Analysis of Chlorinated Paraffins in Environmental Matrices: The Ultimate Challenge for the Analytical Chemist

• Gregg T. Tomy
Chapter
Part of the The Handbook of Environmental Chemistry book series (HEC, volume 10)

Abstract

Commercial chlorinated paraffins (CPs) are derived from the free radical chlorination of n-alkane mixtures. Starting mixtures used in the synthesis fall into three categories: C10–C13 (short); C14–C17 (medium) and C20–C30 (long). This results in complex mixtures containing significant numbers of constitutional and optical isomers. It is this complexity that makes analysis of CPs extremely challenging. Modern analytical methods employ either single or multi-dimensional gas chromatography coupled to mass spectrometric detectors operated in the negative ion mode. This chapter discusses the advances that have been made in the analysis of CPs in environmental samples with a focus on modern analytical techniques.

Keywords

Chlorinated paraffins Short chain chlorinated paraffins (SCCPs) Medium chain chlorinated paraffins (MCCPs) Analytical methods Mass spectrometry

1 Introduction

Chlorinated paraffins (CPs) were first produced in the 1930s for medicinal purposes. Scheer (1944) reported that a commercial antiseptic solution in the form of a chlorocosane, i.e., a solution containing 20 carbon atoms, was used during World War I [1]. Later, during World War II, CPs were used as flame retardants and were applied to tent canvases and other textile materials [2].

Today, commerical CP formulations are synthesized by the crude chlorination of n-alkane feedstocks with molecular chlorine under forcing conditions, e.g., high temperatures and/or UV irradiation. The extent and conditions of the chlorination employed ultimately depend on the desired application [1, 3]. Because the n-alkane feedstocks are derived from petroleum fractions, the end product is a mixture of carbon chain lengths and chlorination. Commercial CP mixtures fall into three categories: C10–C13 (short), C14–C17 (medium), and C20–C30 (long). The mixtures are further subcategorized on the basis of their weight content of chlorine: 40–50%, 50–60%, and 60–70%.

Owing to the varying carbon chain length and chlorine percentages of technical mixtures, CPs provide a range of properties for different applications. In general, CPs are used where the demand for chemical stability is high [4]; common applications include high temperature lubricants, plasticizers, and flame retardants, and as additives in adhesives, paints, rubber, and sealants [5, 6].

This chapter will attempt to cover the analytical methods used for measuring CPs in environmental samples. A recent survey of the literature has revealed that four comprehensive reviews on analytical methodologies for CP analysis were written and published in the last 3 years [7, 8, 9, 10]. Some overlap between this work and the contents of the review papers is therefore unavoidable.

1.1 Complexity of Industrial Mixtures

Free radical halogenation of alkanes under forcing conditions is typically reactive, and the substitution of a hydrogen atom by a chlorine atom is generally not site-specific. Furthermore, the n-alkanes used as starting materials in the industrial process generally consist of a mixture of homologues. The resulting synthetic mixture contains a range of carbon chain length and varying degrees of chlorination. It is this inherent complexity that makes the analysis of CPs in environmental samples extremely challenging.

To illustrate the complexity numerically, Tomy et al. (1997) and Shojania (1999) derived mathematical equations to calculate the theoretical number of possible constitutional isomers of CPs [11, 12]. For a CP of the general formula, CnH2n+2−zClz, and assuming no more than one chlorine atom on any carbon, the number of constitutional isomers is given by:
$$N = \raise.5ex\hbox{\scriptstyle 1}\kern-.1em/ \kern-.15em\lower.25ex\hbox{\scriptstyle 2} [\{ n!/z!\left( {n - z} \right)!\} + s]$$
(1)
where s = the number of symmetrical isomers. Four different cases arise:
1. 1.

n even, z even: s = {½ n}!/{½ z}!{½ n−½ z}!;

2. 2.

n even, z odd: s = 0;

3. 3.

n odd, z even: s = {½ (n−1)}!/{½ (n−1)−½ z}!;

4. 4.

n odd, z odd: s = {½ (n−1)}!/{½ (z−1)}! {½ (n−1)–½ (z−1)}!

One caveat of the equations is that there can be no more than one chlorine atom bound to any carbon atom. This was imposed because, although free-radical chlorination has low positional selectivity, a second chlorine atom does not readily substitute for a hydrogen at a carbon already bound to chlorine [13, 14, 15]. Table 1 shows the number of constitutional isomers for a number of chlorinated alkanes.
Table 1

The number of constitutional isomers calculated for CnH2n+2−zClz by assuming no more than one bound Cl atom on any C atom

n→

z↓

10

11

12

13

1

5

6

6

7

2

25

30

36

42

3

60

85

110

146

4

110

170

255

365

5

126

236

396

651

6

110

236

472

868

7

60

170

396

868

8

25

85

255

651

9

5

30

110

365

10

1

6

36

146

11

1

6

42

12

1

7

13

1

For a technical short chain chlorinated paraffin (SCCP) mixture containing 60% chlorine by weight, the theoretical number of congeners (defined as constitutional isomers and homologues) is 4,200 [11, 16]. It should be noted that the complexity would actually be an order of magnitude greater than that indicated in Table 1 because chlorine substitution at a secondary carbon atom usually produces a chiral carbon atom so that enantiomers and diastereoisomers would be generated.

The complexity can be further illustrated by the appearance of a chromatogram of technical formulations derived using a capillary gas chromatograph column. Figure 1 shows the total ion chromatogram of technical SCCP (top panel) and medium chain CP (MCCP, bottom panel) mixtures containing 60 and 53% of chlorine by weight, respectively, obtained using high resolution gas chromatography (GC).

Based on the large number of congeners present in CP mixtures and in environmental samples, it is not too surprising that the analysis of CPs can be quite challenging. In fact, Coquery et al. (2005) described the short-chain chlorinated paraffins (SCCPs) as the most challenging group of substances to analyze and quantify [17].

To put the number of congeners of CPs into perspective, the numbers of congeners of other environmentally relevant contaminants are shown in Table 2.
Table 2

Number of congeners for a suite of environmentally relevant contaminants

Name

# Congeners

Polychlorinated biphenyls

209

Polychlorinated naphthalenes

75

Polychlorinated bornanes

32,768a

Polychlorinated dibenzo-p-dioxins

75

Polychlorinated dibenzofurans

135

Polybrominated diphenyl ethers

209

aIn reality, the actual # of compounds found in the technical mixture is closer to 1,000 (excluding enantiomers) [71]

2 Sample Extraction and Clean-Up

In general, the extraction and clean-up of CPs in environmental matrices is similar to those of other lipophilic organo-chlorine and -bromine compounds. Furthermore, precautions taken in trace laboratory analysis of other organohalogented compounds like cleanliness of glassware, use of high purity solvents, heat treating adsorbent materials like silica gel and Florisil, and eliminating the use of plastic, also apply to CPs.

Apolar solvents like dichloromethane (DCM), hexane (Hex), and mixtures of DCM:Hex and of ethyl acetate/cyclohexane, Hex/acetone have all been used successful to extract CPs from biota and sediment samples [11, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27].

Soxhlet extraction remains the most popular means of extracting CPs from solid samples [11, 18, 19, 23, 25, 28]. This classical technique is robust, easy to use, and inexpensive and has been applied successfully to the analysis of other organohalogenated compounds. Drawbacks to the technique are lengthy extraction times (typically greater than 6 h) and the use of large volumes of solvents. Newer techniques like pressurized fluid extraction (PLE, or accelerated solvent extraction (ASE)) and microwave assisted extraction (MAE) have been shown to mitigate these two factors.

Parera et al. (2004) were the first to compare the extraction efficiencies of SCCPs from sediment using MAE and Soxhlet and showed extraction recoveries to be comparable [29]. Using ASE, Tomy et al. (1999) found recoveries of MCCPs in the range of 79–108% in spiked sodium sulfate (a surrogate for solid samples) [16].. Marvin et al. (2003) also used ASE for extraction of SCCPs in sediment; surrogate recoveries of greater than 75% were reported [20]. Nilsson et al. (2001) also applied ASE to the extraction of household waste and found greater than 90% recoveries of MCCPs [30].

Three methods have been used for extracting or pre-concentrating CPs in water (a) solid-phase extraction (SPE) [19, 23, 31, 32], (b) solid-phase microextraction (SPME) [32, 33] and (c) liquid–liquid extraction [34].

Perhaps the biggest challenge in the analysis of CPs is removal of co-extracted compounds many of which can interfere with the quantification of CPs themselves. Gel permeation chromatography [11, 16, 23, 30, 31, 35, 36, 37, 38], concentrated sulphuric acid treatment [25] or sulphuric acid-silica gel column chromatography [39] have all been used to remove co-extracted lipids from extracts. Adsorption chromatography on Florisil, silica, and alumina have all been used to further separate CPs from interfering compounds [11, 16, 29, 34, 35, 36, 39, 40]. It should be noted that Reiger and Ballschmiter (1995) have cautioned against the use of alumina as they have showed that dehydrochlorination of CPs on alumina during the adsorption process can lead to partial or total destruction of CPs [41]. Using a high energy mercury lamp, Fridén et al. (2005) were able to eliminate halogenated aromatic interfering compounds like DDT, HCB, and higher chlorinated PCB congeners in their extract [38].

3 Instrumental Analysis

3.1 Chromatographic Separation

The complexity of CP mixtures precludes the complete separation of individual congeners using either high performance liquid chromatography (HPLC) or GC. Tomy attempted to resolve a technical SCCP mixture by reverse-phase HPLC using a C18-column [42]. Little separation was obtained using this method likely because of the complexity of technical mixtures and the small number of theoretical plates of HPLC columns.

Even with the improved number of theoretical plates relative to HPLC columns, injection of a technical material onto a single non-polar capillary GC column results in a broad hump, eluting over several minutes (see Fig. 1). Throughout the elution period there can be individual broad peaks. It is thought that the congeners present at greater concentrations in the mixture give rise to the individual broad peaks, while the underlying broad hill result from congeners present at smaller concentrations[11, 43, 44].

Various non-polar stationary phases have been used in an attempt to improve on the separation: 5% phenyl- and 100% -methylpolysiloxane have been reported, and even slightly more polar phases like 35% phenyl-methylpolysiloxane have been used [11, 19, 20, 21, 30, 45, 46]. Looser and Ballschmiter (1999) were unable to attain any appreciable improvement in resolution on a highly thermally stable stationary phase Optima δ-3 capillary column [47].

While most work has been done on capillary columns, 30-m in length, of 0.25 μm film-thickness and 0.25 mm internal diameter, Coelhan (1999) first proposed the use of shorter columns (65 cm to 15 m) [48]. The rationale is that with very short columns, SCCPs can be eluted as a single peak, thereby reducing analysis times and improving analytical sensitivity [48]. Others have recognized the benefit of this approach and have adopted shorter columns in their analytical procedure [21, 49]. The obvious drawback to using shorter capillary columns is that there is an increased risk of co-elution of CPs with other co-extracted organohalogenated compounds. As such, more thorough sample clean-up procedures are necessary.

