Synthesis of atropisomeric chiral MeOBIPHEP analogues via Pd-catalyzed P-C coupling - applications to asymmetric Rh-catalyzed C-C bond formations in water

Article abstract of DOI:10.1016/j.catcom.2015.06.005

Abstract The preparation and catalytic applications of water-soluble MeOBIPHEP-based atropisomeric chiral congener ligands are described. The optimization of the catalytic system to promote the key P-C coupling revealed the superiority of palladium cataly

Full text of DOI:10.1016/j.catcom.2015.06.005

Catalysis Communications 69 (2015) 129–132
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short communication
Synthesis of atropisomeric chiral MeOBIPHEP analogues via Pd-catalyzed
P–C coupling — applications to asymmetric Rh-catalyzed C–C bond
formations in water
Lucie Leseurre a, Florent Le Boucher d'Herouville a, Anthony Millet a, Jean-Pierre Genêt a,
a,
Michelangelo Scalone b, , Véronique Michelet
⁎
⁎
a
PSL Research University, Chimie ParisTech, CNRS, Institut de Recherche de Chimie Paris, 75005 Paris, France
Process Research & Development CoE Catalysis, F. Hoffmann-La Roche AG, CH-4070 Basel, Switzerland
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The preparation and catalytic applications of water-soluble MeOBIPHEP-based atropisomeric chiral congener
ligands are described. The optimization of the catalytic system to promote the key P–C coupling revealed the
superiority of palladium catalysts versus nickel or copper. Spectroscopic studies of the Pd-catalyzed reaction
medium showed the presence of phosphorylated intermediates. The asymmetric Rh-catalyzed 1,4-addition of
boronic acids to enones in water was described in the presence of water-soluble MeOBIPHEP derivatives. The
4-CO2Na-, 3-CO2Na- and 3,5-(CO2Na)2-substituted MeOBIPHEP analogues showed high and complementary
reactivities and enantioselectivities, considering either the enone or the boronic acid.
Received 14 May 2015
Received in revised form 5 June 2015
Accepted 6 June 2015
Available online 9 June 2015
Keywords:
Palladium
Rhodium
© 2015 Elsevier B.V. All rights reserved.
Atropisomeric ligands
Asymmetric catalysis
1,4-addition
Water-soluble ligands
1. Introduction
because it generally relied on complicated multistep syntheses [13,14].
On the other hand, due to its obvious economic and environmental ad-
The demand for efficient synthetic methodologies proceeding with
atom economy has been increasing constantly along with the need of
optically active compounds from the chemical industry [1]. In this
field, chiral ligands, and more specifically atropisomeric ligands since
the discovery of atropisomeric binaphthyl BINAP, have proven to be
outstanding chiral inducers [2,3]. The main strategies for their synthesis
involved a Grignard addition to a phosphane intermediate [4], a Pd- or
Ni-catalyzed phosphorus–carbon bond formation starting from
bistriflate derivatives [5,6] or a chiral phosphonate and an aryl Grignard
reagent [7,8]. These three strategies have been highly successful and
have led to the development of an important number of new ligands.
However, drawbacks such as resolution of racemate and reduction of
phosphane oxide often cannot be avoided. Our ongoing research pro-
gram on asymmetric metal-catalyzed reactions [9,10] prompted us to
develop a novel synthesis of atropisomeric ligands starting from simple
chiral primary bisphosphonate derivative [11,12]. Whereas a flurry of
chiral ligands has been developed by original methods, the develop-
ment of chiral hydrophilic ligands has been less studied, most probably
vantages, the use of water as a reaction medium is still challenging in
modern chemistry [15–17]. Chiral diphosphanes bearing sulfonated
water-soluble moieties [18,19], in particular MeOBIPHEP analogues
developed at Roche [7] have for example been used in aqueous media
hydrogenation reactions. We wish to report therein that our study on
the optimization of the catalytic system enables to promote the synthe-
sis of water-soluble MeOBIPHEP analogues and their catalytic activities
towards asymmetric metal-catalyzed carbon–carbon bond formation
reactions in water.
