When all C–C breaks LO–Ose

Abstract

Organic peroxides are becoming popular intermediates for novel chemical transformations. The weak O–O bond is readily reduced by transition metals, including iron and copper, to initiate a radical cascade process that breaks C–C bonds. Great potential exists for the rapid generation of complexity, originating from the ability to couple the resulting free radicals with a wide range of partners. First, this review article discusses the history and synthesis of organic peroxides, providing the context necessary to understand this methodology. Then, it highlights 91 examples of recent applications of the radical functionalization of C–C bonds accessed through the transition metal-mediated reduction of organic peroxides. Finally, we provide some comments about safety when working with organic peroxides.

Significance of organic peroxides

Deconstructive and editing techniques are becoming increasingly popular in organic synthesis, supplementing the constructive processes used traditionally. Among these techniques, C–H functionalization (see Glossary) is an attractive tool for late-stage diversification, converting C–H bonds into C–R bonds. In contrast, C–C activation has received much less attention, perhaps because it is counterintuitive to cleave a C–C bond when synthesizing organic molecules [1]. Nevertheless, C–C bond breaking can be a non-obvious strategy for realizing novel transformations unavailable to traditional constructive processes.

Organic peroxides are powerful precursors for C–C activation. They are readily available high-energy species primed for C–C bond cleavage through O–O scission. The weakness of the O–O bond (ca. 45 kcal/mol, 298 K) provides an opportunity for alkoxy radical formation through homolysis or reduction [2]. Transition metals are excellent mediators of peroxide reduction, producing alkoxy radicals for subsequent C–C scission. This review article showcases methods developed recently for transition metal-mediated reduction of organic peroxides that lead to the breaking and functionalization of C–C bonds. To begin, it provides a brief historical background of radical–peroxide chemistry. The syntheses of organic peroxides are then summarized, followed by contemporary examples of C–C activation. To end, safety concerns relating to peroxides are addressed and future perspectives are put forth.

Historical background

The foundation for radicals, peroxides, and their relationship with C–C scission originated with the discovery of synthetic organic peroxides, the preparation of organic radicals, elucidation of the decomposition products of organic peroxides, and the gradual discovery of metal–peroxide redox chemistry. Synthetic organic peroxides emerged in the 1850s, as did the discovery of radicals. The first synthesis of an organic peroxide occurred through the acylation of sodium peroxide with benzoyl chloride, reported in 1858 by Brodie (Figure 1) [3]. Chemists of the 1850s were also publishing articles related to a new idea: radicals [4,5]. These two concepts, radicals and peroxides, were combined and published a decade later by Brodie [6–8]. Thereafter, in 1900, Gomberg serendipitously prepared triphenylmethyl radical, the first isolation and characterization of a free radical [9]. Formal studies of the generation and reactions of free radicals had begun.

Figure 1.

Peroxide chemistry milestones. See [3,9,12,21,115].

The late 19th century was also pivotal for the chemistry of metals and peroxides. In 1868, Parnell reported the reduction of hydrogen peroxide mediated by ferrous sulfate [10]. Fenton would clarify this observation in 1894 by chemically altering tartaric acid through the actions of hydrogen peroxide and ferrous salt [11]. Then, in 1932, Häber and Weiss formalized the mechanism of ferrous salt oxidation and peroxide reduction [12]. With the discovery of organic radicals and the redox relationship between metals and peroxides established, it became possible to use metal salts as mediators for the formation of radicals from peroxides.

Between the 1930s and 1960s, an explosion of research occurred related to the chemistry of organic peroxides. Initial studies involved the thermal decomposition of peroxides. Seminal papers by Gelissen and Hermans described the role of the thermal decomposition of benzoyl peroxide in the formation of polyphenyl derivatives, work that clarified ideas put forth by Lippmann in 1886 [13–16]. The thermal decomposition of peroxides grew beyond acyl peroxides, with Milas’ 1930s series, ‘Studies in organic peroxides’, covering both thermal decomposition and peroxide synthesis [17]. From these initially studied decomposition processes, ideas emerged for a fragmentation mechanism originating from peroxides. Research by Criegee on the ozonation of alkenes to generate peroxide intermediates supplemented these investigations [18–22]. Notably, in 1946, George and Walsh explicitly reported the concept of β-scission as the mechanism of fragmentation of reduced organic peroxides, thereby formalizing the involvement of radicals in this field of chemistry [23]. In the 1950s, the groups of Raley–Rust–Vaughn; Hawkins; Kharash; and Milas pioneered the use of Fenton’s reagent and the Häber–Weiss method for peroxide decomposition, with their results reported in the published series ‘Decompositions of di-t-alkyl peroxides’, ‘Reactions of organic peroxides’, “The chemistry of hydroperoxides’, and ‘Organic peroxides’, respectively [17,24–26]. With the availability of metal salts to mediate peroxide fragmentation, it became unnecessary to heat peroxides to extreme temperatures to initiate their decomposition. From those reports, the predictability of these radical processes was established and methods to control the decomposition were explored [27–29]. The effects of various metals and solvents were also tested. Together, these discoveries laid the groundwork for the application of radicals in organic synthesis [30–33].

With the accumulated knowledge of the mechanism and fission products of peroxides, the focus of peroxide chemistry shifted toward its applications [34]. Successful efforts to trap the radical intermediates generated at the various stages of the decomposition cascade opened up new areas of research [35]. Publications from Kharasch, Kochi, and Minisci began to emerge, demonstrating the use of ferrous and cuprous salts for the reduction, fragmentation, and subsequent functionalization of organic peroxides [36]. In 1980, the utility of this methodology was realized, and popularized, by Schreiber through the total synthesis of (±)-recifeiolide. In the present day, such studies have been reclaimed, and expanded upon, to provide access to a variety of transformations designed to meet synthetic demands.

Synthesis of organic peroxides

Many standard organic precursors, including alcohols, carbonyls, epoxides, alkyl halides, and olefins, can be transformed into organic peroxides. The first synthetic organic peroxides to be generated were peresters. Initially reported by Brodie in 1858, the acylation of sodium peroxide with benzoyl chloride in water is the oldest known synthesis of an organic peroxide (Figure 2A) [3]. Since then, peresters have interested chemists for their unique stability and ease of synthesis. In the late 1940s, Milas used aqueous potassium hydroxide and a variety of acid chlorides at 10–20°C to synthesize tert-butyl peresters from tert-butyl hydroperoxide [37]. A new electrophile, the anhydride, was adopted in the early 1950s by Davies, Foster, and White for the synthesis of peresters. In particular, the reaction of phthalic anhydride and tert-butyl hydroperoxide in pyridine at room temperature yielded the mono-perester [38]. In 1963, peresters were first synthesized from carboxylic acids. Hecht and Rüchardt used N,N-carbonyl-di-imidazole to activate benzoic acid for the acylation of tert-butyl hydroperoxide [39]. More recently, Baj and Chrobok reported the alkylation of alkyl and aryl peracids, using an alkyl sulfonate or oxonium with tetrabutylammonium hydrogensulfate, to generate peresters [40]. These highlighted examples represent only a sampling of the methods available for perester synthesis; for more information, see three reviews on peresters that have been published recently [41–43].

