A general strategy for the synthesis of P-stereogenic compounds

Article abstract of DOI:10.1002/anie.201306628

A great leap forward toward the general synthesis of P-stereogenic compounds: Heating H₃PO₂ with (-)-menthol and paraformaldehyde gives easily crystallized menthyl hydroxymethyl-H-phosphinate (1). From this product, virtually any P-s

Full text of DOI:10.1002/anie.201306628

Angewandte
Chemie
DOI: 10.1002/anie.201306628
Asymmetric Synthesis
Hot Paper
A General Strategy for the Synthesis of P-Stereogenic Compounds**
Olivier Berger and Jean-Luc Montchamp*
Preparing P-stereogenic compounds is one of the biggest
challenges of organophosphorus chemistry.[1] Although vari-
ous methods have been reported for the preparation of
specific P-stereogenic building blocks, based on kinetic
resolution, or on chiral auxiliaries, typically these have
severe limitations in the scope of their application.[1] More
than 40 years ago, Mislow and others pioneered the field of P-
stereogenic compounds and the study of their reactivities.[2]
A
case in point is menthyl phenyl-H-phosphinate PhP(O)-
(OMen)H (1), which has since been employed in various
reactions such as cross-coupling, substitution, or hydrophos-
phinylation.[3] However, enriched diastereomers of 1 remain
difficult to prepare as their isolation requires low-temper-
ature recrystallization (multiple crystallizations below À308C
or at À708C), and yields of isolated products were not
reported. Similar chemistry using MenOPCl2 and aryl
Grignard reagents was reported recently.[4] In the final
analysis, these methods still require cumbersome crystalliza-
tion procedures and are limited in terms of the phosphorus
compounds that are accessible and therefore the final
products (usually P-stereogenic phosphines) that can be
derived from them.
Herein, we report an extremely simple and inexpensive
approach to versatile P-stereogenic building blocks, on multi-
gram scales, and without the need for RPCl2 precursors. The
new intermediates also allow much more flexibility for their
functionalization into a broad variety of useful P-stereogenic
compounds. Compound 2 is prepared from hypophosphorous
acid, paraformaldehyde, and (À)-menthol in 9% yield
(> 6 g), and compound 3 is prepared from phenyl-H-phos-
phinic acid, (À)-menthol, and paraformaldehyde in 26%
yield (> 24 g) (Scheme 1). The two building blocks (RP)-2 and
(SP)-3 are crystallized in high (> 95%) diastereomeric purities
at À188C (in a freezer) or at room temperature, respectively.
While the yields are relatively low, these still compare to the
yields of literature methods, and multigram quantities are
available in a single preparation. The latter reaction was also
scaled up uneventfully to produce 88 g of (SP)-3 (24% yield,
96% de). The structures of (RP)-2 and (SP)-3 were confirmed
by single X-ray crystallography.[15]
Scheme 1. Preparation of 2 and 3.
using our own conditions.[5] Crystallization of the resulting
reaction mixture at room temperature led to 3 in good yield
(23% yield, 97% de; Scheme 1). This diastereomer is
identical to the one obtained directly from PhP(O)(OH)H
(Scheme 1). On the other hand, cross-coupling of crystalline
(RP)-2 with bromobenzene gave (RP)-3 in 68% yield. Thus the
same reaction sequence can be used to make either P confi-
guration simply depending on the starting material: the
mother liquor or the crystalline diastereomer! An overall
yield of 33% of useful P-stereogenic compounds is easily
achieved through the reaction of H3PO2. This yield is much
higher than that of any literature method which employs PCl
intermediates or/and Grignard reagents.
