3
$\begingroup$

Could someone explain in detail as to why exactly recrystallization gives a more pure solid sample? As well as why is would be better 'technique' to allow for the crystal to form slowly?

I was thinking that it would be because being a purification technique it essentially breaks the lattice and allows the structure to reform again. As for why it needs to be done slowly, well, because rapid changes in heat would cause the structure to reform into the impure solid it was.

I'm looking for a more detailed, better expressed, answer to this.

$\endgroup$
  • $\begingroup$ Why recrystallization gives a more pure solid sample than what? Precipitation? Sublimation? Evaporation? Fractional crystallization? $\endgroup$ – long Oct 25 '14 at 4:03
2
$\begingroup$

Recrystallization is a purification method because it is a slow selective generation of a crystal framework, (mostly) free of trapped impurities. It is only effective as a purification method if done properly. Poor recrystallization techniques; rapid cooling, dramatic changes in solubility with temperature, bad choice of solvent, over evaporation of solvent, can all produce crystals of low purity which are not much better than the starting crop.

Impurities in crystals can occur via three mechanisms;

  • inclusion: impurities occupy a lattice site within the crystal framework; that is, they become part of a defective crystal unit themselves

  • occlusion: impurities become physically trapped within the crystal framework. This can be a common occurence for some solvent molecules

  • adsorption: impurities form a weakly bound interaction to the crystal surface.

The faster the crystallization takes place, the greater the chances these mechanisms have of occurring. Crystals are specific odd-shaped units that need to pack in nicely to their neighbours to form a good crystal. This takes time. A new molecule may attach to the growing lattice, but if it is not the right fit (such as because it is an impurity), it will leave in favour of a better suited fit attaching to the lattice. I may be showing my age here, but we used to liken crystal growth to a game of Tetris. Heck you can still play it! Done slowly, there is time to rotate all the pieces and get the perfect fit. As you speed things up, it becomes harder to get the right fit, holes are left, and before long the whole crystal goes to the dogs. That is essentially what is happening with rapid crystal growth. It's why you get smaller crystals and greater levels of impurities.

Of course, there other methods for purifying solids which are effective also, if used correctly. Precipitation, sublimation and fractional crystallization are all methods for purifying solids, which can give high levels of purity, matching purities obtained be recrystallization.

$\endgroup$
0
$\begingroup$

Crystallization is an equilibrium process where we count on the pure compound interacting with itself to be a lower energy state than interacting with impurities. If impurities become trapped in the growing precipitate, we want the system to have enough time and energy to redissolve the surface of the crystal and trapped impurities in the bulk solution. Said another way, effective crystallization achieves the lowest energy state, so we should use thermodynamic conditions: long time and high temperature (i.e., without cooling if possible).

$\endgroup$
  • $\begingroup$ Well, you are counting on the solubility of the impurity being different in the solid and in the melt (which in general, of course, is true). Time can certainly be a factor, because if the impurities segregated to the melt cannot escape the moving crystal front they will be kinetically trapped. Look into 'zone melt refining'. Further on a classic binary alloy diagram, you do have to continue to cool (until you reach the eutectic) or crystal growth will not continue - that assumes you have a very large impurity concentration. $\endgroup$ – Jon Custer Oct 24 '14 at 13:04

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.