Activation of Quinolines by Cationic Chalcogen Bond Donors

^§ Both authors contributed equally

Published as part of the Cluster Organosulfur and Organoselenium Compounds in Catalysis

Abstract

The application of already established as well as novel selenium- and sulfur-based cationic chalcogen bond donors in the catalytic activation of quinoline derivatives is presented. In the presence of selected catalysts, rate accelerations of up to 2300 compared to virtually inactive reference compounds are observed. The catalyst loading can be reduced to 1 mol% while still achieving nearly full conversion for electron-poor and electron-rich quinolines. Contrary to expectations, preorganized catalysts were less active than the more flexible variants.

Key words

chalcogen bonding - Lewis acids - organocatalysis - noncovalent interactions - intermolecular interactions - chalcogens - reduction

In the field of organocatalysis, noncovalent interactions play an important role, which was so far dominated by hydrogen bonding.[1] [2] In the last years, however, less typical noncovalent interactions such as anion-π [3,4] halogen,[5–8] and chalcogen[9] [10] [11] [12] bonding has become a focus of current research and have predominantly been applied in supramolecular assembly and anion recognition processes. To date, both anion-π and halogen bonding (XB) have been introduced in organocatalysis, whereas the application of chalcogen bonding (ChB) in this area is far less studied.

Chalcogen bonding is based on electrophilic chalcogen substituents in organic or inorganic compounds, which act as Lewis acidic centers. The origin of the interaction with Lewis bases can be explained by an n → σ* type orbital interaction (Figure [1]) and by electrostatic attraction, since the chalcogen substituents feature a region of positive electrostatic potential (σ-hole) in the elongation of the R–Ch bond.[13] [14] [15] In comparison to halogen bonding, chalcogen-bond donors possess two of these electron-deficient axes in a perpendicular arrangement. Furthermore, the Lewis acidity may be fine-tuned by variation of both substituents R and R′ (Figure [1]). In solid-state chemistry and crystal engineering, chalcogen bonding has increasingly been utilized in the last 20 years or so.[11,15,16] As an intramolecular interaction, it has also been used in organic synthesis since the pioneering works of Tomoda[17] and Wirth.[18] Pioneering studies on intermolecular interactions in solution were recently published by Taylor[19] [20] and Beer[21] [22] [23] in anion-recognition processes, and in 2017 the group of Matile[24] [25] and our group[26] [27] reported the first applications of intermolecular chalcogen bonding in organocatalysis: quinoline derivatives were activated by neutral ChB donors, while cationic ChB donors were used in halide abstraction reactions. Two further reactions of the latter type were reported subsequently.[28] [29]

However, the application of cationic ChB catalysts in the activation of neutral compounds is still unexplored, even though a significantly stronger activation should be expected compared with neutral catalysts. This is particularly true since in these reactions – in contrast to halide abstractions – no interfering anion (leaving from the substrate) is present.

Figure 1 Chalcogen bonding as n → σ* type interaction between a Lewis base (LB) and a chalcogen bond donor (Ch = S, Se, Te)

As a simple benchmark reaction, the reduction of quinoline derivatives (Scheme [1]) was chosen for several reasons: (a) most importantly, a series of reports is available with data on the performance of halogen or (neutral) chalcogen bonding organocatalysts;[24] [25] [30] (b) the reaction can be easily monitored via ¹H NMR spectroscopy; (c) a large number of quinoline derivatives can be employed; and (d) no background reactivity is observed at room temperature in the absence of activating agent (Table [1], entry 1). It should be noted that, at the same time, the reaction is not without its issues, particularly with regard to halogen bonding, as the action of hidden acid traces such as HI cannot easily be ruled out.[31] [32] This is less relevant for chalcogen bonding, though, as the generation of acid seems much less likely.

As cationic ChB donors, sulfur- and selenium-based imidazolium and benzimidazolium derivatives were chosen, with the selenium compounds already having been applied in halide abstraction reactions.[26] [27]

Scheme 1 Reduction of different quinoline derivatives 1a–d with Hantzsch ester 2 (2.4 equiv) in the presence of sulfur- and selenium-based cationic chalcogen bond donors

The preparation of all ChB donors investigated herein followed the same synthetic route, which is given in Scheme [2] (yields for each step and more details are provided in the Supporting Information). The first step involves the alkylation of the bis-heterocyclic compound, followed by introduction of the chalcogen under basic conditions and a second alkylation of the resulting chalcogenourea to yield the final catalysts as triflate salts.[26]

Scheme 2 Schematic syntheses of different sulfur- and selenium-based chalcogen bond donors. (i) C₈H₁₇OTf (2.5 equiv), anhyd. CH₂Cl₂ (0.1 m), r.t., 24 h; (ii) Ch(2.5 equiv), Cs₂CO₃ (2.5 equiv), anhyd. MeOH (0.1 m), reflux, 24 h; (iii) MeOTf (2.5 equiv), anhyd. CH₂Cl₂ (0.1 m), r.t., 24 h; Ch = S, Se.

