Chemodivergent Csp³─F bond functionalization and cross-electrophile alkyl-alkyl coupling with alkyl fluorides

Abstract

The widespread use of fluorinated organic compounds in the health, agrochemical, and materials sciences is sustained by a steadily growing pool of commercially available fine chemicals. The synthetic utility of the increasingly ubiquitous Csp³─F bond, however, remains to be fully exploited, which is often a difficult task because of its paramount stability and chemical inertness. Here, we demonstrate chemodivergent activation of monofluoroalkyl compounds toward either nucleophilic or electrophilic intermediates. This is accomplished under conditions that are compatible with several reaction types and many functional groups, which drastically widens the current scope of organofluorine chemistry and sets the stage for carbon-carbon and carbon-heteroatom bond formations, stereoselective construction of bisoxindole alkaloid scaffolds via in situ Umpolung, and cross-electrophilic coupling methodology. The selective generation of either nucleophilic or electrophilic species and the possibility of doing so simultaneously or, alternatively, switching molecular polarity enable previously unidentified synthetic opportunities that recognize alkyl fluorides as chemodivergent building blocks.

Chemodivergent activation of the Csp³─F bond for reductive cross-coupling, Umpolung, and other high-yielding reactions is achieved.

INTRODUCTION

The carbon-fluoride bond has a very high bond dissociation energy (BDE >100 kcal/mol) and is the strongest covalent single bond that carbon can form. Practical transformations other than substitution, elimination, and cross-coupling reactions with aryl or alkenyl fluorides are rare (1, 2). The exceptional thermodynamic stability and kinetic inertness of the carbon-fluoride moiety are distinctive properties of organofluorines, which continue to be of increasing utility in the health, agrochemical, and materials sciences (3). The high demand has nourished the development of numerous fluorination methods, and an impressive variety of fluorinated organic compounds is now commercially available or easily prepared. This has raised increasing interest in organofluorine chemistry, and several groups have introduced hydrodefluorination protocols that convert C─F into C─H bonds (4–11). Alternatively, the abundance of fluorinated organic compounds presents opportunities for the development of synthetic methods that use the C─F functionality for C─C bond construction. Substantial progress has been made with aryl fluorides, while aliphatic substrates continue to be very challenging. Csp³─F bond functionalization has been achieved, for example, with reactive Lewis acids or via hydrogen bond assistance, and used for carbon-heteroatom bond formation (12–14), carbon-carbon coupling (15–20), and halide exchange (21–23). With the exception of allylic, propargylic, and benzylic fluorides (24–30), low functional group compatibility, limited synthetic utility and competing rearrangement, elimination, or other side reactions, particularly when harsh reaction conditions need to be applied, have remained persistent problems despite some recently reported remarkable advances (31–37). Moreover, monofluoroalkyl C─F abstraction methods reported to date typically preserve the electrophilic reactivity of the carbon atom and are therefore largely limited to transformations with nucleophiles. We envisioned that selective generation of either nucleophilic or electrophilic species, or possibly both at the same time, via mild Csp³─F bond activation tolerant of common functional groups would substantially alter existing Csp³─F bond functionalization capabilities and enable synthesis that is currently not possible.

Here, we present chemoselective and chemodivergent Csp³─F bond activation methodology that enables carbon-carbon and carbon-heteroatom bond formation using common organofluorines either as proelectrophiles or, based on a reductive fluoride abstraction mechanism that is promoted by abundant metal iodides, as pronucleophilic starting materials (Fig. 1A). This chemistry is broad in scope, adaptable to a notable variety of reactions, scalable, and generally high-yielding (38). The reductive Csp³─F bond activation strategy described here enables Mannich, Michael, aldol, and substitution reactions under base-free conditions and smooth synthesis of chlorides, bromides, selenides, sulfides, hydroxyl amines, and hydrazines at room temperature (Fig. 1B). Alternatively, the reductive carbon-carbon bond formation pathway can be modified to achieve highly stereoselective homocoupling via an iodine-mediated Umpolung mechanism, which generates an important alkaloid motif with two adjacent tetrasubstituted carbon centers (Fig. 1C). Last, an unprecedented reductive Csp³─Csp³ cross-coupling strategy that achieves complementary pronucleophilic and proelectrophilic C─F bond activation with stoichiometric amounts of two different alkyl fluoride compounds under compatible conditions is introduced (Fig. 1D). The diverse chemical space that can now be accessed through pronucleophilic and proelectrophilic C─F bond activation as well as the integration of both pathways into a one-pot procedure is highlighted with more than 65 examples.

