Highly selective markovnikov addition of hypervalent H-spirophosphoranes to alkynes mediated by palladium acetate: Generality and mechanism

Article abstract of DOI:10.1246/bcsj.20100141

Palladium acetate efficiently catalyzes the addition of an H-spirophosphorane (pinacolato)₂PH to alkynes to give Markovnikov addition products highlyselectively. The addition products can be easily converted to the corresponding alkenylphosphonates and phosphonicacids viasimple hydrolysis or thermal decomposition. This new reaction isa general method for the introduction of phosphorus functionality to the internal carbons of terminal alkynes, resolving the problem of the regioselectivity associated with hydrophosphorylation reactions so far reported. Mechanistic studies confirmed that (a) palladium acetate was reduced to metallic palladium by H-spirophosphorane, (b) the P-H bond of H-spirophosphorane could be activated by zero-valent platinum complexes to give the corresponding hydridoplatinum complexes, and (c) an alkenylpalladium species was identified from the reaction of palladium acetate with H- spirophosphorane and diphenylacetylene. These results support a reaction mechanism that palladium acetate was first reduced by H-spirophosphorane to give zero-valent palladium. This zero-valent palladium might insert into the P-H bond of the H-spirophosphorane to give a hydridopalladium species which then added to alkyne via the addition of H-Pd bond to form an alkenylpalladium species with the hydrogen atom added to the terminal carbon of alkynes. Reductive elimination of the alkenylpalladium affords the addition product.

Full text of DOI:10.1246/bcsj.20100141

1086 Bull. Chem. Soc. Jpn. Vol. 83, No. 9, 1086–1099 (2010)
© 2010 The Chemical Society of Japan
Highly Selective Markovnikov Addition of Hypervalent
H-Spirophosphoranes to Alkynes Mediated by Palladium Acetate:
Generality and Mechanism
Li-Biao Han,* Yutaka Ono, Qing Xu, and Shigeru Shimada
National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba 305-8565
Received May 17, 2010; E-mail: libiao-han@aist.go.jp
Palladium acetate efficiently catalyzes the addition of an H-spirophosphorane (pinacolato)2PH to alkynes to give
Markovnikov addition products highly selectively. The addition products can be easily converted to the corresponding
alkenylphosphonates and phosphonic acids via simple hydrolysis or thermal decomposition. This new reaction is a
general method for the introduction of phosphorus functionality to the internal carbons of terminal alkynes, resolving the
problem of the regioselectivity associated with hydrophosphorylation reactions so far reported. Mechanistic studies
confirmed that (a) palladium acetate was reduced to metallic palladium by H-spirophosphorane, (b) the P-H bond of
H-spirophosphorane could be activated by zero-valent platinum complexes to give the corresponding hydridoplatinum
complexes, and (c) an alkenylpalladium species was identified from the reaction of palladium acetate with H-
spirophosphorane and diphenylacetylene. These results support a reaction mechanism that palladium acetate was first
reduced by H-spirophosphorane to give zero-valent palladium. This zero-valent palladium might insert into the P-H bond
of the H-spirophosphorane to give a hydridopalladium species which then added to alkyne via the addition of H-Pd bond
to form an alkenylpalladium species with the hydrogen atom added to the terminal carbon of alkynes. Reductive
elimination of the alkenylpalladium affords the addition product.
Alkenylphosphinyl compounds (C=C-P(O)Z1Z2, where Z1
and Z2 represent an alkoxy, alkyl, or aryl group, respectively)
are versatile synthetic reagents for the preparation of bidentate
phosphorus ligands,1 flame retardants, and polymers,2 etc.3 For
example, they are the key intermediates for the preparation of
the clinically used antibacterial agent Fosfomycin ((1R,2S)-
(1,2-epoxypropyl)phosphonic acid) and analogs.4 In addition
to their synthetic applications, some alkenylphosphorus com-
pounds also show interesting biological activity.5 Thus,
alkenylphosphonates are potential enzyme inhibitors,6 and
their strong antiproliferative activities were also discovered
recently.7 A representative example of such compounds is
Midafotel ((R)-4-[(2E)-3-phosphono-2-propenyl]-2-piperazine-
carboxylic acid) which is a potent and competitive N-methyl-D-
aspartate (NMDA) antagonist developed for the treatment of
neurodegenerative diseases.8
Although a few methods for the preparation of alkenylphos-
phinyl compounds are available,3 the transition-metal-mediated
regio- and stereoselective addition of P(O)-H bonds to alkynes
(hydrophosphorylation) initiated by us9,10 appears to be one of
the most direct and efficient ways.9,11,12 Some alkenylphos-
phorus compounds are now commercially prepared by this
method.13
generating the anti-Markovnikov regioisomer 2 is general and
the selectivity is excellent, the preparation of 1 is usually
accompanied by the anti-Markovnikov regioisomer 2 which
can even become the major product when bulky substrates are
used. For example, although the PdMe2(PPh2Me)2 catalyzed
addition of hydrogen phosphonates to 1-octyne gave 1 with ca.
90%-96% regioselectivity,9 a similar addition to 3,3-dimethyl-
1-butyne only proceeded sluggishly to give 42% yield of the
adducts with 36% regioselectivity to 1. A similar result was
obtained when a catalyst11f Pd2(dba)3/Ph3P/CF3CO2H was
employed (Scheme 2). On the other hand, the addition to 2-
methyl-3-butyn-2-ol hardly proceeded. A similar phenomenon
was observed in the addition of Ph2P(O)H to alkynes. Thus by
employing PdMe2(PPhMe2)2/Ph2PO2H,12c,12n Ph2P(O)H added
to 1-octyne to give the corresponding adduct 1 in 92% yield
with 90% selectivity. However, under similar reaction con-
ditions, the addition of Ph2P(O)H to 3,3-dimethyl-1-butyne
only gave 1 in 60% selectivity and the addition to 2-methyl-3-
butyn-2-ol produced a complicated mixture. Therefore, the
regioselectivity to 1 is highly substrate dependent and a general
method for its preparation is required.
In addition to substrate limitations, the purification of the
alkenylphosphorus products by conventional silica gel column
chromatography techniques can be difficult, because of the
structural similarity of alkenylphosphorus compounds with the
phosphine oxide impurities, generated from the phosphine
ligands of the catalysts via air oxidation.
In order to overcome these drawbacks of the current metal-
mediated P(O)-H additions, studies on the development of a
new general method for the preparation of the Markovnikov
As summarized in Scheme 1, by using a suitable catalyst,
both the Markovnikov adduct 1 (catalysts: Me2Pd(PPh2Me)2,9
Me2Pd(PPhMe)2/Ph2P(O)OH,12c,12n or Ni(PPhMe2)4/Ph2P-
O2H12i) and the anti-Markovnikov regioisomer 2 (catalysts:
Ni(PPh2Me)4,12i Pd(PPh3)4,12b or RhCl(PPh3)312e) have been
prepared by the addition of the corresponding H-P(O) com-
pound to an alkyne. However, whereas the addition reaction
Published on the web September 2, 2010; doi:10.1246/bcsj.20100141

L.-B. Han et al.
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1087
[P(O)]-H
cat. M'
cat. M
R
+
[P(O)]
R
[P(O)]
R
2
1
[P(O)]-H = (RO)2P(O)H,
Ph(RO)P(O)H or Ph2P(O)H
Selectivity: excellent
Selectivity: poor to high
Scheme 1. Metal-mediated P(O)-H additions to alkynes forming alkenylphosphorus compounds.
P(O)Z2
the best of our knowledge such a metal-mediated addition
reaction of 3 has never been reported.18
R
R
cat. Pd
P(O)Z2
+
R
Z2P(O)H
1
2
O
O
P
R
R
O
O
O
O
Yield/%
Reg. sel.
to 1/%
Z
R
ð2Þ
P H
O
cat. Pd(OAc)2
O
MeO nC6H13
91
96
tBu
3a
4
42
36
trace
60
C(OH)Me2
i-PrO tBu
Results and Discussion
50
Preparation of 3. Hypervalent phosphorus compounds that
contain P-C bonds are known as labile species and their
synthetic procedures are usually complicated. On the other
hand, -5¤5 phosphoranes built up with an oxo-spirocyclic
skeleton are very easy to prepare and handle.14 They can be
obtained by a simple reaction of PCl3 with the corresponding
diols. H-Spirophosphorane 3a was prepared using a modified
procedure of the literature,16 via a slow addition of PCl3 to 2
molar equivalents of pinacol in Et2O at 0 °C in the presence of
Et3N (Scheme 3). A P(III) phosphite intermediate (CMe2O)2-
POCMe2CMe2OH formed during the reaction isomerizes to the
cyclic P(V) 3a having a P-H bond. Thus, at 0 °C, PCl3
Ph
nC6H13
92
90
tBu
C(OH)Me2
89
60
complicated
Scheme 2. Strong substrate dependency of hydrophospho-
rylation reactions.
adduct 1 via a phosphine-free metal-catalyzed selective hydro-
phosphorylation of alkynes were carried out, which we believe,
once achieved can complete the metal-mediated hydrophos-
phorylation of alkynes enabling the selective preparation of
both isomers 1 and 2.
Pentacoordinated H-spirophosphoranes 3 are easily prepared
and handled.14 Although compound 3 was known three decades
ago,14a studies on its reactivity were unexplored. Thus,
only a few Michael-type additions of these compounds were
reported,14c and no transition metal catalyzed transformations
of 3 were known. Being similar to (RO)2P(O)H, compound 3
can exist as a tautomeric mixture of P(V) and P(III) species
with the latter being able to ligate the metal and thus can
potentially deactivate the catalysts (eq 1).15 However, quite
different from (RO)2P(O)H, this equilibrium is easily affected
by the substituents on 3. In fact, it was known that 3a bearing
two pinacolato substituents solely exists in the P(V) form.16
Therefore, we surmised that because of this noteworthy feature
(the absence of a lone electron pair on phosphorus and hence
the inability of the P(V) form of 3 to coordinate a metal)
different behaviors of 3a from (RO)2P(O)H should be expected
in the metal-catalyzed hydrophosphorylation reactions.
O
O
O
O
O
O
O
O
O
P H
O
P H
O
P H
O
3a
3b
3c
Et3N
OH
+
PCl3
HO
- HCl
OH
P O
O
3a
O
O
O
P
O
O
O
O
O
O
O
ð1Þ
HO
HO
P H
O
P H
O
O
+
P Cl
3b
OH
Et2O
THF
O
3
P(V)
3a
P(III)
OH
(Me2N)3P
3c
+
Herein we disclose the details of a general palladium-
catalyzed addition of an H-spirophosphorane 3a to alkynes
leading to the Markovnikov adduct 4 in high yields (eq 2).17 To
HO
Scheme 3. Preparation of H-spirophosphoranes.

