Showing results for 
Search instead for 
Did you mean: 
Team TFS
Team TFS

shutterstock_383191675What is a collision cell?

If you use quadrupole ICP-MS, the chances are you have encountered collision cell technology.  This technology is used to remove interferences that are generated by the plasma gas and the sample constituents.  The collision cell itself is a type of ion lens located between the interface cones of the instrument and the quadrupole mass spectrometer.  When operated in collision cell mode, a flow of gas is passed into the cell.  Ions coming through the interface cones from the plasma interact with the gas and a range of processes then occur depending on which gas is passed into the cell.

What processes occur inside the collision cell?

The main processes that occur in collision cell operation (namely simple chemical reaction, charge transfer and collisional retardation) are shown below, using H2 (in practice, mixed with He for safety reasons) and He as example gases.  For advanced applications, there is an additional process where the ions interact with the gas to form new ionized molecules which are then detected by the mass spectrometer.  This latter class of process will be discussed in part 2 of this blog series.

Cell collision 1

Using a mixture of 8% (v/v) H2 in He (this concentration of H2 is selected as it is below the explosion limit and so safe to routinely use with no special safety precautions), the picture above shows that the ArAr+ interference both reacts with H2 to generate the lighter ArH+ ion along with a neutral ArH species and undergoes charge exchange with H2, thereby removing the ArAr+ interference on Se.  As the ArAr+ signal is very large in ICP-MS, 8% (v/v) H2 by itself is not sufficient to completely remove ArAr+ but goes a long way to help reduce the interference before needing to rely on the third process to finish the job.  This third process, collisional retardation (commonly known as kinetic energy discrimination [KED]), is highly effective in eliminating the remaining ArAr+ interference on Se.  Although KED occurs with H2/He gas mixtures in the cell, for this mode of interference removal, the H2 is not required and pure He on its own is generally used.  KED doesn’t stop at just removing ArAr+ interference on Se though – it is also highly effective at removing a wide range of other polyatomic interferences, ranging from ArCl+ (which interferes with As) to SO2+ (which interferes with Zn) and much more.  The schematic below shows how KED works.

Cell collision 2

Basically, as the analytes and polyatomic interferences travel through the collision cell, they collide with He atoms.  In doing so, both analytes and interferences lose some of their kinetic energy.  As the polyatomic ions have a larger collision cross section area than the monatomic ions with which they interfere, they undergo more collisions as they traverse the cell.  The effect is that by the time the analyte and polyatomic interference ions reach the cell exit, the polyatomic ions have lost a lot more kinetic energy than the analyte ions have.  By simply setting a voltage barrier between the collision cell and the quadrupole analyzer, you can then very effectively filter out the lower energy polyatomic ions, thereby removing the interference.  As the analyte ions have also lost some kinetic energy, some of those ions are filtered out as well (so the sensitivity goes down) but overall the proportion of analyte ions that make it through to the quadrupole is high enough to dramatically increase the analyte signal to polyatomic interference ratio.  This leads to much more accurate results and lower detection limits in KED collision cell mode compared to standard, non-cell mode.

So, how effective is He KED collision cell mode, really?

The answer is, very effective!  The power of He KED collision cell operation is shown in the two pictures below.  The first was a scan of a multi-component solution taken in standard, non-cell mode, and the second was a scan taken of the same solution across the same mass range in He KED mode.  Note that in these scans, 40Ar2+ is not shown as this mass is automatically skipped in the default scan set up.  The scan window can be set up to scan this mass if you want to, but, I admit it, I forgot to do this before making the scans.

Cell collision 3


Cell collision 4

The figures above show a dramatic reduction in a wide range of interferences when going from standard to He KED mode, but they don’t show what the practical effect is on analyte measurements.  The figure and table below illustrate what performance improvement can typically be achieved in practice in a routine, everyday laboratory environment with regular purity reagents.  In the table below, the performance of the cell is measured in terms of the improvement in the blank equivalent concentration (BEC) (i.e. the calibration intercept divided by the calibration slope) when going from standard, non-cell operation to He KED collision cell mode.

Cell collision 5

The above table shows that orders of magnitude lower background equivalent concentrations can easily be achieved in He KED mode, compared to standard mode, for all of the analytes listed.  For V, further reductions can be achieved by employing reactive cell chemistry using gases such as ammonia – more on that in part 2 of this blog.  For Fe, Cu and Zn, higher purity reagents will provide even lower backgrounds.

With He KED mode operation, the lighter the analyte mass, the more kinetic energy it loses as it passes through the cell, which means that very light masses such as Li have much lower sensitivity in this mode.  However, with the right collision cell design and optimization, it is still easily possible to measure these light masses down to parts per trillion levels in He KED mode, which means that full, multi-element analysis can be achieved without having to switch between collision cell and standard mode.  This saves time and simplifies the analysis set up.

If you have any questions about ICP-MS collision cell operation and optimization, or if you’d like to learn more about how Thermo Scientific’s ICP-MS instruments can help meet your needs for trace element analysis, just let us know via the comments box below!

Additional Resources

To learn more about the Thermo Scientific™ iCAP™ RQ ICP-MS, see here.

For further applications information for the Thermo Scientific™ iCAP™ RQ ICP-MS, see here.

Visit our Community pages at to discover more about our solutions for Food and Beverage, Environmental, Pharmaceutical , Biopharmaceutical analysis and other application areas.
Not applicable
How does the KED work with isotope ratios to make the ions going into the quad have the same kinetic energy? Some studies suggest that this will reduce the mass discrimination effect.
Team TFS
Team TFS
Thanks for your question. Using KED operation is generally expected to increase mass discrimination compared to standard, non-cell operation, because it gives higher suppression on the transmission of lighter isotopes of an element as these have lower kinetic energy than the heavier isotopes. It is, however, true that the effect of collisional damping inside the pressurised collision cell (particularly in pressurised cells operated in non-KED mode) has been shown to improve isotope ratio precision, as this process tends to smooth out some of the noise in the ion beam as it traverses the cell.
Not applicable
I've been using Thermo iCap triplequad to measure Li via KED mode (hooked up with laser ablation), and have trouble to resolve Li due to the reasons stated in the text. I wonder how one could approach a right collision cell design and how to optimize the signal?
Team TFS
Team TFS
Hi Karl, In KED mode, Li sensitivity is always significantly reduced because of scattering by the He gas in the cell, and this is hard to improve with any collision cell design. If you can, I would advise running the samples once in KED mode for those elements that benefit from this mode, and then again in standard mode for elements such as Li that aren’t generally significantly interfered and have better sensitivity in standard mode.