Volatile Organic Compound Profiling of Commercial Kombucha Using GC-TOF-MS and GC×GC-TOF-MS

Kombucha is a fermented beverage rapidly growing in popularity. Its chemical composition is shaped by complex microbial metabolism and multiple fermentation pathways, yet its aroma profile remains poorly characterized. Concerns over alcohol content and evolving regulatory oversight have intensified interest in accurate analytical characterization of commercial products.

A joint study between William & Mary (Williamsburg, Virginia) and James Madison University (Harrisonburg, Virginia) analyzed volatile organic compounds (VOCs) from commercially produced and locally sourced kombucha products using gas chromatography–time-of-flight mass spectrometry and flame ionization detection (GC-TOF-MS/FID) and comprehensive two-dimensional gas chromatography coupled with TOF-MS and FID (GC×GC-TOF-MS/FID). The enhanced separation capacity of GC×GC enabled improved resolution and identification of VOCs in these complex matrices compared to conventional GC–MS approaches. VOC data were used to differentiate kombucha products, distinguish tea-derived volatiles from microbial fermentation products, and evaluate analytical performance between techniques. These findings advance chemical understanding of industrial kombucha and support future consumer awareness and regulatory monitoring efforts.

LCGC International spoke with Sarah Foster, lead author of the paper resulting from this study, about the team’s findings and the key takeaways from this study. ⁽¹⁾

What analytical challenges do complex fermented beverages like kombucha present for traditional one-dimensional gas chromatography-mass spectrometry (GC–MS) analysis?

Fermented beverages, such as kombucha, contain many VOCs from a diverse range of compound classes. Because a lot of the compounds in kombucha have similar properties, a conventional GC separation would contain several co-eluting regions. Those co-elutions would complicate compound identification since mass spectra would not be fully resolved.

How does comprehensive two-dimensional gas chromatography (GC×GC) improve VOC separation compared to conventional GC, particularly for co-eluting compounds?

Co-elution results from two or more analytes sharing similar properties and therefore having similar interactions with the stationary phase. GC×GC improves separation by employing two columns with independent retention mechanisms provided by orthogonal stationary phases. A modulator acts as a secondary injector, injecting small plugs of primary effluent into the secondary column. The small plugs of effluent would contain compounds with similar properties as they separated in the primary column. Injecting that plug into a column with a new retention mechanism provides further separation of the analytes via a different chemical property.

Can you explain the role of differing stationary phase chemistries in GC×GC and how they enhance compound class separation in kombucha samples?

GC×GC uses two columns with orthogonal stationary phases to obtain a greater degree of separation between similar analytes. Stationary phases range from very polar to very nonpolar. One common column set up is a non-polar column in the primary dimension and a polar column in the second dimension so that compounds first separate by boiling point in the first dimension, and compounds with similar boiling points can separate by polarity in the second dimension. As compounds separate in the primary column, the effluent is trapped in the modulator, which then injects small plugs of the primary effluent into the secondary column. The modulation system ensures that compounds with similar boiling points that eluted close together in the primary column can be separated by a different mechanism in the secondary column. Kombucha contains many distinct volatile compound classes with overlapping boiling points. The introduction of a polar separation in the second dimension better resolves those compound classes with similar boiling points.

Why is time-of-flight mass spectrometry (TOF-MS) particularly well suited for GC×GC applications compared to other MS detectors?

TOF-MS is often paired with GC×GC because of its high acquisition rate, which is necessary to record the extremely narrow peaks that rapidly elute from the secondary column. TOF-MS also has high resolution and a wide mass range, making it ideal to observe a wide range of compound classes.

What advantages and limitations does flame ionization detection (FID) offer when used alongside TOF-MS in GC×GC volatile organic compound (VOC) profiling?

GC×GC with reverse fill/flush modulation, as used in this study, requires a high flow in the second dimension, exceeding the flow typically capable of being handled by a TOF-MS instrument. Having an FID allows the flow to be split between the TOF-MS and the FID. FIDs have a better range of linearity than TOF-MS instruments, allowing for easier quantification. The higher volume of flow to the FID allows for the detection of lower-level analytes. However, FIDs do not allow for any structural determination as provided by the TOF-MS instrument for identification.

How does modulation (for example, reverse fill/flush modulation) influence sensitivity, resolution, and detector compatibility in GC×GC analyses?

