Nitrile assisted, Bronsted acid catalyzed regio and stereoselective diarylphosphonylation of allyl silyl ethers

Article abstract of DOI:10.1039/c001660h

We have discovered a mild, catalytic protocol for the regio- and stereoselective synthesis of trisubstituted allyl diarylphosphonates from the corresponding disubstituted allyl silyl ethers, circumventing the challenges related to the preparation and availability of stereodefined trisubstituted olefins. A closely related arylation reaction was also discovered during the methodology development. By simply switching the reaction medium, high phosphonylation/arylation ratios and vice versa can be achieved. This may not be a direct result of changing solvent polarity. The allyl diarylphosphonates were evaluated as carboxylesterase inhibitors, and the screening results revealed that the inhibitory efficiency is highly related to the choice of alkenes and aryl substituents. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c001660h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Nitrile assisted, Brønsted acid catalyzed regio and stereoselective
diarylphosphonylation of allyl silyl ethers†
Chun-Yu Ho,*a Chun-Wa Chan,a Siu-Kwan Wo,b Zhong Zuo*b and Lai-Ying Chana
Received 26th January 2010, Accepted 19th May 2010
First published as an Advance Article on the web 8th June 2010
DOI: 10.1039/c001660h
We have discovered a mild, catalytic protocol for the regio- and stereoselective synthesis of
trisubstituted allyl diarylphosphonates from the corresponding disubstituted allyl silyl ethers,
circumventing the challenges related to the preparation and availability of stereodefined trisubstituted
olefins. A closely related arylation reaction was also discovered during the methodology development.
By simply switching the reaction medium, high phosphonylation/arylation ratios and vice versa can be
achieved. This may not be a direct result of changing solvent polarity. The allyl diarylphosphonates
were evaluated as carboxylesterase inhibitors, and the screening results revealed that the inhibitory
efficiency is highly related to the choice of alkenes and aryl substituents.
with high olefin stereoselectivity and b-substituent flexibility,
Introduction
with a view to obtaining a good platform for subsequent b-g-
position asymmetric functionalization. Herein we describe a novel
synthetic strategy for that preparation by using triarylphosphites
and allyl silyl ethers directly obtained from the corresponding
Ni-catalyzed coupling.20,21 Interestingly, by simply switching the
solvent system employed, various stereodefined trisubstituted allyl
arylation products can also be obtained (a Friedel–Crafts like
reaction), which provides mechanistic insights for both reactions
and should simulate further developments in this area (Scheme 1).
Tetracoordinate phosphoryl groups are recognized as excellent
mimics of the tetrahedral transition state of ester and amide
hydrolysis.1,2 Recently, strategies for preparing related stereo-
defined allyl phosphonates have received increasing attention,3–7
providing a good platform from which to introduce chiral elements
close to the phosphonate functionality through a wide range
of enantioselective olefin transformations8–12 and to synthesize
dienes through the HWE reaction.13–15 Although many diarylphos-
phonates have been identified as potent enzyme inhibitors, the
synthesis of allyl diarylphosphonates is much less studied than that
of their dialkyl relatives. Current methods for stereodefined allyl
phosphonate synthesis with concomitant formation of a new C–P
bond and an olefin generally focus on the addition of dialkyl phos-
phites (RO)2P(O)H to acrylate derived Baylis–Hillman adducts
or rearrangement of dialkyl phosphonyl protected adducts.16–18
However, we are aware of no related examples for allyl di-
arylphosphonates, probably because of the lower nucleophilicity
of aryl phosphites, and the synthesis involves a very different
hydrolysis mechanism.3c Also, one of the inherent problems of
the above approaches is that they have limited choices of b-
substituents, generally bearing an electron-deficient group and
yielding acrylic esters. Olefin cross-metathesis provides a powerful
method of making allyl phosphonates, but the stereoselectivity of
the allyl phosphonate group has been reported to be problematic
when unsymmetrical higher substituted olefins were used.19 As
part of a programme evaluating phosphonates for chemical and
biological applications (carboxylesterase inhibitors), we decided to
develop a new method for allyl diarylphosphonate preparations
Scheme 1 Allyl phosphonylation versus arylation. P : A refers to the ratio
of total phosphonylation isomers to total arylation isomers.
