Palladium-catalyzed benzylic substitution of benzyl carbonates with phosphorus nucleophiles

Article abstract of DOI:10.1246/cl.170901

A wide range of benzyl carbonates reacted with dimethyl phosphonate or diphenylphosphine oxide in the presence of the palladium catalyst, [Pd(η³-allyl)Cl]₂DPEphos, to give dimethyl benzylphosphonates and benzyldiphenylphosphine oxides in high yields. The catalytic phosphonylation was applied to the one-pot synthesis of alkenes from the benzyl esters.

Full text of DOI:10.1246/cl.170901

CL-170901
Received: September 24, 2017 | Accepted: October 2, 2017 | Web Released: October 7, 2017
Palladium-catalyzed Benzylic Substitution of Benzyl Carbonates with Phosphorus Nucleophiles
Yusuke Makida,1 Kazumi Usui,1 Satoshi Ueno,1,2 and Ryoichi Kuwano*1
1Department of Chemistry, Faculty of Science, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395
2Department of Applied Chemistry, School of Engineering, Tokyo University of Technology,
1404-1 Katakuramachi, Hachioji, Tokyo 192-0982
(E-mail: rkuwano@chem.kyushu-univ.jp)
A wide range of benzyl carbonates reacted with dimethyl
phosphonate or diphenylphosphine oxide in the presence of the
palladium catalyst, [Pd(η3-allyl)Cl]2-DPEphos, to give dimethyl
benzylphosphonates and benzyldiphenylphosphine oxides in
high yields. The catalytic phosphonylation was applied to the
one-pot synthesis of alkenes from the benzyl esters.
Table 1. Effects of solvent and ligand on the palladium-
catalyzed nucleophilic substitution of 1a with 2aa
5.0% [Pd]
5.5% ligand
Ph
OCOMe
O
Ph
P(OMe)2
+
HP(O)(OMe)2
O
1a
2a
3a
X
PPh2
PPh2
Keywords: Palladium catalyst
| Benzylic substitution |
Ph2P
Fe
PPh2
Phosphorus nucleophiles
n-1
O
Ph2P
PPh2
DPPE (n = 2)
Organophosphorus compounds are of great importance in
various phases of organic chemistry, e.g., Wittig and Horner-
Wadsworth-Emmons reagents,1 chiral ligands,2 enzyme inhib-
itors,3 and emitting dyes.4 A variety of reactions have been
developed for the preparation of organophosphorus compounds.5
Among them, the substitution of organohalides with phosphorus
nucleophiles through transition-metal catalysis is advantageous
in terms of functional group compatibility, because the reactions
proceed well under mild conditions.5g,5h,5k However, the organo-
halides have a great negative impact on the environment as
well as the ecosystem. Toward solution of this drawback, the
unreactive C-O bond is an attractive alternative to the C-X
(X = halogen) bond. However, the use of ethers or carboxylates
is not well studied for catalytic C-P bond formation.6
DPPPent (n = 5)
DPPF
DPEphos (X = 2H)
Xantphos (X = CMe2)
Temp.
