Facile CâF Bond Formation through a ... - American Chemical Society

Article Cite This: Acc. Chem. Res. 2017, 50, 2822-2833

pubs.acs.org/accounts

Facile C−F Bond Formation through a Concerted Nucleophilic Aromatic Substitution Mediated by the PhenoFluor Reagent Constanze N. Neumann† and Tobias Ritter*,†,‡ †

Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, Massachusetts 02138, United States ‡ Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470 Mülheim an der Ruhr, Germany CONSPECTUS: Late-stage fluorination reactions aim to reduce the synthetic limitations of conventional organofluorine chemistry with respect to substrate scope and functional group tolerance. C−F bond formation is commonly thermodynamically favorable but almost universally associated with high kinetic barriers. Apart from PhenoFluor chemistry, most modern aromatic fluorination methods reported to date rely on the use of transition metal catalysts, with C−F bonds often formed through reductive elimination. Reductive elimination chemistry to make C−X bonds becomes increasingly challenging when moving to higher atomic numbers in the periodic table from C−C to C−F, in part because of higher metal−X bond dissociation energies. The formation of C−C, C−N, and C−O bonds via reductive elimination has become routine in the 20th century, but it took until the 21st century to develop complexes that could afford general C−F bond formation. The availability of such complexes enabled the substrate scope of modern fluorination chemistry to exceed that of conventional fluorination. PhenoFluor chemistry departs from conventional reaction mechanisms for aromatic fluorination chemistry. Instead, we have revealed a concerted nucleophilic aromatic substitution reaction (CSNAr) for PhenoFluor that proceeds through a single neutral four-membered transition state. Conceptually, PhenoFluor chemistry is therefore distinct from conventional SNAr chemistry, which typically proceeds through a two-barrier process with Meisenheimer complexes as reaction intermediates. As a consequence, PhenoFluor chemistry has a larger substrate scope than conventional SNAr chemistry and can be performed on arenes as electron-rich as anilines. Moreover, PhenoFluor chemistry is tolerant of protic functional groups, which sets it apart from modern metal-mediated processes. Primary and secondary amines, alcohols, thiols, and phenols are often not tolerated under metal-catalyzed late-stage fluorination reactions because C−N and C−O reductive elimination can have lower activation barriers than C−F reductive elimination. The mechanism by which PhenoFluor chemistry forms C−F bonds not only rationalizes the substrate scope and functional group tolerance but also informs the side-product profile. Fluorinated isomers are not observed because the four-membered transition state necessitates ipso substitution. In addition, no reduced product, e.g., H instead of F incorporation, as is often observed with metal-mediated methods, has ever been observed with PhenoFluor. PhenoFluor chemistry can be used to deoxyfluorinate both phenols and alcohols. PhenoFluor is an expensive reagent that must be used stoichiometrically and therefore cannot replace cost-efficient methods to make simple fluorinated molecules on a large scale. However, PhenoFluor is often successful when other fluorination methods fail. The synthesis of 18F-labeled molecules for positron emission tomography (PET) is one application of modern fluorination chemistry for which material throughput is not an issue because of the small quantities of PET tracers used in imaging (typically nanomoles). The high emphasis on functional group tolerance, side-product profiles, and reliability combined with less stringent cost requirements render PhenoFluor-based deoxyfluorination with 18F promising for human PET imaging.

■

INTRODUCTION

satisfactory, but there are many exceptions: Metabolic decomposition, often by P450 enzymes, can render pharmaceuticals inactive.7−10 The incorporation of a fluorine substituent can block metabolic derivatization. To avoid de novo synthesis of the carbon skeleton once the sites of metabolic attack in a lead compound are known, late-stage introduction of the fluorine substituent is desirable. Further-

Fluorinated materials hold a privileged position in the pharmaceutical and agrochemical industry and in materials chemistry.1−6 Given the difficulties traditionally associated with C−F bond formation, fluorinated materials are often prepared through a “building-block approach” in which fluorine is introduced early in a synthetic sequence to ensure that the fluorination step has to tolerate few functional groups and little structural complexity. For several applications of fluorinated materials, a building-block approach is convenient and © 2017 American Chemical Society

Received: August 23, 2017 Published: November 9, 2017 2822

DOI: 10.1021/acs.accounts.7b00413 Acc. Chem. Res. 2017, 50, 2822−2833

Article

Accounts of Chemical Research

Figure 1. Deoxyfluorination reagents that are not applicable to the synthesis of aryl fluorides or lead to unwanted side reactions in the deoxyfluorination of alcohols.

Figure 2. Suite of commercially available deoxyfluorination reagents developed by the Ritter lab.

more, the utility of 18F-labeled molecules in positron emission tomography (PET) imaging has spiked an increased interest in late-stage fluorination reactions11 because the need for latestage fluorination in 18F chemistry due to its 110 min half-life is well-appreciated.12 Fluorinated arenes are often synthesized through the Balz− Schiemann reaction, which converts potentially explosive aryldiazonium tetrafluoroborate salts into aryl fluorides upon heating.13,14 Arenes bearing strongly electron-withdrawing groups at the ortho or para position with respect to the desired site of fluorination are readily available through nucleophilic aromatic substitution of halides.15,16 The Buchwald group showed that bulky monodentate phosphine ligands could achieve the challenging C−F bond reductive elimination from PdII through which aryl triflates and aryl bromides could be converted to aryl fluorides with nucleophilic fluoride sources.17 The design of specialized ligands curbed the formation of fluorinated constitutional isomers, which complicated product purification in early iterations.18−22 Unlike the case of copper-mediated fluorination reactions of aryl halides,23 hydrodehalogenated products were not detected in significant amounts for the palladium-catalyzed fluorination of aryl bromides and triflates. The development and commercial availability of bench-stable electrophilic fluorinating reagents opened the door to C−F bond formation via reductive elimination from high-valent transition metals.24,25 C−F bond formation from high-valent silver complexes allowed the Ritter group to develop a latestage fluorination reaction of aryl stannanes that can derivatize molecules such as taxol and rifamycin S.26,27 Strategies based on electrophilic fluorinating reagents are generally incompatible with electron-rich amine and thioether functional groups, whereas nucleophilic fluorination methods generally require anhydrous conditions and substrates devoid of protic functional groups. Phenols and alcohols are easily accessed, bench-stable materials, often with an appreciably different polarity from aryl and alkyl fluorides.28 Deoxyfluorination thus constitutes an alternative to cross-coupling for aryl fluoride synthesis that

sources a different pool of potential starting materials, which in some cases are more readily accessible than aryl halides.

