Dianhydrohexitols: new tools for organocatalysis. Application in enantioselective Friedel-Crafts alkylation of indoles with nitroalkenes

A series of hydrogen bonding organocatalysts was synthesized from dianhydrohexitol and was used for the first time in organocatalysis for Friedel-Crafts alkylation of indoles with nitroalkenes. Moderate yields and enantioselectivities were achieved


Results and Discussion
A series of novel organocatalysts 1-5 derived from isomannide and organocatalyst 6 from isoidide have been synthesized (Figure 1).A similar synthetic route could be used, with the formation of the diamines 9 and 11 as key intermediates which were then transformed into their corresponding mono or dithioureas derivatives (Scheme 1 and 2 respectively).For organocatalyst 5, a sulfonamide function was introduced on the amino group of the monothiourea 1.The synthesis of diamine 9 44 was initiated by the ditosylation of the hydroxyl groups of isomannide in the presence of a catalytic amount of DMAP leading to 7 in 91% yield.
Subsequently, the substitution of the two tosylated groups with sodium azide in DMF at 120 °C afforded diazide 8 in 77% yield which was then hydrogenated in almost quantitative yield to form the diamine 9 (Scheme 1).For the synthesis of diamine 11, as the ditosylation of isoidide and the displacement of the two mesylated groups with sodium azide was unsuccessful, 19 the triflate was chosen as a leaving group.In the presence of Tf2O, isoidide led to the corresponding triflate which was then used without further purification.The introduction of the azide groups was then performed to afford the diazide 10 in 26% yield over 2 steps (Scheme 2).
With a series of organocatalysts in hand, we were interested to explore their potential as organocatalysts for enantioselective Friedel-Crafts alkylation of indoles with nitroalkenes Since the nitro group is a strong electron-withdrawing group, the nitroalkenes are very attractive for hydrogen-bond organocatalysis.
Our investigations started with the examination of the reaction between indole 12a and trans--nitrostyrene 13a in the presence of 20 mol% of the organocatalyst 2 in toluene.It was a rather slow reaction which resulted in 10% ee with moderate yield (Table 1, entry 1).Therefore, different solvents were screened in order to improve the yield and the enantioselectivity.The results are summarized in Table 1.As we have observed that the catalyst was not very soluble into toluene, we first performed the reaction in DMSO, DMF, MeCN, and EtOAc (Table 1, entries 2-5).It is worth mentioning that the reaction almost could not proceed in these solvents in which the organocatalyst 2 is perfectly soluble.It is probably due to the formation of a strong hydrogen bond interaction between the organocatalyst 2 and the solvent instead of interactions with the substrate.In chlorinated solvents, the reaction proceeded faster in CHCl3 than in CH2Cl2 resulted in 75% and 20% yield respectively but, it was almost racemic (entries 6 and 7).
Although MeOH can form hydrogen bonding with both the organocatalyst and the substrate, it led to poor yielding Friedel-Crafts alkylation but without enantioselectivity (entry 8).With ethers as solvents (entries 9-12), Et2O and tert-butyl methyl ether (MTBE) gave the best enantioselectivities (Table 1, entries 11 and 12).The better yield observed with MTBE is probably due to a better solubility of the organocatalyst 2 in MTBE than in Et2O.From all the screened solvents, the best combination in term of yield and enantioselectivity was found with MTBE.Thus, this solvent was chosen in order to perform the other organocatalytic experiments.Recently, Herrera and co-workers reported that external Brønsted acids could enhance the efficiency of thiourea catalyst. 40So we envisioned that higher enantioselectivity and reactivity might be achieved by using a suitable additive acids thus, we started to investigate the use of acid additives in this reaction.
Compared to the reaction performed without acid, the yield of the reaction is in general lower, probably due to a competition of the formation of a hydrogen bond between the catalyst and the substrate with the acid.It was observed that there is no correlation between yield, enantioselectivity and pKa value.Nevertheless, the pKa value and the nature of the acid additive, such as non-functionalized carboxylic acid or functionalized ones containing groups which are not able to act as hydrogen bond donor find some correlation (Table 2, entries 2, 3, 7, 8).Moreover, an increase of the enantioselectivity was observed when the pKa decreased, ee reaching 37% and 39% with TCA and TFA (Table 2, entries 7 and 8).With D-and L-mandelic acid, no match or mismatch effect was observed and the chirality of the acid has no influence on the enantioselectivity; the same major isomer being formed (entries 4 and 5).Similar results have been found by Herrera et al. 40 This fact showed that chirality was preferentially controlled by the thiourea catalyst.The additive acids only activated the thiourea moiety rather than participating itself into the transition state.However, the use of sulfonic acid such as D-camphorsulfonic acid and p-TsOH, resulted in similar yields, but p-TsOH, due to strong acid character led to almost racemic product contrary to the D-camphorsulfonic acid (entries 6 and 9).Taking into account mainly the ee, we finally chose TCA as the best additive.a Reaction conditions: 13a (0.1 mmol), 12a (2.0 equiv), solvent (0.4 mL), 2 (20 mol%).b Yield of product isolated after flash chromatography.c Determined by HPLC analysis using a Phenomenex  Lux 5 cellulose-2 column.
In order to determine the best organocatalyst, the reaction was performed at room temperature, in MTBE using 10 mol% of TCA and 20 mol% of different catalysts (Table 3).Compared to organocatalyst 2, monothiourea 1 led to a similar yield, but without enantioselectivity (entries 1 and 2).This may be attributed to the lack of two hydrogen bonding functions.With phenyl thioureas 2, 3 and 4 (entries 2-4), the best organocatalyst is always the compound 2. The difference in behavior for thiourea 2 is probably due to the presence of the two CF3 groups in 2 and 4 positions of the aromatic ring leading to an augmentation of the acidity of the hydrogen located on each thiourea function.Thus, the ability to form hydrogen bonding is enhanced and favored with organocatalyst 2.  Surprisingly, with the dithiourea 6, no reaction occurred.It was supposed that the proximity of the two thioureas groups in endo position prevents the intermolecular hydrogen bonding and favors the intramolecular ones (site-site interactions) (entry 6).
Besides, the effect of the loadings of the additive TCA, the concentration of the substrate and the TCA in the reaction mixture and the temperature were also examined (Table 4).A decrease in the concentration of TCA led to an increase of the yield, but the enantioselectivities were similar (entries 2-5).Nevertheless, similar results in terms of enantioselectivity and yield were found with 10 and 5 mol% TCA (Table 4, entries 4 and 5).A comparison with the experiment performed without TCA at room temperature shows that only 10 mol% TCA permits a slight increase in enantioselectivity and yield (Table 4 entries 1 and 4).Increasing the concentration of the substrate from 0.25N to 0.5N only led to an increase of the yield from 29% to 49% with a slight decrease of ee (entries 3 and 6).In opposite, a decrease of the concentration led to a significant decrease of the yield but the enantioselectivity was slightly improved (entries 7 and 8 vs entries 3 and 4).Using the conditions mentioned in entry 8, a longer reaction time permitted to increase the yield (entry 9).A similar enantioselectivity was observed if the reaction was performed at -10°C, but the yield decreased drastically from 43% to almost 15% (Table 4, entry 10).a Yield of product isolated after flash chromatography.b Determined by HPLC analysis using a Phenomenex  Lux 5 cellulose-2 column.
Finally, with the optimized reaction conditions (Table 4, entry 9) we studied the influence of the electronic effects of the substituent located in the 5-position of the indole and on the βnitrostyrene (Table 5).Compared to the reference reaction (entry 1), an electron-withdrawing atom such as bromine settled in the indole led to a drastic decrease in the reaction rate (5% yield entry 3) contrary to an electron-donating group such as methoxy group which has a good influence on the reaction rate but not on the enantioselectivity (entry 2).
For the reaction with different nitroalkenes bearing either electron-withdrawing groups or electron-donating groups located in the position 4 of the aromatic ring, the same range of yield and enantioselectivity were observed (Table 5, entries 4-7).On the contrary, the presence of a chlorine atom in position 2 led to lower yield and enantioselectivity (Table 5, entry 8). a Reaction conditions: 13a-f (0.1 mmol), 12a-c (2.0 equiv), MTBE (1.0 mL), 2 (20 mol%), and TCA (10 mol%).b Yield of product isolated after flash chromatography.c Determined by HPLC analysis using a Phenomenex  Lux 5 cellulose-2 column.d Absolute configuration was determined by comparison of the optical rotation with the known compounds in the literature.10a