/°C
Time
/h
Yield
/%b
Entry Solvent
Ligand
1
2
3
4
5
DMSO
DMF
tAmOH
EtOAc
THF
DPEphos
DPEphos
DPEphos
DPEphos
DPEphos
DPEphos
Xantphos
DPPF
80
80
80
80
80
80
80
80
80
80
80
80
100
120
5
5
5
11
39
25
2
5
5
5
5
4
5
2
6
7
8
9
10
11c
12d
13d
14d
toluene
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
5
5
16
6
DPPPent
DPPE
We have developed nucleophilic substitutions of benzyl
5
5
0
0
carbonates through palladium catalysis,7,8 which involves
PPh3
electrophilic (η3-benzyl)palladium as a key intermediate.9
A
DPEphos
DPEphos
DPEphos
48
24
5
63
76e
63e
variety of nucleophilic compounds work as the reaction partners
of the benzyl esters. However, the use of phosphorus nucleo-
philes has been virtually unexplored for the benzylic substitu-
tion, while some phosphorus nucleophiles are often used for
the nickel- or palladium-catalyzed allylic and propargylic
substitutions.10 As a related study, the reaction of benzyl
alcohols with hypophosphorous acid had been developed with
a palladium catalyst by Montchamp. The reaction requires an
excess amount of the phosphorus nucleophile to obtain the
desired products in good yields.11,12 Moreover, Chen and Han
recently used a benzyl ester as the electrophilic substrate in their
study on the nickel-catalyzed C-O/P-H coupling reaction, but
the substrate scope was limited to (2-naphthyl)methyl pivalate.6e
Here, we have successfully developed the palladium-catalyzed
benzylic substitution with dimethyl phosphonate or diphenyl-
phosphine oxide. It is noteworthy that the latter nucleophile
efficiently provides a broad range of benzyldiphenylphosphine
oxides, which are useful for the synthesis of phosphine
ligands2,13 or as the substrates of Horner-Wittig reaction.1,14
The reaction of benzyl methyl carbonate (1a) with dimethyl
phosphonate (2a) was carried out in DMSO at 80 °C with
[Pd(η3-allyl)Cl]2-DPEphos catalyst (Table 1, Entry 1). The
reaction conditions are optimal for the palladium-catalyzed
aUnless otherwise noted, all reactions were carried out
with 0.20 mmol of 1a and 0.24 mmol of 2a in 1.0 mL of
solvent under N2; [Pd] = 0.5[Pd(η3-allyl)Cl]2. bDetermined by
c
d
1H NMR analysis. 11.0% of PPh3 was used. 1.0 mmol of 1a
and 1.2 mmol of 2a were used with 1.0% of [Pd] and 1.1% of
e
DPEphos. Yield of isolated 3a.
benzylic sulfinylation reported by us.7c Formation of the desired
benzylphosphonate 3a was detected in the reaction mixture by
1H NMR analysis, but the yield of 3a was only 11%. The yield
was improved by conducting the reaction in other polar solvents,
such as DMF or tert-amyl alcohol (Entries 2 and 3). In
particular, the use of DMF resulted in the highest yield.
Meanwhile, the benzylic C-P bond formation scarcely took
place in less polar solvents (Entries 4-6). The choice of
phosphine ligand on the palladium is crucial for the efficient
production of 3a (Entries 7-11). DPPF-palladium complex can
work as the catalyst for the benzylic phosphonylation. However,
no formation of 3a was observed when the palladium was
modified with a chelate bisphosphine bearing small bite angle or
1814 | Chem. Lett. 2017, 46, 1814–1817 | doi:10.1246/cl.170901
© 2017 The Chemical Society of Japan

Table 2. Nucleophilic substitution of benzyl carbonates 1 with
dimethyl phosphonate 2aa
Table 3. Nucleophilic substitution of benzyl carbonates 1 with
diphenyphosphine oxide 5a
1.0% [Pd]
3.0% [Pd]
1.1% DPEphos
3.3% DPEphos
Ar
OCOMe
O
Ar
P(OMe)2
O
Ar
OCOMe
O
Ar
PPh2
O
+ HP(O)(OMe)2
+
HP(O)Ph2
DMF, 100°C
DMF, 100°C, 24 h
1
2a
3
1
5
6
Time
/h
24
24
24
24
24
24
Yield
/%b
74
Time
/h
Yield
/%b
Entry
1
3
Entry
1
1
6
3b: R = Me
3c: R = OMe
3d: R = CO2Me
3e: R = CF3
3f : R = F
1
2
3
4
5
6
1b
1c
1d
1e
1f
Ph
PPh2
1a
48
91
77
P
O
6a
6b: R = Me
6c: R = OMe
6d: R = CO2Me
6e: R = CF3
6f : R = F
69
71
70
65
2
3
4
5
6
7
1b
1c
1d
1e
1f
24
24
24
24
48
24
96
94
97
94
91
98
R
P
P = P(O)(OMe)2
3g: R = Cl
1g
R
MeO
P(OMe)2
P = P(O)Ph2
7
1h
24
85
O
6g: R = Cl
3h
1g
Me
Me
8
1i
1j
48
96
8
1i
24
77
PPh2
P(OMe)2
O
O
6i
3i
Me
Me
9c
96
50
9c
1j
72
3
68
78
PPh2
P(OMe)2
O
O
6j
Me
Me
Me
Me
3j
10c
11
1k
1l
24
24
88
96
10c
1k
PPh2
P(OMe)2
O
6k
O
3k
PPh2
aUnless otherwise noted, all reactions were carried out with
1.0 mmol of 1 and 1.2 mmol of 2a in 1.0 mL of DMF at 100 °C
under N2; [Pd] = 0.5[Pd(η3-allyl)Cl]2. bYield of isolated 3.
