SnCl2 mediated efficient N,N-dialkylation of azides to tertiary-amine via potential stannaimine intermediate

Article abstract of DOI:10.1016/j.jorganchem.2005.12.017

A base free one-pot conversion of azides to N,N-dialkylamine is described. A two-step reaction pathway has been postulated invoking the intermediacy of stannaimine. This new carbon-nitrogen bond formation strategy adds to the repertoire of tin(II) chemistry.

Full text of DOI:10.1016/j.jorganchem.2005.12.017

Journal of Organometallic Chemistry 691 (2006) 1525–1530
www.elsevier.com/locate/jorganchem
Note
SnCl2 mediated efficient N,N-dialkylation of azides to tertiary-amine
via potential stannaimine intermediate
*
Ujjal Kanti Roy, Sujit Roy
Organometallics and Catalysis Laboratory, Chemistry Department, Indian Institute of Technology, Kharagpur 721302, India
Received 15 November 2005; received in revised form 8 December 2005; accepted 8 December 2005
Available online 19 January 2006
Abstract
A base free one-pot conversion of azides to N,N-dialkylamine is described. A two-step reaction pathway has been postulated invoking
the intermediacy of stannaimine. This new carbon–nitrogen bond formation strategy adds to the repertoire of tin(II) chemistry.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Stannaimine; Sn–N; Tin(II); Tin(IV); Azides; Amine
1. Introduction
cessfully tested with aliphatic and aromatic azides 1–6,
and alkyl, benzyl, allyl, propargyl halides 7–12. As detailed
later, the effect of catalytic additives had also been tested,
and in case of allyl halides Pd(0) catalyst was found to be
promising in further enhancing the product yield.
Prodrugs (6-AAP) for antiviral nucleosides ara-A are
designed in such a way that it utilizes azide reduction bio-
transformation pathway [1]. On the other hand, tertiary
amino group is often embedded as a structural motif in
various biologically active compounds and natural prod-
ucts [2,3]. Intramolecular N-alkylation strategies are often
used for the synthesis of fused polycyclic heterocycles [4].
RCM of diallylamines is also an efficient approach to pyr-
rolidine derivatives [5]. In our laboratory to study the
metathesis catalysts, we required a method for the facile
construction of N,N-diallyl amino compounds under
base-free condition. Note that primary amines can be easily
allylated in the presence of a base or basic additive; how-
ever, such a method may lead to complicacy in case of
base-sensitive functionalities. Thus, we became interested
to look into an one-pot conversion of azide, immediate pre-
cursor of amine, to the corresponding N,N-diallyl amine in
a neutral medium. In this short communication, we wish to
report our preliminary result on SnCl2 mediated N,N-dial-
kylation of azides (Scheme 1). The reaction has been suc-
2. Results and discussion
In our laboratory, we have been exploring a bimetallic
strategy to generate allyl, allenyl, and propargyl organome-
tallic reagents for new carbon–carbon bond formation [6].
Recent reports on the successful utilization of allylstannane
for the preparation of N-allylsulfonamide from sulfonyl
azides [7], and also of allylzinc for the preparation of N-
allylamines from azides [8] prompted us to devise an
approach towards the N,N-bisallylation of azides via
in situ generated allytin. Surprisingly control experiment
with allyltrihalostannane did not lead to expected allyla-
tion [9].
Upon addition of anhydrous SnCl2 to a solution of azi-
dobenzene in DMSO led to a strong nitrogen evolution at
room temperature. The rate of nitrogen evolution is solvent
dependent, and decreases in the order DMSO > DMF
ꢀ DCM, THF. Transition metal catalyst has little effect
on this decomposition. Furthermore, decomposition rate
is found to be in the order aryl azide > aliphatic
*
Corresponding author. Tel.: +91 3222 283338; fax: +91 3222 282252.
E-mail address: sroy@chem.iitkgp.ernet.in (S. Roy).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.12.017

