Selective 15N-labeling and analysis of 13C- 15N J couplings as an effective tool for studying the structure and azide-tetrazole equilibrium in a series of tetrazolo[1,5- b ][1,2,4]triazines and tetrazolo[1,5-a ]pyrimidines

Article abstract of DOI:10.1021/jo1017876

Two general methods for the selective incorporation of an ¹⁵N-label in the azole ring of tetrazolo[1,5-b][1,2,4]triazines and tetrazolo[1,5-a]pyrimidines were developed. The first approach included treatment of azinylhydrazides with ¹⁵

Full text of DOI:10.1021/jo1017876

pubs.acs.org/joc
15
13
15
Selective N-Labeling and Analysis of C- N J Couplings as an
Effective Tool for Studying the Structure and Azide-Tetrazole
Equilibrium in a Series of Tetrazolo[1,5-b][1,2,4]triazines
and Tetrazolo[1,5-a]pyrimidines
,
Sergey L. Deev,* Zakhar O. Shenkarev, Tatyana S. Shestakova, Oleg N. Chupakhin,*
†
‡
†
,†,§
†
Vladimir L. Rusinov, and Alexander S. Arseniev
‡
†
‡
Yeltsin Ural Federal University, 19 Mira Street, 620002 Ekaterinburg, Russia, Shemyakin-Ovchinnikov
Institute of Bioorganic Chemistry, Russian Academy of Sciences, 16/10 Miklukho-Maklaya Street,
17997 Moscow, Russia, and I. Ya. Postovsky Institute of Organic Synthesis, Ural Branch of the Russian
§
1
Academy of Sciences, 22 S. Kovalevskoy Street, 620219 Ekaterinburg, Russia
deevsl@yandex.ru, chupakhin@ios.uran.ru
Received September 11, 2010
1
5
Two general methods for the selective incorporation of an N-label in the azole ring of tetrazolo[1,5-
b][1,2,4]triazines and tetrazolo[1,5-a]pyrimidines were developed. The first approach included
1
treatment of azinylhydrazides with N-labeled nitrous acid, and the second approach was based
5
1
on fusion of the azine ring to [2- N]-5-aminotetrazole. The synthesized compounds were studied by
5
1
13
15
H, C, and N NMR spectroscopy in both DMSO and TFA solution, in which the azide-tetrazole
equilibrium is shifted to tetrazole and azide forms, respectively. Incorporation of the N-label led to
1
5
1
3
15
the appearance of C- N J coupling constants (J ), which can be measured easily using either 1D
CN
1
3
15
13
C spectra with selective N decoupling or with amplitude modulated 1D C spin-echo experi-
ments with selective inversion of the N nuclei. The observed JCN patterns permit unambiguous
determination of the type of fusion between the azole and azine rings in tetrazolo[1,5-b][1,2,4]triazine
15
1
derivatives. Joint analysis of JCN patterns and N chemical shifts was found to be the most efficient
5
way to study the azido-tetrazole equilibrium.
1
2-14
Introduction
different substitution reactions and ring transformations
that make them convenient precursors for the synthesis of
other heteroaromatic compounds, e.g., azoloazines.
Derivatives of 1,2,4-triazines and pyrimidines present an
6-11
15,16
The
1-5
important core in many natural
and synthetic
bio-
logically active compounds. These heterocycles easily undergo
(7) El. Ashry, E. S. H.; Rashed, N.; Taha, M.; Ramadan, E. Adv.
Heterocycl. Chem. 1994, 59, 39–177.
(
1) Elyakov, G. B.; Stonik, V. A.; Makar’eva, T. N. Chem. Heterocycl.
Compd. 1977, 13, 345–359.
2) Wang, H.; Lim, K. L.; Yeo, S. L.; Xu, X.; Sim, M. M.; Ting, A. E.;
Wang, Y.; Yee, S.; Tan, Y. H.; Pallen, C. J. J. Nat. Prod. 2000, 63, 1641–1646.
(8) Cheng, C. C. Prog. Med. Chem. 1969, 6, 67–134.
(9) Suschitzky, J. L.; Sheard, P. Prog. Med. Chem. 1985, 21, 1–61.
(10) Stavenger, R. A. Annu. Rep. Med. Chem. 2008, 43, 87–102.
(11) Rusinov, V. L.; Ulomskii, E. N.; Chupakhin, O. N.; Charushin, V. N.
Russ. Chem. Bull. 2008, 57, 985–1014.
(
(
(
(
3) Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2004, 67, 1216–1238.
4) Roje, S. Phytochemistry 2007, 68, 1904–1921.
5) Smirnov, V. V.; Kiprianova, E. A.; Garagulya, A. D.; Esipov, S. E.;
(12) Alekseeva, N. V.; Yakhontov, L. N. Russ. Chem. Rev. 1990, 59,
514–530.
Dovjenko, S. A. FEMS Microbiol. Lett. 1997, 153, 357–361.
6) Jain, K. S.; Chitre, T. S.; Miniyar, P. B.; Kathiravan, M. K.; Bendre,
V. S.; Veer, V. S.; Shahane, S. R.; Shishoo, C. J. Curr. Sci. 2006, 90, 793–803.
(13) Charushin, V. N.; Chupakhin, O. N.; Van der Plas, H. C. Adv.
Heterocycl. Chem. 1988, 43, 301–353.
(14) Van der Plas, H. C. Tetrahedron 1985, 41, 237–281.
(
DOI: 10.1021/jo1017876
r 2010 American Chemical Society
Published on Web 11/17/2010
J. Org. Chem. 2010, 75, 8487–8497 8487

Deev et al.
JOCArticle
SCHEME 1. Azido-Tetrazole Tautomerism in a Series of
Azido-1,2,4-triazines A (X = CR , Y = CR
1
0
of the tetrazolo[5,1-c][1,2,4]triazines T was described in the
cyclization of benzo derivatives of 3-azido-1,2,4-triazines A
2
, Z = N) and
3
, Y = CR , Z = CR )
2
6,27
Azidopyrimidines A (X = CR
1
2
(
Scheme 1, Z=N, X=Y=benzo(C H )).
6 4
X-ray crystallography is usually used to confirm the azo-
1
7-20
loazine structures
mode in tetrazoloazines.
and for determination of the fusion
2
4,28,29
Unfortunately, this method
only gives information about the structure of compounds in
the solid state. To confirm the structure of such compounds
1
13
in solution, indirect UV-vis and H, C NMR spectro-
scopic methods are usually used, based on comparison with
model compounds. In this case, the spectra of tetrazoloa-
zines are most frequently compared with those of their deaza
transformations of the azine ring by the Dimrorth rearrange-
ment can easily change the type of fusion between the azole
2
7,30-32
analogues.
carbon atoms in azoloazines, the direct spectroscopic studies
Due to the low densities of hydrogen and
1
7-23
and azine rings.
tion of the fusion type in azoloazines becomes an important
Therefore, the unambiguous determina-
1
13
using well established H and C NMR methods (1D spec-
troscopy, 2D HMQC, HMBC, INADEQUATE, etc.) did
not provide enough information for structural characteriza-
tion of these compounds. In addition the computational
methods based on calculations of relative thermodynamic
stability of open-chain and cyclic forms can be used for
studies of azido-tetrazole equilibrium. This approach was
successfully applied for investigation of rearrangement of
linear azido-1,2,4-triazines A (Z=N; X=Y=benzo(C H ))
1
7-20
task in organic chemistry.
A similar problem arises from
the possibility of a structural rearrangement in the azole ring.
In this case, the azido-tetrazole tautomerism between fused
tetrazoles and their open-chain isomers (Scheme 1) should be
taken into consideration. For example, the cyclization of the
azido group attached to the di(tri)azine ring at the position
between nitrogen atoms (structure A) can result in formation
0
of two differently fused tetrazoloazines, T and T (Scheme 1).
6
4
Therefore, two structural parameters should be determined
for the characterization of the fused tetrazoloazines: (1) the
state of azide-tetrazole equilibrium and (2) the mode of fusion
between tetrazole and azine rings.
and azidoqunazolines A (X=CH; Z=Y=benzo(C H )) into
4
6
0
33,34
their angular isomers T .
To the best of our knowledge,
there is no general method to study the azido-tetrazole iso-
merization that allows for the direct experimental determina-
tion of the equilibrium state and the structure of the tetrazole
isomers in solution.
In many previous works, topics of the azido-tetrazole
transformation of azido-1,2,4-triazines similar to A (Z = N,
X=CR , Y=CR ) have been discussed, but several aspects
1
5
1
2
N-isotopic labeling of heterocyclic compounds increases
significantly the breadth of NMR methods that can be used
for determination of molecular structures and mechanisms
2
4-32
remain unclear.