The tested ChB donors and reference compounds are shown in Figure [2]. Each contains a characteristic cationic bis(benz)imidazolium scaffold as well as octylated side chains for improved solubility. Some of the compounds feature a trifluoromethyl group on the central benzene core. This serves two purposes: it allows easy tracking of potential catalyst decomposition via ¹⁹F NMR spectroscopy and it allows the separation of syn- and anti-isomers. Based on the well-known supramolecular concept of preorganization, the former is expected to be more active than analogous compounds with freely rotating chalcogen bonding moieties (such as 8 or 9).

Table 1 Performance of Reference Compounds (Figure [2]) in the Reduction of 2-Phenylquinoline (1a)

Entry	Catalyst	mol%a	Yield of 3a (%)b
1	–	–	<5
2	8H	1	64
3	9H	1	53
4	syn/anti-10H c	1	18
5	anti-10Cl	1	<5
6	syn-10I	1	88
7	11S	1	<5
8	11Se	1	<5
9	12S	1	<5
10	12Se	1	<5
11	syn-13S	1	<5
12	anti-13S	1	<5
13	anti-13Se	1	<5

^a Catalyst loading with respect to the amount of 1a.

^b After 12 h reaction time. Determined via ¹H NMR spectroscopy and relative to TES as internal standard. Averaged values of three measurements. A measurement / analysis error of ±10% is assumed.

^c Inseparable 40:60 mixture of syn-10^H and anti-10^H .

Figure 2 Overview of the tested chalcogen bond donors and reference compounds; Ch = S, Se (except syn-13); X = H, I, Cl (only anti-10).

First orientating experiments were performed with our previously most active ChB catalyst[26] [27] syn-10^SeMe and 2-phenylquinoline (1a) as substrate to find an optimal reaction profile regarding reaction time, catalyst loading, and overall concentration (in view of solubility issues of compound 2). In the presence of 1 mol% of catalyst syn-10^SeMe , full conversion of the reaction in CDCl₃ (0.042 mm) was observed after 12 h.[33] Accompanying ¹⁹F NMR measurements (see the Supporting Information) confirmed the stability of the catalyst under these reaction conditions. This was also corroborated by a second run, in which further starting material 1a was added to a fully converted reaction mixture, giving a virtually identical yield of product 3a and thus ruling out the action of accumulated decomposition products.

With this promising lead result in hand, various reference compounds (Figure [2]) were investigated to dismiss an activation that is not based on chalcogen bonding. Surprisingly, however, the non-selenated compounds syn/anti-10^H , 8^H and 9^H were also active (Table [1], entries 2–4). For instance, the most active reference compound 8^H gave 64% yield of compound 3a after 12 h reaction time. To exclude the possibility that the observed activation is solely driven by cation-π interactions or hydrogen bonds via the backbone protons, the chlorinated reference compound anti-10^Cl was tested.[34] As previous work had shown that its chlorine substituents do not form halogen bonds,[35] [36] its potential modes of action (cation-π and hydrogen bonding) would be shared by the core structure of syn-10^SeMe . The chlorinated compound showed no activity (Figure [3]; Table [1], entry 5), which strongly indicates that the activity of syn-10^SeMe can indeed be ascribed to chalcogen bonding. As a further consequence, the performance of compounds syn/anti-10^H , 8^H and 9^H is likely based on hydrogen bonding. To verify these results, reference compound syn-10^I and urea derivatives 11–13 (Figure [2] and Scheme [2]) were also employed as catalyst candidates. The iodinated compound syn-10^I provided 88% yield of 3a after 12 h reaction time, in line with expectations for this strong XB donor (Table [1], entry, 6). Compounds 11–13 showed no activity (Table [1], entries 7–13), likely due to the lack of chalcogen bonding options and strong π-π interactions, respectively.

