Carbohydrate-Based Chiral Iodoarene Catalysts: A Survey through the Development of an Improved Catalyst Design

Iodoarene catalysts can be applied in versatile reactions, for instance in the construction of complex chiral molecules via dearomatization of simple aromatic compounds. Recently, we reported the synthesis of the first carbohydrate-based chiral iodoarene catalysts and their application in asymmetric catalysis. Here we describe the synthesis of some new and improved catalysts. An account on how we got to the improved catalyst design, as well as the X-ray structure of one of the carbohydrate-based iodoarenes, is given.


Introduction
Carbohydrates are prevalent in nature. For instance, they occur as skeleton substances in plants or are used to store energy in almost all living creatures. Carbohydrates can be isolated from many different sources; therefore, common carbohydrates like d-glucose, d-galactose, or d-fructose are commercially available at low cost [1]. Bearing multiple stereo centers, carbohydrates contain a large amount of stereochemical information. Thus, they have already been widely used as starting materials for chiral auxiliaries as well as for the synthesis of chiral ligands for enantioselective, transition metal catalyzed reactions [2][3][4][5][6][7].
Transition metal catalyzed reactions provide a plethora of selective chemical modifications with a high tolerance against many functional groups [8]. Numerous catalytic reactions proceed in an enantioselective manner [9]. For a long time, metal-catalyzed reactions were nearly the only way of performing enantioselective catalytic reactions in an effective way. Since the beginning of the 2000s, the field of organocatalysis expanded rapidly, and many metal-free protocols for enantioselective catalysis can be found in the literature. Advantages of organocatalysis over transition metal catalysis are, for example, their lower toxicity and lower cost of the catalysts. Furthermore, organocatalysts are often more tolerant against moisture and oxygen [10].
Chiral iodoarenes have gained significant importance as metal-free catalysts for stereoselective oxidation reactions. Oxidative dearomatization, for instance, provides the opportunity to construct complex molecules, like natural products, by starting from simple aromatic molecules [11][12][13]. Kita and coworkers developed a method for the enantioselective dearomatization of naphthole derivatives like compound 1 (Scheme 1) to the corresponding spiro lactone 2 using the chiral iodoarene catalyst 3 ( Figure 1) and stoichiometric amounts of oxidant [14]. In the meantime, the iodoarene catalysts

