Supporting Information

Asymmetric Grignard Synthesis of Tertiary Alcohols through Rational

Ligand Design
Bartosz Bieszczad and Declan G. Gilheany*

anie_201610462_sm_miscellaneous_information.pdf
 Supplementary Information

Table of Contents:

1. General 3
2. Preparation of ligands 4
3. Procedures for asymmetric Grignard synthesis of tertiary alcohols,
including ligand recovery 8
4. Tertiary Alcohol Products from Asymmetric Grignard Synthesis 11
5. Asymmetric synthesis of α-tocopherol precursor 23
6. Comparison of methods for asymmetric nucleophilic addition to ketones 24
7. Mechanistic considerations 27
8. References 30
List of Abbreviations

AcOH Acetic acid

Ar Aryl
DCM Dichloromethane
DMF Dimethylformamide
MeOH Methanol
NBS N-Bromosuccinimide
r.t. room temperature
STAB Sodium triacetoxyborohydride
TBACl Tetrabutylammonium chloride
TFA Trifluoroacetic acid
TFAA Trifluoroacetic anhydride
THF Tetrahydrofuran

2  

 
1. General

All chemicals were supplied by Aldrich apart from natural phytol, which was supplied by
Wuhan ChemFaces Biochemical Co. and both enantiomers of cyclohexane-1,2-diamine-
tartrate provided by Arran Chemical Company, Athlone, Ireland. (7R,11R)-1-(2,5-
Dimethoxy-3,4,6-trimethylphenyl)-7,11,15-trimethylhexadecan-3-one (the ketone precursor
of α-tocopherol precursor A (Scheme 2 in main text) was prepared in seven steps from
natural phytol.1 Diethyl ether, toluene and THF were dried prior using a Grubbs type
solvent system (Innovative Technologies, Model: PS-400-3-MD), other solvents and
reagents were used without further purification. Water content of dried solvents was
determined before use by Karl Fisher titration, on Aquamax v2.6i KF coulometric system.
All reactions that required air/moisture free conditions were carried out under nitrogen
using standard Schlenk manifold techniques. Microwave assisted reactions were
performed using CEM Discover microwave instrument (S-series).

All NMR analysis was performed using 300, 400 or 500 MHz Varian spectrometers. NMR
samples were made up in CDCl3. 1H and 13C NMR chemical shifts are relative to TMS.

HPLC analysis was performed using Agilent Technologies 1260 Infinity system with auto-
sampler, combined with Agilent UV-vis detector. Chiral columns Chiralpak® IA, IB, IC, and
Chiracel® OJH of an internal diameter 4.6 mm, 250 mm length, particle size of 5 µm,
coupled to a guard column of internal diameter of 4.6 mm and length of 50 mm were used
to separate enantiomers. Columns were mounted in an Agilent column switcher device.
HPLC was performed in normal phase, with heptane and ethanol as solvents. Both
solvents were filtered through an acrodisc CR 13 syringe filter with 0.2 µm PTFE before
injection on the column.

Supercritical Fluid Chromatography (SFC) was performed using ACQUITY UPC2 system
with auto-sampler, combined with PDA Detector Type UPLC eLambda 800 nm. Chiral
column Chiralpak® IA-3 of internal diameter of 4.6 mm, 100 mmL with particles size of 3
µm was used to separate diastereoisomers in a single case with methanol as a co-solvent.

HLPC and SFC samples were prepared by diluting crude mixture with heptane in 1 mL
sample vial to approximately 1 mg/mL.

Flash chromatography was performed using silica 60, 40-63 µm of diameter, supplied by
Apollo Scientific. TLC was done with pre-coated silica 60 plates F-254. Realisation of
plates was done under UV light or developed by iodine vapour.
3  

 
2. Preparation of Ligands

Resolution of (±)-trans-1,2-diaminocyclohexane

(±)-trans-1,2-Diaminocyclohexane was resolved using L-(+)-tartaric acid according to the

procedure previously developed in our group2 (optimised method C in ref 1).
To a water (100 mL) / methanol (800 mL) solution of racemic diamine (57 g, 500 mmol)
glacial acetic acid was added with stirring (37.5 mL, 39 g, ~650 mmol). A solution of L-(+)-
tartaric acid (37.55 g, 250 mmol) in methanol was added dropwise and addition funnel was
rinsed with 50 mL of methanol. The mixture was refluxed for 24 h, cooled to room
temperature, the resulting slurry was filtered, washed with methanol (3 x 100 mL) and
dried under high vacuum. The product – (R,R)-1,2-diaminocyclohexane (L)-(+)-tartrate was
recrystallized from minimum amount of hot water to yield colourless crystals (63.5 g, 240.4
mmol, 48% yield, >99.8% ee [HPLC]). Enantiomeric excess was determined by
derivatisation with m-toluoyl chloride using the previously developed protocol.2

All mother liquors combined (approximately 1.5 L) were treated with conc. HCl (50 mL,
~650 mmol) and heated to reflux. Acetone (2.5 L) was added in portions (250 mL) over 1 h
under reflux, which was continued overnight. Mixture was cooled to room temperature.
The white precipitate was filtered, washed with acetone (3 x 100 mL), dried under high
vacuum. Product - (S,S)-1,2-diaminocyclohexane hydrochloride was recrystallized from
minimum amount of hot water to yield colourless, crystalline powder (45.6 g, 137.25 mmol,
28% yield, >99.8% ee [HPLC]).

HPLC conditions: Column: IA, Mobile phase: heptane : ethanol (97:3), enantiomers at
19.11 min (R) and 21.72 min (S). Flow rate: 1 ml / min.