Besides improved crystallization properties (all com-
pounds were obtained after a single crystallization!), these
P-stereogenic building blocks have the advantage of contain-
ing a hydroxymethyl handle, which offers two major ways to
functionalization (Scheme 2)—either with preservation of the
hydroxymethyl carbon atom or through oxidation to the
corresponding H-phosphinate. We recently reported that the
Corey–Kim oxidation of (hydroxymethyl)phosphinates gives
To improve the value of the reaction leading to 2, the
mother liquor was directly cross-coupled with bromobenzene
[*] Dr. O. Berger, Prof. J.-L. Montchamp
Department of Chemistry, Box 298860
Texas Christian University, Fort Worth, TX 76133 (USA)
E-mail: j.montchamp@tcu.edu
[**] We gratefully acknowledge the National Science Foundation (CHE-
0953368) for financial support. We are also very grateful to TCU
Prof. Kayla Green for her help with X-ray crystallography.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201306628.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
the corresponding H-phosphinates in a highly stereoselective
way.[6]
Diastereomer (RP)-2 is clearly the most versatile P-
stereogenic building block to date (Scheme 3). Cross-cou-
pling of (RP)-2 with bromobenzene gives (RP)-5a (= (RP)-3) in
obtaining 2 and 3, and then 1, this approach is clearly
competitive with the synthesis of 1 from PhPCl2 or Men-
OPCl2.[2–4]
The usefulness of compound 1 in asymmetric organo-
phosphorus synthesis is well-established. However, it can only
be used for the synthesis of phenyl-containing products. To
obtain other substitution patterns, crystallization must be
optimized for each case.[4] Therefore the novel building block
(RP)-2 offers much flexibility previously unavailable. Com-
pound (RP)-2 can be viewed as a protected chiral equivalent
of alkyl phosphinates ROP(O)H2, since it can be stereospe-
cifically alkylated to form 4, or cross-coupled to form 5, and
the hydroxymethyl moiety can subsequently be cleaved to
form H-phosphinates like 1 and 6. Also, the presence of the
hydroxymethyl group in both 2 and 3 provides further
opportunities for functionalization since the carbon atom
can be preserved if desired. For example, Mitsunobu reaction
of 3 with phthalimide gives 9 in 70% yield (Scheme 3).
Our method is not limited to compounds 2 and 3. For
example, cinnamyl-H-phosphinic acid[7] can be esterified and
hydroxymethylated in one pot to form (RP)-10 in 32% yield
and > 99% de (Scheme 4) after a single crystallization at
room temperature.[8] Oxidation of (RP)-10 provides menthyl
cinnamyl-H-phosphinate (SP)-11 in 82% yield and > 99% de.
Scheme 2. Functionalization of P-stereogenic (hydroxymethyl)phosphi-
nates.
68% yield, and subsequent oxidative cleavage delivers (SP)-
1 in 81% yield. Compound (SP)-3 can also be oxidized to form
stereospecifically (RP)-1 in 91% yield (Scheme 3). Therefore,
cross-coupling of 2 followed by oxidation of 3 also leads to
either P configuration of 1 using inexpensive (À)-menthol in
all cases. Previous methods by Mislow and Han to prepare
either P stereoisomer of 1 relied on (À)-menthol and
(+)-menthol, respectively.[3c] (+)-Menthol is fifty times
more expensive than (À)-menthol. Because of the ease of
Scheme 4. Synthesis of P-stereogenic cinnamyl derivatives. Step
(a) Corey–Kim: 1) N-chlorosuccinimide (3 equiv), Me2S (3 equiv),
CH2Cl2, À788C, 1 h. 2) 10 (1 equiv), À788C, 30 min. 3) Et3N (5 equiv),
À788C to RT, 1 h.
Table 1 summarizes the oxidation of various (hydroxyme-
thyl)phosphinates. As can be seen the reaction is general and
proceeds in good yields and excellent stereoselectivities. Thus,
the present work also provides a general route to P-
stereogenic H-phosphinates that is much simpler and cheaper
than the desymmetrization of alkyl phosphinates.[9]
An example of the exploitation of the CH2OH moiety is
the Wittig rearrangement[10] (Scheme 5). Allylation of 3 gives
Table 1: Summary of the oxidative cleavage reactions.[a]
Scheme 3. Functionalization of 2 and 3, and stereodivergent synthesis
=
of PhP(O)(OMen)H (1). a) 1) Me3SiN C(OSiMe3)Me (2 equiv),
CH2Cl2; 2) MeI (1 equiv), 08C to RT, 20 h; or allylBr (2 equiv), 08C to
RT, 36 h. b) ArBr (1 equiv), Pd(OAc)2 (2 mol%), xantphos (2.2 mol%),
iPr2NEt (1.3 equiv), DMF/DME or toluene/ethylene glycol (9:1, v/v),
1158C, 24 h. c) 1) N-chlorosuccinimide (3 equiv), Me2S (3 equiv),
CH2Cl2, À788C, 10 min. 2) 3 or 5a or 5c (1 equiv), À788C, 3 h.