1. DEOXYFLUORINATION REAGENTS A number of synthetic methods had been described in the literature for the interconversion of alcohols to alkyl fluorides, but the reactions were often plagued by low reaction efficiency, formation of side products, and the need for a large excess of potentially explosive reagents.29,30 Conventional deoxyfluorination reagents were furthermore not applicable to the synthesis of aryl fluorides that were not already accessible via the Halex reaction (Figure 1).29,31,32 Our group has developed a suite of deoxyfluorination reagents capable of converting densely functionalized and structurally complex alcohols and phenols into the corresponding alkyl and aryl fluorides (Figure 2).31,33−36 Inspiration for the PhenoFluor reagent was drawn from the DFI reagent; the use of an imidazole instead of an imidazoline framework was identified as a key design element for phenol deoxyfluorination. 2. SCOPE OF PHENOFLUOR, PHENOFLUORMIX, AND ALKYLFLUOR A wide range of phenols underwent PhenoFluor-mediated deoxyfluorination, and anilines, protic functional groups, and various heterocycles were tolerated (Figure 3A). The range of alkyl fluorides that can be accessed via deoxyfluorination has been extended by the PhenoFluor reagent (Figure 3B). Deoxyfluorination with PhenoFluor is a “late-stage” reaction, as is apparent from the complexity of suitable substrates and the chemoselectivity observed in many of its transformations.33 A concern with many deoxyfluorinating reagents is their limited thermal stability, which can lead to explosions for reactions conducted at elevated temperatures. An exothermic decomposition was observed by differential scanning calorimetry for PhenoFluor at 213 °C, compared with 155 °C for DAST, 158 °C for Deoxo-Fluor and 205 °C for XtalFluorE.31,37 The solid PhenoFluor reagent hydrolyzes upon contact with moisture and needs to be stored and handled under an inert atmosphere. PhenoFluorMix, a combination of chlori2823


Article


Figure 3. (A) Deoxyfluorination of phenols with PhenoFluor. (B) Deoxyfluorination of alcohols with PhenoFluor. aNo KF was used. (C) PhenoFluorMix matches the substrate scope of PhenoFluor but lacks its sensitivity to moisture. bReactions were performed in xylene at 140 °C. (D) Deoxyfluorination with AlkylFluor. cCsF was used instead of KF, preheated with AlkylFluor at 100 °C in toluene. d10 equiv of KF was used. 2824


Article


Figure 4. Features of the PhenoFluor-mediated aliphatic deoxyfluorination reaction. (A) Exceptions to the usually observed complete inversion. (B) Chemoselective deoxyfluorination of sterically exposed hydroxyl groups not engaged in strong intramolecular hydrogen bonding. (C) An SN2 reaction was favored over an SN2′ pathway.

Figure 5. Conversion of a hydroxide into a leaving group.

nated IPr·HCl and an excess of CsF, can directly be used for the deoxyfluorination of phenols (Figure 3C). In contrast to PhenoFluor, PhenoFluorMix is not moisture sensitive. We developed the AlkylFluor reagent to engender the same operational convenience to alcohol deoxyfluorination that PhenoFluorMix brought to phenol deoxyfluorination (Figure 3D). While PhenoFluor itself can deoxyfluorinate both alcohols and phenols, the use of PhenoFluorMix for alcohols induces an undesired deoxychlorination reaction that competes with the productive deoxyfluorination pathway. AlkylFluor is structurally similar to PhenoFluorMix but does not contain other nucleophilic anions as counteranions. AlkylFluor can be conveniently synthesized on a large scale and is insensitive to moisture. The use of dry fluoride in apolar solvents can lead to E2 elimination, but because KF is virtually insoluble in toluene, base-induced elimination occurs at a low rate. Aliphatic deoxyfluorination with the PhenoFluor reagent proceeded with inversion of configuration in all but two of the cases examined (Figure 4A). Retention was observed for cyclic

hemiacetal 1, possibly because of the formation of an oxycarbenium intermediate, and less than 10% retention occurred for allylic alcohol 2. Chemoselective deoxyfluorination of a single hydroxyl group was frequently observed in substrates featuring several hydroxyl groups (Figure 4B). In general, OH groups engaged in strong intramolecular hydrogen bonds, such as in 1,3-hydroxyketones, were resistant to deoxyfluorination, as were β,β′-substituted hydroxyl groups. In certain cases, the preference for deoxyfluorination of primary alcohols over secondary alcohols was sufficient for a chemoselective transformation (Figure 4B). Several allylic secondary substrates were subjected to PhenoFluor-mediated deoxyfluorination, and a large preference for an SN2 mechanism over an SN2′ mechanism was observed (Figure 4C). While most primary and secondary alcohols reliably underwent deoxyfluorination, tertiary alcohols were resistant to the transformation apart from some tertiary allylic substrates. Interestingly, an SN2 mechanism was still favored over an SN2′ mechanism for the displacement in the case of tertiary allylic substrates. 2825


Article


Figure 6. Hammett plot for phenol deoxyfluorination with PhenoFluor.

Figure 7. Concerted nucleophilic aromatic substitution features a single activation barrier and leads to a decreased barrier height for electron-rich phenol substrates.