1,2,4]triazines T was reported for the isomerization of 3-azido-
,2,4-triazines A (Scheme 1, Z=N, X=CH, Y=CAr; Z=N,
Exclusive formation of tetrazolo[1,5-b]-
[
3
5-37
1
of chemical rearrangements in solution.
In the case of
24,25
15
X=COH, Y=CMe/CPh).
Atthe same time, the formation
tetrazoloazines, the chemical shifts of the labeled N atoms
can be an indicator for either the azide or tetrazole forms of
3
8-44
1
15
13
15
(
15) El Ashry, E. S. H.; Rashed, N. Adv. Heterocycl. Chem. 1998, 72,
27–224.
16) Charushin, V.; Rusinov, V.; Chupakhin, O. In Comprehensive
the molecule.
Heteronuclear H- N and C- N
1
J couplings (JHN and JCN) give direct information about
(
3
chemical structure. Moreover, the vicinal couplings ( J) are
Heterocyclic Chemistry III; Katritzky, A. R., Ramsden, C. A., Scriven,
E. F. V., Taylor, J. K., Eds.; Pergamon Press: Oxford, UK, 2008; Vol. 9,
Chapter 9.02, pp 95-196.
45
useful in conformational studies of molecules in solution. These
favorable properties of JHN and JCN are used widely in NMR
studies of proteins and nucleic acids, in which heteronuclear
J couplings form the basis of so-called “triple-resonance” experi-
(
17) Loakes, D.; Brown, D. M.; Salisbury, S. A. J. Chem. Soc., Perkin
Trans. 1 1999, 1333–1337.
18) Rusinov, V. L.; Zyryanov, G. V.; Egorov, I. N.; Ulomskii, E. N.;
Aleksandrov, G. G.; Chupakhin, O. N. Russ. J. Org. Chem. 2004, 40, 85–89.
19) Vas’kevich, R. I.; Savitskii, P. V.; Zborovskii, L. Y.; Staninets, V. I.;
Rusanov, E. B.; Chernega, A. N. Russ. J. Org. Chem. 2006, 42, 1403–1408.
20) Carballares, S.; Cifuentes, M. M.; Stephenson, G. A. Tetrahedron
Lett. 2007, 48, 2041–2045.
21) Medwid, J. B.; Paul, R.; Baker, J. S.; Brockman, J. A.; Du, M. T.;
(
46,47
(
ments and play important role in conformational analysis.
(
(33) Asaad, A,N.; El Ashry, E. S. H. Z. Naturforsch. 1996, 51a, 1012–
1018.
(
Hallett, W. A.; Hanifin, J. W.; Hardy, R. A., Jr.; Tarrant, M. E.; Torley,
L. W.; Wrenn, S. J. Med. Chem. 1990, 33, 1230–1241.
(34) Alkorta, I.; Blanco, F.; Elguero, J.; Claramunt, R. M. Tetrahedron
2010, 66, 2863–2868.
(
22) Cameron, M.; Foster, B. S.; Lynch, J. E.; Shi, Y.-J.; Doling, U.-H.
Org. Process Res. Dev. 2006, 10, 398–402.
23) M u€ hlst €a dt, M.; Krausmann, H.; Fischer, G. J. Prakt. Chem. 1970,
12, 254–262.
24) Goodman, M. M.; Atwood, J. L.; Carlin, R.; Hunter, W.; Paudler,
W. W. J. Org. Chem. 1976, 41, 2860–2864.
25) Goodman, M. M.; Paudler, W. W. J. Org. Chem. 1977, 42, 1866–
869.
(35) Farr ꢀa s, J.; Fos, E.; Ramos, R.; Vilarrasa, J. J. Org. Chem. 1988, 53,
887–891.
(36) Chupakhin, O. N.; Rusinov, V. L.; Tumashov, A. A.; Sidorov, E. O.;
Karpin, I. V. Tetrahedron Lett. 1992, 33, 3695–3696.
(37) Coursindel, T.; Farran, D.; Martinez, J.; Dewynter, G. Tetrahedron
Lett. 2008, 49, 906–909.
(38) Th ꢀe taz, C.; Wehrli, F. W.; Wentrup, C. Helv. Chim. Acta 1976, 59,
259–264.
(39) Hull, W. E.; K u€ nstlinger, M.; Breitmaier, E. Angew. Chem. 1980, 92,
957–959; Angew. Chem., Int. Ed. Engl. 198019, 924-926.
(40) Chom, P.; Korczak, H.; Stefaniak, L.; Webb, G. A. J. Phys. Org.
Chem. 1999, 12, 470–478.
(41) Cmoch, P.; Wiench, J. W.; Stefaniak, L.; Webb, G. A. J. Mol. Struct.
1999, 510, 165–178.
(42) Cmoch, P. Magn. Reson. Chem. 2002, 40, 507–516.
(43) Cmoch, P.; Stefaniak, L.; Webb, G. A. Magn. Reson. Chem. 1997, 35,
237–242.
(44) Fruchier, A.; Pellegrin, V.; Claramunt, R. M.; Elguero, J. Org.
Magn. Reson. 1984, 22, 473–475.
(45) Bystrov, V. F. Prog. NMR Spectrosc. 1976, 10, 41–82.
(
3
(
(
1
1
1
8
2
(
26) Messmer, A.; Haj oꢀ s, G.; Benk oꢀ , P.; Pallos, L. J. Heterocycl. Chem.
973, 10, 575–578.
(
27) Messmer, A.; Haj oꢀ s, G.; Tam ꢀa s, J.; Neszm ꢀe lyi, A. J. Org. Chem.
979, 44, 1823–1824.
28) Karczmarzyk, Z.; Mojzych, M.; Rykowski, A. J. Mol. Stuct. 2007,
29, 22–28.
(
(
29) Mojzych, M.; Karczmarzyk, Z.; Rykowski, A. J. Chem. Crystallogr.
005, 35, 151–155.
(
(
30) Stevens, M. F. G. J. Chem. Soc., Perkin Trans. 1 1972, 1221–1225.
31) Castill oꢀ n, S.; Mel ꢀe ndez, E.; Pascual, C.; Vilarrasa, J. J. Org. Chem.
1
982, 47, 3886–3890.
32) Castill oꢀ n, S.; Vilarrasa, J. J. Org. Chem. 1982, 47, 3168–3169.
(
8
488 J. Org. Chem. Vol. 75, No. 24, 2010

Deev et al.
JOCArticle
1
5
SCHEME 2. Synthesis of Selectively N-Labeled 2* and
Associated NMR JCN Data Confirming Its Structure
isotopomers 4*aT and 4*bT in 5:1 ratio was obtained in 51%
total yield (Scheme 3, Method 1; please note that the identical
numeration of atoms in triazines 2*, 9*, 12*, and pyrimi-
dines 4* instead of the IUPAC numbering was used for the
a
4
9,50
convenience). The Dimroth rearrangement
is one of the
most probable sources of the observed isomerization (see
Scheme 3).
The second method involved the cyclization of pyrimidine
1
5
or 1,2,4-triazine rings to labeled [2- N]-5-aminotetrazole 5*
and its derivatives. Synthesis of 5* was based on the Thiele
method, which included the interaction of aminoguanidine
salts with nitrous acid (see Supporting Information).
1
5
Because the N-aminotetrazole 5* exists in two tautomeric
forms in solution the reaction of this compound should result
in formation of a mixture of isotopomers. As expected, the
condensation of tetrazole 5* with 1,1,3,3-tetramethoxypro-
pane led to a mixture of tetrazolopyrimidines 4*aT and 4*bT
in a 1:1 ratio (Scheme 3, Method 2).
15
Cyclization of diazonium salt 6*, derived from N-la-
beled aminotetrazole 5* (see Supporting Information), with
ethyl cyanoacetate 7 gave a mixture of tetrazolo[1,5-b][1,2,4]-
triazines 9*aT and 9*bT in 60% yield and a 1:1 ratio (Scheme 4).
Similarly, the 1:1 isotopomer mixture of tetrazolo[1,5-b]-
a
The azide-tetrazole equilibrium is shifted to the tetrazole/azide form
in DMSO/TFA solution. The observed JCN couplings from the N7
nucleus are shown by red arrows. The expected but unobserved coupling
1
5
2
(
presented in Table 1.