Fig. 1. C─F bond functionalization and cross-electrophilic alkyl-alkyl coupling.

(A) Chemodivergent C─F bond activation for selective transformations with either electrophiles or nucleophiles. (B) Reductive C─F bond cleavage and reactions with heteroatom electrophiles. (C) In situ Umpolung and homocoupling. (D) Reductive cross-coupling based on complementary alkyl fluoride activation.

RESULTS AND DISCUSSION

Reductive C─F bond activation and Csp³─Csp³ bond formation with electrophiles

Because the cleavage of the C─F bond is highly endothermic, the relatively low BDE of a typical C─C bond (85 kcal/mol) does not provide the necessary thermodynamic driving force to achieve C─F functionalization under mild conditions. Reports on proelectrophilic C─F bond activation with ytterbium, silicon, and lanthanide Lewis acids (4, 5, 12, 14, 18, 21, 22, 30) and C─F bond cleavage observed in our laboratory during the synthesis and derivatization of fluorinated oxindoles, nitriles, nitromethyl ketones, as well as alkyl, benzyl and aryl fluorides (19, 39–42) encouraged us to investigate whether broadly applicable, chemodivergent C─F bond cleavage using alkyl fluorides either as proelectrophiles or as pronucleophiles is possible when coupled with the generation of a stable inorganic fluoride salt. Following our previous work with fluorinated oxindoles and through serendipitous screening of reaction conditions, we found that the fluorooxindole 1 undergoes reductive defluorination in the presence of trimethylsilyl iodide (TMSI). This results in pronucleophilic C─F bond activation, which is in stark contrast to the proelectrophilic C─F bond cleavage previously reported by Hilmersson and others (21, 22). We were able to trap the in situ generated nucleophile as the Michael addition product 3 in 76% yield (Fig. 2A, entry 1). The reductive C─F bond cleavage is accompanied by the formation of iodine or hypervalent iodine fluorides and trimethylsilyl fluoride (TMSF) (Si-F BDE: 135 kcal/mol), which was identified by ¹⁹F nuclear magnetic resonance (NMR) spectroscopy. We continued to examine a variety of metal halides that form stable fluoride salts (see the Supplementary Materials). The best results were obtained with LiI, YbI₃, and YI₃ in chlorinated solvents or ethers (entries 2 to 8). These are safe, practical, and sustainable defluorinating agents and may be generated in situ or used in substoichiometric amounts (entries 9 and 10). Ytterbium and yttrium are more earth abundant than tin, which is frequently encountered in the form of aryl stannanes in Stille couplings. The yield and rate of the C─F bond functionalization are not affected by equimolar amounts of (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO), and all mechanistic observations favor a two-electron transfer process over a radical mechanism. This assumption is in agreement with the result of a radical clock experiment using compound 4, which gave the expected Michael addition product 5 in quantitative amounts, while the radical cyclization product 6 was not detected. In addition, we were able to trap the intermediate nucleophile generated from fluorooxindole 7 by a deuteriodefluorination reaction with D₂O, which gave 8 in 98% yield.

Fig. 2. Selective Csp³─F bond functionalization.

(A) Development of reductive C─F bond cleavage and use of alkyl fluorides as pronucleophiles. (B) Scope of the carbon-carbon bond formation based on pronucleophilic C─F bond activation. (C) Applications in carbon-heteroatom bond formation.

