Due to the lability of the P-N bond to hydrolysis in cyclic
species 10,16 we reasoned that we could employ the
phosphorous acid diamide moiety as a P(III)-temporary tether
in a one-pot RCM/hydrolysis procedure (Scheme 2). Opti-
mization of the previously reported conditions6b provides
acyclic RCM precursors 8 in 80-90% yield. Subsequent
RCM utilizing the second generation Grubbs catalyst 915c,17
in refluxing benzene, followed by facile cleavage18 of the
P-tether with methanolic HCl, results in quantitative yields
of C2-symmetric 1,4-diamine 1 with complete stereochemical
and geometrical integrity. Furthermore, the RCM reaction
is complete within several minutes, reaction scale is a
nonissue, and the RCM/hydrolysis sequence is a single-pot
event.
Scheme 3a
a Reagents and conditions: (a) RP(O)Cl2 (12a,R ) OPh; 12b,
R ) Ph), Et3N, DMAP, CH2Cl2, reflux, R ) OPh, >95%, R ) Ph,
84%; (b) i. 9, benzene, reflux, ii. HCl/H2O/THF, 50 °C, R ) OPh,
91%, R ) Ph, 70%.
A number of other temporary tethers were also investi-
gated,19 including various metals9a,c (Cu, Fe, Mn, Mg, and
Ni), as well as carbon (CO) and boron9b (BPh). Thus far,
none have allowed this facile “di-amine” binding/metathesis
sequence to occur. Our group previously reported an RCM
strategy to generate cyclic sulfamides analogous to 10;20
however, the inability to effectively cleave the sulfamide
linkage (R2NSO2NR2) under mild conditions limits their
utility in the production of 1,4-diamines such as 1-3.
Moreover, while temporary silicon tethers11 have been
employed in the RCM reaction to access 1,4-diols, all of
our attempts to prepare 1 from 7 utilizing silicon tethers
(SiPh2, SiMe2, and SiCl2) have been unsuccessful. We have
found that not only does phosphorus appear to be the sole
nucleus in which this 1,4-diamine chemistry is successful
but the efficiency and ease of the sequence is extraordinary.
With this temporary bridging strategy in hand, we turned
our attention to the synthesis of C2-symmetric 1,4-diamine
2, containing branching at the allylic positions (Scheme 3).
Previously, we found that less sterically encumbering
R-branched primary amines, such as L-valine-derived 11,21
readily couple twice with P(V)-dichloride 12a to give 13a
in high yield.6b In addition, we and others22 have shown that
the reaction between phosphorus oxychloride (POCl3) and
3 equiv of an R-branched primary amine, such as 11, is facile
to afford the corresponding phosphoramide.23 Therefore, it
was crucial in the synthesis of diamine 2 to use RP(O)Cl2
(R * Cl), where R serves as an ancillary blocking group to
prevent the formation of the triply coupled product. Subse-
quent RCM using catalyst 9, followed by in situ hydrolysis
of the P(V)-tether under slightly more forcing conditions (50
°C), generates 1,4-diamine 2.
To extend the scope of utilizing temporary P-tethers, we
directed our efforts toward the synthesis of unsymmetric,
differentially substituted 1,4-diamines such as 3 (Scheme 4).
Scheme 4a
(15) For the first generation Grubbs catalyst, see: (a) Schwab, P.; Grubbs,
R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118, 100-110. (b) Schwab,
P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew. Chem., Int. Ed. Engl.
1995, 34, 2039-2041. For the second generation Grubbs catalyst, see: (c)
Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953-
956.
a Reagents and conditions: (a) 7a, Et3N, DMAP, CH2Cl2, reflux,
>95%, ds ) 1.1:1.0;24 (b) 11, Et3N, DMAP, CH2Cl2, 0 °C, 88%,
ds ) 6.6-13.2:1.0; (c) i. 17, CH2Cl2, reflux, ii. methanolic HCl,
50 °C, 97%.
(16) 1,3,2-Diazaphosphepine 2-oxide 10 hydrolyzed after prolonged
storage at 0 °C (2-3 weeks).
(17) RCM with the traditional Grubbs benzylidene catalyst15a,b occurs
in excellent yields with most substrates if the reaction was performed on
small scale (<500 mg). Reaction times varied from 1 to 24 h.
(18) No transesterification was observed during the tether cleavage
procedure when benzyl esters were employed, as in 1c and 1d.
(19) Details of the unsuccessful attempts with other tethers are provided
in the Supporting Information.
(20) Dougherty, J. M.; Probst, D. A.; Robinson, R. E.; Moore, J. D.;
Klein, T. A.; Snelgrove, K. A.; Hanson, P. R. Tetrahedron 2000, 56, 9781-
9790 and references therein.
Prior work in our laboratory revealed that only 1 equiv of
an N-allylated, R-branched amino ester, such as 7a, couples
with P(V)-dichlorides, such as methylphosphonic dichloride
(14), to give an ∼1.1:1.0 diastereomeric mixture of phos-
phonamidic monochloridates 15.24 We reasoned that this
monochloridate, 15, would serve as an ideal intermediate in
the production of the differentially substituted 1,4-diamine
3. Therefore, addition of primary amine 11 to the diastere-
(21) (a) Fehrentz, J.-A.; Castro, B. Synthesis 1983, 676-678. (b) Saari,
W. S.; Fisher, T. E. Synthesis 1990, 453-454.
(22) (a) Unpublished results from our laboratory. Our findings are in
agreement with Wills and co-workers who have reported that 3 equiv of
(R)-R-methyl benzylamine couple readily with POCl3 to provide the
corresponding phosphoramide, see: (b) Burns, B.; Studley, J. R.; Wills,
M. Tetrahedron Lett. 1993, 34, 7105-7106.
(23) This is in sharp contrast with our report that the addition of
R-branched secondary amines such as 7 to POCl3 occurs only once to give
the phosphonamidic dichloridate, see ref 6b.
(24) Sprott, K. T.; Hanson, P. R. J. Org. Chem 2000, 65, 4721-4728.
Org. Lett., Vol. 3, No. 24, 2001
3941