Results and Discussion
Our first attempt to improve the enantioselectivity of our carbohydrate-based iodoarene catalysts was to change the linkage between the monosaccharide and the iodine-substituted aromatic ring. Instead of linking the carbohydrate moiety, via its position 6, and a phenyl ether, like in 8, we now accomplished the linkage via a corresponding benzyl ether. An advantage of a benzylic connection of the sugar part over a phenolic connection is that the bond formation is not solely limited to position 6 at the monosaccharide, and other positions could be used as well. For this purpose, we chose 1,3-bis(bromomethyl)-2-iodo-benzene (10) as the iodoarene building block. The latter was prepared from 1,3-dimethyl-2-iodobenze according to a literature protocol [21]. For the introduction of the carbohydrate moiety, 1,2:5,6-di-O-isopropylidene-α-d-glucofuranose 9 and 1,2:4,5-di-O isopropylidene-β-d-fructopyranose 13 were prepared in one step from d-glucose and d-fructose, respectively, as was previously described [22,23]. Deprotonation of the hydroxyl group at position 3 in compounds 9 and 13 followed by subsequent addition of iodoarene 10 provided catalysts 11 and 14 in high yields (Scheme 2). We also changed the substitution pattern of 11 and 14 as follows. First, the isopropylidene groups at positions 5 and 6 in 11 and positions 4 and 5 in 14 were removed under acidic conditions, respectively. Next, the hydroxyl groups were benzylated using sodium hydride and benzyl bromide. Glucose-based catalyst 12 was, thus, isolated in 62% yield and the fructose-based catalyst 15 in 80% yield.
With four new carbohydrate-derived iodoarenes at hand, we next tested their potential as catalysts in enantioselective spirolactonization, for which we chose the identical conditions reported previously for catalyst 8 [20]. At −20 • C none of the new benzyl ether linked catalysts showed any conversion of 1 to 2 (Table 1, entries 2, 4, 6, and 8). At room temperature, catalysts 11 and 14 gave low yields of products (12% and 13%, respectively), though with a slight enrichment of the R-enantiomer in the product (Table 1, entries 3 and 5). Catalysts 12 and 15 were not at all capable of catalyzing the spirolactonization of 1. Instead, a six-membered lactone between the naphtholic hydroxy and the acid group was formed ( Next, we tested whether decreasing the catalyst's symmetry could lead to higher enantioselectivities because examples of C 1 symmetric catalysts like 5 and 6 ( Figure 1) are known to work well in asymmetric catalysis [16,19]. C 1 symmetric catalyst 17 was prepared in high yield by deprotonation of 9 and subsequent addition of 1-(bromomethyl)-2-iodo-3-methylbenzene 16 (Scheme 3). The latter was obtained as a side product in the synthesis of 10. For the preparation of catalyst 20, a Mitsunobu protocol was used [24]. Condensation between commercially available 2-iodophenol 19 and known methyl d-glucoside 18 [25] afforded the respective carbohydrate-substituted iodoarene 20 in excellent yield. We selected 18 as the carbohydrate substituent for the catalyst because this substituent worked best in our previously reported catalyst 8.
Again, when 17 was applied as a catalyst in Kita's spirolactonization, no conversion was observed at −20 • C (Table 1, entry 10). At room temperature, a small amount of 2 was isolated ( Table 1, entry 11). Compared to the corresponding C 2 symmetrical catalyst 11 (Table 1, entries 3 and 11), 20 also gave a slightly higher yield than 8d, but the enantioselectivity was much lower in this case (Table 1, entries  1 and 12).
From those experiments, it can be generally concluded that C 2 symmetric catalysts, having the sugar moiety linked to the iodoarene through a phenyl ether bond, gave better yields and higher enantioselectivities than C 1 symmetric catalysts where the sugar was bound via a benzylic linkage. Therefore, we decided to focus next on catalysts having the same or similar structures as catalyst 8 but now varying the sugar and its anomeric configuration instead. For the above-mentioned purpose, we prepared derivatives of methyl-β-d-glucopyranoside, methyl-α-d-mannopyranoside, and methyl-α-d-galactopyranoside with a similar substitution pattern like 18. First, commercially available methyl glycosides 21, 22, and 23 were reacted with trityl chloride in pyridine to selectively protect the primary alcohol at position 6 (Scheme 4). Next, the remaining hydroxyl groups were benzylated using sodium hydride and benzyl bromide. Selective removal of the trityl group under acidic conditions gave catalyst precursors 30, 31, and 32.
The substituted iodoarenes 34, 35, and 36 were then applied as catalysts in Kita's spirolactonization of 1-hydroxy-2-naphthalenepropionic acid (1). β-Configurated 34 gave 73% yield and an enantiomeric ratio of 70:30 (Table 1, entry 13). Compared to catalyst 8d, which had an α-configuration at the anomeric center, the yield was in a similar range, and the enantioselectivity was only slightly lower. Mannose-derived catalyst 35 had an axial substituent at position 2, whereas the respective substituent in 8d was equatorial. Catalyst 35 gave a lower yield and slightly lower enantioselectivity when applied in Kita's spirolactonization (Table 1, entry 14) than 8d ( Table 1, entry 1). Galactose-derived catalyst 36 had an inverted configuration at position 4 compared to 8d. The yield when using 36 as a catalyst was in a similar range as that of 35. However, the enantiomeric ratio decreased significantly. The almost complete loss of stereoinduction by changing the configuration at position 4 led us to the assumption that this stereocenter is of major importance for the catalyst design.  In order to verify our assumption of the importance of position 4 of the carbohydrate moiety further, we first took a closer look at the molecular structure of the C 2 symmetrical phenyl ether linked catalysts. For that purpose, we tried to crystallize catalyst 8 as well as 34, 35, and 36. However, we could not get useful crystallographic data on catalysts 8a, 8b, 8d, 34, 35, and 36. Fortunately, we were able to obtain crystals of compound 8c that were suitable for X-ray crystallography. Compound 8c bears methyl groups instead of benzyl groups at positions 2, 3, and 4 like 8d does. Compound 8c gave an er of 69:31 and 61% yield by its application in Kitas spirolactonization, which was slightly lower than 8d (78:22; 67% at the same reaction conditions) [20]. Despite the fact that different protective groups might cause changes in the conformation of the catalyst, the X-ray structure of 8c can be used as a model system to get a better idea of the molecular structure of our catalysts. The molecular structure of 8c is shown in Figure 2. The molecule crystallized in the orthorhombic space group P2 1 2 1 2 1 [27]. The molecular structure showed that the substituents at C-4 and C-4 were both located close to the iodine where the catalytic reaction took place. The substituents at the anomeric centers did not come as close as the substituents at position 4, and the groups at positions 2 and 3 even pointed in different directions. Thus, we assumed that a very bulky substituent at position 4 was be beneficial for the enantioselectivity and that the other substituents were not that important. In order to investigate the influence of the substituent at position 4 in greater detail, catalysts with very bulky substituents at this position and "small" substituents (methyl groups) at positions 1, 2, and 3 were prepared. For the substituents at position 4 we chose 3,5-dimethyl-benzyl, 3,5-di-tert-butyl-4-methoxy-benzyl and 1-naphthylmethyl since they were sufficiently bulky, were stable under the reaction conditions, and also could be easily introduced by using the corresponding benzylic bromides. The latter bromides were either commercially available or were prepared as described previously [3].
We started the preparation of the known 1,2,3-trimethyl-6-trity-glucose (37) [28]. Reaction of the latter with different benzyl bromides under basic conditions afforded glycosides 38, 39, and 40 in good to virtually quantitative yields (Scheme 6). Removal of the trityl protective group at position 6 with trifluoroacetic acid afforded primary alcohols 41, 42, and 43 in good yields as well. Finally, alcohols 41−43 were coupled with 2-iodoresorcinol (33), using DIAD and PPh 3 as reagents, to afford carbohydrate-substituted iodoarenes 44, 45, and 46. Catalysts 44, 45, and 46 were than applied in Kita's spirolactonization of naphthole derivative 1. The catalysts with the substituted benzyl groups at position 4 gave similar enantiomeric ratios (79:21 for 44 and 81:19 for 45 respectively) and yields ranging between 70% and 80% (Table 1, entries 16 and 17). Enantioselectivities and yields were in a similar range as for the previously published catalyst 8d. 1-Naphthylmethyl-substituted α-d-glucose-based iodoarene 46, however, resulted in a yield in the same range as for 44 and 45 but in a slightly higher enantioselectivity (er 86:14, Table 1, entry 18), which proved our assumption that the substituents at position 4 of the sugar moiety were critical. Finally, we also prepared a catalyst with sterically demanding 1-naphthylmethyl groups at all positions of the sugar moieties. If position 4 is indeed the main origin of stereoinduction in our catalyst design, such fully 1-naphthylmethyl protected iodoarenes should provide similar selectivities like 46.
Indeed, position 4 at the monosaccharide derivative is most important for the stereoinduction of our carbohydrate-based iodoarene catalyst, but the other positions seem to have some influence too.

