5(6)-anti-Substituted-2-azabicyclo[2.1.1]hexanes. A Nucleophilic Displacement Route

Abstract

Nucleophilic displacements of 5(6)-anti-bromo substituents in 2-azabicyclo[2.1.1]hexanes (methanopyrrolidines) have been accomplished. These displacements have produced 5-anti-X-6-anti-Y-difunctionalized-2-azabicyclo[2.1.1]hexanes containing bromo, fluoro, acetoxy, hydroxy, azido, imidazole, thiophenyl, and iodo substituents. Such displacements of anti-bromide ions require an amine nitrogen and are a function of the solvent and the choice of metal salt. Reaction rates were faster and product yields were higher in DMSO when compared to DMF and with CsOAc compared to NaOAc. Sodium or lithium salts gave products, except with NaF, where silver fluoride in nitromethane was best for substitution by fluoride. The presence of electron-withdrawing F, OAc, N₃, Br, or SPh substituents in the 6-anti-position slows bromide displacements at the 5-anti-position.

Introduction

Pyrrolidines 1, especially those with hydroxy,¹ amino,² fluoro,³ or thio⁴ substituents in a 1,2-relationship β to the nitrogen atom, are a valuable source of biologically significant molecules. One strategy in the search for new bioactive molecules is to incorporate key pharmacophoric units into inflexible structures.^5–7 Viewed in this light methanobridged pyrrolidines 2 (2-azabicyclo[2.1.1]hexanes) that display their functionalities in defined spatial orientations are of interest. Such molecules may prove to be valuable scaffolds for incorporation into proteins,⁸ for drug discovery, or for other purposes.^3b To realize this potential there is a need for practical methods to introduce a diverse array of heteroatom substituents onto these structures.⁹

Heteroatoms at C₅(C₆) of N-acyl-2-azabicyclo[2.1.1]hexanes have been introduced by rearrangement routes (X = syn- or anti-halogen, hydroxyl),¹⁰ nucleophilic ring closure of cyclobutanes (X = syn-SePh),^6a or a thermal 2+2 cycloaddition (X/H = difluoride).¹¹ There are a few examples of N-acyl monoheteroatom-substituted anti-5-hydroxy-2-azabicyclo[2.1.1]hexanes 2a (X = OH) formed by reductive dehalogenation of 5-anti,6-anti-bromohydrins. ^3b,10a,10d

Recently, we described the preparation of N-BOC-5-syn- and 5-anti-carboxy-2-azabicyclo[2.1.1]hexanes 2e (R = H, X = COOH), isolated mainly as the syn-5-carboxy isomers,¹² and their use for introduction of heteroatom functionality into this ring system.^2c The Curtius rearrangement was especially useful for the stereospecific conversion of 5-syn- and 5-anti-acids to the corresponding 5-syn- and 5-anti-amines, isolated as their carbamates 2b [R = H, X = NHCOOBn(Et)]. The major 5-syn-carboxylic acid 2e has been used to introduce other 5-heteroatoms by radical decarboxylative substitutions of either the acid or its Barton ester. Reactions led to mainly 5-syn-chloro, 5-syn-bromo, 5-syn-iodo and 5-syn-pyridylthioether substitution products admixed with only minor amounts of the 5-anti isomers (0–11%); an exception was the iodide that gave 17–44% 5-anti-isomer. Yields were generally poor and this non-stereospecific method is not of general utility as a source of halide or thioether substitution. The rearrangement route remains the most useful for introduction of 5-anti-hydroxy and 5-anti-bromine groups as in 2a and 2e. Thus, methods to displace these substituents by other nucleophiles are welcome.

Nucleophilic substitution reactions of 5-tosylbicyclo[2.1.1]hexanes, the parent carbon bicycle of structures 2, provide insights into the reactivity of 5-substituents in this strained ring system. The substitution reactions of 5-syn substituents occur fairly easily, proceed with retention of configuration, but are accompanied by rearrangement products. The syn-5-alcohol 3a and phosphorus tribromide afforded a product assigned as the 5-syn-bromide 3b (17%) admixed with some 4-bromocyclohexane, while syn-5-tosylate 3c reacts with tetrabutylammonium chloride at 5 °C for 29h to afford a product assigned as 5-syn-chloro[2.1.1]hexane 3d (43%) with retained stereochemistry.^13a Both these products are suggestive of neighboring group participation by the neighboring methanobridge.^13b On the other hand nucleophilic displacement of 5-anti-substituents, our goal, is more difficult. The syn-tosylate 3c reacted at a rate 3 × 10⁶ that of the 5-anti-tosylate 4.^13b,c In order to induce a reasonable rate for acetic acid solvolysis of the anti-tosylate 4, a temperature of 164 °C was required. More importantly, none of the products retained the bicyclo[2.1.1]hexane structure. Acetolysis of 4 produced 4-cyclohexenyl tosylate (80%), 4-cyclohexenyl acetate (8%), and bicyclo[3.1.0]hex-2-yl acetate (8%) as the major products.^13c

Nevertheless, there are two examples of nucleophilic displacement reactions of 5-anti-substituents in N-acyl-2-azabicyclo[2.1.1]hexanes by fluoride in which products have been isolated that maintain the integrity of the heterobicyclic structure. The conversion of alcohol 5a to fluoride 6a was carried out using bis(2-methoxyethyl)aminosulfur trifluoride [BAST or Deoxo-Fluor] in refluxing methylene chloride (63%) (eq 1).^3b Limited success was observed with the replacement of the 5-anti-iodo substituent of 5b by fluoride using AgF/nitromethane at 80 °C/4h to give 6b (19%) (eq 2).^2c Retention of stereochemistry was observed in both cases of displacement reactions of C₅-anti-substituents. We have not been successful in nucleophilic displacements of 5-anti-bromo substituents in N-acyl-5-anti-bromo-2-azabicyclo[2.1.1]hexanes. (See supplemental material.)

eq 1

It occurred to us that replacement of the N-acyl substituent by an N-alkyl group might facilitate nucleophilic substitution reactions. Malpass and White¹⁴ have shown that a free amine can facilitate displacements in a normally slow reacting 7-norbornyl position.¹³ The 7-bromo group in N-benzyl-anti-7-bromo-2-azabicyclo[2.2.1]heptane 7 was displaced by various nucleophiles at 100–110 °C in DMF to give products 8. Only products with retained stereochemistry were observed, presumably the result of neighboring group participation.¹⁴

eq 2

The key objective of this work was to see if replacement of N-acyl groups by N-benzyl in readily available 5-anti-bromo- and 5-anti-,6-anti-dibromo-2-azabicyclo[2.1.1]hexanes would allow for displacement of the bromine atom by useful heteroatom nucleophiles.¹⁵ Specifically, dibromide 9, monobromide 10, and fluorobromide 11 have been chosen as substrates. We now describe conditions that enable preparation of 5(6)-anti-substituted-2-azabicyclo[2.1.1]hexanes with halogen, nitrogen, sulfur, and oxygen containing groups X starting from these bromides. Since preparation of 5-anti-bromides with compatible substituents at other ring positions is feasible,^3b,9,10 the functional group modifications described for these bromides should prove useful in the preparation of more highly functionalized 5(6)-substituted-2-azabicyclo[2.1.1]hexanes 2. Some of these structures can be precursors of methanoprolines 2 (R = COOH);^3b,16 we especially desired to prepare novel 5,6-dihydroxy-, 5,6-difluoro-, 5,6-diamino-, and mixed 5,6-hydroxyfluorides that are not available by other synthetic routes.^7b

Results and Discussion

The requisite N-benzyl dibromides 9–11 were prepared from the N-alkoxycarbonyl dihydropyridine photoproduct 12 (Scheme 1).^7b The known dibromide 13^7b was selectively monodebrominated using (TMS)₃ SiH/toluene/70 °C to give monobromide 14 (74%). Conversion of alkene 12 to bromofluroride 15 (53%) was carried out using NBS/nitromethane/Et₃N·3HF.¹⁷ Hydrogenolysis of the carbobenzyloxy protecting groups of 13–15 with H₂/Pd(OH)₂/MeOH and subsequent benzylation with benzyl bromide/Et₃N/CH₃CN afforded the N-benzyl compounds 9–11 in 51–69% overall yields for the two steps.

SCHEME 1.

Synthesis of N-Benzylbromides.

Our first attempts to effect nucleophilic displacements of dibromide 9 (Scheme 2) were carried out in DMF under conditions used successfully by Malpass¹⁴ for halide displacements with bromide 5 (R = Bn, Y = H). The results are shown in Table 1. Dibromide 9 was slowly converted to its diacetate 16a using excess cesium acetate¹⁸ (entry 1). But, displacement of the second bromine was difficult. Even after 5 days there was unreacted starting material and a large amount of bromoacetate 17 in the reaction mixture. For the stereochemical assignment of diacetate 16, the protons H₅/H₆ are identical and appear as a singlet in the ¹H NMR spectrum. The retained 5-anti-6-anti stereochemistry is apparent from the absence of coupling between H₁ or H₄ and their vicinal syn protons H₅/H₆. In this ring system these syn protons characteristically do not show vicinal coupling.^2c For bromoacetate 18, there is the characteristic W-plan coupling between the syn protons H₅ and H₆ (J = 7.2 Hz).^7b Methanolysis of the diacetate 16a afforded diol 16b. Attempted preparation of difluoride 18 from dibromide 9 using AgF/DMF formed instead the pyrrole aldehyde 19 (entry 2).¹⁹

SCHEME 2.

Reactions of dibromide 9.

Table 1.

