Industry Roundup: Rethinking Headspace SPME: Parallel Extraction for MS Workflows

For decades, headspace solid-phase microextraction (SPME) has been the workhorse sample preparation technique for gas chromatography-mass spectrometry (GC–MS) analysis of trace-level volatiles in food, beverages, and fragrances. Its one-fiber, one-sample format has also made it a bottleneck for laboratories facing growing sample loads.

Quantera Analytical, a Cornell University spin-out, aims to change that with the Curio HS, a planar cartridge-based platform that extracts dozens of samples in parallel. The platform builds on the SPMESH technique developed in Gavin Sacks's laboratory, and Quantera is preparing an onsite pilot with E. & J. Gallo Winery. Extracts can currently be analyzed by GC–MS or by chromatography-free techniques such as direct analysis in real time–mass spectrometry (DART–MS), and liquid chromatography workflows are in development.

LCGC International spoke with Terry Bates, founder and CEO of Quantera, about the engineering behind the platform and how methods transfer from fiber SPME to GC–MS on the Curio. He also discusses how novel sorbents not available as SPME coatings could expand chromatographic selectivity, when skipping chromatography makes sense and when separation remains essential, and how extracting samples in the field could reshape the way laboratories schedule their chromatography work.

SPME fibers have been the backbone of headspace GC–MS for decades. What specific limitations in fiber-based extraction convinced you that a planar, cartridge-based format was the right way forward, and what were the biggest technical hurdles in making it work?

Headspace solid phase microextraction is the foundation of the Curio. Fibers have served chromatographers well for decades, and for those running fewer samples they remain a great fit. We kept the principle and rethought the format for labs that need to parallelize.

My background is in ambient ionization. Working with DART showed me that once an instrument can analyze a sample in seconds, sample preparation becomes the bottleneck. That raised an obvious question. Why not bring parallel extraction to the chromatography that labs already run, with more extractions, less handling and simpler operation?

The groundwork was SPMESH, developed by Madeleine Bee and Jillian Jastrzembski in Gavin Sacks’s laboratory at Cornell,¹ which showed parallel headspace extraction with DART could deliver fast, trace-level results. Turning it into a product meant solving three problems:

• Removable sorbent. In SPMESH the sorbent also served as the gasket, so it couldn't go to a GC, and only elastomeric sorbents worked. A separate cartridge opened the path to GC and a much wider range of sorbents.
• Sealing. Every well must hold the same pressure with no crosstalk. That took well plate flatness and gasket design beyond what an academic laboratory could do.
• Fit. The Curio had to work with systems labs already run, including standard well-plate footprint, so a laboratory changes as little as possible.

The Curio HS lets laboratories use novel sorbents not available as SPME coatings. How does access to these materials change the selectivity and sensitivity chromatographers can achieve, and which analyte classes do you expect to benefit most?

The advantage is access to sorbents that aren’t available as SPME coatings. Because the sorbent sits in its own cartridge rather than being coated onto a fiber, the geometry makes it simpler to integrate new and emerging sorbents. The cartridge can hold materials with a wide range of properties, which gives a chromatographer more ways to tune selectivity toward the compounds they care about and to choose a material that fits the application. That matters most in complex matrices, where being selective about what gets extracted makes the analysis cleaner. Where a better matched sorbent retains a target more strongly, that can also help sensitivity, though it depends on the analyte and the material.

As for which analytes benefit most, small, polar volatiles are a clear candidate. Today’s coatings handle them with trade-offs. Nonpolar phases barely retain them, and the adsorbent phases that do can suffer from competitive displacement and a narrow linear range. A cartridge can hold more sorbent, and sorbents chosen for those compounds, which gives us room to improve recovery and quantitation there.

It also means the sorbent can be matched to the job and the budget. A laboratory that needs an inexpensive, broad screen in the field and a laboratory that needs a highly specific extraction can use the same platform with different cartridges. Our work on specific sorbents and analyte classes is underway, and we’ll share results once they’re published.

You report that established headspace SPME methods transfer with minimal re-optimization. For a laboratory running a validated GC–MS method today, what does that transition involve, and how closely do chromatographic results such as peak profiles, recoveries, and detection limits match between the two approaches?

For a laboratory running a validated headspace SPME GC–MS method, most of the method carries over. Sample preparation and matrix modification, such as salting, stay essentially the same, though our wells are smaller than a 20 mL vial, so sample and headspace volumes are adjusted. Because the sorbent can differ from the lab's fiber coating, extraction time and temperature need to be re-optimized, which is quick, since a full plate runs in parallel. The mass spectrometry method carries over, apart from the solvent delay. Today, the GC route is solvent elution and liquid injection, and we’re optimizing injection modes for sensitivity. From there, a laboratory can build a full calibration on a single plate and generate its revalidation data in a few plates. Our team will work with select early-access partners on method transfer.