1
H NMR (300 MHz, CDCl3): δ 4.12 (s, 2H), 3.22-3.13 (m, 2H), 2.05-1.93 (m, 2H), 1.74-1.27
(m, 2H), 1.42-1.28 (m, 2H), 1.26-1.10 (m, 2H) ppm.

4  

 
Release of free (1R,2R)-cyclohexane-1,2-diamine

Free amine was released using a modification of the literature procedure.3 (1R,2R)-
Cyclohexane-1,2-diamine-tartrate (30.2 g, 114 mmol, 1 eq.) was suspended in DCM (250
mL) and NaOH (11 g, 276 mol, 4.8 eq.), dissolved in water : saturated brine mixture (152
mL, 1:1 ratio) was added slowly. The mixture was left to stir at room temperature for 1h
under a stream of nitrogen. The organic layer was separated and the aqueous layer was
washed with DCM (100 mL x 4). The combined organic layers were dried using anhydrous
Na2SO4 and the solvent was evaporated by rotary evaporator to yield (1R,2R)-
cyclohexane-1,2-diamine as crystalline, white solid (8.4 g, 65% yield). NMR shifts match
those in the literature.4

1
H NMR (300MHz, CDCl3): δ 2.32-2.21 (m, 2H), 1.89-1.78 (m, 2H), 1.72-1.64 (m, 2H), 1.43
13
(s, 4H), 1.36-1.21 (m, 2H), 1.21-1.01 (m, 2H); C NMR (101 MHz, CDCl3): δ 74.6, 33.1,
24.5.

Mono-BOC protection of (1R,2R)-cyclohexane-1,2-diamine

Tert-butyl (1R,2R)-2-aminocyclohexylcarbamate was prepared according to the literature

procedure.5 HCl (37% aq., 7.6 mL, 1.0 eq.) was dissolved in methanol (80 mL) and cooled
to 0 ˚C. To this, (1R,2R)-cyclohexane-1,2-diamine (8.4 g, 74 mmol, 1.0 eq.) was dissolved
in methanol (10 mL) was added dropwise. The mixture was allowed to warm to r.t. Di-tert-
butyl dicarbonate (16.0 g, 74 mmol, 1.0 eq.) was dissolved in methanol (90 mL) and added
slowly to the diamine hydrochloride mixture dropwise over 2h and the mixture was stirred
for additional 2h. The solvent was then removed under reduced pressure and washed by
5  

 
diethyl ether (3 x 100 mL). The remaining solids were then treated with 2M NaOH (250
mL) and the free diamine was extracted with DCM (2 x 150 mL). The organic extracts were
combined, washed with DIW (2 x 100 mL), brine (100 mL) and dried over anhydrous
Na2SO4. The solvent was removed to yield a white solid. (11.25 g, 71% yield).

1
H NMR (300MHz, CDCl3): δ 4.56 (s, 1H), 3.13 (s, 1H), 2.29 (broad, 1H), 2.06-1.94 (m,
13
2H), 1.67-1.53 (m, 2H), 1.43 (s, 9H), 1.15-1.07 (m, 4H), 1.07-0.98 (m, 2H); C NMR (101
MHz, CDCl3,): δ 156.3, 79.2, 57.8, 55.9, 35.4, 33.2, 28.6, 25.4, 25.3. (Values consistent
with the literature data.5)

General Procedure for Ligand Synthesis

Tert-butyl (1R,2R)-2-aminocyclohexylcarbamate (1.0 eq) was dissolved in MeOH (15 vol.)

and salicylaldehyde (1.0 eq.) was added to the solution. The mixture was stirred for 5h and
then solid sodium borohydride (2.0 eq.) was added slowly. The mixture was stirred for
another 12h and quenched with NaHCO3 (aq., sat., 10 vol.). The product was extracted
with diethyl ether (3 x 20 vol.) and the organic phase was washed with water (3 x 20 vol.).
The solvent was removed under vacuum to yield a crude product which was dissolved in
DCM / TFA (1:2, 10 vol.) and stirred for 12h. The mixture was basified with concentrated
KOH to pH 12 (with cooling) and the product was extracted with diethyl ether (3 x 20 vol.)
and washed with water (3 x 20 vol.) and dried over anhydrous sodium sulfate. The solvent
was removed under vacuum to yield solids, which were dissolved in DCM (20 vol.). Acetic
acid (glacial, 0.5 vol.) was added, followed by formaldehyde (aq., 37%, 2.0 vol.) and STAB
(6.0 eq.). The mixture was stirred for 12h at r.t., quenched with NaHCO3 (aq., sat., 10 vol.).
The product was extracted with diethyl ether (3 x 15 vol.) and washed with water (3 x 15
vol.). The solvent was removed under vacuum to yield a crude ligand 1a-1b. The product
was purified by silica column chromatography, using cyclohexane : ethyl acetate mixture

6  

 
(80:20) as an eluent and recrystallized from hot IPA/water (2:1) to yield a purified ligand
1a-1b.

