Use a Uniform Spread Design

When the objective is either to make a carful screening for finding a good candidate for future development, or to study whether a gradual change in the performance of the reaction could be traced to the properties of the reaction system, a design which affords a selection of test items which are uniformly spread in the score plot should be employed. An example of this principle is given in the study of the Fisher indole reaction below. 

4 Use a Statistical Design to Explore Several Dimensions of the Reaction Space 

it is of interest to study the joint influence of varying the substrate and/or the reagent(s) and/or the solvent. This implies that several dimensions of the experimental space should be explored. As interaction effects are to be expected, it is necessary to use a design in which all principal properties are varied simultanously. This can be accomplished by factorial designs or fractional factorial designs in the principal properties. 

Complete multi-level factorial designs would usually yield too large a number of test systems for a first approach to new reaction systems. It is possible to reduce the number of test systems and yet achieve a selection which covers a large part of the entire reaction space. This can be achieved by a selection made from a two-level fractional factorial design. The principles are illustrated by an example provided by the Willgerodt-Kindler reaction. 

---