On-tubing fluorescence measurements of the band broadening of contemporary injectors in ultra-high performance liquid chromatography

Abstract

We report on a detailed study of the injection contribution to band broadening in contemporary UHPLC-instruments, using either flow-through needle or fixed loop injection (full loop). Using on-tubing fluorescence measurements at the outlet of the injector valve, very localized and undisturbed measurements were obtained. Varying both the flow rate and the injected volume allowed to split the injection variance (σ) in a volumetric component (related to the amount injected) and a hydrodynamic component (related to the flow rate). For the flow-through needle injector and for the small injection volumes (<2 μL) typically used in UHPLC, it was found that the volumetric contribution (i.e. the part of σ, that increases with increasing injection volume) is given by a value of σ = 0.8 to 1·V rather than by the value of 0.125 to 0.2·V that is normally assumed in literature. For the hydrodynamic contribution to σ (i..e, the part which remains present even for very small injection volumes), a clear increase in dispersion with flow rate is found, reaching a plateau around 0.8ml/min of 0.6 μL² or 1.2 μL² for the 75 μm and 120 μm needle seat capillaries respectively. The difference between both shows the clear advantage of using a low dispersion 75 μm injection needle seat capillary. For a loop-type injector operated in a full-loop mode, the increase in peak variance with the injection volume is much less pronounced, leading to a total injector variance given by σ = 0.34 μL² + 0.12·V over the entire range of investigated injection volumes of 1.1 μL up to 4.5 μL when using 120 μm or narrower ID loops. This expression was nearly completely independent of the flow rate. For larger ID sample loops, a clear increase of peak variance with flow rate at fixed injection volume was observed (σ increases with 20% for a 170 μm ID loop and with 70% for a 220 μm ID loop from 0.3 to 1 ml/min).

Extra-column band broadening effects in contemporary liquid chromatography: Causes and solutions

“…In a recent experimental study, our group found that for small injection volumes (Vinj<1-2µL) a value 1/inj=1 is a more correct value for a flow through needle type injector. This value decreases to a value of 1/inj=1/8 for the largest investigated volumes (Vinj=10µL) [61]. In another study, 1/inj=0.8 was found for an injection volume of 0.5µL, although this was determined indirectly [13].…”

“…( 1). Albeit slightly cumbersome, this issue can be resolved by calculating the actual flow rate during injection from the measured pressure profile and replotting the chromatogram in volume units to calculate the volumetric peak variance [61].…”

“…For example, it is straightforward to minimize the contribution of the injection volume by injecting only very small volumes (provided the amplitude of the signal in the detector remains sufficiently high) [61].…”

“…Considering the ECBB introduced by the injection process, it is important to distinguish between the dispersion originating from the physical volume of the injected sample and that originating from the hydrodynamic dispersion experienced by the analytes when passing through the injection device before entering the connection tubing. Assuming both contributions are independent of each other (which is only true to a certain degree), the volumetric contribution of the injection process can be written as [61]:…”

Measurement of the Band Broadening of UV Detectors used in Ultra-high Performance Liquid Chromatography using an On-tubing Fluorescence Detector

“…In this contribution, the whole detector cell variance will be measured using the on-capillary LIF approach recently used by our group to measure the dispersion directly after the valve of a sample injector [19]. In this methodology, the sample band width will be measured directly before entering (of after exiting) the detector using an on-capillary LIF detector (see Fig.…”