Based upon these findings, all remaining catalyst candidates shown in Figure [2] were tested in the benchmark reaction (Table [2], entries 1–8). Both sulfur- and selenium-based bisbenzimidazolium catalysts 9–10 were similarly active and induced nearly quantitative conversion of compound 1a after 12 h. This is somewhat unexpected given the higher polarizability of the heavier chalcogen, which should form stronger ChB donors. It is also interesting to note that the more flexible core structure in catalysts 9^SMe and 9^SeMe is slightly superior to more preorganized framework structures such as in catalysts 10. For the latter, the difference in activity between the syn- and anti-atropisomer is negligible in the observed timeframe. More precisely, catalyst 9^SeMe yielded 95% of compound 3a, whereas syn-10^SeMe and anti-10^SeMe provided 91% and 90% of the product, respectively (Table [2], entries 4, 6 and 8). These findings could potentially be explained by the fact that syn-10^SeMe may not bind in a clean bidentate fashion to the substrate. Its core structure, it should be stressed, was developed for halogen bond donors interacting with halides. Thus, it seems that the more flexible backbone of 9^SeMe is better able to adapt to the coordination of the target. The corresponding sulfur-based catalysts show the same trend (Table [2], entries 3, 5 and 7).

Table 2 Overview of the Cationic Sulfur- and Selenium-Based Catalysts (Figure [1]) in the Reduction of 2-Phenylquinoline (1a)

Entry	Catalyst	Mol%a	Yield of 3a (%)b
1	8SMe	1	15
2	8SeMe	1	21
3	9SMe	1	94
4	9SeMe	1	95
5	syn-10SMe	1	92
6	syn-10SeMe	1	91
7	anti-10SMe	1	90
8	anti-10SeMe	1	90

^a Catalyst loading with respect to the amount of 1a.

^b After 12 h reaction time. Determined via ¹H NMR spectroscopy and relative to TES as internal standard. Averaged values of three measurements. A measurement error of ±10% is assumed.

In contrast, a significant decrease in the conversion of 1a was observed when catalysts 8^SMe and 8^SeMe , with the less polarizing imidazolium moieties, were used (Table [2], entries 1 and 2). After 12 h reaction time, only 21% of compound 3a were obtained with catalyst 8^SeMe and only 15% with catalyst 8^SMe . Even after 24 h reaction time, both catalysts yielded only approximately 50% of compound 3a. This observation is in accordance with previous results concerning the weak performance of imidazolium-based ChB donors in carbon–chlorine bond activation.[27]

Finally, relative rate constants were determined for selected catalysts (Table [3] and Figure [3]; shown for the first 5 h of reaction time, full kinetic profile is given in the Supporting Information). To this end, the least active reference compound anti-10^Cl served as standard for comparison (k _rel = 1). For the most potent catalysts 9^SMe and 9^SeMe , relative rate accelerations of about 2000 and 2300 were determined (Table [3], entries 6 and 7). In both cases, the related syn-atropisomers were markedly less active, with rate accelerations being 1.7 (9^SeMe vs. syn-10^SeMe ) or 1.8 times (9^SMe vs. syn-10^SMe ) higher for the more flexible catalysts (entries 4 and 5). The analogous hydrogen-bond donors syn/anti-10^H and 9^H showed 5–15 fold less rate acceleration than the best ChB donors (entries 2 and 3).

Table 3 Relative Rate Constants for Selected ChB Donors, with Reference Compound anti-10^Cl as Standard.

Entry	Catalyst	krel a
1	anti-10Cl	1
2	syn/anti-10H b	150
3	9H	450
4	syn-10SeMe	1200
5	syn-10SMe	1300
6	9SeMe	2000
7	9SMe	2300

^a Normalized rate constant (relative to anti-10^Cl as standard).

^b Inseparable 40:60 mixture of syn/anti-10^H .

The strong performance of the sulfur-based catalysts may in fact not be entirely unexpected, as there are several examples in the literature in which intramolecular sulfur-based chalcogen bonds play a major role in the stabilization of transition states of organocatalyzed reactions.[37] [38] [39] [40] [41] [42] [43]

Figure 3 Kinetic profile of the reduction of 2-phenylquinoline (1a) in the presence of selected ChB, HB and XB donors. Kinetic profile for 12 h reaction time is given in the Supporting Information.

