Catalytic Azido-Hydrazination of Alkenes Enabled by Visible Light: Mechanistic Studies and Synthetic Applications

Article abstract of DOI:10.1002/adsc.201901041

A visible-light-enabled catalytic intermolecular azido-hydrazination method for unactivated alkenes is developed via an orderly radical addition sequence. This transformation features metal-free and redox-neutral conditions and is applicable to a wide range of alkenes with commercially available reagents. Mechanistic and kinetic studies reveal that the efficient generation of azide radical enabled by fluorenone under visible-light is critical to this methodology. The β-azido alkyl hydrazines prepared with this reaction can be conveniently derived to valuable synthetic building blocks, and one of the products has been successfully applied in the total synthesis of (±)-ibrutinib, which is used to treat B cell cancers. (Figure presented.).

Full text of DOI:10.1002/adsc.201901041

Title: Catalytic Azido-Hydrazination of Alkenes Enabled by Visible
Light: Mechanistic Studies and Synthetic Applications
Authors: Peng Wang, Yunxuan Luo, Songsong Zhu, DENGFU LU, and
Yuefa Gong
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201901041
Link to VoR: http://dx.doi.org/10.1002/adsc.201901041

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Catalytic Azido-Hydrazination of Alkenes Enabled by Visible
Light: Mechanistic Studies and Synthetic Applications
Peng Wang, Yunxuan Luo, Songsong Zhu, Dengfu Lu* and Yuefa Gong*
a
School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, 1037 Luoyu Rd.,
Wuhan, Hubei, 430074, China
E-mail: dlu@hust.edu.cn; gongyf@hust.edu.cn
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract. A visible-light-enabled catalytic intermolecular The β-azido alkyl hydrazines prepared with this reaction
azido-hydrazination method for unactivated alkenes is can be conveniently derived to valuable synthetic building
developed via an orderly radical addition sequence. This blocks, and one of the products has been successfully
transformation features metal-free and redox-neutral applied in the total synthesis of (±)-ibrutinib, which is used
conditions and is applicable to a wide range of alkenes with to treat B cell cancers.
commercially available reagents. Mechanistic and kinetic
studies reveal that the efficient generation of azide radical
enabled by fluorenone under visible-light is critical to this
methodology.
Keywords: β-azido hydrazine; azodicarboxylate; diamines;
azide radical; radical addition cascade
Introduction
Organic hydrazines represent an important type of
compounds due to their ubiquitous occurrence in
pharmaceuticals, pesticides, and dyes.[1] Their
applications are also associated with heterocycles
such as pyrazoles and pyridazines.[2] Particularly,
hydrazine-containing heterocycles with a vicinal
functional group are of great interests in the
development of drug molecules (Figure 1). For
example, ibrutinib is used to treat B cell cancers like
mantle cell lymphoma and chronic lymphocytic
leukemia,[3] while baricitinib is a drug for the
treatment of rheumatoid arthritis.[4] A direct and
catalytic preparation of vicinal-functionalized
aliphatic hydrazines is highly desirable in the realm
of organic and medicinal chemistry, however, such
transformations still remain challenging.
Scheme 1. Hydrazination of alkenes with azodicarboxylate
as a reagent.
Dialkyl azodicarboxylates are widely utilized in
organic synthesis as electrophiles and terminal
oxidants.[5]
When
hydrazination/amination
used
for
electrophilic
purposes,[6]
azodicarboxylates are known to react with
nucleophiles such as carbanion equivalents, dienes,
and electron-rich alkenes.[7] Enolates and enamines
were reported to form [2+2] adducts with
azodicarboxylates, which could be consequently
Figure 1. Representative drug molecules containing
vicinal-functionalized hydrazine heterocycle moieties.
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
opened up by nucleophiles to yield difunctionalized
products (Scheme 1a).[8] Alternatively, the carbanion
intermediates formed in the nucleophilic additions to
electron-deficient double bonds could also be
captured by an azodicarboxylate, providing vicinal-
For safety and solubility considerations, TMSN3 was
employed as the azide source.[25] Fortunately, di-tert-
butyl azodicarboxylate 2 showed decent stability
towards TMSN3 in a few typical organic solvents at
room temperature. Since the azo-ene reaction is a
potential competing pathway,[10a] we deliberately
selected but-3-en-1-yl benzoate (1a) that possesses
functionalized
hydrazines
(Scheme
1b).[9]
Nevertheless, the hydrazination of unactivated
alkenes with azodicarboxylates is left underdeveloped, allylic C–H bonds as a model substrate for the
possibly due to the following reasons: 1) regular
alkenes are generally not electron-rich enough; 2)
alkenes with allylic C–H bonds undergo azo-ene
reactions to produce allylic hydrazines.[10]
catalyst discovery and condition optimization.
Current methods that convert TMSN3 to azide
radicals under visible light require sophisticated
photocatalysts such as Ir-complexes,[28a,b,e] acridinium
salts,[28f] Rose Bengal[28d] and sulfonium iodate.[28c]
Following our recent discoveries, simple ketone
photo-sensitizers were first attempted to generate
azide radical due to their availability. Gratifyingly,
the desired β-azido hydrazine 3a was obtained in
moderate yield with fluorenone 4a (6 W blue LEDs).
However, a reproducibility issue emerged when the
reaction was conducted in anhydrous solvents. After
extensive condition screening, a proton source turned
out to be crucial to this transformation. The results of
several key experiments are summarized in Table 1.
Beside the ionic pathways, azodicarboxylates are
also known to act as radical acceptors. Recently, a
few examples of radical hydrazinations were reported
based on C–H activation[11] or decarboxylation.[12]
Azodicarboxylates have been used to trap the alkyl
radicals formed in cyclization/fragmentation as well,
affording intramolecular hydrazination products.[13]
The pioneering studies on the intermolecular
functionalization of unactivated alkenes with
azodicarboxylates were reported by Carreira et al.
