Asymmetric Grignard Synthesis of Tertiary Alcohols through Rational Ligand Design

Article abstract of DOI:10.1002/anie.201610462

A simple, general and practical method is reported for highly enantioselective construction of tertiary alcohols through the direct addition of organomagnesium reagents to ketones. Discovered by rational ligand design based on a mechanistic hypothesis, it has an unprecedented broad scope. It utilizes a new type of chiral tridentate diamine/phenol ligand that is easily removed from the reaction mixture. It is exemplified by application to a formal asymmetric synthesis (>95:5 d.r.) of vitamin E.

Full text of DOI:10.1002/anie.201610462

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201610462
German Edition: DOI: 10.1002/ange.201610462
Asymmetric Synthesis
Asymmetric Grignard Synthesis of Tertiary Alcohols through Rational
Ligand Design
Bartosz Bieszczad and Declan G. Gilheany*
Abstract: A simple, general and practical method is reported
for highly enantioselective construction of tertiary alcohols
through the direct addition of organomagnesium reagents to
ketones. Discovered by rational ligand design based on
a mechanistic hypothesis, it has an unprecedented broad
scope. It utilizes a new type of chiral tridentate diamine/
phenol ligand that is easily removed from the reaction mixture.
It is exemplified by application to a formal asymmetric
synthesis (> 95:5 d.r.) of vitamin E.
M
ore than a century after Victor Grignard first described
his alcohol synthesis,[1] and despite its common use in
academia and industry,[2,3] it still lacks a general asymmetric
version. This is despite the very high desirability of a process
that could take advantage of the benign nature of magnesium,
the high reactivity of the Grignard reagent, the simplicity of
its formation, ease of handling and disposal, and the
impressive substrate scope of the synthesis. This versatility
has been further increased through the introduction by
Knochel and coworkers of functionalized Grignard
reagents.[4]
The lack of a general, asymmetric variant of the Grignard
synthesis, where an organomagnesium is added to a carbonyl
with high enantioselectivity and broad scope, is especially
relevant to the synthesis of chiral tertiary alcohols, which are
a common structural motif in natural products[5a] and
pharmaceuticals.[5b] Chiral tertiary alcohols cannot be
formed by well-established methods of asymmetric hydro-
genation[5,6] so that the only direct synthesis is through C C
À
Scheme 1. Asymmetric nucleophilic addition to ketones: present
bond forming by nucleophilic additions to ketones. Inventive
alternatives[7] to nucleophilic addition have been reported
but, being indirect, they are inherently inefficient. There are
no general enzymatic resolutions.[8]
Scheme 1 shows the current status of asymmetric nucle-
ophilic addition to ketones. The present routes involve
complexes of elements other than magnesium as the source
of asymmetry,[9,10] the most recent being from the group of
Harutyunyan and Minnaard with copper,[11] and Hoveyda and
co-workers with boron, for the specific case of allyl
addition.[12] However, with introduction of other elements,
the advantages of direct Grignard synthesis are either
status.
compromised or lost. Most notably, substrate scope is
substantially reduced, making some important chiral scaffolds
unavailable. For example, the copper-based method men-
tioned is limited to branched Grignard reagents and partic-
ular types of ketone (Scheme 1).
In what we have termed “the road less traveled”, direct
asymmetric organomagnesium addition to carbonyl is poorly
developed.[13] There is only a single successful approach: use
of a diol introduced by Weber and Seebach in 1992.[14–16] Their
work, however, has significant drawbacks and limitations
(Scheme 1) and only recently was a practical variant of their
method reported.[17] We contend that, to date, this road has
only been explored by trial and error, taking little account of
reaction mechanism. This was then our starting point in
a fresh approach to this long-standing problem.
