环境科学  2019, Vol. 40 Issue (10): 4319-4329   PDF    
聊城市冬季PM2.5中水溶性化合物的昼夜变化特征及来源解析
衣雅男1, 侯战方1,2, 孟静静1,2, 燕丽3, 王心培4, 刘晓迪1, 伏梦璇1, 魏本杰1     
1. 聊城大学环境与规划学院, 聊城 252000;
2. 中国科学院地球环境研究所, 黄土与第四纪地质国家重点实验室, 西安 710061;
3. 生态环境部环境规划院, 北京 100012;
4. 华东师范大学地理科学学院, 地理信息科学教育部重点实验室, 上海 200062
摘要: 为探究聊城市冬季PM2.5中水溶性物质的昼夜变化特征及其来源,于2017年1~2月进行PM2.5样品采集,对其水溶性无机离子、乙二酸和左旋葡聚糖等水溶性化合物进行分析,并采用主成分分析-多元线性回归模型(PCA-MLR)对其来源进行解析.结果表明,采样期间聊城市PM2.5平均质量浓度为(132.6±65.4)μg·m-3,是国家二级标准的1.8倍,且夜晚PM2.5的污染程度略高于白天.SNA(SO42-、NO3-和NH4+)是聊城市PM2.5中最主要的水溶性离子,在白天与夜晚占总离子的质量分数为73.4%和77.1%,说明聊城市冬季二次污染较严重.白天与夜晚阴阳离子平衡当量比值(AE/CE)都小于1,说明PM2.5呈碱性,且夜晚PM2.5的酸性比白天强.无论在白天还是晚上,NH4+的主要存在形态均为NH4HSO4和NH4NO3.通过相关性分析,证实了乙二酸是在液相中经酸催化的二次氧化反应形成的,且受生物质燃烧的影响很强.通过PCA-MLR模型分析可知,聊城市冬季PM2.5中的水溶性化合物主要来自机动车尾气及其二次氧化、生物质燃烧,而受矿物粉尘与煤炭燃烧的影响较小.
关键词: 无机离子      乙二酸      左旋葡聚糖      PM2.5      聊城     
Diurnal Variations and Source Analysis of Water-soluble Compounds in PM2.5 During the Winter in Liaocheng City
YI Ya-nan1 , HOU Zhan-fang1,2 , MENG Jing-jing1,2 , YAN Li3 , WANG Xin-pei4 , LIU Xiao-di1 , FU Meng-xuan1 , WEI Ben-jie1     
1. School of Environment and Planning, Liaocheng University, Liaocheng 252000, China;
2. State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, Xi'an 710061, China;
3. Chinese Academy for Environmental Planning, Beijing 100012, China;
4. Key Laboratory of Geographic Information Science, Ministry of Education, School of Geographic Sciences, East China Normal University, Shanghai 200062, China
Abstract: To investigate the diurnal variations and sources of water-soluble compounds in Liaocheng City, PM2.5 samples were collected between January and February 2017. The PM2.5 samples were analyzed for the compositions, concentrations, and sources of water-soluble inorganic ions, oxalic acid, and levoglucosan. The sources of these chemical compound were investigated using principal component analysis (PCA) and multiple linear regression (MLR) modeling. The results showed that the mass concentrations of PM2.5during the nighttime were higher than those during the daytime, and the average concentrations exceeded the National Ambient Air Quality Standard (GB 3095-2012) by more than 1.8 times. Moreover, atmospheric pollution was worse during the day than during the night. SNA (SO42-, NO3-, and NH4+) were the dominant species among the inorganic ions, the relative abundance of which with respect to the total concentrations of inorganic ions was 73.4% and 77.1% during the daytime and nighttime, respectively. The ratios of anion to cation equivalents (AE/CE) were less than one, suggesting that the PM2.5 was slightly alkaline, and the degree of acidity at night was stronger than during the day. The results of the correlation analyses suggested that aqueous-phase oxidation was the major formation pathway of oxalic acid, which is driven by acid-catalyzed oxidation. The oxalic acid was mainly influenced by biomass burning during the winter in Liaocheng City. The results of the PCA-MLR model suggested that water-soluble compounds in Liaocheng City were mostly from vehicular emissions and secondary oxidation, biomass burning, while the impacts of mineral dust and coal burning were relatively minor.
Key words: inorganic ions      oxalic acid      levoglucosan      PM2.5      Liaocheng City     

近年来, 以PM2.5及光化学污染为特征的霾天气在我国已成为一种公害, 成为制约我国经济和社会可持续发展的重要因素之一[1].水溶性化合物是PM2.5的重要组分, 其浓度水平及化学组成不仅能直接影响大气气溶胶的吸湿性和酸碱性[2, 3], 还能间接影响云凝结核的形成与二次气溶胶的形成机制[4, 5], 进而改变大气气溶胶的光学特性及大气能见度, 从而导致重污染天气和霾天气等大气污染事件的发生[6].

大气中的水溶性化合物主要包括无机离子、二元羧酸和糖等化学物质[7].无机离子是PM2.5中浓度最高的一类水溶性组分, 因其具有强吸湿性, 能够在低于水的饱和蒸气压条件下形成雾滴, 影响云凝结核(CCN)的形成及云的寿命, 进而导致区域降水和地球-大气系统能量平衡的变化[8~10].乙二酸是二元羧酸中分子量最小但浓度最高的二元羧酸, 在总二元羧酸浓度的占比高达60%[5].有研究表明, 相对于一次排放源(如机动车尾气、煤炭燃烧与生物质燃烧等)而言, 乙二酸主要是由气态有机前体物及长链二元羧酸(含碳数>2)经二次氧化反应产生的[5, 11, 12]. 3D模式研究结果表明, 全球范围内79%的乙二酸来自异戊二烯、单萜烯等生物源碳氢化合物的光化学氧化反应[13].因此, 乙二酸可用来表征大气气溶胶的老化程度[14].左旋葡聚糖是植物纤维素燃烧的热解产物, 在高于300℃条件下, 经过转糖基作用、裂解等复杂的燃烧热解过程而形成的焦油状无水单糖[11].在由生物质燃烧的烟羽中, 左旋葡聚糖是浓度最高的一种脱水糖类化合物, 其次为甘露聚糖和半乳聚糖[15].基于外场监测[16]及烟雾箱实验[17, 18]分析, 结果表明左旋葡聚糖可与羟基自由基(·OH)在液相中发生氧化反应形成高分子量化合物.但是, 相较于其他生物质燃烧的示踪物(如碳同位素与K+), 左旋葡聚糖具有排放因子高、分析成本廉价、特异性好且化学稳定性强的优点[19], 因此被广泛作为生物质燃烧的重要分子指示物[20].由此可以得出, 研究PM2.5中的水溶性化合物对于探讨我国大气污染特征、来源及其气候效应具有重要的实际意义与科学意义.

