环境科学  2025, Vol. 46 Issue (2): 786-795   PDF    
引用本文
徐泽伟, 李慧珍, 游静. 珠江广州段水体中紫外吸收剂的污染特征及生态风险[J]. 环境科学, 2025, 46(2): 786-795.
XU Ze-wei, LI Hui-zhen, YOU Jing. Occurrence and Ecological Risk of Ultraviolet Absorbents in Water from the Guangzhou Reach of the Pearl River[J]. Environmental Science, 2025, 46(2): 786-795.
珠江广州段水体中紫外吸收剂的污染特征及生态风险
徐泽伟, 李慧珍, 游静     
暨南大学环境学院, 广东省环境污染与健康重点实验室, 广州 510632
收稿日期: 2024-03-07; 修订日期: 2024-04-20
基金项目: 国家自然科学基金委-广东省联合基金项目（U1901220）
作者简介: 徐泽伟（1998~）, 男, 硕士研究生, 主要研究方向为新污染物的水生态风险评估, E-mail：xuzewei@stu2021.jnu.edu.cn
通信作者: 李慧珍, E-mail：lihuizhen@jnu.edu.cn
摘要: 紫外吸收剂（UVAs）是个人护理产品和工业产品中常见的一类添加剂. 因其伪持久性、生物累积性和毒性, UVAs的环境污染和风险已引起人们的广泛关注, 然而目前对城市水环境中UVAs污染特征和生态风险的了解仍然较少. 为探究水环境中UVAs的赋存、组成和空间分布特征, 分析了珠江广州段40个水样（包括水相和悬浮颗粒相）中常见UVAs的浓度, 并利用风险商评估其潜在水生态风险. 结果表明, 珠江广州段水相和颗粒相中UVAs被广泛检出, 两相中总浓度(∑8UVA, 范围及平均值±标准差, ng∙L-1)分别为6.52~85.8（2.80 ± 6.34）和5.69~461（6.57 ± 34.1）. UVAs在水相中的占比范围为16%~99%, 疏水性强的化合物更易于分配在颗粒相. 珠江广州段下游水体中UVAs的浓度高于上游, 娱乐场所、工业区和污水处理厂附近水体中UVAs检出浓度较高. 个人护理品添加剂2-氰基-3, 3-二苯基丙烯酸-2-乙基己酯（OCR）在水相和颗粒相中均是最主要的UVA, 检出浓度分别占总浓度的50%和60%. 风险商的结果显示珠江广州段水体中目标UVAs的水生态风险处于低到中等水平, 亟需关注其长期暴露风险.
关键词: 珠江      水环境      紫外吸收剂（UVAs）      分布特征      生态风险     
Occurrence and Ecological Risk of Ultraviolet Absorbents in Water from the Guangzhou Reach of the Pearl River
XU Ze-wei , LI Hui-zhen , YOU Jing     
Guangdong Key Laboratory of Environmental Pollution and Health, School of Environment, Jinan University, Guangzhou 510632, China
Abstract: Ultraviolet absorbents (UVAs) are common additives in personal care and industrial products. Due to their pseudo-persistence, bioaccumulation, and toxicity, environmental pollution and the risks associated with UVAs have raised widespread concern. However, current understanding of the pollution characteristics and ecological risks of UVAs in urban aquatic environments is still limited. To investigate the occurrence, composition, and spatial distribution characteristics of UVAs in the aquatic environment, eight UVAs in 40 paired water and suspended particulate matter (SPM) samples in the Guangzhou reach of the Pearl River were analyzed, and their potential aquatic ecological risks were further assessed using risk quotients. The results showed that UVAs were widely detected in the aqueous and SPM phases of the Guangzhou reach of the Pearl River, with total concentrations (∑8UVA, range and mean ± standard deviation, ng∙L-1) of 6.52-85.8 (2.80 ± 6.34) and 5.69-461 (6.57 ± 34.1), respectively. The concentrations of individual UVAs in the aqueous phase accounted for 16% to 99% of the sum concentrations in the aqueous and SPM phases. Specifically, two UVAs (2-ethylhexyl 4-(dimethylamino)benzoate (ODPABA) and 2-(2-hydroxy-5-methylphenyl)benzotriazole (UV-P)) were almost exclusively (97% and 99% on average, respectively) distributed in the aqueous phase, one UVA (2-(2H-benzotriazol-2-yl)-4-(1, 1, 3, 3-tetramethylbutyl)phenol (UV-329)) was mostly (84% on average) distributed in the SPM, and the remaining UVAs were similarly distributed to the aqueous and SPM phases, with concentrations in the aqueous phase accounting for 42%-74% of the sum concentrations in both phases. Concentrations of UVAs in the downstream were generally higher than those in the upstream. Compared with that in other regions, the pollution of UVAs in the waterbodies from the Guangzhou reach of the Pearl River was at a moderate to high level. UVAs were detected at higher concentrations in waterbodies near recreational sites, industrial areas, and wastewater treatment plants. The personal care additive 2-cyano-3, 3-diphenyl acrylic acid-2-ethylhexyl ester (OCR) was the most dominant component in the aqueous and SPM phases, with detection frequencies of 98% and 100%, respectively, and its concentrations accounting for 50% and 60% of the sum concentrations in both phases, respectively. The risk quotients of individual UVAs in the aqueous phase from the Guangzhou reach of the Pearl River ranged from < 0.1 to 1.08, indicating low to moderate risk. Three UVAs, i.e., ODPABA, OCR, and 2-(2H-benzotriazol-2-yl)-4, 6-di-tert-pentylphenol (UV-328), were the potential chemicals at risk. Moreover, it is urgent to pay attention to their long-term risk to aquatic organisms.
Key words: Pearl River      aquatic environment      ultraviolet absorbents (UVAs)      distribution characteristics      ecological risk     

