You didn't indicate how the electrode is used and what the sample solution consists of.  If it used in a two-electrode system and sufficient charge is passed, you could be polarizing the electrode.  If it is a three-electrode system, this should not be a problem.  I agree that components of the sample solution could block the liquid-junction frit and/or the formation of AgCl precipitate due to dilution.  In general, it is recommended the a minimum of 3.5M KCl be used as the filling solution, but less than ca. 4.2M (saturated KCl).  The problem with the saturated solution is that, while it is easier to prepare and better for maintaining the equitransferent nature of the liquid junction, the small crystals can spontaneously grow to the point where the inside surface of the frit is blocked.

Hi dear Abdellatif

1-Ag/AgCl normally filled with KCl 3M which generates some soluble Ag species like AgCl2-,AgCl32-, etc. which diffuse slightly through electrode's ceramic frit. These species quickly turn into very insoluble AgCl when you use your electrode in a very low ionic content water such as deionized water . A similar problem is likely to happen when you use your electrode in sulfide-containing water and protein bearing sample(CERAMIC FIRIT CLOGING).To solve this problem use slightly acidic 7% thiourea and leave your electrode in this solution for several hours to remove the precipitate.   

2-Replace KCl 3M with freshly prepared one and ensure that the AgCl in your electrode is immersed in KCl.

3-Always keep closed the KCl reservoir to avoid water evaporation and changing the concentration of KCl and forming the KCl crystals.

A.Rabiei

Thanks so much Dr À. Rabei for your valuable answer. 

Abdellatif 

You have a Ag/AgCl in KCl 3M. If nothing works, you can anodize again the silver wire in chloride medium and then fill with the electrolyte and let stabilize.

Good luck!

Teresa

Thanks so much Dr Tereza Fernandez-Abedul.

You didn't indicate how the electrode is used and what the sample solution consists of.  If it used in a two-electrode system and sufficient charge is passed, you could be polarizing the electrode.  If it is a three-electrode system, this should not be a problem.  I agree that components of the sample solution could block the liquid-junction frit and/or the formation of AgCl precipitate due to dilution.  In general, it is recommended the a minimum of 3.5M KCl be used as the filling solution, but less than ca. 4.2M (saturated KCl).  The problem with the saturated solution is that, while it is easier to prepare and better for maintaining the equitransferent nature of the liquid junction, the small crystals can spontaneously grow to the point where the inside surface of the frit is blocked.

Dear Dr Richard A Durst,

Thank you so much for your answer. I'am working with a traditional three electrode system, with Ag/AgCl as reference electrode.

My very best regards 

Abdellatif, did you solve the problem? 

Dear Aisha,

I tried to solve the problem following the above recommended procedures. The potential shifting persist.

Thanks

Abdellatif,

Dear Abdellatif,

You have received just about every advice regarding Ag/AgCl electrodes that I can think of. It seems that your problem hasn't been solved by adhering to them.

Maybe we have misunderstood your question. You're talking about a potential shift of 200 mV. Is this a shift against a potential that you have measured before or against an expected potential?

In the latter case it would be the potential difference between an Ag/AgCl (3M KCl) vs. a Normal Hydrogen Electrode (NHE) that's roughly 0.2V. See for example: http://www.consultrsr.net/resources/ref/refpotls.htm#ssce or have a look at the links of the following discussion: https://www.researchgate.net/post/Can_anyone_please_explain_me_the_difference_between_NHE_RHE_and_SHE_in_a_simple_way.

https://www.researchgate.net/post/Can_anyone_please_explain_me_the_difference_between_NHE_RHE_and_SHE_in_a_simple_way

http://www.consultrsr.net/resources/ref/refpotls.htm#ssce

Dear Peter Sculthess,

Thanks for your answer. Actually the shifting of 200 mV is compared to a previous measured value using the same reference electrode in the same experimental conditions.

Kindest regards

Abdellatif,

I am very puzzled by your problem. The 200 mV is a very large shift in potential -- equivalent to more than three orders of magnitude in concentration. Could you provide more information on the actual system that you are working with and experimental details? Then, perhaps, I could make a more informed suggestion as to how you might resolve this anomaly.