My suggestion would be to decrease the flow rate.

At a 50 µL.min^-1 flow rate and a 1 mm channel length I calculate the velocity to be 0.26 m.s^-1, with a Reynolds number of 13.9. Pressure would be around 28.5 mbar.

I assume when you mention the metal layer thicknesses that you mean nm and not µm. 10-30 nm seems thick for a Cr adhesion layer. Also it is worth mentioning that annealing at the temperature will cause diffusion between the layers, and could make the electrochemical response inferior.

You could try a Ti adhesion layer instead, 3-5 nm is a good range to use, usually the Ti evaporation also helps reduce the pressure in the evaporation chamber.

The buffer solution with lower conductivity is expect to prevent micro electrodes detachment.

@jules thanks for you comprehensive answer , My device works based on inertial microfluidics so I even need to increase Reynolds up 200, I will check effect of reducing the thickness of my Cr layer to 5 nm and will share my results.

Hi Xiangyu Chen, Unfortunately I'm unable to dilute the solution since the sensitivity of my device will decrease dramatically, but anyway thank you for information I was not aware about this.

Hi Navid,

If you work with PBS, and then you have a medium with chloride, Au will probably form AuCl4- (is first oxidized to Au(III)) and then is detached, is dissolved forming the complex. This depends on the pH, being easier at higher pH (e.g., at pH 8 happens easier than 7). It also depends on the adhesion of the layer gold (Cr, Ti...), but can be a reason. If you can decrease pH and chloride concentration, maybe it can help.

Best regards

Maybe, HEPES with same conductivity would be helpful.