[*] Dr. B. Bieszczad, Prof. Dr. D. G. Gilheany
Centre for Synthesis and Chemical Biology, School of Chemistry
University College Dublin
Belfield, Dublin 4 (Ireland)
E-mail: declan.gilheany@ucd.ie
The task of rendering asymmetric the Grignard synthesis
of tertiary alcohols is challenging: 1) the mechanism may be
radical or concerted,[18,19] 2) there is competition from
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
http://dx.doi.org/10.1002/anie.201610462.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Asymmetric Grignard synthesis of various acyclic tertiary
alcohols mediated by (R,R)-L1.[a]
background non-stereoselective reaction and from reduction/
enolization side-reactions,[20] 3) there is reduced prochiral
face discrimination (as compared to aldehydes), and 4) most
significant, is the presence of complex Schlenk and aggrega-
tion equilibria.[2,3]
We focused on this last issue and initially hypothesized
that Schlenk-type dynamic interconversion of key stereose-
lective complexes is at least partly responsible for reduced
selectivity and so should be locked by inhibiting ligand
exchange. Following Sharpless,[21] we then considered the Mg
coordination spheres in the simplest halogen-bridged aggre-
gate[18] within which the addition occurs (Figure 1). Crucially,
Entry
Product
R
Yield
[%][c]
ee
alcohol[b]
[%][d]
1
2
Et
Et
80
53
78
76
3
iPr
60
78
4
5
Et
Et
99
71
54
60
Figure 1. Design concept: adaptation of the previous mechanistic
hypothesis showing the three potentially available coordination sites
on each metal center (a–f) and the tridentate ligands selected to test
the concept.
6
7
1-Np
Et
64
75
70
we saw that the maximum number of available coordination
sites for either magnesium is three. Therefore we decided to
test a novel additive class—tridentate ligands—as opposed to
bidentate additives, many of which were previously reported
to give low enantioselectivity.[13]
68[e]
8
Et
Et
74
65
88
95
9[f]
We selected half-salan ligands L1[22] and L2. The “priv-
ileged” inexpensive diaminocyclohexane skeleton is best
known linked with the Jacobsen di-tert-butylsalicyl motif.[23]
However, we showed that other substitution patterns could be
more effective;[24a] notably o-CF3.[24b] We were vindicated by
very promising initial results for asymmetric Grignard
syntheses of various acyclic tertiary alcohols with L1
(Table 1). Consistent selectivities were obtained across
a wide variety of substrate and ketone types to give
dialkylaryl, diarylalkyl, and especially trialkyl alcohols;
including primary, secondary, and aryl Grignard reagents,
and aromatic and non-aromatic ketones.
To demonstrate that these initial results can be optimized,
2-phenyl-2-butanol was selected as a test substrate (Table 2).
Firstly, the three-way disconnection was probed with ligand
L1; the route utilizing PhMgBr gave poorer ee whereas those
with MeMgX resulted in higher ee values (entries 1–5).
Ligand L2 further improved stereoselection (entries 6 and 7
vs. entries 1 and 2), ultimately to > 90%. Notably, use of
MeMgX gives rise to the opposite configuration of product
consistent with our proposed mechanism (see below) in which
the same sense of addition (from “the back” for SS, from “the
front” for RR) leads to the opposite enantiomer.
10
b-PhEt
75
86
[a] Conditions (see Supporting Information): toluene/ether (6:1), ketone
(0.1 mmol, overall concentration 0.08m). [b] Arbitrary configurations
shown. [c] Yield of isolated products, except where noted. [d] Determined
by chiral stationary phase high-performance liquid chromatography
(HPLC; see Supporting Information). [e] Yield determined by NMR
spectroscopy. [f] TEMPO (1 equiv) added.
methyl tert-butyl ether (MTBE) and dimethyltetrahydrofuran
(DMTHF) restore selectivity (Table 2, entries 11 and 12).
We also examined cyclic aromatic ketones (Table 3)
where selection is especially good (> 90% ee with L1). With
secondary RMgX, a nice bonus was suppression of reduc-
tion[20] (entries 10 and 12 vs. entries 11 and 13), allowing
synthesis of novel tertiary alcohols.
