A friend wanted some potassium bromate, for making bromine (which, in turn, is for vapor-phase extracting platinum group metals, as the volatile bromides, from catalytic converter substrate). Not having the time to do it himself, I asked him to send me some bromide and I'll upgrade it with my new platinum anodes. A week later, four pounds of granular sodium bromide arrived and I got to work.
It's a straightforward process, working just the same as a chlorate cell. The major difference is probably that elemental bromine and hypobromite, OBr-, are important intermediates in the reaction. At the anode, elemental bromine is produced, having a slight preference over dissolving in solution. Bromine is very dense and sinks, making a deep colored solution at the bottom of the cell. Between stirring (due to convection and hydrogen bubbling) and diffusion, the alkaline solution from the cathode causes bromine to disproportionate to the transient hypobromite, Br2 + OH- = H+ + Br- + OBr-. The Br- of course goes back for more, while the hypobromite disproportionates readily to bromate and bromide: 3 OBr- = 2 Br- + BrO3-. Adding to the color of the solution, chromate is added to prevent reduction at the cathode, which occurs to a far greater extent than in a chlorate cell. Very little bromate is produced in equilibrium without it. See also: Synthesis of KBrO3. In the picture above, the brown color towards the bottom is rich with bromine (don't let its delicious color fool you, it smells like chlorine, but spicier and decidedly unpleasant to be exposed to!), while the yellow-green to white color towards the top is more representative of the chromate color.
Whoda thunk it. It turns out bromine and titanium don't get along together. Suprising, as the metal completely and wholly ignores just about everything else on the periodic table! This is the surface after, oh, a few hours, which left a pile of white crap in the cell. I wirebrushed and washed off the gunk, then annealed it to about 700°C, giving a deep color to the titanium's oxide surface. Apparently, the platinum plating is still intact.
Opposite side. After annealing, I "anodized" in a 20% sulfuric acid solution (which didn't act like anodizing, due to the platinum giving oxygen in solution, but I can only assume the titanium surface was anodized a little), then put it back in the cell, in the hopes that the oxidized surface would stand up a bit better. It didn't, and in the morning I was greeted by a yellow-white mud and the remains of the anode, consisting of swiss cheese. Hey, on the plus side, I've found one of titanium's "kryptonites".
In case you were wondering, I finished the bromate with graphite anodes. When the bromide concentration runs low, the graphite starts oxidizing and swelling (like it does when "anodized" in a sulfuric acid solution, unlike its behavior in a chlorate cell, where it tends to just wear away, rather than swell up), so I went through a few pieces of graphite too. That could be avoided by running a continuous process, as I do with sodium chlorate.