Novel, stereoselective and stereospecific synthesis of allenylphosphonates and related compounds via palladium-catalyzed propargylic substitution

Article abstract of DOI:10.1002/adsc.201100119

We have developed a novel method for the synthesis of allenylphosphonates and related compounds based on a palladium(0)-catalyzed reaction of propargylic derivatives with H-phosphonate, H-phosphonothioate, H-phosphonoselenoate, and H-phosphinate esters. The reaction is stereoselective and stereospecific, and provides a convenient entry to a vast array of allenylphosphonates and their analogues with diverse substitution patterns in the allenic moiety and at the phosphorus center. Some mechanistic aspects of this new reaction were also investigated. Copyright

Full text of DOI:10.1002/adsc.201100119

FULL PAPERS
DOI: 10.1002/adsc.201100119
Novel, Stereoselective and Stereospecific Synthesis of Allenyl-
phosphonates and Related Compounds via Palladium-Catalyzed
Propargylic Substitution
Marcin Kaleka and Jacek Stawinskia,b,*
a
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, S-106 91 Stockholm, Sweden
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61-704 Poznan, Poland
b
Fax : (+46)-8-154-908; e-mail: js@organ.su.se
Received: February 16, 2011; Published online: June 30, 2011
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adcs.201100119.
Abstract: We have developed a novel method for logues with diverse substitution patterns in the allen-
the synthesis of allenylphosphonates and related ic moiety and at the phosphorus center. Some mech-
compounds based on a palladium(0)-catalyzed reac- anistic aspects of this new reaction were also investi-
tion of propargylic derivatives with H-phosphonate, gated.
H-phosphonothioate, H-phosphonoselenoate, and H-
phosphinate esters. The reaction is stereoselective Keywords: allenes; allenylphosphinates; allenyl-
and stereospecific, and provides a convenient entry phosphonates; H-phosphonate diesters; palladium;
to a vast array of allenylphosphonates and their ana- propargylic substitution; SN2’ reactions
Introduction
and S-nucleophiles,[9] as well as selective total[10] or
partial[11] hydrogenation, radical reactions,[12] as well
In the last decade, the chemistry of allenes has as Diels–Alder[13] and other cycloadditions.[14] Allenyl-
emerged as a flourishing research area in organic phosphonates can be further functionalized by metal-
chemistry.[1] In addition, due to the discovery of this catalyzed reactions[15] and the phosphonate group in
structural fragment in many natural products, these these compounds can be employed in the Horner–
chemical entities have started to attract also the in- Wadsworth–Emmons olefination.[16] Activation of the
creasing attention of pharmaceutical research.[2] With allene moiety with electrophilic reagents leads to cy-
the development of novel synthetic methods,[3] with a
clization producing oxaphospholanes,[17] and many of
ACHTUNGTRENNUNG
prominent position of metal-catalyzed reactions,[4] al- the above reactions can be combined in tandem pro-
lenes became useful intermediates in the chemical cesses[18] enabling, for instance, the synthesis of com-
synthesis, frequently enabling a rapid molecular com- plex heterocycles.[7c,15b,19]
plexity increase in a single reaction step.[3a,5] The
Phosphorus-containing compounds have always
transfer of the axial chirality to a newly formed ste- been attractive targets in biological and medicinal
reogenic center in a product observed for such reac- chemistry,[20] as they may serve as analogues of bio-
tions,[1,6] opens new possibilities in stereoselective syn- logically important compounds, e.g., nucleic acids,
thesis.
phospholipids, phosphorylated sugars, or be used as
As part of our interest in the synthesis of biologi- enzyme inhibitors. Although an allene moiety has
cally important phosphorus compounds, we embarked been extensively used in pharmacologically active
recently on investigations of allenylphosphonates and compounds,[2] allenylphosphonates have not been ex-
related compounds. The attractive feature of these plored yet in this context. We believe that this can be
compounds is diverse reactivity that, with substituent- attributed to the lack of universal synthetic methods
loading capability and axial chirality of the allene meeting the requirements of preparation of complex
unit, makes them potentially useful synthetic inter- natural products analogues.
mediates. Due to the electron-acceptor nature of a
Except for some special cases,[13b,21] the synthesis of
phosphonate group, allenylphosphonates undergo allenylphosphonates has been dominated by [2,3]-sig-
facile and selective additions of various N-,[7] O-,[8] matropic rearrangement reactions.[1] Discovered in
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1741

Marcin Kalek and Jacek Stawinski
FULL PAPERS
two identical, usually simple alkyl, substituents at the
phosphorus center.
In order to overcome the limitations of these meth-
ods based on the sigmatropic rearrangement, we
turned our attention to an in this context completely
unexplored transition metal-catalyzed propargylic
substitution (SN2’) reaction as a means for the synthe-
sis of allenylphosphonates. Although, this is a well es-
tablished synthetic approach to a variety of al-
Scheme 1. Synthesis of allenylphosphonates via [2,3]-sigma-
tropic rearrangement.
the early 1960s, these involve propargylic phosphite lenes,[1,24] the reaction has never been used for the for-
esters that spontaneously rearrange to allenylphosph- mation of a carbon-phosphorus bond. An appealing
onates (Scheme 1)[22] with complete center to axial feature of this type of palladium- or copper-catalyzed
chirality transfer from the propargylic alcohol to the reaction is that often stereoselectivity and chirality
allene.[23] The method has a very broad scope with re- transfer from the propargylic substrate to the allene
spect to the precursor propargylic alcohols used and it moiety is observed.[1] Since Pd-catalyzed C P bond
À
enables the synthesis of allenylphosphonates with var- formation has been shown to work well[25] in the syn-
ious substitution patterns in the allene moiety. De- thesis of biologically important phosphorus com-
spite these advantages, the method suffers from a seri- pounds,[26] and mechanistic aspects of the reaction
ous drawback, namely, the limited number of phos- have been studied in depth,[27] we expected that these
phonate derivatives that can be prepared. Phosphorus may lend themselves to a new methodology for the
compounds that are used for the synthesis of propar- construction of complex allenylphosphonate deriva-
gylic phosphites are phosphorochloridites, highly reac- tives.
tive compounds, prone to hydrolysis and oxidation,
There is also an interesting mechanistic aspect of a
and thus only simple alkyl derivatives can be used for Pd-catalyzed synthesis of C-allenes vs heteroatom-
the preparation of allenylphosphonates. This feature substituted allenes (e.g., allenylphosphonates). Apart
limits the applicability of the procedure to the synthe- from a transmetallation of the equilibrating allenyl-
sis of only symmetrical allenylphosphonates bearing and propargyl-palladium species[28] (Scheme 2, paths a
Scheme 2. Possible reaction pathways during Pd-catalyzed propargylic substitution with phosphonate nucleophiles.
1742
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
and c; formation of an allenyl- and propargyl-palladi- gylphosphonates.[31] Since the degree of the conver-
um-phosphonate intermediates, respectively), a ꢁsoftꢂ sion with propargyl bromide was rather low (ca.
heteroatom nucleophile may attack a central carbon 30%), apparently due to its reaction with triethyl-
atom of the ligand (Scheme 2, path b).[24,29] For phos-
ACHTUNGTRENaNGNU mine that was used as a base, in the subsequent stud-
phorus nucleophiles, in the first instance, an allenyl- ies less reactive propargylic derivatives (chlorides, to-
phosphonate (or propargylphosphonate) should be sylates, etc., Table 2) were employed.
formed after reductive elimination, while in the
second scenario, formation of either diene (via b-hy-
dride elimination, path b1) or bisphosphonate (path Optimization of the Reaction Conditions
b2) might be favoured. As shown by the work of
Kozawa et al. on amide nucleophiles, the outcome of In order to evaluate different ligands for their efficacy
the reaction can be controlled by the supporting li- to promote the formation of allenylphosphonates, the
gands used.[30]
reaction between propargyl chloride (1a) and diethyl
Herein, we report studies on a palladium-catalyzed H-phosphonate (2) was studied in THF as a solvent at
propargylic substitution reaction with H-phospho- 688C, in the presence of Et3N as a base (Table 1).
nates and related compounds as nucleophiles, aiming
The screening revealed that only bidentate ligands
À
at the development of a new methodology for C P with large bite angles were able to promote a conver-
bond formation, and to expand the scope of accessible sion into the allenylphosphonate (Table 1, entries 4–
allenylphosphonates, particularly those of potential 7), and the highest reaction rate was observed for
biochemical relevance. A preliminary account of this DPEPhos. To our delight, the desired allenylphospho-
work has recently been published as a short commu- nate 3 was always the sole product of the reaction, ir-
nication.[31]
respective of the bidentate ligand used.
To investigate and optimize further the experimen-
tal conditions, the reactions with primary (1) and sec-
ondary (4) propargylic derivatives bearing different
leaving groups [Cl, MeOC(O)O, TsO, AcO] were car-
Results and Discussion
Our exploratory studies on the reaction of propargylic ried out. In all cases 3 mol% palladium loading was
halides with H-phosphonate diesters in the presence used together with DPEPhos ligand. For some entries
of Pd(0) as a catalyst revealed formation of allenyl- also an effect of a base (Et3N) and anionic additives
phosphonates rather than the corresponding propar- was studied (Table 2).
The data from Table 2 provide practical guidelines
for designing a synthetic method based on this reac-
tion and also bear some mechanistic hints (see
below). As it is apparent from entry 1, the propargylic
chlorides showed relatively high reactivity, that did
not depend on the presence or absence of an alkyl
substituent on C-1 (primary vs. secondary propargylic
derivatives). As expected, this reaction required the
presence of a stoichiometric amount of a base
(Table 2, entry 1 vs. entry 2). Propargylic carbonates,
on the other hand, did not require an external base,
and exhibited significant differences in rates between
primary vs. secondary substrates (entry 3), with the
later one reacting much faster (30 min vs. 7 h). The
tosyl derivatives (1c and 4c, Table 2, entry 4) were
less reactive than the corresponding chlorides and in
case of a primary propargylic derivative (1c), a partial
isomerization of the product to 1-alkynylphosphonate
derivative[32] was observed, apparently due to pro-
longed exposure to the base, necessary for reaching
the full conversion. Finally, propargylic acetates were
investigated (entries 6 and 7). These were found to be
rather poor substrates for the reaction and in the
presence of Et3N, a large fraction of the products iso-
merized to the corresponding 1-alkynylphosphonates
due to the very long reaction times (Table 2, entry 6).
In the absence of a base, the reactivity improved, but
Table 1. Evaluation of ligands.[a]
Entry
Ligand
Pd(PPh3)4
PPh3
dppp
Reaction time[b]
[c]
1
2
3
4
5
6
7
G
no reaction
no reaction
no reaction
3 h
1.5 h
2 h
dppf
DPEPhos
Xantphos
BINAP
6 h
[a]
Abbreviations:
dppp=1,3-bis(diphenylphosphino)pro-
pane, dppf=1,1’-bis(diphenylphosphino)ferrocene, DPE-
Phos=bis(2-diphenylphosphin)diphenyl ether, Xant-
phos=9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene,
BINAP=2,2’-bis(diphenylphosphino)-1,1’-binaphtyl (rac-
emic). Reaction conditions: 0.28 mmol 1a, 0.25 mmol 2,
0.30 mmol Et3N, 1.5 mol% Pd2ACHTUNTRGNEUNG(dba)3·CHCl3, 3 mol% bi-
dentate or 6 mol% monodentate ligand, 1 mL THF
(0.25M), 688C.
[b]
[c]
>95% conversion (31P NMR).
No Pd2ACHTUNGTRENNUNG(dba)3·CHCl3 was used.
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1743

Marcin Kalek and Jacek Stawinski
FULL PAPERS
Table 2. Effect of the leaving group and additives.[a]
Entry
Leaving group
Additive[b]
Reaction time[c]
R=H
R=n-C5H11
1
2
3
4
5
6
7
8
9
Cl
Cl
Et3N
–
–
Et3N
–
Et3N
1.5 h
1.5 h
no reaction
7 h
no reaction
30 min
5 h
OCO2Me
OTs
OTs
OAc
OAc
OTs
OTs
10 h[d]
very slow reaction
very slow reaction
48 h[d]
24 h[d]
–
24 h[e]
8 h[e]
Et3N, ClÀ
Et3N, OAcÀ
1.5 h
45 min
1.5 h
45 min
[a]
Reaction conditions: 0.28 mmol 1 or 4, 0.25 mmol 2, 1.5 mol% Pd
688C.
2ACHTUNGRTNE(NUNG dba)3·CHCl3, 3 mol% DPEPhos, 1 mL THF (0.25M),
[b]
[c]
[d]
[e]
0.30 mmol Et3N or 0.075 mmol (30 mol%) (n-Bu)4N+ anionÀ.
>95% conversion (31P NMR).
Partial isomerization of the product to 1-alkynylphosphonate (ca. 50%).
Formation of 1,3-dienylphosphonate and other by-products (total, ca. 40%).
acidification of the reaction mixtures by the AcOH tion. Since propargylic chlorides and carbonates gave
formed during the course of the reaction, led to for- the best results without the necessity of using any ad-
mation of multiple side-products (Table 2, entry 7).[33] ditives, these two groups of substrates were used in
We also studied possible effects of anionic additi- the further investigations (Table 3).
ves[26f,27a] on the outcome of the reaction with propar-
The scope of the reaction with regard to a substitu-
gylic tosylates. Since tosylate anions are weakly bond- tion pattern in propargylic substrates was investigated
ing to the palladium(II) center, their replacement using diethyl H-phosphonate (2) as a P-nucleophile.
with the added anions (chloride or acetate) could po- As it is apparent from entries 1–19 (Table 3), the fol-
tentially affect the transmetallation step. Indeed, addi- lowing trends in reactivity can be observed. The pres-
tion to the reaction mixture of tetrabutylammonium ence of one or two substituents (R1 and R2) at the C-
chloride shortened down the reaction times to 1.5 h 1 carbon greatly increased the reaction rate of the
for both primary and secondary propargylic deriva- propargylic carbonates, while it had only a minor
tives (Table 2, entry 8), a value identical to that ob- effect on the reaction times of the corresponding
served when propargylic chlorides were used as sub- propargylic chlorides. On the other hand, any sub-
strates (Table 2, entry 1). Addition of external acetate stituent other than hydrogen in the terminal position
anions [(n-Bu)4NOAc] resulted also in a remarkable of the alkyne (R3) dramatically slowed down the reac-
shortening of the reaction times to 45 min (Table 2, tion for both the leaving groups. These two opposite
entry 9), and this indicated that the observed poor re- effects appeared to be approximately additive. Impor-
activity of propargylic acetates in this reaction tantly, good conversions in case of the starting materi-
(Table 2, entries 6 and 7) was apparently due to an in- als containing the R3 substituent could only be ach-
efficient oxidative addition step.
