Synthesis of Regiospecifically Fluorinated Conjugated Dienamides

Modular synthesis of regiospecifically fluorinated 2,4-diene Weinreb amides, with defined stereochemistry at both double bonds, was achieved via two sequential Julia-Kocienski olefinations. In the first step, a Z-α-fluorovinyl Weinreb amide unit with a benzothiazolylsulfanyl substituent at the allylic position was assembled. This was achieved via condensation of two primary building blocks, namely 2-(benzo[d]thiazol-2-ylsulfonyl)-2-fluoro-N-methoxy-N-methylacetamide (a Julia-Kocienski olefination reagent) and 2-(benzo[d]thiazol-2-ylthio)acetaldehyde (a bifunctional building block). This condensation was highly Z-selective and proceeded in a good 76% yield. Oxidation of benzothiazolylsulfanyl moiety furnished a second-generation Julia-Kocienski olefination reagent, which was used for the introduction of the second olefinic linkage via DBU-mediated condensations with aldehydes, to give (2Z,4E/Z)-dienamides in 50%–74% yield. Although olefinations were 4Z-selective, (2Z,4E/Z)-2-fluoro-2,4-dienamides could be readily isomerized to the corresponding 5-substituted (2Z,4E)-2-fluoro-N-methoxy-N-methylpenta-2,4-dienamides in the presence of catalytic iodine.

Although the dioxolane derivative of 2 could be readily prepared from 2-(bromomethyl)-1,3-dioxolane and the sodium salt of 2-mercapto-1,3-benzothiazole, attempts at deprotection of 2-[(1,3-dioxolan-2yl)methylthio]benzo[d]thiazole under various conditions proved unsuccessful. Therefore, synthesis via the dimethyl acetal was considered (Scheme above Table 1). Various conditions were tested to unmask the aldehyde functionality (Table 1). Upon reaction of dimethyl acetal 3 with I 2 (entry 1), CBr 4 (entry 2), or PTSA (entry 3), no hydrolysis was observed. Reaction of 3 with 4 M HCl at 40 °C resulted in complete consumption of 3, but aldehyde 2 was isolated in a low 20% yield after column chromatography (entry 4). Subsequently, we found that compound 2 is unstable under chromatography conditions, on silica gel and alumina [52]. The yield of crude 2 after hydrolysis with 4 M HCl, but without chromatography, was 56%. Hydrolysis with 12 M HCl at 50 °C was complete within 30 min, yielding crude 2 in 81% yield (entry 6). However, due to the solubility of 2 in water, we obtained inconsistent results in repeat experiments. After extensive experimentation we found that crude 2 could be isolated in consistent yields when aqueous workup was avoided. Briefly, acetal 3 was reacted with 12 M HCl in acetone-H 2 O (10:1) at 50 °C for 40 min (entry 7), solid NaHCO 3 was added portion-wise at 5 °C to neutralize the acid, and excess water was removed by addition of anhydrous Na 2 SO 4 . The solution was then passed through a bed of anhydrous Na 2 SO 4 and the solvent was evaporated to afford 2 in >80% yield. When acetone alone was used as solvent, complete hydrolysis of 3 occurred, but the crude product showed the presence of an unidentified byproduct that could possibly result from the condensation of acetone and 2. The use of water as a co-solvent therefore seems to be crucial in order to minimize the formation of the byproduct.
With both desired building blocks in hand, i.e., the Julia-Kocienski reagent 1 and aldehyde 2, we tested reaction conditions for the olefination reaction ( Table 2). All condensation reactions were performed at −78 °C in the presence of LHMDS, and gave (Z)-4-(benzo[d]thiazol-2-ylthio)-2-fluoro-N-methoxy-N-methylbut-2-enamide (4) as the only stereoisomer. Comparably, exclusive Z-selectivity has also been observed in NaH-mediated condensations of 1 with aldehydes [31]. In the reactions herein, the molar ratio of sulfone 1, aldehyde 2, and LHMDS was critical for obtaining a good yield of 4 ( Table 2). When aldehyde 2 was used as a limiting reactant (entry 1), or in an equimolar amount (entry 2), enamide 4 was obtained in low yield. On the other hand, with excess aldehyde 2 and LHMDS, a substantial yield improvement was observed. Thus, product 4 was isolated in 76% yield when 2 molar equiv of 2 and 3 molar equiv of LHMDS were used (entry 4). Since the desired product was obtained with exclusive Z-selectivity and in a good yield, we did not attempt to use other bases, such as KHMDS or NaHMDS. In order to obtain the second generation Julia-Kocienski reagent 5, sulfide 4 was oxidized using H 5 IO 6 and catalytic CrO 3 . Sulfone 5, obtained in 63% yield, was then used for the screening of reaction conditions for the olefination with 2-naphthaldehyde (Table 3). Table 3. Conditions tested for olefination reactions using the second generation Julia-Kocienski reagent 5 and 2-naphthaldehyde.
The relative ratio of isomers in the crude reaction mixtures was determined by 19 F-NMR prior to isolation.
No change in the relative ratio was observed after purification; b Yield is of isolated and purified product 6a; c No product was detected either by 19 F-NMR or by TLC.
Both selectivity and product yield depended upon the reaction conditions. No product formation occurred when LHMDS was used as base (entries 1 and 2), or with DBU as base in THF at room temperature (entry 3). Similarly, Cs 2 CO 3 in either THF or CH 2 Cl 2 at 0 °C did not show product formation (entries 5 and 7). Product 6a was obtained in a low 35% yield and with a moderate 4E selectivity in an overnight reaction with DBU in THF, at −78 to 0 °C (E/Z 57/43, entry 4). When the condensation reaction was allowed to run overnight at 0 °C (entry 6), product 6a was isolated in a better 55% yield, but with a reversed selectivity as compared to entry 4 (E/Z 43/57). Yield and selectivity increased when the condensation reaction was performed overnight using DBU as base in CH 2 Cl 2 , at 0 °C (66%, entry 8).
Using these conditions, the generality of condensation reactions of Julia-Kocienski reagent 5 with other aldehydes was tested. Table 4 shows yields, the 4E/4Z ratios, and 19 F-NMR data of the products. Table 4. Reactions of reagent 5 with aldehydes: yields, E/Z ratios, and 19 F-NMR data. Moderate to high 4Z selectivity was obtained with electron-rich aryl and heteroaryl aldehydes, with yields ranging from 50%-66% (entries 1, 2 and 4). The electron-deficient p-nitrobenzaldehyde gave product 6c in a good 74% yield, but with poor 4Z selectivity (entry 3). Reaction of 5 with 3-phenylpropanal gave product 6e in a moderate 51% yield and with high 4Z selectivity (entry 5). In the 19 F-NMR spectra of all products, the doublet from the (4E)-isomer appears more upfield as compared to the doublet from the (4Z)-isomer (Table 4, entries 1-5).
Next, we considered isomerization of the (2Z,4Z)-isomer to the (2Z,4E)-isomer. Several techniques were evaluated to effect this isomerization. Overnight exposure of the 4E/4Z isomer mixture to light (20 watt bulb) did not cause any isomerization. Treatment of the isomer mixtures with silica powder in CHCl 3 at room temperature or at 0 °C showed the desired isomerization, but the isomerization did not proceed to completion. A convenient method for the isomerization using catalytic I 2 in CHCl 3 at room temperature has been reported [53]. Using this method, complete isomerization of (2Z,4E/Z)-6a-d to (2Z,4E)-6a-d was achieved (Table 5).

