Fluoro-Julia Olefination as a Mild, High-Yielding Route to α-Fluoro Acrylonitriles

Abstract

Synthesis of a novel, stable reagent (1,3-benzothiazol-2-ylsulfonyl)fluoroacetonitrile from readily available ethyl α-(1,3-benzothiazol-2-ylsulfanyl)- α-fluoroacetate is reported. Aldehydes undergo condensations with (1,3-benzothiazol-2-ylsulfonyl)fluoroacetonitrile in the presence of DBU leading to α-fluoro acrylonitriles in high yields and with good Z-stereoselectivity. Lowering of reaction temperature increases the Z selectivity.

INTRODUCTION

Interest in fluorinated organic molecules stems from the altered biological and other properties excerted by fluorine atom introduction.¹ As a consequence, a significant number of pharmaceuticals currently in the market contain fluorine atom.² A convenient approach towards the synthesis of fluoroorganic compounds is through modular assembly via fluorinated building blocks.³ Functionalized fluoroolefins are attractive building blocks and in many instances they are biologically useful entities.⁴ For example, fluoroolefins act as peptide isosteres,^1,5 whereas examples of enzyme inhibitors^1,4 include fluorovinyl nucleoside,^6a amino acid⁴ or steroid^6b,cbased derivatives. Several methods have therefore been developed for the synthesis of variously functionalized vinyl fluorides, such as α-fluoro acrylates⁷ or fluorovinyl sulfones.^6a,8 Synthetic routes to α-fluoro acrylonitriles, on the other hand, are scarce. Patrick and Nadji⁹ reported in situ preparation of diethyl α-cyano-α-fluoromethanephosphonate from fluoroacetonitrile that reacted as expected with aromatic aldehydes. On the other hand, aliphatic aldehydes were reported to give complex reaction mixtures resulting from aldol type reactions.⁹ Xu and DesMarteau¹⁰ prepared and isolated the Horner-Wadsworth-Emmons (HWE) reagent by fluorination of diethyl cyanomethanephosphonate using N-fluorobis(trifluoromethanesulfonyl)imide, a fluorinating agent that is not available commercially. Although the HWE reagent prepared in this manner was isolated, its use in condensation reactions with aldehydes and ketones was successful only when prepared in situ. Van der Gen et al.¹¹ prepared (diphenylphosphinoyl)fluoroacetonitrile in situ from the fluoroacetonitrile anion, but attempts at its isolation were unsuccessful.¹¹ Condensations proceeded with aldehydes as well as ketones in reasonable yields. Common to these methods is the necessity for in situ prepared reagent. Herein, we report preparation of a novel, stable reagent and its reactivity in α-fluoro acrylonitrile synthesis.

Modified Julia olefination¹² has been reported as an alternative to Horner-Wadsworth-Emmons olefination for the preparation of fluoroalkylidenes.^13,14 In this context, we have recently developed novel reagents for the synthesis of fluorinated stilbene and styrene derivatives,¹⁴ α-fluoro-α,β-unsaturated esters^7j and phenyl sulfones⁸ⁱ via Julia olefination. Specifically, mild conditions, combined with high yields obtained in the synthesis of the latter two prompted our interest in exploring Julia fluoroolefination for the preparation of α-fluoro acrylonitriles.