ieved when carbonate was the leaving group (Table 3,
entries 9, 11, 16, 17, and 19). However, in line with
what was stated above, primary propargylic carbo-
nates 9b and 10b with terminal substituents reacted
very slowly (Table 3, entries 9 and 11, respectively),
and in the latter case a microwave irradiation was re-
Synthesis of Allenylphosphonates and
Allenylphosphinates
Having established the reactivity of different propar- quired to drive the reaction to completion. Hence,
gylic substrates under various experimental condi- except for the simple unsubstituted propargyl system
tions, we decided to examine the scope of the reac- (Table 3, entries 1 and 2), the propargylic carbonates
1744
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1745

Marcin Kalek and Jacek Stawinski
FULL PAPERS
displayed higher reactivity than the corresponding
chlorides. Overall, the Pd-catalyzed reaction fully
matched the scope of the method based on a sigma-
tropic rearrangement (Scheme 1), enabling synthesis
of unsubstituted (entries 1 and 2), and mono- (en-
tries 3–11), di- (entries 12–17), and tri-substituted (en-
tries 18 and 19, all in Table 3) allenylphosphonates.
Interestingly, in a separate experiment we found that
also chloroallene (8, Table 3, entry 7) can be efficient-
ly used as a substrate in an analogous cross-coupling
reaction.
Next, various phosphorus nucleophiles were tested,
including different H-phosphonate diesters (Table 3,
entries 20 and 21), and three phosphinates (en-
tries 22–24). They all could be efficiently transformed
into the corresponding allenylphosphonates and alle-
nylphosphinates, although it became apparent that
the reaction was sensitive to steric hindrance (com-
pounds 17 and 20, Table 3, entries 21 and 24). One
should note, however, that due to essentially neutral
pH of the reactions mixtures involving propargylic
carbonates as substrates, long reaction times did not
result in any deterioration (isomerization) of the
products.
Stereochemical Aspects of Allenylphosphonate
Formation
As it was mentioned in the introduction, the synthesis
of allenylphosphonates via a sigmatropic rearrange-
ment occurs with stereospecificity of the allenyl
moiety formation. Taking into account an easy access
to highly enantiomerically enriched propargylic alco-
hols[34] (e.g., from enzymatic kinetic resolution[35] or
stereoselective addition to aldehydes[36]) this feature
would be beneficial in the synthesis of complex alle-
nylphosphonate derivatives. Therefore, we set out to
investigate if the palladium-catalyzed reaction also
can compete with the sigmatropic rearrangement
method in this respect. In order to evaluate this possi-
bility, enantioenriched propargylic chloride 4a and
propargylic carbonate 4b (obtained from the corre-
sponding alcohols), were allowed to react with diethyl
H-phosphonate 2 under the developed conditions.
As it is apparent from the results presented in
Scheme 3, the stereochemical outcome of the reaction
depended on the leaving group present in the starting
material. Using enantiomeric propargylic chlorides
(R)- and (S)-4a as substrates resulted in allenyl-
phosphonates (S)- and (R)-5, respectively (see below
for the absolute stereochemistry assignment), howev-
er, the products ee was reduced to ~93% of that of
the reactants. In contradistinction to this, propargylic
carbonates (R)- and (S)-4b were transformed into the
corresponding enantiomers of 5 with complete (within
the experimental error) stereoselectivity.
1746
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
relating the absolute configuration of the carbon- and
hydrogen-substituted allenes with the sign of their op-
tical rotation. Since the applicability of these rules to
phosphorus-substituted allenes is unknown, we estab-
lished the absolute configuration of allenylphospho-
nate 5 by a stereochemical correlation method. To
this end we took advantage of the fact that chiral alle-
nylphosphonate 5 can be synthesized stereospecifi-
cally via an alternative method, namely by the sigma-
tropic rearrangement reaction, that due to its concert-
ed character must occur in a syn fashion (Scheme 1).
Since the allenylphosphonates 5 synthesized from the
same enantiomer of the propargylic alcohol by the
Scheme 3. Reaction of the enantioenriched propargylic
chlorides (4a) and carbonates (4b) with H-phosphonate 2.
To obtain a deeper insight into origin of these dif- sigmatropic rearrangement and the Pd-catalyzed
ferences, the stereochemical stability of allenyl- propargylic substitution had opposite configurations
phosphonate 5 under different experimental condi- (Scheme 5), we tentatively concluded that the latter
tions was examined. This revealed that compound 5 reaction occurs with an anti-stereoselectivity – similar-
was prone to racemization by prolonged heating (in ly to the previously reported stereochemistry for the
THF, at 688C), and this process was substantially ac- synthesis of C-allenes.
celerated by nucleophilic species, such as ClÀ, OAcÀ,
Apart from the center to axis chirality transfer de-
and palladium(0) [but not palladium(II)] complexes scribed above, the other important aspect of the alle-
(see the Supporting Information). Such a behaviour nylphosphonate synthesis is chirality at the phospho-
of phosphonate 5 was not surprising since an allene rus center. Since it has been already shown that palla-
moiety may undergo reversible nucleophilic addition dium-catalyzed propargylic substitution provides an
at the b-carbon, resulting in racemization[37] as shown easy access to allenylphosphonates with four different
in Scheme 4. Since the reactions with propargylic substituents at the phosphorus atom (Table 3, en-
chlorides 4a took longer time than those with propar- tries 22–24), therefore we decided to study the stereo-
gylic carbonates 4b (1.5 h vs. 30 min, respectively, see chemical course of the investigated reaction when the
Table 3, entry 3 vs. 4), and in the former case also ClÀ starting material contained a stereogenic phosphorus
ions were present in the reaction mixture, these can center. As model compounds for these studies we
account for the observed reduced stereoselectivity chose two diastereomeric dinucleoside H-phospho-
when propargylic chlorides were used as substrates. nates with opposite configurations at the phosphorus
However, one cannot exclude a possibility that other center, (RP)-35 and (SP)-35, that were subjected sepa-
factors may also contribute to a partial racemization rately to coupling with propargylic chloride 1a under
during the course of the reaction.
the developed reaction conditions (Scheme 6). It was
Palladium-catalyzed SN2’ propargylic substitution is found that the formation of dinucleoside allenyl-
known to occur with an anti-stereoselectivity,[38] and phosphonates 36 from the corresponding dinucleoside
the absolute stereochemistry of the allenic products is
assigned on the basis of Lowe–Brewster rules,[39] cor-
Scheme 5. Stereochemical correlation of the allenylphospho-
nates 5 synthesized by the sigmatropic rearrangement and
Pd-catalyzed propargylic substitution.
Scheme 4. A plausible racemization mechanism of allenyl-
phosphonate 5 in the presence of nucleophilic species.
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1747

Marcin Kalek and Jacek Stawinski
FULL PAPERS
Scheme 6. Stereochemical course of the palladium-catalyzed propargylic substitution with P-chiral reactants (Thy=thymin-
1-yl, DMT=4,4’-dimethoxytrityl, TBDMS=tert-butyldimethylsilyl).