R
A
B
de[%]
config.
yield [%]
de[%]
config.
Ph
Ph
95
95
>99
94
SP
RP
RP
RP
RP
91
81
61
87
82
95
>99
96
94
>99
RP
SP
SP
SP
SP
=
3) Et3N (5 equiv), À788C to RT, 1 h. d) 1) Me3SiN C(OSiMe3)Me
Me
1-naphthyl
cinnamyl
(2 equiv), CH2Cl2; 2) MeI (2 equiv), 08C, 2 h; or allylBr (2 equiv), RT,
4 days. e) 1-octene (1 equiv), Et3B (1 equiv), hexane, RT, air, 20 h.
f) phthalimide (1.3 equiv), PyPPh2 (1.3 equiv), DIAD (1.3 equiv),
CH2Cl2, RT, 24 h. DIAD: diisopropyl azodicarboxylate.
>99
[a] For details, see the Supporting Information.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
para,para’-di-tert-butylbiphenylide) reduction[12] or substi-
tuted with organometallic entities. The latter method was
used in the historic synthesis of the P-stereogenic bidentate
phosphine ligand DiPAMP.[2a] The groups of Pietrusiewicz[13a]
and Buono[13b,4] have developed a very nice method to convert
H-phosphinates into secondary phosphine–borane com-
plexes. Finally, many methods are established to convert
tertiary phosphine oxides into the corresponding P-stereo-
genic phosphines through either retention or inversion of
configuration.[1] To date, the major limitation in using the
reactions shown in Scheme 6 has been the preparation of the
required starting materials 15 and 16. This fact has also
prompted many different methods to access P-stereogenic
compounds.[1] For example, the enzymatic resolution of
(hydroxymethyl)phosphinates requires significant optimiza-
tion and the use of expensive lipases/esterases.[14] In contrast,
our approach employs one of the cheapest chiral alcohols
available: (À)-menthol. With the present strategy, accessing
a wide variety of compounds 15 and 16 is straightforward and
inexpensive; so our method should be applicable to the
synthesis of virtually any P-stereogenic phosphine.
Scheme 5. Stereospecific Wittig rearrangement. a) 1) NaH (1.5 equiv),
1
2
=
THF, 08C to RT; 2) R R C CHCH2Br (1.2 equiv), THF, RT, 3 h. b) sBuLi
(2 equiv), THF, À788C, 4 h.
intermediate 12. Subsequent treatment of 12 with sBuLi
delivers the rearranged product 13. In both instances, a single
diastereomer is obtained. Stereochemical assignment was
established on 13b through X-ray crystallography.[15]
H-Phosphinate (RP)-2 was converted into menthyl thio-
phosphonic monoester 14 in quantitative yield [Eq. (1)]. In
the past, P-stereogenic compounds have been prepared by
resolving thiophosphonic acid monoesters with chiral bases
(quinine, brucine),[11] which requires optimization for each
case.
In summary, we have prepared numerous versatile and
inexpensive P-stereogenic phosphinate building blocks. The
reaction of H3PO2 or RP(O)(OH)H with (À)-menthol and
paraformaldehyde is apparently a general method. Menthyl
(hydroxymethyl)phosphinates display favorable crystalliza-
tion properties and synthetic versatility. These are obtained in
multigram quantities through simple and practical crystalli-
zation conditions (most often at room temperature!). The
method does not rely on any chlorophosphine intermediate.
One illustration of the synthetic flexibility is the stereo-
divergent preparation of both (RP)-1 and (SP)-1 from the
same (À)-menthol auxiliary. The presence of the hydroxy-
methyl group not only eases the crystallization process, but
also offers the possibility to maintain the methylene carbon
atom in other P-stereogenic derivatives, or to be cleaved to P-
stereogenic H-phosphinates. Compound (RP)-2 represents
a novel chiral version of hypophosphorous esters, from which
virtually any organophosphorus compound can be synthe-
sized. We believe that our method represents a great leap
forward toward the general synthesis of P-stereogenic com-
pounds. This should provide a revival of older menthol-based
methods and promote the development of novel phosphine
ligands for asymmetric synthesis. Further work to improve the
yields of (RP)-2 and related compounds through additional
reactions of the mother liquor, and to develop, expand, and
apply this method, is currently underway.
