GC is a high-resolution technique for separation of volatile compounds in a multicomponent sample [1], [2], [3]. The hyphenation with mass spectrometry (MS) offers analysis of compounds based on their characteristic molecular fingerprints with different mass to charge ratios (m/z). With a wide range of MS spectrum libraries i.e., NIST/NIH/EPA MS database, Wiley Registry of Mass Spectral Data, and MassBank [4], GC−MS is capable of untargeted compound identification in a sample [5], [6], [7]. In addition, comparison with retention index (I) database is often performed to increase confidence in the identification [8]. However, there is a difficulty involving presence of interference signals [9] and similar MS spectra of isomers leading to the incorrect identification [10]. To improve this, tandem mass spectrometry approach (MS/MS) can be employed, which confirms compounds according to characteristic fragmentation pathways of each selected precursor ion under different modes of operations such as product ion scan, precursor ion scan, selected reaction monitoring (SRM), or multiple reaction monitoring (MRM). GC−MS/MS has been employed for targeted analysis of various types of compounds in different samples [11,12].

The untargeted MS/MS analysis has been well recognized with liquid chromatography (LC) especially in combination with high-resolution mass spectrometry (HRMS). This generally involves data dependent acquisition (DDA) or data independent acquisition (DIA) approaches [13], [14], [15], [16]. These have been applied in proteomics and metabolomics analyses and more recently in small molecule analyses, generating comprehensive MS/MS information for each compound. The DDA method acquires full scan data and subsequently performs MS/MS analysis based on selection of the top ‘n’ ranked precursor ions (e.g. using m/z window of ∼1 Da) [17]. Alternatively, all precursor ions within wider predefined isolation windows are fragmented and the MS/MS spectra can be acquired using the DIA approach. While the DDA approach could exhibit the lower analytical reproducibility and the higher rate of analyte under sampling due to the biased selection of the highly abundant species [18], application of the DIA approach could offer several benefits enabling unbiased acquisition of comprehensive MS/MS data. Data of unknown or unexpected compounds in a sample can also be recorded.

Several DIA approaches have been reported using different LC–HRMS/MS instruments [14]. Among these, sequential window acquisition of all theoretical mass spectra (SWATH) has been proven to be superior compared with the other techniques (such as MSAll and MSX). This technique uses narrower isolation windows resulting in fewer co-fragmented ions, reducing spectral complexity, and improving the quantitative analysis accuracy. SWATH method sequentially applies a fixed isolation window of the first Q (e.g. 25 m/z wide) for precursor ion selection across the overall mass range (e.g. 400–1200 m/z). These ions are then fragmented, and the product ion information can be obtained [19], e.g. using the SCIEX TripleTOF5600 MS instrument. Development of such DIA approach with GC is still a challenge.

In this study, a novel DIA-based MS/MS analysis technique, herein abbreviated as DIMS/MS, was developed for GC instrument. The technique includes data acquisition and analysis for improved identification of terpenes and oxygenated compounds in a perfume sample. Approaches for data presentation and confirmation of compound identity according to prediction of the MS/MS spectra were also established. These approaches are expected to offer significant advancements in the field of untargeted analysis with GC.