Webcast report: Hydrogen blending at compression stations

Published on: 

A TMI Editor’s Series webcast discussed what to consider when introducing hydrogen to a compressor station.

In our November issue, we focus on hydrogen compression and how it may represent the foundation of the hydrogen economy. Wanting to explore this market segment further, Turbomachinery International invited Sarah Simons to break down the components and considerations needed for introducing hydrogen to existing compression stations. Simons is a Senior Research Scientist at the Southwest Research Institute and has worked on over 130 compressor installations.

So why should we blend hydrogen? For one, hydrogen is a form of thermochemical energy storage. It can be created by using excess energy from other sources, one of the most popular being ‘green’ hydrogen – or hydrogen created using excess power from renewables. The same result can be achieved via hydrocarbon reformation or coal gasification, which makes ‘blue’ or ‘gray’ hydrogen.

Introducing hydrogen to gas pipelines allows us to make use of the vast existing pipeline network. Pipelines are also a more efficient means of transporting energy compared to power lines. Finally, pipelines will allow for increasingly larger amounts of hydrogen to be introduced to the system over time, keeping pace with the gradual nature of the hydrogen economy.

Many utilities are already running their own hydrogen compression projects. Canadian utility Enbridge is looking to blend up to 2% hydrogen in their natural gas distribution system located in Markham, Ontario. Enbridge wants to evaluate how consumer appliances and the distribution system handles a hydrogen blend. Another project by Hawaii Gas is underway. The company is operating a small steel pipeline with refinery produced syngas containing around 12% hydrogen. The utility even constructed this pipeline with hydrogen blending in mind.


The first thing to consider before introducing hydrogen to a natural gas blend are the differences between the gases. Hydrogen is a much lighter gas, which means that overall mixture density will decrease following its introduction to a natural gas blend. This lower density will result in a higher sound speed /gas flow velocity along with changes to pulsation control design. Hydrogen’s smaller molecules may also raise concerns regarding pipeline embrittlement and leakage.

In addition, blending hydrogen into natural gas will result in changes in viscosity and a lowering of the Joule-Thomson coefficient. Thus, hydrogen blends experience no temperature drop at pressure loss areas (e.g., flow control valves) and show a decreased temperature drop over long distances. There is also an increase in specific heat and compressibility. An important thing to note is that hydrogen blending results in a higher energy density by mass, but a decreased volumetric calorific value. This means that hydrogen blends will require more volume/pressure to transport energy when compared to conventional natural gas.

When calculating the thermophysical properties for a hydrogen blend, note that prior testing is still very limited. Most thermophysical property tests have been performed on binary mixtures, such as hydrogen with methane. Unfortunately, there is little to no testing validation for calculated hydrogen-natural gas mixtures.

Pure hydrogen requires more power to transport the same amount of energy as natural gas – 3 times as much, to be exact. Hence, available power, compressor speed, delivery pressure, and flow velocities must be considered when introducing hydrogen to the pipeline. There could also be an increase in noise, a lack of cooling, and erosional/structural limits.

Simons offered an example of pipeline limitation calculations regarding hydrogen blending. A 2008 study from the Journal of Hydrogen Energy found that if the available power in a pipeline is fixed at a required 5,000 MW and a delivery pressure of 60 bar, then the maximum hydrogen fraction that can be injected for operation is 6.6% on a mass basis. This calculation ensured that the pipeline (and compression station as a whole) did not operate outside of its constraints.


Southwest Research Institute investigated how hydrogen blending interacts with centrifugal compressors. Mass flow and heating value of transported gas decreases as hydrogen content increases. To maintain a compressor’s volumetric energy capacity when using hydrogen, compressor speed, head rise, and power must be increased. One system tested at Southwest Research required a power increase of 20% to facilitate a 10% hydrogen blend.

Luckily, many seals and materials are already compatible with hydrogen. Dry gas seals are typically designed to handle up to 20% hydrogen. O-rings are generally designed with hydrogen-resistant materials. Shaft seals made of PEEK or PTFE are also hydrogen compatible. These materials will usually operate at their recommended hydrogen percentage so long as temperatures stay below 392 degrees Fahrenheit.

Conversely, API 617 dictates that “Materials that have a yield strength in excess of 827 MPa (120 ksi) … are prohibited for use in hydrogen gas service where the partial pressure of hydrogen exceeds 689 kPa (100 psi gauge).” This is most likely to have an impact on the impeller and shaft, since casing and stationary components are typically made of low yield strength carbon steel.

Like centrifugal compressors, many gas turbines can operate with a certain percentage of hydrogen mixed with natural gas. This is usually 5% or less, though some turbines can handle more than 10%. Newer turbines will almost always be able to handle more hydrogen than their older predecessors. Regardless, it’s important to evaluate the entire turbine package for use with hydrogen before proceeding.

A drawback with conventional combustors is that hydrogen’s faster burn and high combustion temperatures result in much higher NOx emissions. There is also increased flashback risk amongst legacy DLE combustors. Higher hydrogen blends (exceeding 4%) also require additional safety measures, such as requiring additional fire and gas detection devices.