2,4-di-tert-butyl-6-((((1R,2R)-2-(dimethylamino)cyclohexyl)(methyl)
amino)methyl)phenol (1a)

Prepared form 3,5-di-tert-butyl-2-hydroxybenzaldehyde. White,

crystalline solids (2.44 g, 77% purified yield); 1H NMR (400 MHz,
CDCl3): δ 7.19 (d, J = 2.3 Hz, 1H), 6.83 (d, J = 2.2 Hz, 1H), 3.92 (d,
J = 12.9 Hz, 1H), 3.21 (broad, 1H), 2.67-2.32 (m, 2H), 2.29 (s, 6H),
2.20 (s, 3H), 2.07-1.96 (m, 2H), 1.86-1.72 (m, 2H), 1.43 (s, 9H),
13
1.26 (s, 9H), 1.24-0.96 (m, 4H). C NMR (101 MHz, CDCl3): δ
154.5, 139.0, 135.3, 124.4, 123.1, 122.5, 64.0, 63.8, 54.5, 39.3, 35.0, 34.0, 31.8, 29.6,
26.9, 25.7, 23.9, 22.1. NMR shifts consistent with the literature.6

2-((((1R,2R)-2-(dimethylamino)cyclohexyl)(methyl)amino)methyl)-6-
(trifluoromethyl)phenol (1b)

Prepared from 2-(methoxymethoxy)-3-(trifluoromethyl)

benzaldehyde, synthesised using modified Christensen
procedure.7 White, crystalline solids (790 mg, 65% purified yield);
1
H NMR (300 MHz, CDCl3): δ 7.43 (d, J = 7.3 Hz, 1H, H1), 7.13
(d, J = 6.9 Hz, 1H, H2), 6.72 (apparent t, J = 7.2 Hz, 1H, H3),
3.95 (d, J = 12.7 Hz, 1H, H4), 3.09 (broad, 1H, H5), 2.80 – 2.43
(m, 2H, H6), 2.30 (s, 6H, H8), 2.21 (s, 3H, H7), 2.15 – 1.91 (m, 2H, H9), 1.91 – 1.73 (m,
13
2H, H10), 1.39 – 1.09 (m, 4H, H11); C NMR (101 MHz, CDCl3): δ 157.0, 133.5, 126.1,
125.9, 123.2, 117.8, 116.4, 64.3, 64.0, 51.6 (broad), 38.9 (broad), 26.9, 25.6, 23.4, 21.9;
19
FMR (282 MHz, CDCl3): δ -62.27 (s); Elemental analysis: calc., found for C17H25F3N2O):
C (61.80, 61.89), H (7.63, 7.56), N (8.48, 8.31).

7  

 
3. Procedures for asymmetric Grignard synthesis of tertiary alcohols, including
ligand recovery.

General Procedure A for asymmetric Grignard synthesis of tertiary alcohols (0.1

mmol scale)

Ligand (0.1 mmol, 1.0 eq.) was added to a flame dried 15 ml Schlenk flask, followed by
addition of dry toluene (0.5 mL) and solution of ketone in toluene (0.25 mL of 0.4 M, 0.1
mmol, 1.0 eq.). The mixture was agitated at r.t. for 5 min then cooled to -78 ˚C in acetone /
dry ice bath. Grignard reagent in toluene/ether (6:1), (0.42 M, 0.5 ml, 0.21 mmol, 2.1 eq.)
was added slowly dropwise. The mixture was agitated at -78 ˚C and monitored by HPLC
until no more product formation was detected. The reaction was quenched with IPA/water
(3:1, 2 mL), followed by sat. NH4Cl (2 mL) and heptane (2 mL) and mixture was allowed to
warm to r.t. The aqueous layer was separated, washed with heptane (3 x 5 mL). The
organic phases were combined. Ligand was recovered using procedure A. The organic
fractions were combined washed with water (3 x 5mL), brine (5 mL) and then dried over
anhydrous MgSO4. The solvent was removed by rotary evaporator. The crude product was
purified by flash chromatography using diethyl ether : cyclohexane mixture as eluent.8

Rapid Ligand Recovery (Small scale - Procedure A)

To the organic crude reaction mixture (1 mL, 0.1 mmol scale) a few drops of acetic acid
(glacial) was added. The mixture was passed through a 2-3 cm silica column in a glass
Pasteur pipette (Figure 1), followed by 5 mL of cyclohexane : ethyl acetate (50:50). The
organic fractions were combined and the solvent was evaporated to yield a crude product.
The ligand was recovered from the silica column by elution with cyclohexane : ethyl
acetate : triethylamine (50:40:10) wash. The solvent was removed by rotary evaporator to
yield a crude ligand, which was further purified by recrystallization from hot water : IPA
(1:1) to yield 75% - 85% of pure, recovered ligand which can be reused in subsequent
reactions.

8  

 
 Figure 1. Rapid set-up for removal & recovery of ligand from the crude reaction mixture

General Procedure B for asymmetric Grignard synthesis of tertiary alcohol (1 g

scale) exemplified for (S)-2-phenyl-2-butanol

Ligand L1 (3.14 g, 8.32 mmol, 1.0 eq.) was dissolved in 30 mL of dry toluene in a
previously flamed dried 250 mL Schlenk flask. Acetophenone (20.8 mL of 0.4 M solution in
dry toluene, 8.32 mmol, 1.0 eq.) was added and the mixture was cooled to -78 ˚C in
acetone/dry ice bath. Ethylmagnesium bromide in toluene/ether (4:1), (0.75 M, 23.3 ml,
17.47 mmol, 2.1 eq.) was added slowly dropwise. The mixture was agitated at -78 ˚C and
monitored by HPLC until no more product formation was detected. The reaction was
quenched with IPA/water (3:1, 50 mL), followed by sat. NH4Cl (50 mL) and heptane (50
mL) and mixture was allowed to warm to r.t. The aqueous layer was separated, washed
with heptane (3 x 50 mL). Organic phases were combined. The ligand was recovered
using Recovery Procedure B. Organic phases were washed with water (3 x 150 mL), brine
(150 mL), dried over anhydrous MgSO4. The solvent was removed by rotary evaporator.
The crude product was purified by flash chromatography using diethyl ether : cyclohexane
mixture as eluent to yield (S)-2-phenylbutan-2-ol (937 mg, 75% yield, 78% ee).