c5.0% of [Pd] and 5.5% of DPEphos were used.
O
6l
PPh2
O
12c
1m
24
92
6m
PPh2
O
monophosphine. As a result of the further optimization of
reaction parameters, the product 3a was isolated in 76% yield,
when the mixture of 1a and 2a was heated at 100 °C for 24 h
with 1% palladium loading (Entry 13). Although the related
catalytic allylic substitutions with phosphorus nucleophiles are
commonly conducted in the presence of a stoichiometric
base,10a,10b,10d,10j the addition of any organic or inorganic bases
failed to improve the present benzylic substitution. The bases
mostly cause a deterioration of the reproducibility. The substrate
1a was completely consumed within 5 h, when the reaction was
conducted at 120 °C. However, the higher reaction temperature
caused the formation of unidentified side products (Entry 14).
As described in Table 2, various benzylphosphonates were
synthesized through the benzylic phosphonylation with 2a by
using the DPEphos-palladium catalyst. Both electron-donating
and electron-withdrawing substituents on the para position have
little effect on the reactivity of the benzylic esters (Entries 1-6).
The halogen functionalities (F and Cl) were tolerant of the
palladium catalysis. The halogenated substrates 1f and 1g were
converted to 3f and 3g in good yields without formation of the
dehalogenated product. As with the para-substituted benzyl
substrates, 3-methoxybenzyl ester 1h gave the phosphonylated
product 3h in 85% yield (Entry 7). The palladium catalyst is
useful for the benzylic phosphonylation of ortho-substituted
benzyl esters (Entries 8 and 9). The substrate having one ortho-
methyl group (1i) was comparable in reactivity to 1a. However,
aUnless otherwise noted, all reactions were carried out with
0.50 mmol of 1 and 0.60 mmol of 5 in 1.0 mL of DMF at
100 °C under N2; [Pd] = 0.5[Pd(η3-allyl)Cl]2. bYield of
isolated 6. c1.2 mmol of 5, 6.0% of [Pd], and 6.6% of
DPEphos were used.
the methyl groups of 1j caused significant decrease in the
reaction rate. The sufficient conversion of 1j to 3j required
longer reaction time (72 h) as well as higher catalyst loading
(5%). Naphthylmethyl ester 1k also reacted with 2a to give the
desired product 3k (Entry 10).
Some phosphorus compounds other than dimethyl phos-
phonate 2a worked as the nucleophile for the catalytic benzylic
substitution. Diethyl phosphonate 2b can react with 1a to give
the benzyl phosphonates in good yields. However, the C-P bond
formation was accompanied by a remarkable amount of trans-
esterification (eq 1). The benzylation of diphenylphosphine
oxide 5 also proceeded well to afford benzyldiphenylphosphine
oxide 6a in 98% yield (eq 2). The catalyst loading can be
reduced to 3% without significant loss of the yield (Table 3,
Entry 1). No substitution was observed when diphenyl phos-
phonate and dibutylphosphine oxide were employed as the
nucleophile. The former phosphorus compound might be
insufficient in nucleophilicity for the attack on the (η3-
benzyl)palladium intermediate because of the more electron-
Chem. Lett. 2017, 46, 1814–1817 | doi:10.1246/cl.170901
© 2017 The Chemical Society of Japan | 1815

withdrawing phenyl group.15 The latter nucleophile might
poison the palladium catalysis because its Lewis basicity is so
strong that the oxygen atom tightly interacts with the metal
center.