Having established the mechanistic underpinnings of the reductive C─F bond activation, a series of pronucleophilic organofluorines and several electrophiles were tested to determine the scope of the Csp³─Csp³ bond formation (Fig. 2B). We were pleased to find that reductive activation of primary, secondary, and tertiary fluorides is possible, and the base-free generated nucleophiles can be used to produce the multifunctional compounds 9 to 24 in 81 to 98% yield at room temperature. The efficient conversion of tertiary C─F moieties into sterically congested all-carbon quaternary carbon centers as exemplified with 3, 5, 9 to 11, and 14 to 21 is particularly noteworthy. The reaction scale can be increased without compromising yields, and we prepared 2 g of 3 in almost quantitative amounts. Many functionalities including Csp²─F, CF₂, and CF₃ moieties are tolerated. While dimethylaminomethylation of 25 gave 24 in 81% yield, 27 and 28 did not react under identical conditions, and starting materials were quantitatively recovered. The C─F bond activation occurs under base-free conditions that streamline crossed aldol reaction chemistry.

Traditionally, the reaction between two enolizable carbonyl compounds requires prior conversion of one into a silyl enol ether or enamine. By contrast, we obtained product 30 directly from 25 and 29 in 72% yield, eliminating the common derivatization and precautions necessary to avoid self-condensation, enolate scrambling, and elimination side reactions. Similarly, a competition experiment between 25 and 31 exhibiting almost identical pK_a values showed that the C─C bond formation outperforms uncontrolled proton transfer between enolizable substrates. The crossed aldol reaction between 25 and 4-nitrobenzaldehyde, 32, in the presence of equimolar amounts of 31 gave 33 in 79% yield, while 34, which would have originated from enolate scrambling, was not observed.

Carbon-heteroatom bond formation

Our reductive C─F bond activation method also enables heteroatom incorporation (Fig. 2C). To demonstrate the diversity of possible functionalizations, we chose to use the nitrogen, halogen, sulfur, and selenium electrophiles 35 to 40. Halogenations with N-bromosuccinimide (NBS) and N-chlorosuccinimide (NCS) gave 41 and 42 in 82 to 86% yield. The tertiary sulfides and selenides 43 to 48 were produced in 81 to 96% yield. Last, the hydrazides 49 and 50 and the hydroxylamine 51 were obtained in 79 to 84% yield. All reactions occurred at room temperature, which underscores the general efficiency and practicality of this method. The variety of transformations and the inherent functional group tolerance are expected to inspire the strategic use of the typically inert C─F bond for selective carbon-carbon and carbon-heteroatom bond construction.

Proelectrophilic C─F bond activation

Since simple alkyl, allyl, and benzyl fluorides have been used as proelectrophilic starting materials (13, 17, 20, 43, 44), we decided to explore whether this is also possible with deactivated α-fluorinated carbonyl compounds. Initial screening of Brønsted acids was rather unsuccessful, but good to high conversion occurred when boron trifluoride, triphenylboron, and tris(pentafluorophenyl)boron were used as catalyst. We found that fluorooxindole 1 reacts smoothly in the presence of 5 mol% of tris(pentafluorophenyl)boron to the Friedel-Crafts alkylation product 52 in 78% at room temperature (Fig. 3). A slight increase of the reaction temperature to 60°C gave 52 in quantitative yields. A radical clock reaction experiment with the cyclopropyl compound 53 is in agreement with an ionic mechanism, and we isolated 55 in quantitative amounts, while the radical ring-opening by-product 56 was not detected. With this protocol in hand, we successfully used several α-fluorinated carbonyl compounds in Friedel-Crafts reactions, producing 57 to 59 in 83 to 99% yield. Heterocyclic substrates can also be used, and we were able to convert several indole, benzofuran, thiophene, and benzothiophene compounds to 60 to 64 in 72 to 94% yield. Alternatively, the intermediate electrophiles can be trapped with O- and S-nucleophiles as the ethers, sulfides, or thioesters 65 to 72, which were obtained in 71 to 92% yields. The chemistry shown in Figs. 2 and 3 demonstrates orthogonal activation of monofluorinated compounds toward either nucleophilic or electrophilic intermediates, which introduces unprecedented reactivity control and chemodiversity that are not accessible with other alkyl halides or from the parental C─H compounds. Because the chemodivergence is accomplished under conditions that are compatible with a wide range of reaction types, it drastically alters the current synthetic utility of organofluorines, and it sets the stage for unprecedented coupling reactions, vide infra.

