I have been making a conscious effort to find ways to use borohydride compounds rather than just default to the mighty gray sledge hammer- LAH. There are numerous reports of diverse and wonderful means of activating borohydride to reduce the more refractory functional groups. Recently I prepared and used Zn(BH4)2 on a new substrate. Initially, it appears to work poorly. The grey wall cake seems to contain metallic zinc. If preferences mattered, I’d prefer see electrons reducing my substrate rather than Zn (II) to Zn.
You can’t always get what you want. M. Jagger
Lamentations on Chemistry 
Borohydride is juiced by strong Lewis acid cations like lithium. One wonders how the magnesium salt would go, re Mg(MnO4)2. Al(BH4)3 boils at 45°C, hydride-bridged and vastly evil.
http://cameochemicals.noaa.gov/chemical/2377
“Butene explodes after an induction period”
Click to access brown-lecture.pdf
http://www.freepatentsonline.com/3305569.html
LiBH4 has always worked well for me. Easier to deal with than LiAlH4.
Ok. So you all can help a discussion we had the other day. Why does LAH ring open epoxides and not NaH?
Would NaBH4? Can you point us to the pertinent literature?
NaH is not a H- donor, it is only a base.
I personally always use Vitride. It is safe (you can pour it in air) and it has the same reducing abilities as LiAlH4.
The work up can be a little tricky, especially hydrolysing the Al-O bonds, but otherwise, it is my go-to reducing agent.
Hi Bill, I think you can open up an epoxide with LiBH4 or Zn(BH4)2. There are a handfull of references to exactly this reaction. One of the interesting aspects of BH4 is the effect of MeOH on a THF solution. MeOH will activate a mixture of ester and NaBH4 to cause the reduction of the ester to the alcohol. It must be generating BH3 in situ.
I don’t think NaH has any Lewis acidity, and the H- isn’t a very good nucleophile, so it doesn’t do anything (unless the epoxide is base-sensitive). I don’t know why LiBH4 opens epoxides rather than NaBH4, though (matching anion/cation sizes?)
Don’t electrochemical reductions usually require lots of coreagents (salts for the medium, etc.) to mediate the electrochemistry? In the absence of that, electrochemistry would probably be nice. Some ketones are supposed to be reducible with sodium in ethano, but I don’t know how much won’t tolerate the conditions (or won’t be reduced).
Electrosynthesis. There is an area I wish I had time to try some experiments. You know, the wall outlets give high purity electrons.
NaH is insoluble for doing hydride attacks. Obvious adjuncts are 15-crown-5 ether or [2.2.2]cryptands. LiH will be more amenable to complexation with reduced core volume seqestration for its stronger Lewis acidity. Doesn’t mean it’s gonna work.