In a series of studies, Korytár et al. (2005a–c) applied comprehensive two-dimensional (2D) GC (GC×GC) to improve on CP separations achieved using single-capillary columns [7, 50, 51, 52]. The authors first tested their GC×GC analysis on a mixture of polychlorinated decanes (C10) with an average chlorine content of 65% by weight [52]. Figure 2 shows the separation achieved using a DB-1×007-65HT column combination. While the separation of the congeners was not complete, an ordered structure with four parallel groups of peaks corresponding to chlorine content was observed.
A follow-up study by Korytar et al. (2005a) on technical SCCP formulations, using a DB-1×007-65HT combination of stationary phases, showed that separation of CP congeners with the same carbon chain length was based on the number of chlorine atoms (see Fig. 3, [50]). Interestingly, when the mixtures were analyzed, ordered structures comprising compounds having the same number of carbon plus chlorine atoms (e.g., C10Cl8 and C11Cl7) were observed. Furthermore, the authors were able to achieve partial separation of short-, medium, and long-chain CPs in environmental samples (see Fig. 4, [50]).

In an extension of their own work, Korytár et al. (2005b) showed that by using a combination of a DB-1 and 65% phenylmethylpolysiloxane capillary columns, it was possible to separate polybrominated diphenyl ethers (PBDEs) and toxaphenes from CPs [51].

The use of 2D-GC is clearly a powerful and promising approach to CP analysis in complex matrices. However, many laboratories have been slow to embrace this type of instrumentation partly because of its specialized nature.

An intriguing approach of carbon skeleton GC has also been applied to the analysis of CPs [44, 53, 54]. In this technique, catalytic reduction of halogenated compounds takes place directly inside the injector port which is packed with a palladium catalyst. Figure 5 shows a typical packing arrangement for the GC-injector [53]. For CPs, the reduction reaction results in the formation of corresponding non-chlorinated n-alkanes (see Fig. 6). While all information on the degree of chlorination is lost during the reductive process, the procedure does allow for the determination of total amounts of the individual n-alkane homologues [44]. Routine detection of the well-resolved n-alkanes can be achieved using simple flame ionization detectors.

3.2 Detection

Owing to their high electronegativity, detection of CPs is usually achieved either by electron capture or by mass spectrometric detectors. While electron ionization (EI) has been applied to the analysis of CPs, electron capture negative ionization (ECNI) has been used much more frequently. This section will cover detection methods that have been used in CP analysis.

3.2.1 Electron Capture Detector

With its high sensitivity to electronegative compounds, low cost, and ease of use, it is not surprising that GC coupled to electron capture detectors (ECDs) have been used to detect CPs [30, 33, 38, 51, 55, 56]. However, the elution of CPs over a broad retention time range and the general lack of selectivity of ECDs to other electronegative compounds require that efficient clean-up procedures be in place to avoid chemical interferences from other co-extractives. Alternatively, by improving on the GC-separation using 2D-GC, Korytár et al. (2005b) were able to employ a micro-ECD as a detector [51]. In general, ECD has not been widely adopted as detector of choice for routine analysis of CPs.

3.2.2 Mass Spectrometric Detectors

The first report on the use of mass spectrometry (MS) for detection of CPs was by Gjøs and Gustavsen in 1982 [57]. Since that time, numerous reports have appeared on the application of MS for detection of CPs. This section will cover the different ionization methods that have been reported on in the literature.

Electron Capture Negative Ionization
The general appearance of an ECNI mass spectrum of a chlorinated alkane is shown in Fig. 7. At low ion source temperatures ca. 100–120°C, mass spectra are typically dominated by the ion cluster corresponding to the [M−Cl] ion fragment. With increasing source temperatures, ions corresponding to Cl 2 (m/z 70) and HCl−⋅ (m/z 71), dominate mass spectra. Perhaps the most important feature of the ECNI mass spectra of CPs is that there is little ion fragmentation. This means that unlike EI where there can typically be many fragment ions, the total ion current imparted to the molecule when first ionized under ECNI conditions is not distributed to many other ions. This leads to enhanced sensitivity under ECNI especially when dominant fragment ion is monitored.

Gjøs and Gustavsen (1982) were the first to report on the use of ECNI-MS on the analysis of CPs [57]. With methane as their moderating gas, the authors found that their ECNI mass spectra were dominated by ions at odd mass with the major fragment ions assigned as [M–H]−⋅ but contributions from [M−Cl] and [M−HCl−Cl] were also present. Introduction of samples into the MS was not reported.

Müller and Schmid (1984) built on the earlier work of Gjøs and Gustavsen by introducing CP mixtures into a low resolution ECNI-MS ion source via a GC fitted with a 15 m capillary column [58]. The appearance of the chromatogram is remarkably similar to what is generated today by most laboratories.

Jansson et al. (1991) developed a low resolution ECNI-MS method in the selected ion mode (SIM) [59]. CPs were first selectively removed from other potential interferences by GPC and detection was based on the response of the Cl 2 −⋅ (m/z 70) ion. As noted earlier, this ion dominates the ECNI mass spectrum of individual CP congeners at high mass temperature [60].

Metcalfe-Smith et al. (1995) employed a low resolution ECNI-MS method for the analysis of CPs in the full-scan mode [61]. The use of the full-scan method enabled the ion response from CPs to be discriminated against the response from other interfering compounds like PCBs.

The first report on the use of ECNI high resolution mass spectrometry (HRMS) was in 1997 [11]. This method was based on measuring the [M−Cl] ions of each CP congener in the SIM mode at a resolving power of 12,000. Because of the large number of ions monitored, retention time windows were used where select ions could be monitored over short time periods as opposed to over the entire elution period. Figure 8 shows an example of the elution profiles and retention times used for an SCCP with 60% Cl by weight. Under these conditions, interferences from other co-extracted organochlorine compounds like, PCBs, chlordane, toxaphene, and other organochlorine pesticides, were not observed. One of the big advantages of using this approach is the ability to measure the relative molar amounts of each formula group. This is particularly important because the molecular composition of samples can be very different to that of commercial mixtures that are used as external standards for quantitation purposes [11, 16, 18, 23, 35].