In comparison to neutral ChB donors, which were reported earlier as catalysts for this reaction, the selenium-based cationic catalysts were approximately twice as active (based on the time needed for full conversion),[25] whereas the sulfur-based compounds were about 24 times more active, even though 1 mol% of catalyst was used compared with the 30 mol% of the earlier reports.[24] This impressively demonstrates the power of charge-assisted chalcogen bonding, particularly for the comparably weakly polarizable sulfur center.

Finally, the most active bisbenzimidazolium based catalysts 9^SMe and 9^SeMe were applied in a small substrate screening involving selected quinolines 1b–d, which had proven to be very challenging for neutral chalcogen bond donors.[24] [25] First, the background reactivity of the quinolones in the absence of any activating reagent was determined (Table [4], entries 1–3) and was found to be very low. In contrast to neutral ChB donors, which provided no or very low yields of products 3b–d, the cationic ChB catalysts induced up to 82% product formation (Table [4], entries 4–9).

Table 4 Yields for the Reduction of Quinolines 1b–d in the Presence of Chalcogen Bond Donors 9

Entry	Catalysta	Product	Yield of 3b–d (%)b
1	–	3b	0
2	–	3c	6 (13)
3	–	3d	0
4	9SMe	3b	57 (62)
5	9SMe	3c	88 (>95)
6	9SMe	3d	56 (73)
7	9SeMe	3b	50 (63)
8	9SeMe	3c	92 (>95)
9	9SeMe	3d	38 (61)

^a 1 mol% catalyst was applied with respect to the amount of 1b–d.

^b After 12 h reaction time. Yield in brackets after 24 h. Determined via ¹H NMR spectroscopy. Averaged values of two measurements. A measurement error of ±10% is assumed.

For both catalysts, the very electron-poor quinoline 3c was reduced most efficiently (Table [4], entries 5 and 8). Selenium-based catalyst 9^SeMe seems less suitable for the reduction of electron-rich quinolines (such as 3d) than its sulfur-based analogue 9^SMe (cf. entries 6 and 9). This could be explained by either a stronger catalyst-substrate coordination (which may result in a less favorable product liberation from the catalyst) or by a competing coordination of the catalyst to the electron-rich methoxy substituent of the substrate.[25]

In conclusion, the first application of cationic sulfur- and selenium-based chalcogen bond donors in the activation of neutral compounds was presented. Based on a series of comparison experiments, alternative modes of activation could be ruled out. Overall, rate accelerations of up to 2300 were observed compared with inactive reference compounds. The strong performance of the cationic ChB donors allowed the activation of very electron-poor or electron-rich substrates, which were hitherto difficult to activate.

Two findings in particular were somewhat unexpected: the better performance of catalysts with flexible core structures in comparison to their conformationally locked syn-atropisomers, and the potency of sulfur-based ChB donors. Given that the latter are easier to synthesize and handle than their selenium-based analogues, it seems that sulfur-containing Lewis acids could have great potential in chalcogen-bonding-based catalysis. Additional studies to further implement such compounds in noncovalent organocatalysis are under way.

