Aromatic C–H amination in hexafluoroisopropanol† †Electronic supplementary information (ESI) available: Detailed experimental procedures and spectroscopic characterization for all new compounds. CCDC 1545194. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c8sc04966a

We report direct amination of electron-poor arenes and evaluate the crucial factors for the enhanced reactivity in hexafluoroisopropanol.


MATERIALS AND METHODS
All manipulations were carried out under ambient atmosphere unless otherwise noted.

Solvents
HFIP was purchased from Oakwood Chemicals and used as received except where noted. Where it is noted that HFIP was distilled and degassed, HFIP was distilled from 3Å molecular sieves and degassed by the freeze-pump-thaw method. Anhydrous diethyl ether, tetrahydrofuran, dichloromethane and acetonitrile were obtained by filtration through drying columns on an mBraun system. 1

Chromatography
Thin layer chromatography (TLC) was performed using EMD TLC silica gel 60 F 254 plates pre-coated with 250 m thickness silica gel and visualized by fluorescence quenching under UV light and KMnO 4 stain.
Preparative TLC was performed using Analtech Uniplates pre-coated with 1000 m thickness silica gel GF with a volume of the mobile phase of ~100 mL. Flash chromatography was performed using silica gel (230-400 mesh) purchased from Silicycle Inc. Detailed flash column chromatography specifications are given for amination of the substrate ethyl 2-thiophenecarboxylate as a representative example.

Spectroscopy and Instruments
NMR spectra were recorded on either a Varian Unity/Inova 600 spectrometer operating at 600 MHz for 1  GaussView5. The conductor-like polarizable continuum model (CPCM) has been used to simulate solvent effects (1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), acetonitrile). 6,7 Ground state energies are given with respect to the thermal free energy correction at 298.15 K. Time-dependent DFT (TD-DFT) calculations have been carried out using the coordinates of the optimized ground state structures. Images of molecular structures and orbital plots were generated using GaussView5 and Chem3D.

Starting materials
All substrates were used as received from commercial suppliers, unless otherwise stated. FeSO 4 ·7H 2 O was ground into a fine powder before use. 1.00 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M) and the arene (if liquid). The reaction mixture was stirred at 60 °C for 15-120 min, or until judged complete by the color and/or TLC. The reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. The residue was purified by column chromatography and/or preparative TLC. Reagent 1 was synthesized by the method of Morandi 9a and Fagnou 9b . Compound S1 (1.50 g, 7.10 mmol, 1.00 equiv) was dissolved in anhydrous diethyl ether (36 mL, c = 0.2 M) in a flame-dried two-neck round bottom flask. The flask was evacuated and backfilled with nitrogen, then cooled in a water-ice bath. Triflic acid (1.07 g, 627 µL, 7.10 mmol, 1.00 equiv) was added using a plastic pipettor. The reaction mixture was stirred at 23 ºC for 2 h, during which time a colorless precipitate formed. Pentane (20 mL)  The title compound was synthesized by the method of Williams. 11 4-(Trifluoromethyl)phenol (400. mg, 2.47 mmol, 1.00 equiv) was dissolved in anhydrous dichloromethane (12 mL, c = 0.2 M) in a flame-dried round bottom two neck flask. Triethylamine (250 mg, 344 µL, 2.47 mmol, 1.00 equiv) was added and the mixture was cooled in a water-ice bath. 2-Nitrobenzenesulfonyl chloride (547 mg, 2.47 mmol, 1.00 equiv) was then added. The reaction mixture was stirred at 23 ºC for 16 h. The reaction mixture was then poured into a separatory funnel containing 1M HCl (aq) (25 mL). The layers were separated and the aqueous layer was extracted with dichloromethane (2 × 15 mL). The combined organic layers were dried over Na 2 SO 4 , filtered and concentrated. The residue was purified by column chromatography on a short plug of silica gel eluting with a solvent mixture of ethyl acetate/pentane (30:70 (v/v)). Purification afforded 747 mg of the title compound as a colorless solid (87% yield). Rf = 0.54 (ethyl acetate/hexanes, 40:60 (v/v)). Reagent 1 (141 mg, 0.540 mmol, 1.80 equiv) and FeSO 4 ·7H 2 O (0.8 mg, 3 µmol, 0.01 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M) and nitrobenzene (36.9 mg, 30.8 µL, 0.300 mmol, 1.00 equiv). The reaction mixture was stirred at 60 °C for 45 min. The orange reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated.  Reagent 1 (141 mg, 0.540 mmol, 1.80 equiv), FeSO 4 ·7H 2 O (0.8 mg, 3 µmol, 0.01 equiv), and phenyl methylsulfone (46.9 mg, 0.300 mmol, 1.00 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M). The reaction mixture was stirred at 60 °C for 90 min. The orange reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue indicated a ratio of 3a:3b:3c = 4.7:1.7:1.0 by integrating the signal of 3a at 6.87 ppm, the signal of 3b at 6.76 ppm, and the signal of 3c at 6.68 ppm. The residue was purified by column chromatography on silica gel eluting with a gradient solvent system, starting with a solvent mixture of diethyl ether/pentane (50:50 (v/v)) and finishing with diethyl ether.
The second fraction was further purified by preparative TLC using a solvent system of ethyl acetate/pentane (30:70 (v/v)) to give 3a (14.6 mg) and 3c (3.6 mg) as separate samples. Characterization data matched previously reported data for 3b 13a and 3c 13b .