In our work so far, chromatography for most analytes with liquid injection has been comparable to fiber desorption. The main differences are with the most volatile compounds near the solvent front. Because only part of the eluate reaches the column, detection limits run somewhat higher on average than a fiber's, as you'd expect from liquid injection. For the compound classes we work on, they’re still well below the levels that matter. Once a method is below the concentration where a decision changes, more sensitivity doesn’t change the answer. Side-by-side recovery comparisons with fiber SPME are part of our validation work now underway. We're also optimizing a solventless introduction mode, which aims to close both gaps with fiber SPME.

The largest gains, up to 60 times the throughput, come from pairing the Curio HS with chromatography-free techniques such as DART–MS. Without a chromatographic separation, how do you handle isomers, co-eluting interferences, and matrix effects, and where do you see GC-MS remaining essential?

Without chromatography, the question is always what level of selectivity the application actually needs.

For isomers, mass spectrometry is sometimes enough. When isomers fragment differently, a well-tuned tandem mass spectrometry (MS/MS) method can tell them apart. When they don't, we can't, and the question becomes whether it matters. Positional isomers like the cresols can't be separated without chromatography, but if a total cresol value answers the question, the method is fit for purpose. The scientist decides whether giving up that dimension of selectivity is worth the gain in speed and simplicity from a chromatography-free workflow. Terpenes are the opposite case. So many isomers exist, and the individual compounds matter so much, that we need chromatography to say how much of each is present. Enantiomers are another example. The two forms of linalool, found in grapes and wine, have different aromas, and telling them apart takes chiral chromatography.

Interferences work the same way. 3-isobutyl-2-methoxypyrazine (IBMP) gives the green bell pepper character in Cabernet Sauvignon, Sauvignon Blanc and related cultivars, and because consumers can detect it at just a few nanograms per liter, quantitating it at trace levels guides harvest timing and vineyard management in those varieties. It’s a great candidate for DART because of its high gas-phase basicity, but we found an environmental interference that persists even through MS/MS, and it limits our detection limit. We showed it was still low enough for the application, and a higher resolution mass spectrometer resolves it. So, the questions are what you need to know, what's chemically possible, and what that answer is worth to you.

For matrix effects, stable-isotope-labeled internal standards, where a labeled standard exists, correct for suppression or enhancement sample by sample.

GC–MS remains essential whenever the individual compounds matter and mass spectrometry alone can't separate them, which includes much of food, agriculture and beverage profiling today. Where we can speed up an analysis, we should. Where we can't, chromatography is the right tool. Matching the analytical test to the data you need is the name of the game, and the Curio works with both.

Your extracts are compatible with GC–MS and chromatography-free MS, and cartridges can be loaded in the field and shipped to the laboratory. How stable are analytes on the sorbent during storage and transport, and how might this decoupling of sampling from analysis change the way food, beverage, and fragrance labs schedule their chromatography work?

On stability, keeping extracted analytes on sorbent for storage and shipment has been explored in prior Cornell work, and characterizing it on the Curio is part of our validation now underway. We'll share data once it's published.

The bigger change is in scheduling. Today, sampling and analysis are tied together. A perishable sample must reach the laboratory and be extracted promptly, so the lab's calendar follows the sample's shelf life. If extraction happens where the sample is collected, you ship a small cartridge instead of a crate of fruit or a refrigerated sample. The laboratory can batch its instrument time and run at capacity, so a seasonal peak or a truckload at the receiving dock no longer turns directly into an instrument backlog.

This builds on earlier testing at Cornell. We’re planning field trials with vineyards and food producers on both the East and West Coasts, which will give us real-world data on extracting where samples are collected and shipping cartridges to the laboratory.

It also widens who can do the first step. Loading a plate and sealing it doesn't take a trained analytical chemist, so extraction can happen in a vineyard, at a receiving dock or on a production floor, with the analysis done by experts in the laboratory.

References

1. Bee, M. Y.; Jastrzembski, J. A.; Sacks, G. L. Parallel Headspace Extraction onto Etched Sorbent Sheets Prior to Ambient-Ionization Mass Spectrometry for Automated, Trace-Level Volatile Analyses. Anal Chem. 2018, 90 (22), 13806-13813. DOI: 10.1021/acs.analchem.8b04465