Supplement of Measurement Report: Important Contributions of Oxygenated Compounds to Emissions and Chemistry of Volatile Organic Compounds in Urban Air
Total Page:16
File Type:pdf, Size:1020Kb
Supplement of Atmos. Chem. Phys., 20, 14769–14785, 2020 https://doi.org/10.5194/acp-20-14769-2020-supplement © Author(s) 2020. This work is distributed under the Creative Commons Attribution 4.0 License. Supplement of Measurement report: Important contributions of oxygenated compounds to emissions and chemistry of volatile organic compounds in urban air Caihong Wu et al. Correspondence to: Bin Yuan ([email protected]) and Min Shao ([email protected]) The copyright of individual parts of the supplement might differ from the CC BY 4.0 License. 24 1. Sensitivity of VOCs in PTR-MS 25 Proton-transfer-reaction time-of-flight mass spectrometry (PTR-TOF-MS) allows 26 the detection of a large number of VOCs in air through proton-transfer reaction with 27 H3O+ reagent ions and detection by a mass spectrometer. Measurement sensitivities can 28 be experimentally determined using calibration gases, but only the sensitivity of few 29 species can be obtained. For other species, the sensitivity can be calculated using the 30 rate constant for the proton-transfer reaction. 31 In PTR-ToF-MS, VOCs that have higher proton affinity than water will be ionized + 32 via proton transfer with H3O to produce the product ions. + + 33 VOC+H3O → VOC·H +H2O 34 According to the reaction rate equation, the concentration of VOC•H can be 35 calculated as follows: + 36 VOC·H=H3O 0(1-exp(-[VOC]∆)) + + 37 Where k is the reaction rate constant, [H3O ]0 is the signal of H3O ions before 38 reaction, [VOC] is the number concentration of the VOC in the drift tube, and Δt is the + 39 reaction time for H3O traversing the drift tube. + 40 It is assumed that only a small amount of H3O has proton transfer reaction with 41 VOCs, the concentration of VOC•H can be calculated as follows: + 42 VOC·HH3O [VOC]∆ + + 43 Where [H3O ] is the signal of H3O ions after the end of the drift tube. 44 Considering the possibility of fragmentation for the protonated ions, a term + + 45 VOC∙H can be introduced to represent the fraction of product ions detected as VOC•H + 46 ions (0 ≤VOC∙H ≤1). The concentration of VOC·H becomes + + 47 VOC·HH3O [VOC]∆VOC∙H + 48 The fraction of H3O that is converted into VOC·H+ ions can be expressed as: VOC·H + 49 + [VOC]∆VOC∙H H3O 50 The signal of VOC·H is related to the transmission efficiency: 51 VOC·H+ VOC·H×VOC·H+ 52 The factors VOC·H+ is the transmission efficiencies for VOC·H . These are 2 53 determined by (i) the extraction efficiency of ions from the drift tube into the mass 54 analyzer, (ii) the transmission efficiency of the analyzer, and (iii) the detection 55 efficiency of the detector for each mass (Sekimoto et al., 2017). 56 The measured sensitivity in PTR-TOF-MS is defined as the ion intensity of 57 VOC·H obtained at a volume-mixing ratio of 1 ppbv (parts per billion by volume; 58 10−9): + + VOC·H + VOC·H + 59 Sensitivity 10 ∆H3O VOC∙H VOC + 10 H3O 60 For a certain VOC, the sensitivity is linearly related to the reaction rate constant. 