DMSO injections

[james little]

I noticed when reading the advertising for Waters new Sunfire column the following statement:

SunFireâ„¢ Prep OBDâ„¢ columns exhibit optimum column efficiencies and lifetime through extended usage (over 1000 DMSO injections).

Is DMSO more of a problem than other solvents in HPLC analyses? (due to viscosity, modification of pH, etc?)

We often inject in DMF and have noted columns lasting only a few hundred injections. When taking apart, note a void in head of column.

Anyone have any insight?

[Chris Pohl]

James,

Yes, this is a common observation. As you suggest, most people believe this problem is caused by the viscosity of DMSO. To be more exact, since the viscosity of DMSO-water mixtures is even greater than that of pure solvent, there can be significant fluctuations in viscosity associated with injection of 100% DMSO due to mixing with the mobile phase at the edges of the sample band. The passage of this relatively narrow high viscosity plug through the column as the solvent passes through the column can significantly disrupt the column bed unless it is very well packed.

[Narendra]

Chrish

DMSO also injected for column cleaning purpose. It means DMSO injection only check for how the column packed ? or it has any other significance

[Mark Tracy]

50% DMSO is the solvent of choice for samples from combinatorial chemistry because is dissolves almost everything. As others have noted, DMSO and other highly viscous solvents are rough on the packed bed of a column. Since Waters wanted to promote Sunfire for combichem, they naturally wanted to address this issue. (And raise the bar on their competitors.)

From the chemistry point of view, DMSO is not damaging to the bonded phase, nor the silica substrate. From a chromatography point of view, samples injected in DMSO tends to produce badly formed peaks, especially in the early part of the chromatogram.