Safe and fast tetrazole formation in ionic liquids

Article abstract of DOI:10.1016/j.tet.2006.10.057

The [2+3] cycloaddition of nitriles and azides is reliable for intramolecular reactions, but the hazards with volatile azides in intermolecular reactions are tremendous. Zinc catalysis in aqueous solution is a magnificent improvement, but requires the removal of the zinc salts from the acidic product. Herein, we report safe solvents featuring low vapor pressure and good solubility of NaN₃. Ionic liquids based on alkylated imidazoles combined with microwave heating turned out to be a solution for the given tasks.

Full text of DOI:10.1016/j.tet.2006.10.057

Tetrahedron 63 (2007) 492–496
Safe and fast tetrazole formation in ionic liquids
*
Boris Schmidt, Daniela Meid and Daniel Kieser
Clemens Scho€pf-Institute for Organic Chemistry and Biochemistry, Darmstadt Technical University, Petersenstr. 22,
D-64287 Darmstadt, Germany
Received 17 May 2006; revised 18 October 2006; accepted 19 October 2006
Available online 13 November 2006
Abstract—The [2+3] cycloaddition of nitriles and azides is reliable for intramolecular reactions, but the hazards with volatile azides in
intermolecular reactions are tremendous. Zinc catalysis in aqueous solution is a magnificent improvement, but requires the removal of the
zinc salts from the acidic product. Herein, we report safe solvents featuring low vapor pressure and good solubility of NaN3. Ionic liquids
based on alkylated imidazoles combined with microwave heating turned out to be a solution for the given tasks.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
compared them to the methods published by Amantini,7
Sharpless6 and Hallberg.9 Commercial ionic liquids10,11
(IL) based on alkylated imidazoles turned out to be aversatile
solution for the given tasks. Two nitriles (1 and 2, Scheme 1)
were selected to explore the influence of the methylimidazo-
lium substituents: methyl, butyl, hexyl and octyl. It turned
out that these substituents have only minor impact on the
outcome of the reaction (Table 1, entries 6–8)12 in compar-
ison to the contribution of the acid/azide ratio. But there is
hidden value in these alkyl chains. The more lipophilic hexyl
and octyl imidazoles are less miscible with water and allow
easier aqueous extraction of the products, which simplifies
the product isolation and improves the isolated yields. The
counter anion of the ionic liquids was varied from: Clꢁ,
Brꢁ, ꢁSO3CF3 to ꢁPF3(C2CF5)3 and have a higher impact.
There was no general trend observed for the two halides,
but the halides are far better than triflate or the perfluorinated
phosphate (Table 1, entries 28 and 29). The ionic liquids
1H- and 2H-Tetrazoles are regarded as isosteric replace-
ments1 of carboxylic acids with improved properties in
drug metabolism and pharmacokinetics. Thus, they are
frequently employed in the lead optimization of ethical
drug candidates to enhance the oral bioavailability. Several
successful examples of this strategy are present in the sartane
drug family, which is used to treat hypertension.2,3 The [2+3]
cycloaddition of nitriles and azides is a reliable method for
intramolecular cases, but volatile azides create tremendous
hazards in intermolecular reactions. Large-scale processes
solve some of the safety issues by using non-volatile tin
azides, but introduce the problem of tin recovery.4,5 The
‘click’ chemistry approach utilizing6 zinc catalysis in aque-
ous solution is a magnificent improvement over previous
methods, but sometimes still requires the tedious removal
of zinc salts from the acidic products. Moreover, the reaction
rates at 100 ꢀC are often insufficient for bulk intermediates.
The economic ‘click’ processes in water require higher tem-
peratures and pressurized vessels. The classical approach
utilizes sodium azide and trialkylamines in refluxing tolu-
ene, but is limited to the laboratory scale because of volatile
ammonium azides, which are prone to sublimation. Further-
more, toluene is not a good solvent for NaN3, which results
in slow turnover, azide deposits and has thus to be regarded
as an unsafe, poor alternative. The solvent free conversion by
tetrabutylammonium fluoride and trimethyl azide7 is very
attractive for small-scale experiments, but the exothermic
azide formation will be hard to control in upscaled reac-
tions.8 We set out to investigate safer solvents featuring
a low vapor pressure and good solubility of NaN3 and
R
R
1) NaN3
2) HCl
R
N
N
X
CN
N
HN
N
N
3 R = OMe
4 R = NO2
1 R = OMe
2 R = NO2
1) NaN3
2) HCl
N
N
N
N
N
OMIM-Cl
130°C 48 h
CN
N
HN
91%
5
6
N
* Corresponding author. Tel.: +49 6151 163075; fax: +49 6151 163278;
e-mail: schmidt_boris@t-online.de
Scheme 1. Tetrazole formation from commercial nitriles.