3-(Methylsulfonyl)aniline (3a)
Rf = 0.17 (ethyl acetate/hexanes, 40:60 (v/v)).  The reaction mixture was stirred at 60 °C for 30 min. The red reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue indicated a ratio of 4a:4b:4c = 2.0:1.6:1.0 by integrating the signal of 4a at 7.00 ppm, the signal of 4b at 6.64 ppm, and the signal of 4c at 6.76 ppm. The residue was purified by column chromatography on silica gel eluting with a gradient solvent system, starting with a solvent ELECTRONIC SUPPLEMENTARY INFORMATION S14 mixture of diethyl ether/pentane (20:80 (v/v)) and finishing with diethyl ether/pentane (50:50 (v/v)). Purification afforded 2-aminobenzonitrile (4c) in one fraction (6.4 mg) and 3-aminobenzonitrile (4a) and 4aminobenzonitrile (4b) in a second fraction (25.2 mg) for a combined yield of 31.6 mg (89% yield). The second fraction was further purified by preparative TLC using a solvent system of acetone/pentane (10:90 (v/v)) to give 4a (10.0 mg) and 4b (9.1 mg) as separate samples. Characterization data matched previously reported data for 4a, 4b, and 4c. 14

3-Aminobenzonitrile (4a):
Rf = 0.50 (ethyl acetate/hexanes, 40:60 (v/v)).   3 µmol, 0.01 equiv), and 1,4dibromobenzene (70.8 mg, 0.300 mmol, 1.00 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M). The reaction mixture was stirred at 60 °C for 15 min. The red reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. The residue was purified by column chromatography on silica gel eluting with a solvent system of ethyl acetate/hexane  Reagent 1 (118 mg, 0.450 mmol, 1.50 equiv) and FeSO 4 ·7H 2 O (0.8 mg, 3 µmol, 0.01 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M) and ethyl 2-thiophenecarboxylate (46.9 mg, 40.3 µL, 0.300 mmol, 1.00 equiv). The reaction mixture was stirred at 40 °C for 15 min. The red reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue indicated a ratio of 6a:6b:6c = 5.9:1.8:1.0 by integrating the signal of 6a at 6.09 ppm, the signal of 6b at 6.39 ppm, and the signal of 6c at 6.55 ppm. The residue was purified by column chromatography on silica gel eluting with a gradient solvent system, starting with a solvent mixture of diethyl ether/pentane (5:95 (v/v)) and finishing with diethyl ether/pentane (50:50 (v/v)).* Purification afforded ethyl 3-aminothiophene-2-carboxylate (6c) in one fraction (2.9 mg), ethyl 5-aminothiophene-2-carboxylate (6a) in a second fraction (18.0 mg), and ethyl 4aminothiophene-2-carboxylate (6b) in a third fraction (6.6 mg) for a combined yield of 30.5 mg (59% yield). *Column specifications: diameter = 3 cm, packing height = 7 cm; total amount of silica used: 18 g; total ELECTRONIC SUPPLEMENTARY INFORMATION S16 amount of eluent mixture to collect all products: 610 mL.
Reagent 1 (235 mg, 0.900 mmol, 3.00 equiv) was then added to the vial. The reaction mixture was stirred at 60 °C for 45 min. The red brown reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue indicated a ratio of 11a:11b = 10:1.0 by integrating the ELECTRONIC SUPPLEMENTARY INFORMATION S21 signal of 11a at 7.25 ppm and the signal of 11b at 6.63 ppm. The residue was purified by column chromatography on basified silica gel (NH 4 OH) eluting with a gradient solvent system, starting with a solvent mixture of methanol/dichloromethane (1:99 (v/v)) and finishing with methanol/dichloromethane (5:95 (v/v)).