H-phosphonates 35 was completely stereospecific Reactions of Diphenylphosphine Oxide with
[Scheme 6, conversion of (RP)-35 into (RP)-36 and Propargylic Derivatives
(SP)-35 into (SP)-36], and occurred most likely with
retention of configuration at the phosphorus center. Exclusive formation of allenylphosphonates or alle-
Such an outcome was not unexpected, since other Pd- nylphosphinates in the reactions discussed above
catalyzed cross-coupling reactions of H-phosphonates might indicate that in the transmetallation step
with electrophiles, such as aryl and benzyl halides, are (Scheme 2) H-phosphonates and H-phosphinates
known to be stereospecific[26c–f,40] and occur with re- acted as hard nucleophiles. To explore a mechanistic
tention of the configuration at the phosphorus part of this Pd-catalyzed reaction, we searched for a
center.[41]
softer phosphorus nucleophile that could attack a cen-
Full control of the stereochemistry at the phospho- tral carbon atom of the palladium complex rather
rus center seems to be a significant advantage of the than palladium itself (path b in Scheme 2).
Pd-catalyzed vs. sigmatropic rearrangement method
Indeed, when diphenylphosphine oxide 37 was sub-
for the synthesis of allenylphosphonates on two jected to the reaction with either propargyl chloride
counts: (i) easy preparation of H-phosphonate die- (1a) or carbonate (1b) in the presence of a palladi-
sters and their analogues with defined stereochemistry um(0) catalyst (Table 4) approximately equimolar
at phosphorus centers, and (ii) high stereochemical amounts of allenylphosphine oxide (38) and bis(phos-
stability of tetracoordinated H-phosphonates vs. ter- phine oxide) (39) were obtained (Table 4, entries 1
valent phosphite derivatives.
and 2). Formation of the latter product could be ex-
mechanism path 2b in
Scheme 6 illustrates also the fact that the palladi- plained by assuming
a
um-catalyzed reaction of propargylic compounds with Scheme 2, a reaction pathway that is favoured by soft
H-phosphonate diesters can be applied to the synthe- nucleophiles (e.g., malonate anion).[29]
sis of complex organic compounds. Since preparation
Since steric hindrance at C-1 should disfavour this
of H-phosphonate diesters is well established,[42] this mechanistic pathway but not those involving a trans-
method may provide a convenient entry to allenyl- metallation, we expected that reaction of C-1 disubsti-
phosphonates bearing biologically important moieties tuted propargylic derivatives would afford mainly the
attached to the phosphorus center.
corresponding allenylphosphine oxides. As it is appar-
ent from entries 3 and 4 in Table 4, the reactions of
dimethyl-substituted propargylic substrates (chloride
11a and carbonate 11b) with diphenylphosphine oxide
1748
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
Table 4. Reactions of diphenylphosphine oxide with propargylic derivatives.[a]
Entry
1
Propargylic substrate
P-nucleophile
Products
Reaction time (isolated yields)
1 h (38: 42%; 39: 31%)
2
3
45 min (38: 53%; 39: 27%)
1.5 h (77%)
4
30 min (84%)
[a]
Reaction conditions: 1.38 mmol propargylic substrate, 1.25 mmol P-nucleophile, 0.019 mmol (1.5 mol%) Pd
2ACHTUNGTRNEN(UNG dba)3·CHCl3,
0.038 mmol (3 mol%) DPEPhos, 5 mL THF (0.25M), 688C. For propargylic chlorides, additionally 1.5 mmol Et3N.
37 indeed afforded exclusively the allenylphosphine isomeric
propargylphosphonothioate
(e.g.,
45;
oxide 40.
Table 5, entries 1 and 2). A similar outcome of the re-
action was also observed for dinucleoside H-phospho-
nothioate 42 (Table 5, entry 6) and for terminally sub-
stituted propargylic carbonate 9b (Table 5, entry 3).[44]
In the absence of the palladium catalyst, no reaction
could be observed between propargylic substrates (1a,
Synthesis of Thio and Seleno Analogues of
Allenylphosphonates
To expand the scope of this reaction further, we 1b, and 9b) and H-phosphonothioate 41. These results
turned our attention to H-phosphonothio- and H- suggested that for H-phosphonothioate nucleophiles
phosphonoselenoate diesters, as possible phosphorus transmetallation of the propargylpalladium(II) com-
nucleophiles, since compounds containing sulfur or se- plexes became a feasible reaction pathway (Scheme 2,
lenium bound to the phosphorus are attractive syn- path c).
thetic targets due to often favorable biochemical
Since isomeric allenyl- and propargylphosphono-
thioates (44 vs. 45, 46 vs. 47, 50 vs. 51) turned out to
properties.[43]
First, the reaction of different propargylic deriva- be inseparable using standard experimental tech-
tives with a model diethyl H-phosphonothioate (41) niques, we attempted to suppress the formation of un-
was carried out using the developed synthetic proto- desirable, in this context, propargylphosphono-
col (Table 5, entries 1–5). Somewhat surprisingly, in thioates. On a mechanistic ground, it seemed likely
contradistinction to diethyl H-phosphonate, that in that transmetallation of propargylpalladium(II) com-
the reaction with simple propargylic derivatives pro- plexes should be sensitive to steric hindrance from
duced exclusively allenylphosphonates, the thio con- substituents on C-1, and thus mono- and di-substitut-
gener 41 diverted the reaction towards propargylphos- ed propargylic derivatives might favour transmetalla-
phonothioate derivatives. For primary propargylic tion of allenylpalladium(II) species, and in conse-
compounds containing chloride or carbonate as a quence, formation of allenylphosphonothioates.
leaving group, the reaction afforded a mixture of the Indeed, the reaction of propargylic derivatives bear-
desired allenylphosphonothioate (e.g., 44) and the ing one or two C-1 substituents (R1 and R2) with H-
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1749

Marcin Kalek and Jacek Stawinski
FULL PAPERS
phosphonothioate 41 resulted in exclusive formation
With this knowledge, we decided to examine if
of allenylphosphonothioates 48 and 49 which were more complex H-phosphonothioates can also be used
isolated in ca. 60% yield (Table 5, entries 4 and 5).
in this reaction. To this end, we subjected two sepa-
rate diastereomers of dinucleoside H-phosphono-
Table 5. Synthesis of allenylphosphonothio- and allenylphosphonoselenoates.[a]
Entry Propagylic sub-
strate
P-nucleophile
Product
Reaction time
(isolated yield)
5 h (insepara-
ble mixture)
1
2
7 h (insepara-
ble mixture)
24 h (insepara-
ble mixture)
3
4
5
30 min (59%)
1.5 h (64%)
1 h (insepara-
ble mixture)
6
2 h [(RP)-52
74%; (SP)-52
71%]
7
1750
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
Table 5. (Continued)
Entry Propagylic sub-
strate
P-nucleophile
Product
Reaction time
(isolated yield)
10 min [(RP)-53
57%; (SP)-53
67%]
8
[a]
Reaction conditions: 1.38 mmol propargylic substrate, 1.25 mmol P-nucleophile, 0.019 mmol (1.5 mol%) Pd2ACTHUNTRGNE(UGN dba)3·CHCl3,
0.038 mmol (3 mol%) DPEPhos, for propargylic chlorides additionally 1.5 mmol Et3N, 5 mL THF (0.25M), 688C; abbre-
viations: Thy=thymin-1-yl, DMT=4,4’-dimethoxytrityl.
thioate 42, with opposite configuration at the phos- H-phosphonothioate diesters, only allenylphosphono-
phorus center, to the reaction with a single enantio- selenoates 53 were formed formation.[45]
mer of propargylic carbonate 4b (Table 5, entry 7). To
our delight, the synthesis of allenylphosphonothioates
(R,RP)-52 and (R,SP)-52 was uneventful and complete- Catalytic Cycle for the Pd(0)-Promoted
ly stereospecific at the phosphorus center, although Allenylphosphonates Formation
due to a longer reaction time (2 h), a slight epimeriza-
tion of the allene moiety occurred (see above).