紫外吸收剂（ultraviolet absorbents, UVAs）作为添加剂被广泛用于个人护理产品和工业产品[1]. 根据不同用途, UVAs主要分为紫外过滤剂和紫外稳定剂[2]. 紫外过滤剂是防晒霜等个人护理产品的重要有效成分, 降低紫外线辐射对人体皮肤的负面影响[3]. 紫外过滤剂在水体中广泛存在, 尤其是在旅游和娱乐活动密集地区, 如沿海旅游水域, 大量使用的防晒霜已成为对水生态系统构成威胁的新污染物[4, 5]. 紫外稳定剂作为添加剂被广泛用于工业产品, 如塑料产品、油漆和粘合剂等, 主要用于保护材料免受紫外线辐射的损害[6, 7]. 近年来, UVAs在全球范围的生产和消费量不断增加, 2014年欧洲、美洲和亚太地区的防晒产品市场零售额已达70亿欧元[8]. 大量使用的UVAs可通过旅游活动和污水排放等途径进入水体, 对藻类、贻贝和珊瑚等水生生物造成不利影响[9 ~ 14]. 由于具有伪持久性、生物累积性和毒性, UVAs水生态风险引起了广泛关注, 部分UVAs已被欧洲化学管理局列为高度关注物质[15], 对水环境的风险不容小觑[16]. 探究紫外过滤剂和紫外稳定剂这两类具有不同用途的UVAs在水环境中的赋存特征和生态风险, 有利于对该类化学品进行合理的监督管理.

多数UVAs具有较强疏水性, 排放进入水环境后倾向于被颗粒物吸附, 改变其在水体中的分布特征, 目前关于水体中UVAs的污染研究主要集中在水相, 对其在水相和颗粒相的分布情况了解甚少[17 ~ 19]. 珠江是中国的第二大河流, 珠江三角洲是中国城市化和工业化程度最高的地区之一[20]. 珠江流域的西江和北江以及珠江河口的水体中UVAs被广泛检出, 总浓度可达720 ng∙L-1, 可能对水生态系统造成低到中等程度的风险[15, 21]. 广州作为珠江三角洲的中心城市之一, 城市化发展程度高, 污水处理厂密集, 人类活动可能造成严重污染, 然而目前珠江广州段地表水中UVAs的污染情况和水生态风险尚不清晰. 因此, 本研究的目的是探究珠江广州段中UVAs（包括紫外过滤剂和紫外稳定剂）的赋存和分布特征, 并利用风险商（risk quotient, RQ）评估其潜在水生态风险, 以期为有效管控UVAs污染和风险提供基础信息.

1 材料与方法 1.1 化学品和试剂

基于前期研究选择了水体中常被检出的8种UVAs作为目标化合物[10, 21 ~ 23], 其基本信息列于表 1. 目标化合物、回收率指示物（苊-d10、菲-d10和-d12）和内标[2-（苯并三唑-2-基）-4-甲基苯酚和6-乙酰-1, 1, 2, 4, 4, 7-六甲基四氢萘-d3]的标准品购于AccuStandard（康涅狄格州, 美国）, 纯度大于98%. 色谱级甲醇、正己烷和二氯甲烷购自上海安谱公司, 亲水亲脂平衡吸附剂（hydrophilic-lipophilic balance, HLB）、弱阴离子交换吸附剂（weak anion exchange, WAX）和弱阳离子交换吸附剂（weak cation exchange, WCX）以及聚丙烯固相萃取管（solid phase extraction, SPE）（6 mL）购于天津博纳艾杰尔科技有限公司.

表 1 紫外吸收剂的基本信息 Table 1 Basic information of ultraviolet absorbents

1.2 样品采集

样品采集于2020年11月, 从珠江广州段的上游至珠江口共设置了40个采样点, 在珠江主河道上布置了9个采样点（S1、S9~S11和S16~S19）, 其余点位分布于周围的典型河涌（图 1）. 从地理上看, 珠江广州段流经商业区、工业区和农业区直至珠江口, 途经区域人口密集, 塑料、纺织等工业众多, UVAs可通过娱乐活动、城市排水和工业废水等多种途径进入珠江广州段[16]. 在每个点位采集4 L水样储存于棕色玻璃瓶中（采样前使用甲醇和超纯水各润洗3遍）, 随后将这些水样运送至实验室4 ℃避光储存, 并在7 d内完成样品前处理.

图 1 珠江广州段采样点示意 Fig. 1 Sampling sites in the Guangzhou reach of the Pearl River

1.3 样品前处理与仪器分析

水样经0.7 μm玻璃纤维滤膜（GF/F, Whatman公司, Maidstone, 英国）过滤后, 获得水相和颗粒相样品. 水相样品加入50 ng回收率指示物（苊-d10、菲-d10和-d12）后, 利用SPE（200 mg, HLB∶WAX∶WCX = 2∶1∶1）进行富集萃取. 萃取前, 加入5 mL甲醇和5 mL纯水活化平衡SPE柱, 随后以2~3 mL∙min-1的速度载入水样, 结束载样后维持真空状态30 min. 利用8 mL正己烷和二氯甲烷混合溶剂（体积比为1∶1）洗脱目标化合物. 洗脱液经氮吹浓缩、置换溶剂为正己烷、过无水硫酸钠除水后, 加内标定容至0.5 mL待仪器分析. 颗粒相样品经冷冻干燥后, 利用快速溶剂萃取仪（ASE-350, Thermo Fisher公司, 美国）萃取. 将含有颗粒物的玻璃纤维滤膜剪碎后加入萃取池, 同时加入硅藻土填满萃取池. 萃取溶剂为混合溶剂（正己烷∶丙酮∶二氯甲烷=2∶2∶1）, 添加50 ng回收率指示物后在温度为100 ℃, 压力为10 350 kPa（1 500 psi）的条件下重复静态萃取3次, 每次5 min. 萃取液浓缩并置换溶剂为正己烷, 使用硅胶氧化铝层析柱净化, 层析柱自下而上依次为6 cm氧化铝、12 cm硅胶和2 cm无水硫酸钠. 利用70 mL正己烷和二氯甲烷混合溶剂（体积比为7∶3）洗脱目标化合物, 洗脱液氮吹浓缩、置换溶剂为正己烷后, 加内标定容至0.5 mL待仪器分析.