3-Aminomoclebomide (11a):
Rf = 0.18 (methanol/dichloromethane, 5:95 (v/v)).  vial, followed by HFIP (1.5 mL, c = 0.2 M). The reaction mixture was stirred at 60 °C for 120 min. The orange reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue indicated a ratio of 12a:12b = 3.3:1.0 by integrating the signal of 12a at 6.91 ppm and the signal of 12b at 6.58 ppm.
The residue was purified by column chromatography on silica gel eluting with a gradient solvent system, starting with a solvent mixture of diethyl ether/pentane (20:80 (v/v)) and finishing with diethyl ether.

Comparison to other amination methods
In comparison to other reported modern amination methods that use an ammoniumyl radical precursor, our method is applicable to a much broader electronic scope of aromatic substrates. The most relevant conditions are compared in Table S1. Our method also works in the absence of an iron salt, albeit with longer reaction times. When other reported methods are used in the absence of an iron salt, essentially no reaction is observed. Bromobenzene was used to compare reactivity in the absence of iron (Table S2). µmol, 0.01 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M) and bromobenzene (47.1 mg, 31.5 µL, 0.300 mmol, 1.00 equiv). The reaction mixture was stirred at 60 °C for 30 min. The purple ELECTRONIC SUPPLEMENTARY INFORMATION S26 reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. Characterization data matched previously reported data for S4a, S4b, and S4c. 9a This work (Table S2). Reagent 1 (86.2 mg, 0.330 mmol, 1.10 equiv) was added to a flame-dried Schlenk tube, followed by HFIP (1.5 mL, c = 0.2 M) and bromobenzene (47.1 mg, 31.5 µL, 0.300 mmol, 1.00 equiv).
The reaction mixture was stirred at 60 °C for 16 h. The brown reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . An ethyl acetate solution containing 1,3,5trimethoxybenzene (0.1 mmol) was added as an internal standard, and the layers were separated. The aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue was taken, and yield was determined based on integration against the internal standard to be 83%. (Table S2). Reagent 1 (313 mg, 1.20 mmol, 4.00 equiv) was added to a flame-dried Schlenk tube, followed by degassed MeCN/H 2 O (2:1, 900 µL, c = 0.33 M) and bromobenzene (47.1 mg, 31.5 µL, 0.300 mmol, 1.00 equiv). The reaction mixture was stirred at 23 °C for 16 h. The colorless reaction mixture was diluted with 1.0 M NaOH (aq) (10 ml) and was poured into a separatory funnel. An ethyl acetate solution containing 1,3,5-trimethoxybenzene (0.1 mmol) was added as an internal standard, and the aqueous layer was extracted with dichloromethane (3 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue was taken, and yield was determined based on integration against the internal standard to be 0%. (Table S2). Hydroxylamine-O-sulfonic acid (33.9 mg, 0.300 mmol, 1.00 equiv) was added to a flamedried Schlenk tube, followed by AcOH/H 2 O (2:1, 500 µL, c = 0.60 M) and bromobenzene (47.1 mg, 31.5 µL, 0.300 mmol, 1.00 equiv). The reaction mixture was stirred at 40 °C for 16 h. The colorless reaction mixture was basified with 1.0 M NaOH (aq) (~10 ml) and was poured into a separatory funnel. An ethyl acetate solution containing 1,3,5-trimethoxybenzene (0.1 mmol) was added as an internal standard, and the aqueous layer was extracted with diethyl ether (3 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue was taken, and yield was determined ELECTRONIC SUPPLEMENTARY INFORMATION S27 based on integration against the internal standard to be 3%.