Nucleophilic Substitutions of Dibromide 9.

entry	substrate	reagent	conditions	product	X	Y	yield (%)
1	9	CsOAc	DMF/60 °C/5d	16a	OAc	OAca	40
				17	Br	OAc	42
2	9	AgF	DMF/50 °C/8h	19	-	-	72
3	9	NaN3	DMF/60 °C/8d	20	N3	N3	34
				21	Br	N3	49b
4	9	CsOAc	DMSO/60 °C/5d	16a	OAc	OAc	89
5	9	NaOAc	DMSO/60 °C/5d	16a	OAc	OAc	14c
				17	Br	OAc	59
6	9	NaF	DMSO/70 °C/5d	19	-	-	22d
7	16b	BAST	CH2Cl2/25 °C/12h	18	F	F	24
				19	-	-	40
8	9	AgF	CH3NO2/50 °C/8h	18	F	F	52
				19	-	-	24
9	9	NaN3	DMSO/60 °C/2d	20	N3	N3	87
10	9	NaSPh	DMSO/60 °C/5h	22	SPh	SPh	37
				23	Br	SPh	29

^aAlso 10% dibromide 9.

^bAlso 17% of dibromide 9.

^cAlso 15% dibromide 9.

^dAlso 50% dibromide 9.

In order to introduce nitrogen functionality bromide 9 was reacted with sodium azide in DMF (entry 3) to give diazide 20 and azidobromide 21. As noted with CsOAc (entry 1), it was difficult to replace the second bromine in azidobromide 21 despite extended reaction times (8d). The symmetrical diazide 20 gave a singlet for H₅/H₆ while bromoazide 21 showed the characteristic W-plan coupling for H₅/H₆ indicating that both structures have 5-anti,6-anti stereochemistry. The long reaction times and low yields in DMF solvent for preparation of desired diacetate 16a and diazide 20, along with the failure to prepare the desired difluoro isomer 18, initiated a search for alternative superior reaction conditions.

Dimethylsulfoxide was found to be a superior solvent for nucleophilic displacement reactions of dibromide 9 (entry 4).²⁰ The substitution of DMSO for DMF, and otherwise the same reaction conditions in entry 1 for the reaction with CsOAc, resulted in complete conversion to the diacetate 16a in a suitable yield after 5 days. The use of cesium acetate was clearly superior to sodium acetate in this reaction (entry 5). Our pleasure was tempered somewhat by the failure of DMSO as solvent to enable dibromide 9 to be converted to a desired difluoride 18 in the presence of NaF (entry 6); again only the pyrrole aldehyde 19 was obtained. But, it was discovered that difluoride 18 could be obtained in small yield (24%) by reaction of diol 16b with BAST (entry 7).^3b The symmetrical difluoride evidenced the expected multiplet AA’XX’ pattern in the ¹H NMR spectrum shown in Figure 1. The main product in the reaction was the oxidized ring cleaved pyrrole aldehyde 19. Later, it was found that the difluoride 18 could be made directly from the dibromide 9 in better yield by reaction with silver fluoride in nitromethane as solvent (entry 8).

Figure 1.

400 MHz ¹H-NMR Pattern for H₅/H₆ Protons in Difluoride 18 (CDCl₃).

The symmetrical diazide 20 also was prepared from dibromide 9 in both vastly improved yield (87%) and in shorter time (2 days) simply by replacing DMF with DMSO solvent (entry 9). With DMSO solvent it was also possible to prepare the symmetrical thiophenyl ether 22, although after 5h some bromothiophenyl ether 23 and unreacted dibromide 9 remained (entry 10).

Reactions of monobromides 10 and 11

We next turned our attention to the monobromide 10 (Scheme 3). Its substitution reactions are tabulated in Table 2. Our initial efforts again focused upon reactions in DMF solvent because of precedent.¹⁴ With silver acetate in DMF bromide 10 gave acetate 24a in moderate yield (entry 1). This was methanolyzed using K₂CO₃/methanol to give alcohol 24b (84%). To show that the benzyl group could be removed without destruction of the strained ring, alcohol 24 was hydrogenolyzed and the resulting amine was protected by reaction with (BOC)₂O to give N-BOC alcohol 25 (92%).

SCHEME 3.

Reactions of Monobromides 10.

Table 2.

Nucleophilic Substitutions of Bromide 10.

entry	substrate	reagent	conditions	product	X	yield (%)
1	10	AgOAc	DMF/60 °C/12h	24a	OAc	54
2	10	AgF	DMF/70 °C/24h	19	-	42a
3	10	NaF	DMF/70 °C/12h	10	Br	88
4	10	NaN3	DMF/70 °C/12h	27	N3	51
5	10	LiNub	DMF/70 °C/8d	29	Nub	55
6	10	NaIc	DMF/70 °C/3d	31	I	50d
7	10	CsOAce	DMSO/70 °C/6h	24a	OAc	90
8	10	NaF	DMSO/70 °C/12h	10	Br	82
9	24b	BAST	CH2Cl2/40 °C/12h	26	F	62
10	10	AgF	CH3NO2/50 °C/12h	26	F	80
11	10	NaN3	DMSO/70 °C/5h	27	N3	88
12	10	NaSPh	DMSO/60 °C/5h	30	SPh	77
13	10	NaIf	Acetone/reflux/4d	31	I	74g

^aBromide 10 was stable in DMF/70 °C/12h; 90% recovery, no 19 formed.

^bNu = N-imidazole.

^c3 Equiv.

^dAdmixed with unreacted bromide 10.

^eTo separate samples of the monobromide 10 in DMSO-d6 there was added 1.5 equiv of either NaOAc (sample A) or CsOAc (sample B). After 6h the conversion to acetate 24a was 39% for sample A and 66% for sample B by NMR analysis.

^f20 equiv.

^gTotal conversion of 10.

Monobromide 10 and AgF in DMF gave the same ring-opened and oxidized pyrrole aldehyde 19 (entry 2) observed upon reaction of dibromide 9 under these conditions. Bromide 10 in DMF did not react with NaF (entry 3), but it did react with NaN₃ and gave azide 27 in moderate yield (entry 4). The azide 27 was reduced using triphenylphosphine/water and the resultant amine was reacted with (BOC)₂O to afford the protected carbamate 28. An N-imidazole ring could be introduced by generation of lithium imidazole in DMF and reaction with bromide 10 to give amine 29 (entry 5). Sodium iodide (3 equiv) effected partial displacement of bromide ion to give an inseparable 50:50 mixture of bromide 10 and iodide 31 (entry 6).

DMSO again proved to be a superior solvent for the replacement of bromide using cesium acetate (entry 7), and bromide 10 produced acetate 24a in high yield. Cesium acetate was found to be a better salt for the displacement than NaOAc. NaF in DMSO did not yield a fluoride with bromide 10 (entry 8). The desired fluoride 26 could be obtained from alcohol 24b upon reaction with BAST (entry 9), but the fluoride 26 was obtained in higher yield from bromide 10 using AgF in nitromethane (entry 10). DMSO was shown to be a better solvent for bromide 10 in its reactions with NaN₃ to give the azide 27 (entry 11) or with NaSPh to give the thiophenyl ether 30 (entry 12). It was possible to convert the bromide 10 to the iodide 31 using excess NaI/acetone after extended reflux (entry 13). ¹H NMR indicated complete conversion of the bromide. The 5-anti stereochemistry for all new compounds in Table 2 was indicated by the observation of W-plan ¹H NMR couplings (J_5,6 = 6.9–7.6 Hz).

The next substrate investigated was the bromofluoride 11 (Scheme 4) and its reactions are tabulated in Table 3. The bromofluoride 11 reacted slowly with CsOAc in DMF to give fluoroacetate 32a (entry 1). A sequence of methanolysis of the acetate 32a to alcohol 32b, then hydrogenolysis followed by acylation with (BOC)₂O, gave a desired fluoroalcohol 33 (86%). Formation of azide 34 from bromofluoride 11 was also a slow reaction (entry 2) and was accompanied by decomposition. The azide 34 was converted to the amine 35 using triphenylphosphine/water and the amine was acylated to give the acetamide 36.

SCHEME 4.

Reactions of Fluorobromide 11.

Table 3.

Nucleophilic Substitutions of Bromofluoride 11.

entry	substrate	reagent	conditions	product	X	yield (%)
1	11	CsOAc	DMF/70 °C/5d	32a	OAc	30a
2	11	NaN3	DMF/70 °C/5d	34	N3	43b
3	11	NaOAc	DMSO/70 °C/5d	32a	OAc	33c
4	11	CsOAc	DMSO/70 °C/5d	32a	OAc	90
5	11	NaN3	DMSO/70 °C/7d	34	N3	67
6	11	NaSPh	DMSO/60 °C/5h	37	SPh	15d
7	11	NaSPh	DMSO/60 °C/9d	37	SPh	69e

^aAlso 64% unreacted 11.

^bAlso 31% unreacted 11.

^cAlso 64% unreacted 11.

^dAlso 66% unreacted 11.

^eAlso 4% unreacted 11.

While fluoroalcohol 32b showed coupling between OH-F (J = 3.9 Hz),²¹ there was no evidence for such coupling in either the fluorine or proton NMR spectra of amine 35 or amide 36. Molecular models indicate the 5-anti and 6-anti substituents are not actually parallel, but point slightly away from each other. The 5-anti,6-anti arrangement of halogen substituents was again indicated by W-plan couplings (J_5,6 = 7.1–7.8 Hz).

Replacement of solvent DMF by DMSO facilitated the displacement reactions of 11 to give fluoroacetate 32a with either NaOAc (entry 3) or more effectively with CsOAc (entry 4). The same solvent effect was observed in the improved yields in formation of azide 34 upon reaction of 11 with NaN₃ in DMSO (entry 5). It was also possible to prepare the fluorothioether 37 using NaSPh in DMSO, although the reaction was quite slow (entries 6,7).