Result and discussion
Using the Ni-catalyzed alkene-aldehyde-silyl triflate coupling
product 1a (Table 1) reported previously by one of the authors here
and Jamison20a with P(OPh)3 as reaction partners, a systematic
investigation led to the identification of the reaction conditions
(Scheme 2). Our first attempts to synthesize the phosphonates
involved heating the two reaction components together, which
is analogous to other procedures commonly used for allyl di-
alkylphosphonate synthesis, summarily failed.16a–16e Inspired by
Spilling’s pioneering work that used a stoichiometric amount
of mild Lewis acidic silylating reagent (BSA) to facilitate the
phosphonylation of acrylate derived Baylis–Hillman adducts with
dialkyl phosphite,16f and the protic acid to accelerate dialkyl
phosphonylation of conjugated imines reported by Stevens,16g
aCenter of Novel Functional Molecules, The Chinese University of Hong
Kong, Shatin, New Territories, Hong Kong SAR. E-mail: jasonhcy@
cuhk.edu.hk; Tel: +852-26096342
bSchool of Pharmacy, Faculty of Medicine, The Chinese University
of Hong Kong, Shatin, New Territories, Hong Kong SAR. E-mail:
joanzuo@cuhk.edu.hk; Tel: +852-26096832
† Electronic supplementary information (ESI) available: Experimental
details including screening of the reaction conditions, preparation of
substrates, characterization of products and phosphonylation inhibitor
screening results. See DOI: 10.1039/c001660h
3480 | Org. Biomol. Chem., 2010, 8, 3480–3487
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Catalytic stereoselective and regioselective synthesis of allyl diarylphosphonates.a
c
2 : 2¢
R1
R2
R3
H
R4
Ph
Yield (%)b
88
2 E : Zc
1
1a
n-Hex
n-Bu
p-C6H4OMe
3.0 : 1
E only
3.0 : 1
E only
3.5 : 1
E only
4.3 : 1
E only
4.0 : 1
E only
1 : 1.2
13.0:1
3.0 : 1
E only
3.3 : 1
E only
4.0 : 1
E only
1 : 1.7
E only
1 : 2.7
E only
2.0 : 1
E only
2.0 : 1
E only
1.5 : 1
1.3:1
2d
80
3e
79
4f
77
5
1b
1c
1d
1e
1f
CH2i-Pr
c-Hex
Ph
90
6
99
7
86
8
PhCH2
BnCH2
n-Hex
78
9
86
10g
11
12g
13g
14g
15g
16h
17
18
19
1g
1h
1i
o-C6H4OMe
o,p-C6H3(OMe)2
p-C6H4Me
Ph
61
85
52
1j
52
1k
1l
p-C6H4Cl
18
m-C6H4OMe
p-C6H4NMe2
p-C6H4OMe
p-C6H4OMe
35
1.5 : 1
3.0:1
1 : 1
E only
1 : 5.7
E only
3.3 : 1
E only
4.0 : 1
E only
1m
1n
1a
92
Me
Ph
H
90
n-Bu
p-C6H4Me
p-C6H4Cl
90
99
See Scheme 2 for structures and Experimental section for details.a Typical conditions: 0.1 mmol substrate, 500 mol% P(OPh)3, 5 mol% p-TsOH·H2O,
2 mL CH3CN. b Isolated yield of phosphonylation (2+2¢, average of at least two runs, separable by column chromatography). c Determined by 1H NMR,
olefin stereochemistry was determined by NOESY. d 3 times the scale. e Using 2 mL PhCN. f Using 2 mL i-PrCN. g Using 100 mol% p-TsOH·H2O, run at
35 ◦C for 18 h. h Using 105 mol% of p-TsOH·H2O.
They can be separated by typical silica gel column chromatography
without any decomposition. This procedure allowed in situ hydrol-
ysis of the phosphonium salts and the typical strictly anhydrous
condition employed for the phosphonylation step can be avoided.
However, this modification alone is not a practical method for
Scheme 2 Catalytic stereoselective trisubstituted olefin synthesis.
allyl diaryl phosphonates synthesis. An unexpected allyl arylation
pathway (or aromatic allylation, giving 3a) is operating as a major
side reaction with phosphonylation : arylation (P : A) equal to
38 : 62 (Scheme 2).22–25
We suspected that the formation of the arylation product could
be attributable to the phenol generated after the phosphonylation
step and the acid catalyzed phosphite hydrolysis.26 Extensive
variation of physical parameters including acidic additives stoi-
chiometry, rate/order of addition and reaction temperature failed
to favor the phosphonylation pathway. We thus required an
p-toluenesulfonic acid was selected as a promoter for our di-
arylphosphonylation. The hope was to enhance the electrophilicity
of our silyl allylic components toward weakly nucleophilic tri-
arylphosphites. To our delight, by using only a catalytic amount
of TsOH in neat conditions and simply at room temperature in
open air, the desired allyl diarylphosphonates were obtained. This
consisted of 2 different regio-isomers (2a, 2¢a), favoring the one
forming an E-trisubstituted olefin (Table 1, see ESI for details).
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3480–3487 | 3481
©

View Article Online
alternative means of overcoming this problem and acquiring
phosphonylation more easily.
CH3NO2 and DMF. Together with the change in phosphonylation
regioselectivity observed when a more bulky nitrile source was
employed as reaction medium (see later section), all of the above
indicated that the superiority of using CH3CN may not entirely
be the result of a higher P(OPh)3 nucleophilicity in a more polar
solvent, suggesting the nitrile functionality may play a substantial
role.
As the phosphite hydrolysis in aqueous organic solution may
increase on account of the interaction of water molecules with
the organic medium,27 we tried to use toluene as the reaction
medium on the presumption that it might help to slow down
phosphite hydrolysis. Unfortunately, this strategy failed to favor
the phosphonylation pathway, and strongly favored the arylation
pathway. Next, we shifted our attention to look for additives
that might facilitate the nucleophilic phosphonylation but that
would also be compatible with the phosphonium ion hydrolysis
step. Noting that CH3CN was frequently employed to modulate
nucleophilic displacement reaction pathways, such as in the
preparation of various glycoside linkages by forming nitrilium
conjugates and the Ritter reaction, we tested CH3CN as an
additive for our phosphonylation.28 We were mindful of the fact
that this approach could result in faster phosphite hydrolysis and
provide a higher phenol availability for the arylation, and could
also incorporate CH3CN into the substrate in a manner similar
to the N-glycosyl amine formation. Nevertheless, the CH3CN
was found to be a superior additive for the phosphonylation.