Reverse fill/flush modulation improves sensitivity by focusing effluent from the primary column into narrow bands in the second dimension which results in improved signal to noise. It also enhances resolution by injecting compressed effluent into the shorter secondary column, making separation fast and efficient. Reverse fill/flush modulation requires high flow rates in the secondary dimension, often requiring the use of a splitter to split the flow between two detectors. Commonly, the flow will be split between an FID and a TOF-MS instrument for quantification and identification.

In your study, GC×GC-TOF-MS detected more analytes and compound classes than GC-TOFMS. What factors contribute to this increased chemical coverage?

The increased chemical coverage observed with GC×GC-TOF-MS compared to conventional GC-TOF-MS is primarily because of the much higher peak capacity provided by comprehensive two-dimensional separation. The orthogonal nature of the two stationary phases allows for compounds with similar properties in the primary column to be further resolved in the secondary column, reducing chemical co-elutions. The enhanced resolution of GC×GC provides more distinct mass spectra, allowing for better identification of both low-level analytes and closely eluting compounds.

How can multivariate statistical tools, such as principal component analysis (PCA), principal coordinates analysis (PCoA), and hierarchical cluster analysis(HCA) complement chromatographic and spectrometric data in differentiating kombucha products?

Multivariate statistical tools like PCA, PCoA, and HCA complement chromatographic and spectroscopic data in differentiating kombucha products by revealing patterns and relationships that are difficult to discern from raw data alone. PCA and PCoA are unsupervised models that reduce the dimensionality of the data using dissimilarity matrices (a square matrix that shows the distinction between a set of objects). The resulting principal components in PCA reveal which analytes or variables drive variance across a data set. The associated loadings plot indicates which features are the most influential, making PCA useful for identifying key compounds responsible for sample differentiation. PCoA visualizes relationships among samples, emphasizing similarities or dissimilarities in overall composition. This makes PCoA great when you are visualizing sample relationships rather than the specific analytes driving variance. Meanwhile, HCA is an unsupervised technique that groups samples based on their nearness or similarity. This means that two samples must lack or contain a compound in similar abundance to cluster closely together. The output of HCA is a dendrogram plot where shorter branches indicate more similar samples and longer branches indicate more dissimilar samples. All three of these techniques can make interpretation of chromatographic and spectroscopic data much easier by revealing patterns that would otherwise be difficult to perceive.

What strategies can be used to distinguish VOCs originating from tea substrates versus those produced through microbial metabolism during fermentation?

One strategy that can be used to distinguish VOCs originating from tea substrates versus those produced through microbial metabolism during fermentation is through data visualization. By comparing the volatiles of a purely microbial source, like a symbiotic culture of bacteria and yeast (SCOBY) sample, to the volatiles of a kombucha tea sample, it is possible to determine the analytes the two have in common. Moreover, a statistical fold change test can be useful to determine analytes specific to one group or another. The goal of a fold change test is to establish a large relative change between two samples or groups of samples, essentially establishing the key volatiles that differentiate one sample group from another. Another strategy that can be used to distinguish VOCs originating from tea substrates versus those produced through microbial metabolism during fermentation is comparing plain tea to fermented tea throughout its fermentation process. Volatiles that increase in abundance throughout fermentation can likely be attributed to microbial metabolism. However, when analyzing commercial products, data analysis strategies are much more accessible. It is also important to note that VOCs are not specific to a single source, so the same VOCs can be coming from both microbial metabolism and come from tea.

How might advanced chromatographic and spectrometric characterization of kombucha VOCs inform regulatory monitoring, quality control, or alcohol compliance testing?

Advanced chromatographic and spectrometric characterization, especially nontargeted analysis, of kombucha VOCs can inform regulatory monitoring and quality control in various ways. By identifying the VOCs present in kombucha products, it is possible to monitor the quality and consistency of commercial products. Batch differences and subtle variations in aroma profile can be tracked and adjusted for. Ethanol content is of particular interest from a regulatory standpoint. GC×GC methods can be developed to specifically track and quantify ethanol levels to determine whether the product complies with legal guidelines. Additionally, advanced VOC profiling can identify the presence of off-flavors or contaminants resulting from microbial overgrowth or unintended fermentation. Lastly, comprehensive chemical data can help establish a standard for kombucha products, allowing for easier regulation and quality control in the future.

Reference

1. Foster, S. C.; Tipton, L.; Perrault Uptmor, K. A. Evaluating Chromatographic Techniques for the Aroma Profiling of Kombucha. Anal. Bioanal. Chem. 2025.DOI: 10.1007/s00216-025-06257-5