The ease of bromide displacements in bromides 9–11 was dependent upon the adjacent substituent X. Monitoring of the disappearance of starting bromides indicated a relative reactivity order monobromide 10 > dibromide 9 > fluorobromide 11 (see Supplemental Section). Nucleophilic substitution reactions with the bromide 10 in DMSO solvent with CsOAc or NaN₃ required hours for completion, with the dibromide 9 a few days, and with the fluorobromide 11 five-seven days. In addition, upon displacement of one of the bromides of dibromide 9 by acetate, azide, or thiophenyl it took longer to displace the remaining bromides of bromoacetate 17, bromoazide 21, or bromo(phenylthio) ether 23 (Table 1). These reactivity orders indicate that all parallel heteroatom substituents in the adjacent methyl bridge, so far investigated, are rate retarding for bromide substitution.

One plausible explanation for the rate retarding effects of heteroatom groups is that electron withdrawal of the nitrogen lone pair by a second atom X reduces the ability for nitrogen atom interaction with the leaving bromide. In molecular orbital terms the nitrogen’s lone pair of electrons could interact with the σ* orbitals of the C-Br bond. On the basis of the electronegativity of the non-reacting C-X bond (H < Br < F), it might be predicted that the n → σ* overlap for the reacting C-Br bond should follow the order monobromide 10 > dibromide 9 > fluorobromide 11. To gain evidence about the substituent effect upon lone pair n → σ* interactions, NBO calculations were performed for structures 9–11.^22,23 The calculations, shown in Table 4, indicate the nitrogen lone pair electrons, front lobe and rear lobe, interact with the σ* orbitals of both C₅ and C₆ substituents. Pictures in Figure 2 show the orbital overlap for the two lobes of the nitrogen lone pair orbital in the dibromide 9.^23d The larger n → σ* interaction in each case occurs from the endo-N-benzyl conformer with the rear lobe (yellow) of the nitrogen atom overlapping (N → C₅Br).²⁴ While it is true that the calculated relative overlap stabilization energies (10 > 9 > 11) are consistent with the observed reactivity order of (10 > 9 > 11) with nucleophiles, the ground state interaction energies are too similar for this factor alone to explain the large relative rate differences. Indeed, the relative stabilization energies may differ appreciably as the corresponding transition state energies are approached. Electron lone pair orbital interactions would be expected to become more important as positive charge is created at C₅.

Figure 2.

n → σ* Orbital Overlaps for C-Br Bonds in Benzyl Dibromide 9.

In order to gain information on charged intermediates derived from N-methyl substrates 38a-c we performed single point energy calculations of N-methyl carbocations 39a-c (Scheme 5) by two different means: (1) Hartree-Fock 6-31+G(d,p)²⁵ and (2) B3LYP/6-31+G(d,p)²⁶ methods/basis sets using the Gaussian 03 suite of computations.²² We then optimized these structures for geometry using the same two methods. In all instances, save one, each optimization of a cation 39 led to an iminium ion 40; exo and endo isomers led to the same ions. The one exception occurred with bromofluoride 38c using method (2) in which the aziridinium ion 41c was the outcome of the calculation. Independently, we optimized iminium ion 40c using method 2. The fluoro aziridinium ion 41c was calculated to be 43.5 kcal/mol less stable than fluoroiminium ion 40c. See the supplemental for details on aziridinium ions 41a,b.²⁷

SCHEME 5.

Calculated Ionic Intermediates for Reactions of N-Methyl-azabicyclic Bromides 38–40.

Considering the large energy preference calculated for the gas phase iminium ions 40 versus the aziridinium ions 41, it is perhaps surprising that the 5-anti-bromo-2-azabicyclo[2.1.1]hexanes 9–11, related to 38a-c, can undergo nucleophilic substitutions to afford unrearranged 5(6)-anti products related to 42. A 1,2-alkyl shift of C₆ from C₁ to C₅ on the anti-face of a cationic species 39 leads to iminium ion 40. Unrearranged products 42 that have retained stereochemistry are consistent with an intermediate ion that under suitable conditions is resistant to rearrangement. Such an ion in solution might be an aziridinium ion such as 41, associated with its counterion.

Solvent effects are consistent with the need to stabilize a transition state leading to charged intermediates, such as 41a-c. The more polar DMSO was a more effective solvent than DMF for the displacement reactions of all substrates 9–11.^20,28 A “cesium ion effect” was also noted for acetate displacements.^28–30 Reactions of fluorobromide 11 with CsOAc and NaOAc are illustrative. This substrate remained 64% unreacted (Table 4, entry 3) with NaOAc in DMSO solvent at 70 °C after 5 days; but with CsOAc and the same temperature/solvent conditions fluorobromide 11 reacted completely (Table 3, entry 4). These reactions were run under heterogeneous conditions, and thus the salt solutions were concentrated. The greater solubility of CsOAc than NaOAc in DMSO and DMF, as well as lesser ion pairing of cesium salts, increases the ionic strength of the CsOAc reaction solutions and facilitates the formation of charged ions in the polar solvents.³⁰

Table 4.

Second order perturbative estimates (NBO basis) for n → σ* interaction energies for bromides 9–11.

entry	substrate	X	n → σ* (kcal/mol)a N → C5-Br N → C6-Xb		substrate	n → σ* (kcal/mol) N → C5-Br N → C6-X
1	9endo-N-Bnc	Br	1.71	0.96	9exo-N-Bn	0.96	1.71
2	10endo-N-Bn	H	1.75	(0.28)	10exo-N-Bn	0.89	(0.55)
3	11endo-N-Bn	F	1.63	(0.69)	11exo-N-Bn	0.92	(1.63)

^aGeometry optimizations, frequency calculations, and NBO analyses were performed at the B3LYP/6-311+G(2d,p) level of theory on 9–11.^22,23

^bThose bonds that do not have leaving groups for substitutions are in parentheses.

^cThe benzyl group has been arbitrarily assigned as endo (or exo) to enable us to distinguish the two bromides as C₅ and C₆ for purposes of this analysis.

Silver ions facilitated bromide displacements. The outcomes of the silver salt reactions we investigated were dependent upon counterion and solvent. Illustrative are the conversions of monobromide 10 in Table 2. AgOAc/DMF gave acetate 20a (entry 1). However, with AgF/DMF the reaction took a different course and an oxidative rearrangement occurred to give pyrrole aldehyde 19 (entry 2). The sodium salt NaF/DMF was unreactive with bromide 10 over 12h (entry 3), but the silver salt AgF/CH₃NO₂ provided fluoride 26 (entry 10).

Reactions of the alcohols 16b (25 °C) and 24b (40 °C) with BAST/CH₂Cl₂ to give fluorides 18 and 26 were markedly easier than conversion of bromides 9 (50 °C) or 10 (70 °C) to the fluorides, even with AgF/CH₃NO₂ (Table 1, entries 7,8 and Table 2, entries 10,11). The hydroxyl groups are activated for displacement by BAST after formation of O-sulfur bonds; fluorination reactions occurred in CH₂Cl₂ in a few hours.

The oxidative ring opening reaction to form the aromatic aldehyde 19 occurred under a variety of conditions. Reactions of dibromide 9 in Table 1 are informative. AgF/DMF afforded the aldehyde 19 (entry 1), but so did NaF/DMSO (entry 6). AgF/CH₃NO₂ gave a mixture of difluoride 18 and aldehyde 19 (entry 8). In two trials with NaN₃/DMF similar to entry 3, but using less pure non-crystalline dibromide 9, a small amount (<6%) of aldehyde 19 was obtained with air as the only recognized oxidant. Aldehyde 19 also formed during reactions of BAST with diol 16b (Table 1, entry 7). For proposed mechanisms to this oxidized ring-cleaved aldehyde 19, see the Supplemental Material.

Conclusion

The novel N-benzyl-5-anti,6-anti-dibromo-2-azabicyclo[2.1.1]hexane nitrogen mustard 9, the bromide 10 and bromofluoride 11 react with nucleophiles to give products with retained stereochemistry. We have observed single bromide displacement reactions, and somewhat slower displacements of two bromides by appropriate oxygen (acetate), nitrogen (azide, imidazole), thioether (phenylthio), and halide (fluoride, iodide) nucleophiles. The present synthetic route describes the first reported examples of 5-anti,6-anti diols, difluorides, diazides, dithioethers, fluoroamines, fluorothioethers, as well as the first 5-anti-imidazoles. We presently envision use of the diols, fluoroalcohols and difluorides as key intermediates for the preparation of methanoproline derivatives, desired in order to study substituent effects on amide conformations.

Experimental Section

N-Benzyl-5-anti,6-anti-dibromo-2-azabicyclo[2.1.1]hexane (9)

