Someone should start up a new journal entitled the Journal of Failed Reactions. I’m serious. The contemporary primary literature is a collection of success stories. Fine. This is how things are supposed to be.
But—every chemist with experience in the field of organic synthesis knows that the vast majority of the reactions we attempt are, in the end, and in one way or the other, unsuccessful. This is not because we are bad scientists. This is because we are always pushing the envelope of what is possible. We cannot discover new things, unless we try new things out.
The problem with having access to a collection of only the successful reactions is that we cannot know whether someone has already tried (and failed with) the next thing we want to try.
A concrete example: A colleague had a boronic ester of a nitrogen containing heterocycle. She wanted to know if it was possible first to oxidize it to the N-oxide, then perform a Suzuki coupling. We thoroughly searched the literature, but we couldn’t find a single example of a boronic ester/acid containing an N-oxide moiety as a reactant in a catalytic cross-coupling reaction.
What does this mean? That no one has tried it, or that a thousand graduate students have spilled their guts already?
We don’t know. We would have known, if there had been a Journal of Failed Reactions.