Fig. 3. Selective Csp³─F bond functionalization and reactivity reversal.

(A) Complementary use of α-fluorinated carbonyl compounds as proelectrophiles. (B) Scope of the carbon-carbon and carbon-heteroatom bond formation with C-, O-, and S-nucleophiles.

Reductive homo- and cross-coupling of alkyl fluorides

Corriu-Kumada, Suzuki, Stille, Negishi, and Hiyama couplings require prior conversion of one reaction partner into a Grignard, boronic acid, stannane, zinc, or silyl compound. To overcome this drawback, reductive and photoredox methods for bond construction with alkyl and aryl chlorides, bromides, iodides, carboxylates, sulfonates, or pseudohalides have been introduced (45–57). The prospect of direct alkyl halide cross-coupling is paradigm-shifting in the realm of organic synthesis because it is substantially more efficient and minimizes cost and waste production. We therefore decided to explore the possibility of cross-electrophile C─C bond formation using organofluorines without the common need for transition metal catalysis with a separately prepared organometallic reagent. We envisioned that the reductive defluorination pathway described above would provide a unique entry toward direct Csp³─Csp³ cross-coupling of two different alkyl fluoride compounds, which is currently not possible.

C─F bond functionalization based on in situ Umpolung

During reaction analysis of the reductive defluorination with fluorooxindoles, we found a highly stereoselective Umpolung pathway that prevails when a suitable electrophile for the consumption of the initially generated nucleophile is absent (Fig. 4A). We first observed that N-methyl 3-fluoro-3-methyloxindole, 7, gives the homochiral dimer 73 carrying two adjacent tetrasubstituted carbon centers in 51% yield in the presence of one equivalent of LiI. Further optimization showed that 73 can be formed with excellent yields and in a diastereomeric ratio greater than 99:1 at room temperature. Nearly quantitative yields and exclusive formation of the homochiral dimers 74 and 75 were obtained using the same protocol. The high-yielding formation of 75 carrying two aryl fluoride moieties underscores the selectivity of this method for aliphatic C─F bonds. Defluorinative in situ Umpolung homodimerization thus addresses a long-standing problem in bisoxindole alkaloid synthesis. The fabrication of the congested alkaloid scaffold in 73 to 75 has been challenging and is typically accomplished through multistep synthesis or by radical oxindole dimerization, which suffers from low diastereoselectivity or low yields (58–60). We postulate that the initially formed intermediate A reacts with concurrently generated iodine toward B, which spontaneously gives the iminium Michael acceptor C. This short-lived species then combines with subsequently generated A to the bisoxindole product. The reaction outcome was confirmed by crystallographic analysis of 73 and 74, and we obtained proof for the generation of intermediate A by quantitative deuteriodefluorination upon addition of D₂O (61). The proposed role and consumption of iodine, which acts as a redox switch in the Umpolung mechanism, is in agreement with the absence of the characteristic I₂ color during the reaction and was verified by a trapping experiment with KPPh₂.

Fig. 4. Reductive homo- and cross-coupling with orthogonally reacting alkyl fluorides.

(A) Carbon-fluoride bond Umpolung chemistry. (B) General cross-coupling protocol, optimization, and mechanistic study. (C) Scope of the reductive alkyl fluoride cross-coupling reaction.

