The cleanliness of an H2 piping system is often based on end use requirements since the purity of the system may impact the end use application. Cleanliness required for end use purity is usually much greater than that required to prevent condensation or oxygen content sufficient to create a safety risk. While not required for the same safety reasons as oxygen safety, specifying a system to be “oxygen clean” for systems where a moderate level of cleanliness is required is a typical practice.

 Documents such as CGA 4.1 provide instruction and inspection guidelines for oxygen cleaning. Systems that require very high purity, such as for electronics, will have custom cleaning specifications for that application. One safety-related cleanliness consideration is filtering of particulates since they can cause regulators and safety valves to not operate correctly. This is a frequent concern on recently assembled lines that can have residue from machining, thread cutting, and welding.

Proper purging procedures should be used when opening a system for maintenance to ensure that a flammable mixture of hydrogen can’t be formed when the system is exposed to air. Connections should be covered when open to avoid entry of particles and moisture. Moisture is a particular concern when opening LH2 lines since water vapor can condense on cold surfaces and result in plugging of lines over time.
 

Yes, these would be ignitable mixtures.  In this case, it does not appear complicated geometry is involved, so 1200 psig pipe should be more than adequate to protect against internal deflagration. The most likely scenario is a "backfire," similar to a car, where ignition occurs too soon and shoots out the open end of the pipe. Consider using an inline deflagration flash arrestor on the supply line to protect upstream piping and equipment. Also, make sure pipe welds, fittings, and instruments have a comparable rating. The pipe needs to withstand an over-pressure resulting from an ignition at the mixing point of H2-O2. This is the issue that necessitates a high pipe pressure rating, preferably high enough to withstand a detonation.

The nominal H2-air detonation peak pressure is about 10-20 times the initial pressure, but there are also possible reflected pressures, a comfortable margin is needed to withstand shock wave pressures above the nominal detonation peak pressure. It’s possible for reflected pressures to get a little higher, but these generally remain within 20X. If a pipe is at atmospheric pressure, 300 psig will result in 20X the pressure, so that is the basis for the 1200 psig pipe having sufficient margin. 

The HSP discourages the use of this type of piping for hydrogen despite the low likelihood of ignition within the pipe itself. The polyethylene film can indeed develop a charge and, depending on the film thickness and substrate wall material, will eventually produce an electrostatic discharge. Since the nominal hydrogen concentration is so high (and assuming the purge-in and purge-out procedures preclude flammable mixture formation), these electrostatic discharges should not produce an ignition in the pipe itself. However, besides this type of pipe not being a good practice for flammable gas flow, there is concern about an electrostatic discharge possibly occurring at the pipe exit and at any junctions between pipe sections. 

One particularly energetic type of electrostatic discharge is a propagating brush discharge, which occurs when a thin film of an insulating material with a high charge retention capacity is attached to a conductive material. The propagating brush discharge occurs over a large section of the insulating film and might extend beyond the pipe exit if the polyethylene film extends to the exit. If the hydrogen is flowing into an air atmosphere, an electrostatic discharge at the pipe exit could ignite the flammable mixture formed near the exit. If the pipe wall itself is combustible, a local ignition from leaking hydrogen could lead to pipe degradation and eventual containment failure of the pipe.

As a good practice, hydrogen piping should be conductive and bonded/grounded to avoid electrostatic charge generation from a variety of mechanisms including atmospheric discharges. The hydrogen gas itself should not develop an electrostatic charge during pipe flow, but the gas flow can entrain pipe scale or other contaminant particulates that develop electrostatic charges that can ignite the hydrogen vented into air. Autoignition during emergency venting of hydrogen at pressures greater than 200 bar can occur due to shock wave compression heating near the vent exit, but it’s doubtful this would occur during controlled filling of balloons because of the lower velocities expected in that application.

The HSP recommends against the use of glycols for pressure tests due to the difficulty of adequately removing all glycol that might be left in a system after a hydrotest. The HSP recommends a pneumatic test at 110% of the system maximum allowable working pressure (MAWP), which is acceptable by code. Due to an increased danger with pneumatics vs hydrotesting, establish a pressure test zone for personnel in the area. 

