kilomentor

Optimization continued

kilomentor | 10 February, 2011 22:50

This particular article part can be found at http://www.kilomentor.com/2011/02/optimization-continued

 

Results B

Experiment # Yield Rank Yield Rank (100-Rank)/(N+1)
1 30 15 1 7.1
2 63 23 2 14.3
3 65 26 3 21.4
4 26 30 4 28.6
5 37 37 5 35.7
6 15 42 6 42.9
7 50 50 7 50.0
8 23 63 8 57.1
9 75 65 9 64.3
10 94 75 10 71.5
11 42 76 11 78.5
12 91 91 12 86.7
13 78 94 13 92.8
Results C
Experiment # Yield Rank Yield Rank (100-Rank)/(N+1)
1 0 0 1 7.1
2 45 0 2 14.3
3 0 11 3 21.4
4 11 15 4 28.6
5 15 23 5 35.7
6 32 24 6 42.9
7 53 32 7 50.0
8 33 33 8 57.1
9 44 42 9 64.3
10 24 44 10 71.5
11 60 45 11 78.5
12 42 53 12 86.7
13 23 60 13 92.8
 

Chemical terms yield

Comments

Remainder of article

kilomentor | 11/02/2011, 12:36

If you press general above you can find the remainder of this article.

Re: Optimization continued

Not Process Chemist | 04/03/2011, 11:55

What is the theory based of probability formula "(100-Rank)/(N+1)"? Is 13th experiments can reflect to the probability of 93%, and whether choosing optimistic reaction conditions by chemist is subject to "random". IMO It's really hard to deal with discrete variables in reactions, too many parameters in "one" reaction (addition sequences, quench time, quench temp, catalyst or reagent purity), needless to say different reagents. it may be more suitable in some robust reactions(coupling or recrystallization?) or reactants(without labile group)for optimization method.

Add comment

authimage

 
Powered by LT - Design by BalearWeb