利用气相色谱-质谱联用仪（GCMS-QP2020, Shimadzu公司, 日本）在电子轰击电离模式下定量分析样品中UVAs, 采用HP-5MS色谱柱（安捷伦公司, 30 m × 0.25 mm, 膜厚0.25 μm）进行分离, 载气为氦气, 流速设定为1.5 mL∙min-1. 采用脉冲不分流模式进样, 压力为250 kPa. 进样器、接口和离子源的温度分别设为260、260和230 ℃. 柱温程序设置如下：初始柱温为60 ℃, 以10 ℃∙min-1的速率升温至200 ℃, 以2 ℃∙min-1的速率升温至214 ℃, 再以5 ℃的速率升温至250 ℃, 随后保持1 min, 以20 ℃∙min-1的速率升温至290 ℃并保持7 min.

1.4 质量保证与控制

在样品前处理过程中, 每20个样品同时处理4个质量控制样品, 包括溶剂空白、基质空白、基质加标和基质加标平行样. 在溶剂空白和基质空白中均未检出目标化合物, 水相基质加标中UVAs的回收率为55%~78%（平均值±标准偏差：65% ± 6%）, 颗粒相基质加标中UVAs的回收率为53%~111%（78% ± 17%）. 所有样品萃取前加入回收率指示物, 以评估样品前处理过程是否合格. 3个回收率指示物（苊-d10、菲-d10和-d12）在水样中的回收率分别为56% ± 12%、56% ± 12%和102% ± 15%, 在颗粒相样中的回收率分别为81% ± 32%、70% ± 19%和76% ± 10%. 为了检查仪器分析过程中仪器的稳定性, 每分析10个样品后分析一个标样, 标样中所有化合物的响应与标准曲线中该化合物的响应差异均低于20%. 仪器方法的检出限（limits of detection, LOD）和定量限（limits of quantification, LOQ）分别定义为当信噪比为3和10时对应目标化合物的浓度（表 2）. 样品中目标化合物的报告检出限（reporting limits, RL）计算如下：
表 2 目标化合物的标准曲线线性范围、相关系数（r2）、仪器方法检出限、定量限以及报告检出限 Table 2 Linear ranges, correlation coefficients (r2), instrument method limits of detection (LOD), limits of quantification (LOQ), and reporting limits (RL) of the target compounds

式中, C为标准曲线的最低浓度（2 ng·mL-1）, V1为用于萃取的水样体积（4 L）, V2为仪器分析时样品萃取液的体积（0.5 mL）. 样品中可检出但浓度低于RL的目标化合物, 在后期数据分析中, 其浓度计为1/2 RL[2].

1.5 数据分析

利用风险商（risk quotient, RQ）评估珠江广州段水体中UVAs的潜在生态风险, 公式如下：

式中, MEC为实测环境浓度（measured environmental concentration）, PNEC为预测无效应浓度（predicted no effect concentration）. 目前UVAs的PNEC多根据代表性模式生物的毒性阈值, 如半数致死浓度（50% lethal concentration, LC50）或半数效应浓度（50% effect concentration, EC50）计算得到. 根据本研究样品特征, 选择藻类、大型溞和鱼类为模式生物. 通过检索文献或利用ECOSAR（Ecological Structure-Activity Relationships Program, https://www.epa.gov/tsca-screening-tools/ecological-structure-activity-relationships-ecosar-predictive-model）预测获得UVAs的LC50或EC50[30]. 根据欧盟风险评估指南, 采用评估因子法（assessment factor, AF）获取PNEC, 公式如下：

考虑种内、种间和急慢性暴露差异, 本研究中AF取1 000[31]. 目标UVAs的毒性阈值和PNEC如表 3所示. 风险级别划分方式如下：RQ≥10表示高风险, 1≤RQ < 10表示中风险, 0.3≤RQ < 1表示低风险, RQ < 0.3表示无风险[32].

表 3 紫外吸收剂对藻类、大型溞和鱼类的毒性阈值和预测无效应浓度1） Table 3 Toxicity thresholds and predicted no effect concentrations (PNEC) of the ultraviolet absorbents to algae, Daphnia magna, and fish

利用IBM SPSS Statistics 22.0软件进行Kruskal-Wallis非参数检验, 确定不同UVAs在水相的占比是否存在显著性差异（α = 0.05）.

2 结果与讨论 2.1 污染水平

珠江广州段水相和颗粒相中目标UVAs被广泛检出, 检出率分别为35%~98%和18%~100%, 两相总浓度(∑8UVA，范围及平均值±标准差，ng∙L-1)为17.0~480（74.9 ± 92.4）（图 2和表 4）. 虽然总浓度(∑8UVA)在水相和颗粒相的占比相当, 但不同UVAs的占比差异较大（表 4）. 两个UVAs（ODPABA和UV-P）主要分布在水相, 水相平均占比分别高达97%和99%；3个UVAs（4-MBC、UV-326和UV-327）倾向于分布在水相, 水相平均占比为63%~74%；两个UVAs（OCR和UV-328）在水相和颗粒相的占比相当, 而UV-329更倾向于吸附在颗粒物, 颗粒相占比的平均值高达84%. 两相分布结果与化合物疏水性相关, 疏水性越强, 越倾向于吸附在颗粒物（表 1）. 监测不同相中污染物的赋存水平有助于更全面地了解污染状况, 避免遗漏潜在的风险物质.

图 2 珠江广州段水相和颗粒相中紫外吸收剂的浓度水平 Fig. 2 Concentrations of ultraviolet absorbents in the aqueous and particulate phases from Guangzhou reach of the Pearl River

表 4 不同紫外吸收剂在水相和颗粒相的总浓度以及水相和颗粒相占比 Table 4 Total concentrations of different ultraviolet absorbents in aqueous and particulate phases and percentage of aqueous and particulate phases

为进一步了解珠江广州段UVAs的污染水平, 总结了不同区域水相中UVAs的赋存情况（表 5）, 由于各研究检测的化合物种类不同, 表中列出了具体UVAs的浓度. 珠江广州段中UVAs的总浓度平均值与中国香港、加拿大圣劳伦斯河及印度卡维里河接近, 但比中国巢湖、中国南海以及美国切萨皮克湾约低一个数量级（表 5）. 从同类型UVAs的赋存情况来看, 珠江广州段ODPABA的浓度高于中国巢湖, 但低于中国南海和中国香港水体中的浓度[10, 21, 37, 38], 4-MBC的浓度与中国巢湖的接近[21, 37], OCR的浓度与中国巢湖、美国切萨皮克湾以及中国香港水体中的浓度接近, 高于加拿大圣劳伦斯河的水浓度[10, 16, 37, 39], UV-P、UV-326、UV-327、UV-328和UV-329的浓度则与印度卡维里河和中国巢湖的水浓度相近[22, 37]. 值得注意的是, 珠江流域不同时间采集样品中UVAs污染水平和组成亦有较大差异, 2020年采集的珠江广州段水体中ODPABA和4-MBC的污染情况相较于2018年珠江河口/南海更为严重（表 5）. 珠江广州段受城市径流和工业废水影响严重, 强调了对流经人流密集区和发达城市的河流进行监测的重要性[15, 19]. 作为常见有机聚合物添加剂的UV-328和UV-329, 在珠江广州段的污染水平比2018年采集的珠江流域（西江和北江）水体低两个数量级[15, 18], 可能是由于西江和北江流经区域的工业类型占比与广州不同[40 ~ 42]. 总体而言, 珠江广州段水体中UVAs的污染处于中等偏上水平, 污染程度受区域产业类型和地理环境影响[43]. 相对于水相污染, 颗粒相中UVAs的研究较少, 珠江广州段水体颗粒物中紫外稳定剂的浓度明显低于中国黄海和东海水体颗粒物中的浓度[44].

表 5 不同区域水相中紫外吸收剂的浓度 Table 5 Concentrations of ultraviolet absorbents in surface water in different areas

2.2 空间分布特征

本研究采样点始于珠江广州段上游, 经商业区、居住区、工业区、农业种植区和水产养殖区等混合区域, 止于珠江口（图 1）. 整体而言, 河流上游（S1~S22）和中游河段（S23~S30）水体中UVAs的含量低于靠近河口的下游河段（S31~S40）（图 2）. 珠江主河道（S1、S9~S11和S16~S19）污染水平较低, S9和S16分别位于沙面公园和琶洲岛下游, 人类娱乐活动可能是其浓度较高的原因. 娱乐活动是UVAs的重要来源之一, 例如, 作为娱乐和旅游胜地的马略卡岛的地表水中, ρ（4-MBC）高达（113 ± 7）ng∙L-1[45], 而在中国合肥塘西河沙滩公园下游, OCR、4-MBC和EHMC等UVAs的总浓度同样超过了490 ng∙L-1[46]. 石井河（S2、S3和S4）、沙河涌（S7和S8）和市桥河（S23、S24和S25）水体中UVAs浓度较低, 而花地河上游（S5）及下游（S6）的污染可能来源于附近公园和城市排水公司. 含有UVAs的工业或生活废水是城市河流的重要污染源, 例如, 加拿大多伦多的降雨和融雪过程中所带来的城市径流导致大量紫外稳定剂流入安大略湖, UV-P在颗粒相中的浓度最高达800 ng·g-1[47]. 车陂涌上游（S12和S13）污染较轻, 下游污染的潜在来源为附近的污水处理厂（S14）和公园（S15）. 与上游（S1~S22）和中游（S23~S30）相比, 下游部分点位（S32、S33和S35）污染更为严重, 这可能与采样点附近的工业园（S32, 冶炼、印染和化工）、污水处理厂（S32）和公园（S35）有关. 珠江广州段污水处理厂附近（S14）水体中以OCR和4-MBC为主, 这两个化合物也是前期研究报道的污水处理厂出水中最常被检出的UVAs[31, 48 ~ 52], 说明污水处理厂出水确实是受纳河流中UVAs的重要来源. 河口汇聚处（S40）水相UVAs浓度最高, 该处汇聚多条河流导致污染物聚集, 与附近湿地公园等娱乐场所共同作用导致了该点的高污染.

2.3 组成特征

根据用途可将UVAs分为紫外过滤剂和紫外稳定剂, 紫外过滤剂主要用于个人护理产品, 包括ODPABA、OCR和4-MBC等, 其中OCR和4-MBC是防晒化妆品的推荐成分, 在中国最大允许剂量分别为10%和4%[37]. 而紫外稳定剂主要用于工业产品, 包括UV-P、UV-326、UV-327、UV-328和UV-329等[7]. 整体而言, 紫外过滤剂的浓度[∑3UVA的水相和颗粒相浓度（范围及平均值±标准差, ng∙L-1）分别为3.06~74.0（19.3 ± 16.2）和2.76~458(47.5 ± 92.2)]显著高于紫外稳定剂[∑5UVA的水相和颗粒相浓度（范围及平均值±标准差, ng∙L-1）分别为0~24.1（3.11 ± 4.07）和0~21.2(5.03 ± 3.91)]. 紫外过滤剂OCR是珠江广州段水相和颗粒相中检出率和浓度最高的UVAs（图 3）, 在两相中的检出率分别为98%和100%, 检出浓度占∑8UVA的比例分别为50% ± 18%和60% ± 26%, 显著高于其他UVAs（P < 0.05）. OCR属于肉桂酸酯类UVAs, 紫外线吸收能力强[53], 在2018年珠江河口水体的相关UVAs研究中同样具有最高的检出浓度（1.0~140 ng∙L-1）[21]. Zhu等[54]估算2012年中国OCR的使用量和排放量分别为206 t和116 t. OCR进入水体后不易被光解、水解（水解半衰期可长达206 a）或生物降解[21, 55], 且疏水性较强（lg Kow为6.9）, 易于吸附在悬浮颗粒物上. 前期研究显示, OCR可对藻类、海胆等海洋生物的生长产生不良影响, 进而破坏水生态系统的正常结构和功能[14]. 水体中广泛赋存且具有环境持久性和毒性的OCR可能威胁珠江水生态健康.

（a）水相, （b）颗粒相 图 3 珠江广州段水相和颗粒相中紫外吸收剂的组成 Fig. 3 Composition of ultraviolet absorbents in the aqueous and particulate phases from Guangzhou reach of the Pearl River

ODPABA是珠江广州段水相中另一具有高频检出率（98%）的UVAs, 检出浓度占水相∑8UVA的比例为23% ± 15%, 但该化合物在颗粒相中检出率仅为18%, 检出浓度占颗粒相∑8UVA的比例仅为1% ± 2%. ODPABA属于对氨基苯甲酸酯类UVAs, 是应用最早的紫外过滤剂之一, 作为护理产品组成成分具有高额消费量[4], 在中国南海和中国香港等水体中被广泛检出[10, 38]. ODPABA在进入水体后会与水中的消毒剂或有机物反应生成具有更高水生态风险的转化产物, 对水环境的威胁同样不容忽视[26]. 4-MBC属于樟脑衍生物类UVAs, 是常用的紫外过滤剂之一[13], 检出浓度在水相和颗粒相的占比分别为12% ± 12%和12% ± 27%. 紫外稳定剂UV-P、UV-326、UV-327、UV-328和UV-329属于苯并三唑类UVAs, 常被用于塑料和合成材料中以降低材料受紫外线辐射的影响[7, 37, 56]. 水体中最常检出的紫外稳定剂为UV-P（78%）, 占比为8% ± 10%, 其次为UV-329（70%）、UV-326（55%）、UV-328（43%）和UV-327（35%）, 占比为1%~2%. 在颗粒相中UV-329被广泛检出（100%）, 占比为19% ± 17%, 仅低于OCR. 可见, 个人护理产品添加剂是该区域水环境中UVAs的主要来源.

2.4 潜在水生态风险评估

利用商值法评估了珠江广州段水体中UVAs的潜在水生态风险, 结果如图 4所示. 利用基于藻类毒性阈值的PNEC评估水生态风险, 10%水样中的ODPABA具有低到中等风险, 最高RQ值为1.08；8%水样中OCR表现低风险, 其他UVAs均无风险. 利用基于大型溞毒性阈值的PNEC评估水生态风险, 5%水样中UV-328表现低风险, 其他UVAs均无风险. 利用基于鱼类毒性阈值的PNEC评估水生态风险, 20%水样中OCR具有低到中等风险, 其他UVAs均无风险. 可见, 珠江广州段中ODPABA、OCR和UV-328可能危害水生态健康. 本研究风险评估结果与前期研究一致, 如珠江河口水体中OCR具有较高的潜在风险[21], 珠江支流（北江和西江）旱季水样中UV-328具有较高风险[15]；巢湖中OCR对藻类具有低到中度风险[36].

（a）预测无效应浓度基于藻类毒性阈值, （b）预测无效应浓度基于大型溞毒性阈值, （c）预测无效应浓度基于鱼类毒性阈值 图 4 珠江广州段水体中紫外吸收剂的风险商 Fig. 4 Risk quotients (RQ) of ultraviolet absorbents from water bodies in Guangzhou reach of the Pearl River

当前UVAs的水生态风险阈值缺乏, 开展水生态风险评估时多通过一种模式生物的毒性阈值外推获得风险阈值, 使用单一物种毒性阈值可能遗漏一些潜在风险物质, 推荐使用多物种毒性阈值进行评估. 针对这一问题, 亟需开展本地物种毒性测试, 建立物种敏感性分布曲线, 获取具有区域特征和环境相关性的水生态风险阈值, 提高UVAs水生态风险评估的准确性.

3 结论

本文探究了珠江广州段水体中典型UVAs的赋存、组成和生态风险, 发现UVAs在珠江广州段水体和颗粒相中被广泛检出, 总浓度(∑8UVA)在两相的占比相近, 但不同UVAs的两相组成存在较大差异, ODPABA和UV-P主要分配在水相, 而UV-329更倾向于吸附在颗粒相；从空间分布看, 珠江广州段下游水相和颗粒相中UVAs的浓度略高于上游, 人类娱乐活动、工业废水和污水处理厂可能是水体的主要污染来源. 个人护理产品添加剂OCR在水相和颗粒相中均为主要组成；利用风险商评估了水相UVAs的水生态风险, 3个UVAs（ODPABA、OCR和UV-328）是珠江广州段水体中的潜在水生态风险物质.

