There can be good accidents as well as the more usual bad ones. Irreproducible results are not always bad; in fact, it is the good result that can neither be accounted for nor duplicated that causes us to figuratively tear our hair out in frustration. It suggests that there is probably something worth knowing that we don’t understand. It is taking advantage of good accidents that more than any other characterizes the inspired process chemist.
For the
process chemist the serendipitous result can arise either in the reaction phase
or in the purification/isolation part. Most
frequently the pleasant surprise comes from some particular efficient or simple
phase shift separation. Niel Anderson in
his book, Practical Process Development,
expresses the point this way,
“When Nature offers a separation, see if it can be exploited to advantage!”
Route
selection already includes some very optimistic assumptions that may be
difficult for reality to surpass. That is why these pleasant surprises are more
likely in the work-up phase.
A strategy
for giving serendipity a chance is what I call the ‘wish reaction’. Ask yourself
what you really wish would happen so long as it is at least theoretically
possible. This single experiment is added at the beginning of an experimental
program. If the experiment fails utterly or is confusing or ambiguous, work can
return to the most probable main route. A simple example involves the use of a
protecting group. First we would try the reaction without protection. If the
experiment shows that protection is needed then we apply protection. Otherwise,
we are heading towards saving both the protection and deprotection steps.
The
Political Experiment
Former R
&D Directors who have been promoted to business executives often like to slip
back into that former role. Thus, if they were in process-chemistry, they
retain interest in its technological problems. They may propose solutions to
you. More precisely, they may repeated propose a particular course of action. It may not be considered promising by the
research team. In this situation I
recommend the ‘political experiment’.
One single simple experiment is done to test the proposal. The
experiment is done with great care, taking extraordinary precautions (dry,
inert, distilled solvents, etc.). The
more likely it is that the result will be an utter failure, the more useful it
is to perform the experiment. When the result is as expected, it can be
reported and the free advice will stop. Do not report the particular care that
was taken. You may be questioned as to whether you did this or that correctly.
This is an occasion for a superior to learn how professional you are! Besides
if some deficiency is found you could be ask to repeat the experiment and you
want to go back to your real experimental program. Of course don’t even mention
that you expected or predicted that the idea would fail. If the experiment
works, be generous in your praise. This boss might know something about this
chemistry that it could be useful to tap further.