This additional modification reversed the selectivity observed when
toluene or P(OPh)3 was used as reaction medium and provided a
new system for allyl diarylphosphonate preparation with good
yield (P : A = 87 : 13). This remarkable result prompted us to
screen other solvents as reaction medium further for completeness,
but none promoted the phosphonylation pathway to the extent
that CH3CN did (Scheme 3, see ESI for details). Other solvents
examined included CHCl3, Acetone, THF, CH3NO2, DMF and
NEt3. In fact, we also could not find any direct correlation between
the product selectivity with the solvent polarity or dielectric
constant. In particular, CH3NO2 and DMF are solvents having
similar polarity or dielectric constant with CH3CN (with er ~
35–37),29 however, experimental results showed that they strongly
favored the undesired arylation pathway, and the phosphonylation
was observed only in minor amounts (P : A = 24 : 76; 14 : 86). In
addition, when PhCN was employed as the medium, which has a
much lower dielectric constant (with er ~ 26), the reaction outcome
(P : A = 81 : 19) sharply contrasted with those obtained with
We propose a possible scenario here which may account for
these unusual observations and this is shown in Scheme 4. The
nitrile employed may first provide a nitrilium ion intermediate
with the allyl silyl ether under a tosyl acid accelerated addition,
which is analogous to the Ritter reaction. The weakly nucleophilic
triarylphosphite may regenerate the nitrile at this point, and
provide the phosphonium ion intermediate. Followed by an in
situ hydrolysis step in open air atmosphere, the desired allyl
diarylphosphonate was obtained. Slightly higher 2 : 2¢ ratios were
observed in reactions using a more bulky nitrile source and
may be a result of a more selective formation of nitrilium ion
owing to steric considerations. This change in phosphonylation
regioselectivity also supports the theory that this is not a direct
SN1 or SN1¢ reaction. It should be noted that we could not observe
any nitrilium ion formation directly by spectroscopic techniques
and could not isolate any Ritter products, thus the reason(s) for
the selectivity improvement (P : A and regioselectivity) remains
elusive. Further studies revealed that P(OPh)3 is the nucleophile
responsible for the phosphonylation. Control experiments showed
that the HP(O)(OPh)2 which resulted from P(OPh)3 hydrolysis
did not produce any desired phosphonylation product under the
standard TsOH/CH3CN conditions, but the arylation product
was obtained in good yield (a result of further hydrolysis of
HP(O)(OPh)2).
Scheme 4 A possible rationale for the unusual observation: nitrile assisted
catalytic allyl diarylphosphonylation.
Scope of allyl diarylphosphonylation
After further optimization, we found that only 5 mol% of
TsOH·H2O was necessary, with a ratio of P : A further improved
to 93 : 7. With this new protocol in hand, the scope and generality
of the system was explored, with the results summarized in
Table 1. This new system generally allowed a highly selective
allyl diaryl phosphonylation over arylation, provided excellent
olefin stereoselectivity and expanded the scope particularly at the
b-position (R1).
Scaling up the reaction or employing more sterically encum-
bered substrates, such as R1 = a-branch, b-branch, benzyl and
aryl, did not compromise the high phosphonylation to arylation
ratio, high olefin stereoselectivity and yield of phosphonylation
(P : A = 7.3–19; E : Z ≥ 93 : 7; 78–99% yield; entries 2–9). Also,
the trisubstituted olefin 2 is generally favored over the terminal
olefin 2¢ by a factor of 3–4 in this series except in case of
Scheme 3 Remarkable observations by using nitriles as medium. P : A
refers to the ratio of total phosphonylation isomers to those in arylation.
3482 | Org. Biomol. Chem., 2010, 8, 3480–3487
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
1c. Moderate improvement in 2 : 2¢ can be obtained when i-
PrCN was employed as reaction medium, but the yield was
diminished considerably (entries 1 and 4, from 3.0 : 1 to 4.3 : 1).
Although an o-anisyl group is required as directing group, the
phosphonylation regioselectivity can be swapped to favor 2¢
without sacrificing the high P : A ratio (entries 1 vs. 10, 11). The
possibility of forming a benzopyran-like intermediate may account
for this interesting observation. We found that the yields of the
desired phosphonylation were decreased considerably when less
electron-rich substrates were employed (entries 12–15), but the
substrate with R2 bearing basic aryl/alkyl amine substituent can
still undergo the phosphonylation reaction with high yield and
required no specialized protecting group (entry 16).
While the effect of additional alkene substituents (R3 π H) in
approaches using Baylis–Hillman adducts on yield and selectivity
was not examined,16 it should be noted that a higher substituted
allyl silyl ether (R3 = Ph) synthesized by Ni-catalyzed alkyne-
aldehyde coupling21 was found also a compatible substrate for
this new method, providing a-branched allyl diarylphosphonate in
good yield and selectivity (entry 17). Furthermore, substituents on
the triarylphosphite are well-tolerated (R4), providing additional
flexibility in synthesis when necessary (entries 18, 19).
Several notable features of this new methodology deserve
further comment. Strictly anhydrous conditions, high reaction
temperature, a stoichiometric amount of additive, or strongly
acidic medium are required most of the time in the allyl phos-
phonylation, particularly for those with concomitant formation
of a new C–P bond and an olefin.1f,16b–16f Our catalytic system
did not require a strictly anhydrous medium to proceed, allowed
in situ hydrolysis of the highly unstable allyl diarylphosphonium
intermediate without isolation or delicate hydrolysis steps, and
can be conducted at r.t. using only a catalytic amount of additive,
overall highlighting their technical advantages and providing good
functional group tolerance. Existing methods of stereoselective
allyl phosphonate synthesis generally focused on acrylate derived
Baylis–Hillman adducts and alkyl phosphites, which results in
the dialkyl phosphonates having an electronically activated Z-
trisubstituted olefin (Scheme 5). The new approach and findings
here complement the existing methods in several respects, allowing
diaryl derivatives to be synthesized with an electronically neutral
E-trisubstituted olefin, and expanding the scope particularly at the
b-position. In addition, subsequent trials showed that the Baylis–
Hillman adduct diarylphosphonylation can also be done with
81% yield by using our tosyl acid–CH3CN combination at 35 ◦C
(2 : 2¢ = 3.2 : 1.). Notably, the olefin stereochemistry preference
(E : Z = 18 : 82) were found opposite to what we observed in
Table 1, and was found parallel to those observed in typical allyl
dialkylphosphonylations, these experimental results highlighted
the important role of R1 substituents in these reactions (see ESI for
details). The stereodefined olefins obtained may offer a correlated
foundation for enantioselective olefin transformation or HWE
reactions.
