From research challenge to standard product

What started as a practical challenge in an ongoing research project gradually developed into a collaboration with the Danish company Unisense and a technical solution that has now become part of the company’s standard product range.

A bottleneck in the research

Gas fermentation is a technology in which microorganisms use gases such as hydrogen, CO₂ or methane as feedstock and convert them into valuable products such as chemicals, fuels or other bio-based compounds. To understand and optimise some of the most promising processes, it is important to be able to measure hydrogen levels directly in the bacterial culture inside the bioreactor.

The challenge was that the reactors are sterilised at 121 °C before use, while available sensors for measuring dissolved hydrogen had limited temperature tolerance and measuring range. They were also not designed to be attached to a bioreactor and used over long periods of time, especially not in potentially explosive atmospheres (ATEX), which any industrial-scale use of hydrogen gas entails. As a result, the researchers lacked a robust solution for continuous measurement under realistic process conditions. Eystein Opsahl explains why such measurements are crucial:

– By following how the hydrogen concentration changes under different operating conditions, we can better understand what the bacteria need in order to increase the reaction rate and use nutrients more efficiently. The sensor has, among other things, been used at pilot scale to produce acetic acid from hydrogen and carbon dioxide. This opens up opportunities for a more circular bioeconomy, where off-gases from other industries can be used to create something valuable. When hydrogen is one of the reactants, it is essential to know how much is available so that we avoid wasting hydrogen or causing the reaction to stop, says Opsahl.

A solution developed through collaboration

To solve the problem, the research group contacted Unisense. Together they developed a solution based on a retractable sensor housing, allowing the sensor to be withdrawn from the reactor during sterilisation and reinserted when the process starts, thereby avoiding the temperature issue altogether. At the same time, the sensor can be sterilised using chemical methods, reinserted once the reactor is ready, and kept separate from the surrounding atmosphere using inert gas. The hydrogen sensor is made of stainless steel, as is normally required in bioprocesses, and withstands the demanding process conditions the researchers were working with. None of this is technologically revolutionary in itself, but the application is all the more important: NORCE is the first to demonstrate that this can be done. Opsahl emphasises that the solution builds on both long-term investments and broad collaboration:

– We are standing on the shoulders of many people here. Projects such as PYROCO2, with major investments in research infrastructure, are only made possible through the efforts of technicians, engineers, researchers, support functions and public funding sources. Through SFI Industrial Biotechnology, we had both the equipment and the opportunity to finance the development work. Unisense took on the challenge without hesitation and has been a competent and curious partner. At the same time, the team in the Industrial Biotechnology group has been absolutely essential in designing, building and operating the gas fermentation facility in Risavika while this work has been underway, says Opsahl.

Tested at pilot scale

The new sensor was installed in a 150-litre pilot reactor at NORCE and connected to the process control system. After successful tests in water pressurised with hydrogen, it was used for continuous monitoring and control of an active gas fermentation process at 60 °C. The sensor delivered stable measurements for more than a week of operation, much to the enthusiasm of the research community.

For the research environment, this opened up new possibilities for following the process in real time and collecting data that are important for the further development of hydrogen-based biotechnology. According to Opsahl, better measurement data may also be highly significant for future process control and scale-up:

– When we are finally able to measure something new, we do not necessarily know what discoveries await us. The most important thing is to start measuring and logging, so that over time we can identify patterns and understand the process better. Today, many decisions in bioprocesses are still based on experience and gut feeling. Artificial intelligence has great potential for predictive process control, but this requires a solid data foundation that makes it possible to correct the process early. Such data are also important when we develop reactor designs and scale up from, for example, 150 to 1,000 litres. This work is now being followed up in SFI Industrial Biotechnology and is planned to result in a scientific publication, says Opsahl.

From NORCE need to standard product

Perhaps the most interesting aspect of the story is that the solution did not remain a customised setup for a single reactor. Following the development work, the sensor in question has been added as a standard product in Unisense’s portfolio and is now used in several process monitoring setups.

This is a good example of how research is often driven by concrete challenges in the laboratory and pilot hall. When researchers, engineers and technology providers collaborate to solve such problems, the result can be solutions that create value far beyond the original project.