o,o-Difluorophenylacetic acid 10.32g (59.95mmol) and 2 drops of DMF was suspended in dichloromethane 25mL in a 250mL round flask. With cooling on ambient water bath, oxalyl chloride 7,9mL (90 mmol) was added in one portion and the mixture was stirred under Drierite-filled tube (as a gas outlet) overnight (18 hours). The reaction mixture was concentrated on rotavap (RT, 150 to 12 Torr). The residue was distilled at 12 Torr using a Kugelrohr apparatus, with the air bath temperature 135C. Y=11.258g (98.5%) of the o,o-difluorophenacetyl chloride asa pale yellow liquid. 1H(CDCl3, 400MHz): 7.314(m, 1H), 6.934(app dd, 8.2Hz, 7.0Hz, 2H), 4.232(t, 1.2Hz, 2H); 13C(CDCl3, 100MHz): 170.03(s, 1C), 161.23(dd, 249Hz, 6,9Hz,2C), 130.50(t, 10.0Hz, 1C), 111.55(dd, 19.2Hz, 6.1Hz, 2C), 108.72(t, 20.0Hz, 1C), 40.06(t, 3.0Hz, 1C); 19F(CDCl3, 376.5MHz): -114.19(t, 6.6Hz, 2F)
Aluminum chloride 8.10g (99.5%+ powder, Aldrich, 60.7 mmol) in an oven-dried 250mL flask under Ar was suspended in anh carbon disulfide 55mL and the slurry was stirred for 5 min. With cooling on ambient water bath, solid oxindole 1.332g (10.0mmol) was added in one portion and the mixture was stirred for 10 minutes. Neat o,o-phenacetyl chloride 1.906g (10.0mmol) was added in one portion to the reaction mixture and stirring was continued for additional 10 minutes. The ambient bath was then replaced with an oil bath and the reaction mixture was stirred under a reflux condenser at 60-70C under Ar for 5 hours. (A sticky bottom gum in the heterogennous reaction mixture eventually solidified and the mixture became stirable again.) After 5 hours the reaction mixture was carefully quenched by addition of 1M HCl 100mL (added through condenser) and the mixture was refluxed for additional 30 min, then cooled, diluted with hexane 100mL and vigorously stirred at RT overnight (18 hours, all sticky chunks gradualy transformed into a white free-floating precipitate). The obtained 2-layer mixture with a precipitate was filtered through a large medium-porosity sintered-glass Buchner funnel, the precipitate was washed thoroughly with hexane and water (repeatedly, alternating water with hexane) and then dried by suction. The obtained crude product was dissolved in acetone (approx 250mL, 10 min stirring), the solution was filtered through the same Buchner funnel (washed with additional acetone) and the filtrates were evaporated. The residue was suspended in a mixture of acetone 20mL and ethanol 40mL, the mixture was heated to reflux and then gradualy diluted with water (about 0.5L). The obtained slurry was allowed to settle in a refrigerator (+4C) for 2 hours. The precipitate was collected by filtration, washed with water and dried on highvac. Y=2.850g (99%) of a cream-colored crystalline solid. 1H(d6-DMSO, 400MHz): 10.811(br s, 1H), 7.998(br dd, 8.2Hz, 1.5Hz, 1H), 7.926(br s, 1H), 7.392(m, 1H), 7.101(app t, 7.8Hz, 2H), 6.947(d, 8.2Hz, 1H), 4.410(s, 2H), 3.582(s, 2H); 19F(d6-DMSO, 376.5MHz): -114.80(t, 6.6Hz, 2F)
The 5-(o,o-difluorophenacetyl)-oxindole 1.669g (5.81 mmol) and NaBH4 704mg (18.61mmol) suspension in anh ethanol 100mL was stirred at 60C for 2 hours. The reaction mixture was cooled to RT, acidified with 4M HCl 10mL (gas evolution), diluted with water 0.7L and allowed to crystallize in a refrigerator overnight (+4C, 14h). The precipitated product (1.301g) was collected by filtration, washed with water and dried on highvac. A second fraction of the pure product (260mg) was obtained by concentrating the supernatants down to 0.5L volume, cooling the mixture on ice and filtering the precipitate. Combined Y=1.561g(93%) of a white crystalline solid. 1H(d6-DMSO, 400MHz): 10.306(s, 1H), 7.268(m, 1H), 7.158(br s, 1H), 6.991(m, 3H), 6.701(d, 8.2Hz, 1H), 5.333(br d, 3.1Hz, 1H), 4.674(br t, 6.0Hz, 1H), 3.427(s, 2H), 2.955(ddABX, 13.3Hz, 8.2Hz, 1H), 2.838(ddABX, 13.3Hz, 5.9Hz, 1H); 19F(d6-DMSO, 376.5Hz): -114.72(t, 6.6Hz, 2F)
2 mL of triethylsilane was stirred with TFA 11mL for 10 min and the resulting homogennous mixture was added to the hydroxyderivative from the previous step, 402.5mg (1.391mmol). The mixture was stirred at RT for 1 hour, the reaction was quenched by addition of hexane 10mL and water 40mL. The stirring was continued for additional 2 hours, the precipitated product was collected from the heterogennous mixture by filtration, washed thoroughly with hexane and water, then dried on highvac. Y=363mg (95.5%) of a white crystalline solid.
1H(d6-DMSO, 400MHz): 10.272(s, 1H), 7.291(m, 1H), 7.032(app t, 8.2Hz, 3H), 6.935(br d, 8.2Hz, 1H), 6.688(d, 7.8Hz, 1H), 3.409(s, 2H), 2.847(br dAB, 8.1Hz, 2H), 2.745(br dd, 8.9Hz, 5.8Hz, 2H); 19F(d6-DMSO, 376.5MHz): -116.29(t, 8.0Hz, 2F)
The oxindole Friedel-Craft with phenacetyl chlorides works well only when the phenacetyl chloride has at least one ortho halogen substituent on the ring and no substitution on the methylene. It is likely that enolisation of the oxindole and the formed aryl ketone takes place with excess of AlCl3. Ar-conjugation and the ortho substitution on the ring supresses polycondensation side-reactions of the Al-enolate complex.
The TES-H + TFA reduction to the ethylene compound can be probably accomplished from the aryl ketone in one step, without need to reduce it first with borohydride to alcohol. In this case, we needed the alcohol intermediate also. Even more powerful (and easier to work up) carbocation reduction system employs triethyl silane in TFA with added BF3 etherate.
Update: As Milo and others have commented, CS2 is a particularly unpleasant solvent: It stinks, it is highly neurotoxic – the damage is cumulative – and its vapors have a particularly low self-ignition temperature (glassware hot from drying oven or a hotplate can ignite CS2 vapors). Take all necessary precautions and keep the stink in the hood. If you can smell CS2, it is bad for you already.