Figure 2.

Organic peroxide synthesis. (A) Acylation. See [3,37]. See also [38–43]; not pictured. (B) Autoxidation. See [44,45,47]. See also [46]; not pictured. (C) Condensation. See [48–50]. (D) Substitution. See [52,56]. See also [38,51,53–55,57–59]; not pictured. (E) Pericyclic. See [21,62]. See also [22,60,61,63–65]; not pictured. (F) Alkene addition. See [67,68]. See also [66,69–73]; not pictured.

In the simplest case, autoxidation involves the insertion of O₂ into a C–H bond to yield an organic hydroperoxide (Figure 2B). This technique is limited to weak C–H bonds and is not widely applicable, but it can be useful in certain situations. The earliest examples of intentional autoxidation were reported simultaneously by Bach (in France) and Engler and Wild (in Germany) in 1897; they mixed carbon monoxide with oxygen over water to form percarbonic acid [44,45]. Since then, many reports of autoxidation have appeared. In 1931, Milas published a review on autoxidation, discussing its criteria, classes, and methods of study [46]. Although dated as a practical process, an understanding of autoxidation is important for peroxide synthesis and general safety. Common solvents, such as tetrahydrofuran, can autoxidize into explosive peroxides, as reported by Robertson in 1948 [47]. If the conditions allow, autoxidation can be used as a clean method to add peroxides to organic molecules.

Peroxides are readily synthesized through condensation chemistry from ketones and aldehydes (Figure 2C). They can form hemi-peracetals and hemi-perketals, or condense water to form perhydroxyhydrates. One of the first reports of this condensation chemistry was by Wolffenstein in 1895: the synthesis of triacetonetriperoxide (TATP) from acetone and hydrogen peroxide at room temperature [48]. Much more recently, in 2021, Terent’ev demonstrated a reaction similar to Wolffenstein’s, using cyclohexanone to form the hemi-perketal and perhydrate [49]. In the same year, Terent’ev also reported the use of dicarbonyl species and ammonium acetate to form endoperoxides, demonstrating the compatibility of imines with the peroxide condensation methodology [50].

Classical nucleophilic substitution methods are applicable to organic peroxide synthesis (Figure 2D). Hydroperoxides, with increased nucleophilicity, may undergo substitution reactions directly through S_N1 and S_N2 mechanisms [51]. The conditions for the S_N1 process are rather harsh and have inherent limitations. For example, primary and secondary alcohols are not very reactive in S_N1 substitution [38]. Tertiary alcohols can be treated with aqueous hydrogen peroxide to form tertiary hydroperoxides, as exemplified by tert-butanol yielding tert-butyl hydroperoxide, as reported by Milas in 1938 [52]. The use of an acid catalyst with hydrogen peroxide and tert-butanol was demonstrated by Davies and White in 1952 to yield the tert-butyl hydroperoxide [53]. A similar process has been used recently by Maruoka to generate tertiary hydroperoxides [54]. For S_N2 reactivity, anionic hydrogen peroxide is typically required. Anionic peroxides are not ideal nucleophiles for S_N2 reactions with alkyl halides because peroxides can inhibit the reaction and promote side reactions [55]. Nevertheless, such reactions are still possible. In 1940, Milas reacted sodium peroxide and nicotinyl chloride in water and diethyl ether to synthesize nicotinyl peroxide [56]. Despite the inherent limitations of substitution reactions, modern techniques allow such routes to organic peroxides that could not be achieved before. In 2013, Dussault reported the alkylation and subsequent hydrolysis of gem-dihydroperoxides with silver oxide, then treatment with refluxing sulfuric acid in methanol, to yield primary hydroperoxides [57]. Peroxides can also readily substitute other types of electrophiles. For example, epoxides can be opened by hydrogen peroxide to form trans-1,2-hydroxy hydroperoxides, as reported in 2021 by Darcel [58]. New methods using an S_H2 mechanism to form organic peroxides are being developed. Zhong published a technique in 2019 for replacing benzyl halides with hydroperoxides [59]. Using an indium mesh in air, a single electron reduction event formed a benzylic radical for capture by molecular oxygen to install the peroxide. Overall, substitution methods to generate alkyl hydroperoxides are viable synthetic options.

Pericyclic reactions can quickly generate complexity and are applicable to peroxide synthesis (Figure 2E). The oldest pericyclic reaction involving peroxide formation is ozonolysis. In the 1830s, Schönbein first speculated that ozone could react with olefins to generate organic peroxides [60]. One of the most important discoveries thereafter was made by Criegee, who, in 1949, found that a (3+2) dipolar cycloaddition and retro-(3+2) cycloaddition of ozone and 9,10-octalin in methanol formed an α-methoxyhydroperoxide [21,22]. In 1955, the capabilities of ozone and olefins for peroxide generation were expanded upon by Milas, such that Criegee intermediates have since been applied widely in synthesis, with subsequent oxidations or reductions used to formally break the σ- and π-bonds of the original olefin and convert it into a di-carbonyl [61]. Cycloadditions with singlet oxygen have been used to form endoperoxides. One example, from Pederson in 2017, involves triplet oxygen being sensitized in situ to singlet oxygen, mediated by a tetraphenylporphyrin (TPP) (or Rose Bengal) and light (100 W), for stereoselective cycloaddition with a diene [62,63]. The ene reaction adds molecular oxygen α to an olefin, thereby allowing peroxide insertion adjacent to olefins. In 2003, Foote and Houk published a Rose Bengal-sensitized ene reaction for the generation of an allylic hydroperoxide [64]. These sensitized pericyclic reactions do not typically require harsh conditions, making them ideal for peroxide installation. Organic peroxides have also been used in click reactions to add further complexity [65]. With their ability to generate complexity rapidly and their general compatibility with other functional groups, pericyclic reactions are excellent tools for synthesizing organic peroxides.