• References and Notes

• 1 Doyle AG, Jacobsen EN. Chem. Rev. 2007; 107: 5713
• 2 Alemán J, Parra A, Jiang H, Jørgensen KA. Chem. Eur. J. 2011; 17: 6890
• 3 Zhao Y, Beuchat C, Domoto Y, Gajewy J, Wilson A, Mareda J, Sakai N, Matile S. J. Am. Chem. Soc. 2014; 136: 2101
• 4 Schottel BL, Chifotides HT, Dunbar KR. Chem. Soc. Rev. 2008; 37: 68
• 5 Bulfield D, Huber SM. Chem. Eur. J. 2016; 41: 14434
• 6 Cavallo G, Metrangolo P, Milani R, Pilati T, Priimagi A, Resnati G, Terraneo G. Chem. Rev. 2016; 116: 2478
• 7 Nagorny P, Sun Z. Beilstein J. Org. Chem. 2016; 12: 2834
• 8 Tepper R, Schubert SU. Angew. Chem. Int. Ed. 2018; 57: 6004
• 9 Vogel L, Wonner P, Huber SM. Angew. Chem. Int. Ed. 2019; 58: 1880
• 10 Lim JY. C, Beer PD. Chemistry 2018; 4: 731
• 11 Gleiter R, Haberhauer G, Werz DB, Rominger F, Bleiholder C. Chem. Rev. 2018; 118: 2010
• 12 Mahmudov KT, Kopylovich MN, Guedes da Silva MF. C, Pombeiro AJ. L. Dalton Trans. 2017; 10121
• 13a Murray JS, Lane P, Clark T, Politzer P. J. Mol. Model. 2007; 13: 291
• 13b Rosenfield RE, Parthasarathy R, Dunitz JD. J. Am. Chem. Soc. 1977; 99: 4860
• 13c Murray JS, Lane P, Clark T, Politzer P. J. Mol. Model. 2009; 15: 723
• 13d Pearson RG. J. Am. Chem. Soc. 1963; 85: 3533
• 13e Ramasubbu N, Parthasarathy R, Murray-Rust P. J. Am. Chem. Soc. 1986; 108: 4308
• 13f Murray-Rust P, Motherwell WD. S. J. Am. Chem. Soc. 1979; 101: 4374
• 13g Politzer P, Murray JS, Clark T. Phys. Chem. Chem. Phys. 2010; 12: 7748
• 13h Politzer P, Riley KE, Bular FA, Murray JS. Comput. Theor. Chem. 2012; 998: 2
• 14 Wang W, Ji B, Zhang Y. J. Phys. Chem. A 2009; 113: 8132
• 15 Bleiholder C, Gleiter R, Werz DB, Köppel H. Inorg. Chem. 2007; 46: 2249
• 16a Mukherjee AJ, Zade SS, Singh HB, Sunoj RB. Chem. Rev. 2010; 110: 4357
• 16b Werz DB, Gleiter R, Rominger F. J. Am. Chem. Soc. 2002; 124: 10638
• 16c Werz DB, Gleiter R, Romiger F. J. Org. Chem. 2002; 67: 4290
• 16d Schulte JH, Werz DB, Romiger F, Gleiter R. Org. Biomol. Chem. 2003; 1: 2788
• 16e Werz DB, Gleiter R, Romiger F. Organometallics 2003; 22: 843
• 16f Gleiter R, Werz DB, Rausch BJ. Chem. Eur. J. 2003; 9: 2676
• 16g Bleiholder C, Werz DB, Köppel H, Gleiter R. J. Am. Chem. Soc. 2006; 128: 2666
• 17 Iwaoka M, Tomoda S. J. Am. Chem. Soc. 1996; 118: 8077
• 18 Wirth T. Angew. Chem. Int. Ed. Engl. 1995; 34: 1726
• 19 Garrett GE, Carrera EI, Seferos DS, Taylor MS. Chem. Commun. 2016; 52: 9881
• 20 Garrett GE, Gibson GL, Straus RN, Seferos DS, Taylor MS. J. Am. Chem. Soc. 2015; 137: 4126
• 21 Lim JY. C, Marques I, Félix V, Beer PD. Chem. Commun. 2018; 54: 10851
• 22 Lim JY. C, Marques I, Thompson AL, Christensen KE, Félix V, Beer PD. J. Am. Chem. Soc. 2017; 139: 3122
• 23 Lim JY. C, Liew JY, Beer PD. Chem. Eur. J. 2018; 24: 14560
• 24 Benz S, López-Andarias J, Mareda J, Sakai N, Matile S. Angew. Chem. Int. Ed. 2017; 56: 812
• 25 Benz S, Mareda J, Besnard C, Sakai N, Matile S. Chem. Sci. 2017; 8: 8164
• 26a Wonner P, Vogel L, Düser M, Gomes L, Kniep F, Mallick B, Werz DB, Huber SM. Angew. Chem. Int. Ed. 2017; 56: 12009