2. Results and discussion
The primary phosphane (R)-1 was available on a 20 gram scale, via a
quantitative reduction of the Roche industrially prepared bisphospho-
nate [12]. Table 1 compiles the attempts for Ni-, Cu- and Pd-catalyzed
P–C couplings of (R)-1, the reactions being followed by 31P NMR
(see supplementary material for selected spectra). The first attempt
for the phosphorus–carbon coupling using a Ni catalyst [20] in the
presence of (R)-1 and 4-methylphenyltriflate gave poor results as only
traces of the desired product was detected (entry 1). Considering the
high temperature for such coupling, we turned our attention to copper
catalysts [21,22]. Under the reported reaction conditions for the
⁎
Corresponding authors.
E-mail addresses: michelangelo.scalone@roche.com (M. Scalone),
veronique.michelet@chimie-paristech.fr (V. Michelet).
http://dx.doi.org/10.1016/j.catcom.2015.06.005
1566-7367/© 2015 Elsevier B.V. All rights reserved.

130
L. Leseurre et al. / Catalysis Communications 69 (2015) 129–132
Table 1
The use of di-isopropylethylamine was beneficial to the (R)-2/(R)-3
Metal-catalyzed P–C coupling.
ratio and the desired product was isolated in 61% yield (entry 20). The
catalytic amount of the ligand was optimized, the best result being ob-
served using 8 mol% of diphenylphosphino ferrocene, 4 equivalents of
di-isopropylethylamine, and 4 equivalents of methyl 4-iodobenzoate
(entries 20–22). Increasing the amount of aryliodide did not give
much better results (entry 23), whereas slow addition of the iodide
partner allowed the formation of the desired ligand in 83% isolated
yield and an excellent selectivity (entry 24). These observations and
the results reported in the literature [23–25] led us to postulate the
mechanism described in Scheme 1. The active Pd(0) species would be
generated by analogy to the Pd(OAc)2/PPh3 system [26], where an ex-
cess of reductant (amine or phosphane) present in the reaction media
would enable the transformation of Pd(II) into Pd(0). The oxidative ad-
dition of the aryl iodide would lead to the intermediate B that would
then undergo a ligand exchange. The nucleophilic assistance of the
base would allow the iodide ligand to be exchanged with a phosphido
group from the primary phosphane 1 to generate the intermediate C.
A reductive elimination step would free the ligand D1 and regenerate
the active Pd(0) specie A. The repetition of this process would allow
the insertion of four aromatic rings on the phosphorus atoms to give
ligand 2. Noteworthy that the intermediate species containing only
1–3 aryl substituents on phosphorus do not have to de-coordinate to
continue in the catalytic cycle but could well remain coordinated.
The formation of the side product 3 could be explained by
a dehydro-coupling step, observed in the literature on simpler
phosphanes [27]. Based on the results described in the literature, the
dehydro-coupling step between the two phosphorus atoms would
occur after the insertion of an aromatic ring on D1 intermediate leading
to complex E, which would evolve towards the bis-phosphido species F.
The steric vicinity between the phosphorus atoms on intermediate F
would then enable the dehydro-coupling leading to 3.
Facing these encouraging results, we determined the optical purity
of ligand 2 and, disappointingly, observed that starting from an
enantiopure (ee N 95%) phosphonate total racemization had occurred
leading to the formation of rac-2. The racemization process seemed
more likely to occur during the P–C coupling step at 80 °C than during
the reduction step of the bisphosphonate into (R)-1 at −78 °C. The
Pd-catalyzed coupling was monitored by 31P NMR (see supplementary
material for details) revealed, as anticipated, the presence of several in-
termediates. A slow and progressive addition of (R)-1 to the reaction
media allowed its rapid consumption with no racemization process.