Olefins are not only useful in pericyclic peroxide-forming reactions (Figure 2F). A simpler transformation is the Markovnikov addition of a peroxide across an olefin in the presence of an acid, but this method suffers from limitations similar to those of S_N1 reactions. Nevertheless, it is still possible to achieve the Markovnikov addition of a hydrogen atom and a peroxide unit across an alkene. In 1989, Isayama and Mukaiyama reported the application of a cobalt(II) catalyst and triethylsilane under an oxygen atmosphere to achieve the novel hydrosilylperoxidation of an olefin [66]. In 2016, Shenvi published a review on this transformation, which has been applied successfully in recent years. In 2018, Woerpel used the conditions of Isayama–Mukaiyama hydrosilylperoxidation to functionalize furans with silylperoxides [67]. Consistent with a finding from Isayama and Mukaiyama, tert-butyl hydroperoxide was added to decrease the induction period for the reaction. Woerpel also demonstrated the applicability of this chemistry with enones as substrates (in 2019) and for the direct hydroperoxidation of olefins (in 2021) [68–70]. Other additions across olefins are also possible. The Li group developed a trifluoromethythiylation–peroxidation of alkenes and allenes in 2020 by using a cupric catalyst and potassium persulfate to add tert-butyl hydroperoxide and trifluoromethylthianate to an olefin [71]. Li also demonstrated the germylperoxidation of alkenes in 2022, using germyl hydride and a cuprous catalyst [72]. Furthermore, in 2020, Wu used sodium molybdate–glycine as a catalyst for hydrogen peroxide to transform allylic alcohols into cis-2,3-diol-trans-hydroperoxides in one pot [73].

C–C scission

Organic peroxides are highly energetic species that are primed for chemical reactions. As opposed to thermal or photolytic processes, catalytic quantities of transition metals, including iron and copper, can offer semi-controlled single-electron pathways through which the energy of peroxides can be harnessed for useful chemical transformations. Because of the redox capabilities of metals, these reactions do not need to be performed stoichiometrically, but rather catalytically, with two benefits. First, the side reactions of the free radicals are limited, because the quantity of radicals generated is restricted by the concentration of the reduced metal ions. Second, lower quantities of total metal are required, thereby decreasing both the expense and environmental impact. This section focuses on reports from the last 5 years regarding metal-mediated reduction of organic peroxides to achieve C–C activation for alkylation [C–C(sp³)], hydrogenation (C–H), amination (C–N), amidation (C–NHCO), azidation (C–N₃), cyanation (C–C≡N), halogenation (C–X), chalcogenation (C–O/C=O, C–S, or C–Se), olefination (C–C=C), arylation (C–Ar), alkynylation (C–C≡C), allylation (C–C–C=C), borylation (C–B), and silylation (C–Si). With such a large variety of transformations available, this methodology is compatible with most synthetic needs.

Mechanism

In general, methodologies involving the reduction of organic peroxides follow the generalized steps in Figure 3A. First, in step A, single-electron transfer (SET) from the metal to the peroxide 1 mediates the formation of the alkoxy radical 2. Subsequent β-scission, or decarboxylation with acyl peroxides, of the alkoxy radical 2 leads to the carbonyl–alkyl radical 3. The regioselectivity of the C–C cleavage is dependent on the stability of the resulting radical [74]. The alkoxide formed during this step coordinates to the oxidized metal ion. Next, in step B, a ligand exchange swaps the alkoxide for the coupling partner R³. Subsequently, in step C, the metal coordinates with the alkyl radical 3 to form the metal–alkyl complex 4. Finally, in step D, reductive elimination releases the product 5 and restarts the catalytic cycle. Some coupling partners do not participate in the ligand exchange step, but can couple directly to the alkyl radical. In these examples, the metal acts simply as a reducing agent and is used in a stoichiometric quantity.

Figure 3.

Mechanistic overview. (A) General mechanism. (B) Radical polar crossover. M = Fe(II), Cu(I), Ni(II). R¹ = H, alkyl, aryl, OR. R² = H, SiR₃, C(O)R. R³ = coupling partner. Abbreviations: L, ligand; m, n, integers; Nu, nucleophile; SET, single-electron transfer; X, leaving group.

While most of the methodologies covered in this review article follow the general mechanism, some substrates are known to participate in a radical–polar crossover mechanism (Figure 3B). In step A of this mechanism, SET from the metal reduces the peroxide 6 to an alkoxy radical 7. After β-scission, or decarboxylation with acyl peroxides, the benzylic radical intermediate 8 remains. The catalyst is regenerated in step B when the higher-oxidation-state metal formed during step A oxidizes the benzyl radical intermediate 8 to the benzyl carbocation 9, crossing over from a radical to a polar species. The fate of the carbocation intermediate 9 depends on the presence of a base or a nucleophile. In the presence of a base, the styrenyl product 10 is generated in an E1-type manner. In the presence of a nucleophile, however, the benzylic carbocation 9 is trapped to form the adduct 11. Of note, it is possible that the radical intermediate 8 undergoes radical coupling (e.g., addition to an alkene) prior to oxidation in step B.

Alkylation

Reactions that break and build C–C σ-bonds in one-pot are powerful. As early as 1950, Hawkins noted that alkyl radicals obtained through peroxide degradation could be trapped by carbon species to form new C–C bonds [75]. Upon addition of ferrous sulfate to a methylcycloalkyl hydroperoxide, with no designated radical trap, two separate alkyl radical species dimerized to form a new C–C bond (Figure 4). Many years later, several papers appeared concerning radical alkylation from peroxides. In 2018, Bao and Li reported the use of light to activate a ruthenium complex to an excited state for the reduction of copper(II) to copper(I) [76]. This cuprous species mediates SET to the diacyl peroxide to form an alkyl radical after rapid decarboxylation. The alkyl radical undergoes carbonation with the olefin to form a benzyl radical. Through radical–polar crossover, this radical is then oxidized by ruthenium(III) to form a benzyl cation, which is trapped by nucleophilic fluoride to yield the product. In 2018, Du reported that iron(II) mediates a similar SET process with a perester, inducing alkyl radical post-decarboxylation [77]. After carbonation into a cinnamimide, the benzyl radical is trapped through cyclization to form a dihydroquinolinone radical species. With oxidation induced by iron(III) and base in radical–polar crossover manner, a neutral dihydroquinoline remains. Also in 2018, Sodeoka reported using copper with bis (chlorodifluoroacetyl) peroxides to form allyl chlorodifluoromethyl groups and β-chlorodifluoromethyl aziridines [78]. The proposed mechanism involves a reduction event followed by decarboxylation to yield a chlorodifluoromethyl radical, which adds into an alkene. Subsequent oxidation of the resulting radical returns the copper(I) catalyst and generates a carbocation. Basic conditions promote the formation of allyl chlorodifluoromethyl groups upon deprotonation. Alternatively, aziridines may be formed through addition of a neighboring amine to the carbocation, exemplifying the great synthetic utility of this process.

Figure 4.

Alkylation. *Only the major β-product is shown, the δ-adduct is minor. See [75–78,82–84,90,101]. See also [79–81,85–88]; not pictured.

