The selective functionalization of complex molecules, particularly important in the field of drug discovery, poses a persistent challenge in synthetic chemistry. The precise manipulation of individual atoms within the core molecular skeleton — whether through insertion, deletion, or replacement— is an emerging and powerful tool in the arsenal of organic chemists, which simplifies the synthesis of complex molecular architectures by reducing the often-lengthy synthetic procedures1,2. Among the different skeletal editing approaches, the insertion of a single carbon atom into a molecular scaffold holds significant promise as it can facilitate a smooth incorporation of desired functional groups3. Dating back to 1881, the Ciamician–Dennstedt rearrangement stands as one of the earliest examples, yet contemporary methodologies for inserting single carbon atoms into aromatic ring systems are still limited4. Now, writing in Nature Catalysis, Glorius, Gutierrez and colleagues report a mild and versatile synthetic methodology enabling the synthesis of 2-functionalized naphthalenes from corresponding indenes, through a photocatalysed ring expansion mechanism5 (Fig. 1a).

Fig. 1: Photoredox indene skeletal editing via single-carbon atom insertion.
figure 1

a, General reaction conditions for the synthesis of 2-functionalized naphthalenes. b, Predicted mechanism highlighting the key intermediates in the reaction pathway. c, Examples of the reaction scope. d, Photoredox single-carbon atom insertion process applied to drug molecules. LED, light-emitting diode; Ar, aryl; Alk, alkyl; OTf, trifluoromethanesulfonate; SET, single electron transfer.

Building upon Suero’s research6, the team directed their attention to the utilization of α-iodonium diazo compounds as a source for masked carbynyl radical species, thereby offering a strategic approach for accessing single carbon atoms to be inserted into the indene core. Indeed, other precursors for carbyne species were initially tested but found unsatisfactory. The key design of the presented methodology consists in taming the reactivity of the iodonium diazo compounds with a photocatalyst, Ru(dtbbpy)3(PF6)2, and an inorganic base, Na2CO3, under visible light irradiation, achieving the selective formation of a single reactive carbynyl intermediate among other possible intermediates — carbon radical, carbene, and free carbyne species — by modulating the relative rates of N2 loss and radical generation. The combined use of Stern–Volmer luminescence quenching analysis and cyclic voltammetry experiments revealed that the highly reducing power of the photocatalyst excited state *[Ru]II (E1/2(ii)*/(iii) = –0.81 V) enables a single-electron transfer (SET) event, effectively reducing the hypervalent iodonium diazo compound (Ered = –0.46 V) and generating the corresponding carbon-centred radical (Fig. 1b; ∙C(N2)–R). This key intermediate was indirectly probed through radical trap experiments, allowing for the detection of corresponding intermediates. DFT calculations supported the findings by confirming that, following the SET process, the favourable generation of the C-centred radical occurs through the homolytic III[I]–C bond cleavage. The most likely calculated pathway involves the regioselective addition of the C-centered radical into the indene substrate 1, forming the indene radical (Fig. 1b; 2), which is subsequently oxidized to generate a corresponding cationic species 3 by the initial α-iodonium diazo compound. Thus, formation of a bicyclic intermediate 4 followed by nitrogen extrusion results in C–C bond cleavage, leading to the final product 5.

The methodology reported by Glorius, Gutierrez and co-workers enables the creation of a wide library of 2-naphthalene compounds featuring a diverse array of functional groups. Indeed, 2-ethyl naphthoate derivatives adorned with electron-withdrawing substituents as well as electron-donating groups are prepared in moderate to good yields (Fig. 1c; 6,7). The methodology also demonstrates efficiency in transforming indenes bearing heterocyclic rings, such as thiophene and pyridine, or aliphatic chains, into their respective ring-expanded products (Fig. 1c; 810). These examples underscore the significance of the approach in facilitating the seamless preparation of varied products characterized by distinct structural and electronic properties. The mild reaction conditions coupled with the operational simplicity of the developed protocol allowed the team to also employ a variety of different α-iodonium diazo compounds, successfully incorporating various synthetically useful functional groups into the indene skeleton, such as phosphate ester, sulfone, amide and trifluoromethyl groups (Fig. 1c; 1114). Particularly, the preparation of trifluoromethyl-substituted compounds is an important task in drug discovery, due to the ability of this group to modulate the physicochemical properties and binding affinity of the molecule in which it is contained.

The versatility and robustness of the synthetic approach is then applied to obtaining polycyclic compounds and pharmaceuticals: in a single step one can introduce a single carbon atom, accessing naphthalene derivatives — common motifs extensively explored in medicinal chemistry — for example, modifying the aldosterone inhibitor’s core structure 15 to the corresponding naphthalene analogue 16, or preparing the drug molecule Adapalene ester 18 (Fig. 1d). These applied examples open the way to the precise and easy manipulation of molecules, enabling the quick and concise exploration of chemical space through skeletal editing approach.

Despite its synthetic potential, it is important to point out some limitations arising from the specifics of the reaction pathway. For example, while 2-substituted indenes underwent successful conversion into the desired naphthalenes, 3-substituted indenes provided products in low yields. This behaviour could be attributed to the steric hindrance caused by the group at position 3, which could prevent the intramolecular addition of the intermediate cation to the carbene hindering the cyclopropanation step. Additionally, the reaction is inefficient with indoles, hampering the possibility to access quinoline products via ring expansion protocol. Although this limits the application of this protocol, numerous other works in the literature have already disclosed clever techniques for the incorporation of heteroatoms, notably nitrogen7.

However, the ideal synthetic scenario to extend the described reactivity towards the effective functionalization of both carbo- and heterocyclic systems, regardless of their substitution patterns, although hard, does not seem unrealistic given the series of methods being reported in the literature. Certainly, a good starting point might be the further exploration of the sources of the carbynyl species compatible with a larger array of starting materials. In addition, the crafting of simple and eco-friendly carbon surrogates with the capacity to mitigate waste and by-products is of significant interest. In the current landscape, scientists and researchers are actively engaged in the exploration and establishment of synthetic routes characterized by sustainability and high atom economy.

Unarguably, there has always been an element of aesthetics and beauty in organic synthesis: consequently, minimalistic protocols that open access to valuable organic molecules are highly appreciated by the overall community. This strategy is a good example of clever skeletal editing and hopefully will serve as food for thought for the research groups that are currently dealing with the topic.