大气污染控制中心知识库

      二次有机气溶胶Secondary Organic Aerosol SOA 作为 细颗粒物 PM 2.5
的 重要 组分 对 空气质量、全球气候变化 和 人类健康 具有 重要影响 。 SOA 的 形成机制 和 影响因素 是当前 大气 环境 化学 的 前沿领域。 基于 实验室参数的 空气 质量模式 预测值 普遍低 于 SOA 的外场 观测值 ，其根本原因是对 SOA 生成机制 和 影响因素 缺乏全面认识。 近年来 ，我国 机动车保有量 持续 增加， 机动车相关的污染物 加剧 了我国 大气 复合污染的态势，区域 灰霾 事件频发 。 随着我国 机动车尾气排放标准的不断 严格 与 机动车 尾气排放相比， 机动车 蒸发排放对人为源 挥发性有机物V olatile O rganic C ompounds VOCs 的贡 献率已 不容 忽视 。 因此 研究机动车蒸发排放对 SOA 生成的贡献 及 各种因素 的 影响 对 深入了解 灰霾 的成因 具有 重要的环境意义。
      本研究在完成 对 中国科学院 生态环境研究中心（ RCEES CAS 30 m 3 室内 烟
雾箱 一系列评估 与 表征 的 基础上 利用 该烟雾箱进行 了 一系列汽油 蒸汽的 SOA生成 实验 。 以取自 北京、河北等 地 的 液体汽油 蒸汽替代机动车 的蒸发排放， 利用HR ToF AMS 、 SMPS 、 TD GCMS 、 PTR TOF 和 气体分析仪 对 汽油蒸汽 光反应体系中 颗粒相 和气相物种进行 分析 ，系统 研究 了 芳香烃含量 、 共存 气体 SO 2 、 NH 3以及前体物浓度对 SOA 生成的影响 。

首先，对 RCEES CAS 烟雾箱系统 进行 了全面的 表征 。 本研究 建成 了 配备 较
齐全的 30 m 3 室内 烟雾箱 系统 。该烟雾箱系统 可实现 不同季节 、 不同 时段 光强的模拟 ，最高 NO 2 光解速率 可达 0.55 min 1 与北京夏季正午的光强相当 反应 器内 温度 和 湿度 均 可控 气相 NO x 、 O 3 、 SO 2 、 VOCs 和颗粒相物种的 壁 效应 损失均 处于 较低 水平； 建立了 半挥发性 低挥发性有机物（ S emi volatile organic compounds L ow volatility organic compounds S/LVOCs 壁损失 的 计算 方法 ，为准确 修正 颗粒物 的浓度和 产率 提供方法支撑 。 一系列表征 实验 表明 本 烟雾箱可用 于 SOA 的 生成 研究 。
其次，系统 研究了 芳香烃 含量 对汽 油蒸汽 SOA 生成的影响 。 研究发现， 随
着 芳香烃含量的增加 23 %−50 SOA 产率明显增大 4 6 .7 倍）。这 可能 与芳香烃之间的 协同 效应有关，随着芳香烃 含量的增加 会生成更多 的 S VOCs 和
L VOCs 促进 SOA 的生成 。研究 结果表明， 我国亟需 通过 优化炼油工艺 来 降低汽油中的芳香烃含量， 提高 油品质量， 这将有利于缓解我国细颗粒物污染的现状。
再次，系统 研究了 共存 气体 SO 2 和 NH 3 对 汽油蒸汽 S OA 生成的 影响 。 研究发现 环境条件下共存的 SO 2 和 NH 3 可显著促进 汽油蒸汽 S OA 的生成 ，分别 增加 1.6 3.1 倍 和 1.4−1.7 倍 。 SO 2 和 NH 3 的 存在 均对初始 成核速率 和新粒子 生成New particle formation NPF 具有 促进作用 。 SO 2 主要通过 气态 H 2 SO 4 与羰基化合物 在 颗粒 相 表面的 酸催化非均相反应 和低挥发 性 含硫 有机物的生成 促进SOA 的生成 。 NH 3 则 主要 通过 被 摄取 后 与 小分子 有机 羧酸 反应， 同时 NH 4 离子 可 与羰基化合物（如乙二醛）反应生成含氮有机物和低聚物 ，进而 促进 SOA 的生成。 在 我国 SO 2 燃煤烟气 典型 污染物） 和 N H 3 农业面源 和 机动车尾气典型污染物 等复合 污染条件下 估算 机动车蒸发 排放 对 SOA 的 贡献 为 0.49 ± 0.04Tg yr 1 与全球范围内人为源对 SOA 的贡献（ 2.62 Tg yr 1 ）相比 不容忽视 。 因此，为缓解 我国细颗粒物的 污染 现状 应 实现机动车 蒸发排放 、 燃煤烟气 和 农业面源的协同 减排 。
最后，探究了前体物浓度对 SOA 产率 的影响 。研究 发现 低浓度条件下 的
SOA 产率明显高于 高浓度 条件下产率曲线的预测值 3−4 倍） 。 高浓度 条件下，较高 浓度的前体物会与 中间产物 竞争性消耗 OH 等氧化剂， 不 利于中间产物向LVOCs 的 转化， 不利于 SOA 产率 的增加 低浓度条件下， 中间气相 产物 会 提前转化 为 高级氧化产物 主要产物 为 LVOCs ，易 于 成核 ，进而 促进 SOA 的生成和产率的 提高 。 研究结果表明， 利用高浓度条件下获得的SOA产率可能会低估真实大气条件下SOA的产率。这可能是目前模式研究中SOA浓度的模型预测值普遍低于外场观测值的原因之一。