3. DEOXYFLUORINATION MECHANISM The condensation of PhenoFluor and phenols led to uronium intermediates, in which the hydroxy group on the arene has been converted into a good leaving group (Figure 5). Displacement of leaving groups by nucleophiles commonly proceeds via nucleophilic aromatic substitution, which is a wellestablished mechanistic pathway that is known to be limited to electron-deficient arene substrates.16 PhenoFluor deoxyfluorination did not achieve a minor expansion relative to the substrate scope commonly associated with SNAr reactions but differed so drastically from the expected scope that a stepwise SNAr mechanism seemed unlikely.38−41 We conducted a thorough mechanistic study of the deoxyfluorination.42 The Hammett plot for the deoxyfluorination of phenols gave rise to a good linear fit, which excluded a change in mechanism or rate-determining step as the phenol is varied (Figure 6). We obtained a ρ value of 1.8, indicating that electron-deficient phenols reacted faster than electron-rich phenols, but the dependence of the rate on the σ value of the arene substituent was considerably less pronounced than in the case of other SNAr reactions, which are commonly associated with ρ values between 3 and 8.40 We concluded from the Hammett plot that while negative charge accumulated on the arene during the ratedetermining step of the deoxyfluorination, the extent of charge accumulation was considerably lower than had been observed in the transition states of SNAr reactions. We compared the absolute reaction rates of 18O- and 16Olabeled phenol and observed krel = 1.08. The primary kinetic

isotope effect indicated a significant amount of C−O bond cleavage in the rate-determining step.43 No positional isomers were observed in the case of phenol substrates for which the intermediacy of an aryne would have disfavored overall ipso substitution.44,45 Considering the accumulated experimental and computational data, we proposed that PhenoFluormediated deoxyfluorination proceeds via a concerted nucleophilic aromatic substitution (CSNAr) pathway in which nucleophile attack and leaving group loss happen concurrently.39−42,46−52 The overall transformation consists of three independent reaction steps (condensation of phenol and PhenoFluor to form uronium 3, formation of the tetrahedral adduct 4, and nucleophilic displacement); the rate-determining nucleophilic substitution is concerted, and no intermediates are formed as the tetrahedral intermediate 4 is converted to the aryl fluoride product and urea 5. By virtue of the limited accumulation of charge on the aromatic core, electron-rich phenols can undergo concerted nucleophilic aromatic substitution (Figure 7). In the transition state, the majority of negative charge is localized on the incoming nucleophile and the departing leaving group. In contrast, in stepwise nucleophilic aromatic substitution reactions, a full formal negative charge must be deposited on the arene fragment in the Meisenheimer complex. A density functional theory (DFT) study of the PhenoFluormediated deoxyfluorination reaction revealed a single transition state, the characteristic feature of a concerted reaction (Figure 8).53 An intrinsic reaction coordinate analysis revealed that the 2826


Article


Figure 8. Proposed mechanism of the deoxyfluorination and experimental or computational evidence for the proposed intermediates and transition states. (A) X-ray structure of uronium bifluoride 3. (B) Intrinsic reaction coordinate and optimized DFT structure for transition state TS (B3LYP/631G(d)//6-311++G(d,p), toluene solvent model). (C) 13C−19F HSQC and 19F NMR spectra of an isolated sample of the tetrahedral intermediate 4.

Upon formation of uronium bifluoride 3, the optimized PhenoFluor reagent displayed a hydrogen bond between the C−H bonds of the imidazolium core of the uronium and the bifluoride counteranion, which persisted both in solution and in the solid state (Figure 9).31 PhenoFluor derivative 8 with phenyl substituents on the backbone, the uronium intermediate of which is unable to form hydrogen bonds between the 4- or 5-position of the imidazolium and the bifluoride counteranion, exhibited reactivity comparable to that of PhenoFluor. Therefore, hydrogen-bond formation is a coincidental structural feature of the PhenoFluor reagent that may not be crucial to its reactivity. All reagents capable of deoxyfluorinating electronrich phenols feature a diisopropylphenyl group on both imidazole nitrogen atoms. We postulated that the bulky isopropyl substituents are responsible for positioning the Naryl substituent at an almost 90° angle relative to the imidazolium plane. The perpendicular orientation minimizes resonance stabilization of the positive charge of uronium 3 by

transition state interconnected tetrahedral adduct 4 and the aryl fluoride and urea 5 products. The proposed tetrahedral adduct 4 was synthesized from a silylated phenol precursor and the PhenoFluor reagent and studied by 1H, 13C, 19F, and 1H−19F COSY NMR. An experimentally determined activation enthalpy of 21.8 kcal·mol−1 was associated with the conversion of tetrahedral adduct 4 to the aryl fluoride and PhenoFluor urea 5 reaction products. A computational barrier of 23.9 kcal·mol−1 was calculated by DFT for the conversion of 4 to the aryl fluoride.42

4. CONSEQUENCES OF THE PHENOFLUOR DEOXYFLUORINATION MECHANISM 4.1. Design of the Optimal Reagent Scaffold

We found that 4,5-bischlorination of the imidazolium yielded an inactive deoxyfluorinating reagent 6, Likewise, the use of a saturated imidazolidinium core (7) did not form aryl fluorides. 2827


Article


Figure 9. Despite earlier assumptions, the hydrogen bond of the uronium backbone to bifluoride may not be relevant to C−F bond formation.

Figure 10. High fluoride selectivity may originate from a difference in tetrahedral adduct stability.

the N-aryl substituents and destabilizes uronium 3 relative to the reactive tetrahedral intermediate 4.

C−F bond formation was observed in all of the examined cases of phenol deoxyfluorination.

