A safe and facile route to imidazole-1-sulfonyl azide as a diazotransfer reagent

Article abstract of DOI:10.1021/ol3028708

A facile approach to the diazotransfer reagent of imidazole-1-sulfonyl azide was reported. The procedure was well optimized to clarify potential explosion risks. A high production yield as well as small batch variation was achieved even without careful pretreatment of reagents and solvents. HPLC and NMR methods to monitor the process were provided. These features made this protocol suitable for large scale preparation in academia and industry as well.

Full text of DOI:10.1021/ol3028708

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
A Safe and Facile Route to
Imidazole-1-sulfonyl Azide as a
Diazotransfer Reagent
000–000
†
,‡,#
†,‡,#
†,‡,#
†,‡
Yonghui Liu, Haixin Yuan,
§
Hui Ye,
Weikang Guo, Lifei Zhou, Xuefeng Cao, Hongqi Tian, Jie Shen,* and
Ruihua Liu,
Dongmei Li,
†
,‡
†,‡
†,‡
,†,‡
,
†,‡
Peng George Wang*
State Key Laboratory of Medicinal Chemical Biology and College of Pharmacy,
Nankai University, Tianjin 300071, P. R. China, Tianjin Key Laboratory of Molecular
Drug Research, Nankai University, Tianjin 300071, P. R. China, Tianjin Chase Sun
Pharmaceutical Co., Ltd., Tianjin 300026, P. R. China, and School of Biological
Engineering, Tianjin University of Science and Technology, Tianjin, 300457, P. R. China
jieshen@nankai.edu.cn; pwang@nankai.edu.cn
Received October 18, 2012
ABSTRACT
A facile approach to the diazotransfer reagent of imidazole-1-sulfonyl azide was reported. The procedure was well optimized to clarify potential
explosion risks. A high production yield as well as small batch variation was achieved even without careful pretreatment of reagents and solvents.
HPLC and NMR methods to monitor the process were provided. These features made this protocol suitable for large scale preparation in academia
and industry as well.
Thescientificcommunityhasbeenwitnessing abooming
1
could convert primary amines into the corresponding
2À7
era for azideÀalkyne Huisgen cycloaddition reactions.
azides.
A key challenge in this area is to develop a safe
Organic azides, which in traditional organic syntheses had
mainly been considered as masked amines, became key
intermediates for click chemistry in a wealth of recent
publications. Accordingly, attention was refocused toward
modifying current methodologies to prepare organic azides,
including efforts to improve diazotransfer reactions which
and facile approach to the synthesis of a highly efficient
diazotransfer reagent.
4
Initially reported in 1972 by Shiner’s group and later
5
,6
optimized by Wong’s group, the TfN -driven diazotransfer
3
reaction has been widely applied in academia (Scheme 1).
There are two problems, however, for the preparation of
6
the TfN reagent: a large batch variation in yields and
3
4
explosive risks. Strategies were proposed to overcome
†
State Key Laboratory of Medicinal Chemical Biology and College of
Pharmacy, Nankai University.
‡
Tianjin Key Laboratory of Molecular Drug Research, Nankai University.
Tianjin Chase Sun Pharmaceutical Co., Ltd.
§
(4) Cavender, C. J.; Shiner, V. J., Jr. J. Org. Chem. 1972, 37, 3567–
Tianjin University of Science and Technology.
These authors contributed equally to this work.
1) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
001, 40, 2004.
2) Kauer, J. C.; Carboni, R. A. J. Am. Chem. Soc. 1967, 89, 2633–
637.
3) Anselme, J. P.; Fischer, W. Tetrahedron 1969, 25, 855–9.
3569.
(5) Alper, P. B.; Hung, S.-C.; Wong, C. H. Tetrahedron Lett. 1996, 37,
6029–6032.
(6) Nyffeler, P. T.; Liang, C.-H.; Koeller, K. M.; Wong, C.-H. J. Am.
Chem. Soc. 2002, 124, 10773–10778.
#
(
2
(
2
(7) Goddard-Borger, E. D.; Stick, R. V. Org. Lett. 2007, 9, 3797–
3800.
(
1
0.1021/ol3028708 r XXXX American Chemical Society