9  

 
Standard ligand recovery procedure (B). The organic crude reaction mixture (8.32 mmol
scale) was washed with aq. acetic acid (10%, 4 x 250 mL) until the ligand was no longer
visible on TLC. Washes were combined, neutralized with aq. NaHCO3 (sat.) and the ligand
was extracted with toluene (3 x 250 mL). The combined organic phases were washed with
water (2 x 250 mL), brine (250 mL) and dried over anhydrous sodium sulfate. Solvent was
removed by rotary evaporator to yield a crude ligand, which was further purified by
recrystallization from hot water : IPA (1:1) to yield 75% - 85% of pure, recovered ligand
which can be reused in subsequent reactions.

10  

 
4. Tertiary Alcohol Products from Asymmetric Grignard Synthesis

(S)-2-phenylbutan-2-ol and (R)-2-phenylbutan-2-ol

Colourless liquid (77% yield); 1H NMR (400 MHz, CDCl3) δ

7.50-7.42 (m, 2H), 7.41-7.31 (m, 2H), 7.30-7.22 (m, 1H),
2.06 (s, 1H), 1.94-1.73 (m, 2H), 1.57 (s, 3H), 0.82 (dd, J =
7.7, 7.2 Hz, 3H); 13C NMR (101 MHz, CDCl3): 147.7, 128.2, 126.6, 124.9, 74.8, 36.7, 29.7,
8.3; HPLC conditions: Column: OJH, Mobile phase: heptane : ethanol (95:5), enantiomers
at 11.72 min (R) and 13.94 min (S); Flow rate: 1 ml / min; [α]D20 = −5.2 (c=1.0 in CHCl3)
for (S) enantiomer; NMR shifts and optical rotations are consistent with the literature.9

(S)  -­‐  Table  2,  Entry  1   (R)  -­‐  Table  2,  Entry  2  

11  

 
 (R)  -­‐  Table  2,  Entry  6    

(R)  -­‐  Table  2,  Entry  8  

2-(4-methoxyphenyl)butan-2-ol

Colourless oil (53% yield); 1H NMR (300 MHz, CDCl3) δ 7.33 (d, J = 9.0
Hz, 2H), 6.85 (d, J = 9.0 Hz, 2H), 3.80 (s, 3H), 1.72-1.91 (m, 2H), 1.53
(s, 3H), 0.78 (apparent t, J = 7.4 Hz, 3H); 13C NMR (101 MHz, CDCl3): δ
157.9, 139.7, 125.9, 113.2, 74.5, 55.2, 36.7, 29.5, 8.5. HPLC
conditions: Column: IB, Mobile phase: heptane : ethanol (97:3), enantiomers at 19.24 min
and 20.27 min. Flow rate: 1 ml / min. NMR shifts are consistent with the literature.9

12  

 
2-(4-methoxyphenyl)-3-methylbutan-2-ol

Colourless oil (60% yield); 1H NMR (400 MHz, CDCl3) δ 7.34 (m, 2H),
HO
6.86 (m, 2H), 3.79 (s, 3H), 1.99 (m, 1H), 1.68 (s, 1H), 1.50 (s, 3H),
13
0.85 (d, J = 6.9 Hz, 3H), 0.82 (d, J = 6.9 Hz, 3H); C NMR (101 MHz,
O
CDCl3): δ 157.9, 139.8, 126.4, 113.1, 76.5, 55.2, 38.7, 26.4, 17.4, 17.2.
HPLC conditions: Column: IB, Mobile phase: heptane : ethanol (97:3), enantiomers at 6.71
min and 7.12 min. Flow rate: 1 ml / min. NMR shifts are consistent with the literature.10

2-(4-(trifluoromethyl)phenyl)butan-2-ol

Colourless liquid (99% yield); 1H NMR (400 MHz, CDCl3): δ 7.73-7.37

(m, 4H), 1.99-1.70 (m, 2H), 1.54 (s, 3H), 0.80 (apparent t, J = 7.4 Hz,
13
3H); C NMR (101 MHz, CDCl3): δ 151.7, 125.3, 125.1, 125.0, 125.0,
19
125.0, 120.2, 74.8, 36.6, 29.8, 8.1; F NMR (376 MHz, CDCl3): δ -62.42 (s); HPLC
conditions: Column: OJH, Mobile phase: heptane : ethanol (99:1), enantiomers at 13.09
min and 14.21 min. Flow rate: 1 ml / min. NMR shifts are consistent with the literature.11

13  

 
2-(o-tolyl)butan-2-ol

Colourless liquid (71% yield); 1H NMR (300 MHz, CDCl3): δ 7.52-7.43 (m,
1H), 7.53-7.32 (m, 3H), 2.61 (s, 3H), 2.03-1.94 (m, 2H), 1.71 (s, 3H), 0.90
13
(apparent t, J = 7.4 Hz, 3H); C NMR (101 MHz, CDCl3): δ 144.6, 135.4,
132.4, 126.7, 126.7, 125.3, 76.0, 34.4, 28.7, 22.2, 8.4; HPLC conditions:
Column: IA, Mobile phase: heptane : ethanol (98:2), enantiomers at 7.62 min and 9.60
min. Flow rate: 1 ml / min. NMR shifts are consistent with the literature.12

14  

 
5-ethyl-6,7,8,9-tetrahydro-5H-benzo[7]annulen-5-ol

White solid (73% yield); 1H NMR (400 MHz, CDCl3): δ 7.42-7.33 (m, 1H),
7.33-7.05 (m, 3H), 3.09-2.72 (m, 2H), 2.44-2.02 (m, 2H), 1.95-1.90 (m, 2H),
13
1.90-1.86 (m, 4H), 0.97 (apparent t, J = 7.5 Hz, 3H); C NMR (101 MHz,
CDCl3): δ 143.2, 128.2, 126.6, 124.9, 122.8, 118.8, 84.1, 39.4, 37.6, 33.0, 29.5, 20.8, 8.6;
HPLC conditions: Column: OJH, Mobile phase: heptane : ethanol (96:04), enantiomers at
10.22 min and 11.12 min. Flow rate: 1 ml / min. NMR shifts are consistent with the
literature.13

