It is recommended to add internal standards for correction of physical interferences. The selection of the element(s) to be used depends on the matrix to be analysed. The element to be added should not be naturally present in your matrix. In addition it should be a monoisotope element.

 https://www.epa.gov/sites/production/files/2015-10/documents/method_1640_1997.pdf

please see section 7.3 and onwards

I would recommend using an internal standard particularly if you expect to encounter matrix, digestion, or extraction interferences or inefficiencies.  At least then you can calculate a %Recovery.

The selection of the element(s) to be used depends on the matrix to be analysed. The element to be added should not be naturally present in your matrix. In addition it should be a monoisotope element.

The use of the internal standard in ICP-MS is not mandatory but can help improve the accuracy of the analytical data. Indeed, it allows the correction of physical interferences that, for instance, can affect the samples and not the standards of the calibration curve (due to the matrix differences) or instrumental drifts during the sequence. You can either add the internal standard to all the samples that have to be analyzed or split the inlet into two lines, one for the probe aspiring the samples and the other one immersed in the internal standard solution. As already mentioned, the choice of the proper internal standard depends on the analytes to measure. For quantitative analysis, one or more elements can be considered, especially if a wide mass range has to be covered, as an internal standard with the mass number and ionization characteristics similar to that of the analyte improves the precision (10.1016/0039-9140(92)80088-u). Element concentration in the internal standard solution should be chosen to give a suitable and stable cps signal (usually in the ppb range).

Guidelines (Practical guides to ICP-MS, Robert Thomas) for the selection of the proper elements for the internal standard solution indicate that:

1. They are not present in the sample.

2. The sample matrix or analyte elements do not spectrally interfere with them.

3. They do not spectrally interfere with the analyte masses.

4. They should not be elements that are considered environmental contaminants.

5. They are usually grouped with analyte elements of a similar mass range. For example, a low-mass internal standard is grouped with the low-mass analyte elements and so on up the mass range.

6. They should be of a similar ionization potential to the groups of analyte elements, so they behave similarly in the plasma.

7. Some of the most common elements/masses reported to be good candidates for internal standards include 9Be, 45Sc, 59Co, 74Ge 89Y, 103Rh, 115In, 169Tm, 175Lu, 187Re, and 232Th.