Gradient myths

With the routine use of gradient separations in the laboratory, practices have become normal and ideas fixed, which are not always appropriate or do not always correspond to the reality. In the following, some of these myths are mentioned and briefly commented on. Relevant examples are discussed in the following. 

certainly not always. In several cases, we have found that both the resolution within certain areas of the chromatogram and the peak capacity in general improve if one starts with about 30-40% . A specific literature search on recent publications has confirmed these findings. The significant elution strength of such an initial mixture obviously leads to an early differentiation of the sample components their potentially different properties and thus their different degrees of interaction with the phase material become noticeable with approximately 30-40% , while with approximately 95% water/buffer almost all components sit at the column head, making a separation of relatively early peaks more difficult. 

For samples with a large number of components one should run a long 

The improvement in the separation with a long gradient has its limits. With a shorter gradient the peaks elute earher, but the peak width is narrower. Long gradients often lead to a significantly better resolution only for the late eluting peaks. 

Use more %B at the start, so that the whole separation becomes faster Of course this is correct, however the resolution may also change. 

---

Some LC/MS users adhere to isocratic separation because of the myths around gradient elution (it is complex to develop and transfer between instruments and laboratories, it is inherently slower than isocratic methods because of re-equilibration, and other reasons summarized by Carr and Schelling6). A researcher may have a very good reason to use an isocratic method, for example, for a well defined mixture containing only a few compounds. The isocratic method would certainly not be useful in an open access LC/MS system processing varying samples from injection to injection. 

---