Lignin, a byproduct of paper manufacturing, is burned 98% of the time as opposed to reused. It thus serves an important role as an energy source for the paper industry but it prompts the question of whether lignin – readily available in large amounts – could not prove to be equally if not more valuable as a raw material for other industrial processes. Professor Marie-Pierre Laborie of the Faculty for Forest Biomaterials at the University of Freiburg and her team were very interested in this possible research area. They had the idea to manipulate lignin in such a way (e.g. mixing it with another substance) that it could be used as a raw material in typical industrial processes such as 3D-printing. In order to have this idea come to fruition, the team searched for partners who could assist them in the more technical aspects of the project, such as simulations of highly complex manufacturing processes. The Sustainability Center Freiburg offered their services to find a suitable partner. As a cooperation between the five Fraunhofer Institutes in Freiburg as well as the University, the Sustainability Center had ample connections to offer Professor Laborie. Dr. Georg Ganzenmüller and his team from Fraunhofer EMI and the Department for Sustainable Systems Engineering were thrilled about the idea and gladly joined forces with the forestry experts. The eager colleagues collaborated in a project within the Sustainability Center Freiburg, in a three-year project: “Use of lignin as a source material for a biologically based plastic”.
It was clear from the beginning of the project that the processes would need to be not only experimentally tested but also by means of computer simulations. “It is much easier and more cost-effective to tweak parameters of the various aspects to the processes in a computer simulation than in the lab, where every experiment costs time and money”, explains Dr. Ganzenmüller. “However, we can only vary parameters in the simulation once they have been validated experimentally.” It was thus necessary to have two parallel-working teams, one to handle experimental examinations and the other to oversee the computer simulations. It was quickly determined that instead of the 3D-printing process, the teams would concentrate on extruder processes, similarly characterized by a material being pressed through a pre-determined form to obtain desired products. In the lab, Professor Laborie and her colleagues thoroughly formed and inspected phase-separated mixtures of cellulose derivatives and lignin. This mixture exhibits a far lower viscosity than pure lignin which is a key quality required for an extruder process. A further research point for the lab team was the inspection of the fundamental chemistry underlying the entire process. It was planned for this work to be completed within six months of the three-year time frame.
The research questions originally planned to be addressed had to be modified however. “We noticed very quickly that the experiments as well as the computer simulations were far too time consuming and complex for the short time period we had,” Dr. Ganzenmüller reports. The creation of a suitable phase-separated mixture of cellulose derivative and lignin alone would take up the entire funding period. Various problems occurred during the computer simulations, one concrete example of which was the scaling of the simulation of the phase separation. In order to see the miniscule but vital details in the simulation, a micrometer scale resolution was necessary. However, in order to scale-up the simulation onto a centimeter scale, far more time and computing power would need to be invested as was available. Therefore, the researchers decided to overcome this problem by implementing a mathematical formula which would ideally simplify and homogenize the entire simulation model.
However, the development of such a formula is anything but trivial. “A unique characteristic of the extruder process is that a liquid material is shot into the extruder and becomes a solid material throughout the procedure. Standard simulation methods are already widely used for both liquid and solid phases but they can only describe one phase or another. In order to understand the total process, a so-called ‘net-free’ method had to be applied, in which both phases could be inspected,” explains Dr. Ganzenmüller. Before this unconventional method could be implemented however, more ground-laying work had to be done on the method itself. This was assigned to José Luis Sandoval Murillo, a doctoral student at Fraunhofer EMI. After quite some time of arduous work and effort, Mr. Murillo was able to program a basic simulation, based upon various complex formulas. This simulation was not applied to the research question of lignin as a source material in the extruder process however, due to the hurdles described above. Still, an important result of the invested time and effort was the identification of which phase structures under which conditions in the extruder could be formed. With this knowledge researchers could create optimal, desired structures by specifically altering parameters in the extruder process.