1088 Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010)
Markovnikov Addition of P-H Bonds to Alkynes Mediated by Palladium
Table 1. Optimization of the Addition of 3a to 1-Octyne
Forming 4aa)
(0.05 mol) was added to a mixture of pinacol (0.1 mol) and
triethylamine (0.15 mol) dissolved in dry ether (300 mL). The
reaction mixture was stirred overnight at room temperature.
The reaction mixture was washed with water and recrystalliza-
tion of the crude product gave pure 3a as a white solid in good
yield (65%-75%). Spirophosphorane 3a is stable toward air
and water at ambient temperature, though it is hygroscopic. For
the preparation of other H-spirophosphoranes using catechol
and ethylene glycol, a modified method was used. Thus, 3b
was synthesized from catechol and 2-chloro-1,3,2-benzodioxa-
phosphole, while 3c was prepared by a reaction of (Me2N)3P
with ethylene glycol. These compounds are rather unstable
toward water.
nC6H13
O
P
3a
+
nC6H13
cat. Pd
O
P
O
O
nC6H13
O
O
O
5a
4a
Run
Catalyst (mol %)
Time/h
Yield/%b)
1
Pd(OAc)2 (3)
Pd(OAc)2 (3)
2
2
2
2
2
16
14
2
95
93
45
75
88
0
71
96d)
88f)
2c)
3
Pd(OAc)2 (3)/H2O (10)
Pd(OAc)2 (3)/Et3N (6)
Pd(OAc)2 (3)/Ph3P (6)
Pd(OAc)2 (3)/Et3P (6)
Pd(OAc)2 (0.5)
4
5
6
7
8
9c),e)
We also confirmed the result of Burgada by 31P NMR
spectroscopy that 3a exists exclusively in the P(V) form, while
3b and 3c exist as a mixture of P(V) and P(III) species in
solution.16
Pd(OCOCF3)2 (3)
Pd(OCOCF3)2 (0.1)
Metal-Catalyzed Addition of 3 to 1-Octyne: Catalyst
Screening and Optimization of the Reaction Conditions.
The addition of 3a (0.25 mmol) to 1-octyne (0.25 mmol) was
thoroughly investigated in toluene (0.5 mL) at 80 °C in the
14
a) Reactions were carried out using 3a (0.25 mmol) and
1-octyne (0.25 mmol) with palladium catalyst in toluene
(0.5 mL) at 80 °C under nitrogen. b) Yields were determined
by 31P NMR spectroscopy. c) Reaction was carried out
under air. d) 4a/5a = 78/22. e) In the absence of a solvent.
f) 4a/5a = 89/11.
1
presence of a metal catalyst (10 mol %). As determined by H
and 31P NMR spectroscopy, the simple palladium acetate
produced a quantitative yield of 4a after 0.5 h with ²96%
regioselectivity. Other metals showing catalytic activity were
as follows (catalyst, time/h, NMR yield of 4a): Pd(PPh3)4,
4 (18) h, 18 (98)%; Pd2(dba)3, 18 h, 55%; Pd/Al2O3, 72 h,
54%; Pd(OCOPh)2, 2 h, 88%; Pd(NO3)2, 2 h, 67%. However,
metal-complexes such as PdCl2, PdBr2, PdSO4, Pd(acac)2,
PdCl2(PhCN)2, PdCl2(PPh3)2, PdCl2(P-i-Pr3)2, PdCl2(dppe),
PdCl2(dppb), (©3-C3H5PdCl)2, NiCl2, NiCl2(PPh3)2, Ni(PR3)4
(PR3 = PPh3, PPh2Me, and PPhMe2), Ni(cod)2, RhCl(PPh3)3,
and [Rh(cod)Cl]2 either gave a complicated result or did not
catalyze the addition at all. These results clearly demonstrated
a different chemical behavior of 3a from the 4-coordinate
(RO)2P(O)H. For example, Pd(OAc)2 did not catalyze a similar
addition of (MeO)2P(O)H,9 whereas Ni(PR3)4 which can
efficiently catalyze the addition of (MeO)2P(O)H12i did not
show catalytic activity with 3a.
applicable to a variety of alkynes bearing functionalities. Thus,
as shown by Runs 1-6, the reaction of 3a with alkynes having
cyano, hydroxy, amino, silyl,19 or alkenyl groups smoothly
proceeded to produce good yields of the adduct 4 with
excellent regioselectivity. Aromatic acetylenes such as phenyl-
acetylene also gave 86% yield of the corresponding adduct
selectively (Run 7). The results with bulky alkynes are
particularly noteworthy. Thus, an excellent regioselectivity
and high yields of the adducts were obtained from 3,3-
dimethyl-1-butyne (Run 8), 2-methyl-3-butyn-2-ol (Run 3),
mesitylacetylene (Run 9), and the biologically active ethister-
one (Run 10). It was reminded that similar reactions of these
bulky alkynes with (RO)2P(O)H hardly proceeded to produce a
mixture of terminal and internal adducts in low regioselectivity
(vide supra). Not only terminal alkynes, this addition also took
place efficiently with internal alkynes to give the trans-adducts
selectively. Thus, both dialkyl (Run 11) and diphenyl (Run 12)
acetylenes successfully gave the corresponding adducts in high
yields and selectivity. An alkyne with two electron-withdraw-
ing methoxycarbonyl groups also afforded the corresponding
adduct efficiently (Run 13). Finally, the reaction of 3a with an
atmosphere of industry-grade acetylene gas20 in the presence of
0.3-1 mol % of Pd(OAc)2 produced the corresponding vinyl-
spirophosphorane 4r in nearly a quantitative yield (Runs 17
and 18).
This Pd(OAc)2-mediated addition was optimized (Table 1).
Thus, under nitrogen, by using 3 mol % of Pd(OAc)2 (Run 1),
95% yield of 4a was generated after heating 3a and an
equivalent amount of 1-octyne dissolved in toluene for 2 h.
Interestingly, this Pd(OAc)2-catalyzed addition can be con-
ducted without the pre-exclusion of air. For example, the
Pd(OAc)2-catalyzed addition of 3a to 1-octyne conducted under
air also gave 93% yield of 4a (Run 2). The addition of water
lowered the yield of 4a (Run 3), whereas additives such as Et3N
(Run 4) and Ph3P (Run 5) did not significantly affect the
reaction. However, the addition of Et3P could completely stop
the catalytic addition (Run 6). The amount of Pd(OAc)2-catalyst
used could be reduced, albeit a longer reaction time was
required. For example, the reaction with 0.5 mol % Pd(OAc)2
gave 71% yield of 4a after 14 h (Run 7). Trifluoroacetate
palladium also catalyzed the reaction, which also proceeded
under air, though a considerable amount of 5a was observed
due to the hydrolysis of 4a (vide infra) (Runs 8 and 9).
Scope and Limitations. As demonstrated in Table 2, this is
a highly general regioselective addition reaction which is
From unsymmetrical alkynes, good regioselectivities, fol-
lowing the Markovnikov rule, could be observed when the two
substituents are substantially electronically different. Thus,
methyl 2-octynoate (Run 15) and ethyl 3-phenylpropiolate
(Run 16) gave a mixture of 4p/4p¤ and 4q/4q¤ in 86% and
81% yields, respectively. In both cases, isomers with phos-
phorus attached to the side of the less electron-withdrawing
substituent were predominantly formed. On the other hand, the