A variety of organophosphorus compounds were identified as
potent human blood monocyte carboxylesterase (CBE) inhibitors,
especially O P(OPh)3 and P(OPh)3.30 However, we are aware
=
of no study using allyl phosphonates being reported. Therefore
our newly synthesized allyl diarylphosphonates were evaluated as
carboxylesterase inhibitors in rat plasma (see ESI for details).31,32
The results showed that our newly synthesized allyl diarylphos-
phonates were potent carboxylesterase inhibitors with a broad
range of inhibitory efficiency (IC50 = 4.99–0.52 mM), revealing
that the alkene inclusion (or/and the substituents) may provide a
positive effect on inhibition and providing room for fine tuning
when necessary. Allyl diarylphosphonates with much stronger
inhibitory effects than P(OPh)3 were identified quickly through
our small chemical library developed here and provided us with
new directions for the development of a new generation of
carboxylesterase inhibitors in our programme.
Scope of allyl arylation.
Next, we decided to study the closely related arylation using
the same silyl allyl ether starting material based on the findings
discovered in preliminary screening. This arylation method is
attractive to us, not only because it may provide insight on
how to improve the P : A selectivity, but also because it yields
only one single regioisomer over all the other 5 possible isomers
(terminal/carbinol position of allyl silyl ethers and o-/m-/p- of the
aromatic nucleophile) in a single operation, together with excellent
trisubstituted olefin stereoselectivity, C- to O-allylation ratio and
no multiple allylations observed on phenol or on the allyl silyl
ether (Table 2, entry 1).22–24
Although unable to exclude direct arylation between the allyl
silyl ethers and the phosphite, we believe that the arylation product
was formed mainly from the reaction between the substrates and
phenol. The control experiments showed that using phenol in place
of P(OPh)3 with 1a also provided the arylation product successfully
with no oxy-allylation product observed under otherwise the same
reaction conditions, suggesting that phenol is the actual substrate
for the aromatic allylation and protecting the O with a P atom is not
necessary. Also, in an experiment where P(OPh)3, HP(O)(OPh)2
and phenol were allowed to run in parallel, the reaction employing
phenol provided the highest yield with a reduced reaction time,
which also supports the hypothesis. In view of these control
experiments and the observations of phosphonylation, we believe
the change in the P : A ratio observed after switching the solvent
employed is similar to that incurred by adding two different
reagents to carry out the experiment, one providing phenol and
one accelerated C–P bond formation, rather than a direct result of
the difference in solvent physical properties. Therefore, the change
is not an example of solvent-dependent chemoselectivity between
an sp2 carbon center and a phosphorus atom on the P(OPh)3.
With the genuine aryl nucleophile identified, various allyl silyl
Scheme 5 Complementarity of allyl phosphonylation of acrylates and
allyl silyl ethers. Remarkable substituent effect on olefin selectivity.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3480–3487 | 3483
©

View Article Online
Table 2 Stereoselective and regioselective arylation of allyl silyl ethersa
Yield (%)b
Regioselectivity ratioc,e
ArOH
E : Zc,d
p
o
m
1
1a
1b
1c
1d
1e
1f
P(OPh)3 or (phenol)
85 or 92
E only
93
E only
98
E only
40
Z only
100
E only
91
E only
73
E only
99
E only
54
100
100
21
—
—
77
—
—
—
28
39
47
92
—
—
—
—
—
—
—
—
—
8
2
3
4
100
100
100
72
5
6
7f
8f
9f
10g
1g
1i
61
1l
53
E only
92
E only
1a
—
11g
12g
13g
98
E only
—
—
65
100
100
35
—
—
—
80
E only
99
E only
14h
86
E only
100
—
—
No phosphonylation product was observed in all cases examined when using P(OPh)3; see Scheme 2 for structures and Experimental section for
details.a Typical conditions: 0.1 mmol substrate, 500 mol% P(OPh)3, 20 mol% p-TsOH·H2O, 2 mL toluene. b Combined yield of p-, o- and m-substituted
products (average of at least two runs, isolated and separable by column chromatography). c Determined by 1H NMR analysis. d Determined by NOESY,
for all regioisomers. e Regioselectivity relative to the hydroxyl group. f Using 40 mol% of catalyst. g Using 1.5 equivalents of alcohols in place of phosphite.
h Using 5 equivalent of anisole in place of phosphite.
ethers and several aromatic alcohols were subjected to the TsOH–
toluene conditions (Table 2).
substituted with a meta-Cl group (entry 13). Typical aromatic
allylation substrates such as aryl alkyl ether (entry 14, anisole), was
found less reactive than phenol but was also allylated successfully
in this system, providing an ethereal compound that should also
be accessible from 3a.
The success of stereodefined trisubstituted allyl phenol (or
trisubstituted styrenes) synthesis found here is informative to
the field of allylic alkylation of aromatic compounds. It should
be noted that most of the aromatic allylation reports in the
literature employed cinnamyl alcohol types of substrates (trans-
disubstituted alkene) or symmetrical substrates for aromatic ally-
lation, and leaving the effect of an additional alkene substituent on
yield and selectivity not well-examined (Scheme 6). We surmised
that this observation is probably due to the limited availability of
To our delight, the reactions generally proceeded with high
yields by using a slight excess of aromatic alcohols, with good
arylation regioselectivity and olefin stereoselectivity. Less electron-
rich substrate 1i can be used by increasing the catalyst loading
(entry 8). When the para-position of the phenol was blocked by
an additional substituent or difficult to access (OMe, Me, Cl), the
allylation can still take place with high yield at the ortho-position,
providing structures that were analogous to an allyl aryl ether
thermo Claisen rearrangement without sacrificing the high olefin
stereoselectivity and without increasing the reaction temperature
or catalyst loading (entries 10–12). The allylation was found more
favorable to take place at the para-position when the phenol was
3484 | Org. Biomol. Chem., 2010, 8, 3480–3487
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Experimental
p-Toluenesulfonic acid monohydrate and phosphorous trichloride
were purchased from Acros Organics and used without further pu-
rification. Triphenyl phosphite was purchased from International
Laboratory and used as received. Sodium hydride (60% in mineral
oil) was purchased from Panreac Sintesis. Triethylamine and
phosphorous trichloride were distilled over calcium hydride. All
solvents were purchased from LAB-SCAN and used as received.
