Instrumental and theoretical advancements in pulsed elution-LC × LC: Investigation of pulse parameters and application to wastewater effluent

•Structured chromatograms improve reliability of compound identification.•Improved mass spectral quality in pulsed elution-LC × LC-HRMS.•Better understanding retention in the pulsed elution-LC × LC.•Improved pulse shape control and accuracy at low pulse volumes.•Flexible method development in pulsed elution-LC × LC. Due to the decoupling of the first (1D) and second (2D) dimension in pulsed elution-LC × LC (PE-LC × LC), method development is more flexible and straightforward compared to fast comprehensive LC × LC where the dependencies of key parameters between the two dimensions limits its flexibility. In this study we present a method for pulse generation, which is based on a switching valve alternating between one pump that delivers the gradient and a second pump that delivers low eluotrophic strength for the pause state. Consequently, the dwell volume of the system was circumvented and 7.5, and 3.75 times shorter pulse widths could be generated at flow rates of 0.2, and 0.4 mL/min with satisfactory accuracies between programmed and observed mobile phase composition (relative deviation of 6.0 %). We investigated how key parameters including pulse width and step height, 2D gradient time and flow rate affected the peak capacity in PE-LC × LC. The conditions yielding the highest peak capacity for the PE-LC × LC- high-resolution mass spectrometry (HRMS) system were applied to a wastewater effluent sample. The results were compared to a one dimensional (1D)-LC-HRMS chromatogram. The peak capacity increased with a factor 34 from 112 for the 1D-LC run to 3770 for PE-LC × LC-HRMS after correction for undersampling. The analysis time for PE-LC × LC-HRMS was 12.1 h compared to 67.5 min for the 1D-LC-HRMS run. The purity of the mass spectra improved for PE-LC × LC-HRMS by a factor 2.6 (p-value 3.3 × 10−6) and 2.0 (p-value 2.5 × 10−3) for the low and high collision energy trace compared to the 1D-LC-HRMS analysis. Furthermore, the signal-to-noise ratio (S/N) was 4.2 times higher (range: 0.06–56.7, p-value 3.8 × 10−2) compared to the 1D-LC-HRMS separation based on 42 identified compounds. The improvements in S/N were explained by the lower peak volume obtained in the PE-LC × LC-HRMS.