61 Eq.3 suggests that if the measured sensitivity is corrected for fragmentation and 62 transmission efficiencies, the resulting sensitivity is linearly dependent on the reaction 63 rate constant k. This means that sensitivity can be calculated from the rate constants. + 64 Corrected Sensitivity Sensitivity⁄ /VOC∙H 65 The transmission efficiency of ions is quantified by laboratory experiments. A 66 large variety of high concentrations of VOCs, including methanol, acetonitrile, 67 ethylamine, formic acid, acetone, pyrrole, furan, isoprene, MACR, DMF, 68 hydroxyacetone, phenol, furfural, styrene, benzaldehyde, cresol, guaiacol, naphthalene, 69 pinene, bromobenzene, iodobenzene, octamethylcyclotetrasiloxane (D4), 1,3- 70 diiodobenzene, decamethylcyclopentasiloxane (D5), are selected for introduction into 71 the PTR-ToF-MS, reagent ions are obviously consumed. The transmission efficiency 72 can be obtained according to the decrease of reagent ions and the increase of product 73 ions. Figure S7 shows the transmission efficiency curves based on laboratory 74 experiments. 75 The fragment ratio was obtained by using the relative proportions relationship 76 between the signal intensity of VOC and the fragment ions generated by its fracture at 77 the same time during the transmission efficiency experiment. The fragment ratio of 78 different VOCs are shown in Table S2. 79 2 Calculation of initial isoprene concentration 80 Isoprene is extremely reactive and therefore its concentration is poor to indicate 81 direct biogenic emissions. To correct this atmospheric photochemical loss, we 3 82 extrapolated the isoprene concentration back to the source using measurements of 83 isoprene and its photo-oxidized products, methyl vinyl ketone (MVK) and methacrolein 84 (MACR) (Karl et al., 2009). The reaction process of isoprene being oxidized by OH 85 radicals in the atmosphere is: 86 isoprene OH → 0.32MVK 0.23MACR 87 110 88 MVK OH → Products 1.910 89 MACR OH → Products 3.310 . 90 1 ∆ 1 ∆ (1) 91 From the measured ratio between MVK + MACR and isoprene, the OH exposure 92 since emission can be determined, and the observed isoprene concentration can be 93 extrapolated back to the source(Figure S8c). 94 ∆ (2) 95 Here, we used the measured concentrations from online GC-MS/FID for isoprene, 96 as there are substantial interferences for PTR-TOF measurements of isoprene in urban 97 air (Yuan et al., 2017). MVK+MACR was only measured by PTR-ToF-MS. We 98 observed significant elevation of MVK+MACR concentrations in the evening when the 99 primary emissions are highest, indicating MVK+MACR concentrations measured by 100 PTR-ToF-MS are influenced by primary anthropogenic emissions (e.g. traffic), either 101 due to direct emissions of MVK or MACR from vehicles, or potential interferences 102 from other compounds that are emitted by vehicles. The data during the night time (20: 103 00-6: 00) was selected to determine the emission ratio of MVK + MACR relative to CO 104 (Figure S8a). MVK+MACR concentrations that are solely from biogenic sources can 105 be estimated by the following Eq. 3: 106 (3) 107 Where and are the concentration of MVK+MACR and 108 CO, respectively. is the tropospheric background of CO (100 ppb). -4 109 is the emission ratio of MVK+MACR versus CO (6.0×10 ppb [ppb 110 CO]-1). 