I think the surface state of stainless steel is the most important to get good result.You should characterize the surface state of stainless steel first,such as CV before you do the other experiment.

Hei,

what do you mean with electric charge held?

You probably have a potentiostat setup with a counter electrode and a reference electrode.

If you want to get hydrogen into the sample just polarize it to hydrogen evolution for some time. (I guess if you have problems just use galvanostatic mode and some higher current like 10 mA/cm^2 )

Some of the hydrogen evolved will penetrate into the sample

Are you using a Devanathan Stachurski cell?

http://dx.doi.org/10.1149/1.2425894

Then you have two sides, one side on which hydrogen is evolved and one other on which you oxidize the hydrogen that went through the sample. If you use two potentiostats I think this doesn't work because they will be connected through ground loops or something like that.

Laboratory current source at one side, and potentiostat on the other and then something like -0.3V vs AgAgCl

Normally the side where hydrogen is oxidized is coated with Pd, but it works also without. At least for normal steel. But you probably have to let it passivate first before you switch on the hydrogen evolution on the other side (like 5 h or so)

Sample thickness is also critical, 1mm is maybe still ok but more than that might be already too much.

And of course these experiments take quite some time until steady state is reached (1 night at least or so)

Regarding preventing the hydrogen to diffuse out...

I think you will have a very hard time to do this, because the pressures of hydrogen in the sample can be quite high

And immediately when you remove your sample it will already loose hydrogen

Regarding the hydrogen to get in

If your sample passivates, this might, I think either prevent hydrogen from entering, or slow it down

Cyanides might help to poison the recombination reaction and there was some other, was it Arsenic?

I don't remember

But actually I would try first to use electrochemical hydrogen evolution under galvanostatic conditions and a Devanathan Stachurski type of cell (doesn't need to be that exact, just something that has two separated compartements and an area where the surface area is big compared to the thickness)

Regarding reproducibility

If you use the sample twice the second time it will look totally different. Maybe prior heat treatment at let's say 200C helps. If you have passivation you have to wait until this current (on the -0.3V side) has decayed.

I think the surface state of stainless steel is the most important to get good result.You should characterize the surface state of stainless steel first,such as CV before you do the other experiment. - I will look into that.

what do you mean with electric charge held? - constant current

You probably have a potentiostat setup with a counter electrode and a reference electrode. - yes

If you want to get hydrogen into the sample just polarize it to hydrogen evolution for some time. (I guess if you have problems just use galvanostatic mode and some higher current like 10 mA/cm^2 ) yes, and much higher current was used after several trials, max was around 80 mA/cm^2

We used a cell very similar to the Devanathan Stachurski cell, modified to better fit our testing apparatus

Samples were left to charge for 8 days s at max... this was done due to sample thickness, 4mm diameter

As far as poisons go, I'm looking to avoid ones that also serve as human poisons, we are looking at sulfates and thiocyanides in acidic environments now.

electrochemical hydrogen evolution under galvanostatic conditions and a Devanathan Stachurski type of cell - This is my current process

always using new samples, prior bake to remove hydrogen and give us a controllable baseline. baked at 500C for 24 hours

also looking into platinum counter electrodes as opposed to graphite, and platinum coated titanium

I remember that there was a critical thickness above which it was not anymore possible to detect anything.

4 mm is clearly much too thick .Try 1 mm if you can, common sample thickness is if I understood right 0.3-0.5 mm

but we used also 1mm, but steel not stainless steel.

Whatever counter electrodes you use it should be just the same.. as long as they support enough current at cell voltages less than compliance voltage (usually at least 12V)

In alkaline media it is good to use HgO reference electrode, because AgAgCl is not considered very stable in strong alkaline, but if that helps to get some respons I doubt it.

Sample thickness is much more critical. If you can't get it thinner, I suggest you drill for example some holes to thin the sample locally.

The samples we used back then were also 3 mm thick but the workshop guys machined it down to 1 mm in the middle,

such that we could clamp it still into the "sample holder" but penetration thickness was only 1mm.

You must have quite massive samples, what is you thickness to surface area ratio. Wondered just could that be also part of the problem. We had also quite big ones: 3mm thick thinned in the middle to 1 mm, but area 5 cm in diameter.

Could be also that the whole thing is much more difficult with stainless, because some of the comments were such.

Would look for references with stainless and Devanathan Stachurski cell

http://dx.doi.org/10.1016/S1359-6454(00)00122-1

some stainless steel reference

I am not a specialist in this field but I'm trying to help you as much places,please find attach

I'll update once we do some more test. Thanks !

We found that pre-stressing the metal (especially low carbon high alloy) brings the hydrogen diffusion up 30 fold. The alloy we were using was notoriously bad at hydrogen diffusion and stressing during the diffusion took our numbers from 0.3 ppm to nearly 10 ppm.

Got more success with Pd sputter coating and nitrogen diffusion in the test solution to drive out oxygen. used thinner material and started getting results. oh Pt counter electrode helped too.

Interesting that the Pt counter electrode had an influence.

Would be surprising if you anyway constantly remove the oxygen by bubbling nitrogen, but I remmeber also that some guy in Norway studying hydrogen evolution on Ni based catalysts used a diafragma bag for the counter electrode. This prevented the products of the counter electrode reaction to enter the electrolyte and without that it didnt work. (I think the solution was alkaline).

Regarding all this, might it be that Pt does not generate any strange oxidation products that are soluble in alkaline solution?And what would happen if youd bubble the solution with hydrogen to drive out or reduce all soluble oxydants?

Diafragma is probably difficult bec youd want an even current distribution also for bigger samples

I can imagine this was quite tricky since the currents that we measured on the non reacting side were even not very high (wa it 5-30 µA on a 5 cm diameter sample) and not stainless. And when I calculated the hydrogen concentration on the reacting surface it was something around 10 ppm

I appreciate it very much that you sent some details to the solution of your problem, because it  was long time ago and there might very well being loads of other peoples facing the same problems...