The presented above results suggest that the palladi-
Finally, separate diastereomers of a dinucleoside H- um-catalyzed propargylic substitution involving H-
phosphonoselenoate 43 were used as phosphorus nu- phosphonate diesters and related compounds follows
cleophiles in the reaction with propargyl chloride a mechanistic pathway similar to that of other nucleo-
(Table 5, entry 8). In this instance, a rapid, stereospe- philes. Details of this mechanism are depicted in
cific reaction (10 min) took place, but in contrast to Scheme 7.
Scheme 7. A proposed mechanism for the palladium-catalyzed propargylic substitution with P-nucleophiles.
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1751

Marcin Kalek and Jacek Stawinski
FULL PAPERS
The catalytic cycle starts with an oxidative addition pecially, the experiments involving anionic additives
of a propargylic substrate to Pd(0) complex in an anti were very helpful in that respect – Table 2, entries 8
fashion, and this step is responsible for the overall re- and 9). In particular, the accelerating effect of the
action stereochemistry. The rate of the oxidative addi- acetate additives (Table 2, entry 9) is in agreement
tion depended on the leaving group X, as well as on with our previous findings that the palladium(II) com-
the presence of substituents (R2, R3) at C-1. When plexes bearing an acetate ligand undergo the transme-
the oxidative addition was slow, it became apparently tallation much faster than those bearing chloride.[26-
Notably, palladium(II) complexes with an MeOÀ
f,27a,e]
a turnover-limiting step of the reaction, for example,
for propargylic acetates (Table 2, entries 6 and 7) and ligand displayed the highest reactivity, which made
primary propargylic carbonates (Table 2, entry 3 –1b; the propargylic carbonates the substrates of choice
Table 3, entries 2, 9, 11). In all other cases, the overall for this reaction. The increase in the reaction time,
reaction rates seemed to be determined by efficiency when terminally substituted propargylic derivatives
of the next, and the most mechanistically interesting (Scheme 7, R3) and/or P-nucleophiles bearing bulky
step, i.e., the transmetallation (ligand exchange).
groups (R4, R5) were used as the substrates, is also
It has been shown before by means of kinetic[27e] fully understandable in this context.
and mass spectrometry[46] studies, that formation of
The other aspect possibly connected to the trans-
Pd(II)-phosphonate complexes from H-phosphonates metallation (ligand exchange) step is the formation of
(transmetallation) is a two-step process. First, an H- propargylphosphonothioates as side-products during
phosphonate is, most likely, coordinated to the palla- the coupling of propargylic derivatives with H-phos-
dium via the phosphoryl oxygen, which increases the phonothioates (Table 5, entries 1–3). It has been
acidity of the phosphorus-bound proton and facilitates shown on many occasions that the h1-allenic and h1-
its abstraction by a base. The incipient phosphonate propargylic palladium complexes in solution may
anion is then immediately intercepted by the palladi- exist in an equilibrium.[24a,38b,47] Ma et al. reported de-
um, leading to the exclusive formation of a palladi- tailed studies demonstrating that these species can
A
display different preferences towards transmetallation
We believe that such a course of the transmetalla- with various nucleophiles and that steric hindrance is
tion is a key to the observed selective formation of al- a dominating factor governing the regioselectivity.[28]
lenylphosphonates, in favour of other possible prod- In line with these, it is probable that for H-phospho-
ucts (e.g., paths a and c vs. path b in Scheme 2, re- nothioates the transmetallation is feasible for both of
spectively). The somewhat unexpected behaviour of these isomeric complexes, although introduction of
H-phosphonates as ꢁhardꢂ nucleophiles can be ex- substituents (R2, R3) at C-1, efficiently prevented the
plained by the fact that only negligible amounts of a transmetallation of a propargylic complex, and ulti-
free phosphonate anion are present in the reaction mately formation of the propargylphosphonothioate
mixture, therefore no products due to its external derivative. The reason why H-phosphonoselenoates in
attack on the allenylpalladium species are formed. the reaction with propargylic compounds afforded ex-
The only exception from this trend was the case of di- clusively the corresponding allenylphosphonosele-
phenylphosphine oxide, that apparently, due to lack noates, remains unclear. However, a significantly
À
of, or attenuated, back-donation from the C P bonds, higher reactivity of these compounds in the investigat-
exhibited a higher acidity, enabling formation of ed reaction than that of H-phosphonate and H-phos-
larger amounts of the corresponding anion, that at- phonothioate di
ACHTUNGTRENNeUNG sters, indicates the kinetic impor-
tacked the central carbon atom Pd-complex tance of other factors during the transmetallation
a
(Scheme 2, path b) and led eventually to the bisphos- step.
phine oxide formation (Table 4, entries 1 and 2).
As to reductive elimination, we cannot exclude this
Our previous studies on the palladium-catalyzed as a possible turnover-limiting step of the catalytic
cross-coupling reactions involving H-phosphonate nu- cycle on account of the fact that this Pd-catalyzed
cleophiles revealed that transmetallation (ligand ex- propargylic substitution worked only with bidentate,
change) often determines the overall rate of the reac- large bite angle phosphine ligands, that are known to
tion. We have found that the rate of this mechanistic accelerate reductive elimination from metalphospho-
step is sensitive to a steric bulk,[26f,27e] and strongly de- nate complexes.[27c] In such a scenario, it could be the
pends on the kind of the anion present in the Pd(II) activation energy for the reductive elimination of the
complex.[27a,e] Direct transposition of these findings to product that determine predominant (or exclusive)
the propargylic substitution reaction studied herein formation of allenylphosphonates vs. propargyl-
seems to be a viable action, as it leads to a plausible phosphonates or allenylphosphonothioates vs. propar-
explanation of the experimental data. Thus, the ex- gylphosphonothioates.
pected reactivity order of the allenylphosphonate
Irrespective of these mechanistic aspects, it seems
complexes ligated by different anions shown in that throughout the reaction a stereochemical integri-
Scheme 7, is consistent with the results in Table 2 (es- ty at the phosphorus center was preserved and this re-
1752
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
purifier system. All reactions were carried out using stan-
sulted in a stereospecific outcome of the reaction
(most likely, retention of configuration). Further
mechanistic studies on this reaction are in progress.
dard Schlenk techniques. Column chromatography was pre-
formed on silica gel (Grace Davison, Davsil, 0.035–
0.070 mm). The NMR spectra were registered using a
Bruker Avance II 400 MHz instrument. The chemical shifts
are reported in ppm, relative to solvent peaks (1H, 13C) or
2% H3PO4 solution in D2O (31P NMR). Assignment of the
NMR signals was done on the basis of 2D correlation ex-
periments (COSY, HSQC).
