Orthogonal experiments in the development of organic synthetic processes

• A new strategy is presented for the design of explorative experiments in synthetic chemistry when the objective is to identify the important experimental variables. The methodology is based on Taylor expansion (response surface) models, and the principles are: A grid of possible settings of the experimental variables is laid out in the experimental domain. These experiments define a candidate design matrix, DC. From DC, a candidate model matrix, XC is defined by appending columns for each variable in the Taylor model XC is then factored by singular value decomposition (SVD), and XC = USVT. The rows in XC that are most parallel to the singular column vectors in V are selected, and the corresponding experiments in DC are identified. This gives the experimental design. The selected experiments are nearly orthogonal, and they span the dimensions of the model space. The experiments can be run in sequence, and thus, they allow for a systematic search, one experiment at a time. The design principles are illustrated by an example of the dibromination of an acetal. Four variables were studied, and from 12 experiments, all the main effects and all two-factor interaction effects were estimated. From the response surface model, conditions for quantitative yield were predicted, and a mol-scale synthesis carried out under these conditions afforded 98% yield of the isolated pure, >97% product.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Signalbehandling (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Signal Processing (hsv//eng)

Nyckelord

Signalbehandling

Signal Processing

Publikations- och innehållstyp

ref (ämneskategori)

art (ämneskategori)