In 2018, the Zhu group reported five different types of radical methylation. The first was termed alkoxy methylation [79]. A copper(I) species reduces dicumyl peroxide, producing methyl radicals. This radical adds into a styrene derivative to form a benzyl radical. The resulting copper(II) species oxidizes the benzyl radical to a benzyl carbocation through a radical–polar crossover, with subsequent trapping by a nucleophilic alcohol. The second method uses the same strategy, except that the resulting benzyl carbocation is trapped by lithium azide, achieving azido-methylation. The final three methods are intramolecular ring closing strategies for forming cycloethers, lactones, and cycloamines, where the benzyl carbocation is trapped by alkoxides, carboxylates, and sulfonamides, respectively.

The Bao group, in 2020, reported the use of catalytic iron(II) to methylate 1,1-disubstituted alkenes [80]. The methyl radical, which arises from lauroyl peroxide through β-scission, adds into the terminal side of a 1,1-disubstituted alkene that contains an aryl group and an electron-withdrawing group. Coupling of the resulting benzyl radical to form the azido product is mediated by an iron(III) azido species generated in situ through ligand exchange. In the same year, Bao and Li reported the alkylarylation of styrenes [81]. In these cases, catalytic amounts of a cuprous species were used to reduce the diacyl peroxide to an alkyl radical through β-scission. Similarly, the radical adds into the alkene to form the benzyl radical. Meanwhile, the copper(II) species undergoes ligand exchange with the arylboronic acids and, through the effects of copper(III) and reductive elimination, gives the geminal-diaryl product. Also in 2020, the Guo group reported the iron(II) triflate-mediated opening of cyclopentyl silylperoxides to alkyl radicals. These radicals were, again, trapped by various styrene derivatives for alkylation [82]. In this case, the benzyl radical was oxidized to a benzyl carbocation and trapped by an alkoxy group. Finally, the Gui group demonstrated that radical alkylation could be achieved through intermolecular reductive radical addition when using iron(II) sulfate heptahydrate [83]. For their reaction, singlet oxygen generated from photosensitization with methylene blue (MB) engages in a (4+2) cycloaddition with a furan to form an endoperoxide. In the presence of methanol, a simple substitution yields the hydroperoxide, which serves as the reactive species for reduction. The Gui group used these oxidation conditions for a few other types of reductive functionalizations.

From 2020 to 2022, the Maruoka group released several papers concerning radical alkylation. Their first publication was related to the report from Bao in 2020 regarding the alkyl radical from β-scission being trapped by a styrene, with ligand exchange between the copper(II) species and arylboronic acid resulting in radical alkylation and copper-mediated coupling [84]. The difference was that the Maruoka group used cycloalkyl silylperoxides, rather than diacyl peroxides, with their method being the first instance of a catalyst-controlled enantioselective radical addition. Two other papers featured cycloalkyl silylperoxides. One report, from Kato and Maruoka in 2021, described the use of perfluoroalkyltrimethylsilanes and silyl enol ethers (both cyclic and acyclic) as radical traps [85]. Interestingly, the reactions that used the silyl enol ethers as radical traps required the presence of potassium fluoride and 18-crown-6 to avoid significant side reactions. Also in 2021, Kato, Liu, and Maruoka reported twice that methylene activated by two electron-withdrawing groups could be used to trap alkyl radicals when using a copper(I) catalyst [86,87].

To complete the cases from 2021, the Sodeoka group provided two reports demonstrating perfluoroalkylation. The first is an example of mono-perfluoroalkylation of an alkene with a copper(I) salt to form an allylic perfluoroalkyl group [88]. The perfluoroalkyl radical originates from a perfluoroalkyl anhydride that is transformed into an organic peroxide using urea and hydrogen peroxide. Their subsequent example, from Kawamura and Sodeoka, is similar in that it also uses a perfluoroalkyl anhydride with urea and hydrogen peroxide to form the organic peroxide and uses a copper(I) salt to catalyze radical formation. The difference lies not only in the addition of a 2,2′-bipyridine (bpy) ligand to promote bis-perfluroalkylation, instead of mono-perfluoroalkylation, of alkenes and alkynes but also in a transition from catalytic to stoichiometric copper loading [89]. Moving into 2022, Kato, Liu, and Maruoka reported the use of alkyl radicals that were trapped by styrenes [90]. Upon regeneration of iron(II) through radical–polar crossover by promoting the benzyl radical to a benzyl carbocation, the reaction was completed through trapping with anionic β-keto carbonyl derivatives. Finally, the most recent report from Kato, Liu, and Maruoka demonstrated that the alkyl radical from a cycloalkyl silylperoxide and an iron(II) catalyst could be trapped by a dienone [91]. Because nucleophiles could be trapped on the allylic carbocations, it was proposed that the alkyl radical is oxidized to a carbocation in the manner of a radical–polar crossover.

Hydrogenation

Replacement of a C–C bond with a C–H bond is a common functionalization option. The first reports of radical hydrogenation through peroxide degradation were provided by Kharasch and Hawkins, both in 1950. Kharasch reported that the reduction of cumyl hydroperoxide by iron(II) releases a methyl radical, which could be trapped by a hydrogen atom from dextrose to form methane gas (Figure 5) [92]. Hawkins discovered another hydrogenation product when conducting experiments related to the dimerization mentioned in the ‘Alkylation’ section [93]. When reducing cyclopentyl methyl hydroperoxide with ferrous salt, one of the products purified through distillation was 2-hexanone. These examples of hydrogenation were not intended products.

Figure 5.

Hydrogenation. See [83,92–94].

Almost 70 years later, deliberate hydrogenation was achieved by the Kwon group. In 2019, they realized that the Criegee intermediate from ozonolysis could be trapped with methanol and used as the peroxide for hydrogenation in an iron(II)-mediated process. By bubbling ozone into a methanol solution containing a terpenoid at −78°C, the Criegee hydroperoxide forms; with immediate addition of an iron(II) salt, alkyl radicals form and are trapped by hydrogen atom donors. The first in this series of reports described hydrogenation, where benzenethiol was selected as the hydrogen atom donor. The power of this type of process is highlighted by the synthesis of (−)-2-pupukeanone, where a dealkenylated [2.2.2]-bicyclic ketoester was achieved in one step, in contrast to the six steps required previously [94]. Finally, Gui demonstrated that hydrogenation was possible in butenolide formation when using iron(II) sulfate heptahydrate and a mercaptan as the hydrogen donor [83].

Nitrogenation

Methods for direct coupling to nitrogen atoms are essential for chemical synthesis, particularly because nitrogen is one of the six most common elements for life on Earth and because over 75% of medicines today feature nitrogen atoms. One of the first uses of nitrogen as a radicophile for radical peroxide degradation was reported by Minisci in 1958 (Figure 6A). When a cyclohexyl hydroxyhydroperoxide was reacted with ferrous sulfate, the alkyl radical from β-scission was trapped by nitric oxide, forming an alkyl nitric oxide intermediate. With water as a solvent and an excess of nitric oxide, the alkyl nitric oxide group was converted to an aldoxime in 18% yield [95]. The Maruoka group reinvestigated amination in 2019, using cycloalkyl silylperoxides, a catalytic amount of a cuprous iodide/phenanthroline complex as a reducing agent, and aniline derivatives as radical traps [96]. This approach was explored in 2020 by Gui, with furan oxidation used to form a hydroperoxide for fragmentation with iron sulfate heptahydrate and trapping with di-tert-butyl azodicarboxylate (DBAD) to generate the new C–N bond [83]. The most recent example, from Ouyang and Li in 2021, involves diacyl peroxides and a cuprous/phenanthroline catalyst producing alkyl radicals through decarboxylation for trapping by various arylamine derivatives [97].

