A. Chiminazzo, L. Sperni, F. Fabris et al.
Tetrahedron Letters 70 (2021) 153012
Scheme 3. A) Unsuitable direct synthesis of 4 from DMHP, sodium methylate and
dichloromethane; B) plausible mechanism for the formation of dimethoxy methyl
phosphonate.
Under basic conditions [30] and in the absence of methylating
reagents the formation of dimethoxy methyl phosphonate is unex-
pected, but can be justified by considering the high nucleophilicity
of the dimethylphosphite anion and its attack on the methyl resi-
due of a second molecule of DMHP. The same side reaction cannot
occur with DEHP due to the higher steric hindrance on the methy-
lene units connected to the phosphonate moiety and the lower
nucleophilicity of the corresponding anion.
Fig. 1. 1H NMR spectra from bottom to top: ethyl ester derivative 3a in CDCl3, acid
derivative 9a in D2O and methyl ester derivative 6a in CDCl3.
The 1H NMR spectra of 3a, 9a and 6a are reported for compar-
Since it was not possible to prepare 4 in a single step starting
from DMHP, we decided to investigate the synthesis of this build-
ing block in three consecutive steps: i) synthesis of 1 using the pro-
cedure reported by Hormi and co-workers [25]; ii) deprotection of
the ethyl ester groups with TMSBr followed by hydrolysis to give
the corresponding bisphosphonic acid 7; iii) protection of the acid
moieties with trimethyl orthoformate (TOF) to obtain 4 (Scheme 2)
[32]. TOF is a reagent employed for the esterification of hydroxyl
groups [33] and allowed to obtain 4 from 7 in quantitative yield
without the occurrence of undesired side-reactions or partial re-
protection, even when the reaction scale was on a 1 g scale.
The same synthetic steps comprising of the deprotection of 2 (to
give 8, Scheme 2) with TMSBr and re-protection using TOF were
also applied to the synthesis of 5 in 71% overall yield from 2. Anal-
ysis of the crude reaction mixture by GC–MS showed the formation
of 5 together with smaller amounts of two by-products. These are
presumably due to the reaction with TMSBr, an extremely aggres-
sive reagent, which tends to react not only with the phosphonic
ester groups, but also with the vinyl moiety through a not com-
pletely understood mechanism.
ison in Fig. 1, showing the presence of the aromatic unit and the
diagnostic vinylic proton as a doublet of doublets due to the JP-H
3
couplings with the P atoms in cis and trans positions.
The intermediate acids 7, 8 and 9a-c turned out to be quite reac-
tive and had to be directly reacted with TOF to obtain the corre-
sponding methyl esters 4, 5 and 6a-c, or stored at 4 °C for a few
days to prevent decomposition.
Conclusion
An alternative approach is reported for the synthesis of methyl
ester protected BPs building blocks such as methylene bisphospho-
nate 4, vinylidenebisphosphonate 5 and prochiral vinylidenebis-
phosphonates 6a-c that cannot be obtained directly from DMHP
and dichloromethane. These BP precursors, characterized by
reduced steric hindrance with respect to the most common ana-
logues 1, 2 and 3a-c, will favorably spur the development of stere-
oselective reactions of BPs enabling higher yields.
As an alternative method for the synthesis of 5, we adapted the
procedure described by Degenhardt and co-workers [27] to obtain
2 from 1, starting with the reaction of 4 with formaldehyde and
subsequent water elimination under acid catalysis (Scheme 2).
The reaction was characterized by a conversion of up to 97%; how-
ever the isolation and purification of 5 was rather complex.
Replacement of the ethyl ester groups in 1 with the methyl ester
groups in 4 led to a drastic increase in the hydrophilicity of 5. As
a consequence, during the removal of p-toluenesulfonic by aque-
ous extraction of the reaction mixture, most of 5 moved to the
aqueous phase. Multiple extractions of the aqueous phase with
ethyl acetate were required to recover the desired product 5, but
at the expense of only 45% yield.
Similarly, it was decided to modify the procedure reported by
Lehnert and co-workers [34] for the synthesis of monosubstituted
aromatic products 6a-c. The reaction of 4 with benzaldehyde cat-
alyzed by TiCl4 was initially investigated. Despite high yields in
the condensation reaction using 1, a similar procedure performed
on 4 did not give the desired product 6a. 1H and 31P NMR analysis
of the crude reaction mixture demonstrated the total absence of
reactivity of the reagents.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgements
The authors acknowledge Università Ca’ Foscari Venezia and
MUR for support. AS is grateful to the DSMN colleagues who partic-
ipated to the funding of the 400 MHz NMR.
Appendix A. Supplementary data
Supplementary material containing full experimental condi-
tions and NMR spectra for novel compounds 4, 5 and 6a-c is avail-
able. Supplementary data to this article can be found online at
References
We therefore used the deprotection-reprotection method previ-
ously developed for the synthesis of 4 and 5 for the preparation of
6a-c observing that the double bond remained unaltered during
the reactions. Starting from 3a-c, the deprotection step with TMSBr
and re-protection with TOF gave 6a-c in good yields (Scheme 2).
3