on the intelligibility of speech presented in stationary speech-shaped noise. The model was further tested in conditions with noisy speech subjected to reverberation and spectral subtraction. Good agreement between predictions and data was found in all cases. For spectral subtraction, an analysis of the model's internal representation of the stimuli revealed that the predicted decrease of intelligibility was caused by the estimated noise envelope power exceeding that of the speech. The classical concept of the speech transmission index fails in this condition. The results strongly suggest that the signal-to-noise ratio at the output of a modulation frequency selective process provides a key measure of speech intelligibility.

S1.0 CIMS Sensitivities CIMS sensitivities to the oxidation products were determined in multiple ways. Hydroxyacetone and glycolaldehyde are commercially available and were quantified gravimetrically and by Fourier Transform Infrared Spectroscopy (FT-IR) for CIMS calibration.1 Uncalibrated compounds (glycolic acid and all products identified by m/z) were assigned a generic CIMS sensitivity of 2.5×10 -4 ncts /pptv, and are considered accurate to within a factor of 2. Here, normalized counts (ncts) represent the counts observed at the analyte m/z divided by the reagent ion counts. OH + ISOPOOH à ISOPOOH-OH à ProductsIn Figures Figure S1. The different reaction pathways for the reaction between (1,2)-ISOPOOH and OH radical. Figure S2. The different reaction pathways for the reaction between (4,3)-ISOPOOH and OH radical. 11,12 The general reaction scheme is shown in Figure S3. Figure S3. The reaction scheme as used in the MESMER model (Only an illustration, not the energetically correct picture of the reactions). The ISOPOOH-OH complexes are different for each of the reaction pathways even though they are given with the same energy at this figure. S5 S6 S7In our Mesmer modeling the Lennard-Jones (L-J) parameters of the bath gas were chosen to be a nitrogen gas resembling the atmospheric gas ISOPOOH + OH ISOPOOH-OH complex TS TS TSAbstraction trans-Add1 cis-Add1Add2 trans-IEPOX cis-IEPOX S9We have preformed a sensitivity test of Mesmer input parameters. In our sensitivity test we used three collisional activation/deactivation energies of 50, 100 and 200 cm -1 and two different grain sizes of 25 and 50 cm -1 . We did not observe any significant changes in the reaction rate constants (only changes of a few percent). We have also tested the system with different sizes of grain span, e.g., 10 kT, 20 kT, 30 kT, 40 kT and 50 kT. If a grain span of 30 kT or higher is used, the reaction rate constants do not change. We have therefore used a grain size of 30 kT.The reaction rate constants are sensitive to the choice of the Arrhenius pre-exponential factor (A). Each reaction pathway is a separate Mesmer calculation (See Figure S4 and S5 for the individual reaction pathways) -we have not coupled between the reactions in the fitting of the Arrhenius pre-exponential factor (A). We treat the pre-exponential factor as temperature independent and it is varied between 1.0×10 -12 and 2.0×10 -10 cm 3 molecule -1 s -1 . We use nine different Arrhenius pre-exponential factors to calculate the rates. Three of the factors are from the three reactions of n-butane, 3-methyl-3-butene-1-ol and 1-butene with OH. [14][15][16] The total reaction rate constants (OH +ISOPOOH → Products) of the (1,2)-ISOPOOH and (4,3)-ISOPOOH systems are shown in Table S3 and Table S4, respectively. S3.3 (1,2)-ISOPOOHFor the (1,2)-ISOPOOH + OH reactions, the absolute rate constants of all the different reaction pathways increase with an increase in the Arrhenius pre-exponential factor, and the relative yields (in %) of the reaction pathways also change.The yiel...

Autoxidation by sequential peroxy radical hydrogen shifts (H-shifts) and O2 additions has recently emerged as a promising mechanism for the rapid formation of highly oxidized, low-volatility organic compounds in the atmosphere. A key prerequisite for autoxidation is that the H-shifts of the initial peroxy radicals formed by, e.g., OH or O3 oxidation are fast enough to compete with bimolecular sink reactions. In most atmospheric conditions, these restrict the lifetime of peroxy radicals to be on the order of seconds. We have systematically investigated all potentially important (nonmethyl, sterically unhindered) H-shifts of all four peroxy radicals formed in the ozonolysis of α-pinene using density functional (ωB97XD) and coupled cluster [CCSD(T)-F12] theory. In contrast to the related but chemically simpler cyclohexene ozonolysis system, none of the calculated H-shifts have rate constants above 1 s(-1) at 298 K, and most are below 0.01 s(-1). The low rate constants are connected to the presence of the strained cyclobutyl ring in the α-pinene-derived peroxy radicals, which hinders H-shifts both from and across the ring. For autoxidation to yield the experimentally observed highly oxidized products in the α-pinene ozonolysis system, additional ring-opening reaction mechanisms breaking the cyclobutyl ring are therefore needed. We further investigate possible uni- and bimolecular pathways for opening the cyclobutyl ring in the α-pinene ozonolysis system.

The speech-based envelope power spectrum model (sEPSM) presented by Jørgensen and Dau [(2011). J. Acoust. Soc. Am. 130, 1475-1487] estimates the envelope power signal-to-noise ratio (SNRenv) after modulation-frequency selective processing. Changes in this metric were shown to account well for changes of speech intelligibility for normal-hearing listeners in conditions with additive stationary noise, reverberation, and nonlinear processing with spectral subtraction. In the latter condition, the standardized speech transmission index [(2003). IEC 60268-16] fails. However, the sEPSM is limited to conditions with stationary interferers, due to the long-term integration of the envelope power, and cannot account for increased intelligibility typically obtained with fluctuating maskers. Here, a multi-resolution version of the sEPSM is presented where the SNRenv is estimated in temporal segments with a modulation-filter dependent duration. The multi-resolution sEPSM is demonstrated to account for intelligibility obtained in conditions with stationary and fluctuating interferers, and noisy speech distorted by reverberation or spectral subtraction. The results support the hypothesis that the SNRenv is a powerful objective metric for speech intelligibility prediction.