山东省是我国PM2.5和气体污染物(SO2、NOx)排放浓度最高且重污染天气发生天数最多的省份之一[21], 而聊城市地处鲁西北内陆地区, 是京津冀大气污染重要传输通道城市(“2+26”城市)之一, 且其大气质量一直处于山东省末位, 因此, 聊城市大气污染问题已被广泛关注[22~25].目前, 关于聊城市PM2.5污染特征及其来源的研究, 主要集中于无机离子、碳质组分、无机元素、正构烷烃等化学组分方面, 而关于聊城市PM2.5中的水溶性有机物的研究尚不全面, 尤其关于聊城市大气中乙二酸的研究还未见报道[22~25].因此, 本研究对聊城市2017年冬季PM2.5中水溶性化合物的浓度水平及化学组成进行分析, 运用阴阳离子平衡法及气溶胶无机模型(AIM-Ⅱ)计算其酸碱度, 并结合主成分分析-多元线性回归模型(PCA-MLR)阐明聊城市冬季PM2.5中水溶性化合物的来源, 以期为聊城市大气污染治理决策的制定提供理论依据, 同时为“2+26”城市大气污染的联防联控提供基础数据及技术支撑.

1 材料与方法 1.1 样品采集

于2017年1~2月对聊城市PM2.5进行为期25 d的采集, 所用采样仪器为中流量采样器(青岛崂山电子仪公司生产), 流量为100 L ·min-1.采样点设在聊城大学4号实验楼楼顶(34.42°N, 116.01°E), 距地面约20 m.采样点周围无高大建筑物及明显污染源, 可以客观反映聊城市的大气污染状况. PM2.5样品采集分昼(08:00~19:50)和夜(20:00~次日07:50)进行, 共采集50个样品以及2个环境空白样品.采样前, 将所用滤膜置于450℃马弗炉中灼烧6 h, 以去除吸附性有机污染物.在采样完成之后, 将滤膜用铝箔纸封包并置于-20℃的冰箱中冷冻保存待分析.

1.2 无机离子、乙二酸与左旋葡聚糖分析

将1/4滤膜剪碎并置于样品瓶中, 加入30 mL超纯水(R<18.2MΩ ·cm), 在冰水浴下超声萃取4×15 min后, 置于脱色摇床振荡1 h, 然后用0.45 μm的水系过滤器过滤.所得到的滤液用离子色谱仪(Dionex, ICS-1100; Dionex, DX-80)分别提取10种水溶性离子, 即阴离子(F-、Cl-、SO42-、NO3-、乙二酸)和阳离子(NH4+、Na+、K+、Mg2+、Ca2+).水溶性阴离子采用IonPac ASRS-4抑制柱和IonPac AS11HC×250 mm分析柱, 淋洗液为KOH, 梯度淋洗.水溶性阳离子采用IonPacCSRS-4抑制柱和IonPac CS12A×250分析柱, 淋洗液为18mmol ·L-1甲磺酸.对于左旋葡聚糖的分析, 利用离子色谱仪与脉冲安培检测器联用的实验方法[19, 26].首先将1/4滤膜置于15 mL聚四氟乙烯材质的样品瓶中, 加入10 mL浓度为20mmol ·L-1的叠氮化钠溶液振荡60 min, 然后用0.22微孔尼龙滤膜过滤, 最后经RP净化小柱(Dionex OnGuard Ⅱ RP)后利用Dionex ICS-5000型离子色谱仪进行监测.左旋葡聚糖的检测限为1×10-8 mg ·L-1.在样品分析过程中, 每分析10个样品插入一个标准样品, 以确保仪器正常运行, 每个样品分析两次, 误差小于5%.

1.3 AIM-Ⅱ模型

因为气溶胶的含水量(aerosol liquid water content, ALWC)及实际酸度(in-situ pH, pHis)不能通过实验的方式直接测出, 所以本研究利用气溶胶无机模型(aerosol inorganic model, AIM-Ⅱ)计算获得(http://www.aim.env.uea.ac.uk/aim/aim.php).该模式结合实际环境中的相对湿度与温度数据, 计算SO42--NO3--NH4+-H+平衡体系的组成与LWC.其中pHis的计算公式如(1)所示:

(1)

式中, αH+为颗粒物液相表面H+的活性, 单位为mol ·L-1; γH+为H+的活性系数; nH+为单位体积空气中自由H+的浓度, 单位为mol ·m-3; Va为单位体积环境大气中气溶胶液相表面的体积, 单位为cm3 ·m-3.其中γH+nH+Va和ALWC可通过AIM-Ⅱ模型直接获得.

1.4 PCA-MLR模型

利用主成分分析(principal component analysis, PCA)方法对聊城市冬季水溶性化合物的来源进行识别, 再利用多元线性回归(multiple linear regression, MLR)方法确定不同污染源的贡献率[27, 28]. MLR模型的基本方程为:

(2)

式中, Y表示样品中的化学物质, Xi为PCA分析得到的因子得分变量, mib为MLR的回归系数.为便于直接比较各来源的贡献, 通常对因变量Y标准化之后进行计算.

