# Problems with precipitation of transition metal solution

I am trying to make a solution of "trace elements"--primarily transition metals-- for use in biological media; bacteria require elements like Cobalt, Molybdenum, iron, etc. in trace amounts. I am not a chemist, but my understanding is that these elements tend to form hydroxides in physiological pH ranges (slightly above 7). Therefore the authors of the protocol I am using employ EDTA as a complexing agent in 1:1 stoichiometric proportion with the metals in solution. The problem is that EDTA solubility (and complexing capability) and transition metal stability occur at opposite sides of the pH scale, and finding a balance to make both types of reagents happy is apparently really tricky; I've tried this 4 different times with no success. I am able to get all materials into solution, but any disturbance of my flask, (e.g. further stirring or shaking) inevitably causes precipitation.

Here is my recipe and protocol. I'll describe my observations after that. My question is:

Any thought what the precipitate is? How can I avoid it?

I have scant chemistry experience, so any insight you can provide would be useful.

### Recipe

• 190 mM $\ce{EDTA}$ $\ce{Na4.2H2O}$ (equimolar to subsequent di- and trivalent cations)
• 29 mM $\ce{FeCl3.6H2O}$
• 80 mM $\ce{CaCO3}$
• 6 mM $\ce{MnCl2.4H2O}$
• 0.5 mM $\ce{CuSO4.5H2O}$
• 0.5 mM $\ce{CoCl2.6H2O}$
• 5 mM $\ce{ZnO}$
• 2 mM $\ce{H3BO3}$
• 66 mM $\ce{MgCl2.6H2O}$
• 5 mM $\ce{NaMoO4.2H2O}$

### Protocol (making 1 L)

1. Dissolve EDTA in $\ce{750 mL}$ of water. Bring pH down to about 5.8 with $\ce{HCl}$.
2. Add $\ce{FeCl3}$ and watch pH; don't let it drop below 4.8.
3. Add $\ce{CaCO3}$. Allow to dissolve (i.e. $\ce{CO2}$ to evaporate) by maintaining pH at 4.8 with $\ce{HCl}$ and occasionally shaking vigorously (not sure why the recipe doesn't just call for something more soluble like $\ce{CaCl2}$.)
4. Add remaining elements in order given in the recipe, watching the pH. It should drop to about 4.2 without any addition of acid or base.
5. At this point solution appears clear. Bring volume up to $\ce{1L}$.
6. Autoclave solution (heat under pressure at $123^{\circ}\ce{C}$ for 15 minutes) ---this is required for biological sterilization, and seems to improve solubility for a time.

### Observations

Solution remains transparent throughout preparation, stirring constantly throughout at room temperature. Color is orangish-green, due to iron and copper. Once all ingredients have been added, additional stirring before autoclaving leads to gradual (over 15 minutes) accumulation of a white powdery precipitate. I thought this was EDTA coming out, so I added $\ce{NaOH}$, but this only caused the iron to precipitate, forming a brown precipitate. Letting the solution sit also seems to lead to white precipitation, although less rapidly. When I autoclaved the solution, it remained transparent (color had turned to deep rust-orange) even after sitting overnight, but as soon as I disturbed it by transferring from an Ehrlenmeyer flask to a glass bottle it precipitated again. The pH of the precipitated solution seems stable at 4.2, what it was after all ingredients had been added. I have also tried this preparation adding EDTA after the boric acid and keeping pH below 6 with $\ce{HCl}$, but this caused even more rapid precipitation. I have also tried EDTA (fully protonated) after the boric acid and adding NaOH to deprotonate, but the results were even worse.

Here's an idea. Add the cations as their acetates or nitrates. Control pH with acetate buffer. Fe(III) is a bad player. "Ferric acetate," $\ce{[Fe3O(OAc)6(H2O)3]OAc}$, is water-soluble. You might sparge the solution with argon before sterilizing (sealed vessel) to prevent oxidation, and/or add ascorbate (or erythorbate, often biologically inactive) as anti-oxidant - if that is the problem.
Is it conceivable that the white precipitate is $\ce{Zn(OH)2}$? Usually it precipitates more jelly-like, but anyway...
It will however dissolve both on lowering the pH, forming colourless $\ce{[Zn(H2O)6]^{2+}}$ and on increasing the pH, yielding the likewise colourless $\ce{[Zn(OH)4]^{2-}}$.