To a solution of dibromide 13^7b (1000 mg, 2.67 mmol) in methanol (75 mL) there was added Pd(OH)₂ (150 mg ). The solution was degassed and stirred under a H₂ filled balloon for 1 h at room temperature. The reaction mixture was filtered through Celite, the filtrate was evaporated and the residue was chromatographed on silica gel (9:1 EtOH/MeOH) to give 513 mg (80%) of dibromoamine 13-int at R_f = 0.58 (2:1 EtOH/MeOH); ¹H NMR (400 MHz, CDCl₃) δ 4.11 (s, 2H, H₅ and H₆), 3.87 (d, J = 6.0 Hz, 1H, H₁), 3.35 (br s, 1H, NH), 3.17 (s, 2H, 2H₃), 3.10 (d, J = 6.0 Hz, 1H, H₄); ¹³C NMR (100 MHz, CDCl_3, RT) δ 66.2, 52.3, 51.2, 47.1; HRMS m/z 239.9014, 241.9001, 243.8983, calcd for C₅H₈N^{79/79, 79/81, 81/81}Br₂ (M+H) 239.9023, 241.9003, 243.8983. To a solution of amine (0.50 g, 2.08 mol) in acetonitrile (20 mL) there was added Et₃N (1.69 g, 16.60 mol) and then BnBr (1.42 g, 8.30 mol) dropwise at room temperature. The reaction mixture was stirred at room temperature for 36 h. Solvent was removed in vacuo, ether (75 mL) was added, the mixture was stirred for 10 min at room temperature, then filtered and the residue was washed with ether (25 mL). Solvent again was removed in vacuo to afford crude dibromide 9. This was chromatographed on silica gel (hexanes/ether 4:1) to afford 590 mg (86%) of an off-white solid dibromide 9 at R_f = 0.75 (1:1 hexanes/ether); mp 68–70 °C; ¹H NMR (400 MHz, CDCl₃) δ 7.36-7.27 (m, 5H, Ph), 4.32 (s, 2H, H₅ and H₆), 3.85 (s, 2H, CH₂Ph), 3.61 (d, J = 6.6 Hz, 1H, H₁), 3.12 (dd, J = 6.6, 0.9 Hz, 1H, H₄), 2.92 (s, 2H, 2H₃); ¹³C NMR (100 MHz, CDCl₃) δ 137.8, 128.5, 127.4, 70.1, 58.6, 55.1, 52.4, 51.1; HRMS m/z 329.9482, 331.9472, 333.9453, calcd for C₁₂H₁₄N^{79/79,79/81,81/81}Br₂ (M+H) 329.9493, 331.9473, 333.9453. Hexanes (0.3 mL) were added to a vial with a syringe that contained about 5 mg of dibromide 9 dissolved in ether (0.2 mL). The vial was wrapped with aluminum foil and small holes were made with a syringe and the solution was allowed to sit for 3 d to give crystals suitable for x-ray crystallography.

N-(Benzyloxycarbonyl)-5-anti-bromo-2-azabicyclo[2.1.1]hexane (14)

To a solution of dibromide 13 (693 mg, 1.8 mmol) in toluene (50 mL) was added (TMS)₃SiH (596 μL, 1.9 mmol) and AIBN (40 mg). The resulting solution was allowed to stir at 70 °C for 3 h. The solvent was concentrated in vacuo and flash chromatography gave 407 mg (74%) of the monobromide 14 at R_f = 0.39 (2:1 hexanes/ether); 1H NMR (400 MHz, CDCl₃) δ 7.11 (m, 5H), 4.91 (s, 2H), 4.19 (br d, J = 6.3 Hz, H₅), 3.58 (d, J = 8.4 Hz, H₁), 3.27 (d, J = 9.0 Hz, H₃), 3.23 (d, J = 9.0 Hz, H₃), 2.79 (dm, J = 8.1 Hz, H_6anti), 2.67 (br, 1H, H₄), 1.40 (dd, J = 8.1, 6.3 Hz, H_6syn); ¹³C NMR (100 MHz, CDCl₃) δ 155.5, 136.9, 128.9, 128.5, 128.4, 68.6, 65.2, 55.2, 49.4, 46.3, 39.3 and 25.5; HRMS m/z found 296.0284, calcd for C₁₃H₁₄NO₂Br (M+H) 296.0281, m/z 318.0105 calcd for C₁₃H₁₃NO₂BrNa (M+Na) 318.0105.

N-Benzyl-5-anti-bromo-2-azabicyclo[2.1.1]hexane (10)

To a solution of the monobromide 14 (708 mg, 2.4 mmol) in MeOH (40 mL) was added Pd(OH)₂ (71 mg) and the resulting solution was degassed and allowed to stir for 1 h at RT under hydrogen. After 1 h the catalyst was filtered via Celite and the solvent was removed in vacuo to give 500 mg of the crude amine. Without further purification the amine was dissolved in acetonitrile (20 mL) and to the resulting solution was added Et₃N (405 mg, 4.0 mmol) and BnBr (328 mg, 1.9 mmol). The resultant solution was stirred at RT for 3 days. Solvent was removed in vacuo to give an oil which on flash chromatography gave 351 mg (58%) of the bromide 10 at R_f = 0.40 (1:1 hexane/ether); ¹H NMR (400 MHz, CDCl₃) δ 7.39-7.24 (m, 5H), 4.11 (d, J = 8.4 Hz, 1H, H₅), 3.81 (s, 2H), 3.46 (dd, J = 6.9, 1.9 Hz, H₁), 2.88 (m, 4H, 2H₃, H₄, H_6anti), 1.76 (t, J = 8.1 Hz, H_6syn); ¹³C NMR (100 MHz, CDCl₃) δ 139.0, 128.5, 128.3, 127.0, 68.8, 59.0, 55.3, 54.6, 48.0, 35.8; HRMS m/z found 252.0383, calcd for C₁₂H₁₅NBr (M+H) 252.0383.

N-(Benzyloxycarbonyl)-5-anti-bromo-6-anti-fluoro-2-azabicyclo[2.1.1]hexane (15)

To a solution of 2-azabicyclo[2.2.0]hex-5-ene 12^7b (1.30 g, 0.006 mol) in MeNO₂ (50 mL) was added NBS (2.15 g, 0.012 mol) at 0 °C followed by Et₃N·3HF (2.92 g, 0.018 mol) dropwise over a period of 10 min.¹² The reaction was brought to room temperature and stirred for 20 h. Then the reaction mixture was diluted with CH₂Cl₂ (125 mL) and washed with saturated aqueous NaHCO₃ solution (50 mL). The organic layer was dried over anhydrous Na₂SO₄, concentrated in vacuo and the residue was purified by flash chromatography (1:3 ether/hexanes) to afford 997 mg (53%) of bromofluoride 15 as a colorless oil at R_f = 0.49 (1:1 ether/hexanes); ¹H NMR (400 MHz, CDCl₃) δ 7.41-7.29 (m, 5H), 5.18 (d, J = 12.3 Hz, 1H), 5.14 (d, J = 12.3 Hz, 1H), 4.99 (dd, J = 59.1, 7.3 Hz, 1H, H₅), 4.55 (br d, J = 7.2 Hz, 1H, H₆), 4.11 (dd, J = 7.3, 3.1 Hz, 1H, H₁), 3.61 (ddd, J = 9.1, 3.2, 1.2 Hz, 1H, H₃), 3.51 (dt, J = 9.1, 1.3 Hz, 1H, H_3′), 3.13 (brdd, J = 7.3, 3.6 Hz, 1H, H₄); ¹³C NMR (100 MHz, CDCl₃) δ 154.9, 136.0, 128.5, 128.3, 128.0, 99.7 (J_C,F = 226.8 Hz), 67.4, 64.6, 49.8, 49.1 (J_C,F = 17.5 Hz), 48.1; ¹⁹F NMR (282 MHz, CDCl₃) δ −204.75 (d, J = 58.9 Hz), −205.83 (d, J = 58.9 Hz); HRMS m/z found 336.0014, 338.0005, calcd for C₁₃H₁₃NO₂FBr^{79 and 81}Na (M + Na) 336.0011, 337.9991.

N-Benzyl-5-anti-bromo-6-anti-fluoro-2-azabicyclo[2.1.1]hexane (11)

To a solution of fluorobromide 15 (990 mg, 3.2 mmol) in methanol (25 mL) was added Pd(OH)₂ (99 mg). The solution was degassed in vacuo for 5 min and stirred at room temperature under a H₂ balloon for 1 h. The reaction mixture was then filtered through Celite, the solvent was removed in vacuo, the crude amine was dissolved in CH₃CN (17 mL) and then Et₃N (1.3 g, 4 mmol) followed by BnBr (807 mg, 1.5 mmol) were added dropwise. The solution was stirred at room temperature for 3 days followed by removal of solvent in vacuo to give the residue, which was chromatographed to give 494 mg (58%) of bromoamine 11 as a light orange oil at R_f = 0.79 (1:5 ether/hexanes); ¹H NMR (400 MHz, CDCl₃) δ 7.40-7.25 (m, 5H), 5.17 (dd, J = 60.6, 7.3 Hz, 1H, H₅), 4.39 (dd, J = 7.4, 3.2 Hz, 1H, H₆), 3.88 (d, J = 13.2, 1H), 3.77 (d, J = 13.2, 1H), 3.55 (dd, J = 6.8, 1.8 Hz, 1H, H₁), 3.07 (ddt, J = 6.8, 4.4 and 1.2 Hz, 1H, H₄), 3.03 (dt, J = 9.0 and 1.2 Hz, 1H, H₃), 2.74 (ddd, J = 9.0, 4.0, 1.2 Hz, 1H, H_3′); ¹³C NMR (100 MHz, CDCl₃) δ 137.9, 128.4 (2C overlap), 127.3, 100.1 (J_{C, F} = 222.3 Hz), 68.4 (J_{C, F} = 18.7 Hz, C₁), 58.5, 52.9 (J_{C, F} = 5.0 Hz), 51.4 (J_{C, F} = 17.1 Hz, C₄), 49.4 (J_{C, F} = 3.0 Hz); ¹⁹F NMR (282 MHz, CDCl₃) δ −208.68 (d, J = 60.6 Hz); HRMS m/z 270.0275, calcd for C₁₂H₁₃BrFN (M) 270.0288.