The reactions are particularly clean with reduced yields
fully accounted for by returned starting material, arising from
enolisation. The concentration of the reaction mixture can be
increased to 0.6m with little effect on ee. Most significantly,
the ligand can be fully recovered by acid–base extraction with
acetic acid (see Supporting Information).
Toluene solvent gave the best results but small amounts of
diethyl ether (Et2O) are needed to keep the RMgX in
solution (entries 1–8). Tetrahydrofuran (THF) gives racemic
product (Table 2, entry 10), which we attribute to its strong
coordinating ability. Consistent with this, the more bulky
To explore the issue of concerted versus radical mecha-
nisms,[18,19] we used the radical agent 2,2,6,6-tetramethyl-
piperidinyloxyl (TEMPO). Its use increased ee while lowering
yield (Table 1, entries 8 and 9; Table 2, entries 5 and 8;
Table 3, entries 1 and 3). This suggests that there is a small
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 2: Asymmetric Grignard synthesis of 2-phenyl-2-butanol mediated
by ligands (R,R)-L1 and (R,R)-L2.[a]
Entry
R
X
L
Solvent[b]
Yield
[%][a]
ee [%][a]
(config.)
1
2
3
4
5
6
7
Et
Me
Ph
Et
Me
Me
Et
Me
Et
Et
Br
Br
Br
I
I
I
Br
I
Br
Br
Br
Br
L1
L1
L1
L1
L1
L2
L2
L1
L1
L1
L1
L1
Tol/Et2O
Tol/Et2O
Tol/Et2O
Tol/Et2O
Tol/Et2O
Tol/Et2O
Tol/Et2O
Tol/Et2O
Et2O
80
75
74
76
56
65
57
61
46
86
84
52
78 (S)
88 (R)
22 (S)
78 (S)
88 (R)
92 (R)
87 (S)
91 (R)
68 (S)
0
Figure 2. Working proposal for stereoselecting step (solvent omitted).
8[c]
9
10
11
12
Supporting Information for a full mechanistic proposal).
The carbanion transfer occurs from the non-ligated magne-
sium within a chair-like bridged aggregate with the ligated
magnesium acting as a Lewis acid activator of the ketone. For
the SS-ligand shown, attachment of the nucleophile “from the
back” leads to the observed stereochemistry. Notably the
model is also consistent with the change of product config-
uration described in Table 2 (entries 1 and 2). For that specific
case, shown in Figure 2, the configuration of the product 2-
phenyl-2-butanol changes on switching the reagents used to
prepare it (acetophenone/EtMgBr vs. propiophenone/
MeMgBr), which is equivalent to switching the Et and Me
groups in Figure 2.
THF
MTBE
DMTHF
Et
Et
78 (S)
73 (S)
[a] Procedures, conditions, and analyses as for Table 1; use of (S,S)-L1
gave the opposite enantiomer with the same selectivity. [b] Tol/
Et2O=toluene/ether (6:1); THF=tetrahydrofuran; MTBE=methyl tert-
butyl ether; DMTHF=2,5-dimethyltetrahydrofuran. [c] TEMPO (1 equiv)
added.
Table 3: Cyclic tertiary alcohols prepared by asymmetric Grignard syn-
thesis mediated by ligands (R,R)-L1 and (R,R)-L2.[a]
To demonstrate the utility of the new method, we applied
it to an asymmetric synthesis of vitamin E (a-tocopherol).
The naturally occurring stereoisomer (2R,4’R,8’R) exhibits
the highest activity but remarkably, in contrast to most of the
other vitamins, it has no economic total synthesis.[25] As
established by Cohen, Lopresti, and Neukom,[26,27] the crucial
intermediate is tertiary alcohol A (Scheme 2).
Entry
Product
R
X
L
Yield
[%][a]
ee
alcohol[b]
[%][a]
1
Et
Et
Et
Br
Br
Br
L1
L2
L1
72
65
52
94 (S)
90 (S)
97 (S)
2
3[c]
We were therefore gratified that MeMgI treatment of B in
the presence of (R,R)-L1 generated alcohol A in a d.r. of 96:4,
while use of (S,S)-L1 gave a 5:95 d.r. of the other diastereo-
mer (Scheme 2), which was identified by polarimetry as
(R,R,R)-a-tocopherol precursor.[27] Given that A was synthe-
sized in seven steps from phytol in 18% unoptimized overall
yield (see Supporting Information), this formally constitutes
a workable total asymmetric synthesis of vitamin E.
