Targeted multidimensional gas chromatography for the quantitative analysis of suspected allergens in fragrance products

Abstract

Two approaches are described and compared for the analysis of suspected allergens (SAs) in fragrance products, which are defined by the Scientific Committee of Cosmetics and Non-Food Products (SCCNFP). The first consists of a comprehensive two-dimensional gas chromatography (GC × GC) experiment using both a “conventional” non-polar/polar column combination and an “inverse” polar/non-polar column set. The second approach uses a targeted multidimensional gas chromatography (MDGC) system employing a Deans type pneumatic switch and a longitudinally modulated cryogenic system (LMCS). It was found that the conventional and inverse column sets complement each other well, providing identification of SAs present. Compounds well retained on the second dimension of one column set were the first to be eluted from the other. In some instances SAs co-eluting with matrix components on the second dimension for a given column set were clearly resolved on the other, although this has the disadvantage of requiring two analytical runs. Targeted MDGC with a non-polar/polar column set, successfully separated all SAs identified within a fragrance product. The instrument is set up in a similar fashion to a GC × GC system though with longer second dimension (²D) column, a cryogenic trap at the beginning of the second column, and a pneumatic switch coupling both columns. The data are easier to process than for a GC × GC experiment. The targeted MDGC method has the capacity to deliver far greater efficiency to targeted regions of a primary separation than a GC × GC experiment, whilst still maintaining overall run times similar to those of a conventional one-dimensional (1D) GC experiment. Cryogenic focussing at the beginning of the ²D column delivers enhanced sensitivity, accurate ²D retention times and narrow peak widths; these are responsible for an increased resolution obtained from the fast, relatively short (∼5 m) ²D column. The two column set GC × GC analysis provided a quick and effective means to qualitatively determine the presence of six SAs in a commercially available air freshener, however all were not adequately resolved from matrix components. In contrast, quantitation was straightforward using the targeted MDGC method.

Introduction

A list of 26 raw fragrance materials have been identified by the Scientific Committee on Cosmetic Products and Non-Food Products (SCCNFP) as likely to cause contact allergies when applied to the skin. Following the Seventh Amendment of the European Cosmetic Directive 76/768/EEC in 2003, cosmetic products which contain any of these 26 raw materials, above the prescribed levels of 10 or 100 ppm if intended to remain on the skin or rinsed off respectively, must be declared on the product's packaging. Twenty-four of the 26 suspected allergens (SAs) are amenable to analysis by gas chromatography, however the quantitative determination of these compounds presents a significant challenge owing to the chemical complexity of fragrance products and the low threshold levels set by the European Cosmetic Directive. These difficulties have provided impetus for the development of improved methods for the quantitative analysis of SAs in raw fragrance materials and products containing fragrance ingredients.

Due to their inherent complexity, most fragrance mixtures cannot be quantitatively analysed by simply using just one analytical dimension. Thus, GC/MS has been a common approach, for the analysis of allergenic fragrance ingredients [1], [2], [3], [4], [5], [6]. In GC/MS, the compounds separated on the first dimension are further analysed by the mass spectrometer which acts as a second dimension. Rastogi used GC/MS to identify 11 of the defined SAs from commercial cosmetic products followed by flame ionisation detection (FID) for quantitation [5]. Neglecting the increase in analysis time, this approach is clearly hindered by inaccuracies generated when co-eluting peaks interfere with the FID quantitation. To avoid this Ellendt et al. operated the mass spectrometer in SIM mode, monitoring two individual ions for each compound [6]. However, the quantitation procedure followed in this study is thought to be unacceptable on the basis of analysis time, choice of internal standard and breakdown of standard solutions [7]. A thorough study utilising selected ion GC/MS was conducted by Chaintreau et al. who successfully managed to quantitatively analyse all 24 of the volatile SA, but still required the mass spectrometer to be operated in scan mode to verify the occurrence of compounds [4]. In a recently reported GC/MS approach the chromatograph was fitted with two injectors that each fed different analytical columns [8]. The effluent from the two columns was combined immediately prior to the MS interface, and the chromatograms from each column were collected sequentially. Two analyses were performed for each sample, with the GC/MS operated in full-scan mode, leading to two numerical results for each allergen which helped to minimise false-negative and false-positive results. The major disadvantage of this and similar approaches is the doubling of analysis time. Surprisingly, this recently reported method appears to be based on commonly used approaches in GC (dual parallel column operation), and prompts one to question why previous researchers either overlooked such a straightforward approach, or found such an approach not to be useful for SA analysis. The use of fast GC/MS analysis increases the throughput of SA analyses [9] but this does not address the problems caused by related overlapping compounds commonly found in fragrance separations. Although a great deal of attention has been made in prior studies to minimise false negatives and false positives, the analysis of allergenic fragrance ingredients is still problematic using linear separations. An alternative hyphenated technique is two-dimensional gas chromatography (GC–GC), where the compounds separated on the first dimension are subsequently analysed on an additional chromatographic dimension. Coupling GC–GC to a third MS dimension provides additional ‘mass-based separation’ of the target analytes, delivering improved quantitative results.