An international inter-laboratory study highlighted the importance of using external standards whose formula group profile closely resembles that of the samples [62]. This was further confirmed by Coelhan et al. (2000) who quantitated SCCPs in fish by ECNI low resolution mass spectrometry using several individual CP standards, of varying carbon chain length and chlorine content and also using a technical formulation [63]. Not surprisingly, the results varied by a factor of 10 depending on the standard used. Tomy et al. have shown that it is possible to correct for these differences [11], provided that the formula group profiles are not drastically different between sample and external standard.

Perhaps the biggest advantage to generating formula group profiles is that it enables the fate and behavior of individual formula groups of CPs to be monitored. For example, Houde et al. (2008) were able to examine the trophic transfer and biomagnification of individual formula groups of CPs in two aquatic food webs in Canada [23].

While the ECNI high resolution mass spectrometric method offers unmatched specificity because of the high degree of sophistication and their high cost, these instruments are not available in most analytical laboratories. In light of this, more recent efforts have gone into improving on the early ECNI low resolution mass spectrometric analytical methods.

Reth and Oehme (2004) investigated the mass interferences that can occur in samples under ECNI LRMS conditions [64]. Mass overlap can arise for congeners (in the [M−Cl] and [M−HCl]−⋅ ion cluster) with five carbon atoms more and two chlorine atoms less. For example, there is mass overlap between C11H 17 37 Cl35Cl6 (m/z 395.9) and C16H 29 35 Cl5 (m/z 396.1), which can lead to an overestimation in formula group and/or total CP concentrations. Despite this interference and probable others (e.g., C10H14Cl8 and C15H26Cl6) the authors conclude that careful selection of congener masses, isotope ratios, and retention times can resolve this potential problem [64].

The recognition of potential mass interference in the [M−Cl] and [M−HCl]−⋅ ion clusters of CPs themselves and the lower response of the [M−Cl] ion relative to Cl 2 −⋅ and HCl 2 cluster, led Castells et al. (2004) to propose the use of Cl2 (m/z 70) and HCl 2 (m/z 71) for monitoring purposes [45]. Using GC-ECNI-LRMS, it was concluded that there were no significant mass interferences observed from other organochlorine pesticides, toxaphene, PCBs, and polychlorinated naphthalenes [45]. To achieve this, the authors optimized the ECNI ion-source conditions so that the formation of Cl 2 −⋅ and HCl 2 ions from co-eluting compounds were negligible. While some organochlorine pesticides gave a small response under the optimized conditions, these compounds were resolved chromatographically and did not affect the quantitation of total SCCPs [45]. Information on individual formula groups is lost using this method.

Negative Ion Chemical Ionization

The underlying difference between ECNI and negative ion chemical ionization (NICI) is that under ECNI conditions there is no ion chemistry between the moderating gas and target analytes in the ion source. Instead, the moderating gas in ECNI is used to thermalize electrons in the ion source [65]. However, in NICI, the reagent gas and target analyte undergo some type of chemical reaction in the ion source.

Methane and argon are the most common moderating gases used for ECNI CP ionization methods. Under these conditions, the common ion cluster monitored is from a loss of chloride radical and HCl from the molecular ion. Zencak et al. (2003) proposed the use of a mixture of methane and DCM (80:20) as a reagent gas for NICI [46]. The addition of the CH2Cl2 into the reagent mixture induces the exclusive formation of the [M + Cl] adduct ion (see Fig. 9). The authors report that the interferences from other CPs and other chlorinated contaminants were significantly reduced and that overall sensitivity improved relative to the conventional methane moderating gas ECNI approach [46]. Taken together, this allowed LRMS to be used in the detection. In addition, the NICI method resulted in similar response factors for congeners with different degrees of chlorination and also enabled the detection of lower chlorinated CP congener, i.e., tri- and tetra-CPs. One major drawback to using the CH4/CH2Cl2 reagent gas is the formation of polymeric material which coats the ion source leading to a reduction in ionization efficiency over time [7].
Electron Ionization

Junk and Meisch (1993) were the first to report on a low resolution EI-MS method for the detection of CPs in environmental samples [62]. By directly introducing a commercial formulation in the ion source of the MS using an insertion probe, the authors were able to select the m/z 105 ion, which corresponded to the molecular fragment, C5H10Cl+, to be a characteristic ion produced when CPs are ionized under EI conditions [66]. McLafferty (1980) has suggested that this ion is in fact a six-membered cyclic species [67]. The EI fragmentation behavior of CPs has been studied and the stability of some of the proposed cyclic structures assessed using a force-field modeling program and semi-empirical quantum mechanical model [42]. The method by Junk and Meisch was then applied to the analysis of CPs in paving stones from a German metal working industrial plant.

Perhaps the biggest drawback to the use of EI-MS is the large degree of fragmentation it induces on the CP molecule. Numerous fragment ions are formed by consecutive losses of chlorine radicals, by elimination of HCl and by cleavage along the carbon backbone. EI mass spectra of CPs are further characterized by an absence of the molecular ion and by the small abundance of potentially mass-specific high mass fragment ions [42]. Taken together, this suggests that low resolution EI-MS for CP analysis is insufficiently sensitive (large number of fragment ions) and has the potential to suffer from nominal m/z interferences.

More recent developments on EI ion-trap MS have enabled improvements in selectivity and sensitivity for CP analysis [45, 49]. Castells et al. (2004) studied the EI fragmentation of a C10-CP using ion-trap MS and found results that were in general agreement to that of the earlier work of Junk and Meisch (1993) [45, 66]. In an extension of that work, Zencak et al. (2004) reported on fragment ions common to all CPs and proposed selection of specific ion transitions induced using tandem MS (triple quadrupole and ion-trap) [49]. In this approach, the relative response of specific fragment ions of CPs was found to be independent of the chlorine content and carbon chain length. Since all CPs can form these fragment ions independent of their chain length, the method cannot distinguish amongst CPs of varying chain length. While congener and homologue-specific analysis is not possible using this technique, the method is ideally suited when a rapid screening of samples is needed [49].