1526
U.K. Roy, S. Roy / Journal of Organometallic Chemistry 691 (2006) 1525–1530
Table 1
R'
Anhydrous SnCl2
Anhydrous SnCl2 promoted N,N-diallylation of phenyl azide: control
studiesa
R N3
+
2
R'
X
N
R
DMSO, 30-40 oC
R'
R =
Ph
No.
1
R' =
Allyl
No.
7
SnCl2, catalyst
Solvent
13-25
X
Ph-N3
+
Ph N
13
Napth
2
3
4
2-Methylallyl
3-Methylallyl
4-Me-Bn
8
9
1
2
PhCH2
Entry
X
Catalyst (mol%)b
Solvent
Yield (%)
4-OH-C6H4
4-Tol-SO2
4-Me-Bn
10
1
2
3
4
5
6
7
8
9
a
Br
Nil
DMSO
DMSO
DMSO
DMF
H2O
DCM
THF
63
92
64
82
Nil
Trace
Trace
79
Propargyl
Me
5
6
11
12
Br
Cl
Br
Br
Br
Br
Br
Br
Pd2(dba)3 (1)
Pd2(dba)3 (1)
Pd2(dba)3 (1)
Pd2(dba)3 (1)
Pd2(dba)3 (1)
Pd2(dba)3 (1)
CuCl2, 2H2O (30)
CuCl (20)
Scheme 1. SnCl2 mediated N,N-dialkylation of azide.
azide > benzyl azides. We propose the formation of a reac-
tive stannaimine intermediate A for this transformation
(Scheme 2). A similar suggestion is forwarded by Bartra
et al. for the reaction of azide with tin(II)trithiolate in pres-
ence of Et3N wherein intermediate B is postulated (Scheme
3) [10]. Also noteworthy is the isolation of a stannaimine
[(TMS)2N]2Sn@NR (where R = 2,6-iPr2C6H3) by Ossig
et al. from the reaction of tin(II)amide with arylazide
[11]. However, preliminary attempts by us to isolate the
postulated intermediate A by employing coordinated com-
plexes of SnCl2 proved to be futile.
Stannaimines are known to be highly reactive and have
been trapped in a variety of reactions. For example their
reaction with stannylene yields azadistanniridine in a
[2+1] cycloaddition [12] and with azides [2+3] stannatet-
razoles are formed via cycloadditions [12,11]. In absence
of suitable coupling partner dimerization products of stan-
naimines are formed [12,13]. The reactivity of N–Sn(IV) is
also exploited in a number of synthetically useful N–C
bond forming reactions by coupling with various electro-
philes [14].
DMSO
DMSO
71
Condition: azide (1 mmol), SnCl2 (1.2 mmol), allyl halide (3 mmol).
mol% with respect to azide; dba = dibenzylideneacetone.
b
room temperature for 3 h gave rise to N,N-diallyl aniline
13 in 63% yield (Table 1, entry 1). Under identical condi-
tion but in presence of catalytic Pd2(dba)3 (0.01 mmol) a
smooth reaction was observed giving rise to 13 in 92% yield
(entry 2). Among other catalysts, CuCl2 Æ 2H2O and CuCl
were found satisfactory (entries 8 and 9). The effect of tran-
sition metal catalyst in enhancing the yield of allylated
product could well be due to the simultaneous activation
of allyl halide leading to well known p-allyl metal complex,
the latter being a better electrophile compared to allyl bro-
mide itself. Screening of solvents showed that DMSO is
better compared to others (entries 2 and 4–7). Further-
more, allyl bromide was found to be better than allyl chlo-
ride (entries 2 and 3).
This new N-allylation strategy has been further extended
to a number of allyl halides and aryl or benzyl azides giving
rise to good to excellent yields of the corresponding N,N-
diallylated products (Table 2, entries 1–6). Note that in
case of azide 4, selective N-allylation takes place over O-
allylation (entry 6). However, sulfonylazide remain unreac-
tive towards allylation (entry 7).
We made initial attempts to couple the postulated stan-
naimine intermediate A with allyl halide, and the results
from various control experiments have been accrued in
Table 1. Thus, the model reaction of azidobenzene 1
(1 mmol) with anhydrous SnCl2 (1.2 mmol) and allyl bro-
mide 7 (3 mmol) in freshly dry and distilled DMSO at
R
N
N
N
R N N N
SnCl2
R N
Cl
Sn
Cl
-N2
Cl
Cl
Sn
:SnCl2
N R
Cl2Sn
N
R
R N N N
N N
A
Scheme 2. Proposed stannaimine intermediate A from SnCl2 and azide.
N
N
H
N
N
N
N2
-Et3N
R
N N N
R
N
Sn(SPh)3
Sn(SPh)3
R
NH
R
Sn(SPh)3
B
Et3NH
:Sn(SPh)3
Et3NH
Scheme 3. Proposed SnIV–N intermediate from azide and tin(II)trithiolate [10].