They developed
a
cobalt-catalysed hydro-
hydrazination reaction that produces alkyl hydrazines
in Markovnikov fashion (Scheme 1c).[14,15] Herein,
we propose a catalytic strategy that could directly
convert unactivated alkenes to vicinal-functionalized
alkyl hydrazines via radical polarity-controlled
addition cascades as illuminated in Scheme 1d.[16] To
materialize such redox-neutral transformations, a
suitable single-electron redox system has to be
established.
Table 1. Catalyst discovery and condition optimization. a)
Our lab has recently developed an N-oxyl radical
mediated C–H functionalization of ethers enabled by
ketone photosensitizers under visible light.[17] When
this catalytic system was applied with an alkene and
BocN=NBoc, an oxy-hydrazination product was
obtained,[18] revealing the feasibility of an orderly
radical addition cascade among a nucleophile, an
Entry catalyst &
H+
conc.
t
conv. yield
(%)b)
conditions donor (M)
(h)
(%)c)
1
none
0.1
5
5
5
5
5
5
5
5
5
5
5
5
27
75
75
41
78
14
<5
<5
49
19
15
23
23
72
71
21
75
11
<5
<5
47
14
10
20
77
87
65
4a
2
MeOH 0.1
AcOH 0.1
tBuOH 0.1
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.2
MeOH 0.5
4a
3
4a
4
4a
unactivated
alkene
and
an
electrophilic
5
4a
azodicarboxylate with a simple catalyst. These
preliminary results encouraged us to develop this
sequential radical addition mode into a more
synthetically useful protocol. Azide anion was
thereby chosen as the “nucleophile” to continue our
exploration because the anticipated azido-
hydrazination products could be excellent precursors
to vicinal amino hydrazines, azido-amines and
diamine derivatives.[19] Also, azide is a versatile
6
4b
7
8
None
4a, dark
4a, 3W
4c, dark
4d, dark
4e, dark
4a
9d)
10
11
12
13
14e)
15
10 81
10 91
24 69
4a
4e, dark
functional group that serves an excellent linker,[20]
a
a) Unless otherwise stated, the reaction was performed with
1a (0.4 mmol), 2 (0.56 mmol), TMSN3 (0.56 mmol), a
proton source (0.4 mmol) and a corresponding catalyst
(0.04 mmol) in CH2Cl2/CH3CN (4:1, 2 mL) at 35 – 40 ˚C
under the irradiation of 6 W blue LEDs. b) The conversion
of 1a was determined by HPLC with an internal standard.
reliable nitrene precursor[21] and many other roles.[22]
Azide anion and its equivalents are well-known to be
converted to azide radicals by multiple types of
oxidants like redox-active transition metals,[23]
hypervalent iodines,[24] peroxide,[25] electrochemical
mediators,[26] and photosensitizers.[27, 28]
c)
d)
The yield of 3a was measured by HPLC. 3 W blue
LEDs were used as a light source. e) A second portion of
TMSN3 (0.16 mmol) and 2 (0.16 mmol) was added as a
CH2Cl2 solution after 2 h.
Results and Discussion
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
With methanol as a proton source, the reaction
6 d)
7
1.4
1.4
1.8 c) 78
1.4 68
1.8 c) 84
75
72
proceeded smoothly to afford 3a in good yield and
excellent chemo-selectivity (entry 2, 96% yield based
on the recovery of 1a). Notably, there was no allylic
hydrazination product detected for this particular
substrate. Using acetic acid or tert-butanol as proton
sources led to either comparable or inferior outcomes
(entries 3,4). Increasing the concentration to 0.2 M
slightly accelerated the reaction, however, 1a still
could not be fully converted (entry 5). Another
photosensitizer 2,3-butanedione 4b, which was highly
efficient in our previous work, exhibited poor activity
8
9 e)
10
1.4
1.4
1.8 c) 80
1.4 75
1.8 c) 78
67
73
(entry 6). Critically, in the absence of
a
photosensitizer, 3a was not detected at all after 5 h,
wheras the same observation was made when the
reaction was performed with 4a in darkness (entries
7,8). Lowering the intensity of light irradiation also
caused a diminished conversion (entry 9). As a
comparison, a few alternative catalysts that are
reported to be capable of generating azide radical are
also evaluated without light. Under the same
conditions, Mn(OAc)3•2H2O 4c, hypervalent iodines
4d and 4e were far less effective compared to this
visible-light-enabled procedure (entries 10-12).
11
12
a) Unless otherwise stated, the reaction was performed with
an alkene 1 (0.4 mmol), di-tert-butyl azodicarboxylate 2
(0.56 mmol), TMSN3 (0.56 mmol), methanol (0.4 mmol)
and fluorenone (0.04 mmol) in CH2Cl2/CH3CN (4:1, 2 mL)
b)
at 35 – 40 ˚C under the irradiation of 6 W blue LEDs.
c)
Isolated yield. A second portion of TMSN3 and 2 (0.16
mmol for each) were added after 2 h. d) The reaction was
Although the chemo-selectivity for 1a was
excellent, it was found that the conversion and yield
still remained unsatisfactory even when BocN=NBoc
was fully consumed after extended reaction time
(entry 13). After scrutinizing the reaction mixture, we
were able to identified two major by-products:
e)
conducted with 4e (10 mol%) in darkness. An allylic
hydrazination side product was isolated.
