Automation and Experimental Design

Automation is indispensable for high-throughput chemical development. Commercially available automated synthesizers and custom-built instruments, some of which are also useful for chemical process development, have been recently reviewed [7]. Reflecting the diverse requirements of process development in different phases, different types of high-throughput tools have been developed for a variety of modes of process development, since no single instrument can cover all activities. Further information regarding laboratory automation, including commercially available automated systems for chemical process research as of late 1998, is available [5]. Although the review of equipment available is already out of date because of the rapid development in this field, Weinmann provides many excellent examples of how automation can be applied to chemical process development. More recent additions to these instruments that are especially aimed toward chemical process development will be discussed in the following sections.

Some benefits of automation for chemical process development are:

• The time necessary for optimization of the process is shortened, leading to the faster development of the drug candidate.

• A much wider range of variables can be selected, giving an improved chance of success in process screening.

• The quality of data will improve by eliminating human operational errors.

• Better yields can be obtained owing to the more precise control of reaction conditions.

• Large amounts of experimental data are efficiently collected with minimum operator involvement.

• The staff is released from tedious repetitive tasks, allowing more time for creative work.

• Researchers will have more flexibility so that operations involving monitoring reactions over extended time periods are possible.

Automated process development also facilitates the statistical design of experiments (DOE), i.e. full or fractional factorial design of experiments and construction of mathematical models or response surface models of the reactions. The DOE approach often helps to identify crucial factors and their interactions that govern the chemical process. These types of information are useful for developing a robust chemical process [8]. Indeed, the combination of statistical DOE and high-throughput automated experimentation has been proven to be a powerful methodology for the development of a robust process. However, it should be noted that this statistical approach can only be feasible if enough chemical knowledge on the reaction of interest is available, which should be the case at the optimization stage of developmental activities.

Owen et al. describe the benefits of applying experimental design in process research as follows [9]:

• Powerful mathematical models of the chemical process or procedure are produced that allow opportunities or constraints to be fully considered.

• The good quality of data allows better strategic decision-making and faster scale-up of optimized processes into plant.

• The models are obtained for a quantifiable amount of resource.

• It is an efficient and effective method of choosing which experiments to perform.

• The strategy is compatible with running automated reactions in parallel.

• If circumstances change (e.g. the price of reagent increases dramatically or work-up alters), the model can be interrogated in different ways to take the new criteria into account.

• The methodology can be used as a framework to capture and share information between project teams.

Was this article helpful?

0 0

Post a comment