Automated Reaction Optimization

As a suitable model reaction to highlight the steps necessary to successfully translate thermal conditions to microwave conditions, and to outline the general workflow associated with any microwave-assisted reaction sequence, in this section we describe the complete protocol from reaction optimization through to the production of an automated library by sequential microwave-assisted synthesis for the case of the Biginelli three-component dihydropyrimidine condensation (Scheme 5.1) [2, 3],... 

However, a commercial instrument must have a broad appeal and the chemistry regime based on acid/sodium tetrahydroborate offers the most attractive approach. It has a rapid reaction, caters for all of the hydride-forming elements and can be very easily automated. To optimize the procedures, the use of standard Technicon AutoAnalyzer methodologies, i.e. matching blanks, standards and sample matrices, overcomes the difficulties due to matrix interference. Also, the detection levels achieved by a well-designed system allow the samples to be diluted, avoiding any problems still present due to matrix interference. Stockwell [9] has described a system which achieves detection levels that are an order of magnitude better than other systems. 

The time-series array of catalytic spectral data Af-xy can be processed in order to remove the reference spectra, and thereby simplify the reaction spectra. For infrared spectroscopy, this requires proper subtraction of the background moisture and carbon dioxide, the solvent, and some of the reagents. This process can be done manually by trial and error, or it can automated for optimal subtraction. The output is a set of k simplified reaction spectra, Eq. (10). 

An automated and optimal subtraction procedure was previous developed for Eq. (10). The algorithm takes one reaction spectrum after another and optimally subtracts reference signals. Again, the criterion used is minimization of the signal en-tropy (see Section 4.5.2). The recursive equation used appears in Eq. (11), where again is an experimental spectrum, A is a reference spectrum, and y are scalar coefficients. The values of y are easily determined. 

In this chapter, we will focus on the use of CLP techniques for the synthesis of systematic copolymer libraries using high-throughput approaches. Prior to that, automated parallel optimization reactions that have been performed for different CLP techniques will be discussed. At the end of this chapter there will be a highlight on the latest synthetic approaches to synthesize well-defined polymer libraries. 

Boettger [84] developed and implemented a robotic system for the optimization of a palladium-catalyzed reaction by varying the concentration, temperature, and amount and type of catalyst and additives. Porte and co-workers have developed and applied numerous systems for automated process optimization. Representative reactions include the... 

Automation of laboratory experiments robotics to explore reaction optimization. 

A complete library QC or reaction optimization requires approximately 4000 single-bead mass spectra. The time and effort required to manually interpret data from a given Ubrary synthesis or reaction screen would be enormous. To make efficient use of the collected data, the interpretation must be automated as well. In addition, several calculations are required to prepare the library for biological screening. Diversity Sciences wrote several software programs to aid in the process. A brief description of a few of these programs follows. 

Primers (oligonucieotides) are short single-strand DNA sequences, usually synthesized on automated DNA synthesizer. Upstream (forward or sense) primer and downstream (reverse or antisense) primer are needed for each reaction. Optimal primers concentration ranges between 100 nM and 500 nM (0.1-0.5 pM) of each primer. The molar concentration of the primers can be calculated by the following formula ... 

---