General Remarks
Reactions in dry solvents were carried out under an atmosphere of nitrogen using Schlenk techniques. Dry THF and PhMe were distilled from sodium and benzophenone, and dry CH 2 Cl 2 and DMF were distilled from P 4 O 10 . Solvents used for preparative column chromatography were of technical grade and distilled prior to their use. Petroleum ether (PE) refers to the fraction boiling at 60−90 • C. Silica gel "60 M" from Macherey-Nagel (Düren, Germany) was used for preparative column chromatography. For TLC, "Polygram Sil G/U 254 " plates purchased from Machery-Nagel were used. Melting points were measured with a Büchi "Melting Point M-560" apparatus (Flawil, Switzerland). Optical rotations were determined with a Perkin-Elmer "Polarimeter 341" (Wattham, MA, USA). NMR spectra were measured with a Bruker "Avance III HD 400" or a Bruker "Avance III HD 300 NanoBay" spectrometer (Billerica, MA, USA). NMR spectra were calibrated to the solvent signal (CDCl 3 : 1 H 7.27 ppm, 13 13 C-HSQC) were recorded for peak assignment, and the atoms were numbered according to the carbohydrate nomenclature. Chemical shifts were given in ppm. High-resolution mass spectra (HRMS) were recorded on a Bruker "maXis 4G" with electrospray ionization and a time-of-flight detector. Elemental analysis was performed using a HEKAtech "Euro 3000 CHN" (Wegberg, Germany). Enantiomeric ratios were determined by chiral HPLC using a Dr Maisch (Ammerbuch-Entringen, Germany) "Reprosil Chiral-OM, 5 µm, 125 × 4.6 mm" column with n-hexanes/2-propanol, 85/15, as eluent and 0.7 mL/min as the flow rate. The Supplementary Materials contains copies of NMR spectra, HPLC chromatograms, and crystallographic data.

General Procedure A: Benzylic Substitution
Respective carbohydrate derivative (1.0 equiv) was dissolved in dry DMF (0.15 m), and NaH (2.0 equiv for each hydroxyl group; 60% dispersion in mineral oil) was added in small portions. After complete addition, the resulting mixture was stirred for 10 min at room temperature. The reaction mixture was cooled in an ice bath, and the particular benzylic bromide (1.0 to 1.5 equiv for each hydroxyl group) was slowly added. The mixture was allowed to reach room temperature and was stirred until TLC showed complete consumption of the starting material, at which point the reaction was quenched by the addition of MeOH. The solvent was evaporated in vacuo, and the residue was re-dissolved in H 2 O and EtOAc. The aqueous layer was extracted with EtOAc (3 times), and the combined organic layers were dried (Na 2 SO 4 ). The crude product was purified by column chromatography.

