Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

How do I measure the diffusible hydrogen level in my ferritic steel weld?


Frequently Asked Questions

There is no direct way of measuring the diffusible hydrogen in a production weld. However, a test weld can be subjected to a test to determine the diffusible hydrogen evolved over a period of time. This value characterises the amount of diffusible hydrogen that can be expected from welding with the same consumables in the same condition as those used to make the test weld.

Diffusible hydrogen is the monatomic hydrogen that can diffuse from the weld metal during cooling, and at room temperature. Residual hydrogen is that trapped at various sites within the metal, which can only be driven off at elevated temperature. Total hydrogen is the sum of the diffusible hydrogen and the residual hydrogen.

Principal method

The principal method for determining the diffusible hydrogen in a test sample involves allowing hydrogen to evolve by diffusion at room temperature. The hydrogen is collected over a liquid in which it will not dissolve, usually mercury. The recommended practice is to leave the sample until the volume of hydrogen collected does not increase on successive days. The hydrogen evolved, together with the test piece weight, are used to calculate a volume of diffusible hydrogen (ml at standard temperature and pressure) per 100g deposited metal (ml/100g). Sometimes a value of hydrogen in fused metal is calculated.

Collection of diffusible hydrogen over mercury until there is no more hydrogen evolved can take a few weeks, so rapid methods have been developed to determine the diffusible hydrogen in less time. Such methods inevitably involve the evolution of hydrogen from the test sample at a temperature above room temperature, and using a different measurement method, as mercury cannot be used at temperatures above 50°C. The use of mercury is increasingly becoming restricted, and so the principal method is currently under review.

Rapid methods

Rapid methods for determining the diffusible hydrogen use temperatures from 45°C to 400°C, and methods such as vacuum hot extraction, carrier gas methods and gas chromatography are used. Although some less strongly trapped hydrogen is driven off at elevated temperature, methods employing temperatures up to 400°C have been shown to give results very close to values determined by the primary (mercury) method, for a number of ferritic steel weldmetals.

  • Vacuum hot extraction
    Vacuum hot extraction involves heating the specimen, in a vacuum-pumped system, to the required temperature, to allow hydrogen to evolve quickly. The evolved hydrogen is transferred to the analysis volume, along with other gases emitted from the sample or furnace tube. Condensable gases are removed via cold traps, and the pressure of the gases is monitored until hydrogen evolution stops (judged to be when the pressure readings do not change at this temperature). The hydrogen is removed from the analysis volume through a palladium/silver osmosis tube, and the pressure of the residual gases is measured. The difference in pressure is the partial pressure of the hydrogen extracted from the sample. Typical times at varying temperatures for the evolution of diffusible hydrogen are as follows:

    Time for total evolution
    45 72 hours
    150 6 hours
    300 50 mins
    400 30 mins

    Vacuum hot extraction is also used to determine the residual hydrogen, or the total hydrogen of the sample. The method is the same as for diffusible hydrogen measurement, but the temperature is 650°C, to free all trapped hydrogen. The residual hydrogen measurements are taken when the diffusible hydrogen has been determined by another method, or at a lower temperature. Total hydrogen is determined when 650°C is used throughout testing, and takes approximately 30 minutes.

  • Carrier gas extraction
    Carrier gas extraction uses an inert carrier gas (such as Argon), and a detector. The detector output traces a peak on a recorder. The area of the peak is then proportional to the volume of hydrogen that has evolved from the sample.
  • Gas chromatography
    Gas chromatography generally uses dedicated equipment, and hydrogen analysing-only units are available. The output can be in a graphical or numerical print out, and the diffusible hydrogen content can be determined.


All of the rapid methods mentioned can be related back to the mercury method, which is detailed in various standards. The international standard that details the mercury method for diffusible hydrogen determination is EN ISO3690-2012.

It is important to refer to standards when determining diffusible hydrogen, as there are slight differences between national standards, and some national standards may not have mercury as the primary method. One such national standard is JIS Z 3118, 2007, which details the glycerol method, where the test piece is placed in a beaker of glycerol, and the hydrogen evolved at 45°C is measured by the displacement of glycerol. Glycerol does absorb some hydrogen, so low levels of diffusible hydrogen cannot be measured, and any measurement will be slightly lower than for the mercury method. For low levels of diffusable hydrogen (e.g. hydrogen per deposited metal ( ≤ 2ml/100g) gas chromatography is suggested as the method to apply.

Values of hydrogen evolved over mercury at ambient temperature, over a period of three days, are sometimes specified - in an attempt to get results more quickly. This approach is not recommended.

See further information about Materials and Corrosion Management, or please contact us.

For more information please email: