News & Views | Published:

Organic chemistry

A radical approach to diversity

Nature volume 480, pages 184185 (08 December 2011) | Download Citation

Compounds containing the trifluoromethyl group have many uses, but their isomers must often be made using different multi-step routes. Two studies now show how several isomers can be made by the same route. See Letter p.224

The incorporation of fluorine atoms into organic molecules has had a profound impact on the development of new products for numerous industries. This is largely due to the useful properties that fluorine imparts to organic molecules, such as stability and lipophilicity, and the fact that it can modify molecular polarity. Among its many applications, fluorine incorporation has been used in pharmaceutical development to prevent molecules from being metabolized too quickly, thereby allowing a drug to act before it is cleared from the body1. Fluorine has also played an integral part in the development of agrochemicals and high-performance materials2. Two papers, one by Ji et al.3 in Proceedings of the National Academy of Sciences and the other by Nagib and MacMillan4 on page 224 of this issue, now report innovative methods for introducing an important fluorinated group — the trifluoromethyl group, CF3 — into molecules.

Appending fluorinated groups to small organic molecules has long posed a considerable challenge to synthetic organic chemists. The trifluoromethyl group has been particularly difficult to install, in part because the reactive intermediates that are generated during trifluoromethylation reactions are unstable under the conditions necessary for the reactions to proceed. The harsh protocols typically required for these reactions can limit the substrates that can be used and/or cause side-product formation.

Reports highlighting efforts to use milder methods that avoid the common problems associated with trifluoromethylation chemistry have therefore recently garnered great attention5. Although much progress has been made, many methods continue to rely on the use of 'pre-functionalized' reactants containing chemical groups that, in conjunction with transition-metal catalysts or a promoter reagent, direct trifluoromethylations to particular sites in the reactant (Fig. 1a). Alternatively, structurally complex trifluoromethyl-containing molecules can be made from simpler starting materials that already contain a trifluoromethyl group (Fig. 1b). But using these methods comes at a price: the preparation of pre-functionalized and trifluoromethyl-containing starting materials is time-consuming and costly in terms of chemicals used and waste disposal, and overall is inefficient. Furthermore, when access to a diverse range of trifluoromethylated molecules is desired, each molecule requires its own pre-functionalized starting material to be prepared, amplifying the inefficiencies so that they become increasingly prohibitive.

Figure 1: Approaches for making trifluoromethylated compounds.
Figure 1

a, One traditional approach for making compounds appended with trifluoromethyl (CF3) groups involves making a precursor of each isomer that contains a surrogate group (X), which is then converted into a trifluoromethyl group at the end of the synthesis. A different multi-step synthesis is required to make each isomer; only one such route is shown. b, The second traditional approach starts from compounds that already contain a trifluoromethyl group. Again, a different multi-step synthesis is required to make each isomer, only one of which is shown. c, Two papers3,4 now report a different approach: a structurally complex molecule is made, which is then converted into several isomers of the trifluoromethylated analogue in a single step. This strategy reduces the time and labour required to make several trifluoromethylated analogues of structurally complex molecules, and minimizes the overall waste produced.

Despite their drawbacks, existing chemoselective (site-selective) methods are often useful when the best trifluoromethylated molecule for a given application is known. But what happens when the optimal site for trifluoromethylation in a molecule is not known, as is often the case for those working in fields such as drug discovery and agrochemical research? In many instances, it is desirable to append trifluoromethyl groups to a molecule at as many positions as possible, to make lots of different isomers. This approach maximizes the diversity of the molecules obtained and increases the probability of discovering a molecule that has desirable chemical properties and biological activity.

The reports by Ji et al.3 and Nagib and MacMillan4 highlight their efforts to develop versatile methods for the rapid preparation of a diverse range of trifluoromethyl-containing molecules. Both groups sought a general protocol that provides access to trifluoromethylated molecules without the need for pre-functionalized substrates. They targeted a chemical transformation, termed C–H trifluoromethylation, in which the hydrogen of a carbon–hydrogen (C–H) bond is replaced with a trifluoromethyl group. Although ambitious, this approach would provide a large payoff in efficiency by allowing trifluoromethyl groups to be installed at late stages of synthetic routes, without first preparing pre-functionalized or trifluoromethyl-containing substrates. It would therefore decrease the number of synthetic operations necessary to access a variety of trifluoromethylated molecules (Fig. 1c).

The authors3,4 recognized that the development of a successful method hinged on identifying a practical trifluoromethyl source. Both groups independently surmised that the reliable generation of highly reactive trifluoromethyl radicals (·CF3) as intermediates might allow the direct trifluoromethylation of non-functionalized substrates, in what Ji et al.3 call an “innate trifluoromethylation”. In other words, the inherent reactivity of the substrates would dictate the site (or sites) at which the trifluoromethylation reaction occurs.

