Earlier ICP/MS models work well for samples with low background but are limited for use when samples have more complex matrices. For example, samples from marine environments that contain high chlorides can be difficult to analyze with ICP/MS because they form polyatomic ions that can mimic target metals in the detector. The collision cell technology uses kinetic energy discrimination to remove these interferences, making the analysis more rugged and reliable. In addition, the software used to process the data applies correction equations that also help to minimize the impact of the sample matrix.

Bob Strocko, manager of the Metals Analysis Department, is pleased that the new instrument provides improved matrix tolerance. “With the earlier instruments, analysts needed to measure samples at multiple masses and then make complex decisions based on the results obtained. The collision cell technology allows us to make a single measurement and get the right answer,” he explains.

There are other benefits as well. The detection limits from the original ICP/MS units are in the part per trillion range, and the new instrument is capable of similar or better sensitivity. In the past, sample dilution was used as a means of minimizing the negative impact of matrix interference, but there is a down side to this approach: it will cause the detection limits to be increased. Because fewer samples will require dilutions when the collision cell technology is employed, clients can expect better detection limits for some types of samples.

“This new instrument will help us serve clients more efficiently because we can determine multiple elements simultaneously with fewer dilutions or re-runs,” says Strocko. This is especially true when analyzing for arsenic and selenium, which are prone to interference in high chloride samples and were historically determined using GFAA (see accompanying article). “The ability to use ICP/MS to replace GFAA for arsenic and selenium determinations will greatly increase our capacity for large projects that include those analytes, and clients will get more concise data packages when only one technique is used,” adds Strocko.