(3)

式中, Bi为MLR的新回归系数.源i的平均贡献率的计算公式为:

(4)
2 结果与讨论 2.1 PM2.5质量浓度与无机离子的昼夜变化特征

采样期间, 聊城市PM2.5质量浓度变化范围为40.6~342.2μg ·m-3, 平均质量浓度为(132.6±65.4)μg ·m-3(表 1), 是我国环境空气质量标准(GB 3095-2012)二级日均值标准(75μg ·m-3)的1.8倍.以二级标准浓度限值对采样期间污染天气和清洁天气进行划分, 清洁天天数为4 d, 仅占总天数的8%.白天PM2.5的质量浓度[(130.9±63.8)μg ·m-3]与夜晚[(134.4±65.4)μg ·m-3]接近, 昼夜变化特征不显著(t检验, P>0.05), 由于采样期间风速较小(1.4m ·s-1)且主导方向为弱偏南风, 大气扩散条件较差使得静稳天气较多, 从而导致PM2.5浓度的积累.聊城市作为京津冀大气污染传输通道(“2+26”城市)重要城市之一, 其PM2.5的质量浓度高于北京、太原、廊坊、菏泽, 但低于郑州和安阳(表 2).在中国空气质量在线监测分析的366个城市中, 聊城市PM2.5的质量浓度在2017-01-17~2017-02-10期间高于其他337个城市(https://www.aqistudy.cn).与2016年同期相比, PM2.5的质量浓度同比增长了25.7%(https://www.aqistudy.cn).由此可知, 聊城市PM2.5污染很严重, 且大气质量在逐年恶化.

表 1 聊城市PM2.5质量浓度及水溶性化合物的昼夜变化特征/μg ·m-3 Table 1 Diurnal variations in the mass concentrations of PM2.5 and water-soluble compounds in Liaocheng City/μg ·m-3

表 2 2017年冬季聊城市PM2.5质量浓度与同期其它城市的对比1) Table 2 Comparison of mass concentrations of PM2.5 in Liaocheng City with other cities during winter, 2017

表 1可知, 在白天无机离子的浓度由高到低的顺序为NO3->SO42->NH4+>Cl->Ca2+>K+>Mg2+>Na+, 离子总浓度占PM2.5质量浓度的48.4%;在夜晚离子的浓度由高到低的顺序为NO3->SO42->NH4+>Cl->K+>Ca2+>Na+>Mg2+, 离子总浓度占PM2.5质量浓度的47.0%.其中, SO42-、NO3-和NH4+(sulfate、nitrate和ammonium, SNA)被称为二次无机离子, 可表征气溶胶的二次污染程度, 因为这3种离子分别是由SO2、NOx及NH3经过光化学氧化和气-固相分配而形成的[6].在白天与夜晚, SNA的浓度分别占离子总浓度的73.4%和77.1%, 说明聊城市冬季二次污染较严重.

表 1图 1可知, NO3-作为浓度最高的无机离子, 夜晚的浓度是白天的1.2倍.颗粒态的NO3-首先是由NO2与·OH在气相中发生光化学氧化反应, 再经非均相反应进入气溶胶相而形成的[10].另外, 在气溶胶中, NO3-主要是以NH4NO3的形式存在, NH4NO3的热稳定性较差且挥发性较强[29], 夜晚较低的气温条件有利于NH4NO3浓度的累积, 导致NO3-的浓度呈昼低夜高的变化特征. NH4+的浓度也呈昼低夜高的变化特征(表 1), 证明了上述论证的正确性. SO42-主要是由SO2经液相氧化形成的, 且高温、高湿的环境有利于此反应的进行[3, 6].因此, 白天SO42-的质量浓度与在离子总浓度的占比均高于夜晚(表 1图 1). Ca2+和Mg2+都来自于地壳源, 如建筑粉尘、道路扬尘等, 是矿物粉尘的示踪物[30].由表 1图 1可知, 在白天Ca2+和Mg2+的浓度与相对含量均高于夜晚, 说明白天矿物粉尘对无机离子的贡献高于夜晚, 这可能与白天人类活动频繁产生较多的扬尘和建筑尘有关.在夜晚Cl-与K+的浓度分别是白天的1.4倍和1.3倍(表 1).有研究表明, Cl-和K+主要来自海盐、矿物粉尘与生物质燃烧[31].因为聊城市属于内陆城市, 所以PM2.5的化学成分受海洋的影响极小.由表 3可知, 在白天与晚上, Cl-与K+间的相关性很强(R2>0.7, P<0.01), 但二者和Ca2+、Mg2+的相关性均很差(R2<0.3).由此可以得出, 聊城市Cl-与K+主要来自生物质燃烧, 由于夜晚取暖致使生物质燃烧排放源增多, 导致夜晚受生物质燃烧的影响更强.

图 1 聊城市白天和夜晚无机离子占总离子的质量分数 Fig. 1 Comparison of the relative abundance of inorganic ions relative to the total concentration of ions during the daytime and nighttime in Liaocheng City

表 3 聊城市白天和夜晚各离子间的相关性 Table 3 Correlation analysis of inorganic ions during the daytime and nighttime in Liaocheng City

2.2 气溶胶酸碱度分析及NH4+存在形式

阴阳离子平衡常被用来推断PM2.5的酸碱度[32], 阴、阳离子电荷当量分别用AE(anion equivalent)、CE(cation equivalent)表示, 其公式分别如式(5)、(6)所示:

(5)
(6)