4.2. Competitive Deoxychlorination

4.3. Consequences of the Intermediacy of the Tetrahedral Adduct

Unlike AlkylFluor, PhenoFluorMix afforded deoxychlorination as a side reaction in the deoxyfluorination of alcohols. However, deoxychlorination with PhenoFluorMix was never observed in the deoxyfluorination of phenols. In fact, a wide range of cesium, sodium, or tetrabutylammonium salts of various anions other than fluoride could be added to aromatic deoxyfluorination reactions with PhenoFluor with no arene−nucleophile bond formation other than C−F bond formation being observed. Fluorination reactions are essentially always more challenging to accomplish than other halogenation reactions, which makes selective phenol deoxyfluorination with PhenoFluor a case of unexpected chemoselectivity. The halide selectivity may originate in the different stabilities of tetrahedral adducts substituted with different halide anions (Figure 10). The concerted four-membered transition state for aromatic displacement enabled by the PhenoFluor reagent decreases the activation barrier for C−F bond formation. It is reasonable to assume, however, that a similar transition state could be constructed with other nucleophiles and that such a transition state would be associated with a moderate activation barrier, potentially lower than that corresponding to deoxyfluorination. Chemoselective C−F bond formation is observed, however, because the formation of the transition state requires a substantial concentration of a tetrahedral adduct containing the nucleophile of interest. Only fluoride appears to be capable of forming sufficiently stable tetrahedral adducts, and selective

Polar solvents such as dimethyl sulfoxide, N,N-dimethylformamide, and alcoholic solvents increase the solubility of most fluoride salts, but the strong hydrogen-bonding and dipole interactions between fluoride anions and solvent molecules that make fluoride soluble result in or at least contribute to the high reaction barriers typically observed.54 In apolar solvents, partial nucleophile desolvation is associated with a lower energetic barrier, but fluoride solubility can be prohibitively low. The formation of stable fluoride-containing intermediates that bring fluoride into close proximity to the desired coupling partner can circumvent the high solvation energies of fluoride in polar solvents and the high lattice energies of commonly used alkalimetal fluorides. Fluoride capture and solubilization prior to carbon−fluorine bond formation is one of the crucial roles played by transition metals in almost all modern aromatic fluorination reactions. The metal-free PhenoFluor reagent can also provide (with 19F) or capture (with 18F) fluoride and bring it near the desired aryl coupling partner. Palladium-catalyzed C−F bond formation and PhenoFluor-mediated deoxyfluorination share a preference for apolar solvents, in contrast to SNAr reactions, which proceed via a bimolecular ratedetermining C−F bond formation step, for which highly polar solvents are favored.17−19,22 The use of toluene as the reaction solvent favors the formation of a neutral tetrahedral adduct 4 in which the carbon2828


Article


Figure 11. Four- and six-membered transition state structures for an attacking fluoride or bifluoride nucleophile.

formation where fluoride serves as a nucleophile is 24 kcal/mol in the presence of CsF, which is in accord with the experimental value obtained from the Eyring plot for the deoxyfluorination of phenol: ΔG⧧(110 °C) = 23.4 ± 0.19 kcal· mol−1. In the absence of CsF, deoxyfluorination is calculated to occur via attack of bifluoride on the arene with an activation energy of 30 kcal·mol−1. Interestingly, a four-membered transition state is preferred over a six-membered transition state even in the case of bifluoride, where the six-membered transition state appears easily accessible.42 The incoming nucleophile and leaving group in the Newman−Kwart, Chapman, and Schönberg intramolecular rearrangements are constrained by the structure of the reaction precursor to take place via a four-membered transition state. While the activation energy often exceeds 30 kcal·mol−1, the rearrangement’s substrate scope substantially exceeds that traditionally associated with SNAr reactions, and experimental and computational work substantiates a concerted substitution mechanism.38,39 A recently reported unactivated Smiles rearrangement, an intramolecular displacement that could occur via a sixmembered transition state, was shown by computational means to occur in a stepwise fashion and feature a Meisenheimer intermediate.56 It is therefore possible that a four-membered transition state may in fact favor the occurrence of concerted displacement.

bound fluoride functions as an intramolecular fluoride source for the deoxyfluorination. We propose that the energetic contribution of solvation to the deoxyfluorination activation barrier is small because of the intermediacy of tetrahedral adduct 4: neither the associated reaction partners nor the transition state carry an overall charge, and little nuclear motion is required to proceed from tetrahedral adduct 4 to the transition state (Figure 10). Furthermore, the fluoride provided by the PhenoFluor-derived tetrahedral adduct is considerably less basic than anionic soluble fluoride such as TBAF and is primed to undergo productive C−F bond formation in preference to undesired charge-based interactions. The extensive functional group tolerance and large substrate scope of PhenoFluor deoxyfluorination arise because binding of fluoride to the PhenoFluor reagent in the form of a tetrahedral adduct decreases fluoride’s proclivity for side reactions while simultaneously activating it for the desired C−F bond formation reaction. A shift from a two-step SNAr mechanism to a concerted SNAr mechanism not only minimizes charge buildup on the arene during deoxyfluorination but also achieves partial release of the exergonicity of urea expulsion in the transition state, thereby reducing the energy of the transition state. The exergonicity of leaving group loss coupled with the insignificant inherent activation barrier for cleavage of the C−O bond disfavors the formation of a Meisenheimer intermediate. The common use of fluoride as a leaving group for the SNAr reaction is therefore a significant contributor to the prevalence of two-step nucleophilic aromatic substitution mechanisms due to the high C−F bond energy and the high basicity of fluoride.40,50 In fact, some SNAr reactions that feature non-fluoride leaving groups and lack strongly electron-withdrawing groups on the aromatic ring may be misclassified as stepwise reactions.41,55 A concerted displacement avoids the formation of high-energy intermediates, which is enthalpically favorable, but it requires a highly organized transition state involving synchronous nuclear motion, which can carry a steep entropic penalty. The preorganization engendered by the intermediacy of 4 likely reduces the entropic contribution to the activation barrier significantly and permits the occurrence of a concerted displacement at unusually low temperatures.