• 26b General procedure for the synthesis of chalcogeno ureas 11–13: Under an argon atmosphere, the respective bis(benz)imidazolium compound (1 equiv) was added to a Schlenk flask and dissolved in anhydrous methanol (0.17 m; dried for 24 h over molecular sieve). To the solution, elemental selenium powder or sulfur (2.5 equiv) and cesium carbonate (2.5 equiv) were added. The mixture was heated at reflux for 24 h and finally filtered (hot solution) over a short plug of silica and rinsed with CH₂Cl₂. After the solvent was removed, the crude solid was purified by column chromatography (solvents noted for specific compounds). All spectroscopic data are given in the Supporting Information.
• 26c General methylation procedure for the synthesis of compounds 8–10: Under an argon atmosphere, the respective chalcogenourea compound (1 equiv) was dissolved in anhydrous CH₂Cl₂. Then methyl trifluoromethanesulfonate (2.5 equiv) was slowly added at 0 °C and the reaction mixture was stirred for 24 h at room temperature. After the solvent was removed under reduced pressure, the crude residual was diluted in a little acetonitrile and was precipitated by addition of diethyl ether. The solid was filtered off, washed with diethyl ether and pentane and dried under high vacuum. Alternatively, the solvent was removed, and the residue was washed under sonification several times with ether and pentane. Finally, the respective methylated chalcogenated compound was obtained as colorless solid. All spectroscopic data are given in the Supporting Information.
• 26d Spectroscopic data for selected compounds 9^SeMe, 9^SMe and syn-10^SMe: Compound 9^SeMe : Yield: 294 mg (74%); solid; ¹H NMR (300 MHz, CDCl₃): δ = 8.50 (s, 1 H), 8.17 (dd, J = 7.10, 1.80 Hz, 2 H), 8.08 (dd, J = 9.00, 6.80 Hz, 1 H), 7.69 (m, 6 H), 7.43 (dd, J = 7.50, 0.80 Hz, 2 H), 4.67 (d, J = 8.00 Hz, 4 H), 2.41 (s, 6 H), 2.01 (q, J = 7.70 Hz, 4 H), 1.31 (m, 20 H), 0.87 (m, 6 H). ¹³C NMR (101 MHz, CD₃CN): δ = 147.29, 136.10, 135.34, 133.76, 133.48, 131.77, 128.22, 128.59, 128.39, 122.09 (d, J = 321.0 Hz), 114.49, 114.07, 49.88, 32.48, 30.24, 28.79 (d, J = 4.50 Hz), 27.34, 23.34, 14.38, 11.51. ¹⁹F NMR (235 MHz, CD₃CN): δ = −78.53 (s, 6 F). ATR-IR: 3073  (w), 2928  (w), 2857  (w), 1738  (w), 1603  (w), 1501  (m), 1477  (m), 1429  (m), 1358  (w), 1250 (vs), 1223  (s), 1148  (s), 1086  (w), 1028 (vs), 926  (w), 831  (w), 800  (w), 752  (s), 698  (w), 635 (vs), 571  (m), 515  (s), 432  (w) cm^–1. ESI-MS (+): m/z [M-OTf]⁺ calcd. 873.21; found: 872.87; ESI-MS (–): m/z [OTf]^– calcd. 148.95; found 148.80. Anal. calcd. for CHNS: C, 51.82; H, 5.65; N, 6.04; S, 13.83; found: C, 51.79; H, 5.42; N, 5.77; S, 13.72. Compound 9^SMe : Yield: 620 mg (79%); solid; ¹H NMR (300 MHz, CDCl₃): δ = 8.50 (s, 1 H), 8.15 (m, 3 H), 7.68 (sd, J = 8.4, 1.3 Hz, 6 H), 7.43 (d, J = 7.7 Hz, 2 H), 4.64 (d, J = 7.7 Hz, 4 H), 2.51 (s, 6 H), 2.02 (p, J = 7.7 Hz, 4 H), 1.38 (m, 20 H), 0.87 (m, 6 H). ¹³C NMR (101 MHz, CD₃CN): δ = 151.20, 135.35, 134.66, 134.08, 132.97, 131.67, 129.18, 128.83, 127.88, 114.53, 113.93, 48.76, 32.48, 30.11, 29.79 (d, J = 5.40 Hz), 27.35, 23.40, 18.51, 14.37. ¹⁹F NMR (235 MHz, CD₃CN): δ = −79.28 (s, 6 F). ATR-IR: 3073  (w), 2928  (w), 2857  (w), 1738  (w), 1603  (w), 1501  (w), 1477  (w), 1462  (w), 1431  (w), 1364  (w), 1254  (s), 1223  (m), 1150  (m), 1088  (w), 1028 (vs), 847  (w), 802  (w), 752  (m), 698  (w), 634 (vs), 571  (w), 515  (m) cm^–1. ESI-MS (+): m/z [M + OTf ]⁺ calcd. 