U.K. Roy, S. Roy / Journal of Organometallic Chemistry 691 (2006) 1525–1530
1527
Table 2
Anhydrous SnCl2 promoted N,N-diallylation of aryl and benzyl azide R–N3 with allyl bromide R1CH@C(R2)CH2Br using Pd2(dba)3 as catalyst in
DMSOa
Entry
1
Azide R (No.)
Bromide R1/R2 (No.)
Product
N Ph
Pdt. No.
Time (h)
3
Yield (%)
92
Ph (1)
H/H (7)
13
2
Ph (1)
H/Me (8)
14
3
90
Ph
N
3
4
Ph (1)
Me/H (9)
H/H (7)
15
16
3
4
91
87
N Ph
Napth (2)
N
5
6
PhCH2 (3)
H/H (7)
H/H (7)
H/H (7)
17
18
19
5
7
7
72
N Bn
4-OH–C6H4 (4)
81
N
OH
ꢁ
7
4-Tol-SO2 ð5Þ
Nil
O
N S
O
a
Condition: azide (1 mmol), SnCl2 (1.2 mmol), allyl halide (3 mmol), Pd (0.01 mmol).
The generality of the coupling reaction was tested
further with benzyl halides and propargyl bromides as
electrophiles (Table 3). Curiously, these coupling
reactions proceed smoothly even in absence of a transi-
tion metal catalyst leading to high yields of products
(entries 1–5). Alkyl halides are even more reactive, and
in case of methyl iodide reaction proceeds even at 0 ꢁC
(entry 6).
While mechanistic delineation warrants further studies,
a proposal is presented which involves the key stannai-
mine intermediate A discussed earlier (chiefly Scheme
2). According to this proposal (Scheme 4), intermediate
A is involved for the direct electrophilic activation of a
halide (path b), or of a p-allylmetal (path a). Note that
for N-monomethylation using methyl iodide and alkyl
azides in presence of trialkylphosphine, Kato et al. sug-
gested a similar electrophilic attack on iminophospho-
rane intermediate [15].
the key stannaimine intermediate, and explore its reactivity
with other electrophiles.
3. Experimental
1H (200 MHz) NMR spectra were recorded on a
BRUKER-AC 200 MHz spectrometer. Chemical shifts
are reported in ppm from tetramethylsilane with the
solvent resonance as the internal standard (deuterochlo-
roform: d 7.27 ppm). Data are reported as follows: chem-
ical shifts, multiplicity (s = singlet, d = doublet,
t = triplet, q = quartet, br = broad, m = multiplet), cou-
pling constant (Hz). 13C (54.6 MHz) NMR spectra were
recorded on a BRUKER-AC 200 MHz. Spectrometer
with complete proton decoupling. Chemical shifts are
reported in ppm from tetramethylsilane with the solvent
resonance as the internal standard (deuterochloroform: d
77.0 ppm). ESI mass spectra were recorded on a Waters
LCT mass spectrometer. Elemental analyses were carried
out using a CHNS/O Analyzer Perkin–Elmer 2400 Series
II instrument. All reactions were carried out under an
In conclusion, we demonstrated here a base-free proce-
dure for the direct one-pot conversion of azides to N,N-
dialkylated amine. We are currently attempting to isolate

1528
U.K. Roy, S. Roy / Journal of Organometallic Chemistry 691 (2006) 1525–1530
Table 3
Anhydrous SnCl2 promoted alkylation of azide R–N3 with halide in DMSOa
Entry
1
Azide R (No.)
Halide (No.)
Product
Pdt. No.
Time (h)
4
Yield (%)
77
Ph (1)
4-MeC6H4CH2I (10)
20
N
2
4-MeC6H4CH2 (6)
4-MeC6H4CH2I (10)
21
5
71
N
3
PhCH2 (3)
4-MeC6H4CH2I (10)
22
5
80
N
4
5
Ph (1)
HCCCH2Br (11)
HCCCH2Br (11)
23
24
5
5
91
92
N
Napth (2)
N
6b
Napth (2)
Me–I (12)
25
6
51
NMe2
a
Condition: azide (1 mmol), SnCl2 (1.2 mmol), halide (3 mmol).
At 0 ꢁC, MeI (30 mmol).
b
argon atmosphere in flame dried glassware using Schlenk
techniques. Chromatographic purifications were done
with either 60–120 or 100–200 mesh silica gel (SRL).
For reaction monitoring, precoated silica gel 60F254
TLC sheets (Merck) were used. Petroleum ether refers
to the fraction boiling in the range 60–80 ꢁC.
N,N-diallylation of azidobenzene: Azido-benzene 1
(119 mg, 1 mmol) was added to a mixture of anhydrous
SnCl2 (228 mg, 1.2 mmol) in dry DMSO (3 ml) under
argon atmosphere at room temperature and was allowed
to stir (10 min). 3-Bromo-propene 7 (363 mg, 0.26 ml,
3 mmol) and Pd2(dba)3 (10 mg, 0.01 mmol) were added
to it. The solution is stirred for 3 h. Upon completion
(TLC monitoring: silica gel, eluent: n-hexane–EtOAc
9:1), the reaction was quenched by starring with water
and NH4F. The reaction mixture was extracted into ethyl
acetate (3 · 20 ml). The combined ethyl acetate layer was
washed with distilled water (3 · 50 ml.), brine, and dried
over anhydrous magnesium sulfate. Solvent removal
under reduced pressure followed by column chromatogra-
3.1. General procedure
The typical procedures given below were followed in all
cases (please see Supporting information for details). All
products showed satisfactory spectral and analytical data,
and also were compared with authentic compounds wher-
ever possible.