5. DEOXYFLUORINATION WITH 5.1. Metal-Free

18

F

18

F Deoxyfluorination

No-carrier-added radiofluorination requires the absence of exogenous 19F to ensure that the 18F-labeled product is not unnecessarily contaminated with its 19F isotopologue. We prepared fluoride-free uronium chloride precursor 9 and attempted to exchange the chloride counteranion for 18F and achieve efficient deoxyfluorination in the absence of added CsF. 18 F-Fluoride is typically prepared by proton bombardment of 18 O-H2O. We found that productive anion exchange between 18 F and uronium chloride 9 took place on an anion exchange cartridge in the absence of water or exogenous fluoride. Heating of the solution eluted from the anion exchange column yielded the aryl fluoride product with high specific activity. No special care was required to exclude air or moisture from the 18F deoxyfluorination reaction, and the radiolabeled product could conveniently be separated from the positively charged reaction precursor. Several heterocycles underwent deoxyfluorination with 18F, for example, 18F-5-fluorobenzofurazan was synthesized in 27% non-decay-corrected radiochemical yield (RCY) with a specific activity of 3.03 Ci/mol.42

4.4. Fluoride in the Transition State

Upon formation of uronium intermediate 3, the two fluoride atoms originally bound to the PhenoFluor reagent become part of the bifluoride counteranion of the uronium intermediate. Fluorination could thus proceed with either the bifluoride counteranion or a fluoride anion acting as the nucleophile (Figure 11). The calculated activation energy for a trans2829


Article


Figure 12. Deoxyfluorination with 18F-fluoride tolerates a wide range of functional groups.

Figure 13. Limitations on the substrate scope of 18F deoxyfluorination.

5.2. Limits of Metal-Free

18


5.3. Ruthenium-Mediated

18


To efficiently radiolabel electron-rich arenes, we changed the reaction conditions to achieve two objectives: first, to increase the equilibrium constant, KRu (Figure 14B), and second, to lower the activation barrier for fluorination to funnel the small concentration of tetrahedral adduct to the product more efficiently. The incorporation of a positively charged RuCp fragment via η6 complexation to the phenol substrates increased the Coulombic attraction between fluoride and the phenol− reagent complex, which increased the concentration of the tetrahedral intermediate in solution.57 The use of rutheniumactivated phenol substrates furthermore led to a significant reduction of the deoxyfluorination reaction barrier. A DFT study of the reaction mechanism revealed that incorporation of the RuCp fragment stabilized the Meisenheimer complex to such an extent that a classical SNAr mechanism is favored over a CSNAr mechanism even for an electron-rich substrate such as 4-methoxyphenol (Figure 15). The activation barriers for nucleophilic attack and leaving group loss were calculated to be 5.3 and 7.0 kcal/mol for the electronrich 4-methoxyphenol substrate, and decomplexation of the aryl fluoride product is the rate-determining step of the transformation.57 Phenols can be used as labeling precursors, obviating the need for isolation of the uronium intermediates and increasing the synthetic convenience. Given the facility of the C−F bond formation step, the current challenge for 18F deoxyfluorination with PhenoFluor is the formation of phenol− RuCp complexes as well as efficient in situ decomplexation.

A comparison of Figure 3 with Figure 12 illustrates that while deoxyfluorination with 18F rivals deoxyfluorination with 19F in its functional group tolerance, electron-rich substrates are not part of the scope of the 18F deoxyfluorination. We have shown that uronium formation occurs for electron-rich arenes, and anion exchange to incorporate 18F via cartridge elution is efficient with electron-rich substrates. Mechanistic investigation of the C−F bond formation step under 18F conditions is not practical, but given the wealth of data available for reactions with 19F and the insignificant 19F/18F kinetic isotope effect, we assumed that C−18F bond formation would proceed efficiently for electron-rich substrates. The most significant difference between 18F and 19F experiments is that excess 19F fluoride is present in experiments with 19F, whereas fluoride is the limiting reagent in reactions with 18F. Tetrahedral intermediate 4 is in equilibrium with uronium fluoride 10 (Figure 13). In experiments with 19F, electron-rich substrates could be deoxyfluorinated via the small equilibrium concentration of 4 because excess fluoride present in the reaction mixture led to a high overall fluoride concentration and unproductive side reactions were statistically less likely to involve the fluoride anion in 10. The absence of added fluoride under 18F reaction conditions caused the fluoride anion in 18F-10 to react with any fluorophilic species present in the reaction mixture, which depleted the concentration of 18F-10 and consequently 18F-4. 2830


Article


Figure 14. Ruthenium-mediated deoxyfluorination enabled 18F labeling of electron-rich and electron-neutral phenol substrates through stabilization of the tetrahedral adduct intermediate and lowering of the deoxyfluorination barrier compared with the ruthenium-free reaction. *26% non-decaycorrected isolated yield.

Figure 15. (A) Internal reaction coordinates (B3LYP/6-31G(d), toluene solvent model) for the RuCp complex of 4-methoxyphenol. (B) DFT structures and activation barriers for TS1 and TS2 (B3LYP/6-31G(d)//6-311++G(d,p), toluene solvent model). The activation barriers in (B) differ from, and are more accurate than, those shown in (A) because a larger basis set was used.

6. CONCLUSION Deoxyfluorination with the PhenoFluor reagent arrived at a functional-group-tolerant reaction with a large substrate scope using a metal-free approach but features a crucial change in mechanism from Halex chemistry. Concurrent attack of the nucleophile and loss of the leaving group during aromatic deoxyfluorination limits charge buildup in the transition state and allows part of the exothermicity of leaving group loss to be released in the transition state. PhenoFluor and its associated

reagents PhenoFluorMix and AlkylFluor have been thoroughly established to provide convenient and reliable access to a wide range of aryl and alkyl fluorides as well as alkyl aryl ethers. The major limitation of PhenoFluor deoxyfluorination is the need for stoichiometric amounts of an expensive reagent, which prohibits its application to synthesis on a large scale. More economical reagents, such as PyFluor and SO2F2, are preferable whenever less complex substrates are used. However, PhenoFluor deoxyfluorination can succeed when other 2831



■

methods fail and provides direct access to densely functionalized substrates in a reliable and predictable fashion. PhenoFluor deoxyfluorination with 18F holds promise to become one of the radiolabeling methods of choice for the synthesis of tracers for human PET imaging that are not amenable to classic SNAr or SN2 reactions. Several new methods for 18F incorporation are now available. In our opinion, at least the fluorination of boronic esters and PhenoFluor deoxyfluorination have shown the scope, functional group tolerance, robustness, and operational convenience to prove enabling to medical imaging.58,59 However, further development of deoxyfluorination, especially in terms of facilitating ruthenium (de)complexation, are needed to allow 18 F deoxyfluorination to fulfill its full potential. Beyond synthetic utility, PhenoFluor chemistry gives fundamental insight into mechanistic features that permit an expansion of the substrate scope of aromatic nucleophilic displacement. Future endeavors should focus on how the lessons learned with fluoride can be applied to other nucleophiles to permit nucleophilic displacement in general to be feasible even with electron-rich aromatic substrates.