N-Benzyl-5-anti,6-anti-diacetoxy-2-azabicyclo[2.1.1]hexane (16a) and N-Benzyl-5-anti-acetoxy-6-anti-bromo-2-azabicyclo[2.1.1]hexane (17)

General Method A (DMF)

To a solution of dibromide 9 (200 mg, 0.604 mmol) in DMF (20 mL) under argon there was added cesium acetate (696 mg, 3.63 mmol). After stirring at 60 °C for 5 days, the reaction mixture was allowed to reach room temperature. Brine (10 mL) was added and the solvent was extracted with ether (3 × 20 mL). The combined ether extracts were washed with water (20 mL) and dried over Na₂SO_4. The dried ether was evaporated and the residue was chromatographed on silica gel (hexanes/ether 2:1) to give 70 mg (40%) of diacetate 16a as an orange color oil at R_f = 0.27 (1:1 hexanes/ether) as an oil, 78 mg (42%) of bromoacetate 17 as an orange color oil at R_f = 0.53 (1:1 hexanes/ether), and 19 mg (10%) of starting material. For diacetate 16a, ¹H NMR (400 MHz, CDCl₃) δ 7.43-7.23 (m, 5H, Ph), 4.99 (s, 2H, H₅ and H₆), 3.91 (s, 2H, CH₂Ph), 3.56 (d, J = 6.9 Hz, 1H, H₁), 3.04 (d, J = 6.9 Hz, 1H, H₄), 3.03 (s, 2H, 2H₃), 2.10 (s, 6H, 2COCH₃); ¹³C NMR (100 MHz, CDCl₃) d170.8, 138.2, 128.7, 128.4, 127.3, 81.7 (C₅ and C₆), 65.5, 58.6, 52.2, 47.7, 21.0; HRMS m/z 290.1399, calcd for C₁₆H₂₀NO₄ (M+H) 290.1392. For bromoacetate 17, ¹H NMR (400 MHz, CDCl₃) δ 7.41-7.27 (m, 5H, Ph), 4.91 (d, J = 7.0 Hz, 1H), 4.34 (d, J = 7.1 Hz, 1H), 3.94 (d, J = 13.1 Hz, 1H), 3.86 (d, J = 13.1 Hz, 1H), 3.62 (d, J = 6.7 Hz, 1H, H₁), 3.11 (d, J = 6.7 Hz, 1H, H₄), 3.04 (d, J = 8.8 Hz, 1H, H₃), 2.92 (br, 1H, H_3′), 2.13 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) d170.9, 137.7, 128.6, 128.5, 127.4, 82.3, 68.2, 58.4, 53.7, 50.3, 50.0, 21.2; HRMS m/z 310.0410, 312.0424, calcd for C₁₄H₁₇NO₂^{79, 81}Br (M+H) 310.0443, 312.0422.

N-Benzyl-5-anti,6-anti-dihydroxy-2-azabicyclo[2.1.1]hexane (16b)

To a solution of diacetate 16a (50 mg, 0.173 mmol) in methanol (3 mL) under argon there was added triethylamine (175 mg, 1.728 mmol). The solution was stirred at room temperature overnight and concentrated under reduced pressure. Purification of the obtained residue by flash chromatography (9.5:0.5 CH₂Cl₂/MeOH) afforded 27 mg (76%) of diol 16b as a light orange oil at R_f = 0.54 (CH₂Cl₂/MeOH 9:1); ¹H NMR (400 MHz, CDCl₃) δ 7.40-7.28 (m, 5H), 4.59 (s, 2H, H₅ and H₆), 3.89 (s, 2H), 3.57 (br, 2H, 2OH), 3.21 (d, J = 7.0 Hz, 1H, H₁), 2.97 (s, 2H, 2H₃), 2.64 (d, J = 6.9 Hz, 1H, H₄); ¹³C NMR (100 MHz, CDCl₃) δ 137.7, 128.8, 128.5 and 127.4, 81.9 (C₅ and C₆), 67.9, 58.9, 52.7, 50.4; HRMS m/z 206.1173, calcd for C₁₂H₁₆NO₂ (M+H) 206.1181.

N-Benzyl-5-anti,6-anti-difluoro-2-azabicyclo[2.1.1]hexane (18) and N-Benzyl-3-formylpyrrole (19). (From diol 16b)

To a solution of diol 16b (25 mg, 0.122 mmol) in CH₂Cl₂ (0.7 mL) under argon there was added BAST (81 mg, 0.365 mmol) dropwise at −78 °C. The resulting mixture was brought to room temperature and stirred overnight. The solution was diluted with CH₂Cl₂ (1.3 mL) washed with brine (0.5 mL) and water (0.5 mL) and then the CH₂Cl₂ layer was dried over Na₂SO₄. The organic layer was concentrated under reduced pressure and purification of obtained residue by preparative TLC (1:1 hexanes/ether) gave 6 mg (24%) of difluoro compound 18 at R_f = 0.37 (1:1 hexanes/ether) and 9 mg (40%) aldehyde 19 at R_f = 0.13 (1:1 hexanes/ether). For 18, ¹H NMR (400 MHz, CDCl₃) δ 7.35-7.25 (m, 5H), 5.34 (m, AA’XX’ pattern, 2H, H₅ and H₆), 3.82 (s, 2H), 3.45 (dt, J = 7.2, 1.9 Hz, 1H, H₁), 2.99 (m, 1H, H₄), 2.92 (s, 2H, 2H₃); ¹³C NMR (100 MHz, CDCl₃) δ 138.1, 128.4 (2C), 127.3, 100.6 and 98.3 (m, AA’XX’ pattern, 2C, C₅ and C₆), 65.5 (t, J = 18.2 Hz, C₁), 58.7, 50.5 (t, J = 7.3 Hz), 49.2 (t, J = 18.1 Hz, C₄); ¹⁹F NMR (282 MHz, CDCl₃) δ –217.3 (m). HRMS m/z 210.1089, calcd for C₁₂H₁₄NF₂ (M+H) 210.1094.

N-Benzyl-5-anti,6-anti-diazido-2-azabicyclo[2.1.1]hexane (20) and N-Benzyl-6-anti-azido-5-anti-bromo-2-azabicyclo[2.1.1]hexane (21). (DMF)

According to general method A, sodium azide (15 mg, 0.24 mmol) was added to a solution of crystalline dibromide 9 (15 mg, 0.045 mmol) in DMF (2.5 mL) under an air atmosphere. The mixture was allowed to stir at 60 °C for 8 d. Workup and chromatography of the residue using silica gel (5:1 hexane/ether) gave 9.3 mg of a mixture of dibromide 9 (2.6 mg), 2.8 mg (34%) of diazide 20 at R_f = 0.5 (1:1 hexane/ether), and 4.0 mg (49%) of bromoazide 21 at R_f = 0.6. In two trials with non-crystalline dibromide 9 a small amount (<6%) of aldehyde 19 was observed at R_f = 0.35. For diazide 20, ¹H NMR (CDCl₃, 400 Hz) δ 7.25 (m, 5H), 4.14 (s, 2H), 3.75 (s, 2H), 3.31(d, J = 7.0 Hz, 1H, H₁), 2.85 (s, 2H, 2H₃), 2.81(d, J = 7.0 Hz, 1H, H₄); ¹³C NMR (CDCl₃, 100Hz) δ 138.4, 128.9, 128.9, 127.8, 68.4, 67.3, 59.0, 53.9, 48.2; HRMS m/z calcd for C₁₂H₁₄N₇ (M+H) 256.1259, found 256.1263. For bromoazide 22, ¹H NMR (CDCl₃, 400 Hz) δ = 7.25 (m, 5H), 4.22 (d, J = 7.0 Hz, 1H), 4.18 (d, J = 7.0 Hz, 1H), 3.77 (two d, J = 13.2 Hz, 2H, CH₂Ph), 3.41 (d, J = 6.6 Hz, 1H, H₁), 2.93 (m, 2H, H₃+H₄), 2.72 (d, J = 9.1 Hz, 1H, H₃); ¹³C NMR (CDCl₃, 100 Hz) δ 138.3, 128.9, 128.9, 127.8, 69.8, 68.9, 59.0, 54.8, 50.5 (2C); HRMS m/z 294.0483, calcd for C₁₂H₁₄N₄Br (MBr⁷⁹ + H) 294.0481 and 296.0465, calcd for C₁₂H₁₄N₄Br (MBr⁸¹ + H) 296.0461.

N-Benzyl-5-anti-6-anti-di-(phenylthio)-2-azabicyclo[2.1.1]hexane (22) and N-Benzyl-5-anti-bromo-6-anti-(phenylthio)-2-azabicyclo[2.1.1]hexane (23)

According to general method B, to a solution of dibromide 9 (50 mg, 0.15 mmol) in dry DMSO (1 mL) there was added NaSPh (120 mg, 0.906 mmol) under argon and the reaction mixture was maintained at 60 °C for 5 h. The usual workup and chromatography (prep tlc, 1:2 ether/hexanes) gave di(phenylthio) ether 22 (22 mg, 37%) at R_f = 0.53 (1:2 ether/hexanes) and bromo(phenylthio) ether 23 (16 mg, 29%) at R_f = 0.59 (1:2 ether/hexanes) as light orange colored oils and the starting dibromide 9 (2 mg, 4%) at R_f 0.73 (1:2 ether/hexanes) as off-white solid. For di(phenylthio) ether 22, ¹H NMR (400 MHz, CDCl₃) δ 7.44-7.23 (m, 15H), 3.94 (s, 2H, H₅ and H_5′), 3.80 (s, 2H, CH₂Ph), 3.71 (d, J = 6.6 Hz, 1H, H₁), 3.16 (d, J = 6.6 Hz, 1H, H₄), 306 (s, 2H, H₃); ¹³C NMR (100 MHz, CDCl₃) δ 138.5 (br, 2C), 129.2, 129.0, 128.5, 128.4, 127.2, 126.1, 71.3, 58.9, 56.8, 56.6, 51.6; HRMS m/z 390.1361 calcd for C₂₄H₂₄NS₂ (M+H) 390.1345. For the bromo(phenylthio) ether 23, ¹H NMR (400 MHz, CDCl₃) δ 7.37-7.18 (m, 10H), 4.27 (d, J = 7.3 Hz, 1H, H₆), 3.88 (d, J = 13.3 Hz, 1H, CH₂Ph), 3.83 (d, J = 13.3 Hz, 1H, CH₂Ph), 3.74 (d, J = 7.3 Hz, 1H, H₅), 3.64 (d, J = 6.5 Hz, 1H, H₁), 3.13 (d, J = 6.5 Hz, 1H, H₄), 2.97 (d, J = 8.9 Hz, 1H, H₃), 2.94 (d, J = 8.9 Hz, 1H, H₃); ¹³C NMR (100 MHz, CDCl₃) δ 139.1, 138.0, 129.3, 129.0, 128.5, 128.5, 127.4, 126.2, 71.2, 58.8, 58.3, 55.9, 52.7, 52.0; HRMS m/z 360.0433 calcd for C₁₈H₁₉BrNS (M+H) 360.0416.