these defects, including fresh distillation of Tf O, utiliza-
2
all engaged compound 1 as a key intermediate. While we
followed the published protocol to prepare intermediate
1 at 10 g scale, bubbles of hydrochloride were observed
tion of greatly excessive NaN , and direct employment of
3
the in situ generated TfN solution which could be mon-
3
1
9
itoredby F NMR. Althougheffective, these inconvenient
manipulations actually brought one solution while creat-
ing another problem. These inherent defects, particularly
the safety concerns, make Wong’s protocol rarely employed
in scale-up preparations.
which would react with NaN to generate toxic and
3
explosive hydrazoic acid (HN ). In fact, complete removal
3
of hydrochloride from sulfonyl chloride is hardly accom-
plished, especially in a scale-up reaction. Even worse,
under almost exactly the same conditions, a recent pub-
lication indicated that an extremely explosive molecule,
sulfuryl diazide (N ÀSO ÀN , compound 4, Scheme 1),
3
2
1
3
1
could also be generated. These concerns were verified
Scheme 1. Published Protocols To Prepare Diazotransfer
Reagents
1
2
a
by an errant announcement in 2011. Besides, there is
another basic argument: is a reagent containing a sulfonyl
azide bond really suitable for long-term shelf storage?
As anticipated, our computational analysis (Scheme 2)
demonstrated that hydrolysis of the sulfonyl azide bond
was greatly favored thermodynamically. Although some
sulfonyl azides are nonexplosive and nonhydroscopic (e.g.,
1
0
the last two compounds in Figure 1), these kinetically
stable compounds are similar to caged tigers which might
be released by unpredictable factors over long-term
storage. With all these concerns, an updated protocol
is desired to meet the following criteria:
(a) It should avoid the coexistence of NaN3 with
strong acid species.
Scheme 2. Computational Studies on Standard Free Energy
Change of Sulfonyl Azide Bond Hydrolysis
a
a
Note: the blue and red dashed boxes indicated advantages and
disadvantages respectively of the two protocols.
In 2007, Goddard-Borger’s group discovered a new dia-
zotransfer reagent, imidazole-1-sulfonyl azide (ImÀSO ÀN ,
2
3
7
compound 2 in Scheme 1). While maintaining good
diazotransfer activity, this new reagent demonstrated a
reduced tendency for explosion than TfN and could even
3
a
Note: Calculations were carried out with the B3LYP method using
the Gaussian03 program. Detailed descriptions are provided in the
Supporting Information.
be isolated as a solid of the chloride salt. Additionally,
structual modifications ofcompound2 or its salt have been
8À10
investigated.
Listed in Figure 1 are the newly reported
(
(
(
b) It must avoid the employment of excessive NaN3.
c) It should avoid the formation of N ÀSO ÀN .
3
2
3
d) It must utilize a safe and stable reagent, as the
storage compound can be conveniently and effi-
ciently converted into a diazotransfer reagent.
Similar to Wong’s protocol, the generated diazo-
transfer reagent should be utilized in situ, and a
bench accessible monitoring method, which can be
routinely performed by most laboratories, should
be established.
Figure 1. Recently published diazotransfer reagents with
improved properties.
8À10
solid diazotransfer reagents with improved chemical
properties or an increased safety threshold. Neverthe-
less, the synthetic procedures were not optimized, which
Under these guidelines, a safe and facile protocol to
synthesize diazotransfer reagent 2 was described in this
paper (eq 1 in Scheme 3).
To avoid the coexistence of a strong acid with NaN , we
3
(
8) Laus, G.; Adamer, V.; Hummel, M.; Kahlenberg, V.; Wurst, K.;
Nerdinger, S.; Schottenberger, H. Crystals 2012, 2, 118–126.
9) Katritzky, A. R.; Khatib, M. E.; Bol’shakov, O.; Khelashvili, L.;
Steel, P. J. J. Org. Chem. 2010, 75, 6532–6539.
10) Fischer, N.; Goddard-Borger, E. D.; Greiner, R.; Klapotke,
first attempted using sulfuryl diimidazole (compound 5) as
(
(11) Zeng, X. Q.; Beckers, H.; Bernhardt, E.; Willner, H. Inorg.
Chem. 2011, 50, 8679–8684.
(
T. M.; Skelton, B. W.; Stierstorfor, J. J. Org. Chem. 2012, 77, 1760–1764.
(12) Goddard-Borger, E. D.; Stick, R. V. Org. Lett. 2011, 13, 2514.
B
Org. Lett., Vol. XX, No. XX, XXXX