4
5
Me
Me
Br
I
L1
L1
86
89
77
94
6
7
Et
Et
Br
Br
L1
L2
57
45
93
90
8
9
Et
Et
Br
Br
L1
L1
73
45
91
90
In summary we have shown, for the first time, that
a general, practical direct asymmetric Grignard synthesis of
tertiary alcohols is possible with high selectivity. The crucial
10
11
iBu
iBu
Br
Br
L1
–
65
85
n/a
6[d]
12
13
Cpt
Cpt
Br
Br
L1
–
80
76
n/a
0[d]
[a] Procedures, conditions, and analyses as for Table 1, except where
noted. [b] Configurations shown are arbitrary unless given explicitly in
the last column. [c] TEMPO (1 equiv) added. [d] Product mainly
1-indanol.
contribution from the radical pathway leading to erosion of ee
in these cases.
We have adopted the simplest possible working model for
the stereoselectivity (Figure 2), which is derived by adaption
of Figure 1 to reflect the ligand/magnesium stoichiometry 1:2,
and on the basis of simple valence considerations (see
Scheme 2. Formal asymmetric vitamin E synthesis.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Kneisel, P. Knochel, Angew. Chem. Int. Ed. 2005, 44, 4627;
Angew. Chem. 2005, 117, 4703; c) J. Clayden, W. Farnaby, D. M.
Grainger, U. Hennecke, M. Mancinelli, D. J. Tetlow, I. H. Hillier,
M. A. Vincent, J. Am. Chem. Soc. 2009, 131, 3410; d) Y. Yang,
I. B. Perry, G. Lu, P. Liu, S. L. Buchwald, Science 2016, 353, 144 –
150; e) An interesting industrial example making use of a chiral
auxiliary approach is: “Process for asymmetric alkylation of
carbonyl compounds”: M. Albert, H. Sturm, A. Berger, P.
Kremminger, WO2007082771A1, 2007.
breakthrough is the use of a rationally designed tridentate
ligand that can be easily recovered and reused. The method is
of broad scope, gives moderate to excellent yields (ꢀ 90%)
and enantioselectivities (ꢀ 97% ee) of trialkyl-, dialkylaryl-,
and alkyl diaryl carbinols (both cyclic and acyclic). We have
also shown that it can be applied on a gram scale (see
Supporting Information). Together with the three-way dis-
connection flexibility of the Grignard synthesis, this will allow
simplification of synthetic pathways to tertiary alcohols not
previously available in enantiomerically pure form. We have
demonstrated this with an asymmetric synthesis of vitamin E.
The solution to this 100-year-old problem also produced
some intriguing aspects—suppression of reduction and influ-
ence of radical agents—that promise insight into the Grignard
synthesis mechanism. Further investigations are underway
and will be reported in due course.
[8] R. Kourist, U. T. Bornscheuer, Appl. Microbiol. Biotechnol.
2011, 91, 505.
[9] For reviews of other element-based nucleophilic ketone addi-
tions, see: a) J. F. Collados, R. Sola, S. R. Harutyunyan, B. Macia,
ACS Catal. 2016, 6, 1952; b) M. Hatano, K. Ishihara, Synthesis
2008, 1647; c) O. Riant, J. Hannedouche, Org. Biomol. Chem.
2007, 5, 873.
[10] Leading references to methods based on particular elements are
listed. Cu: M. Shibasaki, M. Kanai, Chem. Rev. 2008, 108, 2853;
Zn: P. I. Dosa, G. Fu, J. Am. Chem. Soc. 1998, 120, 445; M.