L.-B. Han et al.
Table 2. Pd(OAc)2-Mediated Addition of 3a with Alkynesa)
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1089
R1
O
O
3 mol% Pd(OAc)2
R1
R2
3a
+
[P]
[P]
P
O
R2
toluene, 80 °C
O
4
Run
1
Alkyne
Time/h
Product
Yield/%b)
[P]
CN
CN
2
4
83 (84)
4b
[P]
OH
2
3
73 (79)
73 (78)
OH
4c
[P]
[P]
5
4d
OH
OH
NBu2
4
4
2
4e
4f
73 (79)
74 (90)
NBu2
5c),d)
[P]
Me3Si
Me3Si
[P]
4g
6
8
76 (76)
[P]
[P]
Ph
7c)
8
Ph
4
4
4h
4i
86 (86)
88 (91)
4j
9
4
7
91 (91)
[P]
OH
OH
[P]
10c)
84d) (84)
4k
O
O
nC3H7
11
12
13
14
nC3H7
nC3H7
1
2
2
2
95 (97)
95 (95)
91 (91)
95 (98)
nC3H7
Ph
4l
[P]
Ph
Ph
Ph
4m
4n
[P]
CO2Me
MeO2C
CO2Me
CO2Me
Me
[P]
Ph
Me
Ph
Me
4o/4o′
Ph
[P]
[P]
[P]
[P]
(48/52)
nC5H11
[P]
CO2Me
nC5H11
nC5H11
CO2Me
22
9
86 (95)
81 (92)
4p/4p′
15
16
CO2Me
CO2Et
(90/10)
Ph
CO2Et
Ph
4q/4q′
(88/12)
Ph
CO2Et
[P]
17e),f)
18e),g)
1.5
16
94 (100)
98 (100)
4r
4r
[P]
[P]
a) Reactions were carried out by heating a toluene solution (0.5 M) of an equimolar ratio of 3a and an
alkyne in the presence of 3 mol % Pd(OAc)2. Regioselectivity for terminal alkynes ²96%. Only trans-
adduct was obtained for internal alkynes. b) Isolated yields (those in parentheses refer to NMR yields).
c) Ph2P(O)OH (1-6 mol %) was added. d) Only trans-adduct was obtained.19 e) Under 1 atm acetylene gas
atmosphere. f) 1 mol % Pd(OAc)2. g) 0.3 mol % Pd(OAc)2.
addition of 3a to 1-propynylbenzene (Run 14) gave an almost
equimolar mixture of the two regioisomers 4o/4o¤.
In sharp contrast to the high reactivity of alkynes, alkenes
such as 1-octene, acrylonitrile, styrene, and ¡-ethylstyrene did
not react with 3a under similar reaction conditions. In addition,
a remarkable difference in reactivity between 3a with 3b (3c)
was observed. Thus, although H-spirophosphorane 3a effi-
ciently reacts with a variety of alkynes to give the addition

1090 Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010)
Markovnikov Addition of P-H Bonds to Alkynes Mediated by Palladium
O
P
O
O
O
O
O
O
P
(HO)2P
EtOH
60 °C, 2 h
AcOH/H2O,
100 °C, 3 h
O
6r, 83%
4r
5r, 86%
nC6H13
O
(HO)2P
nC6H13
O
O
nC6H13
aq. HCl
O
aq. HCl
O
O
P
P
dioxane
RT, 10 min
dioxane
O
O
85 °C, 2.5 h
6a, 85%
4a
100 °C, 2 h
5a, 77%
Scheme 4. Easy conversion of 4 to alkenylphosphonates.
products, no addition products are obtained at all with H-
spirophosphoranes 3b and 3c under similar reaction conditions
even after heating at 80 °C for 24 h. The structural difference
mentioned above that 3a takes the P(V) form while 3b (3c)
exists as a mixture of P(V) and P(III) tautomers,16 was assumed
to be the first reason for this big difference in reactivity.
Hydrolysis and Thermal Decomposition of Alkenylspiro-
phosphoranes 4 to Phosphonate Derivatives.
Though
compound 4 is stable toward moisture and air at room
temperature, the spiro ring collapses smoothly at an elevated
temperature or in the presence of an acid to give the
corresponding phosphonate derivatives. Thus, heating 4r in
EtOH at 60 °C for 2 h produced 1,3-dioxa-2-phosphorane 2-
oxide derivative 5r in 86% yield (Scheme 4). If the reaction is
conducted in an acidic medium 4r can be further hydrolyzed to
give the corresponding vinylphosphonic acid. For example,
after heating 4r in a 1:1 mixture of AcOH/H2O at 100 °C for
3 h, vinylphosphonic acid was obtained in 83% yield. The
hydrolysis can be carried out stepwise under acidic conditions.
Similarly, with hydrochloric acid, alkenylspirophosphorane 4a
produced phosphonate 5a in 77% yield at room temperature,
and produced 85% yield of alkenylphosphonic acid 6a on
heating.
A bulky substituent in 4 accelerates this hydrolysis. Thus,
although other alkenylspirophosphoranes 4 are stable toward
purification on silica gel, 4i with a tert-butyl group was
completely converted to 1,3-dioxa-2-phosphorane 5i when
being stirred with silica gel in hexane at room temperature for
0.5 h. Interestingly, although in the absence of an acid and
water no decomposition of 4i took place at 80 °C, a simple
thermal decomposition of 4i to 5i takes place at an elevated
temperature (eq 3). Thus, heating 4i in toluene at 100 °C for 3
days could generate 5i in 62% yield (a quantitative amount of
5i was obtained after 7 days). The formation of the counterpart
2,3-dimethyl-3-buten-2-ol during this thermal decomposition
could be confirmed by 1H NMR spectrum (ca. 87% NMR
yield).
Chart 1. 31P NMR chart of a reaction mixture of Pd(OAc)2
with 3a in THF-d8 at 25 °C overnight.
Hydrolysis of an alkylspirophosphorane with water afford-
ing the corresponding alkylphosphonate was briefly described
by Burgada in the literature.21 However, such a thermal
decomposition of 4i generating 5i seems to have not been
recognized before.
Reaction of Pd(OAc)2 and Pd(OCOCF3)2 with 3a.
A
mixture of Pd(OAc)2 (0.1 mmol) and 3a (0.1 mmol) dissolved
in THF-d8 (0.5 mL) produced a brown transparent solution.
However, after 2 h at room temperature, no other signals except
that for 3a (¤ ¹44.1) could be detected by 31P NMR spec-
troscopy. Significant changes were observed after standing the
solution overnight, i.e., metallic palladium came out from the
solution and several new signals were detected by 31P NMR
spectroscopy (31P integration ratios: 6/7/3a/8 = 0.5/0.2/10/
HO
−
O
O
O
O
O
O
ð3Þ
P
P
toluene
100 °C, 7 days
O
4i
5i, quant.

L.-B. Han et al.
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1091
O
O
O
O
Pd(OAc)2
- HOAc
heat
O
O
O
O
O
O
O
O
P H
O
P PdOAc
P OAc
O
P O
OAc
- Pd
O
9
8, δ -50.1
H2O
7, δ 2.7
3a, δ -44.1
3a
OAc
O
O
O
O
O
O
P
Pd
P
O
O
O
O
O
O
O
O
H
P OH
O
P O
OH
10
6, δ 4.2
OAc
O
O
O
O
OH
+
P
Pd
P
O
O
O
H
11
Scheme 5. Reduction of palladium acetate by 3a.
1.4) (Chart 1). As expected, heating the solution at 50 °C
accelerated this process (20 min: 6/7/3a/8 = 0.12/0.12/1/0.3
and 3 h: 6/7/3a/8 = 0.29/1.22/1/0.93). Similar results were
observed when excess 3a was used (Pd(OAc)2/3a = 1/2)
except more starting material 3a remained unreacted. The
reaction proceeded faster in toluene: metallic palladium came
out from the initially brown transparent solution after 0.5 h and
the reaction almost completed after 3 h (6/7/3a/8 = 4.6/0.2/
1/7). Compounds corresponding to ¤ 2.74 and ¤ 4.28 were
successfully isolated from the reaction mixture and were
identified as phosphates 7 and 6. The compound corresponding
to ¤ ¹50.1 could not be isolated in pure form because it
gradually decomposed at room temperature to 6 and 7 (vide
infra). As shown below, filtration and evaporation of the above
reaction mixture in toluene gave a yellow oil containing 6, 7,
and 8 with a ratio, initially 6/7/8 = 0.5/0.2/1, and then
became to 6/7/8 = 2.4/2.4/1 after 1 h at room temperature. No
other signals except those for 6, 7, and 8 could be detected.
However, no deposition of metallic palladium was observed
from this mixture even on heating at 100 °C for 1 h, indicating
that 8 is not a palladium-containing species. On the other hand,
as judging from its 31P NMR chemical shift, compound 8
should have a similar spiro structure to compound 3a.
These results agreed with a two-step reaction sequence in
which compound 8 was first generated, and it decomposed
subsequently to give 6 and 7. Thus, the reaction of 3a with
Pd(OAc)2 was rationalized as follows: 3a reacted with
Pd(OAc)2 to give 9 with the liberation of acetic acid (the
formation of AcOH could be identified by 1H NMR); 9
subsequently decomposed, formally via a reductive elimination
pathway, to give 8 and liberate metallic Pd. This spirophos-
phorus compound 8 collapsed to 7 or 6 via hydrolysis with
water.22
Me3P
3a
9a
Pd(OAc)2
1-octyne
+
PMe3
Pd PMe3
OAc
3a
nC6H13
+
O PMe3
O PEt3
O
O
O
O
P Pd PMe3
O OAc
P Pd OAc
O PEt3
9a
9b
Scheme 6. Trapping the intermediate 9 by a phosphine
ligand.
estimated from 31P NMR) could also be observed. Although
this complex has not been isolated, as described below (the
reaction of 3a with (CF3CO2)2Pd), it is reasonable to assume it
has a structure depicted as 11 by comparing to a similar
complex generated from 3a with (CF3CO2)2Pd which was
unambiguously established by an X-ray analysis.
Although palladium complex 9 could not be isolated,
complex 9a and 9b stabilized by a phosphine ligand were
successfully isolated (Scheme 6), supporting the mechanism
proposed in Scheme 5. A high yield of complex 9a was
isolated from a reaction of 3a with Pd(OAc)2 in the presence of
Me3P. Thus, a mixture of 3a, Pd(OAc)2, and PMe3 (3a/Pd/
PMe3 = 1/1/3) in THF at room temperature gave a transparent
yellow solution, in which a new complex 9a was formed.
31P NMR showed that the reaction completed after 4 h and
complex 9a having a cis-structure was generated quantitatively.
THF and other volatiles were removed under reduced pressure
to obtain the crude complex 9a. By recrystallization from
pentane pure 9a was obtained as a white solid in 82% yield.
Similarly was prepared complex 9b by a reaction of 3a with
In addition to the above phosphorus compounds, tiny signals
displayed at 31P NMR at 99.7 ppm (dd, JPP = 90.2 Hz) and 50.3
(dd, JPP = 99.2 Hz) (ca. 2% of all phosphorus compounds as

1092 Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010)
Markovnikov Addition of P-H Bonds to Alkynes Mediated by Palladium
Pd(OAc)2 in the presence of Et3P. In contrast with 9, the
phosphine-ligated complex 9a is thermally stable at 80 °C in
toluene or THF under nitrogen, and no decomposition could be
observed after heating for 20 h. However, it completely
decomposed at elevated temperature (0.05 M in toluene,
110 °C, 3 h) or when exposed to air at room temperature for 2 h.
Heating a mixture of 9a with 1-octyne in toluene at 80 °C for
2 h resulted in the complete disappearance of the complex.
However, the reaction did not produce the addition product
4a but afforded 3a via, formally, a protonolysis of 9a with
1-octyne.
In contrast to the above reaction of 3a with Pd(OAc)2, a
similar reaction with (CF3CO2)2Pd did not produce metallic
palladium, but gave colorless needles of 12 in high yields.
Thus, (CF3CO2)2Pd (0.5 mmol) and 3a (1.0 mmol) dissolved in
CH2Cl2 was mixed at room temperature. The resulting pale
yellow suspension was stirred until solid was dissolved (ca.
24 h). Removal of the solvent followed by recrystallization of
the residue from CH2Cl2 gave complex 12 as white crystals in
83% yield (eq 4). The structure of complex 12 was determined
by X-ray crystallography (Figure 1).23 The complex has a
square-planar configuration surrounding Pd. The bond length
of P2-O7 is 1.483(2) ¡, which is a typical value for P=O bond
rather than a value for a single P-O bond.24 Therefore, the
structure for 12 is depicted as shown below where the spiro
ring of O-P bond was cleft to a P(III) phosphite species bearing
an OH group. The phosphorus atom of the phosphite and the
oxygen atom of the OH group coordinate to the Pd(II) center.
Figure 1. Molecular structure of complex 12 (50% proba-
bility level). Hydrogen atoms bound to carbons are omitted
for clarity. Selected bond lengths (¡) and angles (°): P2-
O7 = 1.483(2), Pd-P1 = 2.1869(6), Pd-P2 = 2.2211(6),
Pd-O4 = 2.1313(2), Pd-O8 = 2.0881(2), P1-Pd-P2 =
94.76(2), P1-Pd-O4 = 90.90(5), P2-Pd-O8 = 86.87(5),
O4-Pd-O8 = 87.42(7).
phosphonate 13 could be generated by the hydrolysis of 3a,14 a
reaction path a in which 13 coordinates to 9¤ via the phosphite
form15 might be proposed. However, as described below, this
possibility seems low since 13 could not be detected during
the reaction. Thus, as shown by 31P NMR spectroscopy, no
detectable five-membered P-H compound 13 could be found
during the reaction, and when an extra amount of 3a was used
in the reactions, it remained unchanged, i.e., 3a does not
decompose to 13 under the present conditions. A separate
experiment using a mixture of 3a and 5 mol % (CF3CO2)2Pd in
toluene also showed that no 13 was formed at room temper-
ature for 24 h. As confirmed by 31P NMR spectroscopy, no
reaction took place between 3a and (CF3CO2)2Pd at ¹30 °C for
2 h. The reaction slowly proceeded at 0 °C to produce a trace
amount of 12 after 1 h as the only product detectable by
31P NMR spectroscopy. On the other hand, the formation of
2,3-dimethyl-3-buten-2-ol was confirmed by 1H NMR spec-
troscopy. Therefore, the formation of 12 may take place via the
coordination of 3a18 to 9¤ which collapsed to give 12 (path b).
O
O
O
P
3a
O
P
- CF3CO2H
O
ð4Þ
+
Pd
O
CF3
O
Pd(OCOCF3)2
OH
O
12
The exact pathway for the formation of 12 is not clear
(Scheme 7). It is reasonable to assume that a palladium species
9¤ was generated first by a similar protonolysis observed for the
reaction of 3a with Pd(OAc)2 (Scheme 5). Since hydrogen
O
HO P
O
O
Pd(OCOCF3)2
13 X
O
O
3a
O
O
12
P PdOCOCF3
P PdOCOCF3
O
O
- CF3CO2H
path a
O
9'
path b
3a
O2CCF3
O
O
O
O
P H
O
O
O
P
OH
P OH
12
+
O
O
O
O
Pd
P
O
O
13
H
Scheme 7. Reaction of 3a with (CF3CO2)2Pd leading to 12.

L.-B. Han et al.
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1093
[P]
O
P
O
O
nC6H13
O
[P]
[P]
nC6H13
O
P
O
nC6H13
CF3
O
O
12
Pd
OH
O
O
P
Pd
nC6H13
50 °C
80 °C
CD2Cl2
25 °C
CF3
O
O
OH
O
15
OH
-
14
[P] = P(O)(OCMe2-Me2CO)
Scheme 8.
Complex 12 catalyzed the addition of 3a to 1-octyne to
produce adducts 4a and 5a as efficiently as (CF3CO2)2Pd
(3 mol % Pd catalyst, toluene, 80 °C, 6 h. Complex 12: 79% 4a,
16% 5a and Pd(OCOCF3)2: 75% 4a, 21% 5a). However, being
similar to Pd(OCOCF3)2, it does not catalyze the addition of
(MeO)2P(O)H to 1-octyne and only sluggishly catalyzed the
addition of the more reactive five-membered P-H compound
13 to produce a trace amount of the corresponding adduct (ca.
3% yield). This fact can be an indication that the formation of
the five-membered adduct 5a obtained from the addition of 3a
to 1-octyne is not due to a simple addition of 13 to alkyne, but
due to other paths such as the protonolysis of the adduct 4a.
Nevertheless, as described below, complex 12 is apparently
not the real catalyst for the formation of adducts 4a and 5a.
Complex 12 reacted with 1-octyne to give quantitatively a new
complex 14 via a cis-insertion of the Pd-P(O) bond of 12 to
1-octyne with P(O) bonding to the terminal carbon and Pd
bonding to the internal carbon, showing that the reaction of 12
with 1-octyne could not lead to adducts 4a and 5a. Thus,
1-octyne (1.9 mg, 0.017 mmol) was added to a suspension of 12
(4.4 mg, 0.007 mmol) in CD2Cl2 (0.5 mL) in an NMR tube. The
suspension turned clear upon standing at room temperature
for 1 h, and a new compound was formed as confirmed by
31P NMR spectroscopy (¤ 55.6 (d, J = 79 Hz) and 100.0 (d,
J = 79 Hz), which was assumed to be an alkyne-coordinated
complex (ratio to 12 = 66/34) (Scheme 8). The ratio of 12 to
the new complex did not change upon further standing for 4 h,
indicating that the reaction reached equilibrium. However,
interestingly, when this solution was heated to 50 °C, a new
complex 14 was formed quantitatively. The structure of
complex 14 was unambiguously determined by X-ray crystal-
lography (Figure 2).
Figure 2. Molecular structure of complex 14 (40% proba-
bility level). Hydrogen atoms except the vinyl proton are
omitted for clarity. Selected bond lengths (¡) and angles
(°): P1-O7 = 1.467(3), P2-O1 = 1.583(7), Pd-P2 =
2.717(11), Pd-C2 = 1.977(5), Pd-O4 = 2.150(3), Pd-
O8 = 2.110(3), P2-Pd-C2 = 88.57(15), P2-Pd-O4 =
94.46(8),
C2-Pd-O8 = 89.04(18),
O4-Pd-O8 =
87.92(13).
starting materials 3a and alkyne was observed at room
temperature to produce a transparent solution, where, in sharp
contrast to reactions in the absence of an alkyne, no decom-
position to metallic palladium was observed after 20 h.
Although no valuable information on the possible reactive
catalytic species was obtained with the former three alkynes,
fortunately, an alkenylpalladium species 15¤ (see below for its
structure assignment) was clearly observed with MeO2C¸
CCO2Me and PhC¸CPh (Scheme 9). Thus, at room temper-
ature, a mixture of Pd(OAc)2, 3a, and diphenylacetylene (0.02
mmol for each substrate) was dissolved in 0.5 mL toluene-d8 to
get a pale yellow transparent solution (Charts 2 and 3). As
Heating 14 at 80 °C for 36 h in CD2Cl2 in a sealed tube
resulted in the formation of bisphosphinylalkene 15 in 75%
yield accompanied by the generation of 81% yield of 2,3-
dimethyl-3-buten-2-ol as determined by 1H NMR spectroscopy.
Reactions of Pd(OAc)2 with 3a in the Presence of an
1
shown in Chart 2, in H NMR, a doublet was observed at ¤
Alkyne: Determining the Real Catalysts.
In order to
6.30 (JHP = 12.2 Hz), assignable to an alkenylpalladium spe-
cies 15¤ (see below for its assignment), which increased as the
reaction progressed for 30 h. Formation of the addition product
4m was also observed (a doublet at ¤ 7.21) which increased
constantly. The solution remained transparent during these
times. Noted however, as 3a was completely consumed after
ca. 50 h, precipitation of metallic palladium was observed,
and the doublet at ¤ 6.30 also gradually disappeared. Upon
heating the solution at 60 °C for 2 h, complex 15¤ completely
determine the real catalysts in this palladium acetate-mediated
additions of 3a to alkynes, a similar reaction to Scheme 5 in
the presence of an alkyne (molar ratios of Pd(OAc)2/3a/
alkyne = 1/1/1 to 1/4/4) was conducted and the reaction was
1
monitored by H and 31P NMR spectroscopies. The following
five alkynes were used as substrates: 1-octyne, CH2=C(Me)C¸
CH, 4-octyne, MeO2C¸CCO2Me, and PhC¸CPh. As indicated
1
by H and 31P NMR spectra, in all cases, consumption of the