Figure 6.

Nitrogenation. (A) Amination. (B) Amidaton. (C) Azidation. See [54,83,85,95–101].

Another class of C–N bond formation is amidation (Figure 6B). Two examples have been reported by the Maruoka group, each using a cuprous iodide-mediated process to quench alkyl radicals from both cyclic and acyclic alkyl silylperoxides with primary arylamides. In reports from 2017 and 2019, they provided many examples of amidation by varying the silylperoxide and the amide [96,98]. These experiments demonstrated the versatility of radical amidation to provide diverse products.

The final class of nitrogenation is azidation (Figure 6C). The first example was reported by Minisci in 1959, where cyclic hydroxyhydroperoxides were used with ferrous or cuprous salts and sodium azide to achieve radical azidation [99]. Azidation reappeared when the Gui group, in 2020, trapped their alkyl radical with copper(II) azide [83]. In 2021, Li and Bao demonstrated an enantioselective decarboxylative azidation involving a chiral ligand, ferrous triflate, and trimethylsilylazide [100]. Remarkably, this reaction achieved enantiomeric excesses of up to 92%. Kato and Maruoka further studied azidation in 2021, using cyclic silylperoxides. Through the standard cuprous/phenanthroline mechanism, azidation from trimethylsilyl azide was successful in trapping the alkyl radical [85]. Their next two papers on this subject, appearing in 2021 and 2022, also concerned azidation, but the starting material was a cyclic hydroperoxide that was transformed into an alkyl silylperoxide in situ from the trimethylsilylazide, which doubled as the azide source [54,101].

Cyanation

Minisci, in 1959, was the first to add nitriles to the alkyl radicals formed through transition metal-mediated reduction of hydroperoxides (Figure 7A). Here, cuprous cyanate and potassium cyanate were used to achieve this transformation [99]. This approach remained dormant for over six decades, until Gui and Kato, Liu, and Maruoka reported their studies of cyanation in 2020 and 2021, respectively. Displaying versatility for butenolide formation, the resulting alkyl radical was trapped using tosyl cyanide [83]. For Kato and Maruoka, copper was the metal of choice, with the nitrile originating from trimethylsilyl cyanide [85]. Another two papers from Kato, Liu, and Maruoka, one in 2021 and the other in 2022, used the same reagents, but with an alkyl hydroperoxide used to generate the reactive intermediate in situ [54,101]. The further application of these products, after introduction of cyanate or thiocyanate groups, appears promising.

Figure 7.

Cyanation and halogenation. (A) Cyanation. See [54,83,85,99,101]. (B) Halogenation. See [54,83,101–105].

Halogenation

Organic chlorides, bromides, and iodides participate in a variety of chemical reactions: they are good leaving groups, they can be converted into nucleophiles through metal–halogen exchange, and they are common precursors of alkyl radicals. In addition, their installation is relevant for certain medicinal purposes. In turn, the alkyl radicals generated from β-scission of alkoxy radicals are readily functionalized into alkyl halides. Minisci published seminal work in this area in 1959 and 1960, using halohydric acids as halogen pools and ferrous or cuprous salts to reduce cycloalkyl peroxides (Figure 7B) [102–104]. More recently, acids, inorganic salts, and organic halogens have been used as halogen donors, providing reactivity equivalent to that obtained through reduction of peroxides. In 2020, the Gui group demonstrated chlorination, bromination, and iodination by trapping the radicals with copper(II) chloride, N-bromosuccinimide, and N-iodosuccinimide, respectively [83]. In 2021, Duan and Guo reported hydrochloric acid, magnesium bromide, and zinc iodide as halogen sources [105]. In the same year, and in 2022, Kato, Liu, and Maruoka used triethylamine with trimethylsilyl chloride, bromide, and iodide as halogen sources to obtain the desired products from alkyl hydroperoxides [54,101]. These newer methods avoid the use of strong acids and are more compatible with labile functional groups. Halogenation in this manner offers a pathway to simultaneously install a leaving group while breaking C–C bonds.

Chalcogenation

Chalcogenation is one of the more conceptually straightforward types of reactivity stemming from peroxide fragmentation. Two of the most well-established radical trapping reagents are diphenyl disulfide and 2,2,6,6-tetramethypiperid-1-yloxy (TEMPO), both of which readily trap alkyl radicals arising from C–C scission, forming C–S and C–O bonds, respectively. Minisci published, in 1959, the first example of a carbon–chalcogen bond formed through radical peroxide degradation (Figure 8A) [106]. Here, ferrous and cuprous salts were used to promote the formation of alkyl radicals, with a thiocyanate group as the radical trap. In 2019, Kwon reported that this reactivity resulted from ferrous sulfate heptahydrate-mediated reduction of α-methoxyhydroperoxides generated by ozonolysis and trapping with diphenyl disulfide [107]. In 2020, the Gui group demonstrated fluoromethylthiylation, thiylation, and thiocyanation when using various sulfur-containing radical traps [83]. Using diphenyl disulfide, Guo reported (in 2021) that ferric tosylate could reduce tertiary hydroperoxides, achieving a similar result with ring-opening capabilities [108].

Figure 8.

Chalcogenation. (A) Thiylation. See [83,106–108]. (B). Oxygenation. See [83,96,109–113]. (C) Selenylation. See [83,108].

