【作者】 杨征军；

【导师】 曾荣今；

【作者基本信息】 湖南科技大学 ， 有机化学， 2007， 硕士

【摘要】 卤代多环芳烃类有机污染物在环境中分布广泛，对生态系统有极大的危害。在评价卤代多环芳烃类化合物在环境中的传播时，它的理化性质对于卤代多环芳烃类化合物的环境暴露评价、生态风险和人类健康评价，具有重要意义。由于标准物缺乏，很多卤代多环芳烃化合物的理化性质实验数据难以获得，只能依靠理论预测。本项研究希望通过建立多卤代多环芳烃类化合物QSPR模型，预测卤代多环芳烃化合物的性质，为研究多卤代多环芳烃类化合物的环境行为提供可借鉴的方法和手段，为此类化合物的生态风险评价提供数据基础，为进一步探讨分子的结构与性质之间的关系提供理论参考。本文以多氯联苯（PCBs）、多氯联苯醚（PCDEs）、多溴联苯醚（PBDEs）、多氯代二苯并二恶英（PCDDs）和多氯代二苯并呋喃（PCDFs）为研究对象，并简要叙述了多卤代多环芳烃类持久性有机污染物的一般性问题。卤代多环芳烃类化合物的QSPR／QSRR研究。首先，获取分子的结构参数：1)从化合物的分子图出发，用分子中基团（分子碎片）的第一电离能对分子图顶点着色，构建分子的邻接矩阵，解矩阵，得特征根，参照休克尔分子轨道理论，指定其中的四个特征根分别为LOMO、HOMO、LUMO和HUMO；2)从化合物的分子图出发，构建分子的距离矩阵，用基团的均衡电负性来对分子顶点着色，构建行矩阵，结合行矩阵和距离矩阵，同时考虑分子中各原子或基团间的作用情况，计算得到一参数，我们定义为拓扑距离电负性指数（TDEI）。其次，采用拓扑距离电负性指数（TDEI）和HOMO、LUMO，结合本工作组前面提出的参数，采用线性回归方法建立PCBs、PCDE、PBDE、PCDDs和PCDFs类化合物的理化性质——过冷液体蒸汽压（P_L⁰）、水溶解度(S_W,L)和辛醇／水分配系数（Kow）等的定量结构-性质相关（QSPR）模型及PCBs的辛醇／空气分配系数（Koa）与坏境温度、分子结构相关（QRSETP）模型。再次，为了考查本文结构参数的适用性，把结构参数应用到PBDE类、PCDDs类和PCDFs类化合物在不同气相色谱柱上的定量结构-色谱保留相关（QSRR）中，结果令人满意。部分模型结果如下：PCBs的辛醇／空气分配系数：logKoa=-15．7714+10．7417TDEI-0．3115HOMO R=0．9913，s=0．1117，F=196．89，n=10，R_CV=0．9685，S_CV=0．1285．PCDE的蒸汽压：-logP_L⁰=-39．5142+16．0350TDEI R=0．9918，s=0．1486，F=6356．32，n=107，R_CV=0．9830，S_CV=0．1501PBDEs的logKoa：logKoa=-65．8659+29．5682 TDEI R=0．9974，s=0．0982，F=3885．27，n=22，R_CV=0．9937，S_CV=0．1039PCDDs的蒸汽压：-logP_L⁰=-27．2814+10．6544TDEI+0．0023Sij R=0．9992，s=0．0499，F=2281．87，n=10，R_CV=0．9970，S_CV=0．0588．PCDFs的正辛醇／水分配系数：logKow=-15．7795-9．8916lnSR+10．1355TDEI+1．4100 HOMO R=0．9518，s=0．1534，F=150．70，n=51PBDEs在DB-XLB固定相上的QSRR：RRT=10．0647+0．0401（no．o-Br）+0．0439（no．m-Br）+0．0281（no．p-Br） +4．9659TDEI+0．0010 Sij-3．1954 lnS_R R=0．9963，s=0．0195，F=2639．89，n=126，R_CV=0．9911，S_CV=0．0207PCDDs在OV-1701固定相上的QSRR：RRT=155．0172+3．5307TDEI+0．0121Sij-23．7134lnS_R R=0．9978，s=0．0108，F=838．99，n=15，R_CV=0．9874,S_CV=0．0165．PCDFs在DB-5固定相上的QSRR：RRT=-2．9884+0．0026S_R-0．0384HOMO R=0．9944，S=0．03812，F=968．56，n=25，R_CV=0．9906，S_CV=0．0465其中R、s、F、n分别为相关系数、回归方程的标准偏差、Fischer值和样本数，R_CV和S_CV为采用去一法LOO（Leave-One-Out）对方程的稳定性和预测能力进行检验所得的交叉验证参数（Cross-Validation）。研究表明，从分子图最基本的结构要素，对分子的顶点进行着色，可以获得计算简单、能表达分子结构特征的拓扑参数，用这些参数来进行多环芳烃的QSPR／QSRR研究，与文献报道的方法的结果相近或更好。更多还原