      As an important component of fine particulate matter (PM2.5), secondary organic aerosol (SOA) has an important impact on air quality, climate change and human health. The formation mechanism and influencing factors of SOA have become the frontier areas in current atmospheric environmental chemistry. Due to the lack of comprehensive understanding of SOA formation mechanisms and influencing factors, the SOA concentration in field observations is always underestimated by air quality models based on laboratory parameters. In recent years, vehicle ownership in China has increased rapidly, and vehicle-related pollutants have exacerbated the severity of compound atmospheric pollution, and caused frequent regional haze events in China. Meanwhile, with the implementation of tailpipe exhaust emission control measures, the proportionate share of vehicular evaporation emissions to this pollution has grown, so that they have become non-negligible contributors. Therefore, the study of the contribution of VOCs emitted by gasoline evaporation to SOA formation and the influence of various factors have important environmental significance for understanding the causes of haze events.

      In this thesis, the 30 m 3 indoor smog chamber at the Research Center for Eco
Environment Sciences, Chinese Academy of Sciences (RCEES CAS) was evaluated in detail, then a series of experiments were conducted in this chamber to investigate SOA formation from gasoline vapors . L iquid gasoline s collected from different gas stations
located in Beijing and Hebei province etc. were used as a proxy for gasoline evaporation emissions. A series of instruments such as a high-resolution time-of-flight aerosol mass spectrometer (HR-ToF-AMS), scanning mobility particle sizer (SMPS), thermal desorption coupled with gas chromatograph/mass spectrometer (TD-GCMS), proton-transfer-reaction time of flight mass spectrometry (PTR-TOF) and gas analyzers were employed to perform real-time online monitoring of the particulate and gas phase species. Combined with in-depth analysis of the relevant data, the effects of aromatic hydrocarbon content, sulfur dioxide (SO2), ammonia (NH3) and precursor concentration on the formation of SOA were systematically studied.
Firstly, the RCEES-CAS smog chamber system was comprehensively characterized. A fully equipped 30 m3 indoor smog chamber system has been built. It can be used to simulate the different light intensities observed in different seasons and different periods, and the highest NO2 photolysis rate can reach 0.55 min-1, which could be comparable to the irradiation intensity at noon in Beijing. Temperature and relative humidity in the reactor can be flexibly controlled. In addition, the wall loss rates of both gas-phase (NOx, O3, SO2, VOCs) and particulate-phase species are lower than those of other smog chambers. A calculation method for the wall loss of semi-volatile organic compounds/low-volatility organic compounds (S/LVOCs) was established, which provided methodological support for accurately correcting the concentration and yield of particulate matter. Results of α-pinene/O3 dark experiments and toluene/NOx photo-oxidation experiments revealed SOA yields could be comparable to those derived from other chamber studies. Characterization experiments demonstrated that the RCEES-CAS smog chamber could be used to investigate the formation mechanism of SOA.
Secondly, the effect of aromatic hydrocarbon content on gasoline vapor SOA formation was systematically investigated. This study found that with the increase of aromatic content (from 23 % to 50 %), the SOA yield was significantly enhanced by a factor of 4.0−6.7. This phenomenon might be related to the synergistic effects between precursors in the mixtures. With the increase of aromatic content, higher amounts of LVOCs and SVOCs formed, which is conducive to SOA formation. Our study suggested that it is urgent to reduce the aromatic hydrocarbon content in gasoline by optimizing the refining process and improving the quality of gasoline, which will help to relieve the current situation of secondary fine particulate pollution in China.
Thirdly, the effects of SO2 and NH3 on gasoline vapor SOA formation were systematically investigated. Studies have shown that coexisting of SO2 and NH3 under environmental conditions have a significant effect on the formation of gasoline vapor SOA, with increases of 1.6−3.1 times and 1.4−1.7 times, respectively. The presence of both SO2 and NH3 could promote the initial nucleation rate and new particle formation (NPF). In the presence of SO2, acid-catalyzed heterogeneous reactions between sulfuric acid and carbonyl that take place in the particle phase and the formation of low-volatility sulfur-containing organics might be the main reasons for the increase in SOA formation. On the one hand, when NH3 is present, it can react with small-molecule organic carboxylic acids. On the other hand, NH4+ ions can react with carbonyl compounds (such as glyoxal) to form nitrogen-containing organics and oligomers, thereby promoting SOA formation. Under the compound pollution conditions of SO2 (coal-fired flue gas) and NH3 (agricultural non-point source and traffic emissions), the contribution of vehicle evaporative emissions to SOA formation was estimated to be 0.49 ± 0.04 Tg yr 1 , which could not be ignored when comparing the SOA production
(2.62 Tg yr 1 ) from anthropogenic precursors. Therefore, in order to relieve the secondary PM2.5 pollution, synergistic emission reduction of vehicular evaporation emissions, coal-fired flue gas and agricultural non-point source pollution should be realized.
Finally, the effect of precursor concentration on SOA yield was also investigated. This study found that the SOA yield at low concentrations is significantly higher (3−4 times) than that predicted from the yield curve obtained at high concentrations. At high concentrations, the precursor will compete with the gas-phase intermediates for the consumption of oxidants such as OH radicals, which is not conducive to the conversion of intermediates to LVOCs, nor to the increase of SOA yield. At low concentrations, gas-phase intermediates would be converted into advanced oxidation products more readily. In addition, LVOCs were the main products, which were prone to nucleate, promoting the formation of SOA, and then increasing the SOA yield. Therefore, the SOA yield obtained at high concentrations might underestimate the SOA yield under real atmospheric conditions. This might be one of the reasons that current model predictions generally underestimate the SOA concentration observed in field observations.