图 2所示, 聊城市冬季白天与夜晚阴离子与阳离子间均呈显著的相关性, 其相关系数(R2)均大于0.5.白天与夜晚线性回归方程的斜率(AE/CE)都小于1, 说明聊城市冬季PM2.5呈弱碱性, 与部分城市地区的酸碱性相似(如济南[29]、黄石[33]和邯郸[34]等), 但与高山地区相反(如黄山[35]和华山[36]等).白天回归方程的斜率(0.85)小于夜晚(0.97), 表明夜晚PM2.5的酸性比白天强.由AIM-Ⅱ模型的计算结果可知, 在白天与夜晚PM2.5的pHis分别为4.5±0.6和3.0±0.3(表 1), 进一步表明夜晚PM2.5的酸性更强.由表 1可知, 白天LWC的浓度[(44.2±33.0)μg ·m-3]低于夜晚[(53.0±28.2)μg ·m-3].在AIM-Ⅱ模式中, LWC是通过计算SO42--NO3--NH4+-H+平衡体系的组成并结合相对湿度、温度得到的.因为只有相对湿度高于(NH4)2SO4、NH4HSO4和NH4NO3等具有吸湿性的化学物质时, LWC才是存在的; 否则, LWC是不存在的[37, 38].例如, 在25℃时, NH4HSO4的潮解点是~40%, (NH4)2SO4的潮解点是~80%.因此, 大气气溶胶的LWC是由相对湿度[37]与具有吸湿性的无机离子[38]共同决定的.白天无机离子的总浓度[(63.6±38.6)μg ·m-3]与夜晚[(64.3±36.9)μg ·m-3]接近, 但白天的相对湿度[(39.0±15.3)%]却低于夜晚[(44.6±16.4)%](表 1).因此, 聊城冬季的LWC呈昼低夜高的变化特征.

图 2 白天和夜晚阴离子与阳离子的相关性分析 Fig. 2 Correlation analysis of anion and cation equivalents during the daytime and nighttime

作为浓度最高的碱性离子, NH4+在大气颗粒物中主要以NH4HSO4、(NH4)2SO4、NH4Cl和NH4NO3的形式存在[39].在本研究中, NH4+与SO42-、Cl-和NO3-间都呈较好的相关关系(R2>0.5, P<0.01)(表 3).但是, NH4Cl的挥发性强且热稳定性差, 加之Cl-占离子总浓度的相对含量比SO42-、和NO3-低得多(图 1), 因此本研究认为聊城市冬季PM2.5中的NH4+通过NH4HSO4、(NH4)2SO4和NH4NO3这3种形式存在[40].有研究表明[33, 36], 当NH4+主要以NH4HSO4和NH4NO3的形式存在, 颗粒物中NH4+的浓度可通过公式(7)计算得到; 当NH4+主要以(NH4)2SO4和NH4NO3的形式存在, 颗粒物中NH4+的可通过公式(8)来计算得到:

(7)
(8)

NH4+的实测浓度与计算浓度间的线性关系如图 3所示.在白天, 由式(7)与式(8)得到的NH4+的计算浓度与实测浓度的相关性均很强(R2=0.98), 但由式(7)得到的线性回归方程的斜率(b=0.998)比式(8)(b=0.72)更接近于1, 说明由式(7)计算出的NH4+计算值比式(8)得到的值更接近于实测值.由此可以得出, 聊城市白天NH4+是以NH4HSO4和NH4NO3的形式存在.在夜晚, 由式(7)得到的NH4+的计算浓度与实测浓度的拟合斜率(b=0.87)比后者的(b=0.67)更接近于1, 表明在夜晚NH4+与SO42-的主要存在形式是NH4HSO4.因此, 无论是白天还是夜晚, NH4+均是以NH4HSO4和NH4NO3的形式存在的.

图 3 白天和夜晚NH4+的实测浓度与计算浓度间的对比关系 Fig. 3 Comparison of calculated and measured concentrations of NH4+ during the daytime and nighttime

2.3 乙二酸与左旋葡聚糖的昼夜变化特征

表 1所示, 聊城市冬季乙二酸白天的浓度[(472.2±281.9)ng ·m-3]是夜晚[(314.7±178.4)ng ·m-3]的1.5倍, 昼夜差异明显(t检验, P<0.001).聊城市冬季乙二酸的平均浓度为(393.4±246.7)ng ·m-3, 不仅高于冬季的海洋地区, 如西北太平洋[(97.4±74.1)ng ·m-3)][41]和西北印度洋[(116±65)ng ·m-3)][42], 还高于高山地区, 如中国的黄山[(166±80)ng ·m-3)]和印度奈妮塔尔的喜马拉雅山(353ng ·m-3)[43].另外, 在全球范围内的城市地区中, 高于中国的青岛[(103.8±75.9)ng ·m-3)]、北京[(149±123)ng ·m-3)][12]和蒙古国的乌兰巴托[(107±28)ng ·m-3)][44], 但远低于中国的西安[(1 162±570)ng ·m-3)][45]、菏泽[(987±778)ng ·m-3)][46]和印度的坎普尔(1 375 ng ·m-3)[11].由此可知, 在全球范围内, 聊城市冬季乙二酸的污染程度较严重.

大量研究表明, 乙二酸是由各种前体物(含碳数大于2的二元羧酸、α-二羰基化合物与乙醛酸等)与氧化剂(如自由基、O3等)在液相中经氧化反应形成的[5, 42, 47, 48].为了探究乙二酸的形成机制, 本研究对聊城市冬季白天与夜晚的乙二酸与SO42-、LWC、pHis和O3间的相关性进行了分析(图 4).由图 4(a)可知, 乙二酸与SO42-均呈较强的相关性(R2>0.7), 证实了乙二酸的液相形成机制.另外, 乙二酸与LWC间的相关性也较强[图 4(b)], 表明气溶胶LWC的升高有利于乙二酸的形成, 因为浓度较高的LWC有利于乙二酸的水溶性半挥发有机前体物(如乙二醛、甲基乙二醛)由气相进入液相, 从而生成乙二酸[49]. Meng等[5]和Wang等[50]的研究结果表明, 酸性环境可以促进乙二酸及其前体物的生成.因此, 乙二酸与气溶胶的pHis呈较强的负相关性[R2>0.5, 图 4(c)].与其他城市地区[51]和高山地区相似[5, 52], 乙二酸在白天与O3呈强相关性(R2=0.65), 但在夜晚二者间的相关性很差[(R2=0.01), 图 4(d)].这是因为在白天乙二酸的生成主要来自前体物与O3、·OH的氧化反应, 而在夜晚乙二酸是由前体物与NO3 ·、H2O2等氧化剂的氧化反应生成的[51].