The two groups pursued different approaches to develop their own mild, efficient methods to access reactive trifluoromethyl-radical intermediates. Ji et al. developed a transition-metal-free approach that relies solely on air- and moisture-stable reagents (which are easier to handle than commonly used organometallic reagents and transition-metal catalysts that react with air and moisture). They determined that the Langlois reagent6 (NaSO2CF3) serves as a reliable source of ·CF3 when treated with t-butylhydroperoxide (a commonly used reagent in radical chemistry). Nagib and MacMillan4, however, chose triflyl chloride (ClSO2CF3) as their source of ·CF3 (refs 7, 8). They generated the radicals simply by shining light from a household fluorescent bulb onto a solution of triflyl chloride in the presence of a light-activated ruthenium catalyst. Both methods3,4 generated ·CF3 in the presence of a diverse range of substrates, and provided access to an array of structurally complex trifluoromethylated molecules. Furthermore, both methods worked reliably even for substrates bearing potentially sensitive functional groups (those whose presence in a molecule often adversely affects the course of a range of chemical reactions, including traditional trifluoromethylation reactions).

To demonstrate the versatility of their methods, both groups conducted trifluoromethylations of common pharmaceuticals and natural products. Ji et al.3 successfully achieved C–H trifluoromethylations of the natural-product derivative dihydroquinine and the billion-dollar drug Chantix (varenicline, used to combat nicotine addiction), providing products bearing a single trifluoromethyl group. These examples are significant because accessing CF3–dihydroquinine and CF3–Chantix using previously available trifluoromethylation methods would require laborious, multi-step total syntheses. Notably, the reactions were chemoselective, even though the starting materials contained multiple reactive sites.

Similarly, Nagib and MacMillan4 subjected the cholesterol-lowering drug Lipitor (atorvastatin) to their reaction conditions. This experiment resulted in the non-selective formation of three chromatographically separable isomers of CF3–Lipitor, in which the trifluoromethyl group was appended to different carbons in the molecule. The improved efficiency of this approach for accessing new chemical entities cannot be overstated, given that preparation of these isomers using traditional methods would require three parallel total syntheses, which would be extremely time- and labour-intensive.

Although the new methods3,4 constitute a major contribution to the field of trifluoromethylation chemistry, many challenges remain. For example, it would be desirable to have precise control over the site at which C–H trifluoromethylation occurs. This might be accomplished through the development of reaction conditions that combine the best attributes of the currently available methods. Complementary protocols that selectively produce different trifluoromethyl-containing molecules from a single substrate would also be advantageous, because it would avoid the often difficult task of separating mixtures of isomers generated as products. Ji et al.3 report their initial efforts towards addressing this problem, and their preliminary results are encouraging. This area of research will continue to be a major focus in organic chemistry, and future strategies for trifluoromethylation will certainly be influenced by the current reports3,4.

In the shorter term, these breakthroughs open the door to the preparation of a diverse range of molecules that would previously have been difficult to access. The ability of these streamlined approaches to perform late-stage trifluoromethylations on structurally complex molecules will undoubtedly have an impact on the work of research chemists in the pharmaceutical, agrochemical and other industries.


  1. 1.

    , & Science 317, 1881–1886 (2007).

  2. 2.

    Organofluorine Compounds: Chemistry and Applications (Springer, 2000).

  3. 3.

    et al. Proc. Natl Acad. Sci. USA 108, 14411–14415 (2011).

  4. 4.

    & Nature 480, 224–228 (2011).

  5. 5.

    & Chem. Rev. 111, 4475–4521 (2011).

  6. 6.

    , & Tetrahedron Lett. 32, 7525–7528 (1991).

  7. 7.

    , & Chem. Lett. 19, 649–650 (1990).

  8. 8.

    , , , & J. Chem. Soc. Perkin Trans. 1 1339–1346 (1994).

Download references

Author information


  1. Andrew T. Parsons and Stephen L. Buchwald are in the Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.

    • Andrew T. Parsons
    •  & Stephen L. Buchwald


  1. Search for Andrew T. Parsons in:

  2. Search for Stephen L. Buchwald in:

Competing interests

S.L.B. is a consultant for a company partially owned by David MacMillan. S.L.B. and David MacMillan co-teach a short course that they have given together at pharmaceutical companies.

Corresponding authors

Correspondence to Andrew T. Parsons or Stephen L. Buchwald.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing