Sulfonimidation via ring-opening of 2-oxazolines with acidic sulfonimide nucleophiles

Acidic sulfonimide nucleophiles including dibenzenesulfonimide, o-benzenesulfonimide, dimethanesulfonimide, and N -(methylsulfonyl)-benzenesulfonamide are discovered to open a variety of alkyl-, aryl-and heteroaryl-2-oxazoline rings to provide the sulfonimidation products in refluxing 1,4-dioxane. The electron-rich 2-oxazoline substrates worked well for the nucleophilic ring-opening reactions while no reaction took place for the electron-poor 2-oxazoline substrates.


Introduction
Oxazolines are versatile ligands and directing groups in organic synthesis.Chiral bis-oxazolines have been extensively used as chiral ligands for asymmetric synthesis. 1,2Aryl-2-oxazolines have served as an important class of directing groups for metalation via complex-induced proximity effect (CIPE), also known as directed ortho-metalation (DoM), pioneered by Meyers, Beak, and Snieckus, et al. 3,4 On the other hand, oxazolines are also useful building blocks.For example, 2substituted 1,3-oxazolines, have been used as starting monomers in the cationic polymerization reactions intended for making poly(ethylene imines) of various compositions and molecular weights. 5,6nother approaches is through nucleophilic ring-opening of oxazolines such as 2-phenyl-2oxazoline (1) to give products 2 as depicted in Scheme 1.A variety of nucleophiles have been applied in this type of transformation.Thiophenols 7 and thiols 8 have been used as Snucleophiles; and phenols 9 as O-nucleophiles.C-Nucleophiles in the form of stabilized carbanions have been used to open oxazoline rings with the aid of methyl triflate. 10Halides as nucleophiles generally react with oxazolines in the form of TMS-X. 11Other nucleophiles that have been used to open oxazoline rings also include selenium- 12 and tellurium- 13 -containing nucleophiles.

Results and Discussion
In the process of pursuing our on-going interest in fluorination via C-H activation, we fortuitously discovered that N-fluorobenzenesulfonimide (NFSI) was able to open the oxazoline ring on 1 to afford sulfonimide 5 in 15% yield.It was quickly realized that fluorine was not actually required for the manipulation and dibenzenesulfonimide [HN(SO 2 Ph) 2 ] was able to serve as an acidic nucleophile to convert 1 into 5 in 93% yield in refluxing 1,4-dioxane (Scheme 2).To the best of our knowledge, this type of sulfonimidation has not been reported previously.Considering the importance of sulfonimides as important bioisosteres of carboxylic aicds in medicinal chemistry, [17][18][19][20] we decided to explore the scope and limitation of such a ring-opening reactions.Scheme 2. Ring-opening 2-oxazoline with dibenzenesulfonimide.
Our accidental discovery of sulfonimidation of 2-phenyl-2-oxazoline (1) to give sulfonimide 5 using HN(SO 2 Ph) 2 prompted us to more closely scrutinize this unique nucleophile.Sandwiched between two powerful electron-withdrawing phenylsulfonyl groups, the NH proton on the molecule is exceptionally acidic with a pK a value of 1.45, 21 with a calculated value of 3.24, 22 approximately as strong as phosphoric acid.As a result, we suspected that dibenzenesulfonimide is acidic enough to protonate the nitrogen atom on 1.Indeed, 1 moved to the baseline on TLC upon contact with HN(SO 2 Ph) 2 .
We initially surveyed the solvent effect on transformation 1→5.When 1 was heated with dibenzenesulfonimide in polar solvents including DMF and DMSO, no reaction was observed even at 150 °C for a prolonged time.The reaction took place in alcoholic solvents such as EtOH, n-BuOH, and isopentanol, but stalled at approximately 70% conversion.The reaction went to completion cleanly in ethereal solvents such as methyl t-butyl ether (MTBE), THF, and 1,4dioxane.At the end, 1,4-dioxane was chosen because it has the highest boiling point.Meanwhile, the optimal stoichiometry for HN(SO 2 Ph) 2 was found to be 1.5 equivalents.
With reaction conditions optimized, we investigated the scope and limitations of this methodology with regard to different oxazoline substrates 7. Most of them were prepared via a 2step sequence consisting of amide formation from carboxylic acid 6 with 2chloroethylamine•HCl, 23 followed by ring closure of the resulting amido-chloride with the aid of NaOH (Scheme 3). 24heme 3. Preparation of 2-oxazoline substrates.
However, this methodology did not work for electron-poor substrates.For example, no reaction was observed when 2-(4-nitrophenyl)-2-oxazoline (13) and HN(SO 2 Ph) 2 were heaterd togegther at reflux in 1,4-dioxane.Further, addition of 0.5 equiv of sodium hydride to boost the nucleophilicity and heating up to 150 °C for six hours did not provide any ring-opened product either.The same phenomenon was observed for another electron-poor substrate 2-(pyridin-4-yl)-2-oxazoline ( 14), derived from isonicotinic acid.The aforementioned observations are readily explained by invoking the intermediacy of protonated oxazoline 1′ (vide infra, Scheme 4).Electron-rich 2-oxazoline substrates such as 3, 8, 9, 10, 11 and 12 donate electrons to the oxazoline ring, promoting protonation. 26On the other hand, electron-poor 2-substituents withdraw electrons from the oxazoline ring so that the nitrogen is no longer basic enough to be protonated by HN(SO 2 Ph) 2 and the reaction does not proceed.
As far as the reaction mechanism is concerned, we speculate that the nitrogen atom is protonated by HN(SO 2 Ph) 2 , furnishing ammonium intermediate 1′ and (PhSO 2 ) 2 N anion. 27his was derived from the fact that 1 was protonated upon contact with Brønsted-Lowry acid HN(SO 2 Ph) 2 to form 1ʹ. Meanwhile, the (PhSO 2 ) 2 N anion serves as a nucleophile to open the oxazoline ring in an S N 2 fashion to furnish the imidation product 5. Scheme 4. Proposed reaction mechanism.
We then proceeded to explore additional viable acidic sulfonamide nucleophiles for this ringopening reaction.Three additional nucleophiles were prepared.As shown in Scheme 5, N-(phenylsulfonyl)acetamide ( 27) was prepared by treating a suspension of benzenesulfonamide in CH 2 Cl 2 with acetic anhydride in the presence of a catalytic amount of TiCl 4 . 282,2,2-Trifluoro-N-(phenylsulfonyl)acetamide (28) was synthesized in the same manner using trifluoroacetic anhydride.Finally, N-(methylsulfonyl)benzenesulfonamide (29) was assembled in 35% yield by the reaction between benzenesulfonamide and methanesulfonyl chloride in the presence of Et 3 N and a catalytic amount of DMAP. 29Scheme 5. Preparation of additional acidic sulfonimide nucleophiles.
We then proceeded to examine a variety of nucleophiles for the oxazoline-opening reaction.With a pKa of 14.7, phthalimide was not acidic enough to open the 2-oxazoline ring under optimized reaction conditions; neither was benzenesulfonamide, with a pKa of 10.1. 30When acylsulfonamide 27 (calculated pKa, 5.51) also failed to open the oxazoline ring, 31 we began to suspect that a nucleophile with similar acidity to HN(SO 2 Ph) 2 would be more amenable to such transformations.Indeed, heating 1 with triflimide [HN(COCF 3 ) 2 , pK a , 2.95) and trifluoroacetylsulfonamide 28 (calculated pK a , 3.52) in refluxing 1,4-dioxane provided the desired ring-opening products.Unfortunately, the resulting products were not stable on silica gel and less than 27% of the products were isolated.
Thankfully, dimethanesulfonimide (calculated pK a , 3.43), 22 and mixed sulfonimide 29 (calculated pKa, 3.63), 22 worked well in the ring-opening reaction.Further, both nucleophilesdimethanesulfonimide and mixed sulfonimide 29 -worked on alkyl-, aryl-and heteroaryloxazolines to produce the ring-opening products with yields ranging from 56% to 84% (Table 2).The relatively moderate yields for reactions (Table 2) can be attributed to the fact that alkylsulfonimides are less acidic than aryl-sulfonimides.Regrettably, sterically hindered oxazolines 38 and 39 (Figure 1) did not react in attempted ring openings using this methodology.This is most likely due to the fact that steric hindrance of the α-neopentyl cannot accommodate the bulky nucleophile (PhSO 2 ) 2 N anion.Many efforts were made to react the sulfonimides with other nucleophiles such as azide, but these invariably produced the ring-closure products, the very starting materials to make those linear amide-sulfonimides.This is not completely surprising because once the NH bond on the amide encounters even weak basic conditions, the intramolecular ring-closure prevails because it is very much more kinetically favored than the intermolecular S N 2 substitution, especially considering that sulfonimides are such good leaving groups.Therefore, the synthetic utility of the linear ring-opening products are limited to their corresponding linear forms.