Having established the optimal reaction conditions,
we next evaluated the azido-hydrazination of several
non-conjugated alkenes, preferentially with the
visible-light-promoted procedure. Decent conversions
and isolated yields were generally achieved by using
1.4 equivalent of TMSN3 and azodicarboxylate 2. For
those less reactive alkene substrates, adding an
additional portion of BocN=NBoc and TMSN3 (0.4
equiv for each) after 2 h could effectually improve
the yields to a satisfactory level. Several functional
groups were well-tolerated by this catalytic method,
such as benzylic C–H bonds and carboxylates (entries
1-5). Notably, allyl benzene suffered an unexpected
poor conversion under the photocatalytic conditions,
while its reaction went smoothly with benziodoxole
4e (10 mol %) to afford 3f in 75% yield (entry 6). In
addition, a 1,1-disubstituted and a tri-substituted
olefin also readily participated in the azido-
hydrazination to produce hydrazines attached to
quaternary carbons (entries 7,8). Then, a few cyclic
alkenes were taken into consideration. Compared to
acyclic ones, cyclohexene showed higher reactivity.
However, the desired 2-azido-cyclohexyl hydrazine
3i was only obtained in 67% yield (entry 9) and an
allylic hydrazination side product (29% yield) was
generated, possibly through the azo-ene process. For
cyclooctene, no distinct allylic hydrazine was
observed, and product 3j was isolated in 80% yield
after an extra portion of azido-hydrazination reagents
were added (entry 10). Norbornene was also
applicable with this transformation, furnishing a pair
of diastereomers, both with the azide group on exo-
position (entry 11). For potential synthetic purposes,
a pent-4-en-1-yl tosylate 1l was then applied under
hydrazine
5 and isobutene azido-hydrazination
product 6, both of which were not associated with
starting material 1a. Based on these findings, the
HPLC yield of 3a was finally improved to 87% by
adding another portion of TMSN3 and 2 (0.4 equiv
for each) after 2 h. On the other hand, by increasing
the reaction concentration to 0.5 M, a decent yield of
3a could also be achieved with hypervalent iodine
reagent 4e (10 mol%) at the same temperature after
extended reaction time (entry 15).
Table 2. Azido-hydrazination of isolated alkenes. a)
entry substrate
product
2+TMSN3 yield
(%)
(equiv)
b)
1.4
74
1
2
3
4
5
1.8 c) 84
1.4
1.8 c) 76
1.4 67
1.8 c) 80
66
1.4
1.4
78
68
1.8 c) 88
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
the photo-catalytic condition. Fortunately, the O-tosyl
group remained intact and the desired molecule 3l
was isolated in practical yield (entry 12). The
synthetic application of 3l will be discussed later.
Next, the compatibility of a series of conjugated
alkenes was inspected and the results are summarized
in Table 3. Styrene and β-vinylnaphthalene exhibited
noticeably higher reactivity to afford corresponding
azidohydrazines 3m and 3n in excellent yields
without using an extra amount of reagents (Table 3,
entries 1,2), while 4-fluorostyrene behaved more like
an isolated alkene (entry 3). The reaction of indene
reached completion in 2 h to provide product 3p as
the only diastereomer in exceptional yield (entry 4).
Conjugated internal alkenes, such as a silyl protected
cinnamyl alcohol 1q and anethole 1r, could also be
efficiently functionalized under standard conditions
(entries 5, 6). Due to the electron-deficient nature of
methyl cinnamate 1s, the addition of azide radical to
C=C double bond became less efficient and
BocN=NBoc was catalytically reduced to afford
hydrazine 5 as a major side product under the
photocatalytic conditions. Fortunately, the alkene
functionalization product 3s was successfully
prepared by drastically increasing the reaction
concentration (1.0 M) with the benziodoxole 4e
recipe (entry 7). Notably, this radical azido-
hydrazination features a reversed regioselectivity
compared to a normal nucleophilic addition sequence
as illustrated in Scheme 1b. The reversed polarity
may be attributed to a better radical stabilization
effect provided by the phenyl group. On the other
hand, dihydrofuran 1t, vinyl acetate 1u and N-
vinylpyrrolidinone 1v, which were reported to be
electron-rich enough to attack azodicarboxylate
directly,[8,26] were also compatible with this catalytic
radical cascade, providing hydrazines 3t, 3u, and 3v,
respectively, with reversed regioselectivity (entries 8-
10).
6
1.4
1.8
91
56
7 d)
8
9
1.4
1.4
83
64
10 d)
1.4
85
a) Unless otherwise stated, the reaction was performed with
an alkene 1 (0.4 mmol), di-tert-butyl azodicarboxylate 2
(0.56 mmol), TMSN3 (0.56 mmol), methanol (0.4 mmol)
and fluorenone (0.04 mmol) in CH2Cl2/CH3CN (4:1, 2 mL)
b)
at 35 – 40 ˚C under the irradiation of 6 W blue LEDs.
c)
Isolated yield. An additional portion of TMSN3 and 2
(0.16 mmol for each) were added after 2 h. d) The reaction
was conducted with 4e (10 mol%) in darkness.
Scheme 2. The reactivity comparison between different
types of alkenes.
Table 3. Azido-hydrazination of isolated alkenes. a)
Though this catalytic method showed good activity
towards different types of substrates, conjugated
alkenes (p-π or π-π) were noticed to be generally
more reactive than those isolated ones. For instance,
the reaction of 1a became very slow at its later stage,
with an isobutene azido-hydrazination product 8
competitively produced (Table 1). To tackle this
unproductive pathway, an extra portion of reagents
had to be used to ensure a good yield. For comparison,
most of the conjugated alkenes achieved adequate
conversion in shorter periods without additional
reagents. To understand this reactivity difference, we
first compared the reaction of an isolated alkene (1-
octene, 1d) and a conjugated one (styrene, 1m) side
by side. After quenching both reactions after 1 h, the
product of 1-octene 3d was isolated in only slightly
lower yield compared to 3m (33% versus 45%,
Scheme 2a). Unexpectedly, when 1d and 1m were
subjected to the same reaction condition in one vessel,
only the styrene product 3m was detected (Scheme
entry substrate
product
2+TMSN3 yield
(%)b)
(equiv)
1
2
1.4
1.4
93
95
3
4
1.8c)
1.4
82
95
5
1.4
79
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
2b), implying that styrene exerted an inhibition effect
on 1-octene. These interesting findings intrigue us to
investigate the reaction kinetic profiles and seek for
detailed mechanistic insights.
and fluorenone 4a (0.04 mmol) in CH2Cl2/CH3CN (4:1, 2
mL) at 35 – 40 ˚C under the irradiation of 6 W blue LEDs.
b) The recovery of alkene was determined by GC analysis
d)
with dodecane as an internal standard. c) Isolated yield.
Not determined. e) Not available.
Next, cis-4-octene was used as a probe to further
shed light on the reaction mechanism. Under the
standard conditions, azido-hydrazine 7 was only
isolated as a minor product, accompanied by two
allylic hydrazination products (8, 9) and trans-4-
octene (Table 4, entry 1). A few key control reactions
were consequently carried out. First, none of these
above products were detected without light
irradiation (entry 2). Similarly, in the absence of
fluorenone 4a, neither azido-hydrazination nor allylic
hydrazination products were observed, except for a
small amount of trans-alkene (entry 3). Next, TMSN3
was also proven a necessity for all the above
reactions to occur, including the alkene isomerization
(entry 4). Last, in the absence of azodicarboxylate 2,
trans-4-octene were quickly formed (entry 5).
According to previous reports, the trans-alkene was
formed through the reversible azide radical addition-
elimination sequence.[24a,28e] In short, the above
observations suggested that all the alkene
isomerization, azido-hydrazination and allylic
hydrazination process were triggered by azide
radicals, which were efficiently generated by treating
TMSN3 with fluorenone under blue light irradiation.
Scheme 3. Kinetic studies on the visible-light-enabled
azido-hydrazination of 1a.
First, we explored the effects of proton donors and
light source on the reaction process of the model
substrate 1a. Under the standard light-enabled
condition, product 3a was generated smoothly in the
first 4 hours as indicated by HPLC analysis. But the
reaction decelerated significantly when approaching
its later stage (Scheme 3a), presumably due to the
decreasing concentration of reactants. Interestingly,
switching the proton donor to CD3OD caused a
remarkably slower reaction, and what’s more, longer
induction and poor conversion were observed in the
absence of a proton source. The results of the light
on/off experiment were summarized in Scheme 3b,
clearly demonstrating that this reaction was light-
dependent, as product 3a was only formed during
light irradiation.
Table 4. Using cis-4-octene as a mechanistic probe. a)
entry variations from cis-4-
7c) 8+9 trans-4-
c)
standard cond. octeneb)
octeneb)
1
2
3
4
5
None
in dark
6%
22
40%
ND
ND
21%
<1%
8%
<1%
68%
Figure 2. Isomerization of cis-4-octene under different
conditions. (A) The isomerization process under standard
conditions; (B) The isomerization process in the absence of
BocN=NBoc 2; (C) The isomerization with benziodoxole
4e (10 mol%) as the catalyst; (D) The isomerization in the
presence of styrene (1.0 equiv). For detailed experimental
procedures, see the supporting information.
99% NDd)
89% ND
without 4a
without TMSN3 85% N/Ae) ND
without 2
18% N/A
N/A
a) Unless otherwise stated, the reaction was performed with
cis-4-octene (0.4 mmol), di-tert-butyl azodicarboxylate 2
(0.56 mmol), TMSN3 (0.56 mmol), CH3OH (0.4 mmol)
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
Since the addition of azide radical to alkenes is
Since the azide radical generation and addition
have been proven fast and reversible, the rate of the
hydrazination step may play a pivotal role in the
whole cascade. Although the addition of alkyl
radicals to azodicarboxylates are reported a rapid
process in recent studies,[11,12] few kinetic data are
available. Herein, we tried to use β-pinene as a
radical clock to study the rate of this radical
hydrazination step. As an abundant natural
monoterpene, the radical oxygenation mechanism of
α and β-pinene has been intensively studied.[31]
According to previous reports, the radical addition to
β-pinene would afford a radical species A, and the
cyclobutyl ring (in green) is known to open promptly
to release the strain, affording a more stable tertiary
radical B. The rate constant of this rearrangement
step was documented to be approximately 8 × 105 s-1
at ambient temperature.[32] Applying β-pinene with
the visible light-enabled procedure afforded three
major products: a direct azido-hydrazination product
10, a ring-opening azido-hydrazination product 11
and allylic hydrazination product 12, which is a
background azo-ene reaction product in this case
(Scheme 5, entries 1, 2).
known to be a reversible process, we conceived to
gain some insights on the relative rates of the
azidation and the following hydrazination steps by
studying the cis-4-octene isomerization (equation 1).
As shown in Figure 2A, under the virtual reaction
conditions, the isomerization of cis-4-octene started
from the very beginning and reached a stable
trans/cis ratio in about 40 min. This phenomenon
implied that the reversible azide addition/ elimination
steps outcompeted its following hydrazination step.
When the azide radical merely served as an
isomerization catalyst (in the absence of
BocN=NBoc), trans-4-octene was formed much
faster (Figure 2B). In contrast, the isomerization rate
was determined to be significantly slower under the
benziodoxole 4e promoted conditions (Figure 2C),
demonstrating that this method was far less effective
than the photocatalytic one in terms of azide radical
generation.
According to the above perceptions, given the
generation of azide radical was efficient enough, its
addition to an alkene should be a kinetically very
rapid event. What really limited the overall reaction
rate might be the concentration of the resulted alkyl
radical intermediate, which in turn was controlled by
the thermodynamics of the ∙N3 addition/elimination
equilibrium. To verify this hypothesis, we again
recorded the isomerization process of cis-4-octene in
the presence of styrene (1.0 equiv) and surprisingly
found that the formation of trans-4-octene was
substantially suppressed (Figure 2D). This
phenomenon echoes the inhibition effect observed in
the competing reaction between 1-octene 1d and
styrene 1m (scheme 2b). A plausible explanation is
illustrated in Scheme 4. As azide radicals and alkenes
form fast equilibriums, their constants Keq mainly
depend on the stability of the alkenes, ∙N3 and, most
importantly, the alkyl radical intermediates.
