Water for Hydrogen

Hydrogen production water consumption

Water is one of the key inputs for a green hydrogen plant. A holistic approach to selection of water technologies can make a significant difference in the overall water balance and viability of a project. While water and water treatment typically does not have a very large cost associated with it, finding sustainable water sources and reducing the water consumption for hydrogen production will assist to lead to renewable hydrogen production obtaining a social licence to operate. Given that renewable energy is often available in more arid parts of the world, in particular solar energy, this places even more emphasis on the reduction of water consumption and the sustainable management of water as a core element of a sustainable hydrogen project.

To produce hydrogen from steam methane reforming (grey hydrogen), which is currently the most prevalent pathway to hydrogen production the stoichiometric water consumption is 4.5 L H₂O per kg of H₂ produced. However, this water is demineralised water (boiler feed water quality), so that a reject from producing the demineralised water also has to be taken into account, with the rejection rate being dependent on the quality of the raw water. In addition, steam losses and losses from evaporative cooling also must be included, so that the raw water use for steam methane reforming amounts to 15-40 L H₂O per kg of H₂ produced.

For other reforming technologies, such as partial oxidation or autothermal reforming, the water demand may appear lower at first, but due to the steam demand for water-gas shift following the reformer to optimise hydrogen production from natural gas, the overall water demand will be very similar, and possibly slightly higher, than for steam reforming.

To produce “blue” hydrogen, CO₂ emissions from reforming must be captured. Carbon capture and compression steam and cooling requirements will further increase the overall water demand for hydrogen production from natural gas reforming to approximately 18-44 L H₂O per kg H₂.

For hydrogen production from coal and biomass gasification (brown and black), the water consumption may be around 70 L per kg of H₂ produced in the case of coal feed, and slightly lower at 60 L per kg of H₂ for biomass feed to gasification, mainly due to the higher average moisture content of the biomass feedstock.

For reforming of biogas (which could also be considered green hydrogen), the stoichiometric water consumption would be very similar to that of hydrogen produced from reforming natural gas (4.5 L H₂O/kg H₂), but the overall demand would be slightly higher at approximately 20-45 L H₂O per kg of H₂ produced due to increased heating and cooling requirements to remove CO₂ from the biogas prior to reforming.

Therefore, the typical hydrogen production processes that are prevalent today all require a significant amount of water to produce a kilogram of hydrogen, and a much higher amount then the often quoted stoichiometric needs for electrolysis of green hydrogen from water alone.

Green hydrogen is produced by taking renewable power, high purity water and converting to hydrogen and oxygen gas via electrolysis. The water requirement for green hydrogen is stoichiometrically 9 L of H₂O per kg of H₂ produced. This is higher than for natural gas reforming, where some hydrogen is already present in the feedstock (mainly CH₄). In addition, commonly overlooked water supply and disposal factors include:

• Significant cooling load for electrolysers – which can require additional 30 to 40 kg of water per kg of hydrogen for makeup in evaporative cooled systems. Over time, the stack efficiency of the electrolyser decreases, and most of the efficiency losses report to additional heating of the stack; with the result that the cooling load increases significantly over the lifetime of the stack (typically 8 to 10 years of operational time). The cooling demand for the electrolyser can typically increase by 40 to 70% from beginning of life to end of stack life.
• Other cooling loads – such as the multi-stage compressors with intercooling to compress the produced hydrogen to a suitable pressure for storage or use.
• Raw water feed requiring treatment to meet high purity electrolyser requirements – with around 20-40% of the water sent to waste during the treatment process, depending on the quality of the imported raw water.
• Water disposal – due to the increased concentration of feedwater impurities into the waste streams this water can often not be discharged to the environment and requires connection to a waste treatment facility or onsite treatment or disposal.

These additional loads can lead to as much as 60 to 95 kg of H₂O required per kg of green H₂ produced.

Of this demand, approximately 60 to 70% is attributed to cooling water makeup, with the calculated water consumption for green hydrogen assuming full evaporative cooling. A typical breakdown for green hydrogen water utilisation is shown in Figure 3.

Figure 3 Breakdown for green hydrogen production water demand, ~60 L H₂O per kg H₂

The above shows a typical green hydrogen production water demand for full evaporative cooling and assuming good quality raw water import.

All of the numbers quoted for raw water demand above assume that the raw water import to the site is fresh water of relatively good quality. If the water is brackish, seawater or industrial wastewater, the volume of raw water will increase dramatically and so will the wastewater/brine produced from water treatment at site.