General Information
THF was distilled over LiAlH 4 and then over sodium. CH 2 Cl 2 , EtOAc, and hexanes were distilled over CaCl 2 . For reactions performed under a nitrogen atmosphere, glassware was dried with a heat gun under vacuum. LHMDS (1.0 M in THF) was obtained from commercial sources. Julia Kocienski reagent 2-(benzo[d]thiazol-2-ylsulfonyl)-2-fluoro-N-methoxy-N-methylacetamide (1) was prepared from the known 2-(benzo[d]thiazol-2-ylsulfonyl)-N-methoxy-N-methylacetamide [54], via metalationfluorination using our previously reported procedure [31]. All other reagents were obtained from commercial sources and used without further purification. Thin layer chromatography was performed on Analtech silica gel plates (250 m). Column chromatographic purifications were performed on 200-300 mesh silica gel. 1 H-NMR spectra were recorded at 500 MHz in CDCl 3 and are referenced to residual solvent. 13 C-NMR spectra were recorded at 125 MHz and are referenced to the carbon resonance of the deuterated solvent. 19 F-NMR spectra were recorded at 282 MHz with CFCl 3 as an internal standard. Chemical shifts () are reported in parts per million and coupling constants (J) are in hertz (Hz).

2-(Benzo[d]thiazol-2-ylthio)acetaldehyde 2. To a stirred solution of 2-(2,2-dimethoxyethylthio)benzo
[d]thiazole (3, 1.60 g, 6.26 mmol) in acetone (48 mL), was slowly added a mixture of HCl (12 M, 10.6 mL) and water (5.2 mL) at rt. The mixture was stirred for 40 min at 50 °C. Upon completion of the reaction, as observed by TLC, the mixture was cooled to 5 °C and the reaction was quenched by portion-wise addition of solid NaHCO 3 up to the neutralization point, and then passed through a bed of anhydrous Na 2 SO 4 . The anhydrous Na 2 SO 4 bed was washed with a minimum amount of acetone, the combined eluent was dried over anhydrous Na 2 SO 4 , and concentrated under reduced pressure. Crude product 2 (1.10 g, 84%) was used in the next step without purification. R f (SiO 2 , 20% EtOAc in hexanes): 0.29.  N-methylacetamide (1, 0.700 g, 2.20 mmol) and 2-(benzo[d]thiazol-2-ylthio)acetaldehyde (2, 0.930 g, 4.44 mmol, 2.0 molar equiv.) in dry THF (48.0 mL) at −78 °C (dry ice/iPrOH), was added LHMDS (4.39 mL, 1 M, 4.39 mmol, 2.0 molar equiv.) dropwise under a nitrogen atmosphere. The mixture was allowed to stir at −78 °C (dry ice/iPrOH) for 2 h and checked for the disappearance of 1 by 19 F-NMR (a small sample was removed by syringe and checked by NMR). Since 19 F-NMR showed the presence of 1, more LHMDS (2.19 mL, 1 M, 2.19 mmol, 1 molar equiv.) was added and the mixture was allowed to stir at −78 °C for an additional 1 h at which time complete consumption of sulfone 1 was observed by 19 F-NMR. The reaction was quenched by the addition of saturated aq. NH 4 Cl, the solvent was partially removed under reduced pressure, and the mixture was extracted with EtOAc (3×). The combined organic layer was washed with 5% aq. NaOH, followed by water and brine, and then dried over Na 2 SO 4 . The organic layer was concentrated under reduced pressure and the crude product was purified by column chromatography using 10%, 15%, and 20% EtOAc in hexanes, to afford compound 4 as a yellow wax (0.523 g, 76%). R f (SiO 2 , 30% EtOAc 698 mmol) in CH 3 CN (10.0 mL) was added dropwise to this mixture, resulting in an exothermic reaction and the formation of a yellowish precipitate. After complete addition, the mixture was stirred for 3 h, at which time TLC showed complete consumption of amide 4. The mixture was filtered through a Celite pad, the pad was washed with CH 3 CN, and the filtrate was concentrated under reduced pressure. Water was added to the residue and the mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with saturated aq. NaHCO 3 (5 × 30 mL) and brine (30 mL), and dried over anhydrous Na 2 SO 4 . The organic layer was concentrated under reduced pressure and the crude product was purified by column chromatography using 15%, 20%, and 25% EtOAc in hexanes to afford compound 5 as a white solid (0.152 g, 63%). R f

Condensation Reactions of Julia-Kocienski Reagent 5
General experimental procedure. To a stirred solution of aldehyde (0.20 mmol) in dry CH 2 Cl 2 (10.0 mL) was added DBU (121.7 mg, 0.80 mmol, 4.0 molar equiv.) and the mixture was cooled to 0 °C. A solution of (Z)-4-(benzo[d]thiazol-2-ylsulfonyl)-2-fluoro-N-methoxy-N-methylbut-2-enamide (5, 103.3 mg, 0.300 mmol, 1.5 equiv.) in dry CH 2 Cl 2 (10.0 mL) was then added slowly, dropwise (over about 2 h). The reaction mixture was allowed to stir overnight at 0 °C. After completion of the reaction, the solvent was evaporated under a stream of nitrogen gas, and the 1 H and 19 F-NMR spectra of the crude product mixture were recorded for determination of the E/Z ratio. The combined E/Z product mixture was purified by column chromatography. For eluting solvents see the specific compound headings. The mixture of (4E)-and (4Z)-isomers was analyzed and characterized based on the 1 H-NMR; the assignment and integration of specific olefinic proton(s) allowed for other signals to be assigned based on the integration, along with a comparison to the pure (2Z,4E)-isomer (obtained after isomerization, vide infra). Assignment of the 19 F-NMR signals to the (4E)-and (4Z)-isomers was based on the integration. Isomer ratio of (4E)-6e:(4Z)-6e = 15:85. Column chromatography using 6%, 8%, and 12% EtOAc in hexanes gave a mixture of (2Z,4E/Z)-6e as a colorless oil (26.6 mg, 51%). R f (SiO 2 , 30% EtOAc in hexanes): 0.30. We were unable to resolve the 1 H-NMR signals for both isomers, so the signals reported are for the major (2Z,4Z)-isomer. For the minor isomer, only diagnostic vinylic C4-H could be unequivocally assigned (separately listed, vide infra). 1

Isomerization of the (2Z,4E/Z)-Isomer Mixture of 6a-d to the (2Z,4E)-Isomer
General experimental procedure. To a stirred solution of the (2Z,4E/Z)-isomer mixture 6 in dry CHCl 3 was added I 2 (7-10 mol %) and the mixture was stirred at rt. The reaction was monitored by 19 F-NMR and when only one isomer was observed, the reaction mixture was diluted with EtOAc (30 mL). The mixture was washed with water, saturated aq. Na 2 S 2 O 3 (2 × 10 mL), and dried over Na 2 SO 4 . The organic layer was concentrated under reduced pressure to afford the desired (2Z,4E)-isomer. HMQC data were obtained for (2Z,4E)-6c and (2Z,4E)-6d, and where unequivocally assigned, through-bond