777.32; found: 777.51; ESI-MS (–): m/z [OTf]^– calcd. 148.95; found: 149.00. Anal. calcd. CHNS: C, 51.82; H, 5.65; N, 6.04; S, 13.83; found: C, 51.79; H, 5.42; N, 5.77; S, 13.72. Compound syn-10^SMe : Yield: 323 mg (87%); solid; ¹H NMR (300 MHz, CD₃CN): δ = 8.40 (t, J = 8.1 Hz, 1 H), 8.24 (d, J = 8.1 Hz, 2 H), 8.04 (dd, J = 7.3, 2.0 Hz, 2 H), 7.81 (dtd, J = 13.0, 7.3, 1.4 Hz, 4 H), 7.67 (d, J = 7.7 Hz, 2 H), 4.71 (td, J = 7.4, 1.9 Hz, 4 H), 2.51 (s, 6 H), 2.03 (m, 4 H), 1.28 (m, 20 H), 0.87 (m, 6 H).¹³C NMR (101 MHz, CD₃CN): δ = 151.83, 138.65, 136.31, 135.17, 133.42 (d, J = 1.60 Hz), 132.81, 129.84, 129.26, 127.14 (q, J = 31.0 Hz), 122.12 (d, J = 321.5 Hz), 114.93, 114.37, 49.30, 32.36, 29.88, 29.68 (d, J = 8.80 Hz), 27.09, 23.30, 18.89, 14.34. ¹⁹F NMR (235 MHz, CD₃CN): δ = −56.73 (s, 3 F), −79.32 (s, 6 F). ATR-IR: 2927  (w), 2857  (w), 1547  (w), 1501  (w), 1477  (w), 1433  (w), 1364  (w), 1256 (vs), 1223  (m), 1175  (m), 1140  (m), 1028  (s), 984  (w), 858  (w), 752  (m), 687  (w), 635  (s), 573  (m), 517  (w), 434  (w) cm^–1. ESI-MS (+): m/z [M+OTf ]⁺ calcd. 845.31; found: 845.27; ESI-MS (–): m/z [OTf]^– calcd. 148.95; found: 148.66. Anal. calcd. CHNS: C, 49.49; H, 5.17; N, 5.63; S, 12.89; found: C, 48.55; H, 4.94; N, 5.53; S, 12.37.
• 27 Wonner P, Vogel L, Kniep F, Huber SM. Chem. Eur. J. 2017; 23: 16972
• 28 Benz S, Besnard C, Matile S. Helv. Chim. Acta 2018; 101: e1800075
• 29 Sebastian B, Poblador-Bahamonde AI, Nicolas LD, Stefan M. Angew. Chem. Int. Ed. 2018; 57: 5408
• 30 Bolm C, Bruckmann A, Pena M. Synlett 2008; 900
• 31 Jungbauer SH, Schindler S, Kniep F, Walter SM, Rout L, Huber SM. Synlett 2013; 24: 2624
• 32 Togo H, Iida S. Synlett 2006; 2159
• 33 General Procedure for the Reduction of 1a: In a dry NMR tube, freshly prepared stock solutions of the respective catalyst (100 μL, 2.5 mm), the quinoline derivative (100 μL, 250 mm) and the Hantzsch ester (400 μL, 150 mm) were added, sealed and shaken for 1 min. Afterwards, a series of ¹H NMR measurements were performed with a total duration of 12 h (for selected derivatives 24 h). As internal standard TES (0.125 equiv) was added to the quinoline stock solution.
• 34 The syn-12BIm-Cl atropisomer decomposed during its synthesis.
• 35 Jungbauer SH, Huber SM. J. Am. Chem. Soc. 2015; 137: 12210
• 36 Gliese JP, Jungbauer SH, Huber SM. Chem. Commun. 2017; 53: 12052
• 37 Srivastava K, Chakraborty T, Singh HB, Butcher RJ. Dalton Trans. 2011; 4489
• 38 Srivastava K, Shah P, Singh HB, Butcher RJ. Organometallics 2011; 30: 534
• 39 Nagao T, Hirata S, Goto S, Sano S, Kakehi A, Iizuka K, Shiro M. J. Am. Chem. Soc. 1998; 12: 3104
• 40 Franchetti P, Cappellacci L, Grifantini M, Barzi A, Nocentini G, Yang H, O’Connor A, Jayaram HN, Carrell C, Goldstein BM. J. Med. Chem. 1995; 38: 3829
• 41 Birman VB, Li X. Org. Lett. 2008; 10: 1115
• 42 Leverett CA, Vikram VC, Purohit C, Romo D. Angew. Chem. Int. Ed. 2010; 49: 9479
• 43 Robinson ER. T, Fallan C, Simal C, Slawin AM. Z, Smith AD. J. Am. Chem. Soc. 2013; 4: 2193