Organocuprates (Gilman Reagents): How They’re Made
Last updated: January 24th, 2020 |
How Gilman Reagents (Organocuprates) Are Made
Here’s the summary for today’s post:
Table of Contents
- Organometallics: What About The Rest Of The Periodic Table?
- Can We Make Other Organometallics The Same Way We Make Grignards (And Organolithiums)?
- Formation of Organometallics From Alkyl Halides Involves Reduction
- Direct Formation of Organometallics From Cu Is Much More Difficult Than From Mg or Li
- A Work-Around: Transmetallation
- OrganoCOPPER reagents: Way Less Reactive Than Grignards
- OrganoCUPRATES: The Much-More-Reactive Cousins of OrganoCOPPER Reagents
- Summary: Organocuprates
In this whole series on organometallics so far has covered exactly TWO metals: lithium and magnesium (with a very brief head-nod to sodium and the comparatively useless
Wurzt Wurtz reaction.)
There are dozens of other metals on the periodic table. What about organometallic compounds of them ?
Great question! (That’s also the first thing you say when someone asks you a question you want to dodge.) Much as we’d love to, we just don’t have enough time to get into the full glory of organometallic chemistry in an introductory course. (If you’re curious, may I suggest The Organometallic Reader?)
In this post and the next we’ll give perfunctory treatment of some organometallic compounds of copper, and perhaps in a later post have something cranky to say about palladium, and call it a series.
Let’s get started.
First, recall what we’ve said so far about how to make organometallic compounds: basically, take an organohalide, add magnesium or lithium, and stir.
You might wonder: is it this easy for making other organometallics?
The short answer is: for most metals, it doesn’t work nearly as well.
Remember that formation of organometallic compounds from organohalides is a reduction reaction. Here it’s illustrated for the synthesis of organolithium compounds.
Notice how the polarity on the carbon changed from positive to negative? Not that we normally do such things, but if you keep track of the carbon oxidation state, you’d see that it changed from (–2) to (–4).
This works well for lithium and magnesium because those metals are so easily oxidized. We can quantify this by looking at a table of oxidation potentials.
See how lithium, sodium, and magnesium are near the top of this table of oxidation potentials for metals? They’re easily oxidized – which means they are extremely strong reductants.
[note that these voltages are in aqueous solution, so take these numbers with a grain of salt for reactions performed in organic solvents].
As we move from lithium down the table to metals like nickel and copper, we should expect the reduction reaction should become progressively more difficult. And it is.
In other words, direct formation of an organometallic from the organohalide and that metal [a process we call ‘insertion’ , or “oxidative addition”, FYI] is less favoured.
Look at copper for instance: the “direct reduction” doesn’t work nearly as well as it does for Li and Mg.
I don’t mean to imply that it can’t be done, but it generally requires heat and the addition of extra reagents that influence the oxidation potential of the metal we call “ligands” ( not going to get into that now) in order to get this reaction to go.
Let’s say we really need to make an organometallic compound of copper. Can we get around this difficulty we generally experience in direct reduction? Yes – there’s a workaround.
What we can do instead is to start with a pre-made organometallic (such as an organolithium or Grignard reagent) where the reduction has already occurred. We can then add a copper (I) salt such as CuBr or CuCl.
The result is displacement of the halide at copper by the carbon bound to lithium, and we form an organocopper reagent (plus a lithium salt). [You might wonder: why does this work at all? See note. ]
Yahoo for shortcuts!!
(Although not shown, this also works for Grignard reagents)
Now that we have a way to make organocopper reagents, the next question is: so what? What can we do with them?
In comparison to Grignard and organolithium reagents which are violently destroyed by water (and sometimes air) organocopper reagents are fairly sedate. For instance, there are organocopper reagents that you can leave out in the air without incident, like 1-hexynylcopper.
Also, organocopper reagents don’t add to carbonyls or epoxides like organolithium reagents or Grignard reagents do.
While organocopper reagents are interesting, it turns out that they have a chemical “cousin” that is even more versatile, reactive, and useful, and for that reason we will from this point further focus on these species: “organocuprate” reagents.
In the 1940’s Iowa chemist Henry Gilman discovered that adding one further equivalent of an organolithium reagent to an organocopper compound resulted in an “organocuprate” reagent, with two Cu–C bonds and is also comprised of a positive counter-ion [lithium in this case].
[Note that “-ate” at the end. Notice how many species that end in the name “ate” [e.g. sulfate, nitrate, tosylate] are negatively charged?]
Organocuprates, with the general formula R2CuLi , have the same general pattern of reactivity as organocopper reagents, but are much more reactive. These compounds are commonly referred to as “Gilman reagents” in ol’ H.G.’s honour. [Why are they more reactive? We’ll talk about that in the next post].
Just like organocopper reagents (and in contrast to Grignards) organocuprates do not generally add to aldehydes, ketones, or esters.
However, as we’ll see in the next post, they do participate in substitution and “conjugate addition” reactions – reactions that Grignards and organolithiums reagents typically don’t do. In this way (as we’ll see) they have a somewhat complimentary function.
Let’s sum up. We saw that we generally don’t make organocopper reagents directly, but via organolithium or Grignard reagents. However, there’s an even more reactive “cousin” of organocopper reagents – “organocuprates” – that we can also make if we use a 2:1 ratio of organolithium to copper.
Note again: you need two equivalents of organolithium for every equivalent of copper.
Also note that you can use various different Cu(I) salts, as well as Grignard reagents.
So what’s so great about Gilman reagents, and what can they be used for?
We’ll talk about that in the next post. See you next time!
Bonus question for today. What’s the oxidation state of Cu in Gilman reagents?
Note 1. Why does transmetallation from lithium (or magnesium) to copper work? It’s a good question. In a nutshell, the carbon-copper bond is stronger than the C-Li or C-Mg bond, and that provides the driving force.
Why is the C-Cu bond stronger than C-Li ? That’s a difficult question to answer succinctly, but I’ll try.
You can think of bonding as having two components:
- an electrostatic component, where there is attraction between opposite charges (such as in salts)
- the overlap of molecular orbitals, resulting in shared electron density between atoms.
Bonding between a carbanion R– with Li+ is almost purely ionic (note the electronegativity difference of about 1.5) meaning that a significant portion of the bonding interaction is due to electrostatic interactions.
In contrast, bonding between C and Cu is considerably more covalent. There’s less of an electronegativity difference (Cu 1.9 vs. C 2.5) and the bonding is better described by metal-carbon bond overlap.
[waves hands] Generally, carbon forms stronger bonds through orbital overlap than via electrostatic interactions. This is largely because carbon is significantly more polarizable (the negative charge is “squishy”, or “diffuse” – we often use the shorthand “soft” to describe this behaviour) than a generally non-polarizable (“hard”) atom like lithium or oxygen. Likewise copper is also a fairly “soft” atom.
If you want to get super technical, for C-Cu the covalent term in the Klopman equation is large, and the coulombic term is small.