Home Page | Rust Removal | Gin Traps | Grimspound | Manganese Mining | Hydrogen Embrittlement
Hydrogen embrittlement is the name given to a certain condition affecting steel and some other metals whereby the metal can become brittle and therefore liable to fracture as a result of absorbing hydrogen during certain treatment, notably electroplating. The process by which this occurs is not fully understood, although it is generally agreed that hydrogen diffuses into the metal's surface at an atomic level and tends to collect between the crystal lattice or within defects, where the pressure build-up can be sufficient to wedge the grain apart and in extreme cases cause microscopic fractures to occur which can reduce the load bearing qualities of the piece. In more moderate cases the metal simply looses its ability to flex and becomes brittle and liable to break.
The hydrogen responsible for causing hydrogen embrittlement can come from a number of sources, but this page deals specifically with the effects of rust removal by electrolysis, and this is our source of hydrogen. When the piece to be cleaned is connected up in the manner required for electrolytic rust removal, hydrogen is evolved at the cathode, the negative terminal, and there is a possibility that some of the hydrogen may permeate the metal. This wouldn't be a problem for mild steel, but if the piece contains hardened or spring steel then there is a possibility that damage may occur either due to the metal developing micro fractures or due to an embrittled spring being operated.
There are many sites dealing with the problem of hydrogen embrittlement, and a simple search will produce many results so I won't attempt to elaborate on the detailed science of this phenomenon any further; However, the effects seem to be difficult to predict as they are dependent on many variables, but importantly the effects are generally considered to be temporary.
Because of the generally vague nature of the details of the effects listed on other sites, I decided there was only one sensible course of action, and that was to attempt to produce a piece of high carbon steel which exhibited the effects of hydrogen embrittlement in order to test it to determine its severity, and most importantly to test just how long the metal took to return to normal when exposed to the air, if indeed this did occur. The details and results of my experiments are set out below.
PLEASE NOTE THIS PAGE IS STILL UNDER DEVELOPMENT, AND THERE ARE NO RESULTS AS YET.
A jig would be designed to enable a measurable bending force to be applied to the test component. A number of untreated components would be tested first to determine their point of fracture in order to establish a control group, and a base figure for MPDF obtained. (Maximum Possible Deflection before Fracture)
Next, a sample of test components would be connected as cathodes and immersed for a measured time in an electrolytic cell with a known electrolyte strength with sufficient, and measured, current passed to cause copious quantities of hydrogen to be produced on their surfaces. This batch of test components would then be tested for MPDF and the difference in MPDF noted. There may be a need for further batches to be tested using different hydrogen exposure times until a suitable embrittlement condition was reached.
Finally, further batches would be exposed to hydrogen for times determined by experience gained from previous batches, but these would be removed and left to 'rest' in the atmosphere. They would eventually be tested for MPDF to test for the ability of the metal to regain its former characteristics. Several batches may need to be tested using different 'resting' times before a sufficient time for complete reversion can be established.
Considerable thought was given to my choice of test component; It had to be made of a material which could possibly be susceptible to the effects of hydrogen embrittlement, it had to be similar to metals commonly found on items likely to be subjected to electrolysis for the purposes of rust removal and the metal had to be thin enough to enhance the chances of absorbing hydrogen to a reasonable percentage of its thickness. It also had to be easily testable for possible embrittlement and available in large numbers. The item finally decided on was a roll pin, which for those not familiar with engineering, is basically a metal dowel made by rolling a strip of carbon steel into a tubular form, which is used for securing components where a shear force exists. The inherent springiness of the roll pin allows it to be a very tight fit within the hole it is pushed into, ensuring it stays in place, and gives the dowel a high degree of resistance to the shear forces it will experience. The roll pin size chosen was 3mm in diameter and 40mm long with the thickness of the steel sheet the pin was made from being about 0.5mm, an example of one being shown magnified in photo 1. This small size allowed for easy placement on a jig to test its resistance to an applied bending force without that force having to be too large.
The roll pins as they came, straight out of the box, were coated in a layer of protective oil. The roll pins destined for exposure to hydrogen would need to have the oil removed before immersion in the electrolyte, so in the interests of consistency it was decided to treat all roll pins the same whether destined for the untreated control group or destined to be exposed to hydrogen, and as such were all washed in detergent to remove all traces of the oil and then quickly dried. The roll pins to be subjected to electrolysis were then connected up in a 10% solution of washing soda, consistent with the strength I normally use for de-rusting iron, and a current was then passed which equated to (current yet to be decided) per roll pin ensuring a good stream of hydrogen bubbles were produced. The setup for this can be seen in photo 2. The pins were left like this for exactly one week, and then removed and immediately tested for brittleness and the results compared to the untreated control batch. Further batches were treated for one week as before, but this time washed, dried and exposed to air at room temperature for different lengths of time, initially for twenty eight-days, before being tested for brittleness.
The jig I designed for applying a repeatable and measurable deflection to the roll pin is shown here in photo 3. To apply the force to the pin, it is loaded on the jig so that the centre clevis arrangement is set to pull on the centre portion of the pin, whilst the outer ends of the pin bear against the two sturdy stops. As the bolt head is turned, it screws into the body of the clevis therefore shortening the effective length causing the centre of the pin to be pulled inwards forcing it to bend. As the bolt is tightened and the roll pin bent, there is a considerable twisting force on the clevis, but this tendency to twist is dealt with adequately by the clevis being of square section and a snug fit between the sides of the angle-iron which forms the stops. It is therefore prevented from turning as the bolt is tightened, which is important as this could alter the stress point conditions possibly leading to inconsistencies and resultant errors in the readings.
The degree of deflection can easily be measured by first tightening up the bolt until it just fully engages the test roll pin to the point of the elimination of all slack movement; This in effect calibrates the jig and compensates for wear on the components. The bolt can then be tightened, ideally one flat at a time until the roll pin fractures or full available deflection is reached. The number of turned 'flats' counted when the pin breaks is then recorded. The actual deflection could be calculated if necessary by knowing that the thread pitch of the bolt is 1.5mm, and there are six 'flats' to one turn.
No results available at this time - experiments still underway.
If you have any comments, or suggestions for additions or corrections to this page
please feel free to e-mail me at this address:
© Andrew Westcott 2003 - 2006