参考文献
[1] Santos A J M, Miranda M S, Esteves Da Silva J C G. The degradation products of UV filters in aqueous and chlorinated aqueous solutions[J]. Water Research, 2012, 46(10): 3167-3176. DOI:10.1016/j.watres.2012.03.057
[2] Hu H, Li Y, Lu G Y, et al. Spatiotemporal trends of ultraviolet absorbents in oysters from the Pearl River Estuary, South China during 2015-2020[J]. Environmental Pollution, 2023, 323. DOI:10.1016/j.envpol.2023.121298
[3] Lam T K, Law J C F, Leung K S Y. Hybrid radical coupling during MnO2-mediated transformation of a mixture of organic UV filters: chemistry and toxicity assessment[J]. Science of the Total Environment, 2024, 915. DOI:10.1016/j.scitotenv.2024.170121
[4] Tovar-Sánchez A, Sánchez-Quiles D, Rodríguez-Romero A. Massive coastal tourism influx to the Mediterranean Sea: the environmental risk of sunscreens[J]. Science of the Total Environment, 2019, 656: 316-321. DOI:10.1016/j.scitotenv.2018.11.399
[5] 朱小山, 黄静颖, 吕小慧, 等. 防晒剂的海洋环境行为与生物毒性[J]. 环境科学, 2018, 39(6): 2991-3002.
Zhu X S, Huang J Y, Lü X H, et al. Fate and toxicity of UV filters in marine environments[J]. Environmental Science, 2018, 39(6): 2991-3002.
[6] Nakata H, Shinohara R I, Murata S, et al. Detection of benzotriazole UV stabilizers in the blubber of marine mammals by gas chromatography-high resolution mass spectrometry (GC-HRMS)[J]. Journal of Environmental Monitoring, 2010, 12(11): 2088-2092. DOI:10.1039/c0em00170h
[7] Fent K, Chew G, Li J, et al. Benzotriazole UV-stabilizers and benzotriazole: antiandrogenic activity in vitro and activation of aryl hydrocarbon receptor pathway in zebrafish eleuthero-embryos[J]. Science of the Total Environment, 2014, 482-483: 125-136. DOI:10.1016/j.scitotenv.2014.02.109
[8] Osterwalder U, Sohn M, Herzog B. Global state of sunscreens[J]. Photodermatology, Photoimmunology & Photomedicine, 2014, 30(2-3): 62-80.
[9] 郑少奎, 李晓锋. 城市污水处理厂出水中的药品和个人护理品[J]. 环境科学, 2013, 34(8): 3316-3326.
Zheng S K, Li X F. Pharmaceuticals and personal care products (PPCPs) in the effluent of sewage treatment plants[J]. Environmental Science, 2013, 34(8): 3316-3326.
[10] Tsui M M P, Lam J C W, Ng T Y, et al. Occurrence, distribution, and fate of organic UV filters in coral communities[J]. Environmental Science & Technology, 2017, 51(8): 4182-4190.
[11] Wang L, Zhang J J, Sun H W, et al. Widespread occurrence of benzotriazoles and benzothiazoles in tap water: Influencing factors and contribution to human exposure[J]. Environmental Science & Technology, 2016, 50(5): 2709-2717.
[12] Tangtian H, Bo L, Wenhua L, et al. Estrogenic potential of benzotriazole on marine medaka (Oryzias melastigma)[J]. Ecotoxicology and Environmental Safety, 2012, 80: 327-332. DOI:10.1016/j.ecoenv.2012.03.020
[13] Cuccaro A, De Marchi L, Oliva M, et al. Ecotoxicological effects of the UV-filter 4-MBC on sperms and adults of the mussel Mytilus galloprovincialis [J]. Environmental Research, 2022, 213. DOI:10.1016/j.envres.2022.113739
[14] Giraldo A, Montes R, Rodil R, et al. Ecotoxicological evaluation of the UV filters ethylhexyl dimethyl p-aminobenzoic acid and octocrylene using marine organisms Isochrysis galbana, Mytilus galloprovincialis and Paracentrotus lividus [J]. Archives of Environmental Contamination and Toxicology, 2017, 72(4): 606-611. DOI:10.1007/s00244-017-0399-4
[15] Hu L X, Cheng Y X, Wu D, et al. Continuous input of organic ultraviolet filters and benzothiazoles threatens the surface water and sediment of two major rivers in the Pearl River Basin[J]. Science of the Total Environment, 2021, 798. DOI:10.1016/j.scitotenv.2021.149299
[16] Castilloux A D, Houde M, Gendron A, et al. Distribution and fate of ultraviolet absorbents and industrial antioxidants in the St. Lawrence River, Quebec, Canada[J]. Environmental Science & Technology, 2022, 56(8): 5009-5019.
[17] Huang W X, Xie Z Y, Yan W, et al. Occurrence and distribution of synthetic musks and organic UV filters from riverine and coastal sediments in the Pearl River Estuary of China[J]. Marine Pollution Bulletin, 2016, 111(1-2): 153-159. DOI:10.1016/j.marpolbul.2016.07.018
[18] Wick A, Jacobs B, Kunkel U, et al. Benzotriazole UV stabilizers in sediments, suspended particulate matter and fish of German rivers: new insights into occurrence, time trends and persistency[J]. Environmental Pollution, 2016, 212: 401-412. DOI:10.1016/j.envpol.2016.01.024
[19] 彭珂醒, 李瑞飞, 周亦辰, 等. 地表水悬浮态多环芳烃时空变化特征及主要输入源响应机制[J]. 环境科学, 2022, 43(7): 3645-3655.
Peng K X, Li R F, Zhou Y C, et al. Spatiotemporal distribution and source apportionment of suspended polycyclic aromatic hydrocarbons in surface water[J]. Environmental Science, 2022, 43(7): 3645-3655.
[20] 吕晓立, 刘景涛, 韩占涛, 等. 快速城镇化进程中珠江三角洲硝酸型地下水赋存特征及驱动因素[J]. 环境科学, 2021, 42(10): 4761-4771.