For C–O bond formation, the most common radicophile has been TEMPO, which is used widely as an experimental tool to prove the presence of radicals. The first instance of an alkyl radical originating from a peroxide being trapped with oxygen-containing species, aside from TEMPO, was achieved by the Maruoka group in 2019 (Figure 8B). Here, the reactivity was very similar to that in the amidation described previously by the Maruoka group, except that they used free carboxylic acids as nucleophiles, rather than primary amides [96]. The intentional use of TEMPO for C–O and C=O bond formation was reported by the Kwon group in 2020 [109]. Hydroperoxides formed through ozonolysis in methanol, followed by reduction with ferrous sulfate heptahydrate in the presence of TEMPO, created the C–O bonds. Then, under reductive (using zinc dust and acetic acid) or oxidative [with magnesium bis(monoperoxyphthalate) hexahydrate (MMPP)] conditions, the 1-hydroxy-2,2,6,6-tetramethylpiperidine (TEMPOH) unit was removed, leaving behind a hydroxyl or carbonyl group, respectively. The Gui group, in 2020, applied oxygenation in a unique way: oxygenating cycloalkyl-fused furans [83]. Selective oxygenation of the furan system with singlet oxygen, followed by immediate addition of iron(II), led to formation of a butenolide and an alkyl radical, with concomitant oxidation of the ferrous salt. This radical was then trapped by oxygen to give a mixture of an alcohol and an aldehyde. Oxidizing or reducing agents could be used to prepare either the carbonyl or the alcohol selectively [83]. The following year, Bao demonstrated radical carboesterification of dienes. The alkyl radicals were formed through decarboxylative β-scission upon reduction of a diacyl peroxide by copper(I). With a chiral ligand, the carboxylate generated was coupled to the allyl radical intermediate to form the product. Two different chiral ligands were reported to provide each enantiomer with enantiomeric excesses of up to 94% [110]. Also in 2021, Maruoka published two reports on O-alkylation. In the first, coauthored with Matsumoto and appearing in July, various silylperoxyacetals were reduced by the iron(II) species formed in situ from the oxidation of TEMPO by iron(III) triflate. The resulting alkyl radical was trapped by an additional equivalent of TEMPO [111]. In the second, published in November, a pyridinyl-imidazole-based ligand was used with a cuprous species to reduce an alkyl silylperoxide that was trapped with phenol derivatives [112]. To complete this overview of oxygenation processes, in June of 2021 Guo demonstrated that cycloalkanones could be cleaved to yield various keto acids. It was proposed that copper(II) triflate enolizes the ketone, with blue light excitation promoting the formation of a peroxide intermediate. Although the authors suggested two likely mechanistic pathways, the absolute mechanism remains elusive. Overall, reduction of the O–O bond results in two carbonyl species [113].

Incorporation of selenyl groups into organic molecules can occur through radical processes (Figure 8C). In 2020, Gui reported the trapping of an alkyl radical with diphenyl diselenide during butenolide formation [83]. A year later, Guo published the selenylation of alkyl radicals generated through the reduction of tertiary hydroperoxides [108]. In this process, ferric tosylate was used as the catalyst, with diphenyl diselenide as the selenyl source. Instead of the transmetalation step observed for some other modes of reactivity, the alkyl radical could add directly to the Se–Se bond to form a selenyl radical. The selenyl radical could release an electron to the ferric species and form a Se–O bond, thereby restarting the catalytic cycle.

Olefination

Olefins are multipurpose functional groups. They build rigidity through their defined geometry and they are used extensively in organic synthesis because of their compatibility with a vast number of chemical reactions. Pertinent to this review article, excised alkyl radicals have been demonstrated to undergo copper(II)- and iron(III)-mediated oxidation processes, yielding olefins.

The oxidation of alkyl radicals to alkenes is well established. In 1961, De La Mare, Kochi, and Rust outlined the basis of this transformation (Figure 9) [114]. Initially, ferrous salts were used to reduce the peroxide, generating an alkoxy radical for β-scission. The alkyl radical resulting from β-scission was oxidized by a cupric salt in solution, generating an olefin. The application of this reaction to the total synthesis of the natural product (±)-recifeiolide was realized by Schreiber in 1980 [115].

Figure 9.

Olefination. See [83,114–116,118,120–122]. See also [82,117,119]; not pictured.

Since the 1980s, radical olefination from peroxides has been used only sporadically, but the Bao group picked it up again in 2018 [116]. This time, the olefin served as the radicophile. The alkyl radicals formed from decarboxylation of a tert-butyl perester were added into an olefin with the aid of an iron(III) catalyst. After radical–polar crossover, the resulting radical species was oxidized to a carbocation where a base promoted the elimination to yield the olefin product. Li and Bao applied the same methodology for use with tertiary propargyl alcohols, with the novelty stemming from a tandem SET/Lewis acid cycle [117]. The Lewis acid cycle helped eliminate the original alcohol in situ, forming terminal 1,3-enynes, to which the alkyl radical added. Both of these reports featured radical–polar crossover and a base to form the resulting olefin. The Bao group demonstrated the direct use of a 1,3-eneyne in 2019 [118]. Through a similar mechanism, this ligand-mediated copper(I)-catalyzed process first provided a propargyl radical with a stable allenyl radical resonance form, followed by trapping of the alkyl radical with an arylboronic acid to form an allene product. In another report in the same year, Zhang and Bao published a similar reaction in which the radical was trapped by a trimethyl cyanide [119].

The Guo and Kwon groups have also pursued radical olefinations with the use of peroxides and iron(II) catalysts. In 2019, the Guo group used cycloalkylsilyl peroxides and β-scission to form alkyl radicals [120], which then added into the olefinic units of cinnamic acids. Oxidation of these species enabled a decarboxylation pathway to form stable olefin products. In 2020, that group also reported the olefination of alkyl radicals with styrene derivatives [82]. Here, the ferric species in solution could oxidize the persistent benzylic radical to form the olefin and restart the catalytic cycle. The Kwon group used a dealkenylative approach for alkenylation. The generated alkyl radical was added to a β-nitrostyrene, with subsequent β-elimination of the nitro group, forming the olefinated product [121]. In the Gui publication mentioned previously, the largest portion of the substrate scope involved olefination [83]. Using copper(II) acetate and no designated radical trap, the resulting radical was transformed into a monosubstituted alkene. Finally, the Gui group also reported ruthenium(II)-catalyzed endoperoxide fragmentation. From bicyclic endoperoxides, the ruthenium catalyst mediated the O–O bond cleavage to facilitate the formation of two ketones and an olefin. The power of this method was exemplified with a phenanthrene system where the three fused rings were converted into one macrocycle, another demonstration of how peroxides can be used in unique ways [122].

Arylation

Arylation of alkyl radicals resembles cross-coupling reactions in terms of both outcome and relevance (Figure 10). In a traditional sense, cross-couplings involve an aryl component and an organometallic nucleophile. Here, the metal salt serves as a reductant by donating an electron to the peroxide, initiating a cascade to yield an alkyl radical that can be coupled to various aryl sources. In 1970, Minisci demonstrated that an alkyl radical, formed through reduction of an α-methoxyhydroperoxide, could be trapped by pyridine [123]. Recently, the Minisci reaction has been expanded by Bao to include reactions of organic peroxide precursors with arenes. Catalytic iron(III) has been used to promote radical decarboxylation in the presence of an electron-deficient arene, resulting in a delocalized radical intermediate that is oxidized, in a radical–polar crossover manner, to a delocalized carbocation to restore aromaticity, with concomitant reduction of the metal to turn over the catalyst [124]. In January of 2020, Maruoka reported using copper(I) iodide and a bipyridine ligand with potassium bicarbonate to reduce tertiary silylperoxides and provoke the coupling of arylboronic acids with the alkyl radical [125]. In April that same year, Guo demonstrated a similar reaction using a bis(diglyme)nickel bromide with di-tert-butyl dipyridine (dtbpy) and triethylamine [126]. Both the Maruoka and Guo reactions required the use of a base, as in a typical Suzuki–Miyaura coupling. In 2021, the Gao group introduced this mode of C–C activation to the quinoxaline system, using cuprous triflate 4-pyrrolidinopyridine (4-Ppy) as a ligand, expanding upon the known methodologies [127]. By means of radical–polar crossover, they proposed an aminyl radical that was oxidized to a nitrenium cation. Deprotonation led to elimination that formed the imine. The most recent arylation reported from the Maruoka group used a nickel(II) catalyst to reduce cyclic silylperoxides. The resulting alkyl radical was then trapped by benzamide derivatives in the ortho-position relative to the directing amide group [128].