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.10.057

B. Schmidt et al. / Tetrahedron 63 (2007) 492–496
493
Table 1. Tetrazole formation in different ionic liquids at varying temperatures and stoichiometric ratios
Entry
Substrate or product
NaN3
(equiv)
HOAc
(equiv)
Ionic liquid
T/time
Turnover
(%)
Yield
(%)
1
2
3
4
5
6
7
8
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Methoxybenzonitrile (1)
4-Nitrobenzonitrile (2)
4-Nitrobenzonitrile (2)
4-Nitrobenzonitrile (2)
4-Nitrobenzonitrile (2)
2-Pyrazylcarbonitrile (5)
40-Methyl-2-biphenylcarbonitrile (7)
40-Methyl-2-biphenylcarbonitrile (7)
40-Methyl-2-biphenylcarbonitrile (7)
40-methyl-2-biphenylcarbonitrile (7)
40-Methyl-2-biphenylcarbonitrile (7) (10 mmol)
40-Methyl-2-biphenylcarbonitrile (7)
40-Methyl-2-biphenylcarbonitrile (7)
40-Methyl-2-biphenylcarbonitrile (7)
8 (0.75 mmol)
8 (0.75 mmol)
8 (0.75 mmol)
8 (0.75 mmol)
8 (0.75 mmol)
8 (0.75 mmol)
8 (0.75 mmol)
9
7
10
5
11
1.1
1.1
2.2
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
2
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
0.7
1.5
1.5
1.5
1.5
0.5
0.5
1.5
1.5
1.5
1.5
0.5
0.7
0.2
0.7
0.7
0.7
0.7
0.7
0.7
0.7
1
[BMIM]Br
[BMIM]Cl
[HMIM]Cl
[OMIM]Cl
[BMIM]Br
[BMIM]Cl
[HMIM]Cl
[OMIM]Cl
[OMIM]Cl
[BMIM]Br
[MMIM]Cl
[HMIM]Cl
[OMIM]Cl
[OMIM]Cl
[MMIM]Cl
[BMIM]Br
[BMIM]Cl
[HMIM]Cl
[OMIM]Cl
[OMIM]Cl
[BMIM]Cl
[OMIM]Cl
[BMIM]Cl
[BMIM]Cl
[BMIM]Cl
[BMIM]Br
[HMIM]Cl
[BMIM]SO3CF3
[HMIM]PF3(C2F5)3
[BMIM]Cl
[BMIM]Cl
[BMIM]Cl
[BMIM]Cl
[BMIM]Cl
[BMIM]Cl
70 ꢀC/38 h
70 ꢀC/38 h
70 ꢀC/38 h
70 ꢀC/38 h
170 ꢀC/24 h
170 ꢀC/24 h
170 ꢀC/24 h
170 ꢀC/24 h
170 ꢀC/24 h
70 ꢀC/72 h
70 ꢀC/72 h
70 ꢀC/72 h
70 ꢀC/72 h
130 ꢀC/48 h
140 ꢀC/48 h
170 ꢀC/48 h
170 ꢀC/48 h
170 ꢀC/48 h
170 ꢀC/48 h
170 ꢀC/24 h
200 W/1 h
62
28
87
79
98
89
95
97
98
97
94
72
97
>95
93
70
62
71
>95
100
85
44
66
17
43
56
76
94
82
91
94
98
91
89
60
89
91
78
12
1
35
71
59
80
18
57
77
73
48
73
6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
2
1.1
1.1
1.1
1.1
2
3
2
3
3
3
3
3
3
3
3
3
3
170 ꢀC/24 h
140 ꢀC/8 h
140 ꢀC/15 h
140 ꢀC/24 h
140 ꢀC/24 h
140 ꢀC/24 h
140 ꢀC/24 h
140 ꢀC/24 h
91
100
100
85
26
0
100
58
100
90
0
170/0.5 h, mw
89a
40a
97a
77a
26a
88a
1
1
1
1
170/0.5 h, mw
170/0.5 h, mw
170/0.5 h, mw
170/0.5 h, mw
170/0.5 h, mw
3
3
3
n.d.