图 4 乙二酸与SO42-、LWC、pHis、O3的相关性分析 Fig. 4 Correlation analysis of oxalic acid with SO42-, liquid water content (LWC), in-situ acidity of particles (pHis), and O3 during the daytime and nighttime

为研究乙二酸液相形成机制的影响因素, 本文分析了乙二酸与环境温度、相对湿度的相关性.理论上讲, 由于亨利定律的控制, 温度较低的环境条件有利于乙二酸的气相前体物(乙二醛和甲基乙二醛)进入液相, 从而促进乙二酸的生成[13].但是, 有大量研究表明, 温度的升高有利于乙二酸的形成及气溶胶的老化[5, 27, 50].由图 5(a)可知, 在整个采样期间, 温度与乙二酸间的相关性很显著(R2>0.6), 这是因为较高的温度能够促进乙二酸通过氧化反应形成[5, 27].通过多元线性回归分析可得到等式(9):

(9)

式中, RH为相对湿度, RN/S为NH4+与SO42-的量比(mol/mol).

由式(9)可知, 相对湿度的升高使得pHis升高, 即降低了气溶胶的实际酸度.上述论述已证明, 乙二酸是经酸催化反应形成的, 乙二酸的浓度与pHis间呈负相关性, 所以较高的相对湿度不利于乙二酸的生成.但是, 相对湿度的升高能够升高LWC的浓度[53], 而LWC的升高能够使乙二酸的气体前体物进入液相, 致使更多乙二酸的生成.因此, 乙二酸与相对湿度间相关性很差(R2<0.3, 图 5).

图 5 乙二酸与温度、相对湿度的相关性分析 Fig. 5 Correlation analysis of oxalic acid with temperature and relative humidity during the daytime and nighttime

表 1可知, 左旋葡聚糖在夜晚的浓度[(347.8±152.4)ng ·m-3]是白天[(284.3±80.0)ng ·m-3]的1.2倍, 这是因为夜间取暖使得生物质燃烧排放增多, 加之大气稳定度较高且边界层降低导致的[25].有大量研究表明, 生物质燃烧可排放出醛类化合物(如乙二醛、甲基乙二醛等), 再经光化学反应形成乙二酸[46, 53, 54].因此, 乙二酸与左旋葡聚糖间的相关性显著(R2>0.7, 图 6), 表明生物质燃烧是聊城市冬季乙二酸的重要来源之一.另外, 白天乙二酸与左旋葡聚糖间的比值是夜晚的1.9倍(表 1), 与Sorathia等[11]和孟静静等[46]所得的结论是一致的, 表明白天温度高且太阳辐射强的环境条件有利于生物质燃烧排放的前体物经光化学氧化反应生成乙二酸.

图 6 乙二酸与左旋葡聚糖的相关性分析 Fig. 6 Correlation analysis of oxalic acid with levoglucosan during the daytime and nighttime

2.4 来源解析

为进一步分析聊城市冬季PM2.5的来源, 本研究利用PCA-MLR模型进行分析, 其结果如表 4所示.经过筛选测试, 共提取4个主成分.在第1主成分中, NO3-、乙二酸与NO2的载荷值较高(>0.7).城市大气中的NO3-与NO2主要来自汽油车和柴油车的尾气排放[55], 因此这2种物质被认为是机动车尾气的示踪物.另外, 大气中NO3-和乙二酸主要是经二次氧化形成的, 所以第1主成分表征聊城市冬季PM2.5主要来自机动车尾气及其产生的前体物经二次氧化形成的二次污染源.代表生物质燃烧源的K+、Cl-和左旋葡聚糖在第2主成分中的载荷值较高, 表明生物质燃烧是聊城市冬季PM2.5的另一重要污染源.在第3主成分中, Na+、Ca2+和Mg2+的载荷值较高, 说明矿物粉尘是PM2.5的第3大污染源.在城市大气中, SO2主要来自煤炭燃烧源[29]. SO2、SO42-与第4主成分的相关性很强, 说明煤炭燃烧对聊城市大气有一定的影响, 但这种影响不显著.

表 4 聊城市冬季PM2.5中水溶性化合物的PCA-MLR模型分析结果1) Table 4 PCA-MLR model for water-soluble compounds in PM2.5 during winter in Liaocheng City

表 4中的因子得分变量作为自变量, 将化合物的总浓度作为因变量, 利用SPSS进行多元线性回归(MLR), 得到线性回归方程(10):

(10)

式中, Fi为源i的因子得分变量, 再利用式(4)求得最终源贡献率.

PCA-MLR结果表明, 聊城市冬季PM2.5中的水溶性化合物主要来自机动车尾气及其二次氧化和生物质燃烧, 贡献率分别为50.7%和28.5%.矿物粉尘与煤炭燃烧对PM2.5的影响较小, 其贡献率分别为14.1%和6.8%.

3 结论

(1) 聊城市冬季PM2.5的平均质量浓度是国家二级标准的1.8倍, 其中清洁天天数仅占总天数的8%. SO42-、NO3-和NH4+是聊城市冬季PM2.5中主要的无机离子, 在白天与夜晚分别占总离子浓度的73.4%和77.1%, 说明聊城市冬季二次污染较严重.仅SO42-、Ca2+和Mg2+呈昼高夜低的变化特征, 而NO3-、NH4+、Cl-和K+呈昼低夜高的变化特征.

(2) 阴阳离子平衡与AIM-Ⅱ模型的计算结果表明, 夜晚PM2.5的酸度比白天强.无论在白天还是在晚上, NH4+均是以NH4HSO4和NH4NO3的形式存在的.

(3) 由相关性分析可知, 乙二酸主要是在液相中经酸催化的二次氧化反应形成的, 含水量的升高有利于乙二酸的生成.乙二酸与左旋葡聚糖呈强相关性, 表明聊城市冬季PM2.5中的乙二酸主要受生物质燃烧的影响.