Lü X L, Liu J T, Han Z T, et al. Geochemical characteristics and driving factors of NO3-type groundwater in the rapidly urbanizing Pearl River Delta[J]. Environmental Science, 2021, 42(10): 4761-4771.
[21] Fisch K, Zhang R F, Zhou M, et al. PPCPs - A human and veterinary fingerprint in the Pearl River delta and northern South China Sea[J]. Emerging Contaminants, 2021, 7: 10-21. DOI:10.1016/j.emcon.2020.11.006
[22] Vimalkumar K, Arun E, Krishna-Kumar S, et al. Occurrence of triclocarban and benzotriazole ultraviolet stabilizers in water, sediment, and fish from Indian rivers[J]. Science of the Total Environment, 2018, 625: 1351-1360. DOI:10.1016/j.scitotenv.2018.01.042
[23] 李富娟, 高礼, 李凌云, 等. 宁夏第三排水沟中药物和个人护理品(PPCPs)的污染特征与生态风险评估[J]. 环境科学, 2022, 43(8): 4087-4096.
Li F J, Gao L, Li L Y, et al. Contamination characteristics and ecological risk assessment of pharmaceuticals and personal care products (PPCPs) in the third drain of Ningxia[J]. Environmental Science, 2022, 43(8): 4087-4096.
[24] Gago-Ferrero P, Díaz-Cruz M S, Barceló D. An overview of UV-absorbing compounds (organic UV filters) in aquatic biota[J]. Analytical and Bioanalytical Chemistry, 2012, 404(9): 2597-2610. DOI:10.1007/s00216-012-6067-7
[25] Apel C, Joerss H, Ebinghaus R. Environmental occurrence and hazard of organic UV stabilizers and UV filters in the sediment of European north and Baltic Seas[J]. Chemosphere, 2018, 212: 254-261. DOI:10.1016/j.chemosphere.2018.08.105
[26] Studziński W, Gackowska A, Kudlek E. Determination of environmental properties and toxicity of octyl-dimethyl-para-aminobenzoic acid and its degradation products[J]. Journal of Hazardous Materials, 2021, 403. DOI:10.1016/j.jhazmat.2020.123856
[27] Rodríguez-Escales P, Sanchez-Vila X. Modeling the fate of UV filters in subsurface: co-metabolic degradation and the role of biomass in sorption processes[J]. Water Research, 2020, 168. DOI:10.1016/j.watres.2019.115192
[28] Cantwell M G, Sullivan J C, Katz D R, et al. Source determination of benzotriazoles in sediment cores from two urban estuaries on the Atlantic Coast of the United States[J]. Marine Pollution Bulletin, 2015, 101(1): 208-218. DOI:10.1016/j.marpolbul.2015.10.075
[29] Leubner N, Pawlowski S, Salinas E R, et al. Assessment of the bioaccumulation potential of four commonly used phenolic benzotriazoles based on in silico and experimental in vivo data[J]. Journal of Applied Toxicology, 2023, 43(9): 1272-1283. DOI:10.1002/jat.4461
[30] Reuschenbach P, Silvani M, Dammann M, et al. ECOSAR model performance with a large test set of industrial chemicals[J]. Chemosphere, 2008, 71(10): 1986-1995. DOI:10.1016/j.chemosphere.2007.12.006
[31] Tsui M M P, Leung H W, Lam P K S, et al. Seasonal occurrence, removal efficiencies and preliminary risk assessment of multiple classes of organic UV filters in wastewater treatment plants[J]. Water Research, 2014, 53: 58-67. DOI:10.1016/j.watres.2014.01.014
[32] Wang Y Y L, Xiong J J, Ohore O E, et al. Deriving freshwater guideline values for neonicotinoid insecticides: implications for water quality guidelines and ecological risk assessment[J]. Science of the Total Environment, 2022, 828. DOI:10.1016/j.scitotenv.2022.154569
[33] Molins-Delgado D, Gago-Ferrero P, Díaz-Cruz M S, et al. Single and joint ecotoxicity data estimation of organic UV filters and nanomaterials toward selected aquatic organisms. Urban groundwater risk assessment[J]. Environmental Research, 2016, 145: 126-134. DOI:10.1016/j.envres.2015.11.026
[34] Paredes E, Perez S, Rodil R, et al. Ecotoxicological evaluation of four UV filters using marine organisms from different trophic levels Isochrysis galbana, Mytilus galloprovincialis, Paracentrotus lividus, and Siriella armata [J]. Chemosphere, 2014, 104: 44-50. DOI:10.1016/j.chemosphere.2013.10.053
[35] Fent K, Kunz P Y, Zenker A, et al. A tentative environmental risk assessment of the UV-filters 3-(4-methylbenzylidene-camphor), 2-ethyl-hexyl-4-trimethoxycinnamate, benzophenone-3, benzophenone-4 and 3-benzylidene camphor[J]. Marine Environmental Research, 2010, 69: S4-S6. DOI:10.1016/j.marenvres.2009.10.010
[36] Park C B, Jang J, Kim S, et al. Single- and mixture toxicity of three organic UV-filters, ethylhexyl methoxycinnamate, octocrylene, and avobenzone on Daphnia magna [J]. Ecotoxicology and Environmental Safety, 2017, 137: 57-63. DOI:10.1016/j.ecoenv.2016.11.017
[37] Tang Z W, Han X, Li G H, et al. Occurrence, distribution and ecological risk of ultraviolet absorbents in water and sediment from Lake Chaohu and its inflowing rivers, China[J]. Ecotoxicology and Environmental Safety, 2018, 164: 540-547. DOI:10.1016/j.ecoenv.2018.08.045