Metastable Atom Bombardment
Moore et al. (2004) reported on the use of metastable atom bombardment (MAB) ionization and HRMS in the positive ion mode for the detection of CPs [34]. Using an inert gas like argon, MAB ionization generates ions via an electrophilic reaction of a metastable atom with the analyte of interest. Figure 10 shows a comparison of ECNI and positive ion MAB mass spectra.

Using the MAB approach, the two most intense ions in the [M−Cl]+ ion cluster of CP congeners are monitored. A real strength of this method relative to ECNI is the detection of congeners having a small number of chlorine atoms.

Positive Ion Chemical Ionization
Castells et al. (2004) evaluated the suitability of positive ion chemical ionization (PICI) for the detection of CPs [45]. Figure 11 shows a comparison of an EI and PICI mass spectra of a typical CP. Like EI, mass spectra of CPs in PICI are characterized by the absence of the molecular ion and presence of numerous low-mass fragment ions corresponding to losses of HCl and chloride radical. As such, PICI has not been embraced as a tool for detecting CPs in environmental samples.

3.3 Quantitation

Prior to the advent of capillary GC, PCBs were quantified in environmental samples using technical formulations as external standards. Because different technical formulations of PCBs were available, the approach to quantifying PCBs was to select a formulation whose GC-profile most closely resembled the PCB profile in the sample. This approach has been adopted in the analysis of CPs. Like PCBs, because CPs are known to undergo environmental transformations, it can be challenging to get a technical formulation whose profile closely matches that of the sample.

Early analytical methods that relied on ECD detectors used a similar approach to quantify CPs as was done with PCBs. To circumvent or mitigate the contribution of interfering peak areas to that of CPs, Walter and Ballschmiter (1991) first proposed the use of a triangulated method for CP quantitation, i.e., constructing a triangle by drawing its base from the start of the CP elution to its end with its apex at the maximum signal (see Fig. 12, [68]).

Coelhan et al. (2000) quantified the extent to which errors can arise when using SCCP standards whose chlorine content differed from that of the sample [63]. Using three standards of varying Cl content, the authors reported that concentration differences of >1,000% were observed when quantifying the same sample.

Tomy et al. (1997) introduced the numerical correction factor, average molar mass (AMM), to correct for differences in the appearance of the sample and standard [11]. Using the AMM, the authors demonstrated that the error in quantifying one technical mixture using another as an external standard was less than 40% [62].

More recently, Reth et al. (2005) reported on an approach that compensates for the difference in the profiles between sample and standard [69]. The method makes use of the linearity of the ECNI response factor with chlorine content which has only been validated on a quadrupole MS.

Only two interlaboratory studies have been done on SCCPs and none on MCCPs or long-chain CPs (LCCPs) [62, 70]. To reduce as many steps or variables as possible, Pellizzato et al. (2009) supplied six laboratories with an industrial soil extract that was processed in a single laboratory [70]. Identical synthesized standard solutions, now commercially available were use by the laboratories except one laboratory which used an n-alkane external standard. Participating laboratories were asked to quantify SCCPs in the supplied extract using an analytical technique of their choice.

All laboratories used GC with a single capillary column, and of the six participating laboratories, five used MS; four in ECNI and one in EI mode. One laboratory used an atomic emission detector (AED). With the exception of a single outlier laboratory, results of the four laboratories that employed MS were in generally good agreement with a range of only 10 ± 1 to 15 ± 2 ng L−1. The one laboratory that employed an AED reported on a value that was ca. four times greater than the mean of the laboratories that used MS detection. Because of the nonselective nature of the AED, the authors felt that this result would improve with better clean-up steps. The one outlier laboratory, which used MS detection, reported on a value that was ca. 200 greater than the other laboratories. It was unclear why there was such a large discrepancy in this one laboratory. The authors concluded that even by eliminating the extraction and clean-up steps, it is still very difficult to obtain comparable SCCP results. Part of the discrepancy was thought to be due to the choice of detection and quantitative approaches used.

Results from an earlier interlaboratory study on SCCPs by Tomy et al. (1999) involving seven laboratories were also met with mixed success [62]. Similar to the Pellizzato et al. (2009) study, participants in the Tomy et al. (1999) study were supplied with an injection ready extract (biota) along with two technical solutions - one with 60% Cl by weight that was used as an external standard and another with 70% Cl by weight that was supplied at an unstated concentration and treated as a ‘sample’. A fourth solution containing a purified mixture of UV-chlorinated 1,5,9-decatriene products was also supplied to the participating laboratories at an unstated concentration. All laboratories used single capillary GC coupled with MS operating in the ECNI mode; two laboratories used HRMS run at a resolving power of 10,000–12,000. One laboratory also reported results obtained with an ECD detector.

Surprisingly, measurements made on the synthetic C10-solution whose carbon and chlorine profile differed quite markedly from that of the supplied technical mixtures were closer to the stated concentration (% error range: 2–74%) than measurements on the technical mixture that was treated as a sample (% error: 30–310%). Reasons for this discrepancy were unclear but it was thought that if there were differences in the purity of technical formulations this could make preparation of standard solutions questionable [62].

Overall the analytical precision in measurements made on the fish extract was acceptable. Because two fish extracts were supplied to participating laboratories, two separate measurements were made; the average deviation from the mean (ADM) in the first instance was 10.2 while in the latter it was 16.2. The smaller ADM value was thought to be due to laboratories correcting for interferences from other co-eluting interferences. This was not performed in the latter case.