(4) 由PCA-MLR模型分析结果可知, 聊城市冬季PM2.5中的水溶性化合物主要来自机动车尾气及其二次氧化和生物质燃烧, 贡献率分别为50.7%和28.5%.矿物粉尘与煤炭燃烧对PM2.5的影响相对较小, 其贡献率分别仅为14.1%和6.8%.

参考文献
[1] Huang R J, Zhang Y L, Bozzetti C, et al. High secondary aerosol contribution to particulate pollution during haze events in China[J]. Nature, 2014, 514(7521): 218-222. DOI:10.1038/nature13774
[2] Gioda A, Reyes-Rodríguez G J, Santos-Figueroa G, et al. Speciation of water-soluble inorganic, organic, and total nitrogen in a background marine environment:cloud water, rainwater, and aerosol particles[J]. Journal of Geophysical Research, 2011, 116(D05): D05203.
[3] Kundu S, Kawamura K, Andreae T W, et al. Diurnal variation in the water-soluble inorganic ions, organic carbon and isotopic compositions of total carbon and nitrogen in biomass burning aerosols from the LBA-SMOCC campaign in Rond? nia, Brazil[J]. Journal of Aerosol Science, 2010, 41(1): 118-133. DOI:10.1016/j.jaerosci.2009.08.006
[4] Liu Z R, Xie Y Z, Hu B, et al. Size-resolved aerosol water-soluble ions during the summer and winter seasons in Beijing:formation mechanisms of secondary inorganic aerosols[J]. Chemosphere, 2017, 183: 119-131. DOI:10.1016/j.chemosphere.2017.05.095
[5] Meng J J, Wang G H, Hou Z F, et al. Molecular distribution and stable carbon isotopic compositions of dicarboxylic acids and related SOA from biogenic sources in the summertime atmosphere of Mt. Tai in the North China Plain[J]. Atmospheric Chemistry and Physics, 2018, 18(20): 15069-15086. DOI:10.5194/acp-18-15069-2018
[6] Wang G H, Zhang F, Peng J F, et al. Particle acidity and sulfate production during severe haze events in China cannot be reliably inferred by assuming a mixture of inorganic salts[J]. Atmospheric Chemistry and Physics, 2018, 18(14): 10123-10132. DOI:10.5194/acp-18-10123-2018
[7] Agarwal S, Aggarwal S G, Okuzawa K, et al. Size distributions of dicarboxylic acids, ketoacids, α-dicarbonyls, sugars, WSOC, OC, EC and inorganic ions in atmospheric particles over Northern Japan:implication for long-range transport of Siberian biomass burning and East Asian polluted aerosols[J]. Atmospheric Chemistry and Physics, 2010, 10(13): 5839-5858. DOI:10.5194/acp-10-5839-2010
[8] Huang X J, Liu Z R, Zhang J K, et al. Seasonal variation and secondary formation of size-segregated aerosol water-soluble inorganic ions during pollution episodes in Beijing[J]. Atmospheric Research, 2016, 168: 70-79. DOI:10.1016/j.atmosres.2015.08.021
[9] Li X R, Wang L L, Ji D S, et al. Characterization of the size-segregated water-soluble inorganic ions in the Jing-Jin-Ji urban agglomeration:spatial/temporal variability, size distribution and sources[J]. Atmospheric Environment, 2013, 77: 250-259. DOI:10.1016/j.atmosenv.2013.03.042
[10] Zhang R Y, Wang G H, Guo S, et al. Formation of urban fine particulate matter[J]. Chemical Reviews, 2015, 115(10): 3803-3855. DOI:10.1021/acs.chemrev.5b00067
[11] Sorathia F, Rajput P, Gupta T. Dicarboxylic acids and levoglucosan in aerosols from Indo-Gangetic Plain:inferences from day night variability during wintertime[J]. Science of the Total Environment, 2018, 624: 451-460. DOI:10.1016/j.scitotenv.2017.12.124
[12] Zhao W Y, Kawamura K, Yue S Y, et al. Molecular distribution and compound-specific stable carbon isotopic composition of dicarboxylic acids, oxocarboxylic acids and α-dicarbonyls in PM2.5 from Beijing, China[J]. Atmospheric Chemistry and Physics, 2018, 18(4): 2749-2767. DOI:10.5194/acp-18-2749-2018
[13] Myriokefalitakis S, Tsigaridis K, Mihalopoulos N, et al. In-cloud oxalate formation in the global troposphere:a 3-D modeling study[J]. Atmospheric Chemistry and Physics, 2011, 11(12): 5761-5782. DOI:10.5194/acp-11-5761-2011
[14] Ren Y Q, Li H, Meng F, et al. Impact of emission controls on air quality in Beijing during the 2015 China Victory Day Parade:implication from organic aerosols[J]. Atmospheric Environment, 2019, 198: 207-214. DOI:10.1016/j.atmosenv.2018.10.061
[15] Fine P M, Cass G R, Simoneit B R T. Organic compounds in biomass smoke from residential wood combustion:emissions characterization at a continental scale[J]. Journal of Geophysical Research, 2002, 107(D21): 8349.
[16] Mochida M, Kawamura K, Fu P Q, et al. Seasonal variation of levoglucosan in aerosols over the western North Pacific and its assessment as a biomass-burning tracer[J]. Atmospheric Environment, 2010, 44(29): 3511-3518. DOI:10.1016/j.atmosenv.2010.06.017
[17] Hoffmann D, Tilgner A, Iinuma Y, et al. Atmospheric stability of levoglucosan:a detailed laboratory and modeling study[J]. Environmental Science & Technology, 2010, 44(2): 694-699.
[18] Hennigan C J, Sullivan A P, Collett Jr J L, et al. Levoglucosan stability in biomass burning particles exposed to hydroxyl radicals[J]. Geophysical Research Letters, 2010, 37(9): L09806.