Hatano, T. Miyamoto, K. Ishihara, Org. Lett. 2007, 9, 4535; Zn/
Ti: D. J. Ramꢀn, M. Yus, Tetrahedron Lett. 1998, 39, 1239; C.
Garcꢁa, L. K. LaRochelle, P. J. Walsh, J. Am. Chem. Soc. 2002,
124, 10970; E. F. DiMauro, M. C. Kozlowski, J. Am. Chem. Soc.
2002, 124, 12668; Ti: K.-H. Wu, Y.-Y. Kuo, C.-A. Chen, Y.-L.
Huang, H.-M. Gau, Adv. Synth. Catal. 2013, 355, 1001; Zn/Al:
D. K. Friel, M. L. Snapper, A. H. Hoveyda, J. Am. Chem. Soc.
2008, 130, 9942; Ti/Al: D. B. Biradar, S.-L. Zhou, H.-M. Gau,
Org. Lett. 2009, 11, 3386; Mg/Ti: E. Fernꢂndez-Mateos, B. Maciꢂ,
M. Yus, Eur. J. Org. Chem. 2014, 6519; B: L. Huang, J. Zhu, G.
Jiao, Z. Wang, X. Yu, W.-P. Deng, W. Tang, Angew. Chem. Int.
Ed. 2016, 55, 4527; Angew. Chem. 2016, 128, 4603.
[11] A. V. R. Madduri, S. R. Harutyunyan, A. J. Minnaard, Angew.
Chem. Int. Ed. 2012, 51, 3164; Angew. Chem. 2012, 124, 3218; Z.
Wu, S. R. Harutyunyan, A. J. Minnaard, Chem. Eur. J. 2014, 20,
14250; J. Rong, T. Pellegrini, S. R. Harutyunyan, Chem. Eur. J.
2016, 22, 3558.
[12] M. Yus, J. C. Gonzalez-Gomez, F. Foubelo, Chem. Rev. 2011,
111, 7774; D. W. Robbins, K. Lee, D. L. Silverio, A. Volkov, S.
Torker, A. H. Hoveyda, Angew. Chem. Int. Ed. 2016, 55, 9610;
Angew. Chem. 2016, 128, 9762.
Acknowledgements
The Irish Research Council for Science Engineering and
Technology is gratefully acknowledged for an Embark post-
graduate Scholarship to B.B. (RS/2011/845). Enterprise Ire-
land is thanked for continuing postdoctoral funding for B.B.
(grant CF/2013/3321). The research group of Professor
Patrick Guiry (UCD) kindly allowed us to use their micro-
wave reactor and Supercritical Fluid Chromatograph, which
was of great assistance. Arran Chemicals, Athlone, Ireland
are gratefully acknowledged for gifts of resolved trans-1,2-
diaminocyclohexane. We thank Dr. Rachel Hiney, Dr. Victor
Garcia-Ruiz, Dr. Sulaiman Al-Sulaimi, and Orla Murray for
preliminary experiments and Dr. Eoghan McGarrigle, Dr.
Kirill Nikitin, and Dr. Brian Kelly for valuable discussions.
Conflict of interest
[13] For a listing of the many other attempts to render magnesium
nucleophilic reagents asymmetric, see: M. R. Luderer, W. F.
Bailey, M. R. Luderer, J. D. Fair, R. J. Dancer, M. B. Sommer,
Tetrahedron: Asymmetry 2009, 20, 981.
[14] B. Weber, D. Seebach, Angew. Chem. Int. Ed. Engl. 1992, 31, 84;
Angew. Chem. 1992, 104, 96.
The authors declare no conflict of interest.
Keywords: asymmetric synthesis · Grignard synthesis ·
tertiary alcohols · tridentate ligands · vitamin E
[15] For a noteworthy earlier publication, again using an alcohol
(diacetone glucose), see: T. D. Inch, G. J. Lewis, G. L. Sainsbury,
D. J. Sellers, Tetrahedron Lett. 1969, 10, 3657.
[16] For a report on the successful use of chiral crown ether additives,
see: N. G. Lukꢃyanenko, A. V. Lobach, O. N. Leus, Russ J. Org.