Effect of iron source, oxygen, and light on the amination reaction Effect of iron and oxygen
The presence of both iron and oxygen has an effect on the reaction time with the shortest reaction times being observed when both are present. 1,4-Dibromobenzene was used to show the effect of iron and oxygen.
See below (pg. S31) for a description of reaction setup and trace metal analysis for metal-free experiments.

Effect of iron source
Multiple iron(II) and iron(III) sources were observed to promote the amination reaction effectively, and the reaction also works in the absence of iron for multiple substrates. Nitrobenzene was used to show the effect of the iron source. were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M) and nitrobenzene (36.9 mg, 30.8 µL, 0.300 mmol, 1.00 equiv). The reaction mixture was stirred at 60 °C for 45 min. The orange reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . The layers were separated, and the aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated.

Effect of light
Ambient light was found to have no effect on the results of the amination reaction. A reaction run in the dark (wrapped in foil) gave similar results to the standard reaction conditions. Reagent 1 (118 mg, 0.450 mmol, 1.50 equiv), FeSO 4 ·7H 2 O (0.8 mg, 3 µmol, 0.01 equiv), and 1,4dibromobenzene (70.8 mg, 0.300 mmol, 1.00 equiv) were added to a 4-mL vial, followed by HFIP (1.5 mL, c = 0.2 M). The reaction mixture was stirred at 60 °C for 18 min. The red reaction mixture was allowed to cool to room temperature and was concentrated. The residue was dissolved in 10 mL ethyl acetate and poured into a separatory funnel containing 10 mL sat. aq. Na 2 CO 3 . An ethyl acetate solution containing 1,3,5trimethoxybenzene (0.1 mmol) was added as an internal standard, and the layers were separated. The aqueous layer was extracted with ethyl acetate (2 × 10 mL). The combined organic layers were dried over sodium sulfate, filtered and concentrated. A 1 H NMR spectrum of the residue was taken, and yield was determined based on integration against the internal standard.

Trace metal analysis
Reactions performed in the absence of FeSO 4 ·7H 2 O were conducted with the following precautions: All glassware and stirbars were washed with aqua regia solution, rinsed with deionized water and dried in an oven. Solids were handled with glass pipettes. Solvents were distilled before use. dibromobenzene (70.8 mg, 0.300 mmol, 1.00 equiv) were added to a flame-dried Schlenk tube. The vessel was evacuated and backfilled with nitrogen three times. Distilled, degassed HFIP (1.5 mL, c = 0.2 M) was then added. The reaction mixture was stirred at 60 °C for 17 hr. The red reaction mixture was allowed to cool to room temperature. A small aliquot was taken to determine conversion, which was always >90% as judged by 1 H NMR. The bulk of the reaction mixture was transferred to a glass vial, sealed and sent out for ICP-MS analysis. Duplicate samples were analyzed in this manner. One contained <1 ppb Fe and 60 ppb Cu, while the other contained <1 ppb Fe and <1 ppb Cu.

Consumption studies of reagent 1
Reagent 1 is consumed to generate methanesulfonic acid (MsOH) even when an arene substrate is not added to the reaction. MsOH is formed faster in the presence of FeSO 4 ·7H 2 O and/or residual moisture.

S6 S1
Compound S1 (1.00 g, 4.73 mmol, 1.00 equiv) was dissolved in anhydrous diethyl ether (24 mL) in a flamedried round bottom flask. The flask was evacuated and backfilled with nitrogen, then cooled in a water-ice bath. Nonafluorobutanesulfonic acid (1.42 g, 784 µL, 4.73 mmol, 1.00 equiv) was added. The reaction mixture was allowed to stir at room temperature for 2 h, during which time a colorless precipitate formed.
Heptane (15 mL) was added to the flask. The colorless solid was collected on a Buchner funnel, rinsed with heptane (10 mL) and dried under high vacuum to give 1.72 g of the title compound as a colorless solid (88% yield). 13 C NMR peaks for the nonaflate counterion were not observed due to the complex C-F splitting. The presence of a nonaflate counterion was confirmed by 19 F NMR spectroscopy.

Failed substrates
All reactions have been carried out according to the general procedure for substrate amination. In case of low conversion, the reaction has not been further investigated with respect to products.    Figure S4. Optimized structure of 1. DFT results for HFIP Figure S5. Optimized structure of HFIP. 1·HFIP (OMs) is found to be 5.9 kcal/mol higher in energy than 1 and a free HFIP molecule.