P
P
Ar
PR2
O
Pd0
Ar
OCOMe
O
3 or 6
A
1
R2P–
5.0% [Pd]
OR1
O
2 or 5
P
P
OR2
Pd+
5.5% DPEphos
CO2
R
Ph
OCOMe
O
Ph
P
+ HP(O)(OEt)2
MeO–
DMF, 100°C, 24 h
+
O
B
C
ð1Þ
4 (R1=R2=Et), 58%
1a
2b
MeOH
(0.5 mmol)
(0.6 mmol)
R
4’ (R1=Et, R2=Me), 24%
3a (R1=R2=Me), 5%
P
P
P
P
= DPEphos
Pd
5.0% [Pd]
MeO
5.5% DPEphos
Ph
PPh2
O
Ph
OCOMe
+ HP(O)Ph2
ð2Þ
DMF, 100°C, 24 h
O
Scheme 2. Possible mechanism of the palladium-catalyzed
benzylic substitution of 1 with 2 or 5.
1a
(0.5 mmol)
5
6a, 98%
(0.6 mmol)
Diphenylphosphine oxide (5) also reacted with a wide range
of benzyl carbonates 1 to give the phosphinylated products in
high yields. The substrate scope is summarized in Table 3.
Electron-rich and electron-deficient benzyl carbonates 1b-1g
were transformed into the benzyldiphenylphosphine oxides 6b-
6g in high yields (Entries 2-7). The ortho-substituent of 1i had
no effect on the formation of 6i (Entry 8). However, the reaction
of 1j was sluggish and failed to produce 6j in high yield even
with higher catalyst loading (6%) and prolonged reaction time
(96 h) (Entry 9). Naphthylmethyl esters 6k and 6l were also
phosphinylated with 5 in high yields (Entries 10 and 11). Two
phosphinyl groups were installed on the dicarbonate of m-xylene
glycol to give 6m in 92% yield (Entry 12).
In summary, we successfully developed the nucleophilic
substitution of benzyl carbonates with dimethyl phosphonate or
diphenylphosphine oxide by using the palladium catalyst, which
was generated in situ from [Pd(η3-allyl)Cl]2 and DPEphos.
The benzylic C-P bond formation smoothly proceeded in the
presence of a small excess amount of phosphorus nucleophiles
without use of base. In addition, a variety of benzyldiphenyl-
phosphine oxides, which were beneficial synthons for phosphine
ligands13 or used for Horner-Wittig olefination,14 can be
obtained in practically quantitative yields through the palladium
catalysis. Moreover, synthetic utility of the catalytic benzylic
C-P bond formation was highlighted by the one-pot synthesis of
E-alkenes from the benzyl carbonate.
The present catalytic C-P bond formation is applicable to
the one-pot synthesis of β-substituted styrenes from benzyl
carbonate 1c in combination with Horner-Wadsworth-Emmons
reaction (Scheme 1). After the palladium-catalyzed phosphonyl-
ation of 1c with 2a, a solution of NaOtBu in DMF was carefully
added at 0 °C to the mixture containing 3c in the presence of an
aldehyde 7. As a result, the alkenes 8 were obtained from the
above reactions in good yields with excellent E-selectivity.
A possible mechanism of the catalytic substitution of
benzyl carbonates with phosphorus nucleophiles is depicted in
Scheme 2.9 The reaction is initiated by the oxidative addition
of the benzylic C-O bond of 1 to palladium(0) species A with
releasing carbon dioxide. The resulting (η3-benzyl)palladium
intermediate B reacts with deprotonated 2 or 5 by methoxide
anion to give the phosphonylated or phosphinylated product.
The η3-benzyl ligand in B can isomerize to η1-benzyl to form
less electrophilic C in equilibrium. The formation of 3 or 6
should proceeds through the attack of the phosphorus nucleo-
phile to the intermediate B. However, the η3-coodination in B
would be unfavorable for the (benzyl)palladium species gen-
erated from 1j, because its two ortho-methyl groups might
sterically disturb the η3-coodination. Therefore, the reactivity of
1j was much lower than those of other benzyl esters.