RESULTS AND DISCUSSION

Our initial goal was synthesis of an appropriate heteroarylsulfonyl derivative as the reagent for a one-pot Julia-Kocienski olefination.¹⁵ We first focussed on the synthesis of (1,3-benzothiazol-2-ylsulfonyl)fluoroacetonitrile. Reaction of bromoacetonitrile (1) with the sodium salt of 2-mercapto-1,3-benzothiazole resulted in 2a¹⁶ in 88% yield, that was subjected to oxidation to 1,3-benzothiazol-2-ylsulfonyl (BT-sulfonyl) derivative 3 (Scheme 1). The use of m-CPBA gave a low yield of 3, whereas dimethyldioxirane or oxone led mainly to the sulfoxide derivative. Use of MoO₂₄(NH₄)₆.4H₂O/H₂O₂¹⁷ gave 75% of 3, that increased to 90% upon use of H₅IO₆/CrO₃.¹⁸ Fluorination of 3 with NaH and Selectfluor in THF resulted in a mixture of monofluoro derivative 4 and a difluoro amide byproduct 5 (2.2:1 ratio, respectively), along with starting material. Laborius chromatographic purification, that prevented synthesis on larger scales, gave (1,3-benzothiazol-2-ylsulfonyl)fluoroacetonitrile 4 in 32% isolated yield. In order to improve the efficiency of synthesis of 4, alternate synthetic routes were tested. Metalation and fluorination of benzylic nitriles using various strong bases (LDA, NaHMDS, KHMDS, LHMDS, n-BuLi) and NFSI gave fluorinated derivatives in 9-16%, and the reaction of the bases with the cyano functionality was suggested as the possible cause for the low yields.¹⁹ However, the use of t-BuLi and NFSI improved the yield of monofluoro derivative to 60%.¹⁹ Fluorination of 1,3-benzothiazol-2ylsulfanyl derivative 2a using tert-BuLi and NFSI was therefore performed, resulting in a complex reaction mixture. Since metalated 1-phenyl-1H-tetrazol-5-yl sulfone derivatives (PT-sulfones) were suggested to be more stabile and less prone to self condensation,^15,20 PT-sulfide 2b was synthesized²¹ and subjected to tert-BuLi and NFSI fluorination to give 6b²¹ in 45% isolated yield (Scheme 1). Attempts at oxidizing 6b to corresponding sulfone using MoO₂₄(NH₄)₆.4H₂O/H₂O₂, or H₅IO₆/CrO₃, were unsuccessful.

SCHEME 1.

Attempts at an Efficient Synthesis of 4

Since little success was achieved with conversions of BT- or PT-derivatives containing a preinstalled cyano functionality, we decided on an alternate method, wherein fluorine introduction into the molecule would precede that of the cyano group. Olah and coworkers have reported high yielding conversions of amides to cyano derivatives using cyanuric chloride.²² Ethyl α(1,3-benzothiazol-2-ylsulfanyl)-α-fluoroacetate (8, 88% yield, Scheme 2) was therefore synthesized from commercially available ethyl bromofluoroacetate 7.^7k Alternatively, unfluorinated precursor ethyl bromoacetate 9 (Scheme 2) was converted to ethyl (1,3-benzothiazol-2-ylsulfanyl)acetate 10²³ (89% yield), followed by fluorination of 10 using LDA and NFSI in toluene to afford 8 in 57% yield.²⁴ Ester conversion to amide 11 with NH₃ in EtOH proceeded in high 96% isolated yield. Reaction of 11 using cyanuric chloride afforded (1,3-benzothiazol-2-ylsulfanyl)fluoroacetonitrile 6a and the crude reaction mixture was subjected to oxidation with H₅IO₆/CrO₃without prior purification, to give the desired 4 in isolated yields ranging from 38%-42% (over two steps). Reversal of synthetic steps involving initial oxidation of the amide 11 to corresponding sulfone was tested as well, but in this case subsequent conversion to cyano derivative was unsuccessful.

SCHEME 2.