JCN) is indicated by dashed arrows. The measured JCN values are
[
1,2,4]triazines 12*aT and 12*bT was obtained in the reac-
Herein we report an efficient method for the structural
determination of the cyclization products of azido-1,2,4-
triazines A (Scheme 1, X = C-R , Y = C-R , Z = N) and
tion of 6* with ethyl R-formylphenylacetate 10 in 48%
overall yield (Scheme 4). It should be noted that the reaction
of 6* with ethyl phenylacetate did not occur due to the low
acidity of the phenylacetate. The C-formylation reaction was
used previously to activate methylene-active compounds in
reaction with diazonium salts. In this case, the electron-
withdrawing formyl group increases the acidity of the C-H
bond and is eliminated from the product after azacoupling
1
2
azidopyrimidines A (Scheme 1, X=Y=Z=CH) based on
1
selective N isotope labeling of the tetrazole ring, followed
5
1
3
15
15
by joint analysis of C- N J couplings and N chemical
shifts. This method not only allows for confirmation of the
type of fusion between azole and azine rings but also permits
the direct study of azide-tetrazole equilibrium in solution.
5
1
5
2-54
(
Japp-Klingemann reaction).
The synthesis of tetrazoloazines 9*a,bT and 12*a,bT pro-
Results
vides a good illustration of how the azide-tetrazole tauto-
merism changes the pathway of chemical reactions. The
cyclization of 8* and 11* (products of azacoupling of 6*
with 7 and 10) should give tetrazolo[5,1-c][1,2,4]triazines
9*a,bT and 12*a,bT but likely undergoes tetrazole ring
opening to give azides 9*a,bA and 12*a,bA followed by an
alternative closure yielding tetrazolo[1,5-b][1,2,4]triazines
9*a,bT and 12*a,bT (Scheme 4).
1
5
Synthesis. Incorporation of an N-isotopic label in the
tetrazole part of tetrazolo[1,5-b][1,2,4]triazines and tetrazolo-
[1,5-a]pyrimidines was performed by two methods using either
1
K NO or K NO with 86% N enrichment of the parent
5
15
15
0
0
3
2
substances. The first method was based on the interaction of
1
heteroarylhydrazides with N-labeled nitrous acid generated
5
15
from K NO . This method, used for 1,2,4-triazine derivatives,
2
was tested on the known reaction of 2-hydrazino-1,2,4-triazines
2
Characterization of Fused Tetrazoloazines. NMR Spectro-
scopy in DMSO Solution. The synthesized compounds were
studied by NMR methods in both dimethyl sulfoxide and
trifluoracetic acid using samples with concentrations in the range
of 90-160 mM. These solvents were chosen because in most
4,25,28,29
with nitrous acid.
The treatment of compound 1
1
5
with N-labeled potassium nitrite in phosphoric acid gave
tetrazolotriazine 2*T in 56% yield through the spontaneous
cyclization of azide 2*A (Scheme 2). Despite two possible ring
closures, only the linear tetracycle 2*T was obtained rather
than the angular 2*T . (The details of the NMR and crystallo-
graphic analysis of the synthesized compounds confirming
their structures are presented in a separate section of the
manuscript.)
cases the azide-tetrazole equilibrium was shifted to the tetrazole
0
48
55-57
form in DMSO solution, whereas the presence of strong
acid, such as TFA, usually increases the relative concentration of
(49) Spickett, R. G. W.; Wright, S. H. B. J. Chem. Soc. C 1967, 498–502.
(
50) Allen, C. F. H.; Beilfuss, H. R.; Burness, D. M.; Reynolds, G. A.;
Tinker, J. F.; VanAllan, J. A. J. Org. Chem. 1959, 24, 787–793.
51) Ulomskii, E. N.; Shestakova, T. S.; Deev, S. L.; Rusinov, V. L.;
Chupakhin, O. N. Russ. Chem. Bull. 2005, 54, 726–732.
52) Japp, F. R.; Klingemann, F. Justus Liebigs Ann. Chem. 1888, 247,
90–225.
53) Japp, F. R.; Klingemann, F. Ber. Dtsch. Chem. Ges. 1887, 20, 3398–
3401.
(54) Yao, H. C.; Resnick, P. J. Am. Chem. Soc. 1962, 84, 3514–3517.
The same procedure was applied to the incorporation of
1
an N-isotopic label in tetrazolo[1,5-a]pyrimidine. The re-
5
(
1
5
action of 2-hydrazinopyrimidine 3 and N-enriched nitrous
(
acid, however, did not proceed selectively, and a mixture of
1
(
(46) Cavanagh, J.; Fairbrother, W. J.; Palmer III, A. G.; Skelton, N. J.;
Rance, M. Protein NMR Spectroscopy: Principles and Practice, 2nd ed.;
Academic Press: San Diego, CA, 2006.
(55) Wentrup, C. Tetrahedron 1970, 26, 4969–4983.
(56) Denisov, A.Yu.; Krivopalov, V. P.; Mamatyuk, V. I.; Mamaev, V. P.
Magn. Reson. Chem. 1988, 26, 42–46.
(
47) Bax, A.; Vuister, G. W.; Grzesiek, S.; Delaglio, F.; Wang, A. C.;
Tschudin, R.; Zhu, G. Methods Enzymol. 1994, 239, 79–105.
48) Joshi, K. C.; Chand, P. J. Heterocycl. Chem. 1980, 17, 1783–1784.
(
(57) Temple, C., Jr.; Montgomery, J. A. J. Org. Chem. 1965, 30, 826–829.
J. Org. Chem. Vol. 75, No. 24, 2010 8489

Deev et al.
JOCArticle
1
5
a
SCHEME 3. Synthesis of Selectively N-Labeled 4* Made by Two Alternative Methods
a
The NMR JCN data confirming the structures are also shown. The Dimroth rearrangement (shown on the right) is probably involved in the
isomerization of 4*aT to 4*bT during synthesis by Method 1. The azide-tetrazole equilibrium is shifted to the tetrazole/azide form in DMSO/TFA
solution. The observed JCN couplings from the N7 and N8 nuclei are shown by red and blue arrows, respectively. The expected but unobserved
15
15
2
coupling ( JCN) is indicated by dashed arrows. The measured JCN values are presented in Table 1.
1
5
a
SCHEME 4. Synthesis of Selectively N-Labeled 9* and 12* and NMR JCN Data Confirming Their Structures
a
15
15
The azide-tetrazole equilibrium is shifted to the tetrazole/azide form in DMSO/TFA solution. The observed JCN couplings from N7 and N8
2
3
nuclei are shown by red and blue arrows, respectively. The expected but unobserved couplings ( JCN and JCN) are indicated by dashed arrows. The
measured JCN values are presented in Table 1.
40,43,57
the azide form.
presented in Figures 1-5 and Tables 1 and 2. The HNMRdata,
as well as representative 2D C-HMQC and C/ N-HMBC
spectra, are shown in Supporting Information.
13 15
The collected C and N NMR data are
15
Selective incorporation of an N label into tetrazolo[1,5-b]-
1
[1,2,4]triazines 2*T, 9*a,bT, and 12*a,bT and tetrazolo[1,5-a]-
13
1
3
13 15
15
pyrimidines 4*a,bT allowed for the detection of C- N
J-couplings, which became apparent as additional splittings of
8
490 J. Org. Chem. Vol. 75, No. 24, 2010

Deev et al.
JOCArticle
1
3
13
15
a
TABLE 1.