Substituents such as aryl group or hetero atoms that
are able to stabilize the alkyl radical adducts facilitate
the ∙N3 additions and increase the value of the
equilibrium constant Keq. However, normal alkyl
substituents do not stabilize radicals well, thus
leading to a smaller Keq. As a result, when cis-4-
octene and styrene were present in the same reaction
vessel, styrene competitively consumed ∙N3 to form a
thermodynamically more stable benzylic radical,
which explained the inhibited isomerization of cis-4-
octene. The result of the competing reaction in
Scheme 2b could be explained in a similar way.
Scheme 5. Using β-pinene as a radical clock.
Under the reaction conditions, azide radical and β-
pinene formed a fast equilibrium to provide
intermediate A in certain concentration. A may either
rearrange to a more stable radical B (unimolecular, r1
= k1[A]) and eventually give product 13, or be
directly captured by BocN=NBoc 2 (bimolecular, r2 =
k2[A][2]) to afford compound 10. The fact that 10 and
11 were isolated in comparable yield indicated the
capturing of the alkyl radical intermediate is fast
enough to compete with the ring-opening process (r1
≈
r2, k1[A]
≈
k2[A][2]). Knowing that the
concentration of BocN=NBoc [2] ranges from 0.28 to
0.1 M throughout the reaction, the rate constant k2 of
this hydrazination step was roughly calculated to be
between 106 and 107 M-1∙s-1, which is a quite fast
process. As we know, concentration has a smaller
impact on a first-order reaction than on a second-
order one. So an increased ratio of ring-opening
Scheme 4. The addition-elimination equilibrium between
azide radical and different types of alkenes.
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
product 11 was obtained when the reaction was
diluted to 0.05 M (Scheme 5, entry 3). These radical
clock experiments demonstrated that the radical
hydrazination step may have a generally large rate
constant, nevertheless, the specific constant varies
from one substrate to another.
r = k[E][2], where [E] = Keq[alkene][∙N3]
And the equation can be further deduced to:
r = k∙Keq[∙N3][alkene][2]
Based on this equiation, the key factors that control
the rate of this transformation include: a) the structure
of alkene that determines the equilibrium constant Keq
of azidation, as well as the rate constant k of the
hydrazination step; b) the catalytic method and light
intensity that influences the efficiency of ∙N3
generation; c) the concentration of substrates and
reagents. All these conclusions are in excellent
agreements with our experimental observations.[33]
Among these factors, the first and the third one are
unavoidable in order to achieve good conversions on
a general substrate scope. So the real key to this
transformation is the efficiency of catalytic azide
radical generation.
Scheme 7. The derivatization of 2-azido indanylhydrazine
3p. Conditions: (i) PPh3 (1.5 equiv), H2O (3 equiv), THF,
50 ˚C, 10 h; (ii) Ac2O (1.5 equiv), pyridine (1.5 equiv),
THF, rt; (iii) Sodium ascorbate (10 mol%), CuSO4‧ 5H2O
Scheme 6. A plausible mechanistic working hypothesis.
t
(5 mol%), phenylacetylene (1.2 equiv), BuOH:H2O (1:1),
Based on the results of the above mechanistic
studies, we proposed a plausible working hypothesis
for this visible-light-promoted reaction (Scheme 6).
First, the ∙N3 is generated via the catalytic oxidation
of azide source by a photo-excited fluorenone C,
accompanied by the formation of reductive ketyl
radical species D. Then, the azide radical quickly
adds to an alkene to give the alkyl radical
intermediate E. All these earlier steps are reversible
and exist in fast equilibriums, and the concentration
of E largely depends on the equilibrium constant and
the concentration of ∙N3. Next, alkyl radical E is
captured by azodicarboxylate 2, affording an N-
centered radical F. On one hand, F could be quickly
quenched by species D to yield the final product 3
(path I). Notably, D was captured and identified as
compound 13 through radical cross coupling with
styrene and TMSN3 in the absence of BocN=NBoc 2.
On the other hand, F could also be directly reduced
by H–N3 to regenerate an azide radical and propagate
the chain (path II). According to the experimental
results, both pathways are viable or even cooperating
with each other.
12 h; (iv) Cs2CO3 (2.5 equiv), BrCH2CO2Et (2.0 equiv),
CH3CN, 50 ˚C, 2 h; (v) Cs2CO3 (3 equiv), 80 ˚C, 12 h; (vi)
HCl/MeOH (10 equiv), rt, 10 h.
To explore the synthetic applications of the
obtained azido-hydrazination products, we chose the
product of indene (3p) as a model to study the
functional group derivatizations (Scheme 7). First,
through a Staudinger reaction, 3p can be reduced to
an amine, which was isolated after acetylation (14).
Second, a triazole 15 was easily formed to link up the
hydrazine moiety with an alkyne through Cu-
catalysed click chemistry. Next, the hydrazine N–N
bond can be selectively cleaved via a 2-step
procedure to render a β-azido amine 16 in good yield.
Furthermore, the N-protecting groups were readily
removed to quantitatively afford a hydrazinium
chloride 17, which has potential applications in
heterocycle synthesis.
Apparently, the overall reaction rate must be
closely associated with the radical hydrazination step
(from E to F). According to the above analysis, the
rate of this step could be calculated as follows:
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
Scheme 8. The preparation of 2-azido N-heterocycles.