Conclusions
In conclusion, we have developed a new series of hydrogen bonding organocatalysts derived from isomannide and isoidide, used for the first time in organocatalysis.Moderate enantioselectivities and yields were achieved in Friedel-Crafts alkylations of indoles to nitroalkenes.Further investigations on its application to other enantioselective transformations are in progress.

Experimental Section
General.All experiments were carried out with anhydrous solvents in dried glassware.
Commercially available materials (Aldrich or Fluka) were used without further purification.THF and CH2Cl2 were dried using a drying station.Flash chromatography was performed on MERCK silica gel (40-63 µm).Analytical TLCs were carried out on MERCK pre-coated silica gel 60 F254.Melting points were determined on a Reichert Thermoval apparatus and are uncorrected.C NMR (50 MHz, CDCl3) δc:145.3,132.9, 129.9 and 127.9 (4C, phenyl), 79.9, 77.Under argon, at 0 o C, to a solution of isoidide (465 mg, 3.18 mmol) and pyridine (1.6 mL, 16.0 mmol) in dry CH2Cl2 (10 mL) was added trifluoromethansulfonic anhydride (1.34 mL, 7.95 mmol) dropwise.After completion of the addition, the reaction was stirred for another 15 minutes.The residue was then diluted with CH2Cl2, washed with an aqueous HCl solution (1N), water, saturated aqueous NH4Cl solution and brine.The combined organic layers were dried over MgSO4, filtered and concentrated to give crude ditriflate compound.This latter was directly dissolved in DMF (11  mL) and NaN3 (434.3 mg, 6.68 mmol) was added.The reaction mixture was stirred for about 0.5 h at room temperature.Water (15 mL) was added and the mixture was extracted with diethyl ether (5  10 mL).The combined organic layers were dried over anhydrous MgSO4 and concentrated under vacuum.The crude diazide was purified by column chromatography (cyclohexane/EtOAc, 4:1) to afford the product

Table 1 .
Catalytic enantioselective Friedel-Crafts alkylation of indole 12a with trans-nitrostyrene 13a in different solvents a