Compressed hydrogen - A fuel for the future? Dr Roy Irani explains in part 1 of a new series exploring gases in the 21st century.

Hydrogen is the lightest gas, has the smallest atom and is the most abundant element in the universe (almost 75 at .%). Almost every day we hear how hydrogen is going to be our saviour in terms of fuel for the future. The main reason for this optimism is its clean combustion by-product (namely water) and the very high resultant energy (due to the highest electron/neutron ratio). Yet isolating hydrogen, using it and storing it, all present monumental challenges in the practical world.

It is still cheaper to use oil even at well over $100 per barrel, than to separate hydrogen from its compounds such as water or natural gas. Using hydrogen needs precise procedures due to the high flammability risks it presents.

As for storing it, which can be as a compressed gas, a cryogenic liquid or in metal hydrides, none of the current solutions are yet economic, despite a huge effort being devoted in these areas. In fact, the current emphasis seems to be concentrating on storing hydrogen in its compressed form, since its low liquefaction temperature of 20K consumes a lot of energy for its formation and preventing evaporation is expensive - whilst the area of metal hydrides is very much in the research stages.

However, even this favoured approach is fraught with technical issues, on account of hydrogen’s potential embrittling properties, when stored in conventional steel cylinders. Even though hydrogen has been used in wrought iron vessels used to inflate war balloons in the 1880’s, this empirical ‘hydrogen embrittlement’ (HE) mechanism remained nascent, until the late 1960’s when a series of notable failures occurred in steel cylinders right across Europe. HE is absent in aluminium alloy (AA) cylinders, but there is the possibility of the hydrogen being manufactured via an electrolysis route which sometimes has mercury as the electrode. It is known that even parts per million levels of mercury are capable of cracking AA cylinders via the liquid metal embrittlement route, while the high cost of AA cylinders has also prevented their widespread use as large hydrogen vessels.

Cylinder Efficiency
The main difference between the vessels used in the 1880’s and the troublesome type of almost a century later, lies in the mechanical properties of the respective steel used for their construction. It is now well established that the presence of defects in gas cylinders (and all cylinders have a degree of defects albeit within the limits of the specification and modern non-destructive test methods) is closely linked to the mechanical tensile strength of the steel.

A modern 40 tonne trailer used for transporting hydrogen, equipped either with ‘jumbo tubes’ or with banks of gas cylinders, carries around just 400kg of hydrogen. That implies an overall efficiency of about 1%, and that is when the cylinders are full!! Ever since the mass transport of hydrogen became a necessity, when industries such as glass manufacturing and production of hydrogenated fats took off, the gas industry began the quest of improving this poor payload factor. The most obvious way was to minimise the weight of the cylinders for a given gas capacity.

Traditionally this is achieved by increasing the mechanical strength of the cylinder’s material, hence throughout the twentieth century the tensile strength of hydrogen vessels was gradually increased from the 350MPa of the wrought iron vessels to 1100MPa in the 1960’s. The latter used chromium-molybdenum alloy steels in a quenched and tempered condition. This approach, however, accompanied the unexpected high failure rates of hydrogen cylinders.

HE – Failures History
During the period 1968 -1978 more than 100 cylinders failed in European countries (see Fig. 1). Of these, 21 failures occurred in the UK. The incidents were shared between four different gas suppliers and were in conventional cylinders on-board trailers which are frequently filled (sometimes more than 10 times per week). Interestingly 19 of the 21 failures took the form of circumferential cracks in the ‘knuckle’ area of the cylinder (see Fig. 2), with 2 failures in the axial direction along the internal wall.

All 21 HE failures were in locations of high stress:
The ‘knuckle’ area is responsible when the base is poorly formed if there is a sudden change in the curvature of the profile from the parallel wall to the base, thus creating a localised stressed region. This is particularly true for cylinders made from a tube.
The axial direction was the culprit, when deep draw lines form a sharp stress raiser.