1-ethyl-1,2,3,4-tetrahydronaphthalen-1-ol

White solid (57% yield); 1H NMR (400 MHz, CDCl3): δ 7.64-7.46 (m, 1H),
7.44-7.02 (m, 3H), 2.92-2.69 (m, 2H), 2.59-2.46 (m, 2H), 2.17-2.01 (m, 2H),
13
2.01-1.77 (m, 2H), 0.95 (apparent t, J = 7.5 Hz, 3H); C NMR (101 MHz,
CDCl3): δ 142.2, 136.9, 128.9, 127.0, 126.3, 126.2, 72.7, 35.3, 34.9, 29.9, 19.7, 8.6; HPLC
conditions: Column: IA, Mobile phase: heptane : ethanol (99:1), enantiomers at 19.80 min
and 24.69 min. Flow rate: 1 ml / min. NMR shifts are consistent with the literature.14

15  

 
(S)-1-ethyl-2,3-dihydro-1H-inden-1-ol

Colourless needles (52% yield); 1H NMR (300 MHz, CDCl3): δ 7.44 – 7.32
(m, 1H), 7.32 – 7.19 (m, 3H), 3.16 – 2.74 (m, 2H), 2.42 – 2.06 (m, 2H),
13
2.06 – 1.69 (m, 2H), 0.98 (apparent t, J = 7.4 Hz, 3H); C NMR (101MHz,
CDCl3): 147.4, 143.2, 128.4, 126.7, 125.0, 122.9, 84.2, 39.5, 33.1, 29.5, 8.7; ;  [α]D20  =  −3.5  

(c=1.0   in   CHCl3)   for   (S)   enantiomer; HPLC conditions: Column: IA, Mobile phase:
heptane : ethanol (99:1), enantiomers at 12.02 min and 12.85 min. Flow rate: 1 ml / min.
NMR shifts and optical rotations are consistent with the literature.14

16  

 
4-ethylchroman-4-ol

Colorless solid (45% yield); 1H NMR (300 MHz, CDCl3): δ 7.60-7.33 (m,
1H), 7.27-7.08 (m, 1H), 7.02-6.74 (m, 2H), 4.33-4.16 (m, 2H), 2.28-1.80 (m,
13
4H), 0.92 (apparent t, J = 7.5 Hz, 3H); C NMR (75 MHz, CDCl3): δ 154.5,
129.0, 127.5, 126.4, 120.7, 117.2, 68.7, 63.0, 34.0, 34.0, 8.5; HPLC
conditions: Column: OJH, Mobile phase: heptane : ethanol (97:03), enantiomers at 14.62
min and 23.26 min. Flow rate: 1 ml / min; HRMS (m/z): [M]+: calcd. for C11H14O2: 178.0994;
found: 178.0992.

1-methyl-2,3-dihydro-1H-inden-1-ol

White solid (86% yield); 1H NMR (400 MHz, CDCl3) δ 7.42-7.31 (m, 1H), 7.30-
13
7.27 (m, 3H), 3.11-2.75 (m, 2H), 2.31-2.07 (m, 2H), 1.59 (s, 3H); C NMR
(101 MHz, CDCl3): δ 148.3, 142.7, 128.2, 126.8, 125.0, 122.2, 81.2, 42.3,
29.4, 27.3; HPLC conditions: Column: IB, Mobile phase: heptane : ethanol (96:4),
enantiomers at 6.11 min and 6.38 min. Flow rate: 1 ml / min. NMR shifts are consistent
with the literature.14

17  

 
1-isobutyl-2,3-dihydro-1H-inden-1-ol

Colourless solid (49% yield) ; 1H NMR (400 MHz, CDCl3): δ 7.42-32 (m,
1H), 7.32-7.08 (m, 3H), 3.14-2.72 (m, 2H), 2.42-2.03 (m, 2H), 1.97-1.81
13
(m, 2H), 1.76-1.61 (m, 1H), 1.08-0.83 (m, 6H); C NMR (101 MHz,
CDCl3): δ 148.3, 142.8, 128.1, 126.6, 124.9, 122.8, 84.1, 48.6, 40.4, 29.6, 24.9, 24.8, 24.4;
HRMS (m/z): [M]+: calcd. for C13H18O: 191.1435; found: 191.1470; HPLC conditions:
Column: IA, Mobile phase: heptane : ethanol (96:4), enantiomers at 9.19 min and 11.08
min. Flow rate: 1 ml / min.

18  

 
1-cyclopentyl-2,3-dihydro-1H-inden-1-ol

White solid (80% yield); 1H NMR (400 MHz, CDCl3): δ 7.40–7.28 (m, 1H),
7.28–7.13 (m,3 H), 3.05–2.76 (m, 2H), 2.48–2.01 (m, 4H), 1.87–1.74 (m,
13
1H), 1.71–1.47 (m, 5H), 1.36–1.26 (m, 1H); C NMR (101 MHz, CDCl3):
δ 147.4, 143.3, 128.1, 126.5, 124.8, 123.3, 85.8, 48.6, 37.9, 29.7, 28.1, 27.4, 26.1, 25.7;
HRMS (m/z): [M]+: calcd. for C14H18NaO: 225.1255; found: 225.1206; HPLC conditions:
Column: OJH, Mobile phase: heptane : ethanol (96:4), enantiomers at 10.87 min and
12.77 min. Flow rate: 1 ml / min.