Several concerns with glycols are: 1) Freezing in a liquid hydrogen   system that could lead to safety issues such as blockage of lines and instrumentation. There could be dead legs or low spots that cannot be emptied or cleaned easily leaving enough to freeze important components such as small gauge lines, vent stacks, and relief devices. 2) Freezing at cold ambient temperatures or within dispenser piping that chills the dispensed gas can also lead to blockage of lines and instrumentation. 3) Off-spec hydrogen if not well cleaned from the lines, vessels, instrumentation, and dead legs of a system. Glycol can interact with materials such as aluminum and lead to pitting and corrosion.

Another concern for glycol solutions is that they are flammable. A leak generated during pressure testing could generate a spray that is readily ignitable by nearby ignition sources. For automatic sprinkler antifreeze solutions, the requirement is to use UL-listed solutions that have undergone flammability testing. with the pressures used in applicable systems. These pressures are at least an order-of-magnitude lower than what is typically used for hydrogen piping. This safety concern can be mitigated by preventing personnel and ignition sources from being near the piping when tested.

As a general rule, pneumatic testing is preferred over hydrostatic, especially for systems with complex piping geometry, as long as proper precautions are taken for the Pressure-Volume (PV) energy. The HSP recommends using a pneumatic test at 1.1 times the design pressure using a clean, dry, inert gas such as nitrogen. This is a best practice for field testing where cleaning afterward is even more difficult. Helpful resources that describe precautions to take are: 1) UK HSE G4 Safety in pressure testing, says to (a) perform a hazard analysis considering stored energy, blast effects, and missile formation; (b) develop a written procedure; and (c) examine system prior to test. 2) ASME PCC-2, Repair of Pressure Equipment and Piping, says to limit stored energy in any one test loop to 271,000,000 J (200,000,000 ft-lb) (0.07 tons TNT). If the stored energy requirement can’t be met, barriers must be erected or separation distances in excess of 60 m (200 ft) must be used. The HSE document offers broad guidance, while the ASME document provides more specific information about precautions as a function of the stored energy.

The Panel recommends performing a pressure test at 110% of design pressure. This requirement should be applied to all systems regardless of construction type since the intent is to ensure pressure integrity and proper installation. All fitting types have modes of failure during installation. For example, there are numerous examples where compression fittings have had ferrules installed incorrectly, tubing improperly inserted, and have been inadequately tightened. In addition, leak checking and pressure testing should always be done in accordance with the locally adopted piping code.  Examples include ASME B31.3 and the Pressure Equipment Directive. 

 
System pressure relief devices will usually need to be removed for the test and temporarily replaced with higher setpoint devices to protect the system during the pressure test. Also, consider a proportional acting relief device.  Piping systems do not require a pop-acting ASME relief valve that are used for pressure vessels since piping system relief devices are more likely to chatter. Chatter can lead to lower than intended flow rate and damage or failure of the valve to operate correctly.   
For high-pressure systems, pneumatic testing is almost exclusively done given the challenge of removing water from a hydro test from the system after the test. Although one might question the wisdom of pneumatically testing at such high pressures, precautions can be taken to ensure a safe test, such as requiring an exclusion zone during the testing.   
 

The HSP is not familiar with this particular fitting or its potential application in H2 service but can offer engineering judgment. Many similar compression-type swaged fittings are used in H2 service. These fittings must be used within the manufacturer recommendations for pressure, temperature, and fluid service. Deformation is also introduced into stainless steel tubing in other ways, such as creating bends, but these locations are not reported to be more vulnerable to failure in H2 service provided that they are within an acceptable range.   

This is probably 300-series stainless steels, particularly 316, that are resistant to hydrogen embrittlement under static stress, although cyclic stresses add the potential for hydrogen-assisted fatigue cracking. One area of caution when applying the fitting to 304 piping is the potential to form strain-induced martensite (phase with magnetic signature) in the stainless steel. The martensite phase is more susceptible to hydrogen embrittlement than the surrounding matrix, and a widely recognized best practice is to limit the formation of these phases with magnetic signature for stainless steels in hydrogen gas service.