C Chemical Shifts (ppm) and C- N J-Coupling Constants (Hz) of the Studied Compounds
compound
solvent
C2
C4/C6
C5
other signals
2
*T
DMSO
148.92
150.57
141.92
116.06(C10) 146.77(C11) 124.22(C12) 113.49(C13)
123.31(C14) 134.89(C15)
114.54(C10) 144.96(C11) 124.56(C12) 113.76(C13)
2
3
JC-N7 3.1
JC-N7 1.4
143.69
c,d
149.64
d
c
c,d
c,d
TFA
143.17
2
b
3
J
c
JC-N7 < 0.3
C-N7 1.7
136.65(C15) 125.56(C14)
116.02(C10) 142.58(C11) 123.43(C12) 113.76(C13)
d
c
2
4
*A
TFA
155.94
149.07
140.31
3
c
134.97(C15) 126.12(C14)
JC-N7 0.5
e
e
*a,bT
DMSO
154.91
160.25(C4)
135.51(C6)
113.74
2
3
J
C-N7 2.3
J
C-N7 1.0
2
2
f
4
J
JC-N8 2.5
J
C6-N7 3.8
C-N8 0.5
3
g
JC6-N8 < 0.3
4*a,bA
DMSO
TFA
161.80
160.10(C4/6)
158.40(C4/6)
154.67
118.30
116.98
3
h
J
C-N7 0.6
2
h
JC-N8 ND
c
157.35
3
J
C-N7 0.5
2
JC-N8 0.7
9*a,bT
9*a,bA
12*a,bT
DMSO
150.07
132.87
162.05(C10) 14.27(Me) 63.65(OCH
2
)
2
3
J
C-N7 3.0
J
C-N7 1.1
2
4
J
JC-N8 2.2
C-N8 1.0
TFA
144.63
155.68
132.72
160.73(C10) 11.65(Me) 65.32(OCH
2
2
)
)
2
i
i
J
C-N7/8 ∼0.5
J
C-N7/8 ∼1.0
j
j
j
DMSO
TFA
156.58
156.33
133.35
164.56(C10) 14.46(Me) 62.10(OCH
j
JC-N7/8 ND
d
161.23(C10) 11.64(Me) 64.93(OCH
157.87
156.29
154.35
153.04
129.77
)
3
J
C-N7 0.5
2
JC-N8 0.8
DMSO
TFA
146.01
152.14
131.34(C10) 132.00(C13) 128.75(C11/15)
129.98(C12/14)
2
3
J
C-N7 3.3
J
C-N7 1.6
2
4
J
JC-N 8c 2.1
C-N8 0.9
c
143.53
152.99
126.49(C10) 133.07(C13) 129.14(C11/15)
128.05(C12/14)
k
3
JC-N7/8 ND
J
C-N7 1.6
4
J
C-N8 0.7
d
d
c
12*a,bA
TFA
156.85
161.31
149.96
128.24(C10) 133.28(C13) 128.66(C11/15)
128.63(C12/14)
3
J
C-N7 0.5
2
JC-N8 0.8
a
13
Unless otherwise stated, the listed JCN values represent the average between two independent measurements using C line-shape analysis and
amplitude modulated spin-echo experiments (see Supporting Information). Both types of measurements gave identical values of JCN within
1
3
b
experimental error ((0.15 Hz). C chemical shifts were referenced indirectly relative to proton signal of tetramethylsilane (TMS). The JCN coupling
constant was not detected, probably due to broadeningof the corresponding C resonance. The signal demonstratedadditional broadening or splitting,
1
3
c
d
which was not connected with JCN or JCH couplings. The signal partially overlapped with the C signals of TFA-d or with the C signals of fluorine-
13
13
e
containing impurities. The mixture of 4*a/4*b isotopomers (5:1) synthesized by Method 1 of Scheme 3 were used for JCN measurements. The JCN value
f
1
3
13
C
was measured only by C line-shape analysis. Measurement using spin-echo was impossible due to fast transverse relaxation of the corresponding
nucleus. The JCN coupling constant was not detected, probably due to low abundance of the 4*b isotopomer. The JC-N7 value was measured by only
g
h
3
1
3
2
C line-shape analysis. Measurement of the JC-N8 coupling was impossible due to low abundance of the 4*b isotopomer. The measurements using
i
13
spin-echo were impossible due to the low concentration of the azide form of 4*a,bA in DMSO solution. The JCN values were estimated from C line-
shape analysis. The precise measurements were impossible due to fast conversion of tetrazoles 9*a,bT to azides 9*a,bA in TFA solution. The assignment
j
1
of C resonances in the azide fragment and measurement of JCN were impossible due to the low concentration of the azide form 9*a,bA in DMSO
3
k
solution. The presented tentative assignment is obtained by comparison with spectra of 9*a,bA in TFA solution. Measurements of JCN couplings were
1
impossible due to significant broadening of the corresponding C resonance.
3
1
3
13
of C signals in azine moieties of the compounds, where
the corresponding signals in the C NMR spectra. It should be
noted that in some cases the J splittings were unresolved, thus
1
13
13
standard 2D protocols based on H- C J-couplings ( C-
HMQC and C-HMBC) were ineffective due to the absence
CN
1
3
15
requiring double-resonance ( C/ N) spectroscopy for their
detection. To detect and quantitatively characterize the C- N
interactions, two alternative methods were employed (see Sup-
porting Information). In the first approach, the values of J
were extracted from nonlinear fits of the C line shapes in the 1D
spectra acquired with or without band-selective decoupling from
13
13
15
13
of protons. Moreover, the line shape analysis of C multi-
1
5
plets confirmed the 86% excess of the N isotope, expected
1
from the degree of N enrichment in the parent substances.
5
CN
13
13
15
1
One set of signals was observed in the C, N, and H
NMR spectra of 2*T in DMSO solution (Figures 1a and 2a;
Figure S1a in Supporting Information) that indicated the
15
N nuclei. The second method was based on the amplitude
4
7
13
modulated spin-echo technique. In this case, the value of JCN
was extracted quantitatively from the intensities of multiplets in
presence of only one tautomeric form. The 1D C NMR
spectra of 2*T showed JCN splittings for signals from the C2
and C5 nuclei (Figure 1a, Table 1). These coupling constants
proved the formation of the cyclic structure with [1,5-b] type
of fusion between tetrazole and 1,2,4-triazine rings. If the
13
1
D C spin-echo spectra measured with and without selective
15
inversion of the N nucleus of interest in the middle of echo
period. Both methods gave JCN values with comparable accu-
racy (see Table 1 and Supporting Information). The analysis of
the JCN patterns permitted the straightforward identification of
the mode of fusion between azole and azine fragments in tetra-
zolo-1,2,4-triazine derivatives and significantly aided assignment
0
alternative structure 2*T with the [5,1-c] type of fusion were
13
present, then one should observe JCN splittings for all
nuclei (C2, C4, and C5) of the triazine fragment (Scheme 2).
C
Similarly, the spectra of azide 2*A should be characterized
J. Org. Chem. Vol. 75, No. 24, 2010 8491

Deev et al.
JOCArticle
1
3
FIGURE 1. Fragments of proton decoupled 1D C NMR spectra
of 90 mM 2*T in DMSO-d (a) and TFA-d (b). The spectrum in
6
1
3
panel (b) was measured 30 days after dissolving. The C multiplets
demonstrating JCN couplings are enlarged and are shown above the
spectra. The fragments of the spectra acquired with band-selective
decoupling from the N7 nucleus are shown in red. The signal of a
fluorine-containing impurity in the TFA is marked by an asterisk.
1
5
FIGURE 2. 1D N NMR spectra of 90-160 mM 2*T, 4*a,bT,
9*a,bT, and 12*a,bT in DMSO-d (a) and TFA-d (b). The time
passed after dissolution of compounds with TFA and method used
6
15
for synthesis of 4*a,bT are given above the spectra. The spectral
ranges for signals of N7 and N8 nuclei in the tetrazole and azide
1
5
15
by the presence of JCN splittings for only the C2 nucleus.
forms of the compounds are denoted by vertical dashed lines. The
signal of a K NO impurity is marked by an asterisk.
2
Here we assume that all geminal ( J ) and vicinal ( JCN)
3
15
3
(
couplings have detectable amplitude >0.3 Hz.) The signal of
CN
1
5
a
1
5
15
TABLE 2.
compound
2*T
N Chemical Shifts (ppm) of the Studied Compounds
1
the N7 nucleus was observed in the 1D N NMR spectrum
at -30.5 ppm (Figure 2a, Table 2), within the spectral range
reported previously for fused tetrazoloazines (from -33 to
5
N7
15
15
b
solvent
δ
δ
N8
δ
N of other signals
DMSO
TFA
TFA
-30.5
-30.7
-130.4
-31.5
4
0,43
-
24 ppm).
1
Two signals corresponding to N-labeled atoms were
detected in the 1D N NMR spectrum of a mixture of iso-
topomers 4*aT and 4*bT of tetrazolo[1,5-a]pyrimidine in
DMSO solution (Figure 2a, Table 2). These signals indicated
N-label incorporation at the N7 (δ -31.5 ppm) and N8
δ 22.6 ppm) positions, respectively. The assignment of the
N7/8 resonances was confirmed by a 2D N-HMBC
5
2*A
4*a,bT
1
5
DMSO
22.6
-42.7 N9
-
104.0 N3
142.0 N1
-
4*a,bA
9
TFA
-129.3
-30.1
-29.5
-129.2
-28.9
-27.9
-131.0
-153.3
18.0
-13.6
-153.9
5.8
-12.2
-153.6
15
a,bT
DMSO
TFA
TFA
DMSO
TFA
TFA
(
9a*,b*A
12*a,bT
1
5
15
spectrum (Figure S6 in Supporting Information). The inte-
1
5
gral intensity of these signals in the 1D N NMR spectrum
provided the ratio of the 4*a,bT isotopomers formed during
the preparation according to Methods 1 and 2 (Scheme 3).