19  

 
1-(naphthalen-1-yl)-1-phenylethanol

Colourless oil (64% yield); 1H-NMR (300 MHz, CDCl3): δ 7.95–7.25

(m, 12H), 2.10 (s, 3H). HPLC conditions: Column: IA, Mobile phase:
heptanes : ethanol (97:3), enantiomers at 9.800 min and 11.029 min.
Flow rate: 1 mL / min. NMR shifts are consistent with literature
values.15

3-methyl-1-phenylpentan-3-ol

Colourless oil (74% yield); 1H-NMR (300 MHz, CDCl3): δ 7.34–7.18 (m,
5H), 2.69 (m, 2H), 1.79 (m, 2H), 1.58 (m, 2H), 1.25 (s, 3H), 0.97
(apparent t, J = 7.4 Hz, 3H). HPLC conditions: Column: IB, Mobile
phase: heptanes : ethanol (99.5:0.5), enantiomers at 18.100 min and 19.627 min. Flow
rate: 1 mL / min. NMR shifts are consistent with the literature.16

20  

 
2-cyclohexyl-4-phenylbutan-2-ol

Colourless solid (75% yield); 1H-NMR (400 MHz, CDCl3): δ 7.38–7.12 (m, 5H), 2.71 (m,
13
2H), 1.95-1.61 (m, 6H), 1.40 (m, 1H), 1.26 – 0.95 (m, 9H); C NMR (101 MHz, CDCll3): δ
142.85, 128.36, 128.32, 125.66, 74.39, 47.58, 41.89, 29.81, 27.58, 26.89, 26.78, 26.73,
26.51, 23.84. HRMS (m/z): [M]+: calcd. for C16H24ONa: 255.1725;
found: 255.1718; HPLC conditions: Column: IB, Mobile phase:
heptanes : ethanol (99:1), enantiomers at 18.100 min and 19.627
min. Flow rate: 1 mL / min.

21  

 
3-phenyl-1-(piperidin-1-yl)pentan-3-ol

Colourless solid (88% yield); HPLC conditions: Column: IA, Mobile

phase: heptanes : ethanol (95:5), enantiomers at 4.429 min and
4.644 min. Flow rate: 1 mL / min.

22  

 
5. Asymmetric Synthesis of α-Tocopherol precursor A (3R,7R,11R)-1-(2,5-
dimethoxy-3,4,6-trimethylphenyl)-3,7,11,15-tetramethyl hexadecan-3-ol and
(3S,7R,11R)-diastereomer

O O MeMgI O HO
L1 (1.0 eq.)

toluene,
O -78 oC
O (R,R)-L1 - SRR-(-)-product, 92% de
(S,S)-L1 - RRR-(+)-product, 90% de

(7R,11R)-1-(2,5-Dimethoxy-3,4,6-trimethylphenyl)-7,11,15-trimethylhexadecan-3-one1 was
treated with methyl magnesium bromide according to General Procedure A using (S,S)-
L1 for asymmetric Grignard synthesis. Purification by flash chromatography (cyclohexane :
ethyl acetate, 9:1) gave the product as a colourless oil (37 mg, 78% yield). The same
procedure with (R,R)-L1 gave 92% de of SRR diastereoisomer. The absolute configuration
was determined through their optical rotations, compared to the literature values.17

1
H NMR (400 MHz, CDCl3): δ 3.72 (s, 3H), 3.64 (s, 3H), 2.72 (m, 2H), 2.22 (s, 3H), 2.18 (s,
13
6H), 1.70 – 1.02 (m, 26H), 0.89 - 0.82 (m, 12H). C NMR (101 MHz, CDCl3) δ 153.09,
152.64, 132.24, 128.21, 127.92, 127.07, 72.76, 60.87, 60.05, 42.28, 41.81, 41.78, 39.34,
37.69, 37.41, 37.26, 32.81, 32.77, 27.94, 26.71, 26.69, 24.76, 24.48, 22.68, 22.59, 21.67,
21.46, 19.72, 12.84, 12.65, 11.94; α21=−1.57   (c=0.75   in   CHCl3)   for   SRR   diastereoisomer  

and  α21=+1.53  (c=0.75  in  CHCl3)  for  RRR  diastereoisomer.

SCF Conditions: Column: IA; Co-solvent: methanol (gradient 1% to 40%, ref. Table 1);
Column Temperature: 35 ˚C; Flow rate: 3.0 mL/min; Diastereoisomers at 2.9 min SRR and
3.2 min RRR.
Table 1. SCF Gradient Table
Flow Rate
No. Time [min] %MeOH
[ml/min]
1 Initial 3.000 1.0
2 1.0 3.000 1.0
3 5.0 3.000 40.0
4 5.1 3.000 1.0

23  

 
reaction performed with no added ligand w/ (R,R)-ligand: SRR (major) RRR (minor)

6. Comparison of methods for asymmetric nucleophilic addition to ketones

(mentioned in Scheme 1, Main Text)

Table 1 presents the detailed comparison between synthetic methods of stereoselective

formation of tertiary alcohols, developed previously by Weber and Seebach (direct
organomagnesium addition to ketones) called Method 1, Harutyunyan & Minnaard
(organocopper mediated addition of Grignard reagents to ketones (Method 2) and the
current work (Method 3). We considered the following criteria that we found relevant for
the evaluation of the method: (1) active species – this describes the type of a nucleophile
which is present in a stereoselective step and which determines its chemistry; (2) reaction
conditions – temperature and reaction time; (3) chiral ligand – its quantity, availability,
tuneability, cost etc.; (4) scope – both in ketone and Grignard reagent; (5) applications –
this describes the best to our knowledge application of the method to the actual synthetic
target. For the scope evaluation we have selected the threshold of 75% ee as a lowest
practical limit with yields higher that 60% as a potentially workable synthesis.