100
8
1
[BMIM]: 1-butyl-3-methylimidazolium, [MMIM]: 1-methyl-3-methylimidazolium, [HMIM]: 1-hexyl-3-methylimidazolium, [OMIM]: 1-octyl-3-methylimid-
azolium. Concn: 0.5 M sample size 1 mL, if not noted otherwise. The turnover was controlled against internal stand, the optimized yields are isolated yields,
mw¼BiotageÒ InitiatorÔ microwave.
a
Crude yields determined by HPLC.
based on N-sulfonamide anions display sufficient vapor
pressure to allow distillation.13 Thus, they were excluded
from our investigation.
for comparison against the established methods from
Amantini and Sharpless.6,7 All compounds were isolated af-
ter aqueous work up, extraction by ethyl acetate or precipi-
tation, where appropriate. The synthesis in IL turned out to
be superior in isolated yield for 13 out of 20 products. Aman-
tini method provided the other seven products, e.g., 9 and 23
(Fig. 1), in better yields. However, this method requires
a hazardous combination of reagents. The Sharpless method
was sometimes even, but generally lacked in isolated yield.
The difficulty in isolation of water soluble products
(clog P<1) was a major cause for the poor yields of the com-
pounds 11, 24–26 (Fig. 1 and Scheme 2). We selected two
advanced intermediates (7+8, Scheme 3) from the synthesis
of sartane drugs2,3 to explore the applicability to functional-
ized substrates. The commercial biphenylnitrile 7 required
an elevated reaction temperature and careful optimization
of the stoichiometry to result in 14 with a yield of 78%. How-
ever, the necessary reaction time was rather long, which re-
sulted in significant byproduct formation, mainly through
product decomposition. However, the tetrazole formation
can be accelerated by microwave heating of DMF solutions.9
The desired rapid heating was reported for [BMIM]PF6. The
standard alkyl-[MIM] based ionic liquids do not feature
dipole moments suitable for rapid microwave heating,
thus the energy absorption is due to ionic conductivity.14,15
Initially, the temperature range had to be optimized to guar-
antee complete conversion within 24 h. This was achieved at
70 ꢀC with electron deficient substrates such as 2, which is
known to react readily during azide formation.6,8 The less
electron deficient pyrazine-2-carbonitrile 5 demanded
a higher temperature (130 ꢀC) and extended reaction time
of 48 h. The electron rich substrates such as 1 and 8 required
even higher reaction temperatures of 170 and 140 ꢀC,
respectively. Extended reaction time or higher reaction
temperatures resulted in decreased yields for all tetrazoles.
This was due to both product and solvent decompositions.
A crucial role was assigned to the amount of acetic acid
added. Large excess of HOAc versus NaN3 drove the volatile
HN3 out of the reaction mixtures and thus created a safety
issue. Equimolar or substoichiometric amount of acetic
acid in relation to sodium azide provided good yields, suffi-
cient reaction rates and economic utilization of NaN3.
The experiments summarized in Figure 1 provided the stan-
dard procedure (140 ꢀC, 0.5 mmol, 24 h in a screw capped
1 mL vial, 3 equiv NaN3, 1 equiv acid in 100 mg IL) in IL

494
B. Schmidt et al. / Tetrahedron 63 (2007) 492–496
100
90
80
70
60
50
40
30
20
10
0
Amantini et. al.
Sharpless et.al.
IL-Method
9
3
12
13
16
15
17
14
10
18
6
19
20
21
22
grouped by decreasing yield and increasing polarity
23
24
11
25
26
Figure 1. Comparative tetrazole synthesis (Scheme 2 for structure). IL-method: 140 ꢀC, 24 h.