Conclusions
We have developed a novel, general method for effi-
cient preparation of allenylphosphonates and related
compounds via a Pd(0)-catalyzed SN2’ reaction of
propargylic derivatives with H-phosphonate, H-phos-
phonothioate, and H-phosphonoselenoate diesters or
their analogues. Several ligands have been evaluated
General Procedure for the Preparation of
Allenylphosphonates and Allenylphosphinates
Pd2ACHTNUTRGNE(NUG dba)3·CHCl3 (19.7 mg, 0.019 mmol, 3.0 mol% Pd) and
DPEPhos (20.5 mg, 0.038 mmol) were placed in the reaction
vessel. The vessel was sealed and filled with N2, by applying
2 cycles of vacuum, followed by N2. THF was introduced via
the septum (5 mL), followed by triethylamine (200 mL,
152 mg, 1.50 mmol; only for propargylic chlorides), P-nucle-
ophile (1.25 mmol), and propargylic substrate (1.38 mmol).
After heating at 688C for the time indicated in Table 1, the
solvent was evaporated and product purified by silica gel
chromatography (depending on the case, pentane:AcOEt
mixture from 95:5 to 0:1 was used as the eluent).
À
for their ability to catalyze this C P bond formation,
and the most efficient catalytic system was found to
consist of Pd2ACHTUNGTRENNUNG(dba)3·CHCl3 as a palladium source and
bis(2-diphenylphosphino)diphenyl ether (DPEPhos)
as a supporting ligand. Some mechanistic aspects of
the reaction were investigated and this permitted us
to optimize the reaction conditions and suppress side-
product formation. The transformation into allenyl-
phosphonates was stereospecific at the phosphorus
center (most likely retention of configuration) and oc-
curred with a complete center to axial chirality trans-
fer from the propargylic to the allene moiety. Since
both types of substrates used for the reaction, i.e.,
propargylic compounds and H-phosphonate deriva-
tives and their analogues are easily available, complex
organic structures can be generated. This new proto-
col may expand the range of biologically important
C-phosphonate analogues with predefined stereo-
chemistry at the phosphorus center and diverse substi-
tution patterns in the allene moiety, that can be pre-
pared under mild conditions and in high efficiency.
In the case of dinucleoside allenylphosphonates (36, 50,
and 53), due to high molecular weight, lower concentration
of the starting materials was used (0.10M H-phosphonate
diester, instead of 0.25M); i.e., Pd2ACHTNUGTRNEG(UN dba)3·CHCl3 (7.9 mg,
0.0076 mmol), DPEPhos (8.2 mg, 0.015 mmol), triethylamine
(83 mL, 61 mg, 0.60 mmol; if necessary), H-phosphonate
(0.50 mmol), propargylic chloride/carbonate (0.55 mmol),
THF (5 mL). For details, see footnotes in the Table 3,
Table 4 and Table 5. Characterization of the synthesized
compounds is provided in the Supporting Information.
Supporting Information
Characterization data for the synthesized allenylphospho-
nate and allenylphosphinate derivatives (Table 3, Table 4,
1
Table 5) and the H- , 13C-, and 31P NMR spectra, synthesis
of some propargylic precursors, and data on the stereochem-
ical stability of allenylphosphonate 5 are available in the
Supporting Information.
Experimental Section
General
All reagents and solvents were of analytical grade, obtained
from commercial suppliers and used without further purifi-
cation. Propargyl carbonate (1b),[48] oct-1-yn-3-yl 4-methyl-
benzenesulfonate (4c),[49] oct-1-yn-3-yl acetate (4d),[50] 1-
chlorobut-2-yne (9a),[51] but-2-yn-1-yl methyl carbonate
(9b),[48] (3-chloroprop-1-yn-1-yl)benzene (10a),[52] methyl (3-
phenylprop-2-yn-1-yl) carbonate (10b),[53] 3-chloro-3-methyl-
but-1-yne (11a),[54] methyl (2-methylbut-3-yn-2-yl) carbonate
Acknowledgements
Financial support from the Swedish Research Council is
gratefully acknowledged.
(11b),[55] 1-chloro-1-ethynylcyclohexane (12),[56] 4-chloro-4- References
methylpent-2-yne (15a),[57] ethyl methylphosphinate (18),[58]
ethyl phenylphosphinate (19),[59] ethyl (3-methylbut-2-en-1-
yl)phosphinate (20),[60] 5’-O-(4,4’-dimethoxytrityl)thymidin-
3’-yl 3’-O-tert-butylsilylthymidin-5’-yl H-phosphonate (35),[61]
diethyl H-phosphonothioate (41),[62] 5’-O-(4,4’-dimethoxytri-
tyl)thymidin-3’-yl 3’-O-(4,4’-dimethoxytrityl)thymidin-5’-yl
H-phosphonothioate (42),[63] and 5’-O-(4,4’-dimethoxytri-
tyl)thymidin-3’-yl 3’-O-(4,4’-dimethoxytrityl)thymidin-5’-yl
H-phosphonoselenoate (43)[64] were prepared according to
published procedures. THF was dried using a VAC solvent
[1] Modern Allene Chemistry, (Eds.: N. Krause, A. S. K.
Hashmi), Wiley-VCH, Weinheim, 2004.
[2] A. Hoffman-Rçder, N. Krause, Angew. Chem. 2004,
116, 1216–1236; Angew. Chem. Int. Ed. 2004, 43, 1196–
1216.
[3] a) S. Ma, Chem. Rev. 2005, 105, 2829–2871; b) L. K.
Sydnes, Chem. Rev. 2003, 103, 1133–1150.
[4] a) A. S. K. Hashmi, Angew. Chem. 2000, 112, 3737–
3740; Angew. Chem. Int. Ed. 2000, 39, 3590–3593;
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1753

Marcin Kalek and Jacek Stawinski
FULL PAPERS
b) R. Zimmer, C. U. Dinesh, E. Nandanan, F. A. Khan,
Chem. Rev. 2000, 100, 3067–3125; c) J.-M. Weibel, A.
Blanc, P. Pale, Chem. Rev. 2008, 108, 3149–3173; d) Z.
Li, C. Brouwer, C. He, Chem. Rev. 2008, 108, 3239–
3265.
4500–4510; d) T. J. J. Mꢆller, M. Ansorge, Tetrahedron
1998, 54, 1457–1470; e) N. N. Kumar, M. N. Reddy,
K. C. Swamy, J. Org. Chem. 2009, 74, 5395–5404.
[20] a) J. P. Krise, V. J. Stella, Adv. Drug Delivery Rev. 1996,
19, 287–310; b) C. Schultz, Bioorg. Med. Chem. 2003,
11, 885–898; c) P. Guga, Curr. Top. Med. Chem. 2007,
7, 695–713; d) A. V. Nikolaev, I. V. Botvinko, A. J.
Ross, Carbohydr. Res. 2007, 342, 297–344; e) M. W.
Bowler, M. J. Cliff, J. P. Walthob, G. M. Blackburn,
New J. Chem. 2010, 34, 784–794.
[5] a) J. Piera, K. Nꢃrhi, J.-E. Bꢃckvall, Angew. Chem.
2006, 118, 7068–7071; Angew. Chem. Int. Ed. 2006, 45,
6914–6917; b) J. Piera, A. Persson, X. Caldentey, J.-E.
