Laserchrom HPLC Laboratories Ltd
Everything for successful HPLC!
 
     
     
Laserchrom Home Page

HPLC Products

  HPLC Courses  
  HPLC Laboratory Services  
  HPLC Technical Support  
  Troubleshooting  
  Equipment Maintenance  
  Download Centre  
  Useful HPLC Information  
  Buying Guides  
  Lunchtime Seminars  
  HPLC Applications  
  HPLC Job Centre  
  Press  

Laserchrom HPLC Laboratories Ltd

Units B16-18, Laser Quay,

Medway City Estate,

Rochester, Kent. ME2 4HU (United Kingdom)

   

Column Switching as an alternative to

Gradient Elution

 

HPLC requires that a system be at dynamic equilibrium (its all happening, but nothing is changing!) so that any observed peaks can be attributed to the injected sample. Gradient elution continuously disturbs this equilibrium (hence changing baseline, ghost peaks etc), and has the following disadvantages:

  • Solvents (especially water) must be pure to avoid ghost peaks eluting
  • A gradient pump is required (ca £7.5K)
  • Degassing is far more important than with isocratic elution, especially for low pressure mixing.
  • Time must be allowed after each run to re-attain starting conditions
  • A Refractive Index detector cannot be used.

All of these problems can be overcome, in certain circumstances, by using column switching instead. It’s a little more complicated, and less well known, but well worth considering.

Situation 1.

Consider a separation where most of the components elute quickly, but the last component takes considerably longer.

One option is to use a gradient to bring the last component off in a reasonable time, but you could also use a switching valve.

 

 

 

 

 

The valve is timed to switch after peak 4 (in this case). The last peak will have travelled only a short distance down the column, so by reversing the flow it elutes much more quickly.

Note that this method is only appropriate when there is only one late-running peak. If there were more than one, back-flushing would cause them to elute together.

 

Situation 2

Now consider the situation where there are a series of late-running peaks.

Using a gradient we would be able to elute the late-running peaks with a profile as follows:

However we could use two shorter columns and a switching valve:

This allows us the flexibility that the two columns need not be the same. So for example, Column 1 could be C8 and Column 2 C18. We start the run with the valve in position A. Once the first four peaks (in the above example) have eluted into Column 2 we switch the valve to position B. The late-running peaks are now eluted quickly from the shorter, less retentive column, and the valve is switched back to position A, eluting the first four peaks from Column 2:

Note that although the peak order is not the same, the results are remarkably similar to the gradient elution run.

 

Situation 3

Consider this chromatogram, simplified and exaggerated to make the point:

This is clearly a classic case where gradient elution would be the right answer, along the lines of the profile shown below:

But what if you were only interested in the first group of four peaks? The rest still have to be eluted from the column…?

 

Using a switching valve there is a way we can speed up the process considerably. The set-up is as follows:

         

Using Pump 1 a sample is injected by Autosampler 1 and runs through Column A1.  As soon as the first group of peaks is eluted onto Column B, the valve is switched so that all remaining peaks are eluted to waste. However Pump 2 now takes over, and another sample is injected by Autosampler 2 into Column A2. As soon as the first group of peaks is eluted into Column B the valve is switched back.. This process is then repeated, alternating injections from Autosamplers 1 and 2, switching the switching valve each time. As you can imagine the chromatogram will look as follows, with just groups of four peaks, following each other down the column:

 

Columns A1 and A2 should be much shorter and less retentive than Column 2, so that the remaining peaks from each injection are eluted to waste quickly. In any event they must be clear before the next injection can occur.

Elution on Column 2 is more of a continuous process than a series of discreet injections. However the data system should be set with a run time just long enough to elute each group of four peaks, and be triggered each time the switching valve turns.

This method of operation is MUCH quicker than running a gradient, and can analyse more samples than even when using two systems. The HPLC remains at dynamic equilibrium throughout, and the automation is easy, making it very reproducible!

 

Free Technical Advice Available Now on 01634-294001

(09.00-17.00 GMT)

 


Counter Graphic