The Netherlands has set ambitious climate targets, aiming to become climate neutral by 2050, which implies a 100% reduction of greenhouse gas emissions compared to 1990 levels. Solar and wind power are key to replacing fossil fuels in electricity generation, but their intermittent output creates a need for both short-term and seasonal energy storage. At the same time, sectors such as heavy industry, aviation and shipping remain challenging to electrify directly. Renewable hydrogen, also referred to as RFNBO-compliant hydrogen, is emerging as a clean and flexible energy carrier that helps fill the gaps where green electricity alone is not enough. It can be stored, transported, and used in applications where direct electrification is not feasible. Unlike grey hydrogen produced from fossil fuels, renewable hydrogen is made by using electricity from solar or wind to split water into hydrogen and oxygen, significantly reducing carbon emissions.
Producing renewable hydrogen remains expensive due to the high capital costs of electrolysers and the limited availability of cheap renewable electricity. Producing renewable hydrogen in a way that is both economically sustainable and scalable is important to decarbonize global industries and provide renewable seasonal storage. To evaluate and compare the economic performance of hydrogen production methods, the Levelised Cost of Hydrogen (LCOH) is used. LCOH represents the average cost of producing hydrogen over the lifetime of a facility, accounting for capital, operational, and energy costs. As the renewable hydrogen market is not yet completely developed, metrics such as Net Present Value (NPV) are less suitable for determining the competitiveness of the technologies. Therefore, it is important for renewable energy companies, like Eneco, to create renewable hydrogen with the lowest LCOH to be competitive in the hydrogen market.
One approach to lowering costs is to pair electrolysers with batteries, where revenues from storing electricity and selling it to the grid, known as arbitrage revenue, can help offset operational expenses. This can be achieved through two types of systems: the Battolyser, an innovative technology that combines battery storage and hydrogen production in a single device, and a system where a battery and electrolyser are co-located but operate as separate units. These systems have distinct advantages and disadvantages, which have not yet been evaluated and compared. Therefore, the main research question is: How does the Battolyser compare to a system with a separate battery and electrolyser in achieving the lowest levelised cost of hydrogen (LCOH) compliant to RFNBO (EU) standards?

To answer this question, this research analyses four different co-located system configurations. Each configuration combines either an alkaline or PEM electrolyser with an LFP lithium-ion battery or a vanadium redox flow battery, selected for their technological maturity and future potential. Together with the Battolyser, the five systems considered are:
• Battolyser
• Alkaline electrolyser + Lithium-ion battery
• PEM electrolyser + Lithium-ion battery
• Alkaline electrolyser + Redox flow battery
• PEM electrolyser + Redox flow battery

Financial and technological characteristics of the systems have been identified, taking the Dutch market into account. The five systems are simulated by using a novel techno-economic simulation model that has been created for this research. The model simulates the systems in a framework based on Dutch hourly electricity prices between 2030 and 2050.
The simulation model focuses on three main operational strategies: battery charging, battery discharging, and hydrogen production. The systems are simulated on an hourly basis for the period 2030–2050, using a combination of mixed onshore wind and solar power profiles contracted through a Power Purchase Agreement (PPA), together with grid electricity as power input in the base case simulations. The model uses a controller with a three-hour receding horizon to anticipate electricity price fluctuations. The simulation framework ensures an objective comparison by keeping operational strategies and regulatory compliance consistent while varying system-dependent parameters such as electrolyser efficiency, minimum stable load, and system costs.