It is clear from both interlaboratory studies that differences in measurement data on SCCPs (and likely CPs of other chain length) can be notable. The choice of quantitative procedures used by laboratories and also choice of external standard employed (this was certainly the case in the Tomy et al. (1999) study [62]) can contribute to unreliable data. The recent commercial availability of impurity-free synthetic C10–C13 solutions should eliminate the uncertainty in preparation of working standard solutions. However, other confounding variables like choice of extraction and clean-up procedures have not been quantitatively assessed and are likely to contribute to the uncertainty in CP data.

4 Conclusion

Over the last 15 years, considerable progress has been made in the analysis of CPs. Much of the analytical efforts have been invested in the analysis of SCCPs with less on MCCPs and relatively little on LCCPs. Mass spectrometric based detectors have clearly facilitated advances in this area.

Perhaps the pioneering work on multidimensional GC offers an enticing glimpse on the future of CP analysis. However, the simplicity of the carbon-skeleton method certainly makes SCCP and analysis of longer chain homologues amenable to more laboratories especially those laboratories new to this area of research.

Perhaps the most pressing issue that needs urgent attention is that of quality assurance. To date, only two inter-laboratory exercises have been conducted; the results from both studies demonstrated that quantitative measurements can be quite varied. Consensus on the choice of a working standard solution, method of quantitation, and certified reference materials will go a long way to ensure that inter-laboratory measurements are more comparable.

With the recent inclusion of SCCP as a candidate POP under international regulatory conventions, it is likely that reliable environmental measurements will be further sought. Addressing the shortcomings that currently exist in the analysis of SCCPs (and other chain length CPs) will go a long way in generating environmental measurement data that can be used by international organizations that regulate the use and release of chemicals.

