I have wondered this for awhile. Let's say I dissolve, without loss of generality, potassium chloride and sodium chloride in water, then evaporate the water. Would any of the resulting crystals consist of randomly dispersed potassium and sodium cations within the lattice or would each resulting crystal be purely potassium chloride/sodium chloride?


Crystals with randomly intermixed ions only occur when the ions have very similar diameter (really the same spatial charge distribution in its exterior, but alkali ions are in very good approximation spherical) and same charge. That would be called a single phase solid solution.

However, $\ce{Na+}$ and $\ce{K+}$ have quite different diameters ($r=102$ resp. $\pu{138 pm}$). Your mixed solution will crystallise as NaCl with a low amount of potassium ions and KCl with some potassium replaced by sodium ions. In other words, your isotropic solution decomposes into a solid two-phase mixture.

The two substances will surely have a different tendency (also temperature dependent) to include such defect ions, and the two compounds have different solubility, so one will precipitate first, so the concentrations in the residual solution changes greatly during the process, and consequently also the defect concentration in the single crystals.

  • $\begingroup$ To add to Karl's point, there is a process called zone refining. Think of a compound packed in a glass tube. You melt a small zone of the tube going from one end to the other. You can repeat the process. A purer compound results at one end and the contaminates are concentrated at the other end. $\endgroup$ – MaxW Jul 24 '18 at 14:06

I did a lot of x-ray fluorescence. Every sample of NaCl which I had also had detectable amounts of $\ce{K^+}$ as well. It is really really hard (impossible?) to get any salt pure enough that XRF doesn't detect other elements.

Look at the Wikipedia article for Sea salt. It is far from being pure sodium chloride.

  • $\begingroup$ Interesting. However, sea salt is not the best example here, evaporation of all of the solvent can be barely called "crystallization". $\endgroup$ – mykhal Jul 24 '18 at 7:40

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.