Successful Synthesis of 4

With the desired (1,3-benzothiazol-2-ylsulfonyl)fluoroacetonitrile 4 in hand, condensations with carbonyl compounds were undertaken. Initially, DBU mediated condensation of 4 with 2-naphthaldehyde under Barbier conditions¹⁵ was performed. Although αfluoro acrylonitrile was formed, the isolated yield of the product was only 32%. Even lower product formation was observed in the condensation of 4 and 2-naphthaldehyde in the presence of LHMDS, at -78 ºC and under Barbier conditions²⁵ and no product was observed when NaHMDS was used as base.²⁵ A much higher 98% yield of the product was obtained upon a slow, dropwise addition of 4 to solution of DBU and aldehyde (Table 1, entry 1). In a typical procedure, aldehyde (1 molar equiv) and DBU (4 molar equiv) were dissolved in CH₂Cl₂ (5 mL/mmol of aldehyde) and under stirring at room temperature, a solution of 4 (2 molar equiv) in CH₂Cl₂ (7.5 mL/mmol of 4) was slowly added dropwise to the reaction mixture. This resulted in the change of color from clear to black-brown. The conversions were checked by TLC after 1 hour and if the aldehyde was still present, 1 molar equiv of sulfone in CH₂Cl₂ was added and the reaction was allowed to proceed for another 0.5 h, checked by TLC and if starting aldehyde was not consumed, DBU (2 molar equiv) in CH₂Cl₂ was added (Table 1, entries 3, 4). Upon disappearance of the aldehyde (15- 150 min) the crude reaction mixtures were analyzed by ¹⁹F NMR for E/Z ratios of products, partially concentrated, then directly loaded on a dry silica gel column and the products were eluted (for purification of 21, see the Experimental Section). Although E/Z isomers were not separated, we wanted to assess the ease of separation by tlc. For this purpose, tlc analysis was conducted for several product mixtures. In all cases tested the separation of isomers was easily achieved (please see Supporting Information for details). The yields obtained in the condensation reactions with a series of aldehydes and the E/Z ratios are displayed in Table 1. Wherever applicable, comparison to literature data is included.

TABLE 1.

Efficient, Mild Route to α-Fluoro Acrylonitriles

entry	aldehyde	temp	product: yield,a%E/Z ratiob	=C(CN)F: δ (ppm);cJ (Hz)	lit:d yield, E/Z ratio
1		rt	12: 98%, 15:85	-122.1; 36.6	--
		-78 °C	12: 97%, 8:92	-122.5; 15.3
2		rt	13: 93%, 19:81	-121.9; 35.2	54%, 66:3310
				-122.6; 17.6	53%, 21:7911
3		rt	14: 95%, 16:84	-126.2; 36.6	52%, 50:509
				-127.1; 18.3	50%, 32:6811
4		rt	15: 91%, 37:63	-122.3; 15.3	--
		-78 °C	15: 91%, 27:73	-124.1; 36.6
5		rt	16: 94%, 17:83	-120.4; 15.3	--
				-123.0; 36.6
6		rt	17: 72%, 17:83	-115.3; 15.3	45%, 33:679
				-115.7; 33.6
7		rt	18: 60%, 15:85	-117.0; 12.2	62%, 38:629
				-120.1; 30.5
8		rt	19: 91%, 16:84	-118.8; 15.3	--
				-119.8; 36.6
9		rt	20: 96%, 17:83	-122.8; 29.4	--
				-129.0; 11.7
10		rt	21: 59%, 18:82	-117.9; 33.6	--
				-125.3; 15.3
11		rt	22: 86%, 20:80	-119.2; 33.6	--
				-125.9; 12.2
12		rt	23: 92%, 17:83	-128.5; 35.2	--
				-129.6; 17.6
13		rt	24: 81%, 17:83	-126.6; 30.5	38%, 60:4010
				-127.7; 12.2	45%, 44:5611
14	n-heptyl-CHO	rt	25: 97%, 23:77	-123.9; 11.8	Lit. data for n-heptanal:e 30%, 30:7010 50%, 26:7411
		-78 °C	25: 90%, 16:84	-125.8; 35.2
15		rt	26: 80%, 23:77	-122.6; 14.7	51%, 26:7411
		-78 °C	26: 81%, 12:88	-124.2; 32.3
16		rt	27: 76%, 15:85	-122.8; 15.3	--
		-78 °C	27: 77%, 8:92	-125.6; 30.5

^aYields of isolated, purified products (reactions were performed under similar conditions but were not optimized for individual cases).

^bRelative ratio of diastereomers in the crude reaction mixture determined by ¹⁹F NMR prior to isolation. No change in ratio was observed after purification.

^cReferenced to CFCl₃ as internal standard.

