The following examples cover some of our practical experiences on process development for pretreatment, fermentation and downstream processing.
Steam explosion pretreatment
When developing the steam explosion proces for a biomass, following process steps are important to consider: size reduction of biomass, chemical impregnation and the actual steam explosion.
At BPF we are using steam explosion, because it's efficient way to open up the biomass. Moreover, it's also a technical and economical proven process at commercial scale. In a typical steam explosion pre-treatment run the process flow starts with size reduction of the biomass, followed by chemical impregnation and the actual steam explosion. The pretreated feedstock is now more susceptible for enzymatic hydrolysis, which liberates the sugars. Leftover lignin and other solids are then separated and the sugar solution is concentrated.
Scaled-down pilot experiments are needed for the following main reasons:
- Steam explosion cannot really be mimicked on lab scale, testing of steam explosion at labscale is technically not possible
- Flowability of the biomass of different sizes, biomass sludge handling and filterability is strongly dependent on biomass source and pretreatment conditions
- No real theoretical validation exists to predict this behaviour
Limited information is published in the public domain because of IP reasons
So getting pretreatment to work on scale is challenging. Commercial scale technology is feasible but a lot of details in scaling up need to be handled well. Important to realize that a pretreatment process needs to be optimized for each feedstock, each enzyme technology and application.
That’s where piloting comes in. It helps to get process and product details right , to mitigate risk mitigation, experience from the learning curve at pilot scale and finally as well a proof of concept to convince investors / biorefinery operators to invest.
We can kickstart your biomass evaluation based on our lab, pilot and industrial scale knowledge on biomass pretreatment and enzymatic hydrolysis.
Choice of raw materials
Use the final product target as starting point. Should the product be used in Pharma (non-clinical)/Feed/Food/Bulk Chemical? Choose the right raw material grade from the start.
Furthermore, take into account the raw material costs:
Assume production scale is 200 m3. Kanamycin is needed to maintain plasmid stability of the production organism. Its concentration is 50 mg/l and the cost price is 3 USD/g. The process has been developed on lab scale by the customer. The total cost of kanamycin is 1.50 USD at 10 L scale.
However, customer targets to produce at 200 m3 scale. This makes the total cost of Kanamycin at this scale 30.000 USD. A great technical choice at lab scale is not always economically viable at commercial scale depending on the targeted end-product selling price.
Seed train optimization
The seed train is the starting point for a fermentation process. A typical example: 1 culture cryo-tube is added to a flask containing 500 mL of medium.
After 48 hours of incubation the complete flask is transferred to a lab scale fermenter containing 9500 mL medium (5% inoculum ratio). This means on 200 m3 factory scale you will need 20.000 flasks to inoculate your fermenter in order to obtain a similar amount of generations of growth. As you can imagine this is not practical.
If more generations are needed on planned scale of production, make sure to study the effect of more generations and the inoculum strategy on the process on lab scale or pilot scale. Are productivities maintained? Is the plasmid stable? Do viscosities change?
Foaming behaviour is complex to model and in some cases unpredictable. Finding the right balance between aeration and agitation while the fermentation is proceeding without interruption is a challenge.
Extensive foaming interrupts the process and leads to yield loss or loss of the complete fermenter. Dosing anti-foam to the fermenter can keep foaming under control.
Anti-foaming agents are however known to potentially inhibit growth and can affect DSP by, for example, blocking membrane filters or negatively influencing product crystallization. Depending on the final application, the choice of antifoam can give rise to issues with final product registration.
Careful selection of a suitable antifoam type and determination of the maximum amount of antifoam that can be added without hampering DSP needs to be determined on lab or pilot scale.