The detection of JCN couplings for the C2, C5, and C6 nuclei
12*a,bA
a 15
N chemical shifts were referenced indirectly relative to nitro-
methane (MeNO ). The assignment of signals was obtained at natural
2
1 15
N abundance using a 2D H, N-HMBC spectrum.
b
1
5
(
Figure 3a, Table 1) proved the formation of fused tetrazole
1
5
Similarly, two signals corresponding to N-labeled atoms
(
4
Scheme 3). Interestingly, the small amount of azide form
1
3
1
15
15
*a,bA (∼ 5%) was detected in both the C and H NMR
(δ N7 -30.1 ppm and δ N8 18.0 ppm) were observed in
the 1D N NMR spectrum of 9*a,bT in DMSO solution
(Figure 2a, Table 2) indicating the formation of a mixture of
15
N7/8 isotopomers with 1:1 ratio. (Here and in the case of
15
spectra (Figure 3a, Figure S2a in Supporting Information).
The low concentration of this form did not allow observation
15
of the corresponding nitrogen resonances in the 1D N spectra.
Similarly, the JCN splitting, which was only observed for the
C2 nucleus, confirmed the azide structure of 4*a,bA (Figure 3a).
According to the obtained data, the ring opening asso-
ciated with the tetrazole to azide rearrangement of 4*a,bT in
DMSO is characterized by a relatively large downfield shift
of the C2 (Δδ 6.89 ppm), C5 (Δδ 4.56 ppm), and C6 (Δδ 24.59
ppm) resonances. It should be noted that partial rearrange-
ment of the tetrazolo[1,5-a]pyrimidine to the corresponding
12*a,bT assignment of N7 and N8 signals was obtained by
1
5
comparison with N spectra of compounds 4*a,bT and
2*T.) The detection of JCN couplings between the labeled
nitrogen atoms and C2 and C5 nuclei (Figure 4a, Table 1)
proved the formation of the fused tetrazole with a [1,5-b] type
of fusion between tetrazole and 1,2,4-triazine rings. If the
alternative structure 9*a,bT with a [5,1-c] type of fusion were
present, the JCN coupling should be detected for all
nuclei (C2, C4, and C5) of the triazine fragment (Scheme 4).
0
13
C
5
6
azide in DMSO was observed previously by Denisov et al.
8
492 J. Org. Chem. Vol. 75, No. 24, 2010

Deev et al.
JOCArticle
13
FIGURE 3. Proton decoupled 1D C NMR spectra of 160 mM 4*a,
bT in DMSO-d (a) and TFA-d (b). The spectrum in panel (b) was
measured 1 h after dissolving. The C multiplets demonstrating JCN
13
FIGURE 4. Fragments of proton decoupled 1D C NMR spectra
of 130 mM 9*a,bT in DMSO-d (a) and TFA-d (b). The spectrum in
6
13
6
couplings are enlarged and shown above the spectra. The fragments of
13
panel (b) was measured 1 h after dissolving. The C multiplets
demonstrating JCN couplings are enlarged and shown above the
spectra. The fragments of spectra acquired with band-selective
15
15
spectra acquired with band-selective decoupling from N7 and N8
nuclei are shown in red and blue, respectively. The signals of TFA and a
fluorine-containing impurity are marked by asterisks.
15
15
decoupling from the N7 and N8 nuclei are shown in red and
blue, respectively. The signal of a fluorine-containing impurity in
the TFA is marked by an asterisk. Please note that the shown
tentative assignment of 9*a,bA in DMSO was obtained by compar-
ison with spectra of 9*a,bA in TFA.
Analogously to compounds 4*a,bT, a small amount of the
13
1
azide form 9*a,bA (∼4%) was detected in the Cand HNMR
spectra (Figure 4a, Figure S3a in Supporting Information). This
form was also characterized by a relatively large downfield shift
of the C2 (Δδ6.51 ppm) resonance. The low concentration of the
azide form did not allow for detection or measurement of the
corresponding JCN couplings.
In contrast to 4*a,bT and 9*a,bT, but similar to 2*T, a
13
1
single set of resonances was observed in the 1D Cand HNMR
spectra of 12*a,bT in DMSO solution (Figure 5a, Figure S4a in
Supporting Information). In this case, two nitrogen signals
1
5
with identical intensities and chemical shifts (δ N7 -28.9
1
5
and δ N8 5.8 ppm), typical for fused tetratrazoloazines, were
15
detected in the 1D N NMR spectrum (Figure 2a, Table 2).
The above observations indicated that compounds 12*a,bT
are only in the tetrazole form in DMSO solution and represent a
1
5
1
:1 mixture of N7/8 isotopomers (Scheme 4). The observa-
15
1
3
tion of the C- N splitting for the C2 and C5 carbon
1
3
signals in the 1D C NMR spectrum of 12*a,bT unambigu-
ously confirmed the [1,5-b] type of fusion between the
tetrazole and 1,2,4-triazine rings (Figure 5a, Table 1).
The NMR-derived structures of compounds 2*T, 4*a,bT
and 12*a,bT in DMSO-d solution are in good agreement
6
with X-ray diffraction data. Suitable crystals for X-ray
structure determination were prepared in the case of un-
labeled analogues of 2*T, 4*a,bT and 12*a,bT (Figure 6).
The results showed that all compounds existed as the
tetrazole isomers in the crystalline form. The X-ray data also
confirmed the [1,5-b] type of fusion between the tetrazole and
azine rings in the 1,2,4-triazine derivatives.
1
3
FIGURE 5. Fragments of proton decoupled 1D C NMR spectra of
10 mM 12*a,bT in DMSO-d (a) and TFA-d (b). The spectrum in
1
6
13
panel (b) was measured 30 days after dissolving. The C multiplets
demonstrating JCN couplings are enlarged and shown above the spectra.
The fragments of spectra acquired with band-selective decoupling from
N7 and N8 nuclei are shown in red and blue, respectively. The signals
of TFA and a fluorine-containing impurity are marked by asterisks.
Tetrazole to Azide Rearrangement. NMR Spectroscopy in
TFA Solution. According to the NMR data, the tetrazolo-
1
5
15
[
1,5-b]pyrimidines 4*a,bT undergo rapid rearrangement to
J. Org. Chem. Vol. 75, No. 24, 2010 8493

Deev et al.
JOCArticle
FIGURE 6. ORTEP diagrams of the X-ray structure of compounds 2T (a), 4T (b), and 12T (c).
the 4*a,bA azide forms in TFA solution (Scheme 3). As
TABLE 3. Tetrazole-Azide Isomerization of the Studied Compounds
in Different Solvents
1
3
15
expected for the azide form, the presence of C- N
J-couplings was detected only for the C2 nucleus (Figure 3b,
Table 1). All of the carbon resonances of 4*a,bA in TFA
solution were shifted upfield relative to the positions ob-
served for 4*a,bA in DMSO. Nevertheless, these shifts had
moderate amplitude (see Figure 3b), and the C2 and C5
resonances of 4*a,bA in TFA still demonstrate downfield
shifts when compared to the corresponding resonances of
solvent
time after dissolution
with TFA
a
azide:tetrazoloazine
DMSO
0:100
TFA
2
2
2
*A:2*T
*A:2*T
*A:2*T
0:100
67:33
87:13
1 h
16 days
30 days
1 h
1 h
2 h
12 h
1 h
30 days
b
4*a,bA:4*a,bT
9*a,bA:9*a,bT
5:95
4:96
100:0
18:82
40:60
97:3
b
b
9*a,bA:9*a,bT
9*a,bA:9*a,bT
12*a,bA:12*a,bT
12*a,bA:12*a,bT
4
*a,bT in DMSO. The signal of the C4/C6 nuclei of the azide
form in TFA demonstrated significant line broadening
Figure 3b), which can be ascribed to protonation of the N1/
nitrogen atoms of the pyrimidine fragment. The resonances
c
0:100
0:100
60:40
(
3
a
The relative concentrations of the azide and tetrazole forms were
determined from integral intensities of the corresponding signals in the
1
5
15
of the N7 (δ -129.3 ppm) and N8 (δ -153.3 ppm) nuclei
were observed in a characteristic region for azidoazines
1
15
b
1D H and N NMR spectra. Signals of the azide form were not
15 c
4
0,43
detected in the N spectra. Signals of the tetrazole form were not
15
detected in the N spectra.
(
4
from -154 to -129 ppm).
The investigation of the
*a,bT mixture synthesized according to Method 1 of Scheme 2
1
5
and having a 5:1 ratio of N7/8 isotopomers allowed assign-
ment of the N7 and N8 resonances using their integral
the associated coupling constant has significantly smaller
amplitude in TFA (<0.3 Hz) than in DMSO solution (3.1 Hz).