References

1. 1.
Scheer WE (1944) Properties and uses of chlorinated paraffins. Chem Ind 54:203–205Google Scholar
2. 2.
Willis B, Crookes MJ, Diment J, Dobson SD (1994) Environmental hazard assessment: chlorinated paraffins. Toxic Substances Division, Department of the Environment, Garston, WatfordGoogle Scholar
3. 3.
Kirk-Othmer (1980) Chlorinated paraffins. Kirk-Othmer encyclopedia of chemical technology, 3rd edn. Wiley, New YorkGoogle Scholar
4. 4.
Svanberg O, Bengtsson BE, Lindén E, Lunde G, Ofstad EB (1978) Chlorinated paraffins – a case of accumulation and toxicity to fish. Ambio 7:64–65Google Scholar
5. 5.
Zitko V (1980) Chlorinated paraffins. Springer, New YorkGoogle Scholar
6. 6.
Mukherjee AB (1990) The use of chlorinated paraffins and their possible effects in the environment. National Board of Water and the Environment, Helsinki, FinlandGoogle Scholar
7. 7.
Zencak Z, Oehme M (2006) Recent developments in the analysis of chlorinated paraffins. Trends Analyt Chem 25:310–317
8. 8.
Santos FJ, Parera J, Galceran MT (2006) Analysis of polychlorinated n-alkanes in environmental samples. Anal Bioanal Chem 386:837–857
9. 9.
Eljarrat E, Barceló D (2006) Quantitative analysis of polychlorinated n-alkanes in environmental samples. Trends Anal Chem 25:421–434
10. 10.
Bayen S, Obbard JP, Thomas GO (2006) Chlorinated paraffins: a review of analysis and environmental occurrence. Environ Int 32:915–929
11. 11.
Tomy GT, Stern GA, Muir DCG, Fisk AT, Cymbalisty CD, Westmore JB (1997) Quantifying C10–C13 polychloroalkanes in environmental samples by high-resolution gas chromatography/electron capture negative ion high-resolution mass spectrometry. Anal Chem 69:2762–2771
12. 12.
Shojania S (1999) The enumeration of isomeric structures for polychlorinated n-alkanes. Chemosphere 38:2125–2141
13. 13.
Fredricks PS, Tedder JM (1959) Free-radical substitution in aliphatic compounds. Part II. Halogenation of the n-butyl halides. Chem Soc 144–150Google Scholar
14. 14.
Colebourne N, Stern ES (1965) The chlorination of some n-alkanes and alkyl chlorides. J Chem Soc 3599–3605Google Scholar
15. 15.
Chambers G, Ubbelohde AR (1955) The effect of molecular structure of paraffins on relative chlorination rates. J Chem Soc 285–295Google Scholar
16. 16.
Tomy GT, Stern GA (1999) Analysis of C14–C17 polychloro-n-alkanes in environmental matrices by accelerated solvent extraction-high-resolution gas chromatography/electron capture negative ion high-resolution mass spectrometry. Anal Chem 71:4860–4865
17. 17.
Coquery M, Morin A, Bécue A, Lepot B (2005) Priority substances of the European water framework directive: analytical challenges in monitoring of water quality. Trends Anal Chem 24:117–125
18. 18.
Tomy GT, Stern GA, Lockhart LW, Muir DCG (1999) Occurrence of C10–C13 polychlorinated n-alkanes in Canadian midlatitude and Arctic Lake sediments. Environ Sci Technol 33:2858–2863
19. 19.
Nicholls CR, Allchin CR, Law RJ (2001) Levels of short and medium chain length polychlorinated n-alkanes in environmental samples from selected industrial areas in England and Wales. Environ Pollut 114:415–430
20. 20.
Marvin CH, Painter S, Tomy GT, Stern GA, Braekevelt E, Muir DCG (2003) Spatial and temporal trends in short-chain chlorinated paraffins in Lake Ontario sediment. Environ Sci Technol 37:4561–4568
21. 21.
Stejnarová P, Coelhan M, Kostrhounova R, Parlar H, Holoubek I (2005) Analysis of short chain chlorinated paraffins in sediment samples from the Czecg Republic by short-column GC/ECNI-MS. Chemosphere 58:253–262
22. 22.
Stevens JL, Northcott GL, Stern GA, Tomy GT, Jones KC (2003) PAHs, PCBs, PCNs, organochlorine pesticides, synthetic musks, and polychlorinated n-alkanes in U.K. sewage sludge: survey results and implications. Environ Sci Technol 37:462–467
23. 23.
Houde M, Muir DCG, Tomy GT, Whittle DM, Teixeira C, Moore S (2008) Bioaccumulation and trophic magnification of short- and medium-chain chlorinated paraffins in food webs from Lake Ontario and Lake Michigan. Environ Sci Technol 42:3893–3899
24. 24.
Kemmlein S, Hermeneit A, Rotard W (2002) Carbon skeleton analysis of chloroparaffins in sediment, mussels and crabs. Organohalogen Compd 59:279–282Google Scholar
25. 25.
Borgen A, Schlabach M, Mariussen E (2003) Screening of chlorinated paraffins in Norway. Organohalogen Compd 60:331–334Google Scholar
26. 26.
Castells P, Parera J, Santos FJ, Galceran MT (2008) Occurrence of polychlorinated naphthalenes, polychlorinated biphenyls and short-chain chlorinated paraffins in marine sediments from Barcelona (Spain). Chemosphere 70:1552–1562
27. 27.
Pribylová P, Klánová J, Holoubek I (2006) Screening of short- and medium-chain chlorinated paraffins in selected riverine sediments and sludge from the Czech Republic. Environ Pollut 144:248–254
28. 28.
Hüttig J, Oehme M (2005) Presence of chlorinated paraffins in sediments from the North and Baltic Seas. Arch Environ Contam Toxicol 49:449–456
29. 29.
Parera J, Santos FJ, Galceran MT (2004) Microwave-assisted extraction versus Soxhlet extraction for the analysis of short-chain chlorinated alkanes in sediments. J Chromatogr A 1046:19–26Google Scholar
30. 30.
Nilsson M-L, Waldebäck M, Liljegren G, Kylin H, Markides KE (2001) Pressurized fluid extraction (PFE) of chlorinated paraffins from the biodegradable fraction of source separated household waste. Fresenius J Anal Chem 370:913–918
31. 31.
Dick TA, Gallagher CP, Tomy GT (2009) Short- and medium-chain chlorinated paraffins in fish, water and soils from the Iqaluit, Nunavut (Canada), area. Int J Environ Waste ManageGoogle Scholar
32. 32.
Castells P, Santos FJ, Galceran MT (2004) Solid-phase extraction versus solid-phase microextraction for the determination of chlorinated paraffins in water using gas chromatography-negative chemical ionisation mass spectrometry. J Chromatogr A 1025:157–162
33. 33.
Castells P, Santos FJ, Galceran MT (2003) Solid-phase microextraction for the analysis of short-chain chlorinated paraffins in water samples. J Chromatogr A 984:1–8
34. 34.
Moore S, Vromet L, Rondeau B (2004) Comparison of metastable atom bombardment and electron capture negative ionization for the analysis of polychloroalkanes. Chemosphere 54:453–459
35. 35.
Tomy GT, Muir DCG, Stern GA, Westmore JB (2000) Levels of C10–C13 polychloro-n-alkanes in marine mammals from the Arctic and the St. Lawrence River Estuary. Environ Sci Technol 34:1615–1619
36. 36.
Thomas GO, Farrar DG, Braekevelt E, Stern GA, Kalantzi OI, Martin FL, Jones KC (2006) Short and medium chain length chlorinated paraffins in UK human milk-fat. Environ Int 32:34–40
37. 37.