Analytical thin layer chromatography (TLC) was performed using
EM Science silica gel 60 F254 plates. The developed chromatogram
was analyzed by UV lamp (254 nm), ethanolic phosphomolyb-
dic acid (PMA) or potassium permanganate (KMnO4). Liquid
chromatography was performed using a forced flow (flash chro-
matography) of the indicated solvent system on Silicycle Silica Gel
(230–400 mesh).1H, 13C and 31P NMR spectra were recorded on
Bruker 300 or 400 MHz spectrometers in CDCl3. High resolution
mass spectra were obtained on a Finnigan MAT 95XL GC Mass
Spectrometer by Miss H.-Y. Ng.
Scheme 6 Preparation of allyl phenolic compounds: 1,1-disubstituted
olefins as aromatic allylation precursors.
General procedure for allyl diarylphosphonylation and arylation
stereodefined trisubstituted allylic alcohols or their synthetic chal-
lenge. It should be noted that our method avoided that synthetic
challenge, simply forming the targets directly from Ni-catalyzed
coupling products (terminal disubstituted alkenes) through a
conjugated addition with concomitant formation of two new C–
C bonds with excellent regioselectivity and olefin stereoselectivity
outcome. The trisubstituted allyl arylation products may be also
difficult to prepare with high olefin selectivity by dehydration of the
corresponding alcohols or by olefin cross-metathesis (e.g. Table 2,
entry 5, benzyl vs p-hydroxybenzyl).
A mixture of 0.1 mmol substrate with 500 mol% of P(OPh)3 and
5 mol% p-TsOH·H2O in 2 mL of CH3CN (for phosphonylation)
or 20 mol% p-TsOH·H2O in 2 mL of toluene (for arylation) was
◦
stirred at 0 C for 2 h in open air and then stirred at r.t. for 9 h.
Solvent was removed under reduced pressure. The yield and the
selectivity were averages of at least two runs. Purification via flash
chromatography on silica gel afforded the desired product. The
stereochemistry of the olefin was determined by NOESY using
isolated product.
Representative spectroscopic data for allyl diaryl phosphonyla-
tion and arylation (see ESI for details):
Allyl diarylphosphonylation:
Conclusions
2a: 1H NMR (400 MHz, CDCl3) d: 7.33–7.29 (m, 5H), 7.19–7.10
(m, 7H), 6.86 (d, J = 8.7 Hz, 2H), 6.48 (d, J = 6.1 Hz, 1H), 3.82
(s, 3H), 3.05 (d, J = 22.3 Hz, 2H), 2.47–2.42 (m, 2H), 1.54–1.42
(m, 2H), 1.35–1.25 (m, 6H), 0.88–0.83 (m, 3H).
In summary, we have discovered two novel catalytic synthetic
strategies for the intermolecular stereo and regioselective allyl di-
arylphosphonylation and allyl arylation that complements existing
methods under mild conditions. The allyl silyl ethers obtained
from Ni-catalyzed aldehyde-alkene/alkyne coupling were first
identified as valuable precursors and seem generally useful for
highly stereoselective synthesis of functionalized trisubstituted
olefins by forming two new bonds, providing alternatives to
those strategies involving prior preparations of stereodefined
trisubstituted olefins bearing a good leaving group with a
regioselective substitution. The stereodefined alkenes obtained
here may be a useful platform for chiral center introduction at
the phosphonate’s and phenol’s b,g-positions by conventional
catalytic asymmetric olefin functionalization technologies. The
diaryl phosphonylation works best for electron-rich systems and
provided olefin stereoselectivity outcome contrasted with previous
work. The novel use of nitriles in terms of differentiating the
reaction pathways and accelerating phosphonylation is one of
the keys for the above accomplishments. The newly synthesized
allyl diarylphosphonates from this work were first identified as
potent carboxylesterase inhibitors and provided us with important
clues for further development. The results obtained here should
find applications in both chemical transformations and biological
studies.
31P NMR (121 MHz, CDCl3) d: 20.36.
2¢a: 1H NMR (400 MHz, CDCl3) d: 7.42 (d, J = 8.8 Hz, 2H),
7.31–7.02 (m, 8H), 6.86 (d, J = 8.8 Hz, 2H), 6.84 (d, J = 8.8 Hz,
2H), 5.69 (d, J = 2.8 Hz, 1H), 5.19 (d, J = 2.8 Hz, 1H), 4.06 (d,
J = 24.8 Hz, 1H), 3.80 (s, 3H), 2.07–1.90 (m, 2H), 1.41–1.37 (m,
2H), 1.36–1.20 (m, 6H), 0.89–0.83 (m, 3H).
31P NMR (121 MHz, CDCl3) d: 18.42.
Allyl arylation:
1
3a: H NMR (400 MHz, CDCl3) d: 7.15 (d, J = 8.6 Hz, 2H),
7.10 (d, J = 8.4 Hz, 2H), 6.85 (d, J = 8.6 Hz, 2H), 6.76 (d, J =
8.4 Hz, 2H), 6.21 (s, 1H), 5.05 (brs, 1H), 3.80 (s, 3H), 3.39 (s, 2H),
2.13 (t, J = 8.2 Hz, 2H), 1.47–1.41 (m, 2H), 1.28–1.20 (m, 6H),
0.86 (t, J = 7.1 Hz, 3H).