The preparation of a variety of P-stereogenic organo-
phosphorus compounds from 1 and from other menthyl
phosphinic esters is well-known (Scheme 6).[1] For example,
treatment of menthyl H-phosphinates 15 with organometallic
reagents gives the corresponding secondary phosphine oxides
with high stereoselectivity (inversion). Similarly, disubstituted
menthyl phosphinates 16 are alkylated after LDBB (lithium
Received: July 29, 2013
Published online: && &&, &&&&
Keywords: chiral auxiliaries · H-phosphinates · menthol ·
.
organophosphorus compounds · phosphinates
[1] Reviews on P-stereogenic compounds: a) O. I. Kolodiazhnyi,
Tetrahedron: Asymmetry 2012, 23, 1 – 46; b) J. S. Harvey, V.
Gouverneur, Chem. Commun. 2010, 46, 7477 – 7485; c) A.
Grabulosa, J. Granell, G. Muller, Coord. Chem. Rev. 2007, 251,
25 – 90; d) D. S. Glueck, Synlett 2007, 2627 – 2634; e) R. Engel,
Scheme 6. Transformations of P-stereogenic menthyl phosphinates,
described in the literature.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
J. I. Rizzo, Curr. Org. Chem. 2006, 10, 2393 – 2405; f) M. J.
Johansson, N. C. Kann, Mini-Rev. Org. Chem. 2004, 1, 233 – 247;
g) K. V. L. Crꢀpy, T. Imamoto, Top. Curr. Chem. 2003, 229, 1 –
40; h) L. A. Wozniak, A. Okruszek, Chem. Soc. Rev. 2003, 32,
158 – 169; i) O. I. Kolodiazhnyi, Tetrahedron: Asymmetry 1998, 9,
1279 – 1332; j) M. Ohff, J. Holz, M. Quirmbach, A. Bçrner,
Synthesis 1998, 1391 – 1415; k) K. M. Pietrusiewicz, M.
Zablocka, Chem. Rev. 1994, 94, 1375 – 1411; l) “Optically
active phosphines: preparation, uses and chiroptical properties”:
H. B. Kagan, M. Sasaki in Organophosphorus Compounds (Ed.:
F. R. Hartley), Wiley, New York, 1990, pp. 51 – 102 (part of the
series PATAIꢀS Chemistry of Functional Groups); m) H. B.
Kagan in Asymmetric Synthesis, Vol. 5 (Ed.: J. D. Morrison),
Academic Press, New York, 1985, pp. 1 – 39; n) D. Valentine Jr.
in Asymmetric Synthesis, Vol. 4 (Eds.: J. D. Morrison, J. W.
Scott), Academic Press, New York, 1984, pp. 263 – 312.
[2] a) B. D. Vineyard, W. S. Knowles, M. J. Sabacky, G. L. Bachman,
D. J. Weinkauff, J. Am. Chem. Soc. 1977, 99, 5946 – 5952; b) J.
Donohue, N. Mandel, W. B. Farnham, R. K. Murray, Jr., K.
Mislow, H. P. Benschop, J. Am. Chem. Soc. 1971, 93, 3792 – 3793;
c) W. B. Farnham, R. K. Murray, Jr., K. Mislow, J. Am. Chem.
Soc. 1970, 92, 5809 – 5810; d) L. P. Reiff, H. S. Aaron, J. Am.
Chem. Soc. 1970, 92, 5275 – 5276; e) R. A. Lewis, K. Mislow, J.
Am. Chem. Soc. 1969, 91, 7009 – 7012; f) O. Korpiun, R. A.
Lewis, J. Chickos, K. Mislow, J. Am. Chem. Soc. 1968, 90, 4842 –
4846; g) T. L. Emmick, R. L. Letsinger, J. Am. Chem. Soc. 1968,
90, 3459 – 3465; h) O. Korpiun, K. Mislow, J. Am. Chem. Soc.
1967, 89, 4784 – 4786.
[3] a) Y. Zhou, G. Wang, Y. Saga, R. Shen, M. Goto, Y. Zhao, L.-B.
Han, J. Org. Chem. 2010, 75, 7924 – 7927; b) G. Wang, R. Shen,
Q. Xu, M. Goto, Y. Zhao, L.-B. Han, J. Org. Chem. 2010, 75,
3890 – 3892; c) Q. Xu, C.-Q. Zhao, L.-B. Han, J. Am. Chem. Soc.
2008, 130, 12648 – 12655; d) L.-B. Han, C.-Q. Zhao, J. Org.
