Many chemical reactions are heat initiated. Consider a compound X1. Heating this compound causes the reaction:
X1 -> X2
as shown in the figure below. However, the heat energy that causes the above reaction may also initiate:
X2 -> X3.
The result is that the yield of X2 is low and many researchers measure the yield of X2 to try and improve conditions that increase the yield of X2 directly. If this is not achieved the scientist has to come up with ways of separating X2 from X1 and X3. Purification methods are therefore.
Is there a better method?
Conventional method
|
How do we do it?
|
Our strategy is as follows:
Instead of measuring the yield of X2 directly, the researcher should consider:
Y1 = X1
Y2 = X1 + X2
Y3 = X1 + X2 + X3
and convert the problem into minimizing Y1 and maximizing Y2. The new metric is then:
NPM = -10 log (MSD of Y1) - 10 log (MSD of Y2)
as shown in the second figure above and the problem is one of maximization.
|