In summary, we have developed an intermolecular
azido-hydrazination method for a broad range of
unactivated alkenes through a selective radical
addition sequence. According to the results of
mechanistic studies, each of the elementary steps in
this cascade is kinetically a fast process, while the
effective generation of azide radical is crucial to an
efficient transformation. To the best of our
knowledge, fluorenone has been, for the first time,
discovered as a powerful catalyst to generate azide
radicals under visible light. As a type of novel
bifunctional compounds, the azido-hydrazination
products have showcased great synthetic values in the
preparation of 1,2-diamines, β-azidoamines, β-
aminohydrazines, and related N-heterocycles. Most
prominently, this catalytic method was successfully
applied in a concise synthesis of an anti-cancer drug
ibrutinib.
Consequently, hydrazinium salt 17 was treated
with a few dual electrophiles to construct different
heterocycles (Scheme 8). A pyridazine-3,6-dione 18
was smoothly formed by reacting 17 with maleic
anhydride. 1,3-Dicarbonyl compounds are also
excellent reaction partners in the construction of
pyrazole derivatives, which are found in a variety of
pesticides
as
fungicides,
insecticides,
and
herbicides.[34] Acetylacetone readily cyclized with 17
under weakly basic conditions to give a dimethyl
pyrazole 19, whereas ethyl acetoacetate and the
hydrazine reacted regio-selectively, furnishing a
pyrazolone in practical yield.[35] Notably, all the
above cyclization products were obtained as single
diastereomers without epimerization.
Experimental Section
General procedure for the visible light-enabled azido-
hydrazination: To a flame-dried sealable 2-dram vial
equipped with
a stir bar were added di-tert-butyl
azodicarboxylate 2 (130 mg, 0.56 mmol) and 9-fluorenone
(7.2 mg, 0.04 mmol). After the vial was evacuated and
backfilled with N2 three times, anhydrous CH2Cl2 (1.6 mL)
and CH3CN (0.4 mL) were added. Then, an alkene 1 (0.4
mmol), TMSN3 (74 μL, 0.56 mmol) and methanol (16 μL,
0.4 mmol) were added to the solution successively through
syringes. The reaction was then stirred under the
irradiation of blue light (6 W LEDs, the photoreactor was
maintained at between 35 to 40 ˚C by circulating water
cooling). The reaction process was monitored with TLC.
Upon the completion of reaction, the resulted mixture was
quenched with saturated NaHCO3 (aq., 2 mL), extracted
with CH2Cl2 (2 mL × 3), dried over anhydrous Na2SO4 and
concentrated in vacuo to afford the crude product, which
was then purified through a silica gel flash column to
afford a corresponding product 3.
Scheme 9. The synthesis of ibrutinib. Conditions: (i) Pd/C
(10 wt%), H2 balloon, EtOH, 8 h; (ii) CbzCl (2 equiv),
NEt3 (3 equiv), 0 ˚C to rt, 3 h; (iii) HCl (4 M in MeOH), 10
h; (iv) 24 (1.2 equiv), MeOH, 50 ˚C, 1 h; (v) HCONH2,
165 ˚C, 5 h; (vi) HBr, HOAc, 0 ˚C, 0.5 h; (vii) Acryloyl
chloride (1.0 equiv), NEt3 (1.2 equiv), CH2Cl2, 0 ˚C to rt,
0.5 h.
More detailed experimental procedures for reaction
optimization, azido-hydrazination of different alkenes,
mechanistic studies, and new compounds characterizations
can be found in the Supporting Information.
Encouraged by these rudimentary synthetic
applications of azidohydrazines, we envisaged
applying this new methodology to the synthesis of
ibrutinib, which is a prescription medication used to
treat patients with mantle cell lymphoma.[3] The
current preparations of ibrutinib were performed
through the separate making of two heterocycle
segments and then coupled through an SN2
substitution.[36] Starting with the β-azido hydrazine 3l
obtained in Table 2 (entry 12), the reduction of the
Acknowledgements
Financial support from Huazhong University of Science &
Technology and the Natural Science Foundation of Hubei
Committee (2016CFA001) are gratefully acknowledged. We also
thank Huazhong University of Science & Technology Analytical
& Testing Center for characterizing new compounds.
azide
group
directly
afforded
a
3-
hydrazinylpiperidine tosylate 20, which could be used
without purification. After routine protection-
deprotection steps, a hydrazinium salt 21 was purified
as a white solid (63% yield over 3 steps). The
heterocycle moiety was then forged through a 2-step
procedure to furnish 22 in 61% yield.[37] Finally,
switching the N-Cbz group to acryloyl rendered the
target molecule (±)-ibrutinib 23 (Scheme 9).
References
[1] E. F. Rothgery, Hydrazine and Its Derivatives. in Kirk
‐ Othmer Encyclopedia of Chemical Technology,
Wiley: New York, 2004; Vol. 13, pp 562−607.
[2] U. Ragnarsson, Chem. Soc. Rev. 2001, 30, 205-213.
[3] J. A. Burger, J. J. Buggy, Leuk. Lymphoma 2013, 54,
2385-2391.
[4] J. D. Clark, M. E. Flanagan, J.-B. Telliez, J. Med.
Chem. 2014, 57, 5023-5038.
Conclusion
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
[5] a) S. Fletcher, Org. Chem. Front. 2015, 2, 739-752; b)
K. C. K. Swamy, N. N. B. Kumar, E. Balaraman, K. V.
P. P. Kumar, Chem. Rev. 2009, 109, 2551-2651.
[16] For perspectives on polarity-reversed radical addition
cascades and selected recent publications, see: a) E.
Godineau, Y. Landais, Chem. – Eur. J. 2009, 15, 3044-
3055; b) J. M. Tedder, Angew. Chem. Int. Ed. 1982, 21,
401-410; c) D. Zheng, A. Studer, Org. Lett. 2019, 21,
325-329; d) J. Chen, S. Zhu, J. Qin, L. Chu, Chem.