1094 Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010)
Markovnikov Addition of P-H Bonds to Alkynes Mediated by Palladium
Pd(OAc)2
+
+
7
3a
9
Pd
3a
8
6
- HOAc
O
O
O
P
R
P PdH
O
O
O
O
16
R
R
R
O
R = Ph
4m
3a
H-Pd addition
O
R
O
O
P Pd
O
R
AcOH
R
O
O
O
P
Pd
H
15
R
OAc
O
H
H
15'
1H δ 6.30, JHP = 12.2 Hz
31P δ 108.5
Scheme 9. Reaction of Pd(OAc)2 with 3a and diphenylacetylene.
Chart 3. Time course of 31P NMR spectra for a mixture of
Pd(OAc)2, 3a, and diphenylacetylene in toluene-d8.
1
obtained from 31P NMR spectroscopy (73% yield of 4m, a total
yield of 23% for compounds 6 and 7 (6/7 = 1/2)).
Chart 2. Time course of H NMR spectra for a mixture of
Pd(OAc)2, 3a, and diphenylacetylene in toluene-d8.
These results very well reflect a reaction process shown in
Scheme 9: Pd(OAc)2 was reduced by 3a to generate zero-
valent palladium and 8 (which subsequently decomposed to 6
and 7); this palladium reacted with 3a and PhC¸CPh to afford
the adduct 4m via an alkenylpalladium intermediate 15, which
was assumed to be generated through an H-Pd addition of
an hydridopalladium species 16. This complex 16 could be
formed via an oxidative addition of 3a to palladium. Although
direct evidence for its formation was not available, as described
below a similar oxidative addition of 3a to Pt(PEt3)n did take
place to give the corresponding hydridoplatinum species
(eq 6), which supports this possibility.
disappeared to leave the addition product 4m as the sole
product. In agreement with 1H NMR spectroscopy, similar
phenomena were observed with 31P NMR spectroscopy
(Chart 3). The signal at ¤ 108.5 was due to 15¤, others signals
could be assigned to 6, 7, 4m, 3a, and 8, respectively. As the
reaction progressed, 15¤ gradually increased, however, as 3a
was completely consumed after 30 h, 15¤ decreased to deposit
metallic palladium and afforded the adduct 4m (amounts
(¯mol) of 15¤ and 4m calculated on the basis of 31P NMR
(time, 15¤, 4m): 1 h: 15¤, trace; 4m, trace. 20 h: 15¤, 7.1; 4m,
4.6. 30 h: 15¤, 7.3; 4m, 6.3. 50 h: 15¤, 2.0; 4m, 11). As
estimated from 1H NMR spectroscopy, 29% of Pd(OAc)2, 74%
of PhC¸CPh, and 100% 3a were consumed, respectively,
and ca. 70% yield of the adduct 4m was generated from the
above reaction after heating at 60 °C. Similar results were also
As described above, an alkenylpalladium species assigned
to 15¤ was observed by NMR spectroscopies. The structural
assignment was based on the following observations. First, its
31P signal appeared in the low magnetic field at ¤ 108.5

L.-B. Han et al.
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1095
indicating that this alkenylpalladium intermediate has a
structure of 15¤ where a phosphite ligates to Pd rather than a
spirophosphane configuration as shown in complexes 15 and
16. The isolation of this alkenlypalladium was not successful
due to its decomposition at room temperature. However, it was
trapped with Et3P to give an alkenylpalladium complex 17
and 3a (eq 5). The successful isolation of complex 17 from
a mixture of complexes shown in Scheme 9 deduces the
structures of 15 and 15¤.
complexes does proceed at room temperature to give the
corresponding hydridoplatinum complex trans-18 (94% NMR
yield based on Pt(cod)2) (eq 6). Thus, NMR spectroscopy of a
1:1:2 mixture of 3a, Pt(cod)2, and Et3P in toluene after standing
overnight clearly showed the formation of a P-Pt-H species as
the result of oxidative addition of 3a to platinum. The 1H NMR
spectroscopy exhibited a dt signal with Pt satellites centered at
2
¤ ¹8.40 with coupling constants of 1JHPt = 723 Hz, JHP
=
291 Hz, and 2JHP = 19.8 Hz, respectively. In 31P NMR, a set of
a doublet and a triplet displayed at ¤ 17.9 (JPP = 36.0 Hz,
PEt3
Ph
J
PPt = 2837 Hz) and 40.1 (JPP = 36.0 Hz, JPPt = 4055 Hz),
Et3P
AcO Pd
ð5Þ
15'
3a
respectively, with a 2:1 integration ratio. Noted that the H-Pt
coupling constant is relatively small compared to other P(O)-
Pt-H complexes (Table 3). At the same time, the JPPt value of
the spirophosphoranyl group trans to hydride is considerably
larger than that of other P(O)-Pt bonds which ranges from 1308
to 3288 Hz. These facts indicate a larger trans influence of the
spirophosphoranyl group than P(O) groups. The complex was
isolated by recrystallization from pentane to give trans-18 as
colorless plates in 60% yield.
+
Ph
PEt3
H
17
Therefore, Pd(OAc)2 is a precursor for the catalytic addition
of 3a to alkynes. In the catalytic cycle, Pd(OAc)2 may be first
reduced by 3a to form zero-valent palladium which generates a
hydridopalladium catalytic species such as 16 to start the
catalytic cycle. Indeed, an induction period (ca. 5 min) for
the catalytic addition reaction of 3a to diphenylacetylene
was observed (Figure 3), which agrees with the mechanism
proposed.
Reaction of Spirophosphorane 3a with Zero-Valent Pd
and Pt Complexes. No reaction was observed between 3a and
the following metal-complexes at room temperature in toluene:
Pd(PCy3)2, Pd(PEt3)4, Pt(PPh3)3, and Pt(cod)2. However, an
oxidative addition of the P-H bond of 3a to Et3P-ligated Pt(0)
O PEt3
O
ð6Þ
3a Pt(cod)2 +
P Pt H
O PEt3
+
2 Et3P
O
trans-18
Oxidative addition of 3a to Pt(PEt3)4 also took place to give
the corresponding hydridoplatinum complexes, though the
reaction was slow compared to the above reaction using a
combination of Pt(cod)2 and Et3P. Thus a mixture of Pt(PEt3)4
and 4 equivalents of 3a in toluene at room temperature only
produced 26% yield (based on Pt) of the corresponding
hydridoplatinum complexes after 2 days (eq 7). In addition, the
hydrido complex was obtained as a mixture of cis- and trans-
hydridoplatinums with the cis one being predominantly formed
(a ratio of cis/trans = 7/3). After further standing the mixture
at room temperature for 5 days, cis-18 completely disappeared
leaving trans-18 as the sole hydrido complex (31% yield).
Therefore, the cis-18 hydrido complex perhaps is the first
product of the oxidative addition reaction which isomerizes to
the thermodynamically more stable trans-18.
Pt(PEt3)4
3a
+
time/min
O PEt3
O PEt3
P Pt H
O PEt3
O
O
O
ð7Þ
P Pt PEt3
O H
Figure 3. Yields of 4m vs. time for
a reaction of
O
diphenylacetylene (0.2 mmol) with 3a (0.2 mmol) in the
presence of palladium acetate (0.01 mmol) in toluene
(0.5 mmol) followed by 31P NMR spectroscopy.
cis-18
trans-18
Table 3. NMR Spectra Data for Complexes trans-HPt[P][P¤]2
P
P¤
¤
H
JHPt
JHP
JHP¤
¤
P
JPPt
¤
P¤
JP¤Pt
JPP¤
(pinacolato)2P (18)
Ph(MenO)P(O)12n
(EtO)2P(O)9
PEt3
PEt3
PEt3
PEt3
PEt3
PPh3
¹8.40
¹5.23
¹5.23
¹5.74
¹6.75
¹4.59
723
720
737
872
805
812
291
193
226
125
124
120
20
19
18
15
17
13
40.1
100.1
95.1
¹48.7
¹3.7
¹176.7
4055
2803
3288
1440
1575
1308
17.9
20.0
22.3
17.9
16.7
29.7
2837
2671
2630
2572
2648
2892
36
28
33
19
17
15
25
PhPH¢BH3
25
Ph2P¢BH3
26
PH2¢W(CO)5

