Elemental Composition Analysis in MestreNova
Note: This post has been written by Scott Campbell, Chief Scientist at Sierra Analytics
Many high performance mass spectrometers have sufficient resolving power and mass accuracy to determine the elemental composition of mass peaks in a spectrum. High resolving power may be required to separate two or more components at a similar m/z value, otherwise the unresolved m/z value does not represent a single component. The resolving power required is a function of the exact components present and cannot be predicted in advance, but typically a resolving power of 5,000 to 10,000 might be necessary. High mass accuracy is also needed, on the order of a few parts per million. For masses up to about 500 Da, this level of mass accuracy will often result in very few possible chemical formulas. This capability is provided within MestreNova: Mass Analysis – Elemental Composition.
There are several parameters, also called constraints, involved in elemental composition analysis. The picture below shows the constraints pop-up with the default values.

- Element ranges – the table shown above is the set of elements to be considered in the analysis. By default, C, H, N, O, and S are enabled as shown in the table. To add additional elements, one clicks on the “+” sign button. To remove an element, one can click on its row and then click the “X” button. Or, the minimum and maximum values can be set to zero.
- Tolerance – as mentioned above, high mass accuracy, a small ppm or mDa value, is required. In the case that’s shown below, at m/z 353.078, allowing carbon, hydrogen, nitrogen and oxygen with ranges shown in the table above, and from 0 to 2 chlorines and no sulfur, there are 13 possible molecular formulas within 5 ppm. If the higher mass accuracy tolerance of 10 mDa was used, there would be 76 molecular formulas within the tolerance.
- Adduct – if we choose, we can specify an adduct, in this case protonation, and also display in the results the unadducted molecular formulas.

The correct formula, since it’s known in this case, is the formula for griseofulvin, C17 H17 O6 Cl, which is fourth in the list. At this point, there are two refinements to the constraints that can be made. The molecular ion region of the spectrum is shown below.
Since the ionization mode of the spectrum is ESI with protonation, the molecular ion will be an even electron ion. Thus, we can select “even” for electron mode (as opposed to “odd” or “both”). The constraints pop-up at this point follows.

Now the results table shows a list of seven results, down from thirteen, since odd electron ions are excluded.

The final consideration is the molecular ion isotope cluster, shown below. It is quite apparent that the molecular formula contains a single chlorine (or a large number of sulfurs, which is unlikely). This is reflected in the “cluster match” column above, which is a measure of the similarity of the computed isotope clusters of each formula vs. the acquired spectrum. Only the third and fourth formulas match to a high degree, 0.999 and 1.000. The other cluster match values of about 0.85 indicate a significant difference in the computed vs. acquired clusters.