In addition to the line broadening, the C2 resonance of 2*T
demonstrated a relatively large upfield shift (Δδ -5.75 ppm)
upon transition of the compound from DMSO to TFA solution
1
5
intensities in the 1D N NMRspectrum (Figure 2b, Table 2).
All obtained data confirm the full rearrangement of tetra-
zolo[1,5-b]pyrimidine to the azide form under acidic condi-
tions and are in a good agreement with the data reported by
5
7
13
Temple et al.
In contrast to the 4*a,bT heterocycles, 2*T, 9*a,bT, and
2*a,bT did not undergo fast conversion to the azide form in
(Figure 1, Table 1). Similar changes in C chemical shifts of
common carbon atoms in azole and azine rings were reported
4
3
1
previously for tetrazolo[1,5-a]pyridines.
Compound 2*A in TFA was characterized by the N7
15
TFA solution. This difference allowed the simultaneous
NMR detection of both isomeric forms. The position of
the tetrazole-azide equilibrium has been characterized quan-
titatively by the integral intensities of signals of both
3
resonance at -130.4 ppm (Figure 2b) and by the J
C2-N7
coupling constant with a magnitude of 0.5 Hz (Figure 1b,
Table 1). The other JCN interactions were not detected in the
spectra, thus confirming the azide structure of 2*A. The
comparison of chemical shift values between 2*T and 2*A in
TFA indicated that upon ring opening during the tetrazole to
azide rearrangement the C2 nucleus demonstrates a signifi-
cant downfield shift (Δδ 12.77 ppm), although the C4 and C5
resonances exhibit moderate upfield shifts (Figure 1b, Table 1).
Additional splittings were observed also for some signals of the
benzene fragment of both 2*T and 2*A (Table 1). These effects
are possibly connected with proton/deuterium exchange at the
N16 atom.
1
15
isomers in the 1D H and N NMR spectra (Table 3). In
the case of compounds 9*a,bT, the almost complete rearran-
gement of the tetrazole isomers to azides was observed after
1
2 h. Compounds 12*a,bT and 2*T were not transformed
completely to the corresponding azides even after 1 month in
TFA solution. The similar slow isomerization of tetrazole to
40
azide was reported previously for tetrazolo[1,5-a]pyridines.
The NMR spectral parameters of compound 2*T were
determined in TFA solution immediately after dissolving the
tetrazole form, after 16 days, and again after 30 days. During
this period the relative population of the azide form 2*A
increased gradually from 0% to 87% (Table 3). The tetrazole
The rearrangement of 9*a,bT to 9*a,bA in TFA solution
was relatively fast when compared to compound 2*T. The
1
5
1
15
H, N, and C NMR spectra measured at 1 and 12 h after
13
form of 2*T in TFA solution was characterized by the N7
resonance at -30.7 ppm (Figure 2b) and by the presence of a
dissolving 9*a,bT in TFA demonstrated that the tetrazole
and azide isomers of the compounds (Table 3) were the
predominant forms. The assignment of the N7 and N8
3
J
coupling constant with magnitude 1.7 Hz (Figure 1b,
Table 1). These values were in good agreement with data
C5-N7
15
3
15
signals in both isomers was obtained by comparison with N
observed for 2*T in DMSO (δ N7 -30.5 ppm and J
C5-N7
1
[
.4 Hz), thus proving the cyclic structure of 2*T in TFA with the
1,5-b] type of fusion between the azole and azine rings. The JCN
interaction between the C2 and N7 nuclei was not detected
possibly due to broadening of C2 resonance), indicating that
spectra of compounds 4*a,bA and 2*T,A (Figure 2b, Table 2).
The JCN interactions for the C2 and C5 carbon atoms of
1
3
tetrazole 9*a,bT were observed in the C NMR spectra
(Figure 4b, Table 1), thus proving that the tetrazole structure
(
8
494 J. Org. Chem. Vol. 75, No. 24, 2010

Deev et al.
JOCArticle
different nitrogen atoms of the 1,2,4-triazine fragment under
was retained and that the [1,5-b] type of fusion for 9*a,bT
occurred. The precise measurements of JCN values for 9*a,bT
was impossible due to fast conversion of tetrazole to azide.
Simplified line-shape analysis, however, revealed that the
5
8
acidic conditions.
Discussion
2
J
2
and J
interactions in 9*a,bT have significantly
C2-N7
C2-N8
1
3
15
lower values in TFA (e0.5 Hz) than in DMSO solution (3.3/
.1 Hz). Similarly to the observation for compound 2*T, the C2
C/ N NMR Markers Provide a Way To Characterize
Fused Tetrazoles and Their Azides in Solution. Selective
2
1
5
resonance demonstrates relatively large upfield shift (Δδ -5.44
ppm) upon transition of tetrazole 9*a,bT from DMSO to TFA
solution (Figure 4). In addition, a significant upfield shift of the
N8 resonance (Δδ -31.6 ppm) was observed.
incorporation of the N isotope in the azole ring of tetrazolo-
[1,5-a]pyrimidine and tetrazolo[1,5-b][1,2,4]triazine derivatives
allows the structure of these compounds to be characterized
and provides a way to track the state of the azide-tetrazole
equilibrium by NMR spectroscopy. Three types of NMR
markers were found to be the most useful for characterizing
fused tetrazoles and their open-chain isomers. These markers
As expected, the azide isoforms 9*a,bA in TFA solution
1
3
15
were characterized by the presence of C- N J-coupling
constants for only the C2 carbon (Figure 4b, Table 1). Inter-
estingly, comparison of the NMR spectra of 9*a,bA mea-
sured in DMSO and TFA solutions does not reveal large
15
are (1) the N chemical shifts of the nuclei from azole fragment,
1
3
(2) the C chemical shift of the bridge-head carbon C2,
which belongs to both the azole and azine rings (Scheme 1),
and (3) the JCN coupling constants between the nitrogen
nuclei of azole and the carbon nuclei of azine ring.
1
3
changes in the C chemical shifts (Figure 4). The largest dif-
ferences (up to 3.6 ppm) were observed for the C5 atom of the
triazine fragment and the atoms of the ethoxycarbonyl
group, which is covalently linked to C5 (Table 1). The com-
15
In all studied compounds, the resonances of the N7 and
N8 nuclei were found in specific spectral ranges (Table 2,
1
3
1
5
parison of the C chemical shift values between 9*a,bT and
*a,bA in TFA solution indicated that the resonance of the
9
Figure 2). In the tetrazole form, these signals were observed
at δ -31.5 to -27.9 ppm and at δ -13.6 to 22.6 ppm. Upon
ring opening and formation of azides, however, the signals
moved to δ-131.0 to -129.2 ppm and δ-153.9 to -153.3 ppm,
respectively. Despite the relatively large solvent-dependent
C2 atom demonstrates a large downfield shift (Δδ 13.24
ppm) upon opening of the tetrazole ring. In contrast, the
other resonances of the 1,2,4-triazine fragment shift moder-
ately upfield (Figure 4b).
The NMR spectral parameters of compounds 12*a,bT
were determined in TFA immediately after dissolving the
tetrazole form and 30 days after preparation of the solution.
During this period of time the relative population of the
azide forms of 12*a,bA increased from 0% to 60% (Table 3).
The assignment of the N7 and N8 signals in both isomers was
15
shifts observed for the N8 resonance in tetrazoles, the
spectral ranges listed above do not overlap. This lack of
overlap makes it possible to use the chemical shifts and integral
15
15
intensities of the N8 and N7 resonances as characteristic
criteria for the identification of different tautomeric forms
and measurement of their relative populations in solution.
1
5
made by comparison with N spectra of other compounds
Figure 2b, Table 2). The significant broadening of the C2
The obtained data correspond well with results reported
40,43
(
previously for tetrazolo[1,5-a]pyridine.
work, we were able to individually assign the resonances of the
In the present
1
3
signal observed in the C spectra of 12*a,bT in TFA did not
allow for identification of J couplings from this carbon
CN
15
15
N7 and N8 nuclei and to track changes of their chemical
shifts upon the tetrazole to azide rearrangement.