Acknowledgements
We thank the Center of Novel Functional Molecule, The Chinese
University of Hong Kong (2060366, 2060376) for financial sup-
port. C.-W. C. thanks The Chinese University of Hong Kong for
post-graduate studentship (3250049). L.-Y. C. thanks Chung Chi
College for an undergraduate Student Helper Award.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3480–3487 | 3485
©

View Article Online
12 Recent advance in catalytic asymmetric synthesis of a-hydroxy or amino
allyl phosphonates: (a) G. D. Joly and E. N. Jacobsen, J. Am. Chem.
Soc., 2004, 126, 4102; (b) X. Zhou, X. Liu, X. Yang, D. Shang, J. Xin
and X. Feng, Angew. Chem., Int. Ed., 2008, 47, 392; (c) J. P. Abell and
H. Yamamoto, J. Am. Chem. Soc., 2008, 130, 10521; (d) L. Albrecht, A.
Albrecht, H. Krawczyk and K. A. Jørgensen, Chem.–Eur. J., 2010, 16,
28; (e) K. Suyama, Y. Sakai, K. Matsumoto, B. Saito and T. Katsuki,
Angew. Chem., Int. Ed., 2010, 49, 797.
13 Allylphosphonates have been widely used in the preparation of dienes
and polyenes, an advantage of allylphosphonates in the olefination
over the corresponding phosphonium salts is exemplified by their
stereoselectivity and stereospecificity observed in the reactions: (a) B. E.
Maryanoff and A. B. Reitz, Chem. Rev., 1989, 89, 863; (b) R. S.
Edmundson, Chem. Funct. Groups, 1996, 4, 495; Condensation of
an allylic phosphonium ylide: (c) E. J. Corey and B. W. Erickson,
J. Org. Chem., 1974, 39, 821; Recent advance in the synthesis of densely
substituted 1,3-butadienes: (d) S. M. Date and S. K. Ghosh, Angew.
Chem., Int. Ed., 2007, 46, 386; Affording conjugated dienes with high
cis selectivity: (e) Q. Wang, M. El Khoury and M. Schlosser, Chem.–
Eur. J., 2000, 6, 420.
References
1 For general review on the organophosphorus oxide chemistry: (a) H.
Hays and D. J. Peterson, Organic Phosphorus Compounds, 1972, 3,
341; (b) A. K. Bhattacharya and G. Thyagarajan, Chem. Rev., 1981,
81, 415; (c) A. K. Bhattacharya and N. K. Roy, The Chemistry of
Organophosphorus Compounds, 1992, 2, 195; (d) T. Minami, Compre-
hensive Organic Functional Group Transformation, 1995, 2, 819; (e) T.
Minami, Comprehensive Organic Functional Group Transformations II,
2005, 853; (f) D. Enders, A. Saint-Dizier, M.-I. Lannou and A. Lenzen,
Eur. J. Org. Chem., 2005, 29; For arylphosphine oxides synthesis:
(g) C. A. Ramsden, Science of Synthesis, 2007, 31, 2035.
2 For recent reviews on phosphonic acids and their derivatives as
inhibitors: (a) K. Moonen, I. Laureyn and C. V. Stevens, Chem. Rev.,
2004, 104, 6177; (b) F. Palacios, C. Alonso and J. M. de los Santos,
Chem. Rev., 2005, 105, 899; (c) N. J. Wardle, S. W. A. Bligh and H. R.
Hudson, Curr. Org. Chem., 2005, 9, 1803; (d) L. Azema, R. Baron and
S. Ladame, Curr. Enzyme Inhib., 2006, 2, 61; (e) V. D. Romanenko and
V. P. Kukhar, Chem. Rev., 2006, 106, 3868.
3 (a) G. H. Birum, J. Org. Chem., 1974, 39, 209; (b) J. Oleksyszyn, L.
Subotkowska and P. Mastalerz, Synthesis, 1979, 985; (c) Mechanism
revised: M. Soroka, Liebigs Ann. Chem., 1990, 331; (d) H. R. Hudson,
F. Ismail, M. Pianka and C. W. Wan, Phosphorus, Sulfur Silicon Relat.
Elem., 2000, 164, 245; (e) P. Merino, E. Marques-Lopez and R. P.
Herrera, Adv. Synth. Catal., 2008, 350, 1195.
4 Pioneer example in the preparation of dialkyl allylphosphonate from
allyl halides and dialkylphosphite at reflux: (a) G. M. Kosolapoff, J. Am.
Chem. Soc., 1951, 73, 4040; Using (PhO)2P(OTMS): (b) R. Francke
and G. V. Roeschenthaler, Phosphorus Sulfur Relat. Elem., 1988, 36,
125; Modification employing TBAI/CsCO3 with monosubstituted allyl
bromide: (c) R. J. Cohen, D. L. Fox, J. F. Eubank and R. N. Salvatore,
Tetrahedron Lett., 2003, 44, 8617; Phenyl phosphorochloridite with
allyl halide: (d) I. Pergament and M. Srebnik, Org. Lett., 2001, 3, 217;
(e) H. Zhang, R. Tsukuhara, G. Tigyi and G. D. Prestwich, J. Org.
Chem., 2006, 71, 6061; Direct addition of P(OPh)3 to conjugated
imine catalyzed by SnCl4: (f) P. Van der Veken, I. El Sayed, J.
Joossens, C. V. Stevens, K. Augustyns and A. Haemers, Synthesis, 2005,
634.
14 Wittig olefination reaction and modifications involvingphosphoryl-
stabilized carbanions: B. E. Maryanoff and A. B. Reitz, Chem. Rev.,
1989, 89, 863.