■

Article

REFERENCES

(1) Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.; Meanwell, N. A. Applications of Fluorine in Medicinal Chemistry. J. Med. Chem. 2015, 58, 8315−8359. (2) Hagmann, W. K. The Many Roles for Fluorine in Medicinal Chemistry. J. Med. Chem. 2008, 51, 4359−4369. (3) O’Hagan, D. Understanding organofluorine chemistry. An introduction to the C-F bond. Chem. Soc. Rev. 2008, 37, 308−319. (4) Wang, J.; Sanchez-Rosello, M.; Acena, J. L.; del Pozo, C.; Sorochinsky, A. E.; Fustero, S.; Soloshonok, V. A.; Liu, H. Fluorine in pharmaceutical industry: fluorine-containing drugs introduced to the market in the last decade (2001−2011). Chem. Rev. 2014, 114, 2432− 2506. (5) Zhou, Y.; Wang, J.; Gu, Z.; Wang, S.; Zhu, W.; Acena, J. L.; Soloshonok, V. A.; Izawa, K.; Liu, H. Next Generation of FluorineContaining Pharmaceuticals, Compounds Currently in Phase II-III Clinical Trials of Major Pharmaceutical Companies: New Structural Trends and Therapeutic Areas. Chem. Rev. 2016, 116, 422−518. (6) Jeschke, P. The unique role of fluorine in the design of active ingredients for modern crop protection. ChemBioChem 2004, 5, 570− 589. (7) Lynch, T.; Price, A. The effect of cytochrome P450 metabolism on drug response, interactions, and adverse effects. Am. Fam. Physician 2007, 76, 391−396. (8) Daly, A. K. Pharmacogenetics of the cytochromes P450. Curr. Top. Med. Chem. 2004, 4, 1733−1744. (9) Walsh, J. S.; Miwa, G. T. Bioactivation of Drugs: Risk and Drug Design. Annu. Rev. Pharmacol. Toxicol. 2011, 51, 145−167. (10) Park, B. K.; Kitteringham, N. R.; O’Neill, P. M. METABOLISM OF FLUORINE-CONTAINING DRUGS. Annu. Rev. Pharmacol. Toxicol. 2001, 41, 443−470. (11) Ametamey, S. M.; Honer, M.; Schubiger, P. A. Molecular imaging with PET. Chem. Rev. 2008, 108, 1501−1516. (12) Lee, E.; Kamlet, A. S.; Powers, D. C.; Neumann, C. N.; Boursalian, G. B.; Furuya, T.; Choi, D. C.; Hooker, J. M.; Ritter, T. A Fluoride-Derived Electrophilic Late-Stage Fluorination Reagent for PET Imaging. Science 2011, 334, 639−642. (13) Balz, G.; Schiemann, G. Ü ber aromatische Fluorverbindungen, I.: Ein neues Verfahren zu ihrer Darstellung. Ber. Dtsch. Chem. Ges. B 1927, 60, 1186−1190. (14) Laali, K. K.; Gettwert, V. J. Fluorodediazoniation in ionic liquid solvents: new life for the Balz−Schiemann reaction. J. Fluorine Chem. 2001, 107, 31−34. (15) Terrier, F. Modern Nucleophilic Aromatic Substitution; WileyVCH: Weinheim, Germany, 2013. (16) Cheron, N.; El Kaim, L.; Grimaud, L.; Fleurat-Lessard, P. Evidences for the key role of hydrogen bonds in nucleophilic aromatic substitution reactions. Chem. - Eur. J. 2011, 17, 14929−14934. (17) Watson, D. A.; Su, M.; Teverovskiy, G.; Zhang, Y.; GarcíaFortanet, J.; Kinzel, T.; Buchwald, S. L. Formation of ArF from LPdAr(F): Catalytic Conversion of Aryl Triflates to Aryl Fluorides. Science 2009, 325, 1661−1664. (18) Lee, H. G.; Milner, P. J.; Buchwald, S. L. An Improved Catalyst System for the Pd-Catalyzed Fluorination of (Hetero)Aryl Triflates. Org. Lett. 2013, 15, 5602−5605. (19) Lee, H. G.; Milner, P. J.; Buchwald, S. L. Pd-Catalyzed Nucleophilic Fluorination of Aryl Bromides. J. Am. Chem. Soc. 2014, 136, 3792−3795. (20) Maimone, T. J.; Milner, P. J.; Kinzel, T.; Zhang, Y.; Takase, M. K.; Buchwald, S. L. Evidence for in Situ Catalyst Modification during the Pd-Catalyzed Conversion of Aryl Triflates to Aryl Fluorides. J. Am. Chem. Soc. 2011, 133, 18106−18109. (21) Milner, P. J.; Kinzel, T.; Zhang, Y.; Buchwald, S. L. Studying Regioisomer Formation in the Pd-Catalyzed Fluorination of Aryl Triflates by Deuterium Labeling. J. Am. Chem. Soc. 2014, 136, 15757− 15766. (22) Sather, A. C.; Lee, H. G.; De La Rosa, V. Y.; Yang, Y.; Müller, P.; Buchwald, S. L. A Fluorinated Ligand Enables Room-Temperature and