【Abstract】 The halogenated polycyclic aromatic hydrocarbons distribute widely in the environment, and have huge harm to the environment and biogeocenose. Physicochemical properties of an organic chemical compound play an important role in determining its distribution and fate in the environment. Due to not only the time consumption and high expense, but also the unavailability of chemical standards of many compounds, it is nearly impossible to determine experimentally the physicochemical properties for all those compounds. Therefore, alternative approaches are needed. Some previous studies showed that it was indeed feasible to predict the properties with quantitative structure - property relationship models for many organic compounds. QSPR will give a useful method to study its distribution and fate in the environment, at the same time, QSPR afford the date of Physicochemical properties of an organic chemical compound.The Polychlorinated Dipheny （PCBs）、Polychlorinated Diphenyl Ethers （PCDEs）、Polybrominated Diphenyl Ethers （PBDEs）、Polychlorinated Dibenzodioxins （PCDDs） and Polychlorinated Dibenzofurans （PCDFs） were used to develop the QSPR/QSRR models in this thesis. This thesis simply introduced the concept of persistent organic pollutants （POPs）, took the linear regression and topological method to research the Quantitative Structure - Property/chromatographic Retention Relationships （QSPR/QSRR） for Pops. 1): From the molecular graph, we took the IP construct a adjacency matrix, solve the matrix, obtain the latent roots, the four of which were assigned the LOMO、HOMO、LUMO and HUMO. 2): From the molecular graph, we constructed a distance matrix and took the group electronegativity to construct a row matrix, combined two matrix, calculated, treat into and assigned topological distance electronegativity index （TDEI）. The QSPR models were successfully developed on sub-cooled liquid vapor pressures （P_L⁰）, n-octanol/water partition coefficients （Kow） and sub-cooled liquid water solubility (S_W,L) of PCBs、PCDEs、PBDEs、PCDDs and PCDFs; the models of quantitative relationships between structures, environmental temperatures and properties （QRSETPs） of PCBs on octanol-air partition coefficient （Koa） were also successfully developed; and the QSRR models were successfully developed on different gas chromatographic stationary of varying polarity of PBDEs、PCDDs and PCDFs. The results are very well. Some results are expressed as:The Koa of PCBs: logKoa= -15.7714+10.7417TDEI - 0.3115HOMOR=0.9913, s=0.1117, F=196.89, n=10, R_CV=0.9685, S_CV=0.1285.The P_L⁰ of PCDE: -lgP_L⁰=-39.5142 + 16.0350TDEIR=0.9918, s=0.1486, F=6356.32,n=107, R_CV=0.9830, S_CV=0.1501The logKoa of PBDEs: logKoa = - 65.8659 + 29.5682 TDEIR=0.9974, s=0.0982, F=3885.27, n=22, R_CV=0.9937, S_CV=0.1039The P_L⁰ of PCDDs: -logP_L⁰= -27.2814+10.6544TDEI+0.0023SijR=0.9992, s=0.0499, F=2281.87, n=10, R_CV=0.9970, S_CV=0.0588.The Kow of PCDFs:logKow = -15.7795-9.89161nSR+10.1355TDEI+1.4100 HOMOR=0.9518, s=0.1534, F=150.70, n=51The QSRR of PBDEs in DB-XLB:RRT = 10.0647 + 0.0401 （no.o-Br） + 0.0439 （no.m-Br） + 0.0281 （no.p-Br）+ 4.9659TDEI + 0.0010 Sij - 3.1954 lnS_R R=0.9963, s=0.0195, F=2639.89, n=126, R_CV=0.9911, S_CV=0.0207The QSRR of PCDDs in OV-1701:RRT= 155.0172 + 3.5307TDEI + 0.0121Sij - 23.7134lnS_R R=0.9978, s=0.0108, F=838.99, n=15, R_CV=0.9874,S_CV=0.0165.The QSRR of PCDFs in DB-5:RRT= -2.9884 + 0.0026S_R-0.0384HOMO R=0.9944, S=0.03812, F=968.56, n=25, R_CV=0.9906, S_CV=0.0465R、s、F、n are correlation coefficient、standard deviation、Fischer value and the number of swatch. R_CV and S_CVare the parameters of Cross-Validation.The researches show that a topological indices was obtained to express the main information of molecular structure features from the essential structure elements of molecular graph. The calculational methods of the topological index in this paper is much simpler than that of literatures, and its QSPR/QSRR results are close to or even better than the results which are reported by literatures.更多还原