24  

 
Table 1. Detailed comparison of methods for asymmetric nucleophilic addition to
ketones mentioned in Scheme 1 (Main Text).

Weber & Seebach18 Harutyunyan & Gilheany &

19
Minnaard Bieszczad

Method 1 Method 2 Method 3

Active Species Organomagnesium Organocopper Organomagnesium

Temperature -105 oC -78 oC -78 oC
Reaction Times 9-11 h Not specified20 15-30 min
Amount of
3.0 eq. 1.0 eq. 2.0 eq.
Grignard reagent
- Catalytic (6 mol%)
- Stoichiometric
- Two components: - Stoichiometric
- One component:
(S,RP)-1g = 502 € - One component
(R,R)-1g = 116.5 €21
Ligand (R,SP)-1g = 560 € - Easy to prepare (4
(S,S)-1g = 74.8 €
CuBr.Me2S-10g = 73 € steps)
- Easy to prepare
- Difficult to prepare - Easily tuneable
- Difficult to tune
- Difficult to tune
Difficult to separate Difficult to separate. V. easy to separate.
(tertiary alcohol itself). Req. Column Column
Ligand Removal /
Distillation (works only chromatography. chromatography not
Recovery
for small molecules). Cannot be easily necessary.
Can be recycled. recycled. Easily recycled.
20
- Ar-methyl ketones
Ketone Scope
2
- Ar-SiPh2Me ketones22 Ar-alkyl ketones
(ee>75%, sp -methyl ketones
- α-methyl-α,β- Alkyl-alkyl ketones
yield>60%)
unsaturated ketones23
Methyl
Grignard Scope Linear alkyl
24
(ee>75%, Linear alkyl Branched alkyl Branched alkyl
yield>60%) Cyclic alkyl
Naphthyl
α-tocopherol
γ-tocopherol
Applications Unknown (10 steps,
(16 steps, 73% de)25
90% / 92% de)

25  

 
The main advantage of Method 1 is its high selectivity for the addition of linear Grignards
to sp2-methyl ketones (aryl-methyl, heteroaryl-methyl, alkynyl-methyl and α,β-unsaturated-
methyl), using relatively inexpensive TADDOL ligand. The method, however requires
extremely harsh conditions (-105 ˚C), long reaction times (9-11 h) and large amount of
Grignard reagent (3.0 eq.). The ligand is a tertiary alcohol itself therefore it is difficult to
remove from the reaction mixture and the distillation of crude product is only possible for
relatively small molecules. The scope of the reaction in Grignard reagent is limited to linear
Grignards only (ethyl, n-butyl etc.) and ketone scope is limited to sp2-metyl ketones.

Method 2 is a rare example of a catalytic addition of Grignard reagents to ketones,

mediated by organocopper complexes with an expensive Josiphos ligand (6 mol%). The
main disadvantage of this method is that it is not very general - working only for the
addition of branched Grignard reagents (isobutyl, etc.) to a limited and very specific set of
ketones: aryl-methyl ketones, Ar-SiPh2Me ketones24 and α-methyl substituted α,β-
unsaturated ketones. This method was applied for γ-tocopherol precursor synthesis, but
due to scope limitations the synthesis was long (16 steps) and the de was only 73%.

Method 3 belongs to the same group as Method 1 – the direct organomagnesium addition
to ketones, mediated by a stoichiometric amount easily accessible chiral ligand, but unlike
Method 1 it does not require such extreme conditions (temperature) and the reaction is
complete in 30 min. The ligand is tunable and can be easily prepared, removed from the
reaction mixture and recycled. The other advantage is that the method has a much
broader scope than Methods 1 and 2 – both (1) in terms of Grignard reagents: it works for
methyl and ethyl Grignard (ee>90%) – linear, for branched, cyclic and naphthyl (ee>75%);
(2) in terms of a ketone – it give high selectivities and yields for both for aryl-alkyl ketones
(ee>90%) and alkyl-alkyl ketones (ee>90%). Its application to α-tocopherol precursor
synthesis gave the desired alcohol in 90% / 92% de (depending on the sense of
stereochemistry) in 10 steps.

26  

 
 7. Mechanistic considerations

Based on initial observations we have adopted a tentative mechanistic model of a

stereoselective complex, which exemplification is presented in Figure 2. This is the
simplest possible model derived by adaption of Figure 1 (in main MS), taking into account:
(i) the ligand/magnesium stoichiometry (1:2); (ii) that the carbanion and phenoxide could
not be on the same magnesium for valence reasons. The carbanion transfer occurs from
the non-ligated magnesium within a chair-like bridged aggregate with the ligated
magnesium acting as Lewis acid activator of the ketone.

Figure 2. Possible stereoselective complex for ligand (S,S)-L1 for the addition of
ethylmagnesium bromide to acetophenone  . Ether solvents molecules omitted for clarity.
We envisage formation of the complex by a chain of events as presented in Figure 3.

Step 1 is a simple acid-base reaction between Grignard reagent and the acidic proton of a
ligand. This reaction is exothermic – the rise of temperature for the addition of first
equivalent of Grignard is higher than for the second - and faster than 1,2-addition. No
product was formed when 1.0 eq. of Grignard was used. As a result, the complex A is
formed (Step 2), when magnesium atom is complexed by 3 heteroatoms, occupying 3 out
of 4 sections of its coordination sphere, creating a “chiral pocket” in a 4th quarter. Following
Ashby,26 we assume formation of a dimeric species, bridged by 2 halogen atoms. The free
space around magnesiums is assumed to be occupied by ether molecules. Our
hypothesis, which requires further investigation, states that the complex formed between
an ether molecule and the ligated magnesium in a “chiral pocket” is of a higher energy that
the complex formed with an ether molecule of the “free magnesium”.