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected], ritter@mpi-muelheim. mpg.de. ORCID

Tobias Ritter: 0000-0002-6957-450X Funding

We thank the Patty and Michael Phelps Foundation, the National Institutes of Health (NIH) National Institute of General Medical Sciences (GM088237), and the Max-PlanckInstitut für Kohlenforschung for funding. Notes

The authors declare the following competing financial interest(s): T.R. may benefit financially from PhenoFluor or PhenoFluorMix sales. Biographies Constanze N. Neumann completed her undergraduate education at the University of Oxford, where she worked under the supervision of Timothy J. Donohoe. Constanze completed her doctoral studies at Harvard University in the laboratory of Tobias Ritter, where she worked on fluorination chemistry. She is currently a postdoctoral researcher working for Mircea Dincă at MIT. Tobias Ritter was born in 1975 in Lübeck, Germany. He studied chemistry in Braunschweig, Bordeaux, Lausanne, and Stanford. After research with Prof. Barry M. Trost at Stanford, he obtained his Ph.D. in 2004, working with Prof. Erick M. Carreira at ETH Zurich. He then carried out postdoctoral research with Prof. Robert H. Grubbs at Caltech. In 2006, he was appointed as Assistant Professor in the Department of Chemistry and Chemical Biology at Harvard, and he was promoted to Associate Professor in 2010 and to Professor of Chemistry and Chemical Biology in 2012. He is also on the faculty at Massachusetts General Hospital in the Department of Radiology. In 2015 he accepted the position of Director of Organic Chemistry at the Max-Planck-Institut für Kohlenforschung. 2832


Article

Accounts of Chemical Research Regioselective Pd-Catalyzed Fluorination of Aryl Triflates and Bromides. J. Am. Chem. Soc. 2015, 137, 13433−13438. (23) Fier, P. S.; Hartwig, J. F. Copper-Mediated Fluorination of Aryl Iodides. J. Am. Chem. Soc. 2012, 134, 10795−10798. (24) Furuya, T.; Benitez, D.; Tkatchouk, E.; Strom, A. E.; Tang, P.; Goddard, W. A.; Ritter, T. Mechanism of C−F Reductive Elimination from Palladium(IV) Fluorides. J. Am. Chem. Soc. 2010, 132, 3793− 3807. (25) Furuya, T.; Ritter, T. Carbon−Fluorine Reductive Elimination from a High-Valent Palladium Fluoride. J. Am. Chem. Soc. 2008, 130, 10060−10061. (26) Tang, P.; Furuya, T.; Ritter, T. Silver-Catalyzed Late-Stage Fluorination. J. Am. Chem. Soc. 2010, 132, 12150−12154. (27) Furuya, T.; Strom, A. E.; Ritter, T. Silver-Mediated Fluorination of Functionalized Aryl Stannanes. J. Am. Chem. Soc. 2009, 131, 1662− 1663. (28) Neumann, C. N.; Ritter, T. Late-Stage Fluorination: Fancy Novelty or Useful Tool? Angew. Chem., Int. Ed. 2015, 54, 3216−3221. (29) Nielsen, M. K.; Ugaz, C. R.; Li, W.; Doyle, A. G. PyFluor: A Low-Cost, Stable, and Selective Deoxyfluorination Reagent. J. Am. Chem. Soc. 2015, 137, 9571−9574. (30) Bennua-Skalmowski, B.; Klar, U.; Vorbrüggen, H. AmineInduced Decomposition of Perfluoroalkanesulfonyl Fluorides. Synthesis 2008, 2008, 1175−1178. (31) Tang, P.; Wang, W.; Ritter, T. Deoxyfluorination of Phenols. J. Am. Chem. Soc. 2011, 133, 11482−11484. (32) Schimler, S. D.; Cismesia, M. A.; Hanley, P. S.; Froese, R. D. J.; Jansma, M. J.; Bland, D. C.; Sanford, M. S. Nucleophilic Deoxyfluorination of Phenols via Aryl Fluorosulfonate Intermediates. J. Am. Chem. Soc. 2017, 139, 1452−1455. (33) Sladojevich, F.; Arlow, S. I.; Tang, P.; Ritter, T. Late-Stage Deoxyfluorination of Alcohols with PhenoFluor. J. Am. Chem. Soc. 2013, 135, 2470−2473. (34) Fujimoto, T.; Becker, F.; Ritter, T. PhenoFluor: Practical Synthesis, New Formulation, and Deoxyfluorination of Heteroaromatics. Org. Process Res. Dev. 2014, 18, 1041−1044. (35) Fujimoto, T.; Ritter, T. PhenoFluorMix: practical chemoselective deoxyfluorination of phenols. Org. Lett. 2015, 17, 544−547. (36) Goldberg, N. W.; Shen, X.; Li, J.; Ritter, T. AlkylFluor: Deoxyfluorination of Alcohols. Org. Lett. 2016, 18, 6102−6104. (37) L’Heureux, A.; Beaulieu, F.; Bennett, C.; Bill, D. R.; Clayton, S.; LaFlamme, F.; Mirmehrabi, M.; Tadayon, S.; Tovell, D.; Couturier, M. Aminodifluorosulfinium Salts: Selective Fluorination Reagents with Enhanced Thermal Stability and Ease of Handling. J. Org. Chem. 2010, 75, 3401−3411. (38) Jacobsen, H.; Donahue, J. P. Expanding the scope of the Newman−Kwart rearrangement  A computational assessment. Can. J. Chem. 2006, 84, 1567−1574. (39) Lloyd-Jones, G. C.; Moseley, J. D.; Renny, J. S. Mechanism and Application of the Newman-Kwart O→S Rearrangement of O-Aryl Thiocarbamates. Synthesis 2008, 2008, 661−689. (40) Fry, S. E.; Pienta, N. J. Effects of molten salts on reactions. Nucleophilic aromatic substitution by halide ions in molten dodecyltributylphosphonium salts. J. Am. Chem. Soc. 1985, 107, 6399−6400. (41) Glukhovtsev, M. N.; Bach, R. D.; Laiter, S. Single-Step and Multistep Mechanisms of Aromatic Nucleophilic Substitution of Halobenzenes and Halonitrobenzenes with Halide Anions: Ab Initio Computational Study. J. Org. Chem. 1997, 62, 4036−4046. (42) Neumann, C. N.; Hooker, J. M.; Ritter, T. Concerted nucleophilic aromatic substitution with 19F− and 18F−. Nature 2016, 534, 369−373. (43) Sawyer, C. B.; Kirsch, J. F. Kinetic isotope effects for reactions of methyl formate-methoxyl-18O. J. Am. Chem. Soc. 1973, 95, 7375− 7381. (44) Bronner, S. M.; Goetz, A. E.; Garg, N. K. Overturning Indolyne Regioselectivities and Synthesis of Indolactam V. J. Am. Chem. Soc. 2011, 133, 3832−3835.