We selected para- and meta-ester substituted aryl iodides for the prep-
aration of water-soluble ligands (Scheme 2). Under the optimized reac-
tion conditions, ligands (R)-4, (R)-5, and (R)-6 were isolated in good to
excellent yields (76–88%) along with excellent enantioselectivities
(N98% ee) [11].
In the presence of trifluoroacetic acid, the tert-butyl esters of ligands
(R)-4–6 were quantitatively transformed into carboxylic acids [12]. The
corresponding salts were obtained by the addition of sodium hydride
allowing the synthesis of water-soluble ligands (R)-7a–c in high overall
yields. In order to evaluate the influence of the sodium carboxylate
water-soluble groups, we also prepared the corresponding alcohols
and sodium alcoholates (R)-8a–c via a reduction/deprotonation
sequence.
There are still few examples of asymmetric metal-catalyzed carbon–
carbon bond formation reactions in water [12–19], which encouraged
us to test the prepared water-soluble ligands in the Rh-catalyzed addi-
tion of phenylboronic acid to enones (cyclopentenone, cyclohexenone
and cycloheptenone, Table 2) [28,29]. The choice of this Rh-catalyzed
reaction was also motivated by the fact that both organoaqueous and
aqueous conditions were compatible with several atropisomeric ligands
[30,31]. Most of the chiral MeOBIPHEP analogues led to similar results
under classic conditions compared to BINAP ligand [32]. We selected
some conditions involving Na2CO3 in water as we and others have
shown that it can accelerate the reaction kinetic [33–35]. We screened
Entry Catalyst (mol%)
Ligand (mol%)
Solvent Base
2/3
Yield
of 2b
(%)a
1c
2
NiCl2(dppe) (10)
CuI (20)
CuI (20)
CuI (20)
CuI (20)
/
DMF
DABCO
/
/
8
/
DMEDA (40)
DMEDA (40)
DMEDA (40)
DMEDA (20)
/
/
/
/
/
/
/
/
/
/
Toluene Cs2CO3
100/0
0/100
100/0 15
100/0 18
100/0 15
13/87
60/40
85/15 30
97/3
33/67
100/0 24
73/27 17
68/32
3d
4d
5
Toluene
/
Toluene Cs2CO3
Toluene Cs2CO3
Toluene Cs2CO3
6
CuI (20)
7
8
9
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(PPh3)4 (4)
Pd(PPh3)4 (4)
Pd(PPh3)4 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
Pd(OAc)2 (4)
DMF
Et3N
/
/
Toluene Et3N
DMA
CH3CN
DMF
Et3N
Et3N
Et3N
10
11
12
13
14
15
16
17
18
19
20
21
22
23d
24e
31
5
Toluene Et3N
CH3CN
CH3CN
CH3CN
CH3CN
Et3N
n-Bu3N
i-Pr2NEt 85/15 41
Et3N
Et3N
Et3N
Et3N
i-Pr2NEt 90/10 61
i-Pr2NEt 60/40
i-Pr2NEt 40/60
i-Pr2NEt 100/0 57
i-Pr2NEt 94/6 83
/
PPh3 (12)
P(t-Bu)3HBF4 (12) CH3CN
95/5
0/100
47/53
31
/
/
dppe (8)
dppf (8)
dppf (8)
dppf (4)
dppf (12)
dppf (8)
dppf (8)
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
81/19 40
/
/
Determined by 31P NMR.
Isolated yield.
4-Me-C6H4OTf was used instead of ArI, 130 °C.