4 111 3 Calculation of photolysis rates 112 To estimate photolysis rates for OVOCs, we follow the method in de Gouw et al. 113 (2018). It is assumed that the photolysis rates for all OVOCs are reduced relative to 114 their clear-sky rates by the same factor as for jNO2 and use the following equation: 115 (4) 、 116 Where are clear-sky photolysis rates of OVOC 117 and NO2, respectively. Clear-sky photolysis rates for different compounds are 118 calculated in this study from the parameterization used in the Master Chemical 119 Mechanism v3.3.1 (Saunders et al., 2003): 120 lcos sec (5) 121 Where χ is the solar zenith angle and l, m, and n are parameters listed for different 122 photolysis rates. The photolysis rates of NO2, H2O2, and HCHO were measured on-site 123 using a PFS-100 Photolysis Spectrometer (Focused Photonics Inc.) during the 124 campaign. The above methods were used to calculate the photolysis rates of H2O2 and 125 HCHO, and compared with the measurements. It is found that the calculated values 126 show good agreement with the measurements for the two compounds (Figure S9). 127 Based on this, the photolysis frequencies of other OVOCs were calculated and then the 128 corrections to the rate constants of OH reaction for these OVOCs were estimated based 129 on Eq. 3 in the main text. 130 5 131 Table S1. Sensitivities of PTR-ToF-MS for various VOC species calibrated with 132 standard gas and Liquid Calibration Unit (LCU). VOC species Ion formula Sensitivity, cps/ppb Species calibrated with gas standard + Formaldehyde CH2OH 1042 + Methanol CH4OH 629.3 + Acetonitrile C2H3NH 3374 + Acetaldehyde C2H4OH 2767 + Ethanol C2H6OH 99.23 + Acrolein C3H4OH 4107 + Acetone C3H6OH 4299 + Furan C4H4OH 2544 + Isoprene C5H8H 1888 + MVK C4H6OH 3868 + MEK C4H8OH 4467 + Benzene C6H6H 3151 + 2-Pentanone C5H10OH 4510 + Toluene C7H8H 3978 + Phenol C6H6OH 4076 + Furfural C5H4O2H 7460 + Methyl Isobutyl Ketone C6H12OH 3988 + Styrene C8H8H 4289 + O-xylene C8H10H 4241 + m-Cresol C7H8OH 4299 + 1,2,4-Teimethylbenzene C9H12H 4413 + Naphthalene C10H8H 5117 + a-Pinene C10H16H 2332 Species calibrated with the Liquid Calibration Unit (LCU). + Formic acid CH2O2H 856.6 + Acetic acid C2H4O2H 1711 6 + Propionic acid C3H6O2H 2072 + Butyric acid C4H8O2H 2358 + Pyrrole C4H5NH 2842 + Formamide CH3NOH 2871 + Acetamide C2H5NOH 3992 133 134 7 135 Table S2. The fraction of product ions detected as VOC•H+ ions for different VOCs. + VOC species Ion formula VOC∙H + Formaldehyde CH2OH 1 + Methanol CH4OH 1 + Acetonitrile C2H3NH 1 + Acetaldehyde C2H4OH 1 + Ethanol C2H6OH 1 + Acrolein C3H4OH 1 + Acetone C3H6OH 0.97 + Furan C4H4OH 1 + Isoprene C5H8H 0.87 + MVK C4H6OH 0.62 + MEK C4H8OH 0.87 + Benzene C6H6H 1 + 2-Pentanone C5H10OH 0.94 + Toluene C7H8H 1 + Phenol C6H6OH 1 + Furfural C5H4O2H 1 + Styrene C8H8H 1 + O-xylene C8H10H 1 + m-Cresol C7H8OH 1 + 1,2,4-Teimethylbenzene C9H12H 1 + Naphthalene C10H8H 1 + a-Pinene C10H16H 0.62 136 137 8 138 Table S3. Rate constants of OVOCs representing the combined loss to OH oxidation 139 and photolysis. ∗ jOVOC/[OH] f Species 10−12 cm3 molecule-1 s-1 Formaldehyde 9.4 7.86±0.120 17.26 1.84 Acetaldehyde 15 0.426±0.006 15.42 1.03 Propionaldehyde 20 1.89±0.026 21.89 1.09 n-butyraldehyde 24 3.26±0.046 27.26 1.14 i-butyraldehyde 24 5.64±0.080 29.64 1.24 MACR 29 1.99±0.030 30.99 1.07 Acetone 0.17 0.051±0.0007 0.22 1.29 MEK 1.22 0.368±0.005 1.59 1.30 MVK 20 3.22±0.049 23.22 1.16 Glyoxal 11 3.02±0.044 14.02 1.27 Methyl peroxide 5.5 0.628±0.009 6.13 1.11 Methyl nitrate 0.023 0.107±0.001 0.13 5.65 Ethyl nitrate 0.18 0.122±0.002 0.30 1.67 n-Propyl nitrate 0.58 0.163±0.002 0.74 1.28 i-Propyl nitrate 0.29 0.28±0.004 0.57 1.97 t-butyl nitrate 1.6 0.818±0.011 2.42 1.51 140 f represents the ratio of the rate constant representing the combined losses of reaction ∗ 141 with OH radical and photolysis () and the OH rate constant ().