U.K. Roy, S. Roy / Journal of Organometallic Chemistry 691 (2006) 1525–1530
1529
Acknowledgements
- Pd(0)
Pd(0)
X
R
N
Pd
X
We thank CSIR for financial support. Research fellow-
ship to U.K.R. (UGC–SRF) is acknowledged.
- SnX2Cl2
Path a
Appendix A. Supplementary data
R
N
Cl
Cl
Cl
Cl
General procedure, spectral and analytical data of prod-
ucts. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.
2005.12.017.
Sn
Sn
N
R
Path b
References
- SnX2Cl2
R
2R'
R'
X
[1] L.P. Kotra, K.K. Manouilov, E.C. Scott, J.P. Sommadossi, F.D.
Boudinot, R.F. Schinazi, C.K. Chu, J. Med. Chem. 39 (1996) 5202–
5207.
N
R'
[2] Selected examples: (a) Y. Lee, K.Q. Ling, X. Lu, R.B. Silverman,
E.M. Shepard, D.M. Dooley, L.M. Sayre, J. Am. Chem. Soc. 124
(2002) 12135–12143;
Scheme 4. Proposed electrophilic activation of halide from reactive SnIV
N intermediate.
–
(b) B. Pettersson, M. Curvall, C.R. Enzell, Toxicology 23 (1982) 41–55;
(c) B. Pettersson, M. Curvall, C.R. Enzell, Toxicology 18 (1980) 1–15,
and references therein.
[3] Selected examples: (a) E.G. Occhiato, C. Prandi, A. Ferrali, A.
Guarna, P. Venturello, J. Org. Chem. 68 (2003) 9728–9741;
(b) D.L. Boger, W. Yun, H. Cai, N. Han, Bioorg. Med. Chem. 3
(1995) 761–775.
[4] Selected examples: (a) F. Ek, S. Manner, L.G. Wistrand, T. Frejd, J.
Org. Chem. 69 (2004) 1346–1352;
phy over silica gel 60–120 (gradient elution with EtOAc–
hexane 0–1%) afforded diallyl phenyl amine 13, colourless
liquid (159 mg, 92% with respect to azide). 1H NMR
(200 MHz, CDCl3): d 3.92–3.94 (d, 4H, J = 4.9 Hz),
5.14–5.24 (m, 4H), 5.81–5.89 (m, 2H), 6.70–6.74 (m,
3H), 7.17–7.26 (m, 2H). 13C NMR (50.3 MHz, CDCl3):
d 52.74, 112.38, 115.97, 116.32, 129.03, 134.03, 148.67.
ESI-MS for C12H15N [M], [M + H]+ = 174.13. Anal.
Calcd. for C12H15N: C, 83.19; H, 8.73. Found: C,
83.27; H, 8.65%.
(b) E. Boyd, G.S. Coumbarides, J. Eames, R.V.H. Jones, M.
Motevalli, R.A. Stenson, M.J. Suggate, Tetrahedron Lett. 46 (2005)
3473–3478;
(c) A.D.C. Parenty, L.V. Smith, A.L. Pickering, D.L. Long, L.
Cronin, J. Org. Chem. 69 (2004) 5934–5946.
[5] For a recent example please see: Q. Yang, W.J. Xiao, Z. Yu, Org.
Lett. 7 (2005) 871–874.
[6] (a) A. Kundu, M. Prabhakar, M. Vairamani, S. Roy, Organometallics
16 (1997) 4796–4799;
N,N-dipropargylation of azidobenzene: Following iden-
tical method as above using the reagents azido-benzene 1
(119 mg, 1 mmol), anhydrous SnCl2 (228 mg, 1.2 mmol),
DMSO (3 ml), 3-bromo-propyne 11 (0.27 ml, 357 mg,
3 mmol) afforded phenyl-di-prop-2-ynyl-amine 23, colour-
less viscous liquid (154 mg, 91% with respect to azide).
(b) A. Kundu, M. Prabhakar, M. Vairamani, S. Roy, Organomet-
allics 18 (1999) 2782–2785;
(c) P. Sinha, S. Roy, Chem. Commun. (2001) 1798–1799;