15 (a) For selected publications on the Horner–Wadsworth–Emmons
reaction: W. S. Wadsworth, Jr, Organic Reactions, Wiley, New York,
1977, Vol. 25, pp 73; (b) E. L. Eliel, and S. H. Wilen, Stereochemistry
of Organic Compounds, Wiley, New York, 1994, Chapter 14.
16 (a) Recent advance in stereoselective allylphosphonate (phosphono
unsaturated esters) synthesis employing Baylis–Hillman adducts: W. R.
Schoen and W. H. Parsons, Tetrahedron Lett., 1988, 29, 5201; (b) T.
Janecki and R. Bodalski, Synthesis, 1990, 799; (c) D. Basavaiah
and S. Pandiaraju, Tetrahedron, 1996, 52, 2261; (d) M. Matziari,
D. Georgiadis, V. Dive and A. Yiotakis, Org. Lett., 2001, 3, 659;
(e) C. Muthiah, K. Senthil Kumar, J. J. Vittal and K. C. Kumara
Swamy, Synlett, 2002, 1787; (f) P. A. Badkar, N. P. Rath and C. D.
Spilling, Org. Lett., 2007, 9, 3619; (g) Typical condition: BSA, substrate,
dialkylphosphite in 1 : 1 : 1 ratio, heated for reflux for 2–4 days in
THF. The in situ formation of silyl phosphonates was often envisaged
as a key to enhance the nucleophilicity of the dialkylphosphites,
however, the Lewis acidic nature of silylating reagent may also
enhance the electrophilicity of carbonyls towards nucleophilic addition,
especially at higher temperature; (h) for the effect of protic acid on
alkylphosphonylation with conjugated imine: K. Moonen, E. Van
Meenen, A. Verwee and C. V. Stevens, Angew. Chem., Int. Ed., 2005,
44, 7407.
5 From vinylphosphonate via isomerization: base-catalyzed: (a) J. J.
Kiddle and J. H. Babler, J. Org. Chem., 1993, 58, 3572; Prototropic
equilibria of diethyl alkenylphosphonates: (b) J. P. Gerber, T. A. Modro,
C. C. P. Wagener and A. Zwierzak, Heteroat. Chem., 1991, 2, 643.
6 Using sulfone elimination strategy: Y. Vo-Quang, D. Carniato, L. Vo-
Quang, A. M. Lacoste, E. Neuzil and F. Le Goffic, J. Med. Chem.,
1986, 29, 579.
7 Pd-catalyzed hydrophosphorylation of monoallenes and 1,1-
disubstituted allenes: (a) C.-Q. Zhao, L.-B. Han and M. Tanaka,
Organometallics, 2000, 19, 4196; Highly substituted allenes do not
undergo the addition. 1,3-Dienes gave products through least hindered
reductive elimination: (b) F. Mirzaei, L.-B. Han and M. Tanaka,
Tetrahedron Lett., 2001, 42, 297; Atom-economical preparation of
related allyl phosphonic acids from concentrated hypophosphorous
acid and stereodefined trisubstituted allyl alcohols (no b-substituted
allyl alcohol example) using a Pd2dba3-xantphos catalyst at 85–110 ◦C,
please see: (c) K. Bravo-Altamirano and J.-L. Montchamp, Tetrahedron
Lett., 2007, 48, 5755.
8 Epoxidation of allylphosphonate: (a) T. R. Boehlow and C. D. Spilling,
Tetrahedron Lett., 1996, 37, 2717; (b) M. Mitchell, L. Qaio and C.-H.
Wong, Adv. Synth. Catal., 2001, 343, 596.
9 Dihydroxylation and regioselective amination: (a) T. Yamagishi, K.
Fujii, S. Shibuya and T. Yokomatsu, Synlett, 2004, 2505; With AD-mix
reagents: (b) T. Yokomatsu, T. Yamagishi, T. Sada, K. Suemune and
S. Shibuya, Tetrahedron, 1998, 54, 781; (c) T. Yamagishi, K. Fujii, S.
Shibuya and T. Yokomatsu, Tetrahedron, 2006, 62, 54.
10 Highly stereo- and regiocontrolled cyclopent-annulation: (a) G. Agnel
and E. Negishi, J. Am. Chem. Soc., 1991, 113, 7424; Asymmetric
Michael addition reaction: (b) S. E. Denmark and J.-H. Kim, J. Org.
Chem., 1995, 60, 7535; Regioselective alkylation and olefination:
(c) B. S. Lee, J. M. Gil and D. Y. Oh, Tetrahedron Lett., 2001, 42,
2345.
11 As key intermediates in drug synthesis: (a) D. A. Evans and B. T.
Connell, J. Am. Chem. Soc., 2003, 125, 10899; (b) A. A. Fokin, A. G.
Yurchenko, V. N. Rodionov, P. A. Gunchenko, R. I. Yurchenko, A.
Reichenberg, J. Wiesner, M. Hintz, H. Jomaa and P. R. Schreiner, Org.
Lett., 2007, 9, 4379.
17 Stobbe-type condensations providing Z-allyl dialkylphosphonates
analogous to phosphonylation of Baylis–Hillman adducts: (a) H. G.
McFadden, R. L. N. Harris and C. L. D. Jenkins, Aust. J. Chem., 1987,
40, 1619; (b) D. J. Martin, M. Gordon and C. E. Griffin, Tetrahedron,
1967, 23, 1831.
18 Although it received much less attention, we found that acrylonitrile
derived Baylis–Hillman adducts also undergo phosphonylation, pro-
viding the corresponding dialkyl phosphonates with an electronically
activated E-allyl dialkylphosphonates, see ref. 16c.
19 Metathesis strategy: (a) A. K. Chatterjee, T.-L. Choi and R. H.
Grubbs, Synlett, 2001, 1034; (b) A. K. Chatterjee, D. P. Sanders and
R. H. Grubbs, Org. Lett., 2002, 4, 1939; (c) M. J. Comin, D. A.
Parrish, J. R. Deschamps and V. E. Marquez, Org. Lett., 2006, 8,
705; (d) J. Moiese, S. Arseniyadis and J. Cossy, Org. Lett., 2007, 9,
1695.
