SPME troubleshooting: carryover, broken fibres, broad peaks and poor precision

My Ph.D. was on solid-phase microextraction, in the lab of Janusz Pawliszyn, who invented the technique. SPME is simple to describe: a coated fibre sits in a sample or its headspace, analytes partition into the coating, and the fibre is desorbed in a GC inlet or an LC interface. The problems people run into are just as consistent.

This is a symptom-by-symptom guide. Where the advice comes from vendor documents or from practitioners on the Chromatography Forum, I say so, because those are not the same as published data.

Ghost peaks and carryover

Symptom: peaks from the previous sample, or a blank that is not blank.

The coating holds analytes, and a desorption that is too short or too cool leaves some behind. Strongly retained compounds on adsorbent coatings like Carboxen are the usual offenders.

What the evidence says:

• A 2016 study in LCGC North America found that a 2-minute high-temperature conditioning step between runs, in a second injector, eliminated carryover of two dihalomethanes on a Carboxen-PDMS fibre, and made the calibration curves “significantly more linear” (Krisch et al. 2016). One lab and two analytes, so test it on yours.
• EPA Method 8272 runs a carryover blank between every new sample, and stops if the blank exceeds a third of the lower limit of quantitation until the fibre is clean (EPA 8272, p. 17).
• Supelco’s fibre data sheet recommends a blank after conditioning, and a second desorption if extraneous peaks appear; the peaks should drop on the second run (Supelco T794123).

The fibre breaks

Symptom: fibres that snap, often at the vial or the inlet.

The fibre is fused silica, and it is only protected while it is inside the needle. Two moments break most fibres: piercing a septum with the fibre exposed, and pulling the holder out of the inlet before retracting the fibre. A practitioner on the Chromatography Forum put it plainly: pull an exposed fibre back through the septum and “you lose the fiber” (thread 12938).

In another thread, an analyst breaking about five fibres a month on vial septa fixed it by switching to softer septa and vials sold for SPME; another reply asked whether the fibre was touching the bottom of the vial (thread 7321). Check the needle depth setting for that.

Broad, tailing or split peaks

Symptom: peaks much wider than a liquid injection of the same compounds, or “Christmas tree” shapes.

On a GC, the desorbed analytes have to arrive at the column as a narrow band. Practitioners recommend a narrow, low-volume inlet liner without glass wool (thread 12938), and the analyst with Christmas-tree peaks suspected poor focusing on the column (thread 8454). EPA 8272 uses a 2 mm liner with splitless desorption at 320 °C (pp. 5–7).

The usual checks: a narrow liner, splitless mode during desorption, and an initial oven temperature low enough to trap the analytes at the head of the column.

Low recovery for some analytes

Symptom: some compounds barely extract while others are fine.

Two causes cover most cases.

The coating does not suit the analyte. Sigma-Aldrich’s selection guide matches coating to molecular weight and polarity: thick PDMS for volatiles, thin PDMS for larger semivolatiles, polyacrylate for very polar analytes, PDMS/DVB for volatile polar compounds such as alcohols and amines, Carboxen-PDMS for trace volatiles (Supelco fibre selection guide). A 2015 review lists very volatile, polar or thermally unstable analytes as the weak spot of SPME (Merkle et al. 2015, section 1).

The matrix holds the analyte. In a sample with proteins or other binding components, only the free fraction is available to the fibre, and strongly bound analytes extract slowly. I modelled this in my Ph.D. work, and I explain it in why the matrix changes how fast SPME reaches equilibrium.

Poor precision between samples

Symptom: the same standard gives different areas run to run.

Before equilibrium, the amount extracted depends on extraction time, agitation and temperature. Small differences between samples change the result. Our 2018 review shows how agitation changes the diffusion layer around the fibre, and notes that vortex-type agitation, which keeps changing direction, evens out mass transfer on both sides of the device (Reyes-Garcés et al. 2018, Fundamentals).

What helps:

• Automate. EPA 8272 calls an autosampler “much preferred”, with lower and more reproducible blanks (p. 3).
• Fix time, temperature and agitation, or extract to equilibrium where that is practical.
• Add an isotope-labelled internal standard before extraction. This is the single biggest improvement, and I wrote about it separately: SPME without an isotope-labelled internal standard is a guess.
• Watch salt. Adding salt raises extraction for many analytes, but the effect depends on the analyte and the salt concentration (Merkle et al. 2015, section 3.3), so keep it identical across samples and standards.

A damaged coating

Symptom: recovery drifts down over weeks, or the coating looks stripped.

Supelco’s data sheet is direct: never rinse a fibre in chlorinated solvents, which can dissolve the epoxy that holds the fibre; never use nonpolar solvents with non-bonded PDMS; and each coating has a pH range and a maximum temperature, for example pH 2–10 and 280 °C for 100 µm PDMS (Supelco T794123, Table 1). Sigma’s guide adds that 100 and 30 µm PDMS fibres cannot be used with hexane.

Track each fibre like a column: when it went in, how many extractions, and what its blank and a check standard look like. Conditioning temperature and time matter too, and the two main vendors give different advice on that. See conditioning your SPME fibre.

I put these symptoms, plus calibration choice, fibre coatings, conditioning limits and a fibre log to print, into a free 15-page SPME troubleshooting playbook.

Sources

• Krisch MJ et al. Methodology for removing dihalomethane carryover from solid-phase microextraction fibers. LCGC North America 2016; 34(12).
• US EPA SW-846 Method 8272, Revision 0, December 2007.
• Supelco. SPME fiber assemblies data sheet T794123 (1999). Some fibres and part numbers have changed since.
• Sigma-Aldrich. Selection guide for Supelco SPME fibers.
• Merkle S, Kleeberg KK, Fritsche J. Recent developments and applications of SPME in food and environmental analysis. Chromatography 2015; 2:293–381.
• Reyes-Garcés N, et al., including Alam MN. Advances in solid phase microextraction and perspective on future directions. Anal Chem 2018; 90:302–360.
• Chromatography Forum threads 12938, 7321 and 8454 (practitioner reports, not published data).

If your SPME method has a symptom that is not on this list, describe it in one line on LinkedIn and I will tell you where I would look first.