Chem. 2004, 40, 273.
[17] K. Osakama, M. Nakajima, Org. Lett. 2016, 18, 236.
[18] E. C. Ashby, J. Laemmle, H. M. Neumann, Acc. Chem. Res. 1974,
7, 272.
[1] V. Grignard, C. R. Hebd. Seances Acad. Sci. 1900, 130, 1322.
[2] D. Seyferth, Organometallics 2009, 28, 1598.
[3] H. G. Richey, Grignard Reagents: New Developments, Wiley,
New York, 1999.
[4] P. Knochel, Handbook of Functionalized Organometallics,
Wiley-VCH, Weinheim, 2005.
[19] T. Holm, Acta Chem. Scand. 1991, 45, 925.
[20] H. Zong, H. Huang, J. Liu, G. Bian, L. Song, J. Org. Chem. 2012,
77, 4645; M. Hatano, O. Ito, S. Suzuki, K. Ishihara, J. Org. Chem.
2010, 75, 5008; A. Krasovskiy, F. Kopp, P. Knochel, Angew.
Chem. Int. Ed. 2006, 45, 497; Angew. Chem. 2006, 118, 511.
[21] K. B. Sharpless, Chem. Br. 1986, 22, 1. Sharpless has emphasized
in several publications the need to account for the coordination
sphere of the putative active metal species; this publication had
the most influence on our thinking.
[5] a) H. Arimoto, D. Uemura in Quaternary stereocenters, chal-
lenges and solutions for organic synthesis (Eds.: J. Christoffers,
A. Baro), Wiley-VCH, Weinheim, 2005, chap. 1, pp. 1 – 24;
b) J. G. de Vries in Quaternary stereocenters, challenges and
solutions for organic synthesis (Eds.: J. Christoffers, A. Baro),
Wiley-VCH, Weinheim, 2005, chap. 2, pp. 25 – 50.
[6] P. G. Cozzi, R. Hilgraf, N. Zimmermann, Eur. J. Org. Chem.
2007, 5969.
[22] G. Labourdette, D. J. Lee, B. O. Patrick, M. E. Ezhova, P.
Mehrkhodavandi, Organometallics 2009, 28, 1309.
[7] a) J. L. Stymiest, V. Bagutski, R. M. French, V. K. Aggarwal,
Nature 2008, 456, 778; b) H. Leuser, S. Perrone, F. Liron, F.
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[23] E. N. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker, L. Deng, J.
[26] N. Cohen, R. T. Lopresti, C. J. Neukom, J. Org. Chem. 1981, 46,
2445.
Am. Chem. Soc. 1991, 113, 7063.
[24] a) C. P. OꢃMahony, E. M. McGarrigle, M. F. Renehan, K. M.
Ryan, N. J. Kerrigan, C. Bousquet, D. G. Gilheany, Org. Lett.
2001, 3, 3435; b) A. M. Daly, M. F. Renehan, D. G. Gilheany,
Org. Lett. 2001, 3, 663.
[27] S. Takano, T. Sugihara, K. Ogasawara, Synlett 1990, 451.
[25] M. Eggersdorfer, D. Laudert, U. Lꢄtinois, T. McClymont, J.
Medlock, T. Netscher, W. Bonrath, Angew. Chem. Int. Ed. 2012,
51, 12960; Angew. Chem. 2012, 124, 13134.
Manuscript received: October 25, 2016
Revised: February 12, 2017
Final Article published: && &&, &&&&
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Asymmetric Synthesis
B. Bieszczad,
D. G. Gilheany*
&&&& — &&&&
Asymmetric Grignard Synthesis of
Tertiary Alcohols through Rational Ligand
Design
The road less traveled: A simple, general,
direct route for asymmetric Grignard
synthesis of tertiary alcohols was facili-
tated by coordinating tridentate ligands
to magnesium. The method was found by
rational design and has wide scope; it
was applied to a formal asymmetric
synthesis of vitamin E.
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!