[19] 黄绪, 郭云霞, 刘剑斌, 等. 柳州大气PM2.5中糖类物质的分布特征与指示意义[J]. 中国环境科学, 2017, 37(3): 838-843.
Huang X, Guo Y X, Liu J B, et al. The distribution characteristics and implications of carbohydrates in the PM2.5 of Liuzhou[J]. China Environmental Science, 2017, 37(3): 838-843.
[20] Hu Q H, Xie Z Q, Wang X M, et al. Monocarboxylic and dicarboxylic acids over oceans from the East China Sea to the Arctic Ocean:roles of ocean emissions, continental input and secondary formation[J]. Science of the Total Environment, 2018, 640-641: 284-292. DOI:10.1016/j.scitotenv.2018.05.311
[21] Larsen R K, Baker J E. Source apportionment of polycyclic aromatic hydrocarbons in the urban atmosphere:a comparison of three methods[J]. Environmental Science & Technology, 2003, 37(9): 1873-1881.
[22] Zhang J M, Yang L X, Chen J M, et al. Influence of fireworks displays on the chemical characteristics of PM2.5 in rural and suburban areas in Central and East China[J]. Science of the Total Environment, 2016, 578: 476-484.
[23] 张敬巧, 吴亚君, 张萌, 等. 聊城冬季一重污染过程PM2.5污染特征及成因分析[J]. 环境科学, 2018, 39(9): 4026-4033.
Zhang J Q, Wu Y J, Zhang M, et al. PM2.5 pollution characterization and cause analysis of a winter heavy pollution event, Liaocheng City[J]. Environmental Science, 2018, 39(9): 4026-4033.
[24] 张敬巧, 王淑兰, 罗达通, 等. 聊城市冬季PM2.5及水溶性离子污染特征及来源分析[J]. 环境科学研究, 2018, 31(10): 1712-1718.
Zhang J Q, Wang S L, Luo D T, et al. Characterization and source analysis of PM2.5 and water-soluble ions during winter in Liaocheng City[J]. Research of Environmental Sciences, 2018, 31(10): 1712-1718.
[25] 刘晓迪, 孟静静, 侯战方, 等. 聊城市冬季PM2.5中正构烷烃和糖类的污染特征及来源解析[J]. 环境科学, 2019, 40(2): 548-557.
Liu X D, Meng J J, Hou Z F, et al. Pollution characteristics and source analysis of n-alkanes and saccharides in PM2.5 during the winter in Liaocheng City[J]. Environmental Science, 2019, 40(2): 548-557.
[26] 齐炜红, 刀谞, 吕怡兵, 等. 离子色谱快速测定大气颗粒物(PM2.5、PM10)中的三种脱水聚糖[J]. 环境化学, 2016, 35(12): 2521-2527.
Qi W H, Dao X, Lv Y B, et al. Rapid determination of three dehydration lycans in atmospheric particulate matter (PM2.5, PM10) by ion chromatography[J]. Environmental Chemistry, 2016, 35(12): 2521-2527. DOI:10.7524/j.issn.0254-6108.2016.12.2016051905
[27] Meng J J, Wang G H, Li J J, et al. Seasonal characteristics of oxalic acid and related SOA in the free troposphere of Mt. Hua, central China:implications for sources and formation mechanisms[J]. Science of the Total Environment, 2014, 493: 1088-1097. DOI:10.1016/j.scitotenv.2014.04.086
[28] Cao G L, Zhang X Y, Gong S L, et al. Emission inventories of primary particles and pollutant gases for China[J]. Chinese Science Bulletin, 2011, 56(8): 781-788. DOI:10.1007/s11434-011-4373-7
[29] 刘晓迪, 孟静静, 侯战方, 等. 济南市夏、冬季PM2.5中化学组分的季节变化特征及来源解析[J]. 环境科学, 2018, 39(9): 4014-4025.
Liu X D, Meng J J, Hou Z F, et al. Analysis of seasonal variations in chemical characteristics and sources of PM2.5 during summer and winter in Ji'nan City[J]. Environmental Science, 2018, 39(9): 4014-4025.
[30] 赵庆炎, 姜楠, 燕启社, 等. 郑州市夏、秋季大气颗粒物中水溶性无机离子质量浓度及粒径分布特征[J]. 环境科学, 2018, 39(11): 4866-4875.
Zhao Q Y, Jiang N, Yan Q S, et al. Size distribution characteristics of water-soluble inorganic ions during summer and autumn in Zhengzhou[J]. Environmental Science, 2018, 39(11): 4866-4875.
[31] Andreae M O. Soot carbon and excess fine potassium:long-range transport of combustion-derived aerosols[J]. Science, 1983, 220(4602): 1148-1151. DOI:10.1126/science.220.4602.1148
[32] Sun Y L, Zhuang G S, Tang A H, et al. Chemical characteristics of PM2.5 and PM10 in haze-fog episodes in Beijing[J]. Environmental Science & Technology, 2006, 40(10): 3148-3155.
[33] Liu H X, Zheng J R, Qu C K, et al. Characteristics and source analysis of water-soluble inorganic ions in PM10 in a typical mining city, central China[J]. Atmosphere, 2017, 8(4): 74.
[34] Meng C C, Wang L T, Zhang F F, et al. Characteristics of concentrations and water-soluble inorganic ions in PM2.5 in Handan City, Hebei province, China[J]. Atmospheric Research, 2016, 171: 133-146. DOI:10.1016/j.atmosres.2015.12.013
[35] Li L, Yin Y, Kong S F, et al. Altitudinal effect to the size distribution of water soluble inorganic ions in PM at Huangshan, China[J]. Atmospheric Environment, 2014, 98: 242-252. DOI:10.1016/j.atmosenv.2014.08.077
[36] 曹聪, 王格慧, 吴灿, 等. 华山地区PM2.5中无机离子垂直分布特征[J]. 环境科学, 2018, 39(4): 1473-1483.
Cao C, Wang G H, Wu C, et al. Vertical distributional characteristics of inorganic ions of PM2.5 at Mt. Huashan, inland China[J]. Environmental Science, 2018, 39(4): 1473-1483.
[37] Clegg S L, Brimblecombe P, Wexler A S. Thermodynamic model of the system H+-NH4+-SO42--NO3--H2O at tropospheric temperatures[J]. The Journal of Physical Chemistry A, 1998, 102(12): 2137-2154. DOI:10.1021/jp973042r