Cross-electrophilic coupling of orthogonally activated alkyl fluorides

To achieve direct Csp³─Csp³ cross-coupling with different types of alkyl fluorides, complementary activation pathways that do not interfere with one another but selectively generate nucleophilic and electrophilic species under the same reaction conditions are required. The LiI or YbI₃ induced C─F bond activation, and Umpolung chemistry discussed above set the stage for us to succeed with this task. To suppress the Umpolung pathway and other side reactions, we studied the effect of various iodine scavengers using the coupling between N-phenyl 3-fluoro-3-methyloxindole, 1, or 2-fluoro-2-methylindanone, 76, and n-fluorooctane, 77, as a model reaction. We found that both Yb(0) and charcoal are quite effective, and we were able to prepare the cross-coupling products 78 and 79 in 84% yield, a remarkable efficiency for cross-electrophilc coupling methodologies (Fig. 4B). On the basis of our mechanistic analysis discussed above, this reaction is expected to involve an ionic mechanism in which the metal iodide achieves reductive defluorination of pronucleophilic 1 or 76 and parallel activation of the proelectrophilic alkyl fluoride 77 devoid of an electron-withdrawing neighboring group for a direct attack or via fluoride/iodide exchange. The reductive defluorination chemistry including the high-yielding deuteriodefluorination of 7 shown in Fig. 2 is in perfect agreement with the generation of an intermediate anionic species, and we were able to confirm the formation of alkyl iodides from the proelectrophilic fluoride compounds by gas chromatography–mass spectrometry analysis, which was expected on the basis of previous reports from the Hilmersson group (22). LiI and YbI₃ thus fulfill at least two crucial functions by achieving activation of pronucleophilic and proelectrophilic organofluorines in complementary processes that set the stage for subsequent nucleophilic substitution. The exclusion of homolytic C─F bond cleavage pathways is in agreement with the results of radical clock reaction and radical scavenging experiments. When we used 80 and 4-nitrobenzyl fluoride in the reductive cross-coupling protocol, we isolated 82 in 87% yield, while the cyclization product 83 indicative of a radical pathway was not observed. The addition of the radical trap TEMPO to the reaction between 1 and 81 did not show any effect, and we obtained 84 in 88% yield, which is also consistent with an ionic mechanism. To demonstrate the general utility of this chemistry, we selected a variety of proelectrophilic and pronucleophilic fluorides, which are used in stoichiometric amounts (Fig. 4C). The yields of the Csp³─Csp³ cross-coupling products 78, 79, 82, and 84 to 96 were high in all cases, and we did not detect the corresponding homocoupling compounds. Thus, construction of alkyl-alkyl bonds displaying secondary, tertiary, and quaternary carbon atoms directly from two orthogonally activated alkyl fluorides is demonstrated. The integration of distinct C─F bond activation pathways allows selective Csp³─Csp³ bond formation with remarkable functional group tolerance, while yields are not compromised by competing β-hydride elimination, chain-walking, or homocoupling processes that occur during late-transition metal catalysis. We expect that the discovery of the orthogonal reactivity pattern of alkyl fluorides will become a very useful synthetic strategy for selective cross-coupling of multifunctional building blocks that carry a C─F moiety intentionally positioned for chemodivergent functionalization.

In summary, we have introduced broadly applicable C─F bond functionalization chemistry that exploits alkyl fluorides selectively as pronucleophilic or proelectrophilic starting materials, including the possibility of doing so simultaneously in a one-pot protocol with orthogonally activated alkyl fluorides or by redox-switching of molecular polarity, which, together, allows the synthesis of a diverse range of compounds. Chemodivergent C─F bond scission can now be used to achieve carbon-carbon and carbon-heteroatom bond formation via complementary reactivity patterns that are not accessible by other alkyl halides or parental C─H compounds. While proelectrophilic C─F bond activation with strong Lewis acids or fluorophilic lanthanides have been reported, we demonstrate with many examples that reductive defluorination can generate nucleophiles instead. The Csp³–F bond activation pathways presented here generally allow high-yielding, scalable synthesis that occurs with wide functional group tolerance under mild conditions. This enables previously unidentified synthetic methodologies such as highly stereoselective, base-free formation of homochiral alkaloid scaffolds via in situ Umpolung and direct cross-coupling of distinctively activated pronucleophilic and proelectrophilic organofluorines. The variety of transformations found in this study demonstrates that the C─F bond, generally considered inert under conventional synthesis conditions, can be selectively activated and transformed in numerous ways. This drastically alters the synthetic space of the Csp³─F bond and is likely to spur the development of previously unknown reactions and strategies that use alkyl fluorides in complementary nucleophilic and electrophilic reaction mechanisms.

