It is known that Au metal is diffused when it is contacted with copper metal. And Au metal have lattice which consist of atoms. If so, may I think that atom is diassociated from lattice by heating ? If so, it needs to break bond between atoms before diffuion ? I think that it needs melting temperature to break bonding in lattice.


Yes, it takes energy to get an atom off a lattice site. However, as an activated process with an energy barrier, Arrhenius' equation shows you that even with that requirement, some number of atoms will be off their lattice sites in thermal equilibrium, making vacancies (empty sites) and interstitial atoms.

This was realized some time ago by Simmons and Balluffi. They investigated many metals, using precision measurements of the change in length of a bar of material in a furnace, coupled with x-ray diffraction to determine the lattice constant at the furnace temperature. The result is you can back out how many lattice sites are vacant from the difference in the change of length actually measured vs what you would expect from the change in lattice spacing.

The linked article focuses on gold, but has some data on other fcc metals as well. Just at the melting point of gold (not yet melted), the concentration of vacancies is about $7 \times 10^{-4}$, or 0.07% of lattice sites are vacant. The activation energy is given as 0.94 eV.

Bottom line - there are native populations of point defects even at room temperature, and those drive diffusivities. Certain point defects or defect complexes may be more mobile than others, and the concentration of a given defect will vary with temperature.

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.