^dWhere applicable, literature data and references are included for comparison.

^eSince no literature data for n-octanal are available, data for n-heptanal are included for comparison.

The reactions resulted in good to high yields of α-fluoroacrylonitriles. In the case of p- and onitrobenzaldehyde, Barbier conditions¹⁵ were found to give higher yields of products (72% and 60% respectively, entries 6 and 7, Table 1, see the Experimental Section for details).

Imidazole-4-carboxaldehyde and indole-3-carboxaldehyde gave high yields of products only upon N-Boc protection²⁶ (Table 1, entries 10, 11).

In all the reactions studied, the Z-isomer predominated, and stereoselectivity was nearly independent of the structure of the aldehyde. For aromatic aldehydes E/Z ratio ranged from 15/85-20/80, and was independent of the electronic nature of the substituent (entries 1-3, 5-12). The only exception was o-methoxybenzaldehyde (entry 4), where the E/Z ratio dropped to 37/63. At this point, we are unable to offer any explanation for this. We do not believe that this is due to electronic or steric influence, since no changes in E/Z ratios were observed in the case of pmethoxy, o-methyl, o-nitro or o-fluoro substituents (E/Z ratios 16:84, 17:83, 15:85, and 16:84, respectively). In the case of α,β-unsaturated aldehyde a similar E/Z ratio of 17:83 (entry 13) was observed, whereas aliphatic aldehydes gave E/Z ratios of 23/77 for the unbranched (entries 14, 15) and 15/85 for the branched aldehyde (entry 16). The effect of temperature on E/Z ratio was tested in the condensation reactions of 4 with 2-naphthaldehyde, 2-methoxybenzaldehyde, octanal, 3-phenylpropanal and 2-ethylbutanal. In all cases tested, lowering of reaction temperature to -78 °C improved the stereoselectivity, whereas the yield remained unchanged (Table 1, entries 1, 4, 14, 15, 16).

Scheme 3 shows the mechanism that was proposed by Julia (for X = H).^15,27,28 Reactions can proceed via competing pathways, leading to different stereochemical outcomes. The initial anti and syn addition product (I1 and I2) can either undergo retroaddition, if the BT-sulfone anion is stabilized, or collapse to products via the spirocyclic intermediates SI1 and SI2. Intermediate SI2 has been suggested to form faster than SI1 due to steric reasons. Thus, if retroaddition occurs, predominant stereoselectivity leading to cis arrangement of R₁ and R (Scheme 3) can be expected. An additional pathway involving stabilized zwitterionic intermediates Z1 and Z2 has been proposed for aryl and vinyl aldehydes.^15,27,28 In these cases formation of alkenes with a trans arrangement of R and R₁ through Z2 is favored. This is in fact the case with Julia reactions of aryl and vinyl aldehydes leading to the unfluorinated acrylates.²³ On the other hand, in our prior work on fluorinated acrylates a predominant cis arrangement of R and R₁ was observed.^7jThis led us to suggest that in the case of fluorinated sulfones, reaction involving zwitterionic intermediates Z1 and Z2 is disfavored due to the destabilizing effect a β-fluorine substituent has on carbocation stability.

SCHEME 3.

Fluoro-Julia Olefination Stereoselectivities

In connection to our previous work on the fluoro Julia olefination, we were interested in comparing the stereochemistry obtained in DBU mediated olefinations leading to α-fluoro acrylates^7j and to phenyl α-fluorovinyl sulfones,⁸ⁱ to those observed in the present synthesis of α-fluoro acrylonitriles. Interestingly, alkenes with the cis disposition of R (from RCHO) and COOR₂ or SO₂Ph moiety were the major products in the synthesis of acrylates^7j and phenyl sulfones.⁸ⁱ On the other hand, an opposite stereochemical outcome was observed in the present case. Assuming the lesser probability of Z1 and Z2 in fluoro Julia olefinations,^7j due to lower stability of β-fluorocarbocations, the plausible reasons for the present outcome could be as follows. In the event that the initial addition products I1 and I2 do not equilibrate, that is addition/retroaddition of the initially formed β-alkoxy sulfone anion does not occur, then the E/Z product ratio reflects the initial ratio of I1/I2. Alternatively, if equilibration of I1 and I2 does occur, then collapse of adduct I1 through SI1 may not be unfavorable, due to a small steric effect of the cyano functionality. Supportive of the latter, the selectivity in the condensation of the branched aldehyde 2-ethylbutanal (Table 1, entry 16) increased from 15:85 (E/Z) at rt to 8:92 (E/Z) at -78 °C. This may imply a lower energy barrier to Smiles rearrangement for adduct I1 as compared to I2.