6 equivalents of ArI.
a
b
c
d
e
Slow addition of ArI.
synthesis of monophosphanes, we observed the disappearance of
starting material but the desired product was isolated only in a low 8%
yield (entry 2). The presence of the base was crucial as the reaction
led to the formation of another product, biarylphosphine (R)-3 (9% iso-
lated yield), instead of the desired derivative (R)-2 in the absence of
Cs2CO3 (entry 3). Modifying the reaction conditions, such as the number
of equivalents of aryl iodide or DMEDA ligand led to similar results (en-
tries 4–6). The formation of the desired product was observed in con-
comitance with several non-identified species, such that (R)-2 could
be isolated in 15% to 18% yield. We therefore turned our attention to
palladium-catalyzed P–C couplings [23–25]. The use of palladium
diacetate in the presence of triethylamine afforded a mixture of the de-
sired product (R)-2 and the compound (R)-3 in various ratios depend-
ing on the solvent (DMF, toluene, DMA or acetonitrile) (entries 7–10).
The best results were obtained in DMA and acetonitrile, giving the
highest 2/3 ratio, but leading to several by-products and disappointingly
low isolated yields. The use of Pd(PPh3)4 instead of Pd(OAc)2 did
not give better results (entries 11–13). Other bases such as tri-n-
butylamine and di-isopropylethylamine were also tested in lieu of
triethylamine and an encouraging 41% isolated yield was obtained in
the latter case (entries 14–15). The addition of triphenylphosphane or
hindered tri-tert-butylphosphine (entries 16–17) did not lead to better
results. The use of a bidentate ligand such as diphenylphosphinoethane
afforded approximately a 1/1 mixture of phosphanes (R)-2 and (R)-3
(entry 18), whereas the use of diphenylphosphino ferrocene had a
consistent positive influence on the reaction outcome (entry 19).

L. Leseurre et al. / Catalysis Communications 69 (2015) 129–132
131
Scheme 1. Reaction mechanism proposal.
the efficiency of ligands (R)-7a–c and (R)-8a–c to evaluate the
influence of the water-soluble groups and their position relative to the
aromatic ring of the phosphorus ligand.
of the 3,5-disubstituted ligand (R)-7c. The corresponding arylated
ketone 10 was isolated in 96% enantiomeric excess (entry 5). In the
case of cycloheptenone, we performed the addition of the phenylboronic
acid to the enone with all the three carboxylated ligands and observed
that the best ligand was this time the mono-meta-substituted one (R)-
7a (entry 1, compound 11). The 7-membered ring ketone 11 was
isolated in 96% enantiomeric excess instead of 90% in the presence of
(R)-7c (entry 5). Pleasingly, each of the three carboxylated-substituted
ligands could therefore be valorized for Rh-catalyzed arylation of
5-membered to 7-membered ring enones.
We then studied the influence of the aryl boronic acid in the Rh-
catalyzed arylation in water and more specifically the influence of the
electronic and steric effects for the cyclohexenone (Scheme 3). Boronic
acids bearing electron-donating or withdrawing groups (MeO, Cl and
CF3) reacted smoothly under the same conditions with cyclohexenone
The 1,4-addition of phenylboronic acid to enones proved to be very
substrate- and ligand-dependent. In all cases, the higher efficiency of
the ligands bearing a carboxylated group such as (R)-7a–c, was demon-
strated comparing to those bearing an alcoholate moiety (R)-8a–c
(entry 1 versus 2, entry 2 versus 3 and entry 5 versus 6). In the case of
cyclopentenone, the para-substituted ligand (R)-7b was found to be
the best water-soluble ligand as the corresponding ketone was isolated
in 99% yield and 90% ee (entry 3, compound 9). Increasing the hindrance
on the aromatic ring of the phosphorus atoms afforded a slow decrease
of the enantiomeric excess (entries 1 and 5, compound 9). The reaction
of cyclohexenone was less surprising and followed the expected ten-
dency: the highest enantiomeric excess was obtained in the presence
Scheme 2. Synthesis of water-soluble ligands.

132
L. Leseurre et al. / Catalysis Communications 69 (2015) 129–132
Table 2
3. Conclusion
Rh-catalyzed addition of phenylboronic acid to enones.
