488
Corrosion of Copper in De-Aerated Water

Thursday, May 15, 2014: 11:00
Orange, Ground Level (Hilton Orlando Bonnet Creek)
C. Cleveland, M. E. Orazem, and S. Moghaddam (University of Florida)
Impedance spectroscopy is being used to study possible mechanisms for the observed corrosion of nanostructured copper wicks intended for use in heat pipes that provide advanced cooling for high-performance electronics. The heat transfer fluid is deionized and deaerated water.

The experimental techniques employed in this study include impedance spectroscopy and linear sweep voltametry. Experiments were performed in 17.6 MOhm cm water de-aerated by N2 until the oxygen concentration of the gas leaving the cell was measured to be < 2ppb by a Hach Orbisphere 3650 analyzer. Copper working electrodes were polished and degreased prior to experimentation. Impedance data were obtained potentiostatically using a Gamry Ref 3000 at perturbation amplitude of 10mV.

EIS Data & Model: The open-circuit impedance responses of 0.025 cm diameter gold, platinum and copper electrodes are presented in Figure 1. The high-frequency loop can be attributed to the dielectric response of water and has been omitted in Figure 1. All impedance data were scaled with respect to the ohmic resistance Re. This result shows that, while gold and platinum show blocking behavior, the slight curvature of the low-frequency data for the copper electrode indicates the presence of electrochemical reactions. 

A model for the impedance response should include the parallel contributions of anodic and cathodic reactions. A maximum value for corrosion of 2.5 nm/day was found by extrapolation of the impedance data to low frequency by use of a measurement model.1 The corresponding diagram is presented in Figure 2.

Kinetic Model:  In the absence of oxygen, the reactions were assumed to include dissolution of copper, hydrogen evolution, back deposition of copper ions, and oxidation of the hydrogen formed by the reduction of H+. A simulation was performed using kinetic expressions for these reactions coupled with diffusion of reaction products away from the electrode. Spherical diffusion was assumed.  

As kinetic data in pure water could not be found, kinetic parameters were taken from experiments in buffered boric acid and sulphate solutions.2,3 The corrosion current was estimated to be 1. nm/day. While this corrosion rate is too small to have a practical consequence in usual applications, it is large enough to influence the structural integrity of nano-scaled structures. 

 

Acknowledgement

This work was supported by DARPA contract number W31P4Q-12-1-0005.

References

  1. P. Agarwal, M. E. Orazem, and L. H. Garcia-Rubio, J.Electrochem. Soc., 139 (1992) 1917-1927.
  2. Z. Stankovic and M. Vukovic, Electrochem. Acta, 41 (1996), 2529-2535.
  3. S. Shari-Asla and D. D. Macdonald, J. Electrochem. Soc., 160 (2013), H382-H391.