1
3
atom (Figure 5b). The J
4
and J
values for the C5
C5-N7
C5-N8
nucleus were measured easily. The obtained values of 1.6 and
.7 Hz, respectively, correspond nicely to the J-couplings
observed for the C5 carbon of 12*a,bT in DMSO solution
1.6 and 0.9 Hz, respectively, Table 1). This similarity con-
15
D N NMR spectroscopy did not permit detection of the
0
small amount of the azide forms of 4*a,bT and 9*a,bT in
DMSO solution. In this case, the chemical shift of the bridge-
head atom C2 could be used as an additional criterion for the
identification of the different tautomeric forms. The rela-
(
firms the retention of the bicyclic structure with a [1,5-b] type
of fusion between the azole and azine fragments for 12*a,bT
in TFA (Scheme 4). Similar to other compounds, the transi-
tion of tetrazole 12*a,bT from DMSO to TFA solution was
13
tively large downfield shifts of the C2 resonance (Δδ
6
.5-13.3 ppm) upon tetrazole to azide rearrangement were
1
5
13
observed in both DMSO and TFA (Table 1). As a result, the
resonance of this nucleus did not overlap with the spectral
regions corresponding to tetrazoles (δ 143.1-154.9 ppm) or
associated with upfield shifts of the N8 and C2 reso-
nances (Δδ -18.0 and -2.49 ppm, respectively).
1
3
The analysis of the C multiplets of 12*a,bA in the 1D
NMR spectra revealed the presence of JCN interactions for
only the C2 nucleus, which confirmed the azide structure of
1
3
azides (δ 155.9-161.8 ppm) in the 1D C NMR spectrum
Table 1).
Neither the N nor C chemical shifts allow for determi-
(
1
5
13
1
3
C
this compound (Figure 5b). The comparison of the
nation of the type of fusion between the azole and azine rings
in tetrazolo[1,5-b][1,2,4]triazine derivatives 2*T, 9*a,bT, and
chemical shift values between 12*a,bT and 12*a,bA in TFA
solution indicated that the ring opening associated with the
tetrazole to azide rearrangement is characterized by rela-
tively large downfield shifts of the C2 (Δδ 13.32 ppm) and C4
1
3
15
1
terns provides a straightforward way to solve this problem.
2*a,bT. Nevertheless, analysis of C- N J-coupling pat-
The detection of J coupling constants for the C2 and C5
CN
(
(
Δδ 8.27 ppm) signals and a moderate upfield shift of the C5
Δδ -3.03 ppm) resonance (Figure 5b). It should also be
1
atoms in the C NMR spectra of these compounds in
3
1
15
13
DMSO confirmed unambiguously the [1,5-b] type of fusion
0 0
noted that in doubly H, N-decoupled 1D C NMR spec-
trum of a 12*a,bT and 12*a,bA mixture the C11/C15 reso-
nances of phenyl substitutes of both the tetrazole and azide
forms demonstrated additional splitting (data not shown).
These effects can be ascribed to the possible protonation of
(Table 1). The presence of alternative structures 2*T , 9*a,bT ,
(58) Alekseev, S. G.; Torgashev, P. A.; Fedotov, M. A.; Rezvukhin, A. I.;
Shorshnev, S. V.; Belik, A. V.; Charushin, V. N.; Chupakhin, O. N. Chem.
Heterocycl. Compd. 1988, 24, 434–441.
J. Org. Chem. Vol. 75, No. 24, 2010 8495

Deev et al.
JOCArticle
0
and 12*a,bT with the [5,1-c] type of fusion was excluded
by the absence of JCN splitting for the signal of C4 carbon
SCHEME 5. Protonation of Tetrazolo[1,5-b][1,2,4]triazines Is
the Possible First Step in the Tetrazole to Azide Rearrangement
under Acidic Conditions
(see Schemes 2, 3, and 4). Despite the absence of detectable
JCN interactions for the C2 atom of 2*T, 9*a,bT, and 12*a,
bT in TFA solution (see below), these compounds probably
preserve the [1,5-b] structure. This assumption is supported
by the good correspondence between values of the JCN
coupling constants observed for the C5 carbon in DMSO
and TFA and by the absence of JCN couplings for C4
(
Table 1). The presented results indicate that the alternative
0
isomeric structures T with a [5,1-c] type of fusion either do
not form or have very low populations even under acidic
conditions in which the dynamic opening of the azole ring
takes place. At the same time, the formation of a structure
with [5,1-c] type of fusion was observed for azolo-1,2,4-
triazines, which were obtained by cyclization of the azine
The observation of a single but broadened resonance for
1
3
the C2 nuclei of tetrazolazines 2*T, 9*T and 12*T in TFA
(
line-width, Δν ∼2-4 Hz in TFA and ∼0.6-0.8 Hz in
DMSO) indicates that the process of protonation is occur-
ring in the fast to intermediate exchange limit on the NMR
13
time scale. Assuming that the chemical shift of the C2
nuclei in the deprotonated tetrazole is equal to the chemical
shift observed in DMSO, we can conclude that the proton-
ation/deprotonation process is proceeding with a character-
istic time of about or slightly less than 1 ms. This exchange
3
5,59,60
ring to diazotized 5-amino-1,2,4-triazoles or 5-hydra-
6
zino-1,2,4-triazoles. In this case 1,2,4-triazolo[5,1-c][1,2,4]-
1
triazines are thermodynamically more stable than alternative
6
2
bicyclical isomers. Interestingly, the formation of a [5,1-c]
13
15
type of fusion in some of these compounds was confirmed by
qualitative NMR analysis of J
process (proceeding quickly on the time scale of C- N
J-couplings) is most likely responsible for the observed
couplings using selective
N labeling. It is worth noting that structures with the [5,1-c]
CN
15
60
1
suppression of JCN interactions for the tetrazole C2 nuclei
3
type of fusion between azole and azine rings should be
formed transiently during the synthesis of compounds 9*T
and 12*T (Scheme 4).
in TFA solution. It worth noting that attachment of deuterium
(originated from TFA-d) to the N9 atom should lead to
2
13
the appearance of additional H- C J-couplings, which
could also be manifested as visual broadening of the C2
13
As expected, the azide forms were characterized by the
presence of JCN interactions with only the C2 carbon. The
example of 4*a,bT in DMSO (Figure 3a) indicates that diag-
nostic JCN couplings can be detected even for small amounts
of the azide form, which otherwise preclude detection of the
13
resonance. No changes in the line shape of the C2 re-
3
1
sonance were detected in the 1D C spectra measured
with broadband decoupling from deuterium, however, thus
2
13
13
ruling out H- C J-couplings as a possible source of C2
line-broadening.
Taking in mind the relatively large magnitude of solvent-
1
5
corresponding resonances in 1D N NMR spectra. There-
1
3
15
fore, the C- N couplings can be used to determine the
state of the azide-tetrazole equilibrium in a series of tetrazolo-
1
5
dependent shifts of N8 signal, we can propose that species
protonated at the N9 atom (TH, Scheme 5) represent the
major population of the 9*bT and 12*bT tetrazoles in TFA
[1,5-b][1,2,4]triazines and tetrazolo[1,5-a]pyrimidines in both
neutral and acidic media.
41
Protonation of a Specific Nitrogen Atom in the Azole Ring
Is a Possible First Step in the Tetrazole to Azide Rearrange-
ment under Acidic Conditions. The relatively large upfield
solution. According to the data reported by Cmoch et al.,
the azides (A, Scheme 5) in TFA solution also should
experience protonation at some of the nitrogen atoms (N1,
N3, or N6) of the azine ring (AH1, Scheme 5), but the favored
site of protonation for 1,2,4-triazine derivatives was not
1
5
shifts of the N8 signals of tetrazoles 9*bT (Δδ -31.6 ppm)
and 12*bT (Δδ -18.0 ppm) were observed upon changing
solvents from DMSO to TFA (Table 2). According to the
6
3
13
determined unambiguously. The broadening of the C4/C6
signal observed for azidopyrimidine 4*A in TFA solution
confirms the presence of dynamic protonation of N1 and N3
nitrogen atoms. Interestingly the chemical equivalence of
these nitrogens leads to a relatively narrow line (Δν ∼1.0 Hz)
4
3
data reported for tetrazolo[1,5-a]pyridines the shielding of
the N8 nucleus in acidic conditions can be explained by pro-
tonation of a neighboring nitrogen atom in the tetrazole
fragment. Because the N7 nucleus did not exhibit significant
changes in chemical shift upon transition of 2*T, 9*aT, and
1
3
for the C2 resonance. At the same time the observation of
1
3
13
13
1
2*aT from DMSO to TFA, we can conclude that nitrogen
narrow lines (Δν ∼0.6-1.0 Hz) for the C2, C4 and C5
resonances in azides 2*A, 9*A, and 12*A could indicate that
the process of protonation is occurring in the fast exchange
limit on the NMR time scale and/or the deprotonated state
N9 is the most probable site of tetrazoles protonation in
strong acidic TFA (Scheme 5). This assumption is further
supported by the observation of an upfield solvent-depen-
dent shift (Δδ -2.5 to -5.7 ppm) for the resonance of the
neighboring C2 carbon (Table 1). The same changes in
chemical shifts were reported for bridge-head carbon atom
(
A) has vanishing population in TFA solution.