We have extended the methodology developed for the preparation of
MeOBIPHEP-based atropisomeric chiral congener ligands. The optimiza-
tion of the catalytic system to promote the key P–C coupling revealed
the superiority of palladium catalysts versus nickel or copper. During
the study of the asymmetric Rh-catalyzed 1,4-addition of boronic acids
to enones in water, the best ligands were found to be the carboxylated-
substituted ones. The 4-CO2Na-, 3-CO2Na- and 3,5-(CO2Na)2-substituted
MeOBIPHEP analogues showed high and complementary reactivities
and enantioselectivities, regarding either the enone or the boronic acid.
Further studies will be dedicated to the study of other catalytic applica-
tions of substituted ligands based on the MeOBIPHEP skeleton.
Acknowledgments
Entry
1
(R)-L*
9 (% yield, % ee)
10 (% yield, % ee)
11 (% yield, % ee)
This work was supported by the Centre National de la Recherche
Scientifique and ENSCP Chimie ParisTech. L.L. and F.L.B.H. are grateful to
the CNRS and F. Hoffmann-La Roche AG for financial grants. Thanks to Jo-
seph Schneider, Marcel Althaus and teams (all F. Hoffmann-La Roche AG,
Basel) for analytical support.
CO2Na
(R)-7a
CH2ONa
(R)-8a
CO2Na
(R)-7b
CH2ONa
(R)-8b
CO2Na
(R)-7c
CH2ONa
(R)-8c
89, 88
99, 95
83, 96
2
3
4
5
6
95, 76
99, 90
97, 60
99, 82
97, 80
99, 64
99, 93
99, 68
99, 96
99, 84
78, 54
86, 75
85, 38
99, 90
91, 80
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.doi.
org/10.1016/j.catcom.2015.06.005.
References
[1] H.U. Blaser, E. Schmidt, Asymmetric Catalysis on Industrial Scale, Wiley-VCH Verlag
VmbH & Co., Weinheim, 2004.
[2] H. Shimizu, I. Nagasaki, T. Saito, Tetrahedron 61 (2005) 5405.
[3] J.-P. Genêt, T. Ayad, V. Ratovelomanana-Vidal, Chem. Rev. 114 (2014) 2824.
[4] H. Takaya, K. Mashima, K. Koyano, M. Yagi, H. Kumobayashi, T. Taketomi, S.
Akutagawa, R. Noyori, J. Org. Chem. 51 (1986) 629.
[5] D. Cai, J. Payack, D.R. Bender, D.L. Hughes, T.R. Verhoeven, P.J. Reider, J. Org. Chem. 59
(1994) 7180.
[6] Y. Uozumi, A. Tanahashi, S. Lee, T. Hayashi, J. Org. Chem. 58 (1993) 1945.
[7] R. Schmid, E.A. Broger, M. Cereghetti, Y. Crameri, J. Foricher, M. Lalonde, R.K. Muller,
M. Scalone, G. Schoettel, U. Zutter, Pure Appl. Chem. 68 (1996) 131.
[8] V. Michelet, M. Scalone, Electronic Encyclopedia of Reagents for Organic Synthesis, in:
L. Paquette, P. Fuchs, D. Crich, P. Wipf (Eds.), John Wiley & Sons, 2015, http://dx.doi.
org/10.1002/047084289X.rn00832.pub2.
[9] C.-M. Chao, M. Vitale, P.Y. Toullec, J.-P. Genêt, V. Michelet, Chem. Eur. J. 15 (2009) 1319.
[10] C.-M. Chao, D. Beltrami, P.Y. Toullec, V. Michelet, Chem. Commun. 45 (2009) 6988.
