# Why so much color change in synthesis of triphenylmethanol?

In lab today, I prepared a solution of Grignard reagent (phenylmagnesiumbromide) in anhydrous ether. I then added a solution of diphenylmethanone in anhydrous ether (AKA benzophenone) to the Grignard reagent and observed a brilliant array of color change.

Immedietely when the first drops of the benzophenone were added to the cloudy grayish grignard mixture (grey color due to impurities in starting magnesium), the reaction turned a dark purple before quickly changing to a crimson red, before once again changing to a bright pink. As more benzophenone was added, the reaction mixture changed, once more, to a milky white color and promptly hardened.

I understand the white final color and hardness is due to formation of my intended product: triphenylmethoxide magnesium salt...

But WHAT in the world was happening in that reaction with all those fantastic color changes?

• The various shades of blue-purple are a trademark color of benzophenone dianion that could be created by reduction of benzophenone by remaining excess magnesium. Benzophenone and sodium are commonly used in distillation of solvents to produce especially pure and oxygen-free solvents. At lower concentration the anion might appear to have brighter shades. – permeakra Aug 6 '16 at 20:02
• @permeakra, magnesium should be relatively fresh, but yes, that's also a possibility, good catch :) – ChemistryHelpCenter Aug 7 '16 at 5:10

It sounds like you might have generated a species with this structural moiety:

This can happen if your solution was concentrated enough or the boiling of your ether was vigorous, so the local overheating caused the formation of $\ce{MgO}$, the guy above, and $\ce{Br-}$, which gives a very distinct bright red-pink color, which is VERY bright and only requires a bit of the side product to form. Though this carbocation is kinda stable, it will find the $\ce{Br-}$ in the solution and form triphenylmethylbromide as a minute impurity.

• This is a pretty common lab experiment in the undergraduate organic chemistry lab, including in courses I teach. I would love to see a reference for this cation so that I can share it with my students. – Ben Norris Aug 6 '16 at 3:16
• The tri para amino derivative is called crystal violet a well known dye. – porphyrin Aug 6 '16 at 7:50
• the cation you've drawn is bright yellow, see wiki for the corresponding alcohol – permeakra Aug 6 '16 at 19:58
• @BenNorris exactly, every other student in my lab observed similar color array and its described in other protocols for synthesis of triphenylmethanol I found online. Neither the TA's nor professor could pinpoint the exact cause of all the short-lived color change. – Nova Aug 6 '16 at 20:46
• @permeakra, yes, you're right, I didn't mean the cation itself as is, it would need an EDG to give more of a blue hue in low concentrations, or shift towards blue when "stacked" with, say, benzophenone or other aromatix. – ChemistryHelpCenter Aug 7 '16 at 5:05

The colour is because of electron delocalisation over the three rings rather than just one because the cation is planar. (see structure in answer from @ChemistryHelpCenter). Think of particle on a ring, bigger ring , lower energy, thus the absorption of the compound changes from being in the uv (only benzene rings delocalised ) to all three rings, and absorption in the visible. The final product will have a more or less tetrahedral structure so this delocalisation is lost and the colour is again in the uv and outside out visual perception.

• Well, maybe it's again nitpicking, but trityl isn't planar, or one may say only as planar as it can get before steric repulsion of ortho hydrogens blocks further planarisation. – Mithoron Aug 6 '16 at 13:02
• @Mithoron , yes not quite planar as you say, so I should have been more exact in my description. The rings are twisted rather like propellor blades, but nevertheless the delocalisation produces the colour as it were. In fact, the photophysics of these compounds is very interesting showing very non-exponential fluorescence decays and solvent dependent dynamics. – porphyrin Aug 7 '16 at 7:31