Chem. 2005, 70, 10121 – 10123; e) L.-B. Han, C.-Q. Zhao, S.-y.
Onozawa, M. Goto, M. Tanaka, J. Am. Chem. Soc. 2002, 124,
3842 – 3843.
[4] G. Gatineau, L. Giordano, G. Buono, J. Am. Chem. Soc. 2011,
133, 10728 – 10731, and references therein.
[5] O. Berger, C. Petit, E. L. Deal, J.-L. Montchamp, Adv. Synth.
Catal. 2013, 355, 1361 – 1373.
[6] a) O. Berger, L. Gavara, J.-L. Montchamp, Org. Lett. 2012, 14,
3404 – 3407. For the cleavage of (hydroxymethyl)phosphine –
boranes, see: b) K. Nagata, S. Matsukawa, T. Imamoto, J. Org.
Chem. 2000, 65, 4185 – 4188.
[7] a) K. Bravo-Altamirano, J.-L. Montchamp, Org. Synth. 2008, 85,
96 – 105; b) K. Bravo-Altamirano, J.-L. Montchamp, Org. Lett.
2006, 8, 4169 – 4171.
[8] The absolute configuration of (RP)-10 was determined by
comparison with the product obtained from the cross-coupling
of (RP)-2 with cinnamyl acetate.
[9] K. Bravo-Altamirano, L. Coudray, E. L. Deal, J.-L. Montchamp,
Org. Biomol. Chem. 2010, 8, 5541 – 5551.
[10] T. Nakai, K. Mikami, Org. React. 1994, 46, 105 – 209.
[11] a) L. K. Kuo, S. K. Glazier, Inorg. Chem. 2012, 51, 328 – 335;
b) Y. Kobayashi, F. Morisawa, K. Saigo, J. Org. Chem. 2006, 71,
606 – 615; c) R. Allahyari, P. W. Lee, G. H. Y. Lin, R. M. Wing,
T. R. Fukuto, J. Agric. Food Chem. 1977, 25, 471 – 478; d) M.
Mikołajczyk, M. Leitloff, Russ. Chem. Rev. 1975, 44, 670 – 686;
e) H. S. Aaron, J. Braun, T. M. Shryne, H. F. Frack, G. E. Smith,
R. T. Uyeda, J. I. Miller, J. Am. Chem. Soc. 1960, 82, 596 – 598;
f) H. S. Aaron, T. M. Shryne, J. I. Miller, J. Am. Chem. Soc. 1958,
80, 107 – 110.
[12] Y. Koide, A. Sakamoto, T. Imamoto, Tetrahedron Lett. 1991, 32,
3375 – 3376.
ˇ
[13] a) M. Stankevic, K. M. Pietrusiewicz, J. Org. Chem. 2007, 72,
816 – 822; b) D. Moraleda, D. Gatineau, D. Martin, L. Giordano,
G. Buono, Chem. Commun. 2008, 3031 – 3033.
[14] a) T. Yamagishi, J.-i. Mori, T. Haruki, T. Yokomatsu, Tetrahe-
dron: Asymmetry 2011, 22, 1358 – 1363; b) C. Nowlan, Y. Li, J. C.
Hermann, T. Evans, J. Carpenter, E. Ghanem, B. K. Shoichet,
F. M. Raushel, J. Am. Chem. Soc. 2006, 128, 15892 – 15902; c) K.
Shioji, Y. Kurauchi, K. Okuma, Bull. Chem. Soc. Jpn. 2003, 76,
833 – 834; d) P. Kiełbasinski, M. Albrycht, J. Łuczak, M. Miko-
łajczyk, Tetrahedron: Asymmetry 2002, 13, 735 – 738.
[15] CCDC 952809 (C11H23O3P), 952808 (C17H27O3P), and 952807
(C20H31O3P) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data request/cif.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Asymmetric Synthesis
A great leap forward toward the general
synthesis of P-stereogenic compounds:
Heating H3PO2 with (À)-menthol and
paraformaldehyde gives easily crystal-
lized menthyl hydroxymethyl-H-phosphi-
nate (1). From this product, virtually any
P-stereogenic compound can be synthe-
sized (see picture).
O. Berger,
J.-L. Montchamp*
&&&& — &&&&
A General Strategy for the Synthesis of
P-Stereogenic Compounds
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!