Although the original goals of the research team were not reached, a new equally interesting research question was posed. What if the newly developed models were used to obtain exactly determined structures in a completely different medium? The Fraunhofer EMI team had the idea to play with this possibility in form of meat substitute products. Meat substitute products try to imitate meat as closely as possible. This can range from nutritional value to flavor and smell, but also the consistency. Generally meat substitutes consists of denatured vegetative or fungal proteins which present a meat-like consistency after their sent through an extruder and a meat-like flavor is formed due to added aromas. This form of product is momentarily of great interest for science as the demand for vegetarian alternatives for meat products is steadily increasing. The consistency thereof is particularly vital as a meat-like consistency is far more attractive to consumers than the mushy consistency that is sadly often found in such products. Parallel, the university team further dealt with the topic of lignin implementation for the 3D-printing process and was able to show, that the lignin-cellulose derivative mixture could be pressed through a 3D-printer.
The EMI researchers had the idea to further develop and modify their mathematical model to the point that it could be applied to a vegetative protein based meat substitute product. “We wanted to process the vegetative proteins in the extruder system and simultaneously inspect the influence of certain parameters such as viscosity, flow and temperature gradients, and phase separation on the resulting product,” says Dr. Ganzenmüller. The required cooking extrusion procedure is made up of three phases. First, a vegetative protein powder is mixed with water in a twin-screw extruder, second the warming of the mixture in the so-called “cooking zone” of the extruder, and lastly the cooling of the mass, which is pressed through a shaped opening into the open. The resulting product is a mass with a fibrous structure, which when chewed, gives the appearance of being meat.
Pea protein (Pisum sativum L.) was determined to be a suitable plant-based material for the experiments. “Peas are robust and low-maintenance, require very little if any fertilizer, and are grown here,” explains Dr. Ganzenmüller. “They are thereby perfect for a sustainable product.” There are already manufacturers in Germany and Europe that produce similar meat substitute products but certain problems in their production such as deforestation have already been uncovered. Therefore, the demand for a new, sustainable manufacturing process for a meat substitution product is very high. Perfect for this team and their research aims.
The determining mechanism for a meat-like consistency is the phase separation of the protein-water mixture. The experiments proved that one parameter is vital for the resulting consistency namely the temperature. It was determined that the temperature of the mixture at the beginning of the extruder procedure had a direct influence on the structure of the extrudate. During one trial, the starting temperature was 120°C and no visible fibrous structures were formed (see Figure 1, left). With a higher starting temperature of 160°C however, the desired lamellar structures were visible (Figure 1, right). An interesting note is that these lamellar structures were only visible once the water was removed from the product. The flow gradient, particularly the direction thereof, also proved to be a relevant parameter. Due to the friction of the mass on the walls of the tubes of the extruder, a defined flow is present which directly leads to the formation of strips in the resulting mass. Based upon this flow, different strips can be formed. Dr. Ganzenmüller is thrilled with these results “In principle that means that we can mimic different types of meat. On the one hand, we could produce a substitute product with the consistency of a robust steak. Or a delicate chicken breast. Whatever the consumer wishes!”
These experimental results could be verified by simulations, based upon the previous work by Mr. Murillo, undertaken by Fraunhofer EMI colleagues. It was once again numerically verified, how great the influence of temperature and temperature gradients was on the end product. For instance, when the cooling rate of the extrudate was too slow then phase separation did not occur. If on the other hand the cooling rate was too fast, then vertical lamellar structures formed which were also not wanted. A suitable mid-range cooling rate was found which resulted in the desired layers which were parallel to the flow yet perpendicular to the temperature gradient (see Figure 2). “It is this orientation which gives the end product a meat-like consistency,” explains Dr. Ganzenmüller. It was thereby proven that the fibrous structure of a pea-protein mixture could be directly influenced via temperature and flow direction and a “real” meat consistency would be produced.
After three years of research, the team is proud of their results. “Although we had to modify our research goal throughout this period, we were able to showcase new important results which are very applicable in the areas of sustainable research and production, particularly food production,” grins Dr. Ganzenmüller. Even beyond this remarkable implementation, the developed methods have great potential for other areas of research. The interaction between phase separation and typical temperature and flow gradients in an extruder procedure were documented for the first time with help of a simple thermodynamic model. The development of simulation models along with the experimentally gained results and mechanisms could also be utilized in the explanation of different extruder procedures and the resulting end products. The researchers of the University Freiburg and Fraunhofer EMI are therefore already planning a future collaborative project to further develop these mechanisms and simulations. Concretely, the refinement of the available resolution is to be addressed as well as the development of a tool for the simplification of the simulation. This tool could then be used to greatly optimize various industrial production processes. Overall, the collaboration was successful and the partners are happy to have learned so much from one another.