[11] L. Leseurre, F. Le Boucher d'Herouville, K. Puntener, M. Scalone, J.-P. Genêt, V. Michelet,
Org. Lett. 13 (2011) 3250.
substrate in the presence of ligand (R)-7c and led to the corresponding
ketones 12–15 with enantiomeric excesses superior to 80%, the best re-
sult being observed in the case of the methoxy- and trifluoromethyl-
substituted boronic acid (94% and 93% ee respectively). Either lower
enantiomeric excesses or lower yields were observed in the presence
of the para-carboxylated ligand (R)-7b except in the case of a hindered
boronic acids such as the 1-naphthylboronic acid. The use of the
hindered ligand (R)-7c afforded the corresponding ketone 15 in 75%
ee whereas the use of the para-substituted ligand (R)-7b gave higher
yield and enantiomeric excess, 69% and 82% respectively.
[12] L. Leseurre, K. Püntener, J.-P. Genêt, M. Scalone, V. Michelet, Adv. Synth. Catal. 353
(2011) 3269.
[13] D. Sinou, Adv. Synth. Catal. 344 (2002) 221.
[14] K.H. Shaughnessy, Chem. Rev. 109 (2009) 643.
[15] H.C. Hailes, Org. Process Res. Dev. 11 (2007) 114.
[16] P. Dixneuf, V. Cadierno, Metal-catalyzed Reactions in Water, Wiley-VCH, Weinheim,
Germany, 2013.
[17] P. Anastas, C.-J. Li, Handbook of Green Chemistry Series: Water as a Green Solvent,
John Wiley & Sons, New-York, 2011.
[18] H. Kumobayashi, European Patent EP 0544455, 1992.
[19] A.E.S. Gelpke, J.N.J. Veerman, M.S. Goedheijt, P.C.J. Kamer, P.W.N.M. van Leuwen, H.
Hiemstra, Tetrahedron 55 (1999) 6657.
[20] T.F. Knöpfel, P. Aschwanden, T. Ichikawa, T. Watanabe, E.M. Carreira, Angew. Chem. Int.
Ed. 43 (2004) 5971.
[21] D. Gelman, L. Jiang, S.L. Buchwald, Org. Lett. 5 (2003) 2315.
[22] K. Tani, D.C. Behenna, R.M. McFadden, B.M. Stoltz, Org. Lett. 9 (2007) 2529.
[23] C. Korff, G. Helmchen, Chem. Commun. (2004) 530.
[24] O. Herd, A. Heβler, M. Hingst, M. Tepper, O. Stelzer, J. Organomet. Chem. 522 (1996) 69.
[25] T. Oshiki, T. Imamoto, J. Am. Chem. Soc. 114 (1992) 3975.
[26] C. Amatore, A. Jutand, M.A. M'Barki, Organometallics 11 (1992) 3009.
[27] P. Butti, R. Rochat, A.D. Sadow, A. Togni, Angew. Chem. Int. Ed. 47 (2008) 4878.
[28] C.-G. Feng, Z.-Q. Wang, C. Shao, M.-H. Xu, G.-Q. Lin, Org. Lett. 10 (2008) 4101.
[29] Y. Otomaru, T. Senda, T. Hayashi, Org. Lett. 6 (2004) 3357.
[30] G. Berthon-Gelloz, T. Hayashi, in: D.G. Hall (Ed.), 2nd ed., Boronic Acids Preparation and
Applications in Organic Synthesis, Medicine and Materials, 263, Wiley-VHC, Weinheim,
Germany, 2011 (Chapter 5).
[31] P. Tian, H.-Q. Dong, G.-Q. Lin, ACS Catal. 2 (2012) 95.
[32] F. Le Boucher d'Herouville, A. Millet, M. Scalone, V. Michelet, J. Org. Chem. 76 (2011)
6925.
[33] R. Amengual, V. Michelet, J.-P. Genêt, Tetrahedron Lett. 43 (2002) 5905.
[34] R. Amengual, V. Michelet, J.-P. Genêt, Synlett (2002) 1791.
[35] R. Itooka, Y. Iguchi, N. Miyaura, J. Org. Chem. 68 (2003) 6000.
Scheme 3. Influence of the arylboronic acid.