Influence of Sample Stability on Non-target Analysis during Long Batch Acquisitions

Analyte
transformation post-sample collection can introduce uncertainty
and reduce the reproducibility and reliability of results during the
non-targeted analysis (NTA) of environmental samples. Understanding
the type and degree of information lost is critical for accurate interpretation
of the experimental results. Changes in the NTA feature profile of
a sample were assessed over the immediate-to-short term (i.e., the
first 24 and subsequent 120 h) to determine how much information is
lost or changed. A wastewater sample was analyzed within 90 min of
collection using liquid chromatography coupled to a high-resolution
mass spectrometer. The sample remained in the cooled autosampler and
was re-analyzed 47 times over 24 h and daily over the following 5
days (n = 15 data points). Detected features were
classified as “formed,” “increased,” “unchanged,”
“decreased,” or “eliminated.” During the
first 24 h, 7.4% of detected features underwent transformation, increasing
to 58% over the following 5 days. Simultaneously analyzed aliquots
kept at ambient temperature showed changes for 67% of the features.
86 features were (tentatively) identified, including one formed after
sample collection. This shows that even on a short-term scale, considerable
amounts of information can change or be lost while conducting NTA,
potentially affecting reproducibility and reliability.