I am currently attempting to purify technical grade sodium tetraborate decahydrate by recrystallization for use as a titration standard. I don't have a requirement for a specific grade or purity. I'm just looking to improve my results a bit.

I understand that one of its qualities for use as a primary standard is the stable hydration number. However, if I'm drying wet crystals, what would be the best way to ensure that there is no extra water without affecting this? Could heating produce the pentahydrate or anhydrous form?

This is a separate question really, but would it be better to convert to boric acid, recrystallise and then neutralise with a sodium base?

  • 1
    $\begingroup$ It is generally not a good idea to consider salts with crystal water as standards. As, with possibly rate exceptions, you cannot guarantee sufficient moisture elimination and defined composition at the same time. $\endgroup$
    – Poutnik
    May 17, 2020 at 5:40
  • 1
    $\begingroup$ I'm a little confused. Yes, they would need to be dry, but not anhydrous. Sodium tetraborate decahydrate is considered a good primary standard. Is this dried in some way not accessible to a lab? Or purified in some way that doesn't involve water? $\endgroup$
    – eisd
    May 17, 2020 at 7:10
  • $\begingroup$ I suppose borax may be within those exceptions. But the result should be controlled by a commercial primary standard. BTW it is not true decahydrate, as there are 4 OH groups bound to boron. $\endgroup$
    – Poutnik
    May 17, 2020 at 8:42
  • $\begingroup$ Yes, I could just buy a higher purity commercial primary standard, but I'm interested to find out how much closer to that I can get with the supplies I have and basic methods and equipment. The question is what drying method to use. Would there, for instance, be a temperature to stay below to avoid driving of the water of crystallisation? $\endgroup$
    – eisd
    May 17, 2020 at 9:37
  • 1
    $\begingroup$ Unless some differential gravimetry data are available, you may need to experiment to get m=f(T). $\endgroup$
    – Poutnik
    May 17, 2020 at 9:40


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.