(45) Picazo, E.; Houk, K. N.; Garg, N. K. Computational predictions of substituted benzyne and indolyne regioselectivities. Tetrahedron Lett. 2015, 56, 3511−3514. (46) Hunter, A.; Renfrew, M.; Rettura, D.; Taylor, J. A.; Whitmore, J. M. J.; Williams, A. Stepwise versus Concerted Mechanisms at Trigonal Carbon: Transfer of the 1,3,5-Triazinyl Group between Aryl Oxide Ions in Aqueous Solution. J. Am. Chem. Soc. 1995, 117, 5484−5491. (47) Renfrew, A. H. M.; Taylor, J. A.; Whitmore, J. M. J.; Williams, A. A single transition state in nucleophilic aromatic substitution: reaction of phenolate ions with 2-(4-nitrophenoxy)-4,6-dimethoxy-1,3,5triazine in aqueous solution. J. Chem. Soc., Perkin Trans. 2 1993, 1703−1704. (48) Renfrew, A. H. M.; Taylor, J. A.; Whitmore, J. M. J.; Williams, A. Nucleophilic aromatic substitution in heterocycles: alcoholysis and hydrolysis of 2-anilino-4,6-dichloro-1,3,5-triazines. J. Chem. Soc., Perkin Trans. 2 1994, 2389−2393. (49) Renfrew, A. H. M.; Taylor, J. A.; Whitmore, J. M. J.; Williams, A. Timing of bonding changes in fundamental reactions in solutions: pyridinolysis of a triazinylpyridinium salt. J. Chem. Soc., Perkin Trans. 2 1994, 2383−2384. (50) Dub, P. A.; Wang, H.; Matsunami, A.; Gridnev, I. D.; Kuwata, S.; Ikariya, T. C-F Bond Breaking through Aromatic Nucleophilic Substitution with a Hydroxo Ligand Mediated via Water Bifunctional Activation. Bull. Chem. Soc. Jpn. 2013, 86, 557−568. (51) Goryunov, L. I.; Grobe, J.; Le Van, D.; Shteingarts, V. D.; Mews, R.; Lork, E.; Würthwein, E.-U. Di- and Trifluorobenzenes in Reactions with Me2EM (E = P, N; M = SiMe3, SnMe3, Li) Reagents: Evidence for a Concerted Mechanism of Aromatic Nucleophilic Substitution. Eur. J. Org. Chem. 2010, 2010, 1111−1123. (52) Nawaz, F.; Mohanan, K.; Charles, L.; Rajzmann, M.; Bonne, D.; Chuzel, O.; Rodriguez, J.; Coquerel, Y. Temporary Intramolecular Generation of Pyridine Carbenes in Metal-Free Three-Component C−H Bond Functionalisation/Aryl-Transfer Reactions. Chem. - Eur. J. 2013, 19, 17578−17583. (53) Williams, A. The diagnosis of concerted organic mechanisms. Chem. Soc. Rev. 1994, 23, 93−100. (54) Sun, H.; DiMagno, S. G. Room-Temperature Nucleophilic Aromatic Fluorination: Experimental and Theoretical Studies. Angew. Chem., Int. Ed. 2006, 45, 2720−2725. (55) Giroldo, T.; Xavier, L. A.; Riveros, J. M. An unusually fast nucleophilic aromatic displacement reaction: the gas-phase reaction of fluoride ions with nitrobenzene. Angew. Chem., Int. Ed. 2004, 43, 3588−3590. (56) Costil, R.; Dale, H. J. A.; Fey, N.; Whitcombe, G.; Matlock, J. V.; Clayden, J. Heavily Substituted Atropisomeric Diarylamines by Unactivated Smiles Rearrangement of N-Aryl Anthranilamides. Angew. Chem., Int. Ed. 2017, 56, 12533−12537. (57) Beyzavi, M. H.; Mandal, D.; Strebl, M. G.; Neumann, C. N.; D’Amato, E. M.; Chen, J.; Hooker, J. M.; Ritter, T. 18FDeoxyfluorination of Phenols via Ru π-Complexes. ACS Cent. Sci. 2017, 3, 944−948. (58) Tredwell, M.; Preshlock, S. M.; Taylor, N. J.; Gruber, S.; Huiban, M.; Passchier, J.; Mercier, J.; Génicot, C.; Gouverneur, V. A General Copper-Mediated Nucleophilic 18F Fluorination of Arenes. Angew. Chem., Int. Ed. 2014, 53, 7751−7755. (59) Taylor, N. J.; Emer, E.; Preshlock, S.; Schedler, M.; Tredwell, M.; Verhoog, S.; Mercier, J.; Genicot, C.; Gouverneur, V. Derisking the Cu-Mediated 18F-Fluorination of Heterocyclic Positron Emission Tomography Radioligands. J. Am. Chem. Soc. 2017, 139, 8267−8276.

2833


Facile CâF Bond Formation through a ... - American Chemical Society

Recommend Documents

Facile CâF Bond Formation through a ... - American Chemical Society