Figure 10.

Arylation. See [123–128].

Alkynylation and allylation

Alkynylation and allylation of radicals stemming from the reduction of peroxides are more recent discoveries. Alkynes are desirable because of their functional versatility, rigidity, and prevalence in natural products. There have been only three examples of such alkynylations to date (Figure 11A). In 2018, Maruoka reported a catalytic cuprous iodide-catalyzed reduction of tertiary silylperoxides performed in the presence of terminal alkynes or alkylnylsulfones to achieve the desired coupling [129]. An amine base was required to obtain sufficient reactivity, suggesting some similarity to Sonogashira coupling. The versatility of the installed alkyne functionality was demonstrated through syntheses of tetrahydropyrans and enones. Two years later, the Gui group reported alkynylation using an activated (phenylethynyl)sulfone to trap the alkyl radical [83]. In 2022, Kwon published the reduction of the Criegee ozonolysis product in methanol, mediated by iron(II) in the presence of various alkynylsulfones, thereby trapping the alkyl radical while releasing a benzene sulfonyl radical and producing an alkynylated product [130].

Figure 11.

Alkynylation, allylation, boronation, and silylation. (A) Alkynylation. See [83,129,130]. (B) Allylation. See [78,83,131,132]. (C) Boronation and silylation. See [133,134].

For allylation, in 2018 the Sodeoka group reported using copper(II) to reduce bis(chlorodifluoroacetyl) peroxide (Figure 11B) [78]. After subsequent decarboxylation, the alkyl radical added into a terminal alkene. Basic conditions allowed the formation of an allyl chlorodifluoromethane. Further adding to the butenolide functionalization, the Gui group used an α-sulfonylmethyl acrylate to achieve allylation [83]. In 2022, the Gao group reported radical allylation of an alkyl radical originating from a tertiary hydroperoxide [131]. They used catalytic ferric nitrate nonahydrate and an α-sulfonylmethyl acrylate for their allylation. Most recently, the Wang group described a nickel(I)-mediated reduction of a silylperoxide. They trapped the radical with an α-trifluoromethyl styrene. Through a defluorinative mechanism, allylation was achieved to give a tetrasubstituted alkene [132]. With only a few examples of allylation and alkynylation, this area is open for further exploration.

Borylation and silylation

Access to boryl- and silyl- compounds with distal keto groups remains problematic. Boron is an interesting coupling partner because of the versatility of C–B bonds. Silanes have been studied less, but the ability to install silyl groups could be useful for niche applications. In 2019, two papers relating to borylation were published, one by Sakamoto and Maruoka and the other by Guo. Sakamoto and Maruoka were the first to describe borylation through the use of bis(pinacolato) diborane (Figure 11C) [133]. Here, cuprous iodide reduced tertiary silylperoxides to form alkoxy radicals and a cupric species. The cupric species transmetalated with bis(pinacolato)diboron to form a cupric–boron bond and a borate ester. The alkyl radical then added into the copper, forming a species that could reductively eliminate to form a new C–B bond. In the same report, an experiment was performed by the Maruoka group involving a (dimethylphenylsilyl)boronic pinacol ester for transmetalation to form a silyl–cupric species, which went on to forge a C–Si bond [134]. Guo published a boronation similar to that of Maruoka, but using acetone instead of acetonitrile as the solvent, with 4-dimethylaminopyridine (DMAP) as a base/ligand to forge the C–B bond [134].

Safety

A common concern with peroxide chemistry is related to the explosivity of peroxides. It might be easier to simply eschew any reactions involving organic peroxides, to reduce the risk of accidents in laboratory settings, but organic peroxides are not necessarily as hazardous as their reputation suggests. Almost all of the publications cited in this review article describe the use of organic peroxides, so it is imperative that potential safety hazards be addressed.

One method to minimize risk is to conduct risk assessment calculations of molecules before working with them. Historically, simple oxygen balance (OB) calculations have been used to estimate the potential explosive nature of compounds [135]. To calculate this number, the counts of carbon, hydrogen, and oxygen atoms are considered with respect to the overall molecular weight of a molecule (Figure 12A). In essence, lower absolute values correspond to more hazardous compounds. For example, acetyl peroxide has an OB of −95, corresponding to ‘high risk’. Unfortunately, this approach is flawed because some compounds, including water and carbon dioxide, have an OB of zero, but no associated explosive hazards. In addition, tert-butylperoxide has an OB of −252, corresponding to ‘low risk’, but has an observed (obs.) hazard rank of ‘high risk’ [136]. Nevertheless, OB calculations are useful as a baseline estimate for the risk assessment of organic peroxides, but the resulting value should be interpreted with caution.

Figure 12.

Peroxide safety. (A) Oxygen balance. (B) Chemical modification. See [90,125,137–140].

There are several chemical modifications that can improve the stability of organic peroxides, facilitate extractions and purifications, and minimize hazards. Many of the reactions described in this review article use silylperoxides, which behave almost identically to hydroperoxides, but have much larger absolute OB values. Chemically transforming hydroperoxides into silylperoxides is a straightforward and simple process that has been known since 1958. Davies was the first to report the synthesis of a silylperoxide from an organic hydroperoxide, using trimethylsilyl chloride and tert-butyl hydroperoxide in pyridine and pentane at 0°C (Figure 12B) [137]. Even today, this technique is used, for example, in the studies of Maruoka and Guo [90,125]. Silylperoxides have multiple advantages over hydroperoxides. The alkyl substituents on the silicon atom can improve the solubility in organic solvents and the overall stability, potentially yielding bench- and column-stable organic peroxides. Davies also published the functionalization of hydroperoxides with boron (1958), germanium (1959), and tin (1962), using the chlorinated species of these metals to functionalize organic hydroperoxides, although some all of these modified molecules are potentially explosive [138–140].