The base case simulations showed that the Battolyser performs best with an LCOH of 9.96 €/kgH2. Among the four battery-electrolyser combinations, the system with a lithium-ion battery and a PEM electrolyser came closest to matching the performance of the Battolyser, followed by the alkaline and Li-ion system. Although redox flow battery systems have longer lifespans and flexible sizing, their higher costs result in poor LCOH performance, because the technical benefits do not compensate for this. The LCOH values lie close to each other, with a spread of 1.40 €/kgH2 when including redox flow systems, and just 0.40 €/kgH2 when excluding them.
Several analyses, including sensitivity and scenario analyses, were performed to understand the robustness of these findings. A local sensitivity analysis was performed by varying electricity prices by plus and minus 20% compared to the base case, to examine how changes in electricity costs impact LCOH performance. This analysis showed that the PEM + Li-ion system, followed by the ALK + Li-ion system, both perform better than the Battolyser when the electricity prices increase, as the systems can take full advantage of the flexibility by dynamically shifting operation. They can produce hydrogen at nominal capacity during one hour, switch to battery charging the next hour when electricity prices are low, and then sell electricity back to the grid during high-price periods. In contrast, the Battolyser performed significantly worse under these increased conditions, as it cannot separate battery operation from hydrogen production, which limits the system’s ability to fully exploit price fluctuation.
In the global sensitivity analysis, five key parameters (CAPEX, OPEX, electrolyser efficiency, battery capacity, minimum stable load) were varied to investigate the absolute effect on the LCOH, as well as their interaction with other parameters. This analysis shows that varying the electrolyser efficiency has the highest absolute effect on the LCOH. In addition, electrolyser efficiency is present in every statistically relevant interaction term. This shows that electrolyser efficiency is not only a significant factor on its own but also influences the impact of other parameters, making it the most important determinant of LCOH.
This is confirmed in the scenario analysis, where the systems are simulated with a more steady offshore wind power profile, in contrast to the mixed wind on land and solar power profile in the base case. The Battolyser and alkaline systems, both of which have better electrolyser efficiencies than the PEM system, perform better under these steady conditions. In contrast, the PEM-based systems show a decline in performance, requiring significantly lower offshore wind PPA prices to achieve LCOH levels similar to those in the base case. This indicates that under stable power conditions, electrolyser efficiency becomes a more dominant factor in contrast to operational flexibility.
Finally, the Battolyser and co-located systems were compared to standalone alkaline and PEM electrolysers. The standalone systems outperformed the hybrid configurations due to their lower cost structure and ability to channel all available power directly into hydrogen production. The alkaline electrolyser achieved a lower LCOH than the PEM system in the base case, the reduced electricity price scenario, and under an offshore wind profile. This advantage can be attributed to its higher electrolyser efficiency. These findings indicate that adding a battery to an electrolyser does not necessarily lead to lower LCOH. However, the standalone systems lacked operational flexibility, making them perform worse under scenarios with higher and more volatile electricity prices, where the PEM + Li-ion system performed better.

In conclusion, this research explored whether combining battery storage with electrolysis could lower production costs through electricity market arbitrage. The Battolyser was compared with four battery-electrolyser systems for renewable hydrogen production. The results show a small spread in LCOH and that there is no single system that performs best across all situations. Among the hybrid systems, the Battolyser achieves the lowest LCOH in the base case, the reduced electricity price scenario, and under an offshore wind profile. Its strong performance can be attributed to its high electrolyser efficiency, which proves to be a significant factor in determining the LCOH. Nonetheless, standalone electrolysers outperform the Battolyser in these scenarios due to their lower cost structures and direct power use. However, when electricity prices rise and become more volatile, systems with separate batteries and electrolysers perform better, with the PEM + Li-ion configuration achieving the lowest LCOH.
It is recommended that Eneco prioritizes hydrogen production systems with high electrolyser efficiency as a strong option for RFNBO-compliant hydrogen. Standalone electrolysers should be considered when electricity prices are expected to follow patterns similar to the base case. However, Eneco should remain open to using separate battery and electrolyser setups in case when electricity prices are expected to rise. Since the LCOH differences between systems are relatively small, it is further recommended that the system selection should be guided by additional factors such as environmental impact and the availability of technologies within the European market.

While the system choice is important to provide a competitive edge in the hydrogen market, the results also highlight a broader challenge for renewable hydrogen. Eneco expects that the willingness-to-pay for renewable hydrogen will be approximately €9/kgH2. No system is able to produce renewable hydrogen below this threshold and thus a cost gap remains. Therefore, future research should investigate the renewable hydrogen market in the Netherlands and potential subsidy schemes. Moreover, it is recommended to expand the simulation model to allow a more detailed comparison between the systems as this simulation model has some limitations. Possible additions include incorporating optimization horizons that reflect the day-ahead electricity market, enabling dynamic power allocation, exploring participation in additional energy markets, and performing environmental life cycle assessments.