Commun. 2019, 55, 2336-2339; e) J. M. Lear, J. Q.
Buquoi, X. Gu, K. Pan, D. N. Mustafa, D. A. Nagib,
Chem. Commun. 2019, 55, 8820-8823; f) Z. Liu, Z.-Q.
Liu, Org. Lett. 2017, 19, 5649-5652.
[6] C. J. Moody, in Advances in Heterocyclic Chemistry,
Vol. 30 (Ed.: A. R. Katritzky), Academic Press, 1982,
pp. 1-45.
[7] V. Nair, A. T. Biju, S. C. Mathew, B. P. Babu, Chem.
Asian J. 2008, 3, 810-820.
[8] a) R. Dahl, K. K. Baldridge, N. S. Finney, Synthesis
2010, 2010, 2292-2296; b) C. C. Cheng, C. A. Seymour,
M. A. Petti, F. D. Greene, J. F. Blount, J. Org. Chem.
1984, 49, 2910-2916; c) J. H. Hall, M. L. Jones, J. Org.
Chem. 1983, 48, 822-826; d) G. B. Butler, S. R. Turner,
L. J. Guilbault, J. Org. Chem. 1971, 36, 2838-2840.
[17] S. Yang, S. Zhu, D. Lu, Y. Gong, Org. Lett. 2019, 21,
2019-2024.
[18] For details, see the supporting information.
[19] For selected reviews and recent examples on alkene
diamination, see: a) K. Muñiz, C. Martínez, J. Org.
Chem. 2013, 78, 2168-2174; b) S. De Jong, D. G.
Nosal, D. J. Wardrop, Tetrahedron 2012, 68, 4067-
4105; c) D. Wu, S.-S. Cui, Y. Lin, L. Li, W. Yu, J. Org.
Chem. 2019, 84, 10978-10989; d) K. Shen, Q. Wang, J.
Am. Chem. Soc. 2017, 139, 13110-13116; e) Q. Qin,
Y.-Y. Han, Y.-Y. Jiao, Y. He, S. Yu, Org. Lett. 2017,
19, 2909-2912; and the references cited there.
[9] a) S. Yamazaki, Y. Maenaka, K. Fujinami, Y. Mikata,
RSC Adv. 2012, 2, 8095-8103; b) P. Galzerano, F.
Pesciaioli, A. Mazzanti, G. Bartoli, P. Melchiorre,
Angew. Chem. Int. Ed. 2009, 48, 7892-7894; c) A.
Kamimura, Y. Gunjigake, H. Mitsudera, S. Yokoyama,
Tetrahedron Lett. 1998, 39, 7323-7324.
[10] a) N. Derrien, J. S. Sharley, A. E. Rubtsov, A. V.
Malkov, Org. Lett. 2017, 19, 234-237; b) M. A.
Brimble, C. K. Y. Lee, Tetrahedron: Asymmetry 1998,
9, 873-884; c) Y. Leblanc, R. Zamboni, M. A.
[20] M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108,
2952-3015.
Bernstein, J. Org. Chem. 1991, 56, 1971-1972; d) H. M. [21] D. Intrieri, P. Zardi, A. Caselli, E. Gallo, Chem.
R. Hoffmann, Angew. Chem. Int. Ed. 1969, 8, 556-577.
Commun. 2014, 50, 11440-11453.
[11] a) A. Hu, J.-J. Guo, H. Pan, Z. Zuo, Science 2018,
361, 668-672; b) A. Hu, J.-J. Guo, H. Pan, H. Tang, Z.
Gao, Z. Zuo, J. Am. Chem. Soc. 2018, 140, 1612-1616;
c) I. Ryu, A. Tani, T. Fukuyama, D. Ravelli, S.
Montanaro, M. Fagnoni, Org. Lett. 2013, 15, 2554-
2557; d) Y. Amaoka, S. Kamijo, T. Hoshikawa, M.
Inoue, J. Org. Chem. 2012, 77, 9959-9969.
[22] For more featured transformations of azide group, see
reviews: a) D. Huang, G. Yan, Adv. Synth. Catal. 2017,
359, 1600-1619; b) E. F. V. Scriven, K. Turnbull,
Chem. Rev. 1988, 88, 297-368.
[23] a) Y. Chen, T. Tian, Z. Li, Org. Chem. Front. 2019, 6,
632-636; b) A. Hossain, A. Vidyasagar, C. Eichinger, C.
Lankes, J. Phan, J. Rehbein, O. Reiser, Angew. Chem.
Int. Ed. 2018, 57, 8288-8292; c) N. Fu, G. S. Sauer, A.
Saha, A. Loo, S. Lin, Science 2017, 357, 575-579; d)
Y.-A. Yuan, D.-F. Lu, Y.-R. Chen, H. Xu, Angew.
Chem. Int. Ed. 2016, 55, 534-538; e) X. Sun, X. Li, S.
Song, Y. Zhu, Y.-F. Liang, N. Jiao, J. Am. Chem. Soc.
2015, 137, 6059-6066.
[12] a) V. R. Yatham, P. Bellotti, B. König, Chem.
Commun. 2019, 55, 3489-3492; b) M.-J. Zhang, G. M.
Schroeder, Y.-H. He, Z. Guan, RSC Adv. 2016, 6,
96693-96699; c) S. B. Lang, K. C. Cartwright, R. S.
Welter, T. M. Locascio, J. A. Tunge, Eur. J. Org. Chem.
2016, 2016, 3331-3334.
[13] a) J. Davies, N. S. Sheikh, D. Leonori, Angew. Chem.
Int. Ed. 2017, 56, 13361-13365; b) D. Wang, R. Ren, C.
Zhu, J. Org. Chem. 2016, 81, 8043-8049; c) X.-Y.
Duan, X.-L. Yang, R. Fang, X.-X. Peng, W. Yu, B.