N-Benzyl-5-anti-acetoxy-2-azabicyclo[2.1.1]hexane (24a). (DMF)

Following general method A, to a solution of bromide 10 (14 mg, 0.06 mmol) in DMF (8 mL) under argon there was added (72 mg, 0.5 mmol) of AgOAc. The resulting solution was heated for 12 h at 70 °C. Workup and chromatography gave 7 mg (54 %) of acetate 24a at R_f = 0.28 (1:1 hexane/ethyl acetate); ¹H NMR (300 MHz, CDCl₃) δ 7.24 (m, 5H), 4.73 (d, J = 7.5 Hz, H₅), 3.82 (d, J = 13.1 Hz, 1H, CH₂), 3.74 (d, J = 13.1 Hz, 1H, CH₂), 3.39 (dd, J = 6.6, 1.5 Hz, H₁), 2.88 (d, J = 8.7 Hz, H₃), 2.74 (m, 2H, H₄ and H₃), 2.42 (d, J = 7.8 Hz, H_6anti), 2.03 (s, 3H), 1.73 (dd, J = 7.8, 7.5 Hz, H_6syn); ¹³C NMR (100 MHz, CDCl₃) δ 171.1, 130.0, 128.8, 128.4, 127.2, 81.4, 65.5, 58.5, 54.0, 44.4, 32.3, 21.0; HRMS m/z found 232.1315, calcd for C₁₄H₁₈NO₂ (M+H) 232.1315.

N-Benzyl-5-anti-hydroxy-2-azabicyclo[2.1.1]hexane (24b)

To a solution of acetate 24a (3 mg, 0.002 mmol) in methanol (3 mL) was added K₂CO₃ (138 mg, 0.01 mmol) and the solution was stirred at RT for 1 h. After 1 h the base was filtered and the solvent was removed in vacuo to give 2.1 mg (84%) of the alcohol 24b at R_f = 0.20 (1:2 hexane/ethyl acetate); ¹H NMR (300 MHz, CDCl₃) δ 7.36 (m, 5H), 4.18 (brd, J = 7.2 Hz, H₅), 3.78 (d, J = 12.9 Hz, 1H, CH₂), 3.69 (d, J = 12.9 Hz, 1H, CH₂), 3.19 (d, J = 6.8, 1.8 Hz, H₁), 2.96 (d, J = 8.8 Hz, H₃), 2.64 (brd, J = 8.1 Hz, H_6anti), 2.54 (dd, J = 6.8, 3.0 Hz, 1H, H₄), 2.48 (d, J = 8.8 Hz, H₃), 1.74 (dd, J = 8.1, 7.2 Hz, H_6syn); ¹³C NMR (100 MHz, CDCl₃) δ 128.7, 128.3, 126.9, 80.6, 67.3, 59.1, 55.0, 46.1, 32.3; HRMS m/z 190.1212, calcd for C₁₂H₁₅NO (M +H ) 190.1227.

N-(tert-Butoxycarbonyl)-5-anti-hydroxy-2-azabicyclo[2.1.1]hexane (25)

To a solution of alcohol 24b (25 mg, 0.13 mmol) in MeOH (8 mL) there was added Pd(OH)₂ (5 mg) and (BOC)₂O (54 mg, 0.25 mmol). The solution was stirred under 1 atmosphere of hydrogen for 7 h at RT. Afterward, the solution was diluted with 10 mL of MeOH and filtered through Celite. Evaporation of the solvent followed by column chromatography gave 24 mg (92%) of the pure alcohol 25 at R_f = 0.31 (2:1 hexane/ether); ¹H NMR (300 MHz, CDCl₃) δ 4.17 (m, 2H, H₁ and H₅), 3.32 (s, 2H, 2H₃), 2.93 (dm, J = 7.5 Hz, 1H, H₄), 2.70 (m, 1H, H_6anti), 1.81 (br, OH), 1.61 (t, J = 7.5 Hz, 1H, H_6syn), 1.45 (s, 9H); ¹³C NMR (100 MHz, CDCl₃) δ 155.7, 81.4, 79.5, 63.4, 48.3, 44.1, 36.9, 28.5; HRMS m/z found 222.1104, calcd for C₁₀H₁₇NO₃Na (M + Na) 222.1104.

N-Benzyl-5-anti-fluoro-2-azabicyclo[2.1.1]hexane (26). (From bromide 10 using AgF)

To a solution of bromide 10 (33 mg, 0.13 mmol) in CH₃NO₂ (8 mL) at RT under argon there was added AgF (83 mg, 0.65 mmol) and the reaction mixture was heated at 50 °C for 12 h. The AgF was filtered via Celite and the solvent was removed in vacuo to give 20 mg (80%) of pure fluoride 26 at R_f = 0.20 (1:2 hexane/ethyl acetate); ¹H NMR (300 MHz, CDCl₃) δ 7.36 (m, 5H), 4.90 (dd, J = 64, 7.2 Hz, H₅), 3.74 (d, J = 13.8 Hz, 1H, CH₂), 3.65 (d, J = 13.8 Hz, 1H, CH₂), 3.35 (brd, J = 6.4 Hz, H₃ ), 2.88 (b, J = 8.8 Hz, H₃), 2.72 (br, H₄), 2.51 (m, 2H, H₃ and H_6anti), 1.90 (ddd, J = 8.0, 7.2, 2.4 Hz, H_6syn); ¹³C NMR (75 MHz, CDCl₃) δ 138.6, 128.7, 128.4, 127.2, 98.8 (d, J = 208 Hz), 65.8 and 65.6, 58.9, 53.5, 45.3, 32.1; ¹⁹F NMR (282 MHz, CDCl₃) δ –220.1 (d, J = 64.7 Hz); HRMS m/z found 192.1191, calcd for C₁₂H₁₄NF (M + H) 192.1188.

N-Benzyl-5-anti-azido-2-azabicyclo[2.1.1]hexane (27)

Method A. DMF

To a solution of bromide 10 (65 mg, 0.24 mmol) in DMF (10 mL) under argon there was added sodium azide (84 mg, 1.3 mmol). The resulting solution was heated for 12 h at 70 °C. The usual workup and flash chromatography gave 28 mg (51%) of the azide 27 at R_f = 0.39 (1:1 hexane/ether); ¹H NMR (300 MHz, CDCl₃) δ 7.38 (m, 5H), 3.86 (br, J = 7.5 Hz, H₅), 3.72 (d, J = 13.2 Hz, 1H), 3.65 (d, J = 13.2 Hz, 1H), 3.32 (dd, J = 6.3, 2.0 Hz, H₁), 2.86 (brd, J = 9.3 Hz, H₃), 2.67 (m, H₄), 2.61 (d, J = 9.3 Hz, H₃), 2.38 (brd, J = 7.8 Hz, H_6anti), 1.69 (dd, J = 7.8, 7.5 Hz, H_6syn); ¹³C NMR (100 MHz, CDCl₃) δ 129.0, 128.8, 127.5, 69.5, 67.0, 59.2, 54.9, 45.2 and 33.8; HRMS m/z found 215.1291, calcd for C₁₂H₁₅N₄ (M + H) 215.1282; m/z found 256.1551, calcd for C₁₄H₁₈N₅ (M + CH₃CN + H) 256.1556.

N-Benzyl-5-anti-(tert-butoxycarbonylamino)-2-azabicyclo[2.1.1]hexane (28)

To a solution of azide 27 (29 mg, 0.13 mmol) in toluene (8 mL) there was added PPh₃ (71 mg, 0.27 mmol) and water (1 mL) and the resultant solution was heated at 60 °C for 4 h. After allowing cooling, the two layers were separated and the water layer was extracted with CH₂Cl₂ (3 × 5 mL). All the organic layers were combined and dried over Na₂SO₄. Solvent was removed in vacuo and the amine was dissolved in MeOH to which there was added Et₃N (18 μL, 0.26 mmol) and (BOC)₂O (28 mg, 0.13 mmol). Removal of the solvent followed by flash chromatography gave 34 mg (87%) of the BOC protected amine 28 at R_f = 0.37 (1:3 hexane/ethyl acetate); ¹H NMR (300 MHz, CDCl₃) δ 7.44 (m, 5H), 4.92 (br, 1H, NH), 3.82 (s, 2H), 3.76 (br, H₅), 3.40 (br, H₁), 2.86 (d, J = 8.1 Hz, H₃), 2.72 (br, 2H, H₃ and H₄), 2.32 (brd, J = 7.8 Hz, H_6anti), 1.74 (dd, J = 7.8, 8.1 Hz, H_6syn), 1.35 (s, 9H); ¹³C NMR (100 MHz, CDCl₃) δ 156.4, 139.3, 128.9 and 128.8, 127.5, 119.5, 79.9, 66.5, 59.0, 55.2, 44.7, 34.3, 34.2, 28.8; HRMS m/z found 289.1917, calcd for C₁₇H₂₅N₂O₂ (M + H) 289.1911; m/z found 311.1749, calcd for C₁₇H₂₄N₂O₂Na (M + Na) 311.1730.

N-Benzyl-5-anti-imidazol-1-yl-2-azabicyclo[2.1.1]hexane (29)

Butyllithium (90 μL, 2.5 M solution in hexanes, 0.226 mmol) was added dropwise to imidazole (15 mg, 0.226 mmol) in anhydrous DMF (0.5 mL) under argon, and the mixture was stirred at 20 °C for 0.25 h. A solution of bromide 10 (19 mg, 0.075 mmol) in anhydrous DMF (0.5 mL) was added, and after stirring at 70 °C for 8 days workup and chromatography (CH₂Cl₂/MeOH/NH₄OH 90: 10: 1) the imidazolyl compound 29 was isolated as a light orange colored oil (10 mg, 55%) at R_f = 0.60 (CH₂Cl₂/MeOH/NH₄OH, 90:10:1); ¹HNMR (400 MHz, CDCl₃) δ 7.54 (s, 1H), 7.43-7.25 (m, 5H), 7.09 (s, 1H), 6.94 (s, 1H), 4.30 (d, J = 7.3 Hz, 1H, H₅), 3.91 (d, J = 13.2 Hz, 1H, Bn), 3.83 (d, J = 13.2 Hz, 1H, Bn), 3.70 (dbr, J = 6.8 Hz, 1H, H₁), 3.14-3.07 (m, 2H, H₃), 2.79 (d, J = 8.7 Hz, 1H, H_6anti), 2.13 (m, 1H, H₄), 1.88 (two d, J = 8.5, 8.5 Hz, 1H, H_6syn); ¹³CNMR (100 MHz, CDCl₃) δ 138.7, 136.8, 129.4, 128.6, 128.5, 127.3, 119.0, 66.1, 64.4, 58.9, 54.6, 44.1, 33.4; HRMS m/z 240.1495, calcd for C₁₅H₁₈N₃ (M + H) 240.1495.

N-Benzyl-5-anti-phenylthio-2-azabicyclo[2.1.1]hexane (30)

Following method B, to a solution of monobromide 10 (22 mg, 0.087 mmol) in dry DMSO (0.6 mL) there was added NaSPh (35 mg, 0.262 mmol) and the reaction mixture was maintained at 60 °C for 5 h under argon. Workup and chromatography (prep tlc, 1:1 ether/hexanes) afforded phenylthio ether 30 (19 mg, 77%) at R_f = 0.42 (1:1 ether/hexanes) as a light orange colored oil; ¹H NMR (400 MHz, CDCl₃) δ 7.50-7.23 (m, 10H), 3.96 (d, J = 13.3 Hz, 1H, CH₂Ph), 3.91 (d, J = 13.3 Hz, 1H, CH₂Ph), 3.72 (d, J = 8.0 Hz, 1H, H₅), 3.54 (dbr, J = 6.6, 1.8 Hz, 1H, H₁), 2.99 (dd, J = 8.7, 1.0 Hz, 1H, H₃), 2.93 (dd, J = 8.7, 1.0 Hz, 1H, H₃), 2.88 (m, 1H, H_6syn), 2.84 (m, 1H, H₄), 1.81 (t, J = 8.0 Hz, 1H, H_6yn); ¹³C NMR (100 MHz, CDCl₃) δ 138.9, 136.9, 128.9, 128.7, 128.6, 128.4, 127.1, 125.8, 67.3, 58.8, 55.8, 54.4, 45.0, 35.7; HRMS m/z 282.1321 calcd for C₁₈H₂₀NS (M+H) 282.1311.

N-Benzyl-5-anti-iodo-2-azabicyclo[2.1.1]hexane (31)

A solution of NaI (190 mg, 1.269 mmol) in acetone (750 μL) was added to bromide 10 (16 mg, 0.063 mmol) under argon. The reaction mixture was maintained at reflux for 4 days. The solvent was removed in vacuo and the crude was dissolved in CH₂Cl₂ (4 mL) and washed with water (2 mL). The organic layer was separated and the aqueous layer was washed with CH₂Cl₂ (2 × 2 mL). The organic extracts were combined and dried over Na₂SO_4. The solvent was removed in vacuo and the crude was chromatographed (prep tlc, 1:1 ether/hexanes) to give iodide 31 (14 mg, 74%) at R_f = 0.74 (1:1 ether/hexanes) as a light orange colored oil; ¹H NMR (400 MHz, CDCl₃) δ 7.38-7.24 (m, 5H), 3.95 (d, J = 9.0 Hz, 1H, H₅), 3.80 (two d, J = 13.3, 13.3 Hz, 2H, Bn), 3.46 (dd, J = 6.5, 1.8 Hz, 1H, H₁), 2.86-2.77 (m, 4H, 2H₃, H₄ and H_6anti), 1.75 (dd, J = 9.0, 8.0 Hz, 1H, H_6syn); ¹³CNMR (100 MHz, CDCl₃) δ139.0, 128.5, 128.4, 127.1, 69.5, 59.1, 54.1, 48.5, 37.9, 30.1; HRMS m/z 300.0243, calcd for C₁₂H₁₅IN (M + H) 300.0244.

N-Benzyl-6-anti-acetoxy-5-anti-fluoro-2-azabicyclo[2.1.1]hexane (32a)

Method A (DMF)

To a solution of bromofluoride 11 (900 mg, 3.33 mmol) in DMF (55 mL) under argon there was added cesium acetate (1279 mg, 6.66 mmol). The solution was maintained at 70 °C for 5 days. The usual workup and flash chromatography (1:3 ether/hexanes) afforded 578 mg (64%) of unreacted fluorobromide 11 at R_f = 0.61 (1:1 ether/hexane) and 249 mg (30%) (84% BORSM) of fluoroacetate 32a at R_f = 0.44 (1:1 ether/hexanes); ¹H NMR (400 MHz, CDCl₃) δ 7.38-7.22 (m, 5H), 5.29-5.08 (m, ABX pattern, 2H, H₅ and H₆)(See supplemental), 3.88 (d, J = 13.2, Hz, 1H), 3.82 (d, J = 13.2, Hz, 1H), 3.48 (dd, J = 7.1, 2.2 Hz, 1H, H₁), 3.05-2.95 (m, 2H, H₃ + H₄), 2.88 (ddd, J = 9.0, 3.7, 1.2 Hz, 1H, H₃), 2.12 (s, 3H); ¹H NMR (400 MHz, CDCl₃+C₆D₆ 1:1 mixture) δ 7.41-7.22 (m, 5H), 5.26-4.99 (m, ABX pattern, 2H, H₅ and H₆), 3.73 (d, J = 13.2, Hz, 1H), 3.66 (d, J = 13.2, Hz, 1H), 3.48 (dd, J = 7.1, 2.2 Hz, 1H, H₁), 2.86 (t, J = 6.0 Hz, 1H, H₄), 2.79-2.62 (m, 2H, 2H₃), 2.01 (s, 3H); ¹H NMR (400 MHz, C₆D₆) δ 7.44-7.22 (m, 5H), 5.21-4.96 (m, ABX pattern, 2H, H₅ and H₆), 3.61 (d, J = 13.2, Hz, 1H), 3.55 (d, J = 13.2, Hz, 1H), 3.51 (dd, J = 7.1, 2.1 Hz, 1H, H₁), 2.78 (m, 1H, H₄), 2.57 (br d, J = 9.0 Hz, 1H, H₃ ), 2.50 (ddd, J = 9.0, 3.6, 1.2 Hz, 1H, H₃), 1.92 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) δ 171.2, 138.2, 128.5, 128.4 and 127.2, 99.5 (J_{C, F} = 219.0 Hz, C₅), 81.3 (J_{C, F} = 3.6 Hz), 65.5 (J_{C, F} = 18.4 Hz, C₁), 58.7, 51.4 (J_{C, F} = 6.7 Hz), 48.4 (J_{C, F} = 17.7 Hz, C₄), 21.0; ¹⁹F NMR (282 MHz, CDCl₃) δ −214.63 (d, J = 60.9 Hz); HRMS m/z 250.1224, calcd for C₁₄H₁₇FNO₂ (M + H) 250.1238.

N-Benzyl-5-anti-fluoro-6-anti-hydroxy-2-azabicyclo[2.1.1]hexane (32b)

To a solution of fluoroacetate 32a (575 mg, 2.306 mmol) in methanol (35 mL) under argon there was added Et₃N (3212 μL, 23.066 mmol). The solution was stirred at room temperature for 20 h and concentrated under reduced pressure. Purification of the obtained residue by flash chromatography (0.5:9.5 MeOH/CH₂Cl₂) afforded 459 mg (96%) of fluoroalcohol 32b at R_f = 0.62 (1:9 MeOH/CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.40-7.19 (m, 5H), 5.41 (dd, J = 61.8, 8.0 Hz, 1H, H₅), 4.56 (d, J = 8.0 Hz, 1H, H₆), 3.83 (s, 2H), 3.33 (dd, J = 7.2, 2.1 Hz, 1H, H₁ and br, 1H, OH), 2.93 (s, 2H, 2H₃), 2.82 (brdd, J = 7.2, 5.2, Hz, 1H, H₄); ¹³C NMR (100 MHz, CDCl₃) δ 137.8, 128.6, 128.5 and 127.4, 102.2 (J_{C, F} = 208.7 Hz, C₅), 82.4, 66.9 (J_{C, F} = 16.3 Hz, C₁), 58.8, 51.6 (J_{C, F} = 7.8 Hz), 50.2 (J_{C, F} = 16.3 Hz, C₄); ¹⁹F NMR (282 MHz, CDCl₃) δ -213.63 (dd, J = 62.4, 3.9 Hz); the extra 3.9 Hz may be due to H-bonding. Calculated couplings for the related N-methyl fluoroalcohol are 62.53 and 11.8 Hz; HRMS m/z 208.1109, calcd for C₁₂H₁₅FNO (M + H) 208.1132.

