Ligand-Accelerated Gold-Catalyzed Addition of in Situ Generated Hydrazoic Acid to Alkynes under Neat Conditions

Article abstract of DOI:10.1021/acs.orglett.7b01359

The direct addition of in situ generated hydrazoic acid to alkynes is realized without solvent by using a gold catalyst derived from a recently designed remotely functionalized biaryl-2-ylphosphine ligand (i.e., WangPhos). With terminal alkynes, the addit

Full text of DOI:10.1021/acs.orglett.7b01359

Letter
pubs.acs.org/OrgLett
Ligand-Accelerated Gold-Catalyzed Addition of in Situ Generated
Hydrazoic Acid to Alkynes under Neat Conditions
Xiaoqing Li,†,‡ Shengrong Liao,† Zhixun Wang,† and Liming Zhang*,†
†Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106, United States
‡Department of Chemical Engineering, Zhejiang University of Technology, Hangzhou 310014, China
S
* Supporting Information
ABSTRACT: The direct addition of in situ generated hydrazoic acid to
alkynes is realized without solvent by using a gold catalyst derived from a
recently designed remotely functionalized biaryl-2-ylphosphine ligand (i.e.,
WangPhos). With terminal alkynes, the additions are mostly realized with
0.1 mol% catalyst loadings and at 40 °C. With more challenging internal
alkynes devoid of direct EWG substitution, the one-step transformation is
realized for the first time with generally high efficiency at ambient
temperature.
inyl azides are versatile intermediates in the synthesis of
(Scheme 1B). Whereas the first two approaches afford the
electronically deficient subtypes, the last three strategies are
suitable for the preparation of those of electronically neutral or
rich, but are still limited especially in the cases of internal vinyl
azides. For the 1,2-elimination approach, the method developed
by Hassner,3a,6 whereas an iodoazidation of alkenes using
iodine azide is followed by a base-promoted elimination of HI,
is often employed due to its succinct nature (Scheme 1B-3).
However, this method is not ideal as IN3, despite in situ
generated, is still a safety concern due to its explosive nature,
the strongly oxidative/electrophilic nature of the conditions
including the use of ICl significantly limits the scope of
tolerable functional groups,7 and with internal alkene substrates
regioselectivity can be an issue. For the hydroboration/
oxidative azidation approach,4 the reaction requires excess of
Cu(II) salt, uses unstable and freshly prepared Sia2BH, and
leads to formally anti-Markovnikov syn addion of HN3, and the
symmetric case shown in Scheme 1B-4 is the only reported
internal alkyne example.
V
bioactive alkaloids and/or N-heterocycles1 and serve as
precursors to reactive vinyl nitrene and synthetically excep-
tionally valuable 2H-azirines (Scheme 1A).2 They can also react
readily with organometallic reagents to generate metalloimine
with newly formed C−C bonds and with a radical to access an
imino radical.1i Their synthesis1d,e,3 can be mostly achieved via
condensation of aldehydes with α-azido esters/ketones, 1,4-
additions to electron-deficient π systems, 1,2-eliminations,
5
hydroboration/oxidative azidation,4 and 1,2-additions of HN3
Scheme 1. Reactivities of Alkenyl Azides and Their Synthesis
Metal-catalyzed 1,2-additions of HN3, i.e., hydrazoic acid, to
alkynes could offer a straightforward, one-step, atom-economic
alternative. Echavarren and co-workers8 reported a synthesis of
tetrazoles from terminal alkynes and in situ generated HN3 in
the presence of JohnPhosAu(NCMe)+ SbF6 (Scheme 1B-5).