Bꢃckvall, J. Am. Chem. Soc. 2007, 129, 14120–14121;
c) F. Inagaki, K. Sugikubo, Y. Miyashita, C. Mukai,
Angew. Chem. 2010, 122, 1–1; Angew. Chem. Int. Ed.
2010, 49, 1–6.
[21] H. R. Allcock, P. J. Harris, R. A. Nissan, J. Am. Chem.
Soc. 1981, 103, 2256–2261.
[6] a) A. Hoffman-Rçder, N. Krause, Angew. Chem. 2002,
114, 3057–3059; Angew. Chem. Int. Ed. 2002, 41, 2933–
2935; b) N. Bongers, N. Krause, Angew. Chem. 2008,
120, 2208–2211; Angew. Chem. Int. Ed. 2008, 47, 2178–
2181.
[7] a) K. C. K. Swamy, E. Balaraman, N. S. Kumar, Tetrahe-
dron 2006, 62, 10152–10161; b) N. G. Khusainova,
O. A. Mostovaya, E. A. Berdnikov, I. A. Litvinov, D. B.
Krivolapov, R. A. Cherkasov, Russ. J. Org. Chem. 2005,
41, 1260–1264; c) J. M. Santos, Y. Lꢄpez, D. Aparicio,
F. Palacios, J. Org. Chem. 2008, 73, 550–557.
[22] a) A. P. Boisselle, N. A. Meinhardt, J. Org. Chem. 1962,
27, 1828–1833; b) V. Mark, Tetrahedron Lett. 1962,
281–285; c) A. N. Pudovik, I. M. Aladzheva, J. Gen.
Chem. USSR (Engl. Transl.) 1963, 33, 700–701.
[23] M. Muller, A. Mann, M. Taddei, Tetrahedron Lett.
1993, 34, 3289–3290.
[24] a) S. Ma, Eur. J. Org. Chem. 2004, 1175–1183; b) J.
Tsuji, T. Mandai, Angew. Chem. 1995, 107, 2830–2854;
Angew. Chem. Int. Ed. Engl. 1995, 34, 2589–2612.
[25] a) D. Prim, J.-M. Campagne, D. Joseph, B. Andrioletti,
Tetrahedron 2002, 58, 2041–2075; b) A. L. Schwan,
Chem. Soc. Rev. 2004, 33, 218–224; c) F. M. J. Tappe,
V. T. Trepohl, M. Oestreich, Synthesis 2010, 3037–3062;
d) D. S. Glueck, Top. Organomet. Chem. 2010, 31, 65–
100; e) J. L. Montchamp, J. Organomet. Chem. 2005,
690, 2388–2406.
[8] a) C. Mukai, M. Ohta, H. Yamashita, S. Kitagaki, J.
Org. Chem. 2004, 69, 6867–6873; b) V. K. Brel, V. K.
Belsky, A. I. Stash, V. E. Zavodnik, P. J. Stang, Eur. J.
Org. Chem. 2005, 512–521.
[9] M. Chakravarty, K. C. K. Swamy, Synthesis 2007, 3171–
3178.
[26] a) S. Abbas, C. J. Hayes, Synlett 1999, 7, 1124–1126;
b) S. Abbas, R. D. Bertram, C. J. Hayes, Org. Lett.
2001, 3, 3365–3367; c) T. Johansson, J. Stawinski,
Chem. Commun. 2001, 2564–2565; d) G. Lavꢅn, J. Sta-
winski, Collection Symp. Series 2005, 7, 195–199; e) M.
Kalek, J. Stawinski, Collection Symp. Series 2008, 10,
214–218; f) M. Kalek, M. Jezowska, J. Stawinski, Adv.
Synth. Catal. 2009, 351, 3207–3216.
[27] a) M. Kalek, J. Stawinski, Organometallics 2007, 26,
5840–5848; b) A. M. Levine, R. A. Stockland, R.
Clark, I. Guzei, Organometallics 2002, 21, 3278–3284;
c) R. A. Stockland, A. M. Levine, M. T. Giovine, I. A.
Guzei, J. C. Cannistra, Organometallics 2004, 23, 647–
656; d) M. C. Kohler, R. A. Stockland Jr, N. P. Rath,
Organometallics 2006, 25, 5746–5756; e) M. Kalek J.
Stawinski, Organometallics 2008, 27, 5876–5888;
f) M. C. Kohler, T. V. Grimes, X. Wang, T. R. Cundari,
R. A. Stockland, Organometallics 2009, 28, 1193–1201.
[28] S. Ma, A. Zhang, J. Org. Chem. 2002, 67, 2287–2294.
[29] C.-C. Su, J.-T. Chen, G.-H. Lee, Y. Wang, J. Am. Chem.
Soc. 1994, 116, 4999–5000.
[10] V. K. Brel, P. J. Stang, Eur. J. Org. Chem. 2003, 224–
229.
[11] H. Guo, Z. Zheng, F. Yu, S. Ma, A. Holuigue, D. S.
Tromp, C. J. Elsevier, Y. Yu, Angew. Chem. 2006, 118,
5119–5122; Angew. Chem. Int. Ed. 2006, 45, 4997–
5000.
[12] Y.-Q. Mei, J.-T. Liu, Z.-J. Liu, Synthesis 2007, 739–743.
[13] a) Y. Gu, T. Hama, G. B. Hammond, Chem. Commun.
2000, 395–396; b) A. J. Zapata, Y. Gu, G. B. Ham-
mond, J. Org. Chem. 2000, 65, 227–234.
[14] L. S. Trifonov, S. D. Simova, A. S. Crahovats, Tetrahe-
dron Lett. 1987, 28, 3391–3392.
[15] a) S. Ma, H. Guo, F. Yu, J. Org. Chem. 2006, 71, 6634–
6636; b) M. Chakravarty, K. C. K. Swamy, J. Org.
Chem. 2006, 71, 9128–9138.
[16] a) R. S. Macomber, T. C. Hemling, J. Am. Chem. Soc.
1986, 108, 343–344; b) R. S. Macomber, T. C. Hemling,
Isr. J. Chem. 1985, 26, 136–139.
[17] a) J. Yuan, X. Ruan, Y. Yang, X. Huang, Synlett 2007,
2871–2874; b) D. D. Enchev, Heteroat. Chem. 2005, 16,
156–158; c) V. C. Christov, B. Prodanov, Phosphorus
Sulfur Silicon Relat. Elem. 2002, 177, 243–249; d) L. S.
Trifonov, A. S. Orahovats, Heterocycles 1985, 23, 1723–
1728; e) D. D. Enchev, Phosphorus Sulfur Silicon Relat.
Elem. 2000, 165, 273–284; f) F. Yu, X. Lian, J. Zhao, Y.
Yu, S. Ma, J. Org. Chem. 2009, 74, 1130–1134.
[18] a) S. E. Denmark, J. E. Marlin, J. Org. Chem. 1991, 56,
1003–1013; b) F. Yu, X. Lian, S. Ma, Org. Lett. 2007, 9,
1703–1706.
[30] a) Y. Kozawa, M. Mori, Tetrahedron Lett. 2001, 42,
4869–4873; b) Y. Kozawa, M. Mori, Tetrahedron Lett.
2002, 43, 1499–1502; c) Y. Kozawa, M. Mori, J. Org.
Chem. 2003, 68, 8068–8074.
[31] M. Kalek, T. Johansson, M. Jezowska, J. Stawinski,
Org. Lett. 2010, 12, 4702–4704.
