Angewandte
Chemie
perature.[10] Although cyclic B N compounds have been
which is clearly formed during the reaction process). To
À
isolated in some cases, the yield (< 50%) is moderate, and is
overcome this problem, we put our efforts into the develop-
ment of less reactive amine boranes with the aim of slowing
down the decomposition process of the gold catalysts. After
extensive screening of various substrates, the cyano-substi-
tuted amine borane 3 was identified and accessible through
the general synthetic route summarized in Scheme 3.
3
[11]
À
likely due to the poor stability of the C(sp ) B bond. This
limitation decreased the potential of using this method for the
synthesis of cyclic amine boranes.[10d] An amine-directed
alkyne hydroboration will result in the formation of more
2
[11]
À
stable vinyl boranes [C(sp ) B],
and could be more
efficient for cyclic amine borane synthesis. However, to the
best of our knowledge, this transformation has never been
reported. In fact, because of the reduced reactivity of the
ꢀ
alkyne, treatment of the propargyl amine borane 1 (HC
CCH2NH2-BH3) with I2 (2 equiv, Vedejs protocol) gave no
alkyne addition product after 24 hours.[12] The major chal-
lenges associated with this transformation are: 1) low reac-
tivity of the alkyne; 2) the possibility of double addition
rather than desired monohydroboration.
Although transition metal catalyzed hydroboration suf-
fers from catalyst decomposition, successful examples have
been reported.[13] To avoid the undesired catalyst decompo-
sition (caused by borane reduction), almost all of the reported
cases used metal cations which were able to form a metal
hydride (M-H) bond. Meanwhile, the M-H cis addition to the
alkyne led to the formation of the trans isomer, which is not
Scheme 3. Synthesis of the amine boranecarbonitrile 3.
Under the optimal reaction conditions, 3 was prepared in
good to excellent yields (generally > 85%; see the Supporting
Information). This method tolerated various substituents at
the alkyne terminus. Both primary and aliphatic-substituted
À
suitable for the C B bond cyclization (through M-B trans-
À
metallation, Scheme 2B). Thus, transition metal catalyzed
alkyne hydroboration for the synthesis of cyclic amine borane
has not been achieved in the past.
secondary propargyl amines worked well for the B N
formation. The aromatic-substituted propargyl amine (R2 =
Ar) could not give the corresponding complex 3 under the
optimal reaction conditions, which is likely a consequence of
the decreased basicity of the nitrogen atom. Nevertheless, this
method allowed general access to the cyano-stabilized amine
During the past decade, gold cations have been identified
as one of the most effective p acids for alkyne activation.[14]
Based on this activation mode, we hypothesized that cyclic
amine boranes could be synthesized through the intramolec-
ular alkyne hydroboration as shown in Scheme 2C, if an
appropriate gold catalyst (that tolerates borane substrates)
and amine borane precursor could be identified. Two practical
concerns are: 1) the stability of active gold catalysts under the
reductive conditions; 2) the feasibility of proposed Au–B
transmetallation.
Our group has focused on the development of 1,2,3-
triazole-coordinated metal complexes as new catalysts for
challenging chemical transformations.[15] While interesting
reactivity has been observed with the triazole/gold (TA-Au)
complexes, the main advantage of this catalyst is the
significantly improved stability.[16] Thus, we postulated that
the TA-Au catalysts would provide the required p-acid
reactivity toward alkyne activation while maintaining stability
towards the borane, and thus afford a new and efficient
synthesis of cyclic amine boranes under mild reaction
conditions.
To test our hypothesis, various gold and TA-Au catalysts,
[L-Au/TA]+XÀ (L = PPh3, IPr, XPhos; XÀ = TfOÀ, F6SbÀ),
were tested in the hydroboration of 1 (see structure of 1 in
Ref. [12]). Unfortunately, significant gold decomposition
(formation of gold nanoparticle or gold mirror) occurred,
even with the TA-Au catalysts. Corma and co-workers have
recently reported the highly efficient alkyne hydration with
small gold clusters.[17] Their work suggested the in situ formed
Au0 cluster was the active catalyst. Unfortunately, no reaction
occurred after this reduction process by 1 (presumably
because of the formation of catalytically inactive Au0 species,
À
borane containing a less reactive borohydride (B H).
With this new amine borane 3 in hand, we investigated its
reactivity toward gold catalysts. Notably, compared to the
terminal alkyne, the internal alkyne is usually much less
reactive toward p-acid catalysts. Therefore, to better validate
the feasibility of this gold catalyzed intramolecular hydro-
boration, 3b was selected for our initial screening. The results
are summarized in Table 1. Unlike 1 (which caused rapid gold
decomposition), the 3b gave the desired cyclic amine borane
4b using the PPh3 AuCl/AgOTf catalyst, albeit with very low
yield (17%; entry 2). The structure of 4b was determined by
X-ray crystallography. Notably, the reaction gave 3:1 (cis/
trans) isomer mixtures in solution. However, upon crystal-
lization, all of the trans isomer was converted into the
cis isomer. This process has been confirmed by NMR
spectroscopy: the cis isomer was dissolved and a slow equi-
librium was observed (up to 3 h), thus giving a 3:1 cis/trans
mixture. Slowly evaporation of the solvent gave the pure
cis isomer again (structure confirmed by X-ray). Overall, the
success in obtaining 4b greatly supported our hypothesis that
the gold-catalyzed (p-acid) alkyne activation is a feasible
approach for the preparation of cyclic amine boranes. To
improve the reaction performance, we screened other gold
catalysts. As shown in entries 1–4, although 3b was a weaker
reductant, the reaction still suffered from gold decomposition.
Interestingly, the TA/Au catalysts, though giving no reaction,
indicated better tolerance of 3b given the decreased rate of
gold mirror formation. To verify this result, we monitored the
catalyst decomposition (upon treatment with 3b) using
Angew. Chem. Int. Ed. 2014, 53, 5418 –5422
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5419