−
The vinyl azide 1, a Markovnikov adduct, is the likely reaction
intermediate but is apparently susceptible to further reaction
under the requisite thermo conditions (i.e., 80 °C) except in
one case. A corresponding silver catalysis implicated by the
work of Jiao9 and later developed by Bi5 is again conducted
with terminal alkynes at 80 °C, and vinyl azides are formed in
good yields at short duration but could react further after
extended heating to eventually afford nitriles.9 While in these
studies the hazardous nature of hydrazoic acid is mitigated by
Received: May 6, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01359
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
its in situ generation via the reaction between commercially
available TMSN3 and an alcohol, the reactions are limited to
terminal alkynes, and the requisite elevated temperature
appears to diminish the reaction yield. For internal alkynes
leading to internal vinylazides, they are implicated as
intermediates in an Au/Ag catalysis but not detected.10
Table 1. Reaction Optimization
a,b
time
(h)
yield
Recently we11 reported that by functionalizing the privileged
biphenyl-2-ylphosphine framework with an amide group at the
C3′ position the resulting bifunctional phosphine ligand, i.e.,
WangPhos, enables exceptionally expedient acid additions to
alkynes, thereby permitting ppm-level gold catalysis (Scheme
2A). Comparing with JohnPhos, which is sterically and
entry
1
catalyst (mol%)
conditions
(%)
WangPhosAuCl/AgNTf2
(1)
DCM (0.1M),
rt
18
18
2
58
2
3
JohnPhosAuCl/AgNTf2 (1) DCM (0.1M),
rt
3
WangPhosAuCl/AgNTf2
(1)
neat, rt
95
4
5
6
WangPhosAuCl/AgOTf (1) neat, rt
WangPhosAuCl/AgSbF6 (1) neat, rt
2
2
2
82
92
96
Scheme 2. (a) WangPhos and Its Application in Gold-
Catalyzed Acid Addition to Alkyne; and (b) the General
Design of Remotely Functionalized Biphenyl-2-ylphosphine
Specifically Accommodating Au(I) Catalysis
WangPhosAuCl/NaBARF
(1)
neat, rt
7
8
9
WangPhosAuCl (0.1)
NaBARF (1)
neat, rt
10
2
91
c
WangPhosAuCl (0.1)
NaBARF (1)
neat, 40 °C
neat, 40 °C
93
JohnPhosAuCl (0.1)
NaBARF (1)
24
62
25
10
11
IPrAuCl (0.1)/NaBARF (1) neat, 40 °C
24
24
Ph3PAuCl (0.1)/NaBARF
(1)
(2,6-tBu2C6H3)O]3PAuCl
(0.1)
neat, 40 °C
trace
12
neat, 40 °C
24
11
NaBARF (1)
13
BrettPhosAuCl (0.1)
neat, 40 °C
24 h
63
NaBARF (1)
a
Determined by 1H NMR using the terminal methyl group as
electronically comparable but lack of the remote amide
group, this novel ligand accelerates the reaction by estimated
860 times. The design, shown in Scheme 2B and supported by
DFT calculations, entails the positioning of the remote amide
group near the incoming nucleophile by relying on the rather
unique and robust linear structure of L-Au-substrate, thereby
enabling a general base catalysis by the amide group and a quasi
intramolecular nucleophilic attack at the alkyne.
In our attempt to expand the synthetic utility of this novel
ligand and explore the applicability of the design to valuable
nucleophiles beyond carboxylic acids, H2O and aniline,11 we
decided to examine in situ generated HN3, which, however, is
structurally distinctive from our reported nucleophiles due to
its linear azido moiety and could be challenging. Nevertheless,
potential rate acceleration of HN3 addition to alkynes by
WangPhos would offer advantages of high synthetic values,
including (a) low reaction temperatures, which could avoid
product decomposition, (b) low catalyst loadings, (c) a broad
reaction scope including previously unsuitable internal alkynes,
and/or (d) synthetically useful regioselectivity.
b
reference. The remaining 1a and its hydration product account for
the mass balance. 87% isolated yield.
c
−
better than NTf2 . With an order of magnitude lower gold
loading, i.e., 0.1 mol%, the relatively long reaction time in entry
7 was cut down to 2 h by slightly raising the temperature to 40
°C, and the reaction was highly efficient (entry 8). In
comparison, with JohnPhos as ligand, the reaction was
substantially slower, and the yield was only 62% after 24 h
(entry 9). Other often used gold ligands were also examined,
but most were much inferior to JohnPhos (entries 10−12),
with the exception of BrettPhos, which is comparable to
JohnPhos (entry 13).
The reaction scope with terminal alkynes was then examined.
As shown in Table 2, both cyclohexylacetylene (entry 1) and
cyclohex-1-en-1-ylacetylene (entry 2) participate the reaction
smoothly. Alcoholic substrates 1d−1h with different protecting
groups are readily allowed (entries 3−7). For 1g and 1h of
increased steric hindrance and less reactive C−C triple bond,
the catalyst loading is a higher 0.5 mol%. For the sulfonamide
alkyne substrate 1i, an even higher 2 mol% is needed (entry 8).
The reaction with phenylacetylene was uneventful. Notably,
these reactions are all highly efficient, with yields ranging from
80 to 95%.