a mild nucleofuge to react with NaN at ambient tempera-
3
6 to 2, NMR analysis indicated that the content of 2 in
EtOAc solution did not changeafter it was stirred with 1 M
ture. However, no reaction was observed under such
conditions (eq 2 in Scheme 3). Then, we considered com-
1
5
aqueous NaN solution at 0 °C for 2 h. Then, the only
3
1
3
pound 6 as a possibility, as it is a more reactive nucleo-
fuge than 5. In preliminary experiments, 6 showed high
chance to form 4 in our synthetic route is through the
formation of a possible dimethylated side product, com-
1
3
pound 7 (eq 5). Consistent with previous studies, this
possible side product was not detected by NMR in our
experiment either. In case a trace amount of intermediate 7
did form, an extra precaution was designed in the second
a
Scheme 3. Improved Protocol to the Synthesis of Reagent 2
step of eq 1. That is, prior to the addition of NaN , the inter-
3
mediate 6 alone was stirred in cold water for 0.5 h. We
already proved that the majority of 6 could survive such a
manipulation. We presumed that 7, if it existed, should have
quickly decomposed in cold water due to its high reactivity.
Finally, an HPLC method, utilizing UV detection, was
established to quantify the in situ generated diazotransfer
a
Note: The second reaction in eq 5 and the reaction in eq 6 were
putative results inferred from chemistry knowledge, rather than verified
by experiments.
Figure 2. HPLC analysis of in situ generated diazotransfer
reagent 2 and the byproduct N-methyl imidazole. Detection:
UV absorption at 220 nm. Column: Agela Venusil MP C18
(
Flow rate: 1 mL/min.
4.6 mm  250 mm, 5 μm) . Mobile phase: MeOH/H O = 37/63.
2
1
solubility and high stability in cold water (eq 3). We
4
therefore attempted the homogeneous reaction between 6
and NaN in water at 0 °C. Excitingly, the reaction was
quickly accomplished within 1 h. During the whole pro-
cess, the medium is nonacidic, and the employment of
3
reagent 2 in EtOAc solution (Figure 2). This method was
verified by NMR analysis by using isopropyl benzoate as
1
6
an inert and nonvolatile internal standard. For our two-
step protocol (eq 1 in Scheme 3), the yield of reagent 2, by
using compound 5 as the limiting reagent, was consistently
around 88% in multiple batches. Unlike Wong’s protocol,
purification of the key reagent MeOTf proved to be
unnecessary to maintian the consistency of yield. When
the protocol was scaled up to over the 100 g level, the yield
did not dramatically change if the solution in step b was
thoroughly stirred. HPLC analysis, as well as the NMR
data, indicated the existence of side product N-methyl
imidazole as a sole impurity in the final EtOAc solution
of reagent 2. We proved that this impurity did not affect
the diazotransfer reaction of the glucosamine (Scheme 4)
reactant.
largely excessive NaN was unnecessary. Moreover, the
3
product of reagent 2 could almost be completely extracted
from the aqueous phase by adding an equal volume of
EtOAc solvent. This means, besides providing a high yield
via convenient manipulations, this reaction could avoid
not only the formation of the explosive HN molecule but
3
also the existence of a large quantity of NaN in reaction
3
waste. Thus, from 0.8 M compound 6 in H O, 0.7 M
2
reagent 2 in EtOAc solution was obtained.
Another safety concern for Goddard-Borger’s protocol
isthe generation of explosivecompound4. Considering the
inertness of 5 toward NaN (eq 2), we presumed that
3
reagent 2 should also be reluctant to react with NaN3.
(
eq 4) Indeed, under similar conditions used to convert
(
13) Beaudoin, S.; Kinsey, K. E.; Burns, J. F. J. Org. Chem. 2003, 68,
15–119.
14) The hydrolysis process was monitored by NMR, and the experi-
mental data were summarized in the Supporting Information. Caution:
The final pH value for the second reaction in eq 1 was between 7 to 8,
1
Scheme 4. Diazotransfer Reaction of D-Glucosamine
Hydrochloride by Utilizing in Situ Generated Reagent 2
(
3
implying unlikely probability for HN formation. Additionally, com-
pound 2 was not detected in the aqueous phase by NMR analysis at the
end of the conversion. Nevertheless, careful disposal of the aqueous waste
with NaClO solution, as well as the strict compliance with our optimized
protocol, is necessary to avoid the occurrence of an explosion.
Org. Lett., Vol. XX, No. XX, XXXX
C

In summary, a facile and safe protocol (eq 1 in Scheme 3)
to prepare diazotransfer reagent 2 was introduced. The
stable compound sulfuryl diimidazole (5), which could be
prepared from cheap reagents and purified by crystal-
lization in good yield, was employed as a shelf-storage
compound. From this compound, via a two-step process,
the diazotransfer reagent 2 could be prepared in less than 5
h with a consistently high yield. During the whole process,
there was no chromatography purificaiton, no need to
carefully purify involved reagents, and little chance to
generate explosive HN and N ÀSO ÀN molecules. To
3
3
2
3
the best of our knowledge, it was the first safe preparation
of a sulfonyl azide derivative as a diazotransfer reagent at
over the 100 g level. It demonstrated the feasibility of our
ptotocol to be applied to large scale preparation in both
academia and industry.
Acknowledgment. This study was supported by the
National 973 Basic Research Program of China (No.
2010CB529100) and National Natural Science Founda-
tion of China (No. 21072105).
(
15) For detail experimental procedures, please refer to the Support-
ing Information.
16) The NMR method for measuring the yield of reagent 2 was
summarized in Supporting Information.
17) Caution: sulfonyl azides are potentially explosive compounds,
Supporting Information Available. Full experimental
details, NMR spectra, and HPLC. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
(
(
and therefore full compliance with our optimized protocol should be
strictly held, especially for scale up preparation.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX