Design of experiments: An efficient approach to process optimisation

In any chemical reaction, a multitude of variables can impact yield, completion rate, purity and other key properties. Striking the right balance across these factors to optimise reaction conditions may require lengthy experimentation, and this can be especially challenging for more complex reactions.

Design of experiments, or DOE, is a systematic approach to planning and analysing experiments that enables scientists to efficiently establish reaction conditions, reduce variability, accelerate timelines and contain long-term costs. As the margin for error and inefficiency in the pharmaceutical industry has never been lower, DOE enables scientists to harness data to make informed decisions about reaction parameters with high levels of precision and efficiency.

Today, as scientific complexity has risen and DOE software has become more sophisticated and accessible, more and more pharmaceutical organisations are harnessing DOE to improve reaction outcomes without impacting project timelines or adding unnecessary complexity to the project.

In this whitepaper, we’ll take a closer look at DOE and the advantages it delivers, then step through a use case that highlights DOE’s ability to improve reaction outcomes.

Exploring DOE and its advantages

At its core, DOE helps scientists improve reaction conditions relative to their existing approach. Without DOE, scientists can make informed conclusions about where optimal reaction conditions might lie, but it could take dozens, or even hundreds, of experiments to cover the full range of parameters and identify the most favourable reaction conditions.

Conversely, with DOE, scientists can identify these conditions with greater speed and accuracy. DOE software allows scientists to set ranges for reaction parameters like time, temperature, stirring speed, and others. Then, the software generates a limited number of experiments for scientists to run that will help them get closest to their desired results and determine which parameters are most impactful. For example, DOE can elicit conditions that yield a higher percent of a desired product while reducing the concentration of undesired byproducts. ... Read more

Uncovering the advantages

In active pharmaceutical ingredient (API) development, DOE can deliver an number of powerful advantages. By enabling greater efficiency and control for scientific teams, DOE ultimately leads to long-term cost savings and optimised processes for the end customer. Some of the benefits include:

DOE supports robust analysis of reaction parameters and conditions with a small number of experiments, allowing scientists to efficiently cover more ground without independently testing each variable in a given reaction.

For more complex reactions, many different parameters can have a significant impact on results, and some of these parameters may have intervariable relationships with one another. DOE aids scientists in determining the best values for each of these parameters, eliminating guesswork and supporting greater control over the process.

Complex reactions can be especially sensitive to any change in reaction parameters. DOE helps scientists to isolate variables and gain a more complete understanding of the impact of such changes, allowing for greater consistency and a narrower focus as they establish conditions for a particular reaction.

DOE aids in saving long-term costs by optimising resource utilisation in the lab and reducing the amount of experimental work needed to achieve a desired outcome. In addition, DOE can enable scientists to identify conditions that will improve product yield, for example, allowing them to produce a greater volume of product in the same amount of time.

The optimised resource utilisation and reduction in experiments afforded by DOE also helps save time over the course of a project. Scientists can not only more quickly define reaction conditions, but also reap the long-term time savings that come from an optimised process.

Harnessing DOE to navigate complex process steps: A scenario

To demonstrate some of the advantages DOE can deliver in a chemical process, let’s look at an example of how we’re using DOE at Sterling.

The situation

A three-step process involved one particularly complex step that resulted in low yield. The reaction generated five byproducts, all of which were similar in chemical structure to the actual product and in equal proportion, making them challenging to separate out. Since the byproducts could not be removed using traditional column chromatography, the reaction yield remained around 10%.

Beyond the low yield, the reaction leveraged hazardous reagents and required expensive starting materials, presenting opportunities for risk reduction and cost containment using DOE.

The solution

To increase yield of the desired product while reducing the percentage of undesired byproducts, Sterling conducted a DOE exercise to identify reaction conditions that would result in improved yield and higher purity.

The results

Accelerating development with DOE

While DOE has been around for some time, never has it been more relevant. This approach offers a direct response to rising molecule complexity, narrowing margins for waste, and the quest to continually improve reaction outcomes.

DOE empowers scientists to optimise reaction conditions accurately and efficiently by limiting the range of possibilities and delivering high-integrity experimental data. It moves beyond the time and granularity limitations of a manual approach and empowers scientists with a stronger understanding of reaction parameters and their interconnected relationships to help them more accurately define reaction conditions for future processes. ... Read more