Southwest Institute performed a study on hydrogen blending in reciprocating compressors. Blending increasing amounts of hydrogen into a system (while maintaining a fixed speed and pressure ratio) results in minor changes to power and ACFM, but a significant decrease in energy flow. Since reciprocating compressors are rarely designed with speed margins, the best way to maintain a constant energy flow when blending hydrogen is to bring more compressors online.

Reciprocating compressor valves, seals, and materials can handle smaller amounts of hydrogen. Concerns may arise when exceeding 20%. Lubricated compressors can be used if temps are below 350 degrees Fahrenheit and if mole weight is greater than 12 (as per API 618). Non-lubricated machines may require more scrutiny before blending hydrogen.


Most manufacturers find that their gas engines can run on 5-10% hydrogen. Blending hydrogen in gas engines generally results in increased fuel volume consumption due to a decrease in volumetric energy density. Users may need to slow ignition timing to account for this. Flame speed will increase in conditions with less than 20% hydrogen, increasing under in-cylinder peak pressures.

Users should be mindful of increased water from combustion that can become entrained in engine oil. Another cause for concern is increased NOx emissions, which can be managed by adjusting the timing and air-fuel ratio.


One must also consider the effect hydrogen blending has on piping. Introducing hydrogen to a system will result in increased speed of sound and velocity, with more hydrogen resulting a bigger increase. These changes impact acoustic natural frequency and – in the case of centrifugal compressors – create vortex-shedding excitation.

Reciprocating compressors can benefit from the installation of pulsation bottles. These bottles will help reach pulsation attenuation. These bottles will change speed of sound and frequency of a pulsation control system, bringing it in line with the first or second compressor running order.

On the materials front, carbon steels have been used in pressure vessels and piping that have been exposed to hydrogen. These materials general hold up under the conventional temperature/pressure range found during day-to-day operations.

The main cause for concern here is contamination via the hydrogen blending process. If hydrogen is exposed to something like sulfur or phosphorus, it can turn into atomic hydrogen. Atomic hydrogen may enter the pipe crystalline structure, combine with carbons in the piping, and leave pockets of hydrocarbons. These hydrocarbons create gas pockets that weaken materials and make them more susceptible to fractures or cracks.

When considering existing piping, it’s important to establish defect criticality limits to stifle fatigue crack growth. Contamination issues are also present here and can be avoided by providing robust cathodic protection from outside sources. Methods of repair may have to be changed after introducing hydrogen, as piping is often baked. Welds should be pre and post heat treated.

Coolers can run into issues with higher flow velocities brought on by hydrogen blending. They may be under-sized for these increased velocities, and thus unable to cool to desired temperatures. It is also possible that the cooler’s discharge temperature might not be low enough to meet pipeline requirements. Scrubbers, on the other hand, do not have many causes for concern. Like many of the components and materials mentioned before, scrubbers are able to handle up to 20% hydrogen without running into issues.


The most common method for gas analysis is gas chromatography using a helium carrier. In hydrogen applications, however, helium cannot be used. Argon or nitrogen can be used in its place. Keep in mind that the use of these elements can result in longer process times and diminished accuracy compared to helium.

For gas metering, the type of meter used depends on if the hydrogen concentration fluctuates. Systems that do fluctuate in hydrogen percentage can make use of Orifice and Coriolis meters, though they may need resizing. Some concerns regarding gas meters are leakage through polymer membranes, decreased life titanium sensors (in ultrasonic meters), increased leakage through threaded connections, and the life of sealants and lubricants.

Due to hydrogen’s innate properties, its introduction to a compressor station raises some safety concerns. Hydrogen rapidly disperses in air, increases in temperature when it expands, requires a fraction of the ignition energy required by natural gas, and has a high burning velocity. Hydrogen does have the ability to spontaneously combust at pipeline conditions, and it has triple the volumetric leakage of methane.

Despite that, natural gas operations already bring on a slew of safety hazards, many of which are unchanged by introducing small amounts of hydrogen to the system. This is according to an IEA report that assessed risk factors brought on by operating with a 15% hydrogen concentration. Though one should keep in mind that risk assessments for larger blends of hydrogen have yet to be undertaken.


Simons concluded by offering initial steps for evaluation. One should begin by defining the boundaries of the pipeline that will be exposed to hydrogen, based on the hydrogen injection and extraction points. The pipeline’s pressure and energy transport capacity should be determined, and problem points (where contamination and poor cathodic protection may occur) identified.

In addition, a pipeline analysis should be performed to determine compressor and driver limitations. Similarly, gas component and flow measurement constraints should be considered. Users should also re-evaluate the station’s pulsation amplitudes. Gas turbines/engines and piping systems ought to be evaluated and modified (if necessary) for use with hydrogen. Last but not least, there must be a thorough safety review performed in order to account for potential risks brought on by hydrogen’s presence.