27  

 
 Figure 3. Mechanistic hypothesis for the asymmetric synthesis of tertiary alcohols using
tridentate ligands, developed in the course of this project.

Also following Ashby (in the absence of evidence to the contrary), we next assume
displacement of the “less bound” ether molecule by a ketone and the formation of complex
B is the next, slow and rate determining step. This hypothesis is supported by the ether
studies (Table 2, the paper) presented in Figure 4.

Figure 4. Possible explanation of the ether inhibition effect

28  

 
Results obtained with different ethers indicate “all or nothing” type of a situation, when
bulky and less Lewis basic ether provides “ON” mode of enantioselectivity, while THF – a
small ether and strong Lewis base produces a racemate (“OFF” mode). The above
phenomenon can be explained by the strong “inhibition” of the active site on ligated
magnesium atom by THF. This hypothesis is supported by the observation, that the
enantioselectivity is restored, when 2,5-dimethyl THF is used instead, which is
considerably more bulky and produces a complex of a higher energy with a ligated
magnesium atom, therefore it is easier to displace by a ketone.

We assume that the stereoselection occurs because the ketone adopts one of two
possible fixed positions in the product of Step 3 with a fixed orientation of its groups. We
can speculate that the stereoselective complex is formed in Step 3, such that all elements
rapidly adopt the geometrically preferential arrangement in a pseudo-chair conformation in
a six-membered ring (Figure 2) and the carbanion is transferred to the pre-oriented ketone
to yield complex C (Step 5), which is hydrolysed in the workup to chiral tertiary alcohol.

The experimental data presented in Table 2 in the paper is consistent with the above
model. We can see that for example, the (S,S)-ligand influences (R) sense of
stereoselectivity for the addition of ethyl magnesium bromide to acetophenone by addition
of carbanion “from the back” (see Figure 5). We have also observed that the mirror image
of the ligand produces the opposite sense of stereochemistry of the product, as expected

Figure 5. “Back addition” of the carbanion to a pre-chiral ketone of a fixed orientation

influenced by (S,S)-ligand

29  

 
The above observations strongly support the involvement of a single stereoselective
mechanism. Further mechanistic investigations will be reported in due course.

8. References

                                                                                                                       
1 B. Bieszczad, D. G. Gilheany, submitted for publication.
2   H. J. Schanz, M. Linseis, D. G, Gilheany, Tetrahedron: Asymmetry 2003, 14, 2763–2769.  
3 Y. Zhu, J. P. Malerich, R. Viresh, Angew. Chem. Int. Ed. 2010, 49, 153–156.
4 A. M. Lopez-Periago, C. A. Garcia-Gonzalez, C. Domingo, Chem. Comm. 2010, 46, 4315 – 4317.
5 D. W. Lee, H. J. Ha, W. K. Lee, Synth. Commun. 2007, 37, 737–742.
6 G. Labourdette, D. J. Lee, B. O. Patrick, M. E. Ezhova, P. Mehrkhodavandi, Organometallics 2009, 28,
1309–1319.
7 H. Christensen, Synth. Comm. 1975, 1, 61-78
8 Yields were measured as isolated yields and 1H NMR yields with ligand acting as an internal standard
9 S.-J. Jeon, J. Li, C. Garcia, L. K. LaRochelle, P. J. Walsh, J. Org. Chem. 2005, 70, 448-455.
10 H. Zong, H. Huang, J. Liu, G. Bian, L. Song, J. Org. Chem. 2012, 77, 4645-4652.
11 M. Yus, D. J. Ramon, O. Prieto, Tetrahedron: Asymmetry 2003, 14, 1103-1114.
12 C. A. Musser, Jr. H. Richey, J. Org. Chem. 2000, 65, 7750–7756.
13 H. J. Hamann, J. A. Liebscher, J. Org. Chem. 2000, 65, 1873-1876.
14 D. Oezdemirhan, S. Sezer, Y. Soenmez, Tetrahedron: Asymmetry 2008, 19, 2717-2720.
15 C. A. Chen, K. H. Wu, H. M. Gau, Angew. Chem. Int. Ed. 2007, 46, 5373-5376.
16 A. G. Sakhabutdinov, A. G. Usmanova, A. G. Proidakov, B. A. Bazhenov, F. K. Shmidt, Zhurnal
Organicheskoi Khimii 1988, 8, 1691-1696.
17 S. Takano, T. Sugihara, K. Ogasawara, Synlett 1990, 8, 451.
18 B. Weber, D. Seebach, Tetrahedron 1994, 50, 6117-6128,
19 A. V. R. Madduri, S. R. Harutyunyan, A. J. Minnaard, Angew. Chem. Int. Ed. 2012, 51, 3164.
20 However, the reaction time for Grignard addition to α-methyl substituted α,β-unsaturated ketones was
quoted as 5-10 h and for the addition of isobutyl Grignard to Ar-SiPh2Me ketone lasted 2h.
21 All prices from Sigma-Aldrich, 23 November 2016.
22 J. Rong, R. Oost, A. Desmarchelier, A. J. Minnaard, S. R. Harutyunyan, Angew. Chem. Int. Ed. 2014,
53, 1.
23 A. V. R. Madduri, A. J. Minnaard, S. E. Harutyunyan, S. E., Chem. Comm. 2012, 48, 1478–1480.
24 For the specific case of aryl-SiPh2Me ketones, good results (ee>80%) were also obtained for linear
Grignards.
25 Z. Wu, S. R. Harutyunyan, A. J. Minnaard, Chem. Eur. J. 2014, 20, 14250
26 E. C. Ashby, J. Laemmle, H. M. Neumann, Acc. Chem. Res. 1974, 7, 272.

30