CONCLUSIONS

In conclusion, we have developed the synthesis of (1,3-benzothiazol-2-ylsulfonyl)fluoroacetonitrile, a stable, isolable reagent for the preparation of α-fluoro acrylonitriles via a modified Julia olefination. The reagent undergoes reactions with aldehydes under mild, DBU mediated conditions, to give α-fluoro acrylonitriles in good yields and with predominant Z-stereoselectivity. The structure of the aldehyde generally does not influence the stereochemical outcome of the olefination.

EXPERIMENTAL SECTION

(1,3-Benzothiazol-2-ylsulfonyl)acetonitrile (3)

A solution of periodic acid (H₅IO₆, 2.21 g, 9.71 mmol, 4.0 molar equiv) in CH₃CN (39 mL) was allowed to vigorously stir for 30 min at rt. CrO₃ (4.9 mg, 0.05 mmol, 0.02 molar equiv) was added, and the stirring was continued for another 5 min. A solution of 1,3-benzothiazol-2-ylsulfanyl)acetonitrile (2a, 500 mg, 2.43 mmol, 1 molar equiv) in CH₃CN (5 mL) was slowly added dropwise and formation of precipitate was observed. After stirring at rt for 1 h, the reaction mixture was filtered, and the solid was washed with CH₃CN. The filtrate was concentrated under reduced pressure, EtOAc was added to the residue and the organic layer was washed with water and brine. The organic layer was dried over anhydrous Na₂SO₄, and the solvent was evaporated under reduced pressure. The crude product was purified by column chromatography on silica gel using CH₂Cl₂ as eluting solvent to yield 3 as a white solid (523 mg, 91% yield). Mp (recrystallized from 50% EtOAc in hexanes) 173-174 °C. ¹H NMR (500 MHz): δ 8.27 (d, 1H, Ar-H, J = 7.9), 8.07 (d, 1H, Ar-H, J = 7.6), 7.69 (m, 2H, Ar-H, J = 7.6), 4.56 (s, 2H, CH₂CN). HRMS (positive ion ESI): calcd. for C₉H₆N₂O₂S₂Na⁺ (M⁺ + Na) 260.976290, found 260.976192.

Ethyl (1,3-Benzothiazol-2-ylsulfanyl)fluoroacetate (8) via Fluorination of Ethyl (1,3-Benzothiazol-2-ylsulfanyl)acetate (10)

A stirred solution of ethyl (1,3-benzothiazol-2-ylsulfanyl)acetate 10 (200 mg, 0.791 mmol, 1 molar equiv) in 1.3 mL of dry toluene, was cooled to -80 ºC (dry-ice/iso-PrOH) under nitrogen. LDA (0.455 mL, 0.909 mmol, 1.15 molar equiv of a 2 M solution in heptane/THF/EtPh) was added to the reaction mixture and the stirring was continued at -80 ºC. After 1.5 h solid NFSI (312 mg, 0.988 mmol, 1.25 molar equiv) was added, the reaction mixture was allowed to stir at -80 ºC for another 1 h and then warmed to room temperature over 1 h. Sat aq NH₄Cl was added to the mixture and the layers were separated. The aqueous layer was extracted with EtOAc (3 x), and the combined organic layer was washed with sat aq NaHCO₃ and brine. The organic layer was dried over Na₂SO₄ and the solvent was evaporated under reduced pressure. The crude reaction mixtures was purified by column chromatography on silica gel using CH₂Cl₂ to yield 8 as a clear thick pale-yellow liquid (123 mg, 57%). ¹H NMR (500 MHz): δ7.99 (d, 1H, Ar-H, J = 7.9), 7.83 (d, 1H, Ar-H, J = 7.9), 7.49 (t, 1H, Ar-H, J = 7.6), 7.39 (t, 1H, Ar-H, J = 7.6), 6.94 (d, 1H, CHF, ²J_FH = 51.0), 4.34 (q, 2H, OCH₂, J = 7.1), 1.33 (t, 3H, CH₃, J = 7.1). ¹⁹F NMR (282 MHz): δ -161.7 (d, ²J_FH = 47.0). HRMS (positive ion ESI): calcd. for C₁₁H₁₀FNO₂S₂Na⁺ (M⁺ + Na) 294.002919, found 294.002247.

For synthesis of 8 from 7, please see Supporting Information.

(1,3-Benzothiazol-2-ylsulfanyl)fluoroacetamide (11)

Into a solution of ethyl (1,3-benzothiazol-2-ylsulfanyl)fluoroacetate 8 prepared from 7 (3.88 g, 14.3 mmol) in ethanol (50 mL) at room temperature was bubbled a gentle stream of NH₃ gas until disappearance of the starting material was observed by TLC (SiO₂, 50% EtOAc in hexanes), and after 1 h the solvent was evaporated under reduced pressure. The product was dried overnight under vacuum to yield 3.33 g (96%) of 11 as a white solid that did not require any further purification. Mp (recrystallized from 50% EtOAc in hexanes) 134-135 °C. ¹H NMR (500 MHz): δ 7.98 (d, 1H, Ar-H, J = 8.2), 7.82 (d, 1H, Ar-H, J = 7.9), 7.47 (t, 1H, Ar-H, J = 7.9), 7.39 (t, 1H, Ar-H, J = 7.9), 7.01 (d, 1H, CHF, ²J_FH = 51.9), 6.53 (br s, 1H, NH₂), 5.84 (br s, 1H, NH₂). ¹⁹F NMR (282MHz):δ -158.4 (d, ²J_FH = 52.8). HRMS (positive ion ESI): calcd. for C₉H₇FN₂OS₂Na⁺ (M⁺ + Na) 264.987604, found 264.987469.

(1,3-Benzothiazol-2-ylsulfonyl)fluoroacetonitrile (4)

Step 1

To a stirred solution of cyanuric chloride (3.89 g, 21.1 mmol, 1.7 molar equiv) in DMF (57 mL), was added (1,3-benzothiazol-2-ylsulfanyl)fluoroacetamide (11, 3.00 g, 12.4 mmol, 1 molar equiv) at rt. The stirring was continued at rt for 1.5 h, until TLC (SiO₂, 50% EtOAc in hexanes) showed disappearance of starting material. The reaction mixture was diluted with water and extracted with EtOAc (3x). The combined organic layer was thoroughly washed with water, brine, dried over anhydrous Na₂SO₄ and the solvent was evaporated under reduced pressure to yield 2.49 g of the crude product 6a (see Supporting Information for analytical data on a purified sample), that was subjected to oxidation without further purification.

Step 2

A solution of periodic acid (H₅IO₆, 10.1 g, 44.5 mmol, 4.0 molar equiv) in CH₃CN (100 mL) was allowed to vigorously stir for 30 min at rt, CrO₃ (22.3 mg, 0.222 mmol, 0.02 molar equiv) was added, and the stirring was continued for another 5 min. A solution of 1,3-benzothiazol-2-ylsulfanyl)fluoroacetonitrile (2.49 g, 11.1 mmol, 1 molar equiv, crude reaction mixture from step 1) in CH₃CN (20 mL) was slowly added dropwise and formation of precipitate was observed. After stirring at rt for 1 h, the reaction mixture was filtered and concentrated under reduced pressure. EtOAc was added to the residue and the mixture was washed with water and brine. The organic layer was dried over anhydrous Na₂SO₄, and the solvent was evaporated under reduced pressure. The crude product was purified by column chromatography on silica gel using CH₂Cl₂ as eluting solvent to yield 4 as an off white solid (1.21 g, 38% yield over two steps). Mp (recrystallized from 50% EtOAc in hexanes) 149-150 °C. ¹H NMR (500 MHz): δ 8.33 (d, 1H, Ar-H, J = 7.9), 8.09 (d, 1H, Ar-H, J = 7.6), 7.73 (m, 2H, Ar-H), 6.24 (d, 1H, CHF, ²J_FH = 46.7). ¹⁹F NMR (282 MHz): δ-179.4 (d, ²J_FH = 47.0). HRMS (positive ion ESI): calcd. for C₉H₅FN₂O₂S₂Na⁺ (M⁺ + Na) 278.966868, found 278.966867.

Subsequent repetition of the two-step procedure using 5.00 g of 11 resulted in a 42% yield of 4.

(1,3-Benzothiazol-2-ylsulfonyl)fluoroacetonitrile (4) via Fluorination of (1,3-Benzothiazol-2-ylsulfonyl)acetonitrile (3)

A suspension of NaH (49.9 mg, 2.08 mmol, 1.1 molar equiv) in dry THF (6 mL) was cooled to 0 °C under nitrogen and a solution of (1,3-benzothiazol-2-ylsulfonyl)acetonitrile (3, 450 mg, 1.89 mmol, 1 molar equiv) in dry THF (6.5 mL) was added dropwise with stirring. After the addition, the reaction mixture was allowed to warm to rt and left to stir for 45 min. The mixture was then cooled to 0 °C, Selectfluor (804 mg, 2.27 mmol, 1.2 molar equiv) was added, the cooling bath was removed and the reaction mixture was allowed to stir at rt for 1.5 h, and then quenched with aqueous NH₄Cl solution. The aqueous layer was extracted with EtOAc (3x), the combined organic layers were washed with aqueous NaHCO₃, water and brine, dried over Na₂SO₄ and the solvent was evaporated. Separation of the crude reaction mixture by column chromatography on silica gel using CH₂Cl₂ as eluting solvent afforded 4 as an off white solid (155 mg, 32% yield).

General Procedure for Synthesis of 12-16, 19-20, 22-27 via Condensation of Aldehydes with Fluorinated Sulfone 4 at rt

A solution of 4 (2 molar equiv) in CH₂Cl₂ (7.5 mL/mmol of 4) was added slowly, dropwise to a stirring solution of aldehyde (1 molar equiv) and DBU (4 molar equiv) in freshly distilled CH₂Cl₂ (5 mL/mmol of aldehyde) at rt. Upon addition of 4, the reaction mixture turned black brown. The conversions were checked by TLC after 1 hour and if the aldehyde was still present, a solution of 1 molar equiv of sulfone in CH₂Cl₂ (7.5 mL/mmol of 4) was added dropwise and the reaction was allowed to proceed for another 0.5 h, checked by TLC and if starting aldehyde was still not consumed, DBU (2 molar equiv) in CH₂Cl₂ (1 mL/mmol of DBU) was added. The stirring was continued at rt until complete consumption of the aldehyde was observed by TLC (SiO₂, CH₂Cl₂), usually 15-150 min. The E/Z ratio was determined by ¹⁹F NMR of an aliquot, the reaction mixture was concentrated to about 2 mL and directly loaded onto a dry silica gel column (200-300 mesh). The E/Z product mixture was obtained by elution with CH₂Cl₂ and the solvent was evaporated (except in the case of 21, please see detailed experimental procedure below).