Comprehensive two-dimensional gas chromatography is a two-dimensional separation approach in which the whole sample is subjected to analysis in the two dimensions (columns). GC × GC delivers superior separation and increased sensitivity to the whole sample within the same time as a conventional single column analysis. These advantages make GC × GC highly suited to the analysis of fragrances and fragrance ingredients such as essential oils [10]. A GC × GC chromatogram is typically displayed as a three-dimensional surface or a two-dimensional contour plot. For a complex sample the contour or surface plot can be viewed as a unique qualitative fingerprint for each individual sample. Suitably fast scan rates for GC × GC–MS analysis can be achieved by the use of a time-of-flight mass spectrometer [11] or by using a rapid scanning quadrupole mass spectrometer with a reduced mass scan window [12]. With inexpensive faster quadrupole mass spectrometry (qMS) systems becoming readily available, the possibility to acquire data at 33 Hz in scan mode has enabled satisfactory conditions for SA analysis [13]. Debonnville and Chaintreau described a GC × GC–qMS method with a primary non-polar column for quantitation of SAs in fragrances [14].

A GC × GC system consists of an orthogonal column set coupled together through a modulator interface. In order to obtain maximum peak capacity in a GC × GC experiment, orthogonality between the two separation steps must be maximised. In some cases however, maximum orthogonality results in long retention times on the short second column. Much of the GC × GC literature to date employs a non-polar primary column coupled to a polar secondary column (conventional column set) though, recently more interest in the usage of a polar/non-polar column set (inverse column set) has arisen. Adahchour et al. found that for the analysis of diesel oil the use of both a conventional and an inverse column set gave complementary results aiding in the identification of target compounds and unknowns [15]. This approach was also adopted by Ryan et al. for the analysis of coffee [16]. Although a very challenging/varied matrix, the analysis benefited greatly from the extra selectivity provided by the combination of the two column sets.

In this paper GC × GC analysis is shown to also complement a novel heart-cutting two-dimensional GC system. Heart-cut multidimensional gas chromatography (MDGC) is a powerful approach to improve the separation of selected regions of a ¹D separation. Heart-cutting is the process of transferring selected portions of the primary separation to a secondary column for further analysis. Unlike GC × GC, a classical heart-cutting approach only applies further separation to selected regions of the primary separation, rather than the whole sample. MDGC can provide greater peak capacity for the selected regions as it is not restricted to the short second column used in GC × GC methods. The use of a longer ²D column also alleviates some detection issues because it will result in wider peaks, which are more compatible with full-scan qMS detection. Unfortunately in the past MDGC suffered from many technical issues such as surface activity, dead volume, peak broadening, instrument complexity and the possible affect upon first dimension separation subsequent to a heart-cut event [17], [18], [19], [20]. Recently, this group published results of a MDGC system utilising state of the art technology to overcome the many hurdles mentioned above. Employing mechanical switching and cryofocusing for rapid release of heart-cuts the system proved to be a powerful yet versatile technique [21], [22]. In addition the effectiveness of an alternative approach to heart-cutting termed “selective zone compression pulsing” (SZCP) has been investigated [23]. SZCP works by simply coupling two columns together with a cryotrapping modulator positioned between them (identical to a GC × GC system). The modulator operates throughout the whole run and is moved between the trap and release positions at specific times. The major benefit of SZCP is the simplicity of the system. The instrument is operated no differently to a single column experiment, there is no obstruction to the primary column flow and the system can still perform in a GC × GC fashion if needed. This paper investigates the suitability of both a conventional column set arrangement and an inverse polarity column set GC × GC analysis, contrasted with the new MDGC approach which employs a fast second dimension column for the determination and quantitation of SAs in a commercial air freshener.