Protonation of the N9 atom of tetrazole should weaken
4
0
the N1-N7 bond, and as a result this shifts the equilibrium
toward the open-chain azide form. At the same time the
protonation of the N1 atom in azide should protect the open-
chain form of the compound from recyclization. Thus we can
envisage the tetrazole to azide conversion in TFA solution as
a multistep process (Scheme 5), in which the initial fast
4
3
in tetrazolo[1,5-a]pyridines.
(
59) Rusinov, V. L.; Petrov, A. Y.; Chupakhin, O. N.; Klyev, N. A.;
Aleksandrov, G. G. Chem. Heterocycl. Compd. 1985, 21, 576–582.
60) Chupakhin, O. N.; Ulomsky, E. N.; Deev, S. L.; Rusinov, V. L.
Synth. Commun. 2001, 31, 2351–2355.
61) Gray, E. J.; Stevens, M. F. G. J. Chem. Soc., Perkin Trans. 1 1976,
492–1496.
62) Schwalbe, C. H.; Lowe, P. R.; Gray, P. R.; Stevens, M. F. G.; Elder,
M. Acta Crystallogr. 1978, B34, 3409–3411.
(
(
1
(
(63) Charushin, V. N.; Alexeev, S. G.; Chupakhin, O. N.; Van der Plas,
H. C. Adv. Heterocycl. Chem. 1989, 46, 74–135.
8
496 J. Org. Chem. Vol. 75, No. 24, 2010

Deev et al.
JOCArticle
(ESI) m/z calcd for C H N N [M þ H]þ 213.0655, found
15
6
protonation of N9 atom is followed by slower opening of the
azole ring. In this case ring opening should give N9 proto-
nated azide (AH), which is certainly unstable and undergoes
a quick prototropy so that the proton would appear to one of
the nitrogen atoms of the azine ring. The partial protonation
of the N1 atom should stabilize the opened form, which
additionally shifts the equilibrium toward azide. In this case
the different stability of tetrazoles 2*T, 4*T, 9*T, and 12*T
in TFA solution (Table 3) could be ascribed to different degree
of N1 protonation in corresponding azides. The observation of
small amounts of azides of 4*A and 9*A in DMSO solution and
very fast ring opening of corresponding tetrazoles in TFA
permit speculation that mechanisms of tetrazole-azide inter-
conversion in both solvents are similar. In this case either pro-
tonation of terazole or protonation of azide or both of these
reactions could be responsible for shift of the azide-tetrazole
equilibrium to the opened form in TFA solution.
9
5
213.0651.
15
Mixture of [ N]-Tetrazolo[1,5-a]pyrimidines (4*a,bT). Method 1.
A solution of 2-hydrazinopyrimidine (3) (0.100 g, 0.91 mmol)
in an HOAc/H O mixture (3:1, 5 mL) was cooled to 0-5 °C, and
aqueous potassium [ N]-nitrite (0.157 g, 1.82 mmol in 2 mL of
water) was added dropwise so that the temperature of the mix-
ture was kept below 5 °C. After stirring for 2 h at 5 °C, 10 mL of
water was added. The resulting solution was extracted with
chloroform (3ꢀ10 mL). Evaporation of the solvent afforded a
mixture of compounds 4*a and 4*bT (6:1, correspondingly;
0.057 g, 51%).
Method 2. Malonaldehyde bis(dimethyl acetal) (0.031 mL,
.24 mmol) was added to a solution of [2- N]-5-aminotetrazole
monohydrate (5*) (0.025 g, 0.24 mmol) in 1 mL of acetic acid.
The reaction mixture was refluxed for 15 min. After cooling to
room temperature, the precipitate was filtered off and dried to
afford a mixture compounds 4*a and 4*bT (1:1, 0.012 g, 41%):
2
15
1
5
0
þ
mp 120 °C; MS (EI) m/z 122 [M ]. Anal. Calcd (%) for C H
4
3-
1
5
N
4
N: C 39,33, H 2.48, N 58,19. Found: C 39.26, H 2,44, N
Conclusion
5
7.98.
Mixture of [ N]-Tetrazolo[1,5-b][1,2,4]triazines (9*a,bT). A
1
5
We reported, by the example of tetrazolo[1,5-b][1,2,4]-
triazines and tetrazolo[1,5-a]pyrimidines, a method describ-
mixture of ethyl cyanoacetate (7) (0.21 mL, 1.93 mmol) and
NaOAc (1.5 g, 18.30 mmol) in water (5 mL) was added to the
resulting solution of diazonium salt 6*. The reaction mixture
was stirred at room temperature for 64 h. The precipitate formed
was filtered off and dried to give a mixture of compounds 9*aT
1
5
ing the synthesis of selectively N-labeled heterocyclic com-
pounds using simple isotopically enriched parent substances.
1
The selective incorporation of the N-label into the azolo
5
1
3
15
þ
core of tetrazoloazines leads to the appearance of C- N
J-couplings, which can be measured easily using conventional
and 9*bT (0.24 g, 60%): mp 171 °C; MS (EI) m/z 210 [M ];
1
5
þ
HRMS (ESI) m/z calcd for C
found 211.0704.
6
H
8
O
N
2 6
N [M þ H] 211.0709,
1
NMR instruments equipped with triple-resonance ( H-
1
5
13
15
Mixture of [ N]-Tetrazolo[1,5-b][1,2,4]triazinones (12*a,bT).
A mixture of ethyl phenyl(formyl)acetate 10 (0.40 g, 2.10 mmol)
and sodium carbonate (0.6 g) in water (3 mL) and ethanol (2 mL)
was added to the resulting solution of diazonium salt 6*. The
resulting solution was stirred at room temperature for 2 h, and
concentrated HCl (1 mL) was added. The reaction mixture was
filtered off from the precipitate formed, which was then dis-
solved in 2 mL of acetic acid. The solution was refluxed for 2 h
and cooled. The precipitate formed was filtered off and dried to
C- N) probes. The obtained data indicate that J cou-
CN
1
5
plings, together with N chemical shifts, provide a valuable
source of additional structural information that can be used
during structure verification and for studies of mechanisms
of chemical transformations (e.g., the tetrazole to azide
rearrangement under acidic conditions). The selective
1
5
N-labeling extends the applicability of direct NMR methods
for studies of heterocyclic compounds with a high abundance
of nitrogen nuclei, in which conventional NMR techniques
give a mixture of compounds 12*aT and 12*bT (0.2 g, 48%): mp
2
þ
1
13
13
13
25 °C; MS (EI) m/z 215 [M ]; HRMS (ESI) calcd for
based on H- C and C- C J-couplings (HMQC, HMBC,
INADEQUATE, etc.) are ineffective due to low proton and
carbon densities.
1
5
þ
C
9
7
H ON
5
N [M þ H] 216.0651, found 216.0647.
Acknowledgment. This work was supported by the Federal
Target Program “Scientific and Science-Educational Personnel
of Innovative Russia” (State contract P2444), Russian Foun-
dation for Basic Research (project no. 10-03-01007), and
President’s of Russian Federation grant (Grant for Leading
Scientific Schools Nsh-65261.2010.3).
Experimental Section
48,64
Compounds 1 and 3 were prepared as previously described.
1
5
0
0
[
1- N]-Tetrazolo[1 ,5 :2,3][1,2,4]triazino[5,6-b]indole (2*T). A
solution of 3-hydrazino-1,2,4-triazino[5,6-b]indole (1) (0.100 g,
.50 mmol) in 4 mL of polyphosphoric acid was cooled to
0
15
0
2
-5 °C, and aqueous potassium [ N]-nitrite (0.090 g, 1.05 mmol in
Supporting Information Available: Detailed experimental
procedures, synthesis of compounds 5* and 6*, crystallographic
information and X-ray data for 2, 4, and 9 in CIF format,
Figures S1-S6, measurement of JCN coupling constants, table
mL of water) was added dropwise so that the temperature of the
reaction mixture was kept below 5 °C. After stirring for 30 min at
room temperature, 5 mL of water was added. The precipitate was
filtered off, washed with cool water, and dried to afford compound
1
1
þ
containing H chemical shift data, 1D H, and representative 2D
2*T (0.060 g, 56%): mp 290 °C; MS (EI) m/z 212 [M ]; HMRS
1
3
13 15
C-HMQC, 2D C-HMBC and 2D N-HMBC spectra of
synthesized compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
(64) Jean-Jacques, C.; Jianmin, F.; Rajender, K.; Kashinath, S.; Zaihui,
Z. Xenon Pharmaceuticals Inc; PCT Int. Appl. WO 2008/121861, 2008.
J. Org. Chem. Vol. 75, No. 24, 2010 8497