1096 Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010)
Markovnikov Addition of P-H Bonds to Alkynes Mediated by Palladium
Table 4. Bond Lengths (¡) and Angles (°) for Complexes trans-HPt[P][P¤]2
P
P¤
Pt-P
Pt-P¤
P-Pt-P¤
P¤-Pt-P¤
(pinacolato)2P (18)
PEt3
PEt3
PEt3
PPh3
2.351(5)
2.3477(14)
2.331(3)
2.359(2)
2.272(3)
2.2863(13)
2.253(3)
2.278(2)
97.56(10)
2.2771(14) 98.87(5)
164.87(17)
168.26(5)
160.22(12)
161.43(8)
25
PhPH¢BH3
92.83(5)
99.10(9)
96.87(8)
25
Ph2P¢BH3
2.256(2)
2.282(2)
98.72(9)
98.01(8)
26
PH2¢W(CO)5
was heated at 80 °C for 18 h. The complex trans-18 completely
disappeared to generate an alkynylplatinum complex 19 (86%
yield) and a quantitative yield of 3a (Scheme 10). The expected
alkenylplatinum species via the addition of H-Pt bond to the
alkyne could not be observed. Such a protonolysis can even
proceed faster with acetic acid. Thus, 75% yield of complex 20
was generated after mixing 18 and an equivalent amount of
acetic acid in toluene at room temperature for 10 min. On the
other hand, heating a mixture of diphenylacetylene with trans-
18 at 80 °C for 16 h produced complex 21, in which, no
alkenylplatinum species by the addition of the H-Pt bond could
be recognized either.
Summary and Conclusion
Palladium acetate catalyzed the addition of spirophospho-
rane 3a with alkynes to give the Markovnikov isomers
predominantly. The reaction is general and is tolerant toward
a variety of functionalities. The reaction is also easily carried
out and it could even be performed under air. There is a short
induction period for the addition reaction of 3a with alkynes
catalyzed by palladium acetate, during which the addition
reaction almost did not take place. Rapid addition occurred
after this short induction period.
A stoichiometric reaction of Pd(OAc)2 with 3a gave metallic
palladium which was formed first by the substitution of an
OAc of Pd(OAc)2 by 3a to give a spirophosphoranylpalladium
9 and AcOH. Spirophosphoranylpalladium 9 subsequently
decomposed to give metallic palladium and other phosphorus
compounds (Scheme 5). This reduction of palladium acetate to
zero-valent metallic palladium corresponds to the induction
period observed for palladium acetate catalyzed addition of 3a
with alkynes.
An alkenylpalladium species 15¤ was identified from
the reaction of Pd(OAc)2 with 3a and diphenylacetylene
(Scheme 9). In addition, the P-H bond of spirophosphorane
3a could be activated by zero-valent platinum complexes to
give the corresponding hydridoplatinum complexes 18 (eq 6).
These facts indicate a reaction mechanism that Pd(OAc)2 was
first reduced by 3a to produce zero-valent palladium. This
palladium may insert into the P-H bond of spirophosphorane
3a to give a hydridopalladium species. This hydridopalladium
16 added to alkynes via the addition of H-Pd bond to give an
alkenylpalladium species 15 with the hydrogen atom added to
the terminal carbon of terminal acetylenes. Reductive elimi-
nation of complex 15 gave the addition product 4.
This new reaction reported provides a general methodology
for the introduction of phosphorus functionality to the internal
carbon of a terminal alkyne, resolving the problem of the
regioselectivity associated with hydrophosphorylation reac-
tions so far reported. The adduct 4, when necessary, can be
readily converted to the corresponding alkenylphosphonates
Figure 4. Molecular structure of trans-18 (40% probability
level). Hydrogen atoms are omitted for clarity. Selected
bond lengths (¡) and angles (°): Pt1-P1 = 2.351(5), Pt1-
P2 = 2.272(3), P1-O1 = 1.654(7), P1-O2 = 1.751(5),
P1-Pt1-P2 = 97.56(10), P2-Pt1-P2* = 164.87(17), P2-
Pt1-P1-O1 = 46.0(2).
PEt3
nC6H13
Pt H + 3a
nC6H13
PEt3
19
O PEt3
P Pt H
O PEt3
PEt3
O
O
AcOH
+
3a
3a
AcO Pt H
PEt3
20
Ph
trans-18
PEt3
Ph
Ph
Pt
+
PEt3
Ph
21
Scheme 10. Reaction of trans-18.
X-ray crystallographic analysis of the crystal of trans-18
unambiguously confirmed the structure (Figure 4). To the best
of our knowledge, this is the first isolation of a transition metal
hydride complex having a spirophosphorane framework.18 The
complex has a slightly distorted square-planar geometry in
which the spirophosphoranyl group and PEt3 are placed
at the mutually cis-positions with bond lengths being Pt1-
P1 = 2.351(5) ¡ and Pt1-P2 = 2.272(3) ¡, respectively. Inter-
estingly, the Pt1-P1 bond length has a similar value to other
Pt-P bonds (Table 4).25,26
Protonolysis rather than addition of the Pt-H bond were
observed when a mixture of trans-18 with 1-octyne in toluene