Therefore, we investigated the heat rate of the selected pure
IL in comparison to pure DMF under otherwise identical
conditions to select suitable candidates. Up to 250 W of
microwave irradiation was required to achieve a similar
temperature profile as for DMF (Fig. 2).9 The depicted
[BMIM]Cl displayed an induction phase for the energy ab-
sorption and ionic conductivity, which increased signifi-
cantly above 60 ꢀC. Thus, it is due to the melting and the
onset of ionic conductivity. A similar effect was observed
for [BMIM]Br and [HMIM]Cl. Preliminary investigations
of the three ILs by differential scanning calorimetry did
not reveal any phase transitions or decomposition between
ꢁ100 and 200 ꢀC, except for the melting points of
[BMIM]Cl at 41 ꢀC (Fig. 3) and [BMIM]Br at 77 ꢀC. The
1
ILs were stable for one heat cycle as judged by H NMR
spectroscopic analysis. However, repeated heating (3ꢂ)
1
resulted in partial decomposition (15%) as judged by H
NMR spectroscopy. This thermal decomposition has been
reported previously.16,17
Regardless of the IL decomposition, the reaction time was
dramatically shortened using microwave heating at 200 W
N
HN
N
N
N
HN
N
N
N
N
N
NH
N
HN
N
N
N
N
HN
N
O
N
N
N
O
O
N
O
O
11 CLogP: 0.09
9 CLogP: 1.74 10 CLogP: -0.66
3 CLogP: 1.55
6 CLogP: -0.7
N
N
N
N
N
N
N
N
N N
HN
N
HN
N
HN
N
HN
N
HN
N
NO2
16 CLogP: 2.87
Br
15 CLogP: 2.38
12 CLogP: 1.93
13 CLogP: 1.60
14 CLogP: 3.42
N
HN
N
N
N
N
N
NH
N
HN
N
N
N
HN
N
N
N
HN
N
N
N
N
HN
N
O
O
O
17 CLogP: 2.04 18 CLogP: 1.36 19 CLogP: 1.23 20 CLogP: 1.15 21 CLogP: -0.09 22 CLogP: 1.16
N
HN
N
N
N
N
N
H
N
N
N
N
N
N
N
NH
HN
N
N
N
NH
N
N
N
H
N
N
24 CLogP: 2.98
25 CLogP: 0.17
26 CLogP: -0.376
23 CLogP: 0.21
Scheme 2. Tetrazole formation from commercial nitriles.

B. Schmidt et al. / Tetrahedron 63 (2007) 492–496
495
1) NaN3
2) AcOH
R
N
N
N
NC
X
HN
Microwave 200W
60min
N
N
7
14 80%
O
O
1) NaN3
2) AcOH
N
CO2Bn
CO2Bn
N
R
N
N
X
N
27
Δ 73%
8
NC
HN
microwave 88%
N
N
Figure 3. Differential scanning calorimetry of [BMIM]Cl.
Scheme 3. Synthesis of advanced sartane intermediates.
therefore, selected for further evaluation. The [BMIM]Cl
turned out to be superior again, mainly due to rapid con-
version and product purity (Table 1, entries 30–35). Six
representative compounds were submitted to a standard pro-
tocol: 170 ꢀC for 30 min, 0.5 mmol nitrile, 1.5 mmol NaN3,
0.5 mmol HOAc. This protocol is similar in reaction time to
the Hallberg9 conditions, but uses a temperature maximum
instead of a fixed power input to reduce product decomposi-
tion. The yields were superior in five cases, when compared
to thermal heating of IL solutions. The yield improvements
are less spectacular than reported for microwave heating of
DMF solutions.9 Product 7 (Table 1, entry 31) requires ex-
tended reaction time for full conversion, therefore an inade-
quate 40% yield was observed under these suboptimal
conditions.
setting in a burst sealed reaction vessel (entry 21). No at-
tempts were initially made to control the temperature at
the small scale of the reaction (0.75 mmol) in a household
microwave oven. The microwave heated reaction was com-
plete within 60 min and far more selective (compare entries
15–22). It furnished the product in 80% yield. The more de-
tailed investigation was conducted in a Biotage InitiatorÔ,
which allowed to monitor the pressure and temperature
over time. The ILs were screened for rapid heating, which
in some cases varied significantly due to water impurities.