MATERIALS AND METHODS

General information

Commercially available organofluorines, oxindoles, isatins, metals, metal salts, reagents, and solvents were used as purchased without further purification. Solvents were stored over 4-Å molecular sieves before use. YbI₃(THF)₃ (22) and 3-fluorooxindoles (19, 62–64) were synthesized by following the literature procedure. NMR spectra were obtained at 400 MHz (¹H NMR), 376 MHz (¹⁹F NMR), and 100 MHz (¹³C NMR) in deuterated solvents. All reaction products were purified by column chromatography on silica gel (particle size, 40 to 63 μm). All unknown compounds were fully characterized by ¹H NMR, ¹³C NMR, and, if applicable, ¹⁹F NMR spectroscopy and either high-resolution mass spectrometry or elemental analysis (see the Supplementary Materials). The NMR spectra of the previously reported products 2,3-diphenylpropanenitrile (22) (65), 3-hydroxy-1,5-diphenylpentan-1-one (30) (66), 3-hydroxy-3-(4-nitrophenyl)-1-phenylpropan-1-one (33) (67), and methyl 2-(3-methylbenzofuran-2-yl)-2-phenylacetate (63) (68) were in accordance with the literature.

General procedure for pronucleophilic Csp³─F bond activation and functionalization with carbon electrophiles

To a solution of an organofluorine (0.1 mmol) and an electrophile (0.1 to 0.5 mmol) in dry CH₂Cl₂ (0.5 ml) was added YbI₃(THF)₃ (0.15 mmol) under nitrogen atmosphere. The mixture was stirred at 25°C for 20 hours in the dark. The crude products were purified by flash chromatography on silica gel using mixtures of hexanes and ethyl acetate with a ratio ranging from 1:1 to 9:1 (v/v) as mobile phase.

General procedure for halogenation reactions

YbI₃(THF)₃ (0.15 mmol) was added to a solution of 2-fluoro-2-methyl-2,3-dihydro-1H-inden-1-one (0.1 mmol) in dry CH₂Cl₂ (0.5 ml) under nitrogen atmosphere and stirred at 25°C for 8 hours in the dark. An N-halosuccinimide (0.5 mmol) was added, and the reaction mixture was stirred for another 12 hours. The crude product was purified by flash chromatography on silica gel using hexanes–ethyl acetate (4:1, v/v) as mobile phase.

General procedure for the synthesis of thio- and selenoethers

To a mixture of an organofluorine (0.1 mmol) and an electrophile (0.15 mmol) in dry CH₂Cl₂ (0.5 ml) was added YbI₃(THF)₃ (0.15 mmol) under nitrogen atmosphere. The mixture was stirred at 25°C for 20 hours in the dark. The crude product was purified by flash chromatography on silica gel using hexanes–ethyl acetate (4:1, v/v) as mobile phase.

General procedure for the formation of C─N bonds

YbI₃(THF)₃ (0.15 mmol) was added to a solution of an organofluorine (0.1 mmol) in dry CH₂Cl₂ (0.5 ml) under nitrogen atmosphere and stirred at 25°C for 4 hours in the dark. Diethyl azodicarboxylate (0.2 mmol) was added, and the reaction mixture was stirred for another 16 hours. The crude product was purified by flash chromatography on silica gel using hexanes–ethyl acetate (1:1, v/v) as mobile phase.