L.-B. Han et al.
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1097
and phosphonic acids via a simple hydrolysis or a thermal
decomposition process. Therefore, this new reaction is a perfect
complement to the hitherto known hydrophosphorylation
reactions.
for C20H39O4P: C, 64.14; H, 10.50%. Found: C, 63.95; H,
10.49%.
This research was partly supported by New Energy and
Industrial Technology Development Organization (NEDO) of
Japan (Industrial Technology Research Grant Program in
2004).
The new reaction is featured by its high generality and
simplicity: it does not require a phosphine-palladium complex
but the simple palladium acetate; moreover, in contrast to
hitherto reported metal-catalyzed P(O)-H additions9,11,12 that
require oxygen-free reaction conditions, this new addition
reaction could even be conducted under air. In addition, these
alkenyl spirophosphoranes 4 are a new class of phosphorus
containing heterocyclic compounds and there are no alternative
ways for their preparation.17 The high potential of these
compounds as starting reagents for the synthesis of diverse
novel phosphorus heterocyclic compounds, which are of
current interest,14c though the chemical transformations of
their reactive carbon-carbon double bonds can be readily
expected.
Supporting Information
Determination of the regioselectivity, experimental proce-
dure, copies of NMR spectra and analytical data for adducts 4
and other compounds. This material is available free of charge
on the Web at: http://www.csj.jp/journals/bcsj/.
References
1
a) C. R. Johnson, T. Imamoto, J. Org. Chem. 1987, 52,
2170. b) K. M. Pietrusiewicz, M. Zabłocka, W. Wisniewski,
Phosphorus, Sulfur Silicon Relat. Elem. 1990, 49, 263. c) L.
Horner, H. Lindel, Phosphorus Sulfur Relat. Elem. 1984, 20, 161.
d) K. M. Pietrusiewicz, W. Wisniewski, M. Zabłocka, Tetrahedron
1989, 45, 337. e) D. J. Collins, L. E. Rowley, J. M. Swan, Aust. J.
Chem. 1974, 27, 841. f) Y. M. Polikarpov, B. K. Sacherbalkov,
T. Y. Medved, M. I. Kabachnik, Izv. Akad. Nauk SSSR, Ser. Khim.
1978, 2114. g) K. M. Pietrusiewicz, M. Zabłocka, Tetrahedron
Lett. 1988, 29, 1991. h) H. Brunner, S. Limmer, J. Organomet.
Chem. 1991, 417, 173. i) D. Cavalla, S. Warren, Tetrahedron Lett.
1982, 23, 4505. j) M. I. Kabachnik, T. Y. Medved, Y. M.
Polikarpov, K. S. Yudina, Izv. Akad. Nauk SSSR, Otd. Khim. Nauk.
1962, 1584. k) H. Brunner, S. Limmer, J. Organomet. Chem. 1991,
413, 55.
Experimental
General. Unless otherwise noted all reactions were carried
out under dry nitrogen atmosphere. Solvents were dried and
purified under nitrogen before use by standard procedures.
Chemical reagents were purchased and used as received.
1H, 13C, and 31P NMR spectra were recorded on a JEOL
LA-500 instrument (500 MHz for 1H, 125.4 MHz for 13C,
and 201.9 MHz for 31P NMR spectroscopy). Unless otherwise
noted, CDCl3 was used as the solvent. Chemical shift values
for 1H and 13C were referred to internal Me4Si (0 ppm), and that
for 31P was referred to H3PO4 (85% solution in D2O, 0 ppm).
Mass spectra were measured on a Shimadzu GCMS-QP2010
spectrometer (EI). HRMS analysis was performed by the
Analytical Center at the National Institute of Advanced
Industrial Science and Technology.
Crystallographic data have been deposited with Cambridge
Crystallographic Data Centre: Deposition number CCDC-
765608 for compound trans-18. Copies of the data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the Cambridge Crystallographic Data
Centre, 12, Union Road, Cambridge, CB2 1EZ, U.K.; Fax:
+44 1223 336033; e-mail: deposit@ccdc.cam.ac.uk).
2
a) J. Green, in Fire Retardancy of Polymeric Materials, ed.
by A. F. Grand, C. A. Wilkie, Marcel Dekker, New York, 2000,
pp. 147-170. b) F. J. Welch, U.S. Patent 3312674, 1967. c) R.
Rabinowitz, Ger. Patent 1241995, 1967; R. Rabinowitz, Chem.
Abstr. 1967, 67, 44417d. d) R. S. Cooper, U.S. Patent 3340477,
1967. e) C. V. Juelke, L. E. Trapasso, U.S. Patent 3813454, 1974.
f) R. W. Stackman, Ind. Eng. Chem. Prod. Res. Dev. 1982, 21, 328.
g) A. J. Spak, A. J. Yu, U.S. Patent 4088710, 1978. h) J. C.
Goswami, U.S. Patent 4560618, 1985. i) B. Bingöl, W. H. Meyer,
M. Wagner, G. Wegner, Macromol. Rapid Commun. 2006, 27,
1719. j) J. Jin, J. C. H. Hua, U.S. Patent 3925506, 1975. k) M.
Fujimatsu, M. Miura, Japan Patent 86-250155, 1988; M.
Fujimatsu, M. Miura, Chem. Abstr. 1989, 110, 9644k. l) B. S.
Fuller, W. D. Ellis, R. M. Rowell, U.S. Patent 5683820, 1997.
m) J. R. Shukla, U.S. Patent 4241125, 1980. n) C. M. Welch, E. J.
Gonzales, J. D. Guthrie, J. Org. Chem. 1961, 26, 3270.
Pd(OAc)2-Catalyzed Addition of 3a to Alkynes: A
Representative Procedure. 1-Octyne (27.6 mg, 0.25 mmol),
3a (66.1 mg, 0.25 mmol), and Pd(OAc)2 (1.7 mg, 3 mol %) were
dissolved in 0.5 mL of dry toluene. The mixture was heated at
80 °C for 2 h in a sealed tube and evaporated in vacuo. The
residues were passed through a short silica gel column (CH2Cl2
solvent) to remove the catalyst, and was subjected to bulb-to-
bulb distillation under vacuum (1.2 © 10¹3 mmHg) to obtain
pure 4a as colorless oil (82.4 mg, 0.22 mmol, 88% yield).
4a: Colorless oil; bp 128-130 °C/1.2 © 10¹3 mmHg (bulb-
3
a) T. Minami, J. Motoyoshiya, Synthesis 1992, 333. b) M.
Mikolajczyk, P. Balczewski, Top. Curr. Chem. 2003, 223, 161.
c) V. M. Dembitsky, A. A.-A. Al Quntar, A. Haj-Yehia, M.
Srebnik, Mini-Rev. Org. Chem. 2005, 2, 91.
4
a) N. N. Girotra, N. L. Wendler, Tetrahedron Lett. 1969, 10,
4647. b) E. J. Glamkowski, G. Gal, R. Purick, A. J. Davidson, M.
Sletzinger, J. Org. Chem. 1970, 35, 3510. c) X.-Y. Wang, H.-C.
Shi, C. Sun, Z.-G. Zhang, Tetrahedron 2004, 60, 10993. d) E.
Öhler, E. Zbiral, M. El-Badawi, Tetrahedron Lett. 1983, 24, 5599.
e) E. Öhler, M. El-Badawi, E. Zbiral, Chem. Ber. 1985, 118, 4099.
f) C. E. Burgos-Lepley, S. A. Mizsak, R. A. Nugent, R. A.
Johnson, J. Org. Chem. 1993, 58, 4159. g) G. D. Duncan, Z.-M.
Li, A. B. Khare, C. E. McKenna, J. Org. Chem. 1995, 60, 7080.
h) P. C. B. Page, M. J. McKenzie, J. A. Gallagher, Synth. Commun.
2002, 32, 211. i) Y. Ono, L.-B. Han, Tetrahedron Lett. 2006, 47,
1
to-bulb distillation); H NMR: ¤ 0.88 (3H, t, J = 6.9 Hz), 1.14
(12H, s), 1.24-1.39 (6H, m), 1.26 (12H, s), 1.52 (2H, quint,
J = 7.5 Hz), 2.38 (2H, dd, JHP = 10.2 Hz, JHH = 6.7 Hz), 5.38
(1H, dd, JHP = 55.3 Hz, JHH = 1.5 Hz), 5.76 (1H, dd, JHP
25.0 Hz, JHH = 0.9 Hz); 13C NMR: ¤ 14.0, 22.6, 23.7 (d, JCP
=
=
7.1 Hz), 24.4 (d, JCP = 5.1 Hz), 28.1 (d, JCP = 8.3 Hz), 29.1,
31.8, 32.8 (d, JCP = 11.4 Hz), 78.2, 120.5 (d, JCP = 8.3 Hz),
150.3 (d, JCP = 204.7 Hz); 31P NMR ¤ ¹31.6. Anal. Calcd

