Substantial improvements in a multidimensional dynamicsurface tension detector (DSTD) are presented. Rapid, on-line calibration and measurement of the dynamic surfacetension for high-performance liquid chromatography separations is achieved. Dynamic surface tension is determined by measuring the differential pressure across theliquid-air interface of repeatedly growing and detachingdrops. Continuous surface tension measurement throughout the entire drop growth (50 ms to 2 s) is achieved, foreach eluting drop, providing insight into the kineticbehavior of molecular
orientation processes at the liquid-air interface. Three-dimensional data are obtained, withsurface tension first converted to surface pressure, whichis plotted as a function of elution time axis versus droptime axis. Two key
innovations will be reported. First, anovel calibration procedure is described and implemented. Differential pressure signals from three drops(mobile phase, standard in mobile phase, and analyte inmobile phase) are utilized to make the dynamic surfacetension measurement, thereby eliminating the need foroptical imaging, and viscosity and hydrostatic pressurecorrections, as required by other methods. Only pressuresignals from one mobile-phase drop and one standarddrop pressure signal are required, while the analyte droppressure signal is measured along the chromatographictime axis. Second, corrections for drop elongation are notrequired, because the drops are precisely detached by anair burst actuation method in a regime were the surfacetension forces significantly dominate gravitational forces.Drops that would fall with a volume of ~10
![](/images/entities/mgr.gif)
L due togravity are precisely and repeatedly detached earlier at avolume of 2
![](/images/entities/mgr.gif)
L. The sensitivity and unique selectivity ofthe DSTD opens up new possibilities in the analysis ofsmall molecular weight polymers of varying degrees ofsurface activity, as illustrated for the size-exclusion chromatography analyses of complex poly(ethylene glycol)(PEG) samples. Using partial least squares for dataanalysis, polydispersity of complex PEG samples is determined at a relative precision of ~1%.