Return to The Network

Lab of the Future & Synthetic Route Selection

The number and power of reactions developed in synthetic chemistry over the past 30 years are staggering, particularly in the area of asymmetric synthesis and the use of transition metals. Despite that, the time and effort required to make complex molecules has not dramatically fallen. The key reason must be that the selection of which reactions to use in a synthesis has fallen behind. Typically for reasonably complex molecules the ratio between the number of steps in a successful total synthesis, and the number of reactions performed to achieve that synthesis may be less than 1:100. Both the need to re-optimise reactions and to find alternative routes when proposed steps fail contributes to the poor ratio. If we could choose the right reactions and conditions first time a huge step would have been taken towards Dial-a-Molecule. The use of data on reaction conditions and outputs is thus central to Dial-a-Molecule.

Synthesis, particularly in academia, is still largely carried out manually, and documented in a paper laboratory notebook. Analytical data is still mostly collected ‘offline’ after work-up of reactions. Automated and high throughput equipment is not generally available.

To achieve the Dial-a-Molecule goals synthesis needs to become a data driven discipline that makes full use of the revolution in computing and automation that has taken place in the past 20 years.

Lab of the Future is about how synthesis should be performed, and Synthetic Route Selection is about the prediction of which reactions to use. The two are so closely linked that they are dealt with as one theme. The theme was divided into four areas to allow focused discussion, but all are strongly interdependent and indeed many of the same ideas came to the fore in each group although below they are generally discussed in one.

This focus area is principally about collecting and making available data. A key enabling step to Dial-a-Molecule would be to collect and make available information on all reactions carried out, including failed reactions. The necessity for more complete information is also a contributory factor in the common difficulty to repeat, or at least to require substantial re-optimisation of, literature reactions. A Smart Laboratory should automatically collect and make available all such data, as well as enabling the precise repetition of reaction conditions. For the immediate future the promotion of e-lab note books in academia together with the infrastructure to allow sharing of the data produced is an important step towards Dial-a-Molecule. This focus area aims to promote development of the technologies outlined, as well as their widespread adoption in academia.

ORRD is about how to use data to predict the outcome of reactions and hence allow optimum synthetic route selection as well as the data analysis side of reaction optimisation. Making full use of past data on reaction outcomes is crucial, but theoretical methods are also of importance. In addition, the theme covers methods to optimise reactions as they are carried out, and to efficiently experimentally define the scope and robustness of reactions. The selection of the best synthetic route under various constraints (scale, cost, sustainability, time, and available equipment) is another strand of the focus area. In each case developing suitable models and performance measures that can be optimised under realistic constraints provides a unified approach.

This focus area explores how the technology used to carry out synthesis should change to achieve the Dial-a-Molecule goals. Meeting the challenges set by Dial-a-Molecule requires a wide spread adoption of technology to allow for a step change in synthetic capability. The aim of NGRP is to look at the prospects for, and likely impact of, new technology for carrying out synthesis. It also promotes the design and adoption of such new technology. Equally important is to consider both near-term (what affordable changes would have the greatest impact) and longer term (e.g. fully automated general synthesis machines) developments.

RRA is concerned with  the challenge of collecting full analytical data on reactions. The capability to rapidly, ideally in real time, analyse the components in a reaction is critical to the achievement of Dial-a-Molecule, for example by enabling the auto-optimisation of reactions. It requires both dramatic reductions in size and cost of analytical / spectroscopic instruments, and huge advancement of the data analysis techniques used to obtain the required information.