N-(t-Butoxycarbonyl)-5-anti-fluoro-6-anti-hydroxy-2-azabicyclo[2.1.1]hexane (33)

To a solution of fluoroalcohol 32b (250 mg, 1.206 mmol) in MeOH (10 mL) there was added palladium hydroxide (20 wt % Pd on carbon) (38 mg) and (Boc)₂O (316 mg, 1.447 mmol). The resulting solution was stirred at RT under hydrogen for 6 h. Then the solution was filtered through Celite and washed with MeOH (10 mL). The filtrate was evaporated to give an oily solid, n-heptane (20 mL) was added to the residue, and solvent was again evaporated. Then n-heptane (30 mL) was added to the residue, and after stirring at room temperature for 2 h, the separated solid was filtered and dried under reduced pressure to afford 237 mg (91%) of fluoroalcohol 33 as an off-white solid at R_f = 0.71 (1:9 MeOH/CH₂Cl₂); mp 95–97 °C; ¹H NMR (400 MHz, CDCl₃) δ 5.10 (dd, J = 60.8, 7.8 Hz, 1H, H₅), 4.28 (d, J = 7.7 Hz, 1H, H₆), 4.22 (br s, 1H, H₁), 3.47 (d, J = 9.1 Hz, 1H, H₃), 3.40 (d, J = 9.0 Hz, 1H, H_3′), 3.08 (br s, 1H, OH), 2.83 (br t, J = 6.1 Hz, 1H, H₄), 1.43 (s, 9H); ¹³C NMR (100 MHz, CDCl₃) δ 154.9, 101.9 (J_{C, F} = 214.2 Hz, C₅), 84.1, 80.4, 63.8 (br), 48.1 (J_{C, F} = 16.3 Hz), 46.0 (br), 28.3; ¹⁹F NMR (282 MHz, CDCl₃) δ −209.3 (d, J = 57.6 Hz), −210.3 (d, J = 57.6 Hz) (No F-HO splitting was observed.); HRMS m/z 240.1018, calcd for C₁₀H₁₆FNO₃Na(M + Na) 240.1012.

N-Benzyl-6-anti-azido-5-anti-fluoro-2-azabicyclo[2.1.1]hexane (34)

Method A (DMF)

Sodium azide (144 mg, 2.22 mmol) and tetrabutylammonium chloride (30 mg) were added to a solution of fluorobromide 11 (200 mg, 0.740 mmol) in dry DMF (15 mL) under argon. The reaction mixture was maintained at 70 °C for 5 days. Workup and flash chromatography (1:4 ether/hexanes) afforded 74 mg (43%) (62% BORSM) of fluoroazide 34 as an oil at R_f = 0.59 (1:1 ether/hexanes) and 62 mg (31%) of starting material 11 at R_f = 0.69; after two column separations, for 34; ¹H NMR (400 MHz, CDCl₃) δ 7.35-7.27 (m, 5H), 5.21 (dd, J = 60.9, 7.1 Hz, 1H, H₅), 4.31 (dd, J = 7.1, 2.9 Hz, 1H, H₆), 3.85 (d, J = 13.1, 1H), 3.79 (d, J = 13.2, 1H), 3.43 (dd, J = 7.1, 2.0 Hz, 1H, H₁), 3.00 (dt, J = 9.0, 1.3 Hz, 1H, H₃), 2.96 (ddt, J = 7.1, 4.7, 1.2 Hz, 1H, H₄), 2.85 (ddd, J = 9.1, 3.7, 1.2 Hz, 1H, H_3′); ¹³C NMR (100 MHz, CDCl₃) δ 138.0, 128.5, 128.4 and 127.4, 99.6 (J_{C, F} = 220.2 Hz, C₅), 67.7 (J_{C, F} = 4.2 Hz), 66.3 (J_{C, F} = 18.1 Hz), 58.7, 52.1 (J_{C, F} = 7.1 Hz), 48.4 (J_{C, F} = 17.3 Hz); ¹⁹F NMR (282 MHz, CDCl₃) δ −216.11 (d, J = 60.4 Hz); HRMS m/z 233.1202, calcd for C₁₂H₁₄FN₄ (M + H) 233.1202.

N-Benzyl-6-anti-amino-5-anti-fluoro-2-azabicyclo[2.1.1]hexane (35)

To a solution of fluoroazide 34 (70 mg, 0.301 mmol) in toluene (20 mL) and water (2.5 mL) there was added triphenylphosphine (166 mg, 0.633 mmol). The reaction mixture was heated to 60 °C for 5 h. After cooling to room temperature the organic layer was separated and the aqueous layer was extracted with methylene chloride (2 × 5 mL). The combined organic layers were dried over Na₂SO₄. Filtration, removal of solvent, and purification by flash chromatography (1–10 % methanol in methylene chloride) afforded 53 mg (85%) of fluoroamine 35 at R_f = 0.40 (1:9 MeOH/CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.40-7.23 (m, 5H), 5.27 (dd, J = 62.3, 7.4 Hz, 1H, H₅), 3.83 (d, J = 13.3, 1H), 3.78 (d, J = 13.3, 1H), 3.73 (dd, J = 7.5, 1.3 Hz, 1H, H₆), 3.20 (dd, J = 7.0, 2.1 Hz, 1H, H₁), 2.89-2.83 (m, 2H, H₃), 2.68 (ddt, J = 7.1, 5.2, 1.2 Hz, 1H, H₄), 2.20 (br s, 2H, NH₂); ¹³C NMR (100 MHz, CDCl₃) δ 138.6, 128.4, 128.2 and 127.0, 101.9 (J_C,F = 212.1 Hz, C₅), 67.7 (J_C,F = 16.2 Hz), 63.8 (J_C,F = 2.1 Hz), 58.8, 52.9 (J_C,F = 8.2 Hz), 50.3 (J_C,F = 16.3 Hz); ¹⁹F NMR (282 MHz, CDCl₃) δ −213.55 (brd, J = 63.1 Hz); HRMS m/z 207.1301, calcd for C₁₂H₁₆FN₂ (M + H) 207.1298.

N-Benzyl-6-anti-acetamido-5-anti-fluoro-2-azabicyclo[2.1.1]hexane (36)

DMAP (44 mg, 0.3636 mmol) was added to the solution of fluoroamine 35 (25 mg, 0.1212 mmol) in dry methylene chloride (3 mL) under argon. The resulting solution was cooled to 0 °C and acetyl chloride (26 μL, 0.3636) was added dropwise. The reaction mixture was allowed to room temperature and stirred for 3 h. The reaction mixture was then diluted with CH₂Cl₂ (7 mL), washed with water (3 × 5 mL) and dried over Na₂SO₄. Filtration, removal of solvent and purification by preparative thin layer chromatography (1:9 MeOH/CH₂Cl₂) afforded 19 mg (63%) of fluoroacetamide 36 at R_f = 0.54 (1:9 MeOH/CH₂Cl₂); ¹H NMR (400 MHz, CDCl₃) δ 7.41-7.22 (m, 5H), 6.54 (br s, 1H, NH), 5.34 (dd, J = 62.5, 7.4 Hz, 1H, H₅), 4.79 (ddd, J = 9.3, 7.4, 1.8 Hz, 1H, H₆), 3.91 (d, J = 13.2, 1H), 3.84 (d, J = 13.2, 1H), 3.33 (dd, J = 7.1, 2.5 Hz, 1H, H₁), 3.07 (br d, J = 9.1, Hz, 1H, H₃), 2.85 (br dd, J = 9.1, 4.1 Hz, 1H, H_3′), 2.78 (ddt, J = 7.1, 5.7, 1.2 Hz, 1H, H₄), 2.02 (s, 3H); ¹³C NMR (100 MHz, CDCl₃) δ 170.2, 138.1, 128.6, 128.4 and 127.3, 101.8 (J_{C, F} = 209.0 Hz, C₅), 67.0 (J_{C, F} = 18.6 Hz), 58.8, 58. 6 (J_{C, F} = 3.2 Hz), 52.5 (J_{C, F} = 7.1 Hz), 49.2 (J_{C, F} = 16.4 Hz), 23.7; ¹⁹F NMR (282 MHz, CDCl₃) δ −211.80 (brd, J = 63.0 Hz); HRMS m/z 249.1414, calcd for C₁₄H₁₈FN₂O(M + H) 249.1403.

N-Benzyl-5-anti-fluoro-6-anti-phenylthio-2-azabicyclo[2.1.1]hexane (37)

According to method B, to a solution of bromofluoride 11 (26 mg, 0.096 mmol) in dry DMSO (0.6 mL) there was added NaSPh (38 mg, 0.289 mmol) and the reaction mixture was maintained at 60 °C for 9 days under argon. The usual workup and chromatography (prep tlc, 1:3 ether/hexanes) afforded fluoro(phenylthio) ether 37 (20 mg, 69%) at R_f = 0.26 (1:3 ether/hexanes) as a light orange colored oil; ¹H NMR (400 MHz, CDCl₃) δ 7.36-7.18 (br, 10H), 5.13 (dd, J = 61.8, 6.7 Hz, 1H, H₅), 3.88 (d, J = 13.2 Hz, 1H), 3.88-3.84 (m, 1H, H₆), 3.81 (d, J = 13.2 Hz, 1H), 3.54 (dd, J = 6.8, 2.1 Hz, 1H, H₁), 3.11 (dt, J = 9.0, 1.2 Hz, 1H, H₃), 3.01 (m, 1H, H₄), 2.78 (ddd, J = 9.0, 3.9, 1.2 Hz, 1H, H_3′); ¹³C NMR (100 MHz, CDCl₃) δ 138.2, 129.1, 129.0, 128.5, 128.4, 127.3, 126.1, 99.9 (d, J = 216.4 Hz), 68.5 (d, J = 18.7 Hz), 59.9, 54.6 (d, J = 4.6 Hz), 53.6 (d, J = 6.4 Hz), 50.3 (d, J = 17.0 Hz); ¹⁹F NMR (282 Hz, CDCl₃) δ −210.2 (d, J = 62.4 Hz); HRMS m/z 300.1226 calcd for C₁₈H₁₉FNS (M+H) 300.1217.