This work was supported by Grant-in-Aid for Young
Scientists (B) (JSPS KAKENHI Grant Number JP17K14450).
We thank the Cooperative Research Program “Network Joint
Research for Material and Devices” for HRMS measurements.
Supporting Information is available on http://dx.doi.org/
10.1246/cl.170901.
This paper is dedicated to the late Professor Yoshihiko Ito
on the occasion of the 10th anniversary of his sudden death.
References and Notes
1
2
B. E. Maryanoff, A. B. Reitz, Chem. Rev. 1989, 89, 863.
a) R. Noyori, Angew. Chem., Int. Ed. 2002, 41, 2008.
b) D. H. Valentine, Jr., J. H. Hillhouse, Synthesis 2003, 2437.
c) R. C. J. Atkinson, V. C. Gibson, N. J. Long, Chem. Soc.
Rev. 2004, 33, 313.
3
a) P. Kafarski, B. Lejczak, Phosphorus, Sulfur Silicon Relat.
Elem. 1991, 63, 193. b) M. Collinsova, J. Jiracek, Curr. Med.
Chem. 2000, 7, 629.
4
5
T. Baumgartner, R. Réau, Chem. Rev. 2006, 106, 4681.
a) F. Alonso, I. P. Beletskaya, M. Yus, Chem. Rev. 2004, 104,
3079. b) A. L. Schwan, Chem. Soc. Rev. 2004, 33, 218. c) F.
Palacios, C. Alonso, J. M. de los Santos, Chem. Rev. 2005,
105, 899. d) J.-L. Montchamp, J. Organomet. Chem. 2005,
690, 2388. e) D. Enders, A. Saint-Dizier, M.-I. Lannou, A.
Lenzen, Eur. J. Org. Chem. 2006, 29. f) S. Van der Jeught,
C. V. Stevens, Chem. Rev. 2009, 109, 2672. g) C. S.
Demmer, N. Krogsgaard-Larsen, L. Bunch, Chem. Rev.
2011, 111, 7981. h) D. Zhao, R. Wang, Chem. Soc. Rev.
Ar
OCOMe
E/Z >99:1
O
R
O
1.0% [Pd]
1.1% DPEphos
7
1c
R
H
+
MeO
3c
NaOtBu
HP(O)(OMe)2
8a (R = Ph), 78%
8b (R = tBu), 71%
2a
Scheme 1. One-pot synthesis of β-substituted styrenes 8.
1816 | Chem. Lett. 2017, 46, 1814–1817 | doi:10.1246/cl.170901
© 2017 The Chemical Society of Japan

2012, 41, 2095. i) H. Yorimitsu, Beilstein J. Org. Chem.
2013, 9, 1269. j) J.-L. Montchamp, Acc. Chem. Res. 2014,
47, 77. k) I. Wauters, W. Debrouwer, C. V. Stevens, Beilstein
J. Org. Chem. 2014, 10, 1064.
Eur. J. Org. Chem. 2003, 4640. e) K. Bravo-Altamirano, Z.
Huang, J.-L. Montchamp, Tetrahedron 2005, 61, 6315. f) K.
Bravo-Altamirano, J.-L. Montchamp, Org. Lett. 2006, 8,
4169. g) K. Bravo-Altamirano, I. Abrunhosa-Thomas, J. L.
Montchamp, J. Org. Chem. 2008, 73, 2292. h) P. Butti, R.
Rochat, A. D. Sadow, A. Togni, Angew. Chem., Int. Ed.
2008, 47, 4878. i) M. Kalek, T. Johansson, M. Jezowska,
J. Stawinski, Org. Lett. 2010, 12, 4702. j) A. Duraud, O.
Jacquet, J.-C. Fiaud, R. Guillot, M. Toffano, ChemCatChem
2011, 3, 883. k) L. Zhang, W. Liu, X. Zhao, Eur. J. Org.