[32] B. Iorga, F. Eymery, D. Carmichael, P. Savignac, Eur. J.
Org. Chem. 2000, 3103–3115.
[33] In separate experiments when compounds 3 and 5 were
treated with acetic acid in THF, at 688C, similar de-
composition products were formed.
[34] M. Leclꢇre, A. G. Fallis, Angew. Chem. 2008, 120, 578–
582; Angew. Chem. Int. Ed. 2008, 47, 568–572.
[19] a) V. K. Brel, Synthesis 2002, 1829–1832; b) A. Panossi-
an, N. Fleury-Brꢅgeot, A. Marinetti, Eur. J. Org. Chem.
2008, 3826–3833; c) M. Chakravarty, N. N. B. Kumar,
K. V. Sajna, K. C. K. Swamy, Eur. J. Org. Chem. 2008,
1754
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1741 – 1755

Novel, Stereoselective and Stereospecific Synthesis of Allenylphosphonates and Related Compounds
[35] a) T. Ohtani, H. Nakatsukasa, M. Kamezawa, H. Tachi-
bana, Y. Naoshima, J. Mol. Catal. B: Enzym. 1998, 4,
53–60; b) R. Kourist, P. Dmoꢈnguez de Marꢈa, U. T.
Bornscheuer, ChemBioChem 2008, 9, 491–498.
SN2’ substitution. This possibility, however, was ruled
out since in a control reaction without the catalyst no
product formation was observed even after 2 h heating
at 608C.
[36] D. E. Frantz, R. Fꢃssler, E. M. Carreira, J. Am. Chem.
[46] M. Andaloussi, J. Lindh, J. Sꢃvmarkar, P. J. R. Sjçberg,
M. Larhed, Chem. Eur. J. 2009, 15, 13069–13074.
[47] J. M. A. Wouters, R. A. Klein, C. J. Elsevier, L.
Hꢃming, C. H. Stam, Organometallics 1994, 13, 4586–
4593.
Soc. 2000, 122, 1806–1807.
[37] a) M. Node, K. Nishide, T. Fujiwara, S. Ichihashi,
Chem. Commun. 1998, 2363–2364; b) J. A. Marshall, J.
Liao, J. Org. Chem. 1998, 63, 5962–5970.
[38] a) C. J. Elsevier, P. M. Stehouwer, H. Westmijze, P. Ver-
meer, J. Org. Chem. 1983, 48, 1103–1105; b) C. J.
Elsevier, H. Kleijn, J. Boersma, P. Vermeer, Organome-
tallics 1986, 5, 716–720; c) T. Konno, M. Tanikawa, T.
Ishihara, H. Yamanaka, Chem. Lett. 2000, 1360–1361.
[39] a) G. Lowe, J. Chem. Soc. Chem. Commun. 1965, 411;
b) J. H. Brewster, Top. Stereochem. 1967, 2, 33–39.
[40] G. Lavꢅn, J. Stawinski, Synlett 2009, 225–228.
[41] Y. Xu, J. Zhang, J. Chem. Soc. Chem. Commun. 1986,
1606–1606.
[42] a) J. Stawinski, R. Strçmberg, Deoxyribo- and Ribonu-
cleoside H-Phosphonates, in: Current Protocols in Nu-
cleic Acid Chemistry, (Eds.: S.L Beaucage et al.), John
Wiley & Sons, New York, Chapter 2.6.1–2.6.15, 2001;
b) J. Stawinski, R. Strçmberg, Di- and oligonucleotide
Synthesis Using H-Phosphonate Chemistry, in: Oligonu-
cleotide Synthesis: Methods and Applications, (Ed.: P
Herdewijn), Humana Press, Totowa, NJ, Vol. 288,
pp 81–100, 2004; c) J. Stawinski, M. Thelin, J. Org.
Chem. 1991, 56, 5169–5175.
[48] I. Minami, M. Yuhara, H. Watanabe, J. Tsuji, J. Orga-
nomet. Chem. 1987, 334, 225–242.
[49] C. Jonassonm, A. Horvath, J.-E. Bꢃckvall, J. Am.
Chem. Soc. 2000, 122, 9600–9609.
[50] R. J. Detz, M. M. E. Delville, H. Hiemstra, J. H. van
Maarseveen, Angew. Chem. 2008, 120, 3837–3840;
Angew. Chem. Int. Ed. 2008, 47, 3777–3780.
[51] M. G. Ettlinger, J. E. Hodgkins, J. Am. Chem. Soc.
1955, 77, 1831–1836.
[52] A. R. Pereira, J. A. Cabezas, J. Org. Chem. 2005, 70,
2594–2597.
[53] S. Ma, A. Zhang, J. Org. Chem. 2002, 67, 2287–2294.
[54] B. Stulgies, P. Prinz, J. Magull, K. Rauch, K. Meindl, S.
Rꢆhl, A. de Meijere, Chem. Eur. J. 2005, 11, 308–320.
[55] H. Ito, Y. Sasaki, M. Sawamura, J. Am. Chem. Soc.
2008, 130, 15774–15775.
[56] H. Hopf, D. Gottschild, W. Lenk, Isr. J. Chem. 1985, 26,
79–87.
[57] H. Mayr, I. K. Halberstadt-Kausch, Chem. Ber. 1982,
115, 3479–3515.
[43] a) H. Kaur, R. B. Babu, S. Maiti, Chem. Rev. 2007, 107,
4672–4697; b) F. Eckstein, Annu. Rev. Biochem. 1984,
##53##54, 331–366; c) N. K. Sahu, G. Shilakari, A.
Nayak, D. V. Kohli, Curr. Pharm. Biotechnol. 2007, 8,
291–304; d) N. Carrasco, J. Caton-Williams, G. Brandt,
S. Wang, Z. Huang, Angew. Chem. 2006, 118, 100–103;
Angew. Chem. Int. Ed. 2006, 45, 94–97; e) J. Kowalska,
M. Lukaszewicz, J. Zuberek, E. Darzynkiewicz, J. Je-
mielity, ChemBioChem 2009, 10, 2469–2473.
[58] L. Maier, Helv. Chim. Acta 1963, 46, 2667–2676.
[59] Y. R. Dumond, R. L. Baker, J. L. Montchamp, Org.
Lett. 2000, 2, 3341–3344.
[60] K. Bravo-Altamirano, J.-L. Montchamp, Org. Lett.
2006, 8, 4169–4171.
[61] K. Zmudzka, T. Johansson, M. Wojcik, M. Janicka, M.
Nowak, J. Stawinski, B. Nawrot, New J. Chem. 2003, 27,
1698–1705.
[62] W. W. Brand, J. M. Gullo, M. C. Carr, Phosphorus,
Sulfur Relat. Elem. 1981, 2, 183–184.
[44] An attempted separation of compounds 44 vs. 45, 46 vs.
47, and 50 vs. 51 failed and these products could not be
isolated in a satisfactory pure form.
[63] R. Zain, J. Stawinski, J. Org. Chem. 1996, 61, 6617–
6622.
[45] A very fast reaction involving H-phosphonoselenoate
43 might have suggested an uncatalyzed propargylic
[64] M. Bollmark, M. Kullberg, J. Stawinski, Tetrahedron
Lett. 2002, 43, 515–518.
Adv. Synth. Catal. 2011, 353, 1741 – 1755
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1755