[38] Fountoukis C, Nenes A. ISORROPIA Ⅱ:a computationally efficient thermodynamic equilibrium model for K+-Ca2+-Mg2+-NH4+-Na+-SO42--NO3--Cl--H2O aerosols[J]. Atmospheric Chemistry and Physics, 2007, 7(17): 4639-4659. DOI:10.5194/acp-7-4639-2007
[39] Zhang R, Jing J, Tao J, et al. Chemical characterization and source apportionment of PM2.5 in Beijing:seasonal perspective[J]. Atmospheric Chemistry and Physics, 2013, 13(14): 7053-7074. DOI:10.5194/acp-13-7053-2013
[40] 张云峰, 于瑞莲, 胡恭任, 等. 泉州市大气PM2.5中水溶性离子季节变化特征及来源解析[J]. 环境科学, 2017, 38(10): 4044-4053.
Zhang Y F, Yu R L, Hu G R, et al. Seasonal variation and source apportionment of water-soluble ions in PM2.5 in Quanzhou City[J]. Environmental Science, 2017, 38(10): 4044-4053.
[41] Gowda D, Kawamura K. Seasonal variations of low molecular weight hydroxy-dicarboxylic acids and oxaloacetic acid in remote marine aerosols from Chichijima island in the western north Pacific (December 2010-November 2011)[J]. Atmospheric Research, 2018, 204: 128-135. DOI:10.1016/j.atmosres.2018.01.007
[42] Bikkina S, Kawamura K, Sarin M. Secondary organic aerosol formation over coastal ocean:inferences from atmospheric water-soluble low molecular weight organic compounds[J]. Environmental Science & Technology, 2017, 51(8): 4347-4357.
[43] Hegde P, Kawamura K. Seasonal variations of water-soluble organic carbon, dicarboxylic acids, ketocarboxylic acids, and α-dicarbonyls in central Himalayan aerosols[J]. Atmospheric Chemistry and Physics, 2012, 12(14): 6645-6665. DOI:10.5194/acp-12-6645-2012
[44] Jung J, Tsatsral B, Kim Y J, et al. Organic and inorganic aerosol compositions in Ulaanbaatar, Mongolia, during the cold winter of 2007 to 2008:dicarboxylic acids, ketocarboxylic acids, and α-dicarbonyls[J]. Journal of Geophysical Research, 2010, 115(D22): D22203. DOI:10.1029/2010JD014339
[45] Cheng C L, Wang G H, Zhou B H, et al. Comparison of dicarboxylic acids and related compounds in aerosol samples collected in Xi'an, China during haze and clean periods[J]. Atmospheric Environment, 2013, 81: 443-449. DOI:10.1016/j.atmosenv.2013.09.013
[46] 孟静静, 刘晓迪, 侯战方, 等. 菏泽市冬季PM2.5中二元羧酸类SOA的昼夜变化特征[J]. 环境科学, 2018, 39(11): 4876-4884.
Meng J J, Liu X D, Hou Z F, et al. Diurnal variation of dicarboxylic acids and related SOA in PM2.5 from Heze City in winter[J]. Environmental Science, 2018, 39(11): 4876-4884.
[47] Fu P Q, Kawamura K, Chen J, et al. Secondary production of organic aerosols from biogenic VOCs over Mt. Fuji, Japan[J]. Environmental Science & Technology, 2014, 48(15): 8491-8497.
[48] Pavuluri C M, Kawamura K. Enrichment of 13 C in diacids and related compounds during photochemical processing of aqueous aerosols:new proxy for organic aerosols aging[J]. Scientific Reports, 2016, 6: 36467. DOI:10.1038/srep36467
[49] Deshmukh D K, Kawamura K, Deb M K, et al. Sources and formation processes of water-soluble dicarboxylic acids, ω-oxocarboxylic acids, α-dicarbonyls, and major ions in summer aerosols from eastern central India[J]. Journal of Geophysical Research, 2017, 122(6): 3630-3652.
[50] Wang J Y, Wang G H, Gao J, et al. Concentrations and stable carbon isotope compositions of oxalic acid and related SOA in Beijing before, during, and after the 2014 APEC[J]. Atmospheric Chemistry and Physics, 2017, 17(2): 981-992. DOI:10.5194/acp-17-981-2017
[51] He N N, Kawamura K, Okuzawa K, et al. Diurnal and temporal variations of water-soluble dicarboxylic acids and related compounds in aerosols from the northern vicinity of Beijing:implication for photochemical aging during atmospheric transport[J]. Science of the Total Environment, 2014, 499: 154-165. DOI:10.1016/j.scitotenv.2014.08.050
[52] Mochizuki T, Kawamura K, Miyazaki Y, et al. Secondary formation of oxalic acid and related organic species from biogenic sources in a larch forest at the northern slope of Mt. Fuji[J]. Atmospheric Environment, 2017, 166: 255-262. DOI:10.1016/j.atmosenv.2017.07.028
[53] Kawamura K, Okuzawa K, Aggarwal S G, et al. Determination of gaseous and particulate carbonyls (glycolaldehyde, hydroxyacetone, glyoxal, methylglyoxal, nonanal and decanal) in the atmosphere at Mt. Tai[J]. Atmospheric Chemistry and Physics, 2013, 13(10): 5369-5380. DOI:10.5194/acp-13-5369-2013
[54] Kawamura K, Tachibana E, Okuzawa K, et al. High abundances of water-soluble dicarboxylic acids, ketocarboxylic acids and α-dicarbonyls in the mountaintop aerosols over the North China Plain during wheat burning season[J]. Atmospheric Chemistry and Physics, 2013, 13(16): 8285-8302. DOI:10.5194/acp-13-8285-2013
[55] Wang G, Kawamura K, Xie M, et al. Size-distributions of n-alkanes, PAHs and hopanes and their sources in the urban, mountain and marine atmospheres over East Asia[J]. Atmospheric Chemistry and Physics, 2009, 9(22): 8869-8882. DOI:10.5194/acp-9-8869-2009