New Analytical Methods for Investigating the Fate of Chlorinated Paraffins and Their Transformation Products

Abstract

Chlorinated paraffins (CPs) are industrial chemicals widely applied as plasticizers, flame retardants and additives in metal working fluids. CPs are produced by nonspecific chlorination of \textit{n}-alkane mixtures. Therefore, they are a complex class of compounds of tens of thousands of different isomers. CPs can differ in carbon chain length (C10-C30) and degree of chlorination (30-70 m/m%Cl}). They are categorized according to their chain length into short-chain (SCCPs, C10-C13), medium-chain (MCCPs, C14-C17) and long-chain CPs (LCCPs, C≥18). CPs are high production volume chemicals (>1000000 t/a) of emerging environmental concern. SCCPs are persistent, bioaccumulative, subject to long-range transport and toxic. In 2017, SCCPs have been listed as persistent organic pollutants (POPs) under the United Nations Stockholm Convention. Due to their complexity, the analysis of CPs is challenging. Several issues are associated with CP analysis, including (a) chromatographic co-elution of the many isomers, (b) unknown composition of CP standard materials, and (c) broad and complex isotopic clusters often resulting in interfered mass spectra. These issues get worse in presence of CP transformation products. Up to now, CP research mainly focused on reporting SCCP levels in different matrices. Only few studies dealt with their environmental toxicity and the identification of transformation pathways and products. Even fewer studies, reported respective data for MCCPs. Almost no studies are available for LCCPs. Risk assessments and discussions about the regulation of CPs are impeded by the many analytical uncertainties and, hence, the lack of robust data. It is crucial that analytical issues are addressed and solved, and comprehensive data about the fate of CP is generated. The focus of this thesis is on the development of new analytical methods to investigate CPs and potential transformation pathways and transformation products. CPs are considered persistent in the environment and, hence, their abiotic and biotic transformation is slow. This thesis considered thermal exposure during the application of CPs as an important transformation pathway and this subject was herein investigated. In the first part of this thesis, thermal transformation products of CPs were identified by mass spectrometric (MS) analysis using thermally aged chlorinated tridecane mixtures. Direct liquid injection coupled to chloride-enhanced atmospheric pressure chemical ionization (APCI) was the method of choice. This method forces the formation of pseudo-molecular chloride-adduct ions without fragmentation and, therefore, is highly suited to investigate CPs and their transformation, when applied in full scan mode. It was found that thermal exposure of CPs causes the elimination of HCl, resulting in chlorinated olefins (COs) with one or more double bonds. Isotope clusters of CP and CO isomers of the same chain length and degree of chlorination strongly interfere even at a moderately high mass resolution of R≈10000. This so called CP/CO problem was tackled by developing a mathematical deconvolution method that resolves interfered mass spectra of CPs and COs. This method derives a linear combination of theoretical CP and CO isotope clusters that best describes the measured cluster. This information was then used to deduce non-interfered CP/CO data. The deconvolution method was later extended to resolve contributions of chlorinated diolefins as well, which is needed to describe highly aged materials, too. In the second part of this thesis, a variety of analytical methods were assessed and compared for their capability to deal with the CP/CO problem. This assessment was based on a common set of thermally exposed chlorinated tridecanes. It was found that novel methods based on liquid chromatography (LC) or direct liquid injection coupled to soft ionization MS methods are suited best to study transformation processes of CPs. The LC-based methods rely on adduct ion formation and, thus, suppress fragmentation reactions that could interfere with the detection of CP transformation products. High-resolution MS (HRMS), with mass resolution of R≈100000, was identified as the preferred method to assess CP/CO mixtures with strongly overlapping isotopic patterns. If HRMS at R≈100000 is not available, the presented deconvolution method was proven to be a reliable alternative. Conventionally, CPs are analyzed by gas chromatography (GC) coupled to electron capture negative ionization (ECNI) MS. This thesis could demonstrate that ECNI produces a variety of CP fragment ions, which results in complicated mass spectra, and induces a substantial in-source fragmentation of CPs to COs. Thus, it is not possible to distinguish COs, that are present in the sample, from COs, that are formed in the ion source, due to chromatographic co-elution of CPs and COs. Hence, GC-ECNI-MS is not suited to investigate transformation reactions of CPs that could lead to a formation of COs. Furthermore, it was demonstrated that the widely used selected ion monitoring method (SIM) can overlook severe CP/CO mass interferences. Depending on the mass accuracy, this can lead to an over- or underestimation of CP levels. Only full scan methods could reveal CP/CO mass interferences. The third part of this thesis assessed commercially available CP standard materials for their capability to quantify CPs and for their application in CP transformation and toxicity studies. Mostly complex mixtures, containing CPs of different chain lengths, are used for quantification. However, herein, it was demonstrated that CP chain length patterns of standard materials and samples can substantially differ, affecting a correct quantification. An improved quantification method based on pattern deconvolution with single-chain CP mixtures was presented. However, such standard materials are only available for SCCPs. It was highlighted that MCCPs, LCCPs and even very long-chain CPs (vLCCPs, C≥21) can dominate environmental samples. However, no single-chain CP mixtures and almost no constitutionally defined CP standard materials are available for MCCPs and LCCPs. So far, for vLCCPs, no analytical materials are available at all. It was discussed how CP materials of lower complexity, e.g., single-chain CP mixtures and constitutionally defined CPs, can be applied for studying the toxicity and transformation of CPs. With such materials, the influence of carbon chain length, degree of chlorination, chlorine substitution pattern and stereochemistry can be investigated. In the final part of this thesis, the new analytical methods and deconvolution procedures were applied. A single-chain CP mixture of chlorinated tridecanes was used to study the thermolysis of CPs at different temperatures in detail. A LC-based MS method, applying chloride-enhanced soft ionization, was applied to study non-exposed and exposed material. Non-interfered CP/CO data was derived by the mathematical deconvolution of interfered isotope clusters. It was found that thermolysis of CPs follows first-order kinetics. Determined degradation rates and half-lives suggested that CPs with a higher degree of chlorination degrade faster when exposed to heat. Degradation rates increased with increasing temperatures (160-220 °C). The same approach was also applied to study the degradation of chlorinated tridecanes during metal working applications. COs could be identified as CP transformation products formed during metal drilling, when CPs are in contact with freshly cut, hot metal surfaces. Overall, this thesis provides detailed insights into the thermal degradation of CPs to COs. Analytical issues related with the CP/CO analysis were assessed and solutions were presented. This thesis identified the lack of representative CP standard material as the major limitation when studying the fate of longer-chain CPs and their transformation products, which will be needed for a comprehensive risk assessment of MCCPs, LCCPs and vLCCPs in the future.