Several ILs displayed the desired rapid heating and were,
Finally, the valsartan precursor 8 was synthesized according
to literature4,5 and subjected to tetrazole formation. The ad-
justment of the NaN3/HOAc ratio and a gentle thermal heat-
ing at 140 ꢀC provided the product in 73% yield at complete
conversion (Table 1, entries 23–29). Any extension of the re-
action time resulted in slow decomposition of the tetrazole
27. The microwave heated conversion of this important com-
mercial intermediate 8 resulted in significant improvement
of the isolated yield (88%, Table 1, entry 35) and much re-
duced reaction time (24 h vs 30 min). The upscaling of the
reaction, improved the work up procedures such as extrac-
tion by CO2sc and stability of the IL are subject to further
investigations.
Acknowledgements
We thank the Fonds der Chemischen Industrie and Dr. U.
Welz-Biermann, Merck KGaA, Germany, for support of
this work. We thank B. Albert and H. Wolf, TU Darmstadt,
for differential scanning calorimetry.
References and notes
1. Herr, R. J. Bioorg. Med. Chem. 2002, 10, 3379–3393.
2. Wexler, R. R.; Greenlee, W. J.; Irvin, J. D.; Goldberg, M. R.;
Prendergast, K.; Smith, R. D.; Timmermans, P. B. M. W. M.
J. Med. Chem. 1996, 39, 625–656.
3. Schmidt, B.; Schieffer, B. J. Med. Chem. 2003, 46, 2261–2270;
Schmidt, B.; Drexler, H.; Schieffer, B. Am. J. Cardiovasc.
Drugs 2004, 4, 361–368.
Figure 2. Rates of DMF and [BMIM]Cl microwave heating.

496
B. Schmidt et al. / Tetrahedron 63 (2007) 492–496
4. B€uhlmayer, P. Preparation of N-aryl amino acid carboxyalkyl
12. All tetrazoles mentioned in this publication were indepen-
dently synthesized by published methods to guarantee the iden-
tity of the products isolated from reactions run in ionic liquids.
The isolated products were controlled for identity by co-
injected HPLC and NMR analyses. The reactions were moni-
tored by HPLC against internal standard.
13. Earle, M. J.; Esperanca, J. M. S. S.; Gilea, M. A.; Canongia
Lopes, J. N.; Rebelo, L. P. N.; Magee, J. W.; Seddon, K. R.;
Widegren, J. A. Nature 2006, 439, 831–834.
ester derivatives as angiotensin II antagonists. PCT Int. Appl.
WO9730036, 1997; pp 36; Novartis A.-G., Switz.
5. B€uhlmayer, P.; Ostermayer, F.; Schmidlin, T. Preparation
of [(tetrazolylbiphenyl)methyl]amines and analogs as angio-
tensin II antagonists. Eur. Pat. Appl. EP 443983, 1991; pp 64;
Ciba-Geigy A.-G., Switz.
6. Demko, Z. P.; Sharpless, K. B. J. Org. Chem. 2001, 66, 7945–
7950.
7. Amantini, D.; Beleggia, R.; Fringuelli, F.; Pizzo, F.; Vaccaro, L.
J. Org. Chem. 2004, 69, 2896–2898.
14. Vallin, K. S. A.; Emilsson, P.; Larhed, M.; Hallberg, A. J. Org.
Chem. 2002, 67, 6243–6246.
€
8. Butler, R. N. Compr. Heterocycl. Chem. II 1996, 4, 621–678;
Compr. Heterocycl. Chem. II 1996, 4, 905–1006.
9. Alterman, M.; Hallberg, A. J. Org. Chem. 2000, 65, 7984–7989.
10. Wasserscheid, P.; Keim, W. Angew. Chem., Int. Ed. 2000, 39,
3772–3789; Angew. Chem. 2000, 112, 3926–3945.
11. Welton, T. Chem. Rev. 1999, 99, 2071–2084.
15. Hoffmann, J.; Nuchter, M.; Ondruschka, B.; Wasserscheid, P.
Green Chem. 2003, 5, 296–299.
16. Ngo, H. L.; LeCompte, K.; Hargens, L.; McEwen, A. B.
Thermochim. Acta 2000, 357–358, 97–102.
17. Chan, B. K. M.; Chang, N.-H.; Grimmett, M. R. Aust. J. Chem.
1977, 30, 2005–2013.