HE – The Technical Fix
International studies established a new specification to eliminate HE for cylinders. Hence for the prevalent manufacturing limitations (in particular the resultant defects) if the tensile strength of steels was limited to 950MPa, then the HE mechanism can be suppressed. Additionally, a more stringent ultrasonic inspection regime was introduced.

These requirements have now been incorporated into the main cylinder design specifications ISO 9809-1 and the one for metallic materials compatibility, namely ISO 11114-1.

These requirements, particularly limiting the actual maximum tensile strength to 950MPa, have been totally successful in eliminating the HE phenomenon in gas cylinders, but at a cost of transporting hydrogen in inefficient, heavy cylinders. Hence at a time when fuel costs are soaring, the gases industry has constantly been seeking ways of increasing the efficiency of hydrogen cylinders without risking HE.

To objectively make decisions on new materials, a new standard ISO 11114-4 was developed and has now been published. Here three potential test regimes (A, B and C) are specified to evaluate the HE risks. Such is the nature of HE that different parts of the world are comfortable with some but not all the tests in the standard. But in an international forum, as ISO standards have to be, an initial compromise position was taken by the committee to promote the publication of this standard, by accepting as candidate tests, all respectable test methods used around the world.

So whilst the Disc test (Method A) and a Fracture Mechanics Test (Method C) are primarily used in France and North America respectively, a further Fracture Mechanics Test (Method B) was successfully employed by the UK.

However, compromises do not always work!! ISO 11114-4 is currently under revision since it has been established that varying results are obtained for the same steel when subjected to the three tests. Such a position cannot be tolerated in an international standard with a high safety orientated mission.

Nevertheless cylinder manufacturers are keen to exploit recent awareness in steelmaking technologies, improved manufacturing practices and ultrasonic inspection techniques. Thus limited, confidential use of ISO 11114-4 has been made by some manufacturers with the result that the value of 950 MPa for the tensile strength can be exceeded, making way for lighter and efficient cylinders.

Advances with New Materials
Increasing the strength of seamless steel cylinders from 950 MPa is certainly one way forward to improve the trailer’s payload of hydrogen. But then to use such higher strength steels to form the liner of a composite cylinder (see gasworld issue Aug 2007 p.48) is a major breakthrough. Such a move can easily double the efficiency of a cylinder for a hoop-wrapped (Type 2) configuration and there are already several such trailers in hydrogen transport service. Finally, why worry about the risks from HE at all? Why not use a non-metallic (plastic) liner in a fully-wrapped (Type 4) composite geometry? After all the gas carrying capacity of such a cylinder is almost four times higher than the current steel equivalent!

Though technically superior, these Type 4 vessels still need some further developmental work. They are not impervious, allowing seepage of flammable hydrogen! The interface at the neck of the cylinder consists of three dissimilar materials (the metallic boss to accept the valve, the carbon reinforced plastic and the non-metallic liner) that all meet at the same junction. They all have drastically different mechanical characteristics and the constant pressurising and depressurising creates a source of weakness resulting in a potential leakage site. Finally the very high cost of such vessels, coupled with often a limited life, also make them currently unattractive except for the most prestigious projects.

Concluding Remarks
Leading manufacturers all over the world are developing hydrogen powered vehicles, at a frantic pace (see Fig.3). In line with these advances, gas cylinder makers are also making new breakthroughs with lightweight cylinders at working pressures of up to 850 bar. But to get such high pressure cylinders within a time comparable to filling a petrol driven vehicle, the temperature increases by around 150˚C (a temperature at which the composite material can be adversely affected) and involves filling pressures over 1100 bar.

To avoid these complications work is underway to study the effects of cooling the hydrogen gas and perhaps even the cylinder during the filling cycle.

Hence in the coming years, efficient hydrogen gas cylinders coupled with cold filling technology will replace the current hydrocarbon-based auto industry. But the key question remains…can the hydrogen be produced economically? Roll on FUSION power!!