[38] Tsui M M P, Chen L G, He T T, et al. Organic ultraviolet (UV) filters in the South China Sea coastal region: environmental occurrence, toxicological effects and risk assessment[J]. Ecotoxicology and Environmental Safety, 2019, 181: 26-33. DOI:10.1016/j.ecoenv.2019.05.075
[39] He K, Hain E, Timm A, et al. Occurrence of antibiotics, estrogenic hormones, and UV-filters in water, sediment, and oyster tissue from the Chesapeake Bay[J]. Science of the Total Environment, 2019, 650: 3101-3109. DOI:10.1016/j.scitotenv.2018.10.021
[40] 张国俊, 王珏晗, 庄大昌. 广州市产业生态化时空演变特征及驱动因素[J]. 地理研究, 2018, 37(6): 1070-1086.
Zhang G J, Wang J H, Zhuang D C. The characteristics and driving forces of spatial and temporal evolution of industrial ecology in Guangzhou[J]. Geographical Research, 2018, 37(6): 1070-1086.
[41] 严骁, 李淑圆, 王美欢, 等. 电子垃圾拆解工人的肝功能和肾功能健康状况及影响因素分析: 以清远市龙塘镇为例[J]. 环境科学, 2018, 39(2): 953-960.
Yan X, Li S Y, Wang M H, et al. Liver and kidney function of e-waste dismantling workers and potential influencing factors[J]. Environmental Science, 2018, 39(2): 953-960.
[42] 李强, 曹莹, 何连生, 等. 典型冶炼行业场地土壤重金属空间分布特征及来源解析[J]. 环境科学, 2021, 42(12): 5930-5937.
Li Q, Cao Y, He L S, et al. Spatial distribution characteristics and source analysis of soil heavy metals at typical smelting industry sites[J]. Environmental Science, 2021, 42(12): 5930-5937.
[43] Peng X Z, Zhu Z W, Xiong S S, et al. Tissue distribution, growth dilution, and species-specific bioaccumulation of organic ultraviolet absorbents in wildlife freshwater fish in the Pearl River catchment, China[J]. Environmental Toxicology and Chemistry, 2020, 39(2): 343-351.
[44] Zhao M L, Ji X, He Z, et al. Spatial distribution, partitioning, and ecological risk assessment of benzotriazoles, benzothiazoles, and benzotriazole UV absorbers in the eastern shelf seas of China[J]. Water Research, 2024, 248. DOI:10.1016/j.watres.2023.120885
[45] Tovar-Sánchez A, Sánchez-Quiles D, Basterretxea G, et al. Sunscreen products as emerging pollutants to coastal waters[J]. PLos One, 2013, 8(6). DOI:10.1371/journal.pone.0065451
[46] 韩雪, 李广辉, 钟伏勇, 等. 合肥市典型入湖河流有机紫外吸收剂污染特征及生态风险[J]. 环境科学学报, 2018, 38(4): 1569-1578.
Han X, Li G H, Zhong F Y, et al. Contamination and risk of organic ultraviolet filters in main into-lake rivers in Hefei[J]. Acta Scientiae Circumstantiae, 2018, 38(4): 1569-1578.
[47] Parajulee A, Lei Y D, Kananathalingam A, et al. Investigating the sources and transport of benzotriazole UV stabilizers during rainfall and snowmelt across an urbanization gradient[J]. Environmental Science & Technology, 2018, 52(5): 2595-2602.
[48] Ramos S, Homem V, Alves A, et al. A review of organic UV-filters in wastewater treatment plants[J]. Environment International, 2016, 86: 24-44. DOI:10.1016/j.envint.2015.10.004
[49] Moeder M, Schrader S, Winkler U, et al. At-line microextraction by packed sorbent-gas chromatography-mass spectrometry for the determination of UV filter and polycyclic musk compounds in water samples[J]. Journal of Chromatography A, 2010, 1217(17): 2925-2932. DOI:10.1016/j.chroma.2010.02.057
[50] Rodil R, Moeder M. Development of a method for the determination of UV filters in water samples using stir bar sorptive extraction and thermal desorption-gas chromatography-mass spectrometry[J]. Journal of Chromatography A, 2008, 1179(2): 81-88. DOI:10.1016/j.chroma.2007.11.090
[51] Cunha S C, Pena A, Fernandes J O. Dispersive liquid-liquid microextraction followed by microwave-assisted silylation and gas chromatography-mass spectrometry analysis for simultaneous trace quantification of bisphenol A and 13 ultraviolet filters in wastewaters[J]. Journal of Chromatography A, 2015, 1414: 10-21. DOI:10.1016/j.chroma.2015.07.099
[52] Li W H, Ma Y M, Guo C S, et al. Occurrence and behavior of four of the most used sunscreen UV filters in a wastewater reclamation plant[J]. Water Research, 2007, 41(15): 3506-3512. DOI:10.1016/j.watres.2007.05.039
[53] Duis K, Junker T, Coors A. Review of the environmental fate and effects of two UV filter substances used in cosmetic products[J]. Science of the Total Environment, 2022, 808. DOI:10.1016/j.scitotenv.2021.151931
[54] Zhu Y, Price O R, Kilgallon J, et al. A multimedia fate model to support chemical management in China: a case study for selected trace organics[J]. Environmental Science & Technology, 2016, 50(13): 7001-7009.
[55] Rodil R, Moeder M, Altenburger R, et al. Photostability and phytotoxicity of selected sunscreen agents and their degradation mixtures in water[J]. Analytical and Bioanalytical Chemistry, 2009, 395(5): 1513-1524. DOI:10.1007/s00216-009-3113-1
[56] 刘舒娇, 丁剑楠, 石浚哲, 等. 太湖塑料添加剂时空分布和生态风险评价[J]. 环境科学, 2022, 43(5): 2557-2565.
Liu S J, Ding J N, Shi J Z, et al. Spatiotemporal distribution and ecological risk assessment of plastic additives in Taihu Lake[J]. Environmental Science, 2022, 43(5): 2557-2565.