.
Angewandte
Communications
DOI: 10.1002/anie.201206533
Staudinger ligation
Phosphine-Based Redox Catalysis in the Direct Traceless Staudinger
Ligation of Carboxylic Acids and Azides**
Andrew D. Kosal, Erin E. Wilson, and Brandon L. Ashfeld*
À
À
The synthesis of amide C N bonds through nucleophilic acyl
turnover without interfering with C N bond formation.
substitutions constitutes one of the most fundamental trans-
formations in chemical synthesis.[1] Recently, the Staudinger-
type ligation[2] of carboxylic acid derivatives (e.g., acid
chlorides, anhydrides, acyl selenides, and thioesters) and
Given the reactivity of silanes[6] toward the chemoselective
reduction of phosphine oxides, our initial efforts focused on
this class of hydride donors.[7] However, the aqueous con-
ditions required for amidophosphonium hydrolysis in the
conventional Staudinger ligation are incompatible with
silanes.[8] A catalytic design is further complicated by the
tendency of silanes to reduce aza-ylides to their correspond-
ing silyl amines. Competitive aza-ylide reduction would
hinder formation of an activated phosphonium carboxylate
and thus the subsequent acyl substitution.[7a] Therefore, we
speculated that oxygen transfer directly from the carboxylic
acid to phosphorus under anhydrous conditions would be
compatible with a silane reductant. However, recent reports
by van Delft and co-workers showed that while phospholane
oxides and dibenzophospholes are effective redox catalysts,
their corresponding aza-ylides are readily reduced by silane-
s.[7a,b] In contrast, Ph3P-derived aza-ylides were found to
reduce at a slower rate in the presence of Ph2SiH2, which
prompted us to begin our study toward the first phosphine-
catalyzed Staudinger ligation using this reagent combination.
By treating benzoic acid 1a and benzyl azide (2a) with PPh3
(10 mol%) and Ph2SiH2 (1 equiv), the ostensible phosphoni-
um carboxylate 4 led to benzyl amide 3a in 25% yield
[Eq. (1)]. In spite of the modest yield, we were encouraged by
the catalytic behavior of PPh3 in the Staudinger ligation.
À
azides has become a preeminent strategy for amide C N bond
construction (Scheme 1a).[3] However, the generation of
Scheme 1. Staudinger ligation approach to amide synthesis.
=
stoichiometric by-products (e.g., R3P O) often complicates
efforts to isolate products, leads to waste disposal issues, and
limits the overall synthetic efficiency of this method.[4] Thus,
a phosphine-catalyzed Staudinger ligation involving the direct
coupling of carboxylic acids and azides would concomitantly
minimize the formation of undesired by-products while
avoiding the need for an additional acid derivatization
(Scheme 1b). We envisioned a PIII/PV-redox-driven cycle
wherein an acid/base reaction of the carboxylic acid and
intermediate aza-ylide would form an activated phosphonium
À
carboxylate in situ and thus enable catalytic C N bond
formation.[5] Herein, we report a conceptually new approach
toward the phosphine-catalyzed Staudinger ligation for the
chemoselective, direct conversion of carboxylic acids to
amides.
At the outset, we sought to identify an appropriate
phosphine/reductant combination that would enable catalyst
We began our reaction optimization by examining the
efficacy of various silanes to facilitate catalyst turnover
(Table 1). Although known to efficiently reduce phosphine
oxides, alkoxy silanes (MeO)3SiH and (EtO)2MeSiH, and
Cl3SiH failed to provide amide 3a (Table 1, entries 1–3).[7d,9]
However, use of PhSiH3 improved the yield to 70%, while
a slight excess of azide 2a provided 3a in 98% yield (Table 1,
entries 4 and 5). Interestingly, employing either 0.5 or
1.5 equiv of PhSiH3 led to a reduction in the yield of 3a
(Table 1, entries 6 and 7). The observation that 4 equiv of
PhSiH3 prevented amide formation indicates that minimizing
competitive aza-ylide reduction is critical (Table 1, entry 9).[10]
We next turned our attention toward examining catalyst
structure and loading. Unfortunately, reducing the amount of
[*] A. D. Kosal,[+] E. E. Wilson,[+] Prof. Dr. B. L. Ashfeld
Department of Chemistry and Biochemistry
University of Notre Dame
251 Nieuwland Science Hall, Notre Dame, IN 46556 (USA)
E-mail: bashfeld@nd.edu
[+] These authors contributed equally to this work.
[**] We thank Prof. Xavier Creary for helpful discussions. Financial
support for this work was received from the NSF (CHE-1056242)
and the University of Notre Dame.
Supporting information for this article is available on the WWW
12036
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 12036 –12040