A summary of water demands for various hydrogen production pathways is shown in Table 2.

Table 2 Hydrogen Production - Water Demand Accounting for Evaporative Cooling

Replacing blue hydrogen with green hydrogen could lead to an increase in water consumption of between approximately 35 to 100% per kilogram of produced hydrogen. Given the projected dramatic growth in demand of 70 exajoule hydrogen per annum by 2050, the water consumption could be in the order of 35,000 to 55,000 GL/annum to produce all of the hydrogen from water electrolysis. While this is a relatively small volume compared to other users such as 2,800,000 GL/annum for global agriculture, 800,000 GL/annum for industrial users and 470,000 GL/annum for municipal users, most of this would signify a new additional demand and therefore increase pressure in water security as mentioned before in many parts of the world.

Therefore, reducing the water demand for green or renewable hydrogen would be beneficial. The impact that brine production and treatment or disposal would have on the local ecosystem would have to be considered as well.

To put water consumption into perspective, when green hydrogen is produced from a 10 MW electrolyser unit, 4 tpd of hydrogen is roughly produced, requiring approximately 0.24+ ML/day (240+ m3/day) of raw water. For a 1 GW green hydrogen installation, this would increase to 24 + ML/day (24,000+ m3/day), producing 400 tpd of green hydrogen.

Hydrogen carriers

In addition to hydrogen production, hydrogen is typically converted to a carrier of choice, such as ammonia, liquefied hydrogen or liquid organic hydrogen carrier. Each of these carriers require a conversion step with associated steam, boiler feed water and/or cooling water demand.

While these individual water demands should not be the main factor in selecting the carrier, it is important to consider as part of the overall water balance associated with the plant. For example, for ammonia production, including the ammonia synthesis unit and air separation unit to produce nitrogen as feedstock to the ammonia synthesis unit, the cooling load more than doubles compared to the production of gaseous compressed hydrogen alone.

However, as a large portion of this cooling duty is met through generating steam from recovery of process heat, cooling water losses from the combined plant is not proportional to the increase in cooling duty. This generated steam is in turn utilised to drive the main compressors in the ammonia synthesis plant, leading to an almost energy -neutral process unit.

As explained above, high pressure boiler feed water rather than cooling water is utilised for the ammonia synthesis unit to meet the cooling duty / recover process heat. To maintain the quality of the boiler feed water, a blowdown is required, which has to be made up from raw water import. Newer plants include blowdown water treatment, where the blowdown is processed through reverse osmosis (RO) and largely recycled to the plant, reducing the volume of make-up water (and thus raw water import) required.

Adopting ammonia as a carrier introduces an additional layer of interaction and complexity that must also be understood to fully assess the water needs of an integrated hydrogen project from a water perspective.

Reducing water demand to enable green hydrogen

To reduce the water demand for renewable hydrogen, the focus should be on reducing the cooling water make-up demand, since this parameter dominates the overall water demand. Cooling water make-up is required due to evaporative cooling losses from the cooling tower (accounting for 75% of losses from the cooling water system), and blowdown (accounting for 15% of losses), with additional small other losses.

Air cooling and alternatives

To reduce cooling water make-up demand, air cooling can be considered however its applicability is subject to location and local climactic conditions. Air cooling is typically more expensive than evaporative cooling and has a larger footprint and power demand. However, while the capital cost and power demand of the cooling system would increase, these systems have a small contribution to the overall plant cost and power demand of the project, with the electrolyser dominating both capital cost and power demand. Air cooling is limited to 40 or 50°C only, depending on the site conditions, and therefore not all wet cooling can be substituted for air cooling, but up to 50 to 60% of the overall cooling duty for the renewable hydrogen plant could be met through air cooling. This could reduce the overall water demand by 30 to 40% overall in some locations.

In addition to a smaller water demand, less cooling water blowdown would be produced, resulting in a lower volume wastewater/brine production and reduced brine management challenges and costs.

Alternatively, some electrolyser and hydrogen compressor vendors offer closed loop cooling, using chiller systems with very little water losses. In reality however the practical and economical application of these for large scale plants (with a large number of electrolyser units employed) may be a struggle. In addition, these result in additional power consumption and carbon footprint, although the power consumed is small when compared to the electrolyser power.