General procedure for proelectrophilic Csp³─F bond activation

To a solution of an organofluorine (0.2 mmol) and a nucleophile (0.4 mmol) in nitromethane (0.2 ml) was added B(C₆F₅)₃·H₂O (5 mol%). The mixture was stirred at 60° to 100°C. At completion, the reaction mixture was directly dry-loaded onto silica, and the crude product was purified by flash chromatography on silica gel using hexanes–ethyl acetate ranging from 1:1 to 9:1 (v/v) as mobile phase.

General procedure for Umpolung and homocoupling

To a stirred solution of a 3-fluoroindolin-2-one compound (0.1 mmol, 1.0 eq.) in dry CH₂Cl₂ (1.0 ml) was added YbI₃(THF)₃ (0.15 mmol, 1.5 eq) under nitrogen atmosphere. The reaction mixture was stirred at 25°C until completion in the dark unless noted otherwise. The solvent was removed, and the crude products were purified by silica flash column chromatography using hexanes–ethyl acetate ranging from 1:1 to 9:1 (v/v) as mobile phase.

General procedure for the reductive cross-coupling of two alkyl fluorides

To a mixture of a pronucleophilic organofluorine (0.2 mmol), a proelectrophilic organofluorine (0.3 mmol), and Yb powder (0.1 mmol) in dry CH₂Cl₂ (0.5 ml) was added YbI₃(THF)₃ (0.5 mmol) under nitrogen atmosphere and stirred at 25°C for 3 hours in the dark. Then, 1.0 ml of dry THF was added to the mixture and heated to 70° to 90°C for another 18 hours in the dark. After cooling, the solvent was removed, and the products were purified by silica flash column chromatography using mixtures of hexanes and ethyl acetate with a ratio ranging from 1:1 to 9:1 (v/v) as mobile phase.

Crystallographic analysis

Single crystal x-ray analysis was performed at 100 K using a Siemens platform diffractometer with graphite monochromated molybdenum (Mo)-Kα radiation (λ = 0.71073 Å). Data were integrated with the Bruker SAINT program. Structure solution and refinement was performed using the SHELXT/PC suite and ShelXle. Intensities were corrected for Lorentz and polarization effects, and an empirical absorption correction was applied using Blessing’s method as incorporated into the program SADABS. Nonhydrogen atoms were refined with anisotropic thermal parameters, and hydrogen atoms were included in idealized positions. We were able to grow a single crystal of 1,1′,3,3′-tetramethyl-[3,3′-biindoline]-2,2′-dione (73) by slow evaporation of a solution of hexanes and dichloromethane (4:1). Crystal data: C₂₀H₂₀N₂O₂, M = 320.38, colorless block, 0.265 by 0.224 by 0.083 mm³, monoclinic, space group P2₁/n, a = 12.345(2), b = 9.1739(16), c = 15.708(3) Å, V = 1718.6(5) ų, Z = 4. A single crystal of 3,3′-dimethyl-1,1′-diphenyl-[3,3′-biindoline]-2,2′-dione (74) was obtained by slow evaporation of a solution of hexanes and dichloromethane (4:1). Crystal data: C₃₀H₂₄N₂O₂, M = 444.51, colorless block, 0.332 by 0. 288 by 0.276 mm³, triclinic, space group P-1, a = 9.0931(11), b = 10.4352(12), c = 12.6056(15) Å, V = 1131.6(2) ų, Z = 2. A single crystal of 2-methyl-2-(4-nitrobenzyl)-2,3-dihydro-1H-inden-1-one (96) was obtained by slow evaporation of a hexanes:ethyl acetate (4:1) solution. Crystal data: C₁₇H₁₅NO₃, M = 281.30, colorless block, 0.441 by 0.390 by 0.370 mm³, monoclinic, space group P2₁/c, a = 15.5244(11), b = 7.4745(5), c = 12.0072(8) Å, V = 1388.93(16) ų, Z = 4.

Supplementary Materials

This PDF file includes:

Tables S1 and S4

Figs. S1 to S140