As with all compression-style fittings, there is a significant risk of incorrect or incomplete installation. Examples include failure to install the ferrules, installation of the ferrules in the wrong order or backwards, incomplete insertion, and incomplete tightening to reach the proper swaging of the tubing to ensure the necessary grip at high pressure. Fittings should always be inspected, either visually or with appropriate tools, and pressure tested prior to putting into service.
 

The Panel considers two approaches to be acceptable. 

1. Stainless steel (corrosion resistant) lines embedded in concrete. This keeps the piping out of direct contact with the earth and provides a degree of physical protection from activities such as digging. 
2. Lines enclosed within a PVC sleeve. The PVC sleeve would usually be directly buried but could also be embedded in concrete for additional protection.
   Cathodic protection is also an available and acceptable means of corrosion protection for underground lines, particularly for carbon steel. Each installation should be evaluated for local hazards and soil conditions to develop a thorough corrosion protection system. 
    

Regarding the concept of introducing hydrogen gas into natural gas pipelines, this is indeed a hot topic and there are recent quantitative treatments of fatigue crack growth driven by pressure cycling and potentially accelerated by hydrogen.  Some analysis has shown that it can be acceptable to operate natural gas pipelines with a hydrogen blend.  However, this is highly dependent upon the pressure and wall stress.  Whereas low pressure distribution lines with low wall stress are more amenable, the higher pressures and wall stresses of major transportation pipelines may not be.  Deep cyclic stresses that might occur from using pipelines as storage  may also create additional issues since hydrogen can accelerate fatigue crack growth, especially for systems where both hydrogen and deep cycles were not anticipated in the original design.  Individual pipelines likely need to be evaluated based on their design and there is likely to be no single answer for this question of pipeline storage. 

    
For industrial propane tanks, the Panel needs more information about structural materials and their properties (including welds). All propane tanks are not built from the same materials or with the same construction techniques, so these tanks likely need to be evaluated on a case-by-case basis. Propane tanks are often built with techniques that leave potential internal features that are susceptible to the initiation of crack growth. In addition, and depending on the application, it’s likely that the cyclic pressure service will be significantly different for a tank in hydrogen service, particularly one that might be cycled deeply on a daily (or more frequent) basis. From the perspective of the low-pressure storage options, while it is tempting to repurpose old LPG vessels for hydrogen service, the Panel cautions against it due to the potential for hydrogen embrittlement in the steel/weld material. Also, a challenge with newer propane tank designs is that they are moving to thinner walls and higher strength steel, which is notably less resistant to embrittlement than older vessels, so counterintuitively, newer tanks might be less amenable to hydrogen than older tanks.   

Another concern for vessels in hydrogen service is the amount of non-circularity, or “peaking” of the longitudinal welds.  This is not as much of a concern for propane so is likely not to be part of a standard design specification. This manufacturing issue can have a severe effect on cyclic life and is well understood when designing and building high cyclic hydrogen vessels, but not propane. Besides the issue described above for the tank itself, there are similar potential embrittlement issues for PRVs, instrumentation, and other accessories on these tanks. Many incidents have been due to component failures in these accessories. 
In conclusion, propane vessels were not designed for H2 cyclic service. Almost by definition, the reason people want to use them is for storage. and that is likely to be over a wide pressure range. Fracture mechanics should be applied to all cyclic H2 service vessels, especially if they start from another service. Propane tank designs vary and unless a tank is specifically intended for H2, there is no guarantee of what might be used terms of materials and construction techniques.
 

Regarding cylinders, it is not necessary to capture the fuse-backed devices which are on the cylinder itself. However, all other relief devices and vent valves must exhaust from a vent system designed in accordance with CGA G-5.5. Also, note that NFPA 2-7.1.17 requires compliance with CGA G-5.5 regardless of storage quantity when the vent system is servicing pressure relief devices. Cylinders in storage that are capped and within an approved cylinder storage area do not require a vent system.