SmI2, "A Reagent Reborn" Notes from a mini-review by Procter

Citation:

Mansell, J. I.; Romano, C.; Procter, D. J. Contemporary Strategies in SmI2 Catalysis: A Reagent Reborn. Angew. Chem. Int. Ed. 2025, e202519678. 

Been trying to write this post for a few weeks but stuck on writer's block. 

Summary Figure: 

 

Introduction 

While you wouldn't necessarily expect one particular lanthanide to be a workhorse organic reagent, samarium diiodide (SmI2) has become one of the most common synthetic reagents for a number of radical-related reactions, ranging from reductions of carbonyls or alkyl halides to more complicated bond forming reactions. SmI2 is fairly simple to make (which is good, because it often performs better when freshly prepared) and can be used catalytically or stoichiometrically. Most reactions are done stoichiometrically (SmI2 is cheap enough), but methods that use SmI2 catalytically are interesting both from a sustainability perspective and a reactivity perspective- how do you regenerate your Sm2+ from Sm3+? This sort of problem has wide ranging implications for other types of reagents. 

Background

Samarium, as a lanthanide, has many electrons in its f-orbitals that remain unpaired. SmI2, the most common reagent for these transformations, is in the 2+ oxidation state, and its electronic structure is 4f6, so there are 6 unpaired electrons that are easily able to do radical reactions. For the most part, this means the one-electron single electron transfer (SET) from Sm(II) to Sm(III) to reduce a substrate is viable. 

 Above are two drawn out examples of this reduction. In the first, SmI2 can reduce an C=O bond, forming a ketyl radical. This can either be quenched (resulting in the reduced product) or coupled with another radical, enabling Barbier type reactions or pinacol type couplings. In the second, SmI2 can reduce an alkyl halide, forming an alkyl radical. This can then do all sorts of common alkyl radical reactions, but in my opinion is most commonly used for intramolecular cyclizations. 

For a good (but slightly outdated) review on SmI2 reactivity, I recommend this review by Nicolau: 

https://onlinelibrary.wiley.com/doi/abs/10.1002/anie.200902151 

How does it work: How do you use SmI2 catalytically?

In the above reactions, once you make Sm(III)X3, the reagent is done. However, the reduction of Sm(III) to Sm(II) is possible, so how do you reduce it while maintaining the same reactivity and functional group compatibility? You can't just throw another reductant at it, because what if that reductant reacts with the substrate?

Here are the classifications the authors make:

1. Use a (different) stoichiometric metal as a reductant. Some of the early examples are Zn(Hg), also called zinc/mercury amalgam, and another example uses 'mischmetal', which is apparently an indiscriminate mixture of metallic early lanthanides and one of the coolest names I have heard. More modern examples use Mg(0) and TMSCl, although the authors note that there are a number of subtleties to the system that make it complex. For example, over the course of the reaction the SmI2 becomes SmCl2, which is a more potent reductant. 

 

2.  Electrochemical reduction. As always, electrochemical reduction is simple on the surface but comes with a host of challenges: how do you align the kinetics of competing redox processes? For example, it is possible for your substrate to be directly reduced by the electrode before SmI2 can reduce it. One benefit to this is that you do not need to add "SmI2" to your reaction mixture; in some cases Sm(3+)I3 is used as the inital reagent, which is reduced in the electrochemical cell. Another option is to use a Sm(0) cathode, which can produce Sm(2+) in situ. Divided cell systems have also been commonly used to control the movement and activity of the SmI2. 

 

3. Photochemical reduction. Using a ligand system for the samarium, you can bind a photosensitizer to the samarium. Often, this binds Sm(III), not Sm(II). Then, after the Sm(III) complex absorbs an electron, you can move an electron from the ligand to the samarium via ligand-to-metal charge transfer (LMCT). This generates an oxidized ligand radical (cationic) and a reduced metal radical (the active Sm(II)). The oxidized ligand complex can then oxidize some sacrificial organic reductant, like a tertiary amine. 

 

4. 'Electron Recycling'. The SET reduction of C=O bonds to ketyl radicals are reversible. If you design the reaction to generate a ketyl radical, then do some downstream chemistry (for example, beta-scission), and then end up with another ketyl radical, that could reduce the Sm(III) back to Sm(II) and render the process catalytic. The authors report some of their own work in this category, doing ring opening/ring expansion of cyclopropyl ketones. 

 

Key Takeaways:

As much as this is an underdeveloped field (SmI2 is fairly cheap and easy to make, so most reactions just use it stoichiometrically), I think the ideas behind how to render it catalytic can be applied elsewhere. If your catalyst needs an electron to turn over the cycle, how do you deliver it? What considerations are needed? What side reactions are possible? I find the authors do a good job delineating some of these strategies.