1098 Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010)
Markovnikov Addition of P-H Bonds to Alkynes Mediated by Palladium
421. j) M. R. Harnden, A. Parkin, M. J. Parratt, R. M. Perkins,
J. Med. Chem. 1993, 36, 1343. k) Y. G. Smeyers, A. Hernandez-
Laguna, F. J. Romero-Sanchez, N. Fernandez-Ibanez, E. Galvez-
Ruano, S. Arias-Perez, J. Pharm. Sci. 1987, 76, 753. l) S. Megati,
S. Phadtare, J. Zemlicka, J. Org. Chem. 1992, 57, 2320. m) H. B.
Lazrek, A. Rochdi, H. Khaider, J.-L. Barascut, J.-L. Imbach, J.
Balzarini, M. Witvrouw, C. Pannecouque, E. De Clercq, Tetrahe-
dron 1998, 54, 3807. n) P. W. Smith, A. J. Chamiec, G. Chung,
K. N. Cobley, K. Duncan, P. D. Howes, A. R. Whittington, M. R.
Wood, J. Antibiot. 1995, 48, 73. o) S. A. Holstein, D. M. Cermak,
D. F. Wiemer, K. Lewis, R. J. Hohl, Bioorg. Med. Chem. 1998, 6,
687. p) L. H. Chance, J. P. Moreau, U.S. Patent 3910886, 1975.
Chem. Soc. 2004, 126, 5080. j) P. Ribière, K. Bravo-Altamirano,
M. I. Antczak, J. D. Hawkins, J.-L. Montchamp, J. Org. Chem.
2005, 70, 4064. k) R. A. Stockland, Jr., A. J. Lipman, J. A. Bawiec,
III, P. E. Morrison, I. A. Guzei, P. M. Findeis, J. F. Tamblin,
J. Organomet. Chem. 2006, 691, 4042. l) N. Dobashi, K. Fuse, T.
Hoshino, J. Kanada, T. Kashiwabara, C. Kobata, S. K. Nune, M.
Tanaka, Tetrahedron Lett. 2007, 48, 4669. m) M. Niu, H. Fu, Y.
Jiang, Y. Zhao, Chem. Commun. 2007, 272. n) L.-B. Han, C.-Q.
Zhao, S. Onozawa, M. Goto, M. Tanaka, J. Am. Chem. Soc. 2002,
124, 3842. o) L.-B. Han, F. Mirzaei, C.-Q. Zhao, M. Tanaka,
J. Am. Chem. Soc. 2000, 122, 5407. p) C.-Q. Zhao, L.-B. Han, M.
Goto, M. Tanaka, Angew. Chem., Int. Ed. 2001, 40, 1929. q) Q.
Xu, L.-B. Han, Org. Lett. 2006, 8, 2099. r) L.-B. Han, Y. Ono, H.
Yazawa, Org. Lett. 2005, 7, 2909. s) L.-B. Han, Y. Ono, S.
Shimada, J. Am. Chem. Soc. 2008, 130, 2752. t) T. Hirai, L.-B.
Han, J. Am. Chem. Soc. 2006, 128, 7422. u) T. Mizuta, C. Miyaji,
T. Katayama, J. Ushio, K. Kubo, K. Miyoshi, Organometallics
2009, 28, 539. v) N. Ajellal, C. M. Thomas, J.-F. Carpentier, Adv.
Synth. Catal. 2006, 348, 1093. w) M. Kalek, J. Stawinski,
Organometallics 2007, 26, 5840. x) M. C. Kohler, T. V. Grimes, X.
Wang, T.-R. Cundari, R. A. Stockland, Jr., Organometallics 2009,
28, 1193. y) M. C. Kohler, R. A. Stockland, Jr., N. P. Rath,
Organometallics 2006, 25, 5746.
13 http://katayamakagaku.co.jp/img/p_chemicalproduct2.pdf.
14 a) L. W. Nesterov, R. A. Sabirova, N. E. Krepisheva, R. I.
Mutalapova, Dokl. Akad. Nauk SSSR 1963, 148, 1085. For general
reviews see: b) Pentacoordinated Phopshorus, ed. by R. R.
Holmes, ACS, Washington, 1980, Vols. 1 and 2. c) J. C. Tebby, in
Phosphorus-Carbon Hererocyclic Chemistry: The Rise of A New
Domain, ed. by F. Mathey, Pergamon, Amsterdam, 2001,
Chap. 6.3. d) T. Kawashima, in Chemistry of Hypervalent
Compounds, ed. by K. Akiba, Wiley, New York, 1999, Chap. 6.
15 (RO)2P(O)H exists in two tautomeric forms, P(=O)H and
P-OH; the former is dominant in equilibrium: L. A. Hamilton,
P. S. Landis, in Organic Phosphorus Compounds, ed. by G. M.
Kosolapoff, L. Maier, Wiley, New York, 1972, Vol. 4, Chap. 11.
16 Depending on the substituents, H-spirophosphoranes can
also exist as a mixture of 5-coodinated P(V) and 3-coodinate P(III)
forms. However, 100% 5-coodinated P(V) form was reported for
3a. D. Bernard, C. Laurenco, R. Burgada, J. Organomet. Chem.
1973, 47, 113.
17 This work was partially presented at the 86th Annual
Meeting of the Chemical Society of Japan, Funabashi, March
27-30, 2006; L.-B. Han, Y. Ono, Jpn. Kokai Tokkyo Koho 2007-
137850 (filled, November 22, 2005; opened, June 7, 2007). Only
a few alkenyl spirophosphoranes having electron-withdrawing
carboxyl groups are known: a) M. Sanchez, R. Wolf, R. Burgada,
F. Mathis, Bull. Soc. Chim. Fr. 1968, 773. b) C. Laurenço, R.
Burgada, Tetrahedron 1976, 32, 2089. c) R. Burgada, A. Mohri,
Phosphorus, Sulfur Silicon Relat. Elem. 1981, 9, 285.
18 For the synthesis of hypervalent phosphoranidometal
complexes: a) K. Kajiyama, Y. Hirai, T. Otsuka, H. Yuge, T. K.
Miyamoto, Chem. Lett. 2000, 784. b) K. Kajiyama, T. K.
Miyamoto, K. Sawano, Inorg. Chem. 2006, 45, 502. c) K. Kubo,
H. Nakazawa, Y. Ogitani, K. Miyoshi, Bull. Chem. Soc. Jpn. 2001,
74, 1411. d) H. Nakazawa, K. Kawamura, T. Ogawa, K. Miyoshi,
J. Organomet. Chem. 2002, 646, 204. e) T. Głowiak, W. K. Rybak,
A. Skarżyńska, Polyhedron 2000, 19, 2667. f) A. Skarżyńska,
W. K. Rybak, T. Głowiak, Polyhedron 2001, 20, 2667. g) K. B.
Dillon, Chem. Rev. 1994, 94, 1441.
5
a) R. K. Haynes, S. C. Vonwiller, T. W. Hambley, J. Org.
Chem. 1989, 54, 5162. b) R. K. Haynes, W. A. Loughlin, T. W.
Hambley, J. Org. Chem. 1991, 56, 5785. c) E. F. Birse, M. D.
Ironside, A. W. Murray, Tetrahedron Lett. 1995, 36, 6309. d) K.
Yamashita, T. Oshikawa, H. Imoto, Japan Patent S62-207281,
1987; K. Yamashita, T. Oshikawa, H. Imoto, Chem. Abstr. 1988,
108, 167691m. e) K. Yamashita, H. Imoto, Y. Tamada, T.
Oshikawa, Japan Patent H01-45390, 1989; K. Yamashita, H.
Imoto, Y. Tamada, T. Oshikawa, Chem. Abstr. 1989, 111, 115592x.
6
a) W. Tian, Z. Zhu, Q. Liao, Y. Wu, Bioorg. Med. Chem.
Lett. 1998, 8, 1949. b) A. A.-A. Al-Quntar, O. Baum, R. Reich, M.
Srebnik, Arch. Pharm. 2004, 337, 76.
7
a) X. Wang, L.-B. Han, Japan Patent 2007-070278, 2007;
X. Wang, L.-B. Han, Chem Abstr. 2007, 146, 330815. b) X. Wang,
L.-B. Han, Japan Patent 2007-137838, 2007; X. Wang, L.-B. Han,
Chem Abstr. 2007, 146, 23739. c) X. Wang, L.-B. Han, Japan
Patent 2007-063168, 2007; X. Wang, L.-B. Han, Chem Abstr.
2007, 146, 317035.
8
a) B. Aebischer, P. Frey, H.-P. Haerter, P. L. Herrling, W.
Mueller, H. J. Olverman, J. C. Watkins, Helv. Chim. Acta 1989, 72,
1043. b) D. A. Lowe, H. C. Neijt, B. Aebischer, Neurosci. Lett.
1990, 113, 315. c) H. Potschka, M. Fedrowitz, W. Löscher, Eur. J.
Pharmacol. 1999, 374, 175.
9
L.-B. Han, M. Tanaka, J. Am. Chem. Soc. 1996, 118, 1571.
10 As to the background of metal-mediated P(O)-H additions
to carbon-carbon unsaturated bonds before our work (Ref. 9), a
U.S. patent claimed the addition of (RO)2P(O)H to alkynes by
Ni(PR3)2X2 at elevated temperatures. a) K. Lin, U.S. Patent
3681481, 1972. Hirao also briefly claimed in his paper describing
the coupling of (RO)2P(O)H with organic halides that (RO)2P(O)H
added to isoprene, though the efficiency was low (10% yield,
150 °C, 20 h). b) T. Hirao, T. Masunaga, N. Yamada, Y. Ohshiro, T.
Agawa, Bull. Chem. Soc. Jpn. 1982, 55, 909.
11 Reviews: a) L.-B. Han, M. Tanaka, Chem. Commun. 1999,
395. b) L. Coudray, J.-L. Montchamp, Eur. J. Org. Chem. 2008,
3601. c) D. Prim, J.-M. Campagne, D. Joseph, B. Andrioletti,
Tetrahedron 2002, 58, 2041. d) A. L. Schwan, Chem. Soc. Rev.
2004, 33, 218. e) D. S. Glueck, Synlett 2007, 2627. f) V. P.
Ananikov, L. L. Khemchyan, I. P. Beletskaya, Synlett 2009, 2375.
12 Selected examples for metal-catalyzed P(O)-H additions:
a) S. Kawaguchi, S. Nagata, A. Nomoto, M. Sonoda, A. Ogawa,
J. Org. Chem. 2008, 73, 7928. b) L.-B. Han, N. Choi, M. Tanaka,
Organometallics 1996, 15, 3259. c) L.-B. Han, R. Hua, M. Tanaka,
Angew. Chem., Int. Ed. 1998, 37, 94. d) C.-Q. Zhao, L.-B. Han, M.
Tanaka, Organometallics 2000, 19, 4196. e) L.-B. Han, C.-Q.
Zhao, M. Tanaka, J. Org. Chem. 2001, 66, 5929. f) S. Deprèle,
J.-L. Montchamp, J. Am. Chem. Soc. 2002, 124, 9386. g) A. Allen,
Jr., L. Ma, W. Lin, Tetrahedron Lett. 2002, 43, 3707. h) M. D.
Milton, G. Onodera, Y. Nishibayashi, S. Uemura, Org. Lett. 2004,
6, 3993. i) L.-B. Han, C. Zhang, H. Yazawa, S. Shimada, J. Am.
19 (Trimethylsilyl)acetylene only gave the trans anti-
Markovnikov adduct, which was apparently not due to steric but

L.-B. Han et al.
Bull. Chem. Soc. Jpn. Vol. 83, No. 9 (2010) 1099
J. F. Marecek, J. Am. Chem. Soc. 1977, 99, 4515.
electronic effect.
20 Acetylene gas for industry use (ca. 96% purity) was used as
received without further purification.
23 X-ray data of complexes 12 and 14 are available from SI of
Ref. 12s.
21 T. Bailly, R. Burgada, Phosphorus, Sulfur Silicon Relat.
Elem. 1994, 86, 217.
24 D. E. C. Corbridge, Top. Phosphorus Chem. 1966, 3, 57.
For recent reported X-ray data of P(O)-Pd (Pt) complexes see
Refs. 12b and 12n.
22 Similar transformations of 8 to 7 and 10 to 6 have been
described in the literature. Similar transformations of 8 to 7: a) D.
Bernard, R. Burgada, Tetrahedron Lett. 1973, 14, 3455. b) A.
Munoz, B. Garrigues, M. Koenig, Tetrahedron 1980, 36, 2467.
Similar transformations of 10 to 6: c) F. Ramirez, M. Nowakowski,
25 C. A. Jaska, H. Dorn, A. J. Lough, I. Manners, Chem.®Eur.
J. 2003, 9, 271.
26 U. Vogel, M. Scheer, Z. Anorg. Allg. Chem. 2001, 627,
1593.