With the success with terminal alkynes, we then turned our
attention to less reactive and hence challenging internal alkynes
lacking direct EWG substitution, which have so far not been
succumbed to metal-catalyzed hydroazidation reactions. In-
deed, their reactions are in general much slower and best run
with 5 mol% catalyst loading, by using more reactive PrOH as
the proton donor and at ambient termperature. Nevertheless,
the desired vinyl azides can be obtained in fair to excellent
At the outset, we studied the hydroazidation by using the
more reactive terminal alkyne 1-dodecyne as substrate. As
shown in Table 1, entry 1, the reaction performed with in situ
t
generated WangPhosAuNTf2 (1 mol%) as catalyst, BuOH as
the proton source and in DCM indeed occurred at ambient
temperature, but did not complete even after 18 h. Never-
theless, WangPhos performed much better than JohnPhos
(entry 2), indicating that the ligand remote amide group
engages accelerating interaction with HN3. After substantial
effort of condition optimization, we discovered that the reaction
was at best run neat. To this end, the vinyl azide 2a was formed
in an outstanding 95% yield in 2 h (entry 3). Counteranion
screening (entries 4−6) reveals that BARF− performed slightly
i
B
DOI: 10.1021/acs.orglett.7b01359
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
a
Table 2. Reaction Scope with Terminal Alkyne Substrates
Table 3. Reaction Scope with Internal Alkyne Substrates
a
Reaction conditions: TMSN3 (2 equiv), tBuOH (2 equiv),
WangPhosAuCl (0.1 mol%), NaBARF (1 mol%), neat, N2, 40 °C.
b
c
0.5 mol% of WangPhosAuCl used. 2 mol% of WangPhosAuCl and 2
mol% of NaBARF used.
yields (Table 3). For example, hydroazidation of the symmetric
6-dodecyne affords (Z)-6-azido-6-dodecene in 79% yield (entry
1). With the sterically biased cyclohexyl methyl acetylene 1l,
the reaction was sluggish, affording the vinyl azide 2l as a
regioisomeric mixture in a moderate 52% yield and in >90%
yield based on substrate consumption (entry 2). This observed
preference of the azido group attacking the less sterically
demanding alkyne end is consistent with our prior oxidation
chemistry.12 Aliphatic alkynes with protected/functionalized
hydroxyl groups were also examined. A moderate regioselec-
tivity was observed in the case of 1m (entry 3), and the major
isomer of the product 2m is favored by sterics and inductive
effect. The inductive effect is enhanced with 1n featuring a
more electron-withdrawing tosyloxy β to the C−C triple bond
and largely responsible for the synthetically useful 5:1 ratio
(entry 4). In the case of an α-EWG, even a MeO group is
inductive enough to dictate a complete regioselectivity (entry
5). With enyne 1p where the C−C triple bond is conjugated to
an alkene, the reaction turned out to be highly regioselective,
despite a moderate yield due to sluggish conversion (entry 6).
We then studied various arylalkynes 1q−1v and interrogated
the impact of arene substituents on reaction regioselectivity.
Somewhat surprisingly, for the phenyl substrate 1q, the
hydroazidation is only moderately selective; moreover, the 2-
azido isomer of 2q is slightly favored over the 1-azido one,
suggesting the phenyl group is a bulky group and possibly also
behaves more of an inductive electron-withdrawing group than
a resonance stabilizer (entry 7). With a 4-MeO group, the 1-
azido isomer is favored in a ratio of 3.1/1 over the 2-azido
a
Reaction conditions: TMSN3 (2 equiv), iPrOH (2 equiv),
WangPhosAuNTf2 (5 mol%), NaBARF (5 mol%), neat, N2, rt.
b
c
Regioisomer ratios determined by 1H NMR. 2 mol% of
d
WangPhosAuNTf2 and NaBARF used. The remaining substrate
accounts mostly for the mass balance. The NMR yield based on
substrate conversion. Only the major isomer was isolated.
e
f
product (entry 8). As expected, an electron-withdrawing ester
group at the same position enhances the preference for the 2-
azido isomer (entry 9). To our surprise, an ortho-Me group
leads to a good regioselectivity (11:1) favoring the 1-azido
product (entry 10). An ortho-Br, however, results in no
selectivity (entry 11). Finally, diphenylacetylene reacted
smoothly to afford the azidostilbene 2v in an excellent yield
(entry 12). Importantly, these reactions are mostly good to
C
DOI: 10.1021/acs.orglett.7b01359
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(3) (a) L’Abbe, G.; Hassner, A. Angew. Chem., Int. Ed. Engl. 1971, 10,
98−104. (b) Collier, S. J. Alk-1-Enyl Azides In Science of Synthesis;
Thieme Chemistry, 2007; Vol. 33, p 541.
(4) Masuda, Y.; Murata, M.; Ikeda, M.; Watanabe, S. J. Chem. Soc.,
Perkin Trans. 1 1998, 1013−1014.