Synthesis of 21 via Condensation of 4 with 1-Boc-imidazole-4-carboxaldehyde

A solution of 4 (522 mg, 2.04 mmol, 2 molar equiv) in CH₂Cl₂ (15 mL) was added slowly, dropwise (for about 10 min) to a stirring solution of 1-Boc-imidazole-4-carboxaldehyde (200 mg, 1.01 mmol, 1 molar equiv) and DBU (618 mg, 4.06 mmol, 4 molar equiv) in freshly distilled CH₂Cl₂ (5.1 mL) at rt. Upon addition of 4, the reaction mixture turned black brown. The stirring was continued at rt until complete consumption of the aldehyde was observed (1.5 h) by TLC (SiO₂, 50% EtOAc in hexanes). The solvent was concentrated under reduced pressure to about 2-3 mL and the reaction mixture was loaded onto a dry silica gel column (200-300 mesh) and the E/Z product mixture was eluted with 50% EtOAc in hexanes. The solvent was evaporated and the E/Z ratio was determined by ¹⁹F NMR (18/82, Table 1, entry 15, the E/Z ratio does not change upon purification). Since the E/Z product mixture was contaminated with 1,3-benzothiazol-2-ol byproduct, the mixture was dissolved in CH₂Cl₂ (30 mL), aqueous NaOH was added (0.5 M NaOH, 6 mL) and the mixture was stirred at rt for 10 minutes, the organic layer was separated, washed with water and brine and dried over anhydrous Na₂SO₄. Upon removal of solvent under reduced pressure, 143 mg (59%) of 21 was isolated as an off-white solid. ¹H NMR (500 MHz, CDCl₃): δ 8.12 (s, 1H, Ar-H, E-isomer), 8.09 (s, 1H, Ar-H, Z-isomer), 7.75 (s, 1H, Ar-H, Z-isomer), 7.71 (s, 1H, Ar-H, E-isomer), 6.94 (d, 1H, ²J_FH = 14.6, E-isomer), 6.63 (d, 1H, ²J_FH = 34.5, Z-isomer), 1.64 (s, 9H, t-Bu, Z-isomer), 1.635 (s, 9H, t-Bu, E-isomer). ¹⁹F NMR (282 MHz, CDCl₃): δ -117.9 (d, ³J_FH = 33.6, Z isomer), δ -125.3 (d, ³J_FH = 15.3, E isomer). HRMS (positive ion ESI) calcd. for C₁₁H₁₂FN₃O₂Na⁺ (M⁺ + Na) 260.080576, found 260.080458.

General Procedure for Synthesis of 12, 15, 25, 26, 27 via Condensation of Aldehydes with Fluorinated Sulfone 4 at -78 °C

A solution of 4 (2 molar equiv) in CH₂Cl₂ (7.5 mL/mmol of 4) was added slowly, dropwise to a stirring solution of aldehyde (1 molar equiv) and DBU (4 molar equiv) in freshly distilled CH₂Cl₂ (5 mL/mmol of aldehyde) that was cooled to -78 °C under N2. Upon addition of 4, the reaction mixture turned black brown. The reaction mixture was stirred at -78 °C for 1 hour and then allowed to warm to rt. In all cases studied, TLC showed complete conversions. Analysis of reaction mixtures by ¹⁹F NMR and isolation of E/Z product mixtures was performed as described for the rt condensations.

General Procedure for the Synthesis of 17 and 18 via Condensation of Aldehydes with Fluorinated Sulfone 4

In the case of p-nitrobenzaldehyde and o-nitrobenzaldehyde, the condensations were performed under Barbier conditions as follows.

Aldehyde (1 molar equiv) and 4 (2 molar equiv) were dissolved in CH₂Cl₂ (19 mL/mmol of aldehyde), and under stirring at rt, DBU (4 molar equiv) in CH₂Cl₂ (1 mL/mmol of DBU) was slowly added dropwise to the reaction mixture. Upon addition of DBU, the reaction mixture turned black brown. The reaction was monitored by TLC and upon consumption of the aldehyde, the E/Z ratio was determined by ¹⁹F NMR. The reaction mixture was concentrated to about 2 mL in vacuo and directly loaded onto a dry silica gel column (200-300 mesh). The E/Z product mixture was eluted with CH₂Cl₂ and the solvent was evaporated.