Conclusions
We have discovered that acidic sulfonimides can serve as nucleophiles to open 2-substituted oxazolines.This methodology of sulfonimidation works more efficiently for oxazoline substrates with an electron-rich 2-substituent, while no reaction took place for oxazolines with electronpoor 2-substituents.The resulting linear amide-sulfonimides may serve as bioisosteres of carboxylic acids in drug discovery.

Experimental Section
General.All reactions were performed in anhydrous solvents under a N 2 atmosphere.Solvents were purchased from Alfa Aesar and utilized without further purifications.2-Phenyl-2-oxazoline (1), 2-ethyl-2-oxazoline (3), dibenzenesulfonimide, bistrifluoroacetamide, and dimethanesulfonimide are commercially available and were used without further purifications.Analytical thin-layer chromatography (TLC) was carried out using Silica G TLC plates, 200 µM with UV254 (SORBENT Technologies), with visualization by UV or iodine.Flash chromatography was performed using standard grade silica gel (60 Å, 230-400 mesh; SORBENT Technologies).Melting Points were taken using Vernier Melt Station LabQuest 2 and were not corrected.NMR spectra were acquired using an Agilent VNMRS spectrometer equipped with one NMR probe (500 MHz for 1 H, 125 MHz for 13 C, 470 MHz for 19 F).Spectra were processed using MNova software (Mestrelab).Chemical shifts are reported in parts per million (ppm), coupling constants (J) in Hz and are calibrated to residual protonated solvent.Infrared spectra of neat samples were acquired using a PerkinElmer Spectrum 100 FT-IR spectrometer, with solid samples analyzed using a Universal ATR (attenuated total reflectance) sampling accessory.GC-MS was performed on a Hewlett Packard HP6890 Series GC System and a 5973 Mass Selective Detector.

Representative procedures for substrate preparation
Step 1 2-Thiophene carboxylic acid (1 g 7.8 mmol) was dissolved in 10 mL of anhydrous CH 2 Cl 2 and cooled to 0 o C; 5 drops of DMF was added to the suspension.Oxalyl chloride (1.98 g, 15.8 mmol) was added dropwise to the reaction mixture.The reaction was warmed to rt and stirred for 3 h.The solvent was removed in vacuo and the residue was dried for 1 h in high vacuum.The residue was dissolved in 5 mL of CH 2 Cl 2 and added dropwise to a suspension of chloroethylamine•HCl (1.08 g, 9.4 mmol), Et 3 N (2.52 g, 24 mmol) in 7 mL of CH 2 Cl 2 at 0 o C. The reaction stirred for 3 h and was warmed to rt.The reaction was diluted with CH 2 Cl 2 and washed once with sat.aq.ammonium NH 4 Cl then concentrated in vacuo.The crude reaction mixture was purified by flash chromatography in a solvent system of hexane:ethyl acetate 2/1, respectively, to afford N-(2-chloroethyl)thiophene-2-carboxamide.
Step 2 N-(2-Chloroethyl)thiophene-2-carboxamide, (1.07 g 5.7 mmol) was added to a solution of NaOH (0.286 g 7.2 mmol) dissolved in EtOH.The reaction was heated at 50 o C for 1 h and cooled to rt.The reaction was diluted with EtOAc and washed twice with brine.The organic layer was dried over MgSO 4 , filtered, and then concentrated.