20 (a) C.-Y. Ho and T. F. Jamison, Angew. Chem., Int. Ed., 2007, 46, 782;
(b) S.-S. Ng, C.-Y. Ho and T. F. Jamison, J. Am. Chem. Soc., 2006, 128,
11513.
21 General reviews on Ni-catalyzed multicomponents coupling: (a) J.
Montgomery, Acc. Chem. Res., 2000, 33, 467; (b) J. Montgomery,
Angew. Chem., Int. Ed., 2004, 43, 3890; (c) R. M. Moslin, K. Miller-
Moslin and T. F. Jamison, Chem. Commun., 2007, 4441; (d) S.-S. Ng,
C.-Y. Ho, K. D. Schleicher and T. F. Jamison, Pure Appl. Chem., 2008,
80, 929.
22 (a) G. A. Olah, Aromatic allylation and Related Reactions, Wiley-
Interscience, New York, 1964, Vol. 2, Part 1; (b) R. M. Roberts,
and A. A. Khalaf, Friedel–Crafts Alkylation Chemistry. A Century
of Discovery, Marcel Dekker, New York, 1984; (c) G. A. Olah,
R. Krishnamurti, G. K. S. Prakash, Friedel–Crafts Alkylations in
Comprehensive Organic Synthesis, ed. B. M. Trost, I. Flemin, Pergamon
3486 | Org. Biomol. Chem., 2010, 8, 3480–3487
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Press, Oxford, 1991; (d) G. A. Olah, J. Org. Chem., 2001, 66, 5943; (e) M.
Bandini, A. Melloni and A. Umani-Ronchi, Angew. Chem., Int. Ed.,
2004, 43, 550.
25 Some potential applications for the aromatic allylation products: (a) B.
Kramer and S. R. Waldvogel, Angew. Chem., Int. Ed., 2004, 43, 2446;
Trimethylenemethane-based ligands: (b) G. Rodriguez and G. C. Bazan,
J. Am. Chem. Soc., 1997, 119, 343.
23 Recent advances on aromatic allylation of Baylis–Hillman adducts
using: AlCl3: (a) D. Basavaiah, S. Pandiaraju and K. Padmaja, Synlett,
1996, 393; Catalytic conc H2SO4 at reflux: (b) D. Basavaiah, M.
Krishnamacharyulu, R. S. Hyma and S. Pandiaraju, Tetrahedron Lett.,
1997, 38, 2141; Reflux with TMSOTf: (c) S. Ravichandran, Synth.
Commun., 2001, 31, 2345; Montmorillonite clay catalyzed: (d) D. Saib
and A. Foucaud, J. Chem. Res., Synopses, 1987, 372; (e) B. Das,
A. Majhi, J. Banerjee, N. Chowdhury and K. Venkateswarlu, Chem.
Lett., 2005, 34, 1492; Microwave conditions: (f) P. Shanmugam and P.
Rajasingh, Chem. Lett., 2005, 34, 1494; For a recent example that may
favor O-allylations over C-allylations, please see: (g) E. Ramesh and R.
Raghunathan, Tetrahedron Lett., 2008, 49, 1125.
26 E. N. Walsh, J. Am. Chem. Soc., 1959, 81, 3023.
27 Mechanism of the hydrolysis of condensed phosphates in aqueous
organic solvents: M. Watanabe, S. Sato and H. Saito, Bull. Chem. Soc.
Jpn., 1976, 49, 2474.
28 For reviews on glycoside synthesis: (a) X. Zhu and R. R. Schmidt,
Angew. Chem., Int. Ed., 2009, 48, 1900; Nitrile effect on glycoside
anomeric stereocontrol: (b) R. R. Schmidt, M. Behrendt and A.
Toepfer, Synlett, 1990, 694; For a recent review on the Ritter
reaction: (c) A. R. E. Brewer, in Name Reactions for Functional Group
Transformations, ed. J. J. Li, E. J. Corey, John Wiley & Sons, 2007.
29 Landolt-Bo¨rnstein: Numerical Data and Functional Relationships in
Science and Technology - New Series; ed. C. Wohlfarth, M. D. Lechner
Group 4: Physical Chemistry; Springer Berlin Heidelberg, 2008,
Vol.17.
30 A. M. Saboori, D. M. Lang and D. S. Newcombe, Structural
requirements for the inhibition of human monocyte carboxylesterase
by organophosphorus, Chem.-Biol. Interact., 1991, 80, 327.
31 G. He, J. Massarella and P. Ward, Clin. Pharmacokinet., 1999, 37,
471.
24 Pioneer work on highly regio- and stereoselective aromatic allylation
of allylic alcohols mediated by iron carbonyl: (a) J. W. Dieter, Z. Li and
K. M. Nicholas, Tetrahedron Lett., 1987, 28, 5415; Recent advances in
arylation of benzylic alcohols using stoichiometric amount of Brønsted
acid: (b) F. Mu¨hlthau, O. Schuster and T. Bach, J. Am. Chem. Soc., 2005,
127, 9348; Using only catalytic amount of Brønsted acid at refluxing
´
DCM or CH3CN: (c) R. Sanz, A. Martinez, D. Miguel, J. M. Alvarez-
Gutie´rrez and F. Rodr´ıguez, Adv. Synth. Catal., 2006, 348, 1841; AuCl3
catalyzed: (d) W. Rao and P. W. H. Chan, Org. Biomol. Chem., 2008, 6,
2426.
32 Q. Chang, M. S. S. Chow and Z. Zuo, Biomed. Chromatogr., 2009, 23,
852.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3480–3487 | 3487
©