Stefan Schlüter and Tim Schulzke on sustainable methanol production

We are currently in the second phase of the research project Carbon2Chem^®, in which, among other things, the large-scale implementation of sustainable methanol production based on metallurgical gases is set to be validated. What are the next steps? What does simulation mean for real plant operation? Interview with Tim Schulzke and Dr. Stefan Schlüter.

First of all, please can you briefly outline the purpose of the Carbon2Chem^® research project?

Stefan Schlüter: The core aim of the project, which was launched in 2016, is to make steel production at thyssenkrupp greener. To do this, the carbon monoxide and carbon dioxide produced in steel production are recycled chemically. This specifically involves the product paths of methanol, higher alcohols, ammonia and polymers. Carbon2Chem^® is aiming to make a large proportion of the German steel industry’s annual carbon dioxide emissions economically useful. This approach is a kind of bridge on the road to genuine green steel, for the production of which sustainable hydrogen will be used instead of coking coal in the future.

Fraunhofer UMSICHT is highly experienced in methanol synthesis and is providing expertise in this matter.

Tim Schulzke: That’s right. We have been working on synthesis gas chemistry at the institute for a relatively long time, specifically on producing methanol and dimethyl ether from synthesis gases. Fraunhofer UMSICHT has several test plants of varying sizes and with different research focuses that we are using in the context of Carbon2Chem^®.

How exactly do the plants differ?

Tim Schulzke: Catalyst behavior is tested at the smallest test plant. Which gas compositions does a catalyst need to deliver reasonable product amounts and product qualities? Which conditions result in more by-products, and which result in less? This gives us detailed information about the function of gas purification in the steel mill. The second test plant is somewhat larger in size and is used to investigate the reactor behavior. We measure the temperature control, the pressure and the temperature profile in the reactor. Finally, we have the demonstration plant in the form of a container. This is where we examine the behavior of the plant, i.e., the interaction between catalyst, reactor, removal of the liquid product and gas recirculation. For example, we can detect negative effects due to the buildup of trace gases that are not observed in the single-pass operation of the smaller test plants. We are investigating these relationships first with cylinder gases and later with real gases from the steel mill operation.

How does gas recirculation work in the demonstration plant?

Tim Schulzke: At the start, the so-called make-up gas enters the plant and is passed through the reactor. When the gas is cooled at the end, methanol and water vapor condense and can be separated from the remaining gas. The result is methanol and a bit of water. However, for thermodynamic reasons, the make-up gas can only be partially converted in the reaction, so a certain amount of hydrogen and a little carbon monoxide and carbon dioxide are left over. These are recycled as recycle gas and mixed into the make-up gas, and the resulting feed gas goes back into the reactor. The mixing ratios mean that there is always a bit more hydrogen than is actually needed for the reaction. This results in a very high hydrogen concentration in the recycled gas, which in turn leads to a high conversion of carbon. The excess hydrogen and inert constituents are given off to the environment via what we call purge gas, so as to avoid a rise in pressure. If a gas mixture containing only hydrogen and carbon carriers is used for the reaction, the purge gas flow rate is relatively small. With blast furnace gas, on the other hand, the flow rate is significantly higher, as the make-up gas contains up to 22 percent nitrogen that has to be given off again.

As well as practical tests, Fraunhofer UMSICHT also focuses on simulation.

Stefan Schlüter: We simulate the entire system. This includes the metallurgical gas purification stages, the compressors, the actual methanol plant and other technical equipment, such as separators. In principle, everything involved in producing a salable product. We are working on two tracks: The previous practical experiments are recalculated using detailed models for the methanol process. Simplified models based on simpler assumptions are then parameterized using this calculated data. We need the simplified models for the overall simulation in which we simulate all the necessary units at the same time over a relatively long period, e.g., one production year. We must have computing times in a range that we can still manage. We aim to simulate the entire year-long process over about one day.

Tim Schulzke: Thanks to the simulations, we can define a technically feasible framework. While physics is limited in practice, we are relatively free in the simulation — for instance in terms of maximum pressures, temperatures or flow rates. And we can model many more points than would be possible in a limited test time.

Which main parameters influence the simulation?

Stefan Schlüter: One main parameter is the gas compositions, or more precisely what kind of metallurgical gases are used: blast furnace gas or converter gas? At present, we mainly use blast furnace gas, as it is likely to be available at the site for a relatively long time. The question is how the purification works. We cannot simulate this with sufficient accuracy and therefore have to rely on the results from the lab tests. The design of the actual methanol process is also crucial. There are a number of control variables here that are also used in traditional methanol plants to optimize the conversion. Since the metallurgical gas streams and also the gas composition are not constant, our models are designed to cope with fluctuating operation. A working group at Fraunhofer UMSICHT is examining the entire process over a prolonged period of time with the focus on economic aspects. When, for example, does it make sense to operate the methanol plant at increased capacity because the electricity for producing hydrogen is cheap or methanol can be sold at a higher price?

How do you verify the results of the simulations?

Stefan Schlüter: We compare our simulations with other calculations and with data provided by the steel mill and by working groups involved in the project at thyssenkrupp. We are also able to compare them with the technical center tests.

In the first half of 2022, the Fraunhofer UMSICHT demonstration plant in Oberhausen is to be relocated to the Duisburg steel mill. Are there any plans for after that?

Stefan Schlüter: In the next phase we are going to test methanol production in the technical center at the steel mill using real gases. By the time the project ends in 2024, we should have developed one process for methanol and one for ammonia/urea in sufficient detail that demonstration plants can then be built. As things stand, we are set to achieve this goal.

Can the processes initially developed for a steel mill also be adapted to other branches of industry?

Tim Schulzke: Alternative sources of carbon form part of a Carbon2Chem^® lighthouse project and there are already two visible partners. Firstly, there is GMVA Gemeinschafts-Müll-Verbrennungsanlage Niederrhein GmbH as part of REMONDIS Assets & Services GmbH & Co. KG, where the flue gas from waste incineration is purified and, as a result, carbon dioxide is obtained for synthesizing methanol and producing other products. Secondly, there is lime specialist Lhoist Germany Rheinkalk GmbH. It should be said that in this case, carbon dioxide is inevitable in the process, i.e., it cannot be avoided. It is produced when using heat to convert limestone (calcium carbonate) into burnt lime (calcium oxide). The next goal therefore is to chemically bind the carbon dioxide from the lime kilns with electrolytic hydrogen to form methanol and to use this as an e-fuel or raw chemical material, for example.

Tim Schulzke (Low carbon technologies department)
... is responsible for adapting the existing demonstration plant to local on-site conditions in the steel mill. He also conducts test campaigns to obtain baseline data and to test the use limits.

Stefan Schlüter (Low carbon technologies department)
... is responsible for simulating the entire system using mathematical methods and programs developed for this purpose.