General Procedure B: Changing Isoproylidene Groups to Benzyl Groups
The respective isopropylidene-protected compound (1.0 equiv) was dissolved in THF (0.03 m), aqueous HCl (10 mL for each mol of isopropylidene-protected compound; 2 m) was added, and the resulting mixture was stirred at room temperature until TLC indicated complete consumption of the starting material. The reaction mixture was neutralized with solid NaHCO 3 , and the solvent was evaporated in vacuo. The solid residue was extracted with 2-propanol, which was evaporated in vacuo afterwards. The crude product was dissolved in dry DMF (0.05 m with respect to the isopropylidene-protected compound), and NaH (4.0 equiv for each removed isopropylidene group, 60% dispersion in mineral oil) was added. The reaction mixture was stirred 10 min at room temperature, and BnBr (3.0 equiv for each removed isopropylidene group) was added slowly. The reaction mixture was stirred at room temperature overnight, quenched with MeOH, and the solvent evaporated in vacuo. The residue was re-dissolved in H 2 O and EtOAc, the aqueous layer was extracted with EtOAc (3 times), and the combined organic layers were dried (Na 2 SO 4 ). The crude product was purified by column chromatography.

General Procedure C: Mitsunobu Reaction
Particular iodophenol (1.0 equiv) and carbohydrate derivatives (1.0 to 1.1 equiv for each hydroxyl group at the iodophenol) were dissolved in dry THF (0.2 m with respect to the iodophenol), and PPh 3 (1.15 equiv for each hydroxyl group at the iodophenol) and DIAD (1.2 equiv for each hydroxyl group at the iodophenol) were added. The resulting mixture was stirred at room temperature for 15 h, PPh 3 (0.3 equiv for each hydroxyl group at the iodophenol) and DIAD (0.3 equiv for each hydroxyl group at the iodophenol) were added, and the reaction mixture was stirred at room temperature for an additional 4 h. The solvent was evaporated in vacuo, and the residue was purified by column chromatography.

General Procedure D: Mitsunobu Reaction
Compound 33 (1.0 equiv) and PPh 3 (3.0 equiv) were dissolved in dry PhMe (0.05 m with respect to 33), DIAD (3.0 equiv) was added, and the mixture was stirred at room temperature for 5 min. A solution of carbohydrate derivative (2.5 equiv) in dry PhMe (0.2 m) was added, and the reaction mixture was heated to reflux for 17 h. The solvent was evaporated in vacuo, and the residue was purified by column chromatography.

General Procedure E: Introduction of the Trityl Group
A solution of methyl glycoside (1.0 equiv) and Ph 3 CCl (1.2 equiv) in dry pyridine was heated to 60 • C until TLC showed complete consumption of the starting material. MeOH was added, and the solvent was evaporated in vacuo. The residue was coevaporated with PhMe (3 times) and purified by column chromatography.

General Procedure F: Cleavage of the Trityl Group
To a solution of trityl-protected compound (1 equiv) in CH 2 Cl 2 (0.3 m), H 2 O (4 equiv) and TFA (2 equiv) were added, and the mixture was stirred at room temperature until TLC showed complete consumption of the starting material. The reaction mixture was diluted with CH 2 Cl 2 and H 2 O. The aqueous phase was saturated with NaCl and extracted with CH 2 Cl 2 (3 times). The combined organic layers were washed with saturated NaHCO 3 solution and dried (Na 2 SO 4 ). The crude product was purified by column chromatography.

General Procedure G: Kita's Spirolactonization
The spirolactonization was performed according to known procedure in the literature, and the enantiomeric ratio was determined as described previously [16,20].

Conclusions
In summary, thirteen new carbohydrate-based iodoarene catalysts were prepared. Three of them gave higher er's in Kita's spirolactonization than the previously reported ones. The way which led us to the improved catalyst design was described in detail. Investigation of the molecular structure and catalysts derived from different monosaccharides provided insights into the origin of the stereoselectivity of the carbohydrate-based iodoarene catalysts.
First, we changed the type of linkage between the carbohydrate moiety and the aromatic ring, and we found that a phenolic connection outmatched a benzylic linkage in terms of enantioselectivity of the catalyst. Next, we investigated if a different catalyst symmetry was beneficial for the catalyst design, and we found that a C 2 symmetrical catalyst worked better than one with C 1 symmetry. Afterwards, we introduced different carbohydrates as substituents as our catalyst. With d-galactose, the catalyst gave poor enantioselectivity by its application in Kita's spirolactonization. The catalyst with d-galactose worked best. This led us to the assumption that position 4 is of major importance for the stereoselectivity. To gain further insight into the origin of stereoselectivity, we prepared crystals of one catalyst in order to obtain S-ray data to get the detailed molecular structure of our catalysts. With these findings at hand, we prepared improved catalysts with bulky substituents at position 4. One of our new catalysts gave a maximum er of 88:12 by its application in Kita's spirolactonization.