More recently, chemists have turned toward calculations to identify the hazards of peroxides. The computer program ‘Chemical Thermodynamic and Energy Release’ (CHETAH) can estimate the explosive risk associated with a compound, by comparing various parameters to a library of known explosive compounds [141]. The CHETAH-calculated energy and detonation values for the known explosives TATP and hexamethylene triperoxide diamine are comparable with their known empirical risks [142]. Density functional theory, complete basis set theory, and Gaussian software have also been applied to calculate the bond dissociation energies of various organic peroxides. It has been demonstrated that intramolecular hydrogen bonding in gem-1,2-hydroperoxide and acyl-hydroperoxides helps to stabilize their O–O bonds and lower the risk of accidental cleavage [2,143]. With the development of these computer programs, conducting quick calculations before running an experiment in a laboratory setting can help chemists to minimize accidents. While this review article is focused on organic peroxides, these types of calculations have broader use with other types of reactions.

Safety is an important factor to consider when creating new methodologies or planning organic syntheses. Organic peroxides, to the uneducated chemist, may present a safety hazard, but with a few quick and easy calculations, and/or chemical modifications, laboratory accidents can hopefully be avoided.

Concluding remarks

The take-home message from this review article is that organic peroxides are relatively overlooked molecular motifs that provide the opportunity for facile C–C activation. Traditionally, peroxides have been viewed as explosive, corrosive, and hazardous, but with proper precautions they can be synthetically useful, allowing access to otherwise difficult-to-prepare complex molecules. When applied to radical chemistry, the issue of undesirable and uncontrolled side reactivity is a legitimate concern. Nevertheless, the moderately high-yielding reactions appearing in this review article prove that it can mostly be avoided. Presumably, the use of catalytic quantities of metals to mediate this reactivity limits the generation of free radical species to a level dependent on the concentration of the oxidizable metal present in solution. Thus, the concentration of radicals available at any time is restricted. Overall, organic peroxides provide an accessible handle for selective C–C activation through a metal-mediated radical cascade, where a variety of functional groups can be coupled for any synthetic need.

There are three foreseeable areas for future improvement (see Outstanding questions). First, developing simpler methods for the tolerable integration of peroxides to make this methodology applicable to a wider range of molecules. Second, asymmetric cyclic peroxides and methods to couple radicals enantioselectively have yet to be investigated. Most of the recent examples provided in this review article involve tertiary peroxides, demonstrating one of the current limitations. Although there are methods available to form secondary peroxides, they require more thorough investigation. Finally, the products isolated after C–C activation have immense potential for further derivatization and application in complex syntheses. The future of C–C activation chemistry, as it relates to peroxide precursors, is here for those ready to seize it.

Outstanding questions.

There are many different radical trapping methods presented in this review article. Does a universal leaving group, such as boron, sulfur, or nitrogen, exist to enable the addition of any desired coupling partner?

Does the reactivity of primary and secondary organic peroxides differ from that of tertiary organic peroxides? What factors, such as solvent, temperature, ligands, or additives, may influence the C–C scission process for secondary organic peroxides to produce a desired outcome? How do each of these factors affect the mechanism? What is the exact mechanism for these processes? How can computational chemistry help provide insight on the mechanism of this methodology?

How large of a role does the redox potential of the chosen metal play in the reduction of organic peroxides. What is the corresponding redox potential of organic peroxides such as hydroperoxides, silylperoxides, and boroperoxides? Is it possible to predict the best reducing metal based on these conditions? What other metals may be applicable for this chemistry?

How will this methodology be utilized in synthetic organic chemistry or total synthesis? When will the stereotype of wild, uncontrollable radical chemistry be eliminated for synthetic chemists to embrace the power of this methodology?

How can the safety techniques for working with organic peroxides be streamlined to minimize risk? Will existing computational programs become free-of-charge and publicly available to quickly ascertain the risk of a particular substance? How can these methods be improved to provide reliable data?

Highlights.

Interest in the radical decomposition of organic peroxides has spiked recently, with many methods for trapping these radicals having been published.

Synthetic methods have evolved to include the regioselective installation of peroxide functionalities in existing organic molecules, thereby enabling radical decomposition processes.

Inexpensive metals, including iron, copper, and, most recently, nickel, can be used, both stoichiometrically and catalytically, to reduce organic peroxides, mediate radical decomposition, and facilitate subsequent cross-coupling reactions.

Peroxide chemistry suffers from the stereotype of being hazardous, but developments in safety and controlled environments have lowered its risks.

Glossary

β-Scission

a mechanistic process involving the cleavage of the σ-bond between the α- and β-positions of a reactive intermediate. In this review article, the fission involves radicals.

C–C activation

a process that converts an inert C–C bond into a functionalized product.

C–H functionalization

a process that converts an otherwise inert C–H bond into a derivatized product.

Chalcogenation

the addition of a group 16 element, particularly an O, S, or Se atom, into an atom or molecule.

Condensation chemistry

chemical methods that produce water as an intentional byproduct.

Cross-coupling reactions

chemical methods for connecting two different fragments together.

Cuprous

the redox states of copper in reduced copper(I) and oxidized copper(II), respectively.

Ferrous

the redox states of iron in reduced iron(II) and oxidized iron(III), respectively.

Hydrosilylperoxidation

a chemical reaction in which a hydrogen atom and a silylperoxide are added across an alkene.

Ligand

an ion or molecule attached to a metal atom through coordinate bonding.

Markovnikov addition

a reaction in which the more-electron-rich portion of the addition partner adds to the side of the alkene containing the fewest number of hydrogen atoms.

Metal–peroxide redox chemistry

a class of chemistry involving the transfer of a single electron from a metal to a peroxide, thereby oxidizing the metal and reducing the peroxide.

Perester

a molecule containing an ester functional group (CO₂R) with an oxygen–oxygen bond (CO₃R).

Pericyclic reactions

chemical transformations involving the cyclic reorganization of electrons in one concerted step.

Peroxide

a molecule containing at least one oxygen–oxygen (O–O) bond.

Radical

an atom, molecule, or ion that has at least one unpaired valence electron. In Lewis dot structures, radicals are denoted with a single dot (•).

Radical–polar crossover

a mechanistic idea in which a radical species crosses over to an anionic or cationic polar species by reduction or oxidation, respectively.

Radical trap

(radicophile); an atom or molecule that serves to capture a free radical by forming a new covalent bond with the radical species.

Reductive elimination

an organometallic mechanistic step involving the reduction of a metal atom by two states while forming a new bond between two ligands.

Terpenoid

a class of terpene, a molecule built from isoprene units (C₅H₈), that also contains at least one oxygen atom.

2,2,6,6-Tetramethypiperid-1-yloxy (TEMPO)

an isolatable persistent radical that is commonly used to test for the presence of free radicals in solution, due to its rapid, intrinsic radical trapping capabilities.

Transmetalation

an organometallic mechanistic step involving the transfer of a ligand from one metal to another.