Chem. 2014, 6846.
6
For selected examples, see: a) D. A. Holt, J. M. Erb,
Tetrahedron Lett. 1989, 30, 5393. b) Y. Luo, J. Wu,
Organometallics 2009, 28, 6823. c) G. Yang, C. Shen, L.
Zhang, W. Zhang, Tetrahedron Lett. 2011, 52, 5032. d) C.
Shen, G. Yang, W. Zhang, Org. Biomol. Chem. 2012, 10,
3500. e) J. Yang, T. Chen, L.-B. Han, J. Am. Chem. Soc.
2015, 137, 1782. f) W. C. Fu, C. M. So, F. Y. Kwong, Org.
Lett. 2015, 17, 5906. g) J. Yang, J. Xiao, T. Chen, L.-B. Han,
J. Org. Chem. 2016, 81, 3911. h) L.-L. Liao, Y.-Y. Gui,
X.-B. Zhang, G. Shen, H.-D. Liu, W.-J. Zhou, J. Li, D.-G.
Yu, Org. Lett. 2017, 19, 3735.
a) R. Kuwano, Y. Kondo, Y. Matsuyama, J. Am. Chem. Soc.
2003, 125, 12104. b) R. Kuwano, Y. Kondo, Org. Lett. 2004,
6, 3545. c) R. Kuwano, Y. Kondo, T. Shirahama, Org. Lett.
2005, 7, 2973. d) M. Yokogi, R. Kuwano, Tetrahedron Lett.
2007, 48, 6109. e) R. Kuwano, H. Kusano, Chem. Lett.
2007, 36, 528. f) R. Kuwano, H. Kusano, Org. Lett. 2008,
10, 1979.
11 L. Coudray, J.-L. Montchamp, Eur. J. Org. Chem. 2008,
4101.
12 With stoichiometric Lewis acid, Wiemer developed the
nucleophilic substitution of benzyl alcohol with triethyl
phosphite, see: a) R. J. Barney, R. M. Richardson, D. F.
Wiemer, J. Org. Chem. 2011, 76, 2875. b) R. M. Richardson,
R. J. Barney, D. F. Wiemer, Tetrahedron Lett. 2012, 53,
6682.
13 For an example, see: I. P. Beletskaya, A. V. Chuchurjukin,
H. P. Dijkstra, G. P. M. van Klink, G. van Koten,
Tetrahedron Lett. 2000, 41, 1075.
7
8
9
For a review, see: R. Kuwano, Synthesis 2009, 1049.
For reviews on metal η3-benzyl complexes, see: a) B. M.
Trost, L. C. Czabaniuk, Angew. Chem., Int. Ed. 2014, 53,
2826. b) J. Le Bras, J. Muzart, Eur. J. Org. Chem. 2016, 2565.
14 a) L. Horner, H. Hoffman, H. G. Wippel, Chem. Ber. 1958,
91, 64. b) L. Horner, H. Hoffman, H. G. Wippel, G. Klahre,
Chem. Ber. 1959, 92, 2499.
15 A reviewer pointed out that the reaction with diphenyl
phosphonate might be affected by the impurities, phenol and
phenyl phosphonate, from the hydrolysis of the phosphorus
nucleophile. In particular, phenol may be exchanged with the
methoxide in B. The resulting phenoxide may be insufficient
in basicity for the deprotonation of diphenyl phosphonate.
10 a) J.-C. Fiaud, J. Chem. Soc., Chem. Commun. 1983, 1055.
b) X. Lu, J. Zhu, J. Huang, X. Tao, J. Mol. Catal. 1987, 41,
235. c) R. Malet, M. Moreno-Mañas, R. Pleixats, Synth.
Commun. 1992, 22, 2219. d) A. Galkina, A. Buff, E. Schulz,
L. Hennig, M. Findeisen, G. Reinhard, R. Oehme, P. Welzel,
Chem. Lett. 2017, 46, 1814–1817 | doi:10.1246/cl.170901
© 2017 The Chemical Society of Japan | 1817