Han, J. Org. Chem. 2013, 78, 10692-10704; d) M. K.
Zhu, Y. C. Chen, T. P. Loh, Chem. Eur. J. 2013, 19,
5250-5254; e) V. A. Schmidt, E. J. Alexanian, J. Am.
Chem. Soc. 2011, 133, 11402-11405.
[24] a) H. Li, S.-J. Shen, C.-L. Zhu, H. Xu, J. Am. Chem.
Soc. 2019, 141, 9415-9421; b) J. M. Lear, J. Q. Buquoi,
X. Gu, K. Pan, D. N. Mustafa, D. A. Nagib, Chemical
Communications 2019, 55, 8820-8823; c) S. Alazet, J.
Preindl, R. Simonet-Davin, S. Nicolai, A. Nanchen, T.
Meyer, J. Waser, J. Org. Chem. 2018, 83, 12334-12356.
[25] a) S.-J. Shen, C.-L. Zhu, D.-F. Lu, H. Xu, ACS Catal.
2018, 8, 4473-4482; b) H. Zhou, W. Jian, B. Qian, C.
Ye, D. Li, J. Zhou, H. Bao, Org. Lett. 2017, 19, 6120-
6123.
[14] a) J. Waser, B. Gaspar, H. Nambu, E. M. Carreira, J.
Am. Chem. Soc. 2006, 128, 11693-11712; b) J. Waser, J.
C. González-Gómez, H. Nambu, P. Huber, E. M.
Carreira, Org. Lett. 2005, 7, 4249-4252; c) J. Waser, E.
M. Carreira, J. Am. Chem. Soc. 2004, 126, 5676-5677;
d) J. Waser, E. M. Carreira, Angew. Chem. Int. Ed.
2004, 43, 4099-4102.
[26] J. C. Siu, G. S. Sauer, A. Saha, R. L. Macey, N. Fu, T.
Chauviré, K. M. Lancaster, S. Lin, J. Am. Chem. Soc.
2018, 140, 12511-12520.
[27] For a review on visible light photocatalysis enabled
olefin difunctionalizations, see: M.-Y. Cao, X. Ren, Z.
Lu, Tetrahedron Lett. 2015, 56, 3732-3742.
[15] Y. Zhang, C. Huang, X. Lin, Q. Hu, B. Hu, Y. Zhou,
G. Zhu, Org. Lett. 2019, 21, 2261-2264.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
[28] a) J.-J. Wang, W. Yu, Chem. Eur. J. 2019, 25, 3510-
3514; b) J. Chen, S. Zhu, J. Qin, L. Chu, Chem.
Commun. 2019, 55, 2336-2339; c) T. R. Reddy, D. S.
Rao, S. Kashyap, Chem. Commun. 2019, 55, 2833-
2836; d) W. Wei, H. Cui, H. Yue, D. Yang, Green
Chem. 2018, 20, 3197-3202; e) B. Yang, X. Ren, X.
Shen, T. Li, Z. Lu, Chin. J. Chem. 2018, 36, 1017-
1023; f) B. Yang, Z. Lu, ACS Catal. 2017, 7, 8362-
8365.
Maillard, J. C. Walton, J. Chem. Soc., Perkin Trans. 2
1985, 443-450.
[33] For instances: 1) Compared to a conjugated alkene,
the radical adduct E of an isolated alkene lacks
stabilization effect (Keq is small), thus undergoes a
slower reaction; 2) The photocatalytic method furnishes
higher concentration of ∙N3 than benziodoxole 4e does,
as a consequence, enables faster reactions; 3) As the
concentration of reactants decrease, the reactions
become sluggish during their late stage (Scheme 2a).
So, for isolated alkenes with a smaller Keq, an
additional portion of reagents are usually demanded to
achieve good conversions.
[29] H.-T. Zhu, L. Arosio, R. Villa, M. Nebuloni, H. Xu,
Org. Process Res. Dev. 2017, 21, 2068-2072.
[30] a) B. J. Fitzsimmons, Y. Leblanc, J. Rokach, J. Am.
Chem. Soc. 1987, 109, 285-286; b) Y. Leblanc, B. J.
Fitzsimmons, J. P. Springer, J. Rokach, J. Am. Chem.
Soc. 1989, 111, 2995-3000.
[34] A. Ansari, A. Ali, M. Asif, Shamsuzzaman, New J.
Chem. 2017, 41, 16-41.
[31] a) T. Berndt, S. Richters, T. Jokinen, N. Hyttinen, T.
Kurtén, R. V. Otkjær, H. G. Kjaergaard, F. Stratmann,
H. Herrmann, M. Sipilä, M. Kulmala, M. Ehn, Nat.
Commun. 2016, 7, 13677; b) M. E. Davis, P. S. Stevens,
Atmos. Environ. 2005, 39, 1765-1774; c) L. Vereecken,
J. Peeters, J. Phys. Chem. A 2000, 104, 11140-11146.
[35] The reaction between hydrazinium salt 17 and ethyl
acetoacetate is described in the supporting information.
[36] US patent: US20160264584A1
[37] A. Kumar, Y. Wang, X. Lin, G. Sun, K. Parang,
ChemMedChem 2007, 2, 1346-1360.
[32] a) A. G. Griesbeck, J. Lex, K. M. Saygin, J.
Steinwascher, Chem. Commun. 2000, 2205-2206; b) B.
10
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901041
Advanced Synthesis & Catalysis
FULL PAPER
Catalytic Azido-Hydrazination of Alkenes Enabled
by Visible Light: Mechanistic Studies and
Synthetic Applications
Adv. Synth. Catal. Year, Volume, Page – Page
Peng Wang, Yunxuan Luo, Songsong Zhu, Dengfu
Lu* and Yuefa Gong*
11
This article is protected by copyright. All rights reserved.