(d) M. Banerjee, S. Roy, Chem. Commun. (2003) 534–535;
(e) P. Sinha, S. Roy, Organometallics 23 (2004) 67–71;
(f) M. Banerjee, S. Roy, Org. Lett. 6 (2004) 2137–2140;
(g) M. Banerjee, U.K. Roy, P. Sinha, S. Roy, J. Organomet. Chem.
690 (2005) 1422–1428.
1H NMR (200 MHz, CDCl3):
d
2.26 (t, 2H,
J = 2.3 Hz), 4.13–4.14 (d, 4H, J = 2.3 Hz), 6.87–7.01
(m, 3H), 7.26–7.35 (m, 2H). 13C NMR (50.3 MHz,
CDCl3): d 40.54, 72.85, 79.38, 115.89, 119.95, 129.26,
147.86. ESI-MS for C12H11N [M], [M + H]+ = 170.1.
Anal. Calcd. for C12H11N: C, 85.17; H, 6.55. Found:
C, 85.31; H, 6.59%.
N,N-bis-4-methylbenzylation of benzyl azide: Following
identical method as above using the reagents benzyl
azide 3 (133 mg, 1 mmol), anhydrous SnCl2 (228 mg,
1.2 mmol), DMSO (3 ml), 4-methylbenzyl iodide 10
(696 mg, 3 mmol) afforded benzyl-bis-(4-methyl-benzyl)-
amine 22 (252 mg, 80% with respect to azide). 1H
NMR (200 MHz, CDCl3): d 2.36 (s, 6H), 3.56 (s, 6H),
7.14–7.47 (m, 13H). 13C NMR (50.3 MHz, CDCl3): d
21.1, 57.45, 57.51 126.8, 128.22, 128.82, 128.93, 136.36,
139.72. ESI-MS for C23H25N [M], [M + H]+ = 316.22.
Anal. Calcd. for C23H25N: C, 87.57; H, 7.99. Found:
C, 87.53; H, 7.82%.
[7] H.S. Dang, B.P. Roberts, J. Chem. Soc., Perkin Trans. I 13 (1996)
1493–1498.
[8] H.M.S. Kumar, S. Anjaneyulu, B.V.S. Reddy, J.S. Yadav, Synlett
(1999) 551–552.
[9] Allyltrihalostannane was generated by the method of van Koten
Thoonen, B.J. Deelman, G. van Koten, Tetrahedron 59 (2003)
10261–10268, Reaction of allyltrihalostannane (3 mmol; in DCM)
with azidobenzene (1 mmol) in DCM did not lead to any N-allylated
product. Unreacted azide largely remained, and a small amount of
aniline was formed.
[10] M. Bartra, P. Romea, F. Urpi, J. Vilarrasa, Tetrahedron 46 (1990)
587–594.
[11] For the structure, generation and reactivity of stannaimine please see:
(a) G. Ossig, A. Meller, S. Freitag, R.H. Irmer, Chem. Commun.
(1993) 497–499;
(b) For an excellent review see:W.P. Neumann, Chem. Rev. 91 (1991)
311–334.
[12] H. Grutzmacher, H. Pritzkow, Angew. Chem., Int. Ed. Engl. 30
(1991) 1017–1018.

1530
U.K. Roy, S. Roy / Journal of Organometallic Chemistry 691 (2006) 1525–1530
[13] N. Wiberg, S.K. Vasisht, Angew. Chem., Int. Ed. Engl. 30 (1991)
(d)M.Taddei,A.Papini,M.Fiorenza, A.Ricci,G.Seconi,Tetrahedron
Lett. 24 (1983) 2311–2314;
93–95.
[14] (a) J.C. Pommier, A. Roubineau, J. Organomet. Chem. 16 (1969) 23–
(e) M. Fiorenza, A. Ricci, M. Taddei, D. Tassi, G. Seconi, Synthesis
(1983) 640–641;
25;
(b) J.C.Pommier,A.Roubineau,J.Organomet.Chem.17(1969)25–27;
(c) J.C. Pommier, A. Roubineau, J. Organomet. Chem. 50 (1973) 101–
111;
(f) A. Papini, A. Ricci, M. Taddei, G. Seconi, P. Dembech, J. Chem.
Soc. Perkin I (1984) 2261–2265.
[15] H. Kato, K. Ohmori, K. Suzuki, Synlett (2001) 1003–1005.