commercial production. It is useful to consider the challenges from the perspective of precision fermentation and cell culture( another widely used engineering biology technique).
At present, cells are typically grown in bioreactors then used to produce high-value, low-volume ingredients. Scaling these methods to optimise the production of highervolume speciality chemicals is the logical next step.
However, the natural variability of biological systems and the sensitivity of bioprocesses to environmental conditions mean that simply increasing bioreactor volume is not the answer. It risks compromising the transport efficiency of oxygen, nutrients and waste products as the bioreactor tank is stirred, impacting quality, yield and titre.
These scale-up challenges are significant, but not insurmountable. They can be mitigated using novel bioengineering methods, process optimisation and genetic engineering along with advances in artificial intelligence( AI) and machine learning( ML). The following three-step process supports the leap from the bench to the bioreactor for cost-effective industrial production of chemicals.
Step 1: Bench-scale prototyping
The first step characterises the process and demonstrates feasibility via metabolic modelling.
It is common for bench-scale engineering biology to focus on demonstrating the feasibility of a given chemical’ s production. However, scale-up requirements should also be considered at this early stage. It is useful to characterise the dependence of the process on various parameters, in terms of underlying metabolic mechanisms as well as outputs.
Acquiring this insight necessitates strategic Design of Experiments, which explores different process parameters. These may be intentionally suboptimal in some respects to generate a range of data surrounding genomic, proteomic, metabolic and
environmental factors. The resulting data can be processed to produce a mathematical metabolic model that guides system development.
Various modelling frameworks can be used to describe fermentation processes, including constraintbased modelling, flux balance analysis and metabolic flux analysis. These mechanistic models provide a useful framework. However, they can be taken further with ML, which is increasingly used to determine parameters from experimental data, yielding a hybrid approach to metabolic modelling. 9
The initial application of the metabolic model is to identify critical environmental parameter ranges, which is a key input for Step 2. It can also help identify critical measurands for process monitoring and control.
Barriers
Feedstock costs & sustainability
Expensive downstream processes
Use of fossil fuels in production or processing
Potential solutions
Step 2: Development of design rules
Using engineering biology to achieve commercial-scale production that rivals traditional chemical synthesis in terms of cost-compatibility with end products is a major difficulty. To help address this, all phases of scale-up should be considered at the outset, not on an incremental basis.
This allows focused use of intermediate phases to establish and de-risk aspects of the full-scale process. For example, the first phase could demonstrate the design principle for maintaining efficient transport of gases, nutrients and waste products, as well as demonstrating productivity. The next could involve additional monitoring and process control, demonstrating robustness.
Source more sustainable, lower cost substrate materials( e. g. agro-industrial water, which is rich in carbohydrates and lipids).
Explore novel technical solutions with lower costs attached, such as the use of gravity separation for purification.
Improve the efficiency of column chromatography. Adopt continuous and automated processing.
Introduce production methods that are more compatible with renewable energy sources.
Figure 2- Barriers to sustainable & cost-efficient production of biobased chemicals
74 SPECIALITY CHEMICALS MAGAZINE ESTABLISHED 1981