Technology development

The cooling duty of the electrolysers is very high. If compressed gaseous hydrogen is produced (rather than a hydrogen carrier), the cooling duty associated with the electrolysers can account for 80 to 90% of the total cooling duty associated with the plant. As electrolyser technology develops further and efficiencies increase, the cooling requirement for electrolysers will decrease, and therefore the overall cooling water demand will decrease significantly as well. If the stack efficiency increases by 15%, the cooling demand will decrease proportionally.

Another option is to reduce the stack lifetime of the electrolyser so that the cooling demand for the stack does not reach its maximum demand (end of life demand). However, compared to water costs, the stack replacement cost and time off-line for the stack replacement are currently too high to justify earlier stack replacement. The most likely future here would be that technology developers find means to extend the life of the stack without significant degradation and decrease in efficiency from beginning of life to end of life.

Sustainable water sources

Regardless of how much the water consumption of renewable hydrogen plants could be reduced by, finding sustainable water sources for hydrogen production is important. There are three most common options; utilising fresh water, utilising wastewater from industrial sources and utilising seawater desalination. Utilising fresh water likely has the lowest treatment cost but is typically not preferred as this diverts water from other economic and social users to hydrogen production.

Hydrogen hubs will typically be located with other industrial activities or settlements and therefore, there may be significant volumes of wastewater produced close to a hydrogen production facility. While it will be more expensive to treat wastewater from domestic or industrial sources to the desired quality (in particular demineralised water for electrolysis), it would not contribute a large difference in capital for the plant. In addition, it is likely that these sources of water would be closer to the plant than other fresh water sources, reducing pipeline and transmission costs. The water would also likely have a relatively low associated supply cost; or the project could be paid to take the water from another industrial site and treat it.

For large installations, it is likely that seawater desalination will be the only truly viable sustainable source of water. Using seawater would increase the raw water intake by 2.5 to 5 times, depending on the recovery ratio compared to fresh water, but seawater presents a large resource and the capital cost and power consumption of desalination is small compared to those for the electrolyser. Seawater desalination however introduces an additional suite of environmental and approvals issues including social licence concerns and planning and approval timeframes associated with ecological assessment of potential sites.

In Table 3, fresh raw water import of relatively good quality (best case) is compared against seawater import and desalination (worst case). Industrial wastewater use is expected to be somewhere between the best and worst cases, depending on the quality of the wastewater.

Table 3: Hydrogen Production - Raw Water Demand depending on Raw Water Quality and Assuming Evaporative Cooling

With seawater or brackish water as raw water import, the wastewater/brine stream also becomes more pronounced and need to be carefully from a management and disposal perspective as part of the hydrogen production project planning and approvals requirements and program.

To put the energy consumption for desalination into perspective, the energy demand for the desalination plant would be less than 0.5 kWh per kg of H₂ produced (for a reverse osmosis plant), compared to around 50 kWh per kg of H₂ for the electrolyser. Modern desalination plants are reverse osmosis based, although other membrane technologies (forward osmosis and membrane distillation, for example) are being commercialised. At very large scale, thermal distillation units are still employed, particularly in the Middle East. The choice of desalination technology will have a fundamental impact on overall energy requirements and lifecycle cost for the production of H₂ and therefore any desalination method must undergo a focused technology selection study.

One of the challenges with seawater desalination is that the facilities would typically have to be built close to the coast in order to use seawater. However, since it is likely that hydrogen products would be exported when produced at large scale, this would be convenient.

The other challenge with seawater desalination is the disposal of the brine, which could have a local impact on the marine ecosystem, and would have to be carefully managed, in particular for large scale (GW) installations. The brine could be further treated to reduce its environmental impact. Management of brine generated by desalination plants is both a techno-economic and environmental challenge. Thermal brine treatment processes (e.g evaporator crystallisers) are energy intensive, have very high associated CAPEX compared to RO and present numerous operational challenges (scaling, water chemistry).

Optimise water requirements needs an integrated approach

Opportunities to optimise a project’s overall site water requirements, and therefore effectively manage and reduce risks associated with sustainable water security and social and environmental concerns can be identified by an early, integrated approach to water supply/disposal, power demand and cooling technologies, project location and consideration of carrier technology.

The right choice of technologies (water treatment, cooling systems, water disposal), including assessment of hybrid solutions, will often vary on a project-by-project basis but all will be required to be addressed ultimately for projects to be successful.

We can therefore expect that as the green hydrogen industry comes to fruition as part of our energy transition, water for hydrogen is and will become an increasingly important part of the water industry across the world.