Ismail N, Gewurtz S, Pleskach K, Whittle DM, Helm PA, Marvin CH, Tomy GT (2009) Brominated and chlorinated flame retardants in Lake Ontario, Canada, Lake Trout (Salvelinus namaycush) between 1979 and 2004 and possible influences of food-web changes. Environ Toxicol Chem 28:910–920
38. 38.
Fridén U, Jansson B, Parlar H (2005) Photolytic clean-up of biological sample for gas chromatographic analysis of chlorinated paraffins. Chemosphere 54:1083Google Scholar
39. 39.
Iino F, Takasuga T, Senthilkumar K, Nakamura N, Nakanishi J (2005) Risk assessment of short-chain chlorinated paraffins in Japan based on the first market basket study and species sensitivity distributions. Environ Sci Technol 39:859–866
40. 40.
Reth M, Zencak Z, Oehme M (2005) First study of congener group patterns and concentrations of short- and medium-chain chlorinated paraffins in fish from the North and Baltic Sea. Chemosphere 58:847–854
41. 41.
Rieger R, Ballschmiter K (1995) Semivolatile organic compounds-polychlorinated dibenzo-p-dioxins (PCDD), dibenzofurans (PCDF), biphenyls (PCB), hexachlorobenzene (HCB), 4, 4′-DDE and chlorinated paraffins (CP)-as markers in sewer films. Fresenius J Anal Chem 352:715–724
42. 42.
Tomy GT (1997) The mass spectrometric characterization of polychlorinated n-alkanes and the methodology for their analysis in the environment. Ph.D Thesis, University of ManitobaGoogle Scholar
43. 43.
Tomy GT, Fisk AT, Westmore JB, Muir DCG (1998) Environmental Chemistry and Toxicology of Polychlorinated n-alkanes. Rev Environ Contam Toxicol 158:53–128
44. 44.
Pellizzato F, Ricci M, Held A, Emons H (2007) Analysis of short-chain chlorinated paraffins: a discussion paper. J Environ Monitor 9:924–930
45. 45.
Castells P, Santos FJ, Galceran MT (2004) Analysis of short-chain chlorinated paraffins: a discussion paper. Rapid Commun Mass Spectrom 18:529–536
46. 46.
Zencak Z, Reth M, Oehme M (2003) Dichloromethane-enhanced negative ion chemical ionization for the determination of polychlorinated n-alkanes. Anal Chem 75:2487–2492
47. 47.
Looser R, Ballschmiter K (1999) Gas chromatographic separation of semivolatile organohalogen compounds on the new stationary phase Optima δ-3. J Chromatogr A 836:271–284
48. 48.
Coelhan M (1999) Determination of short-chain polychlorinated paraffins in fish samples by short-column GC/ECNI-MS. Anal Chem 71:4498–4505
49. 49.
Zencak Z, Reth M, Oehme M (2004) Determination of total polychlorinated n-alkanes concentration in biota by electron ionization-MS/MS. Anal Chem 76:1957–1962
50. 50.
Korytár P, Parera J, Leonards PEG, Santos FJ, de Boer J, Brinkman UAT (2005) Characterization of polychlorinated n-alkanes using comprehensive two-dimensional gas chromatography-electron-capture negative ionisation time-of-flight mass spectrometry. J Chromatogr A 1086:71–82
51. 51.
Korytár P, Leonards PEG, de Boer J, Brinkman UAT (2005) Group separation of organohalogenated compounds by means of comprehensive two-dimensional gas chromatography. J Chromatogr A 1086:29–44
52. 52.
Korytár P, Parera J, Leonards PEG, de Boer J, Brinkman UAT (2005) Quadrupole mass spectrometer operating in the electron-capture negative ion mode as detector for comprehensive two-dimensional gas chromatography. J Chrom 1067:255–264
53. 53.
Koh I-O, Rotard W, Thiemann W (2002) Analysis of chlorinated paraffins in cutting fluids and sealing materials by carbon skeleton reaction gas chromatography. Chemosphere 47:219–227
54. 54.
Sistovaris N, Donges U (1987) Gas-chromatographic determination of total polychlorinated aromates and chloro-paraffins following catalytic reduction in the injection port. Fresenius J Anal Chem 326:751–753
55. 55.
Randegger-Vollrath A (1998) Determination of chlorinated paraffins in cutting fluids and lubricants. Fresenius J Anal Chem 360:62–68
56. 56.
Ballschmiter K (1994) Determination of short and medium chain length chlorinated paraffins in samples of water and sediment from surface water. Abt Anal Chem und UmweltchemGoogle Scholar
57. 57.
Gjøs N, Justavsen KO (1982) Determination of chlorinated paraffins by negative ion chemical ionization mass spectrometry. Anal Chem 54:1316–1318
58. 58.
Müller MD, Schmid PP (1984) GC/MS analysis of chlorinated paraffins with negative ion chemical ionization. J High Resolut Chromatogr Chromatogr Commun 7:33–37
59. 59.
Jansson B, Andersson R, Asplund LT, Bergman Å, Litzén K, Nylund K, Reutergårdh L, Sellström U, Uvemo U-B, Wahlberg C, Wideqvist U (1991) Multiresidue method for the Gas-Chromatographic Analysis of some Polychlorinated and Polybrominated Pollutants in Biological Samples. Fresenius J Anal Chem 340:439–445
60. 60.
Tomy GT, Tittlemier SA, Stern GA, Muir DCG, Westmore JB (1998) Effects of temperature and sample amount on the electron capture negative ion mass spectra of polychloro-n-alkanes. Chemosphere 37:1395–1410
61. 61.
Metcalfe-Smith JL, Maguire RJ, Batchelor SP, Bennie DT (1995) Occurrence of chlorinated paraffins in the St. Lawrence River near a manufacturing plant in Cornwall, Ontario. National Water Research Institute, Department of the Environment, Burlington, ONGoogle Scholar
62. 62.
Tomy GT, Westmore JB, Stern GA, Muir DCG, Fisk AT (1999) Interlaboratory study on quantitative methods of analysis of C10–C13 Polychloro-n-alkanes. Anal Chem 71:446–451
63. 63.
Coelhan M, Saraci M, Parlar H (2000) A comparative study of polychlorinated alkanes as standards for the determination of C10–C13 polychlorinated paraffins in fish samples. Chemosphere 40:685–689
64. 64.
Reth M, Oehme M (2004) Limitations of low resolution mass spectrometry in the electron capture negative ionization mode for the analysis of short- and medium-chain chlorinated paraffins. Anal Bioanal Chem 378:1741–1747
65. 65.
Ong VS, Hites RA (1994) Electron capture mass spectrometry of organic environmental contaminants. Mass Spectrom Rev 13:259–283
66. 66.
Junk SA, Meisch HU (1993) Determination of chlorinated paraffins by GC-MS. Fresenius J Anal Chem 347:361–364
67. 67.
McLafferty FW (1980) Interpretation of Mass Spectra. University Science Books, Mill Valley, CaliforniaGoogle Scholar
68. 68.
Walter B, Ballschmiter K (1991) Quantitation of camphechlor/toxaphene in cod-liver oil by integration of the HRGC/ECD-pattern. Fresenius J Anal Chem 340:245–249
69. 69.
Reth M, Zencak Z, Oehme M (2005) New quantification procedure for the analysis of chlorinated paraffins using electron capture negative ionization mass spectrometry. J Chromatogr A 1081:225–231
70. 70.
Pellizzato F, Ricci M, Held A, Emons H, Böhmer W, Geiss S, Iozza S, Mais S, Petersen M, Lepom P (2009) Laboratory intercomparison study on the analysis of short-chain chlorinated paraffins in an extract of industrial soil. Trends Anal Chem 28:1029–1035
71. 71.
Kucklick JR, Helm PA (2006) Advances in the environmental analysis of polychlorinated naphthalenes and toxaphene. Anal Bioanal Chem 386:819–836