(5) (a) Liu, Z.; Liu, J.; Zhang, L.; Liao, P.; Song, J.; Bi, X. Angew.
Chem., Int. Ed. 2014, 53, 5305−5309. (b) Liu, Z.; Liao, P.; Bi, X. Org.
Lett. 2014, 16, 3668−3671.
(6) Hassner, A.; Levy, L. A. J. Am. Chem. Soc. 1965, 87, 4203−4204.
(7) Hassner, A.; Marinescu, L.; Bols, M. Iodine Azide In Encyclopedia
of Reagents for Organic Synthesis; John Wiley & Sons, Ltd, 2001.
(8) Gaydou, M.; Echavarren, A. M. Angew. Chem., Int. Ed. 2013, 52,
13468−13471.
(9) Shen, T.; Wang, T.; Qin, C.; Jiao, N. Angew. Chem., Int. Ed. 2013,
52, 6677−6680.
(10) (a) Qin, C.; Feng, P.; Ou, Y.; Shen, T.; Wang, T.; Jiao, N.
Angew. Chem., Int. Ed. 2013, 52, 7850−7854. (b) Qin, C.; Su, Y.; Shen,
T.; Shi, X.; Jiao, N. Angew. Chem., Int. Ed. 2016, 55, 350−354.
(11) Wang, Y.; Wang, Z.; Li, Y.; Wu, G.; Cao, Z.; Zhang, L. Nat.
Commun. 2014, DOI: 10.1038/ncomms4470.
high yielding, and should be synthetically useful despite the
moderate regioselectivities in some cases.
In summary, we have applied our designed, remotely
functionalized biaryl-2-ylphosphine ligand, i.e., WangPhos, to
a gold-catalyzed direct addition of in situ generated hydrazoic
acid to alkynes. With terminal alkynes, the reactions are mostly
realized with 0.1 mol% catalyst, in the absence of solvent and
under mild heating (40 °C). With internal alkyne substrates,
this one-step transformation to synthetically valuable internal
vinyl azides is realized for the first time with generally high
efficiency due to the exceedingly mild reaction conditions.
Moreover, some of the reactions exhibit synthetically valuable
regioselectivities. Comparing to electronically and sterically
similar JohnPhos, the significant rate acceleration by WangPhos
is consistent with that the ligand 3′-amide group behaves as a
general base catalyst in promoting HN3 attack of alkyne, despite
the distinctive linear structure of the azido moiety. This
successful accommodation of valuable and structurally diverse
nucleophiles beyond previously studied carboxylic acid, H2O,
and aniline suggests the general utility of WangPhos in
substantially accelerating and improving gold catalysis.
(12) Lu, B.; Li, C.; Zhang, L. J. Am. Chem. Soc. 2010, 132, 14070−
14072.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b01359.
Experimental procedures and compound characterization
and spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*zhang@chem.ucsb.edu.
ORCID
Xiaoqing Li: 0000-0003-4742-2131
Liming Zhang: 0000-0002-5306-1515
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank NSF CHE-1301343 for finanical support and NIH
shared instrument grant S10OD012077 for the purchase of a
400 MHz NMR spectrometer. Z.W. thanks the Mellichamp
Academic Initiative in Sustainability for a fellowship in support
of his research.
REFERENCES
■
(1) (a) L’Abbe, G. Angew. Chem. 1975, 87, 831−8. (b) Hassner, A.
Vinyl Azides and Nitrenes In Azides Nitrenes; Academic, 1984; pp 35−
94. (c) Moody, C. J. Stud. Nat. Prod. Chem. 1988, 1, 163−185.
(d) Brase, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew. Chem., Int.
̈
Ed. 2005, 44, 5188−5240. (e) Banert, K. The Chemistry of Vinyl,
Allenyl, and Ethynyl Azides; John Wiley & Sons Ltd., 2010. (f) Chiba, S.
Chimia 2012, 66, 377−381. (g) Hu, B.; DiMagno, S. G. Org. Biomol.
Chem. 2015, 13, 3844−3855. (h) Ni, H.; Zhang, G.; Yu, Y. Curr. Org.
Chem. 2015, 19, 776−789. (i) Chiba, S. Synlett 2012, 2012, 21−44.
(2) (a) Palacios, F.; de Retana, A. M. O.; de Marigorta, E. M.; de los
Santos, J. M. Eur. J. Org. Chem. 2001, 2001, 2401−2414.
(b) Khlebnikov, A. F.; Novikov, M. S. Tetrahedron 2013, 69, 3363−
3401.
D
DOI: 10.1021/acs.orglett.7b01359
Org. Lett. XXXX, XXX, XXX−XXX