Thiol on silica as a 'catch and release' support for isocyanates to afford ureas

Article abstract of DOI:10.1016/j.tetlet.2007.05.050

Silica-bound thiocarbamates were prepared by Curtius rearrangement of carboxylic acids in the presence of thiol on silica gel. The solid supported thiocarbamates were found to be stable isocyanate equivalents, which upon treatment with amines efficiently afforded di- and tri-substituted ureas. The urea products released from the catch and release support were, in the majority of cases, greater than 95% pure and required no further work up.

Full text of DOI:10.1016/j.tetlet.2007.05.050

Tetrahedron Letters 48 (2007) 4925–4927
Thiol on silica as a ‘catch and release’ support for isocyanates
to afford ureas
Yuri Bolshan, Miroslaw J. Tomaszewski and Vijayaratnam Santhakumar*
Department of Chemistry, AstraZeneca R&D Montre´al, 7171 Frederick-Banting, Montre´al, Quebec, Canada H4S 1Z9
Received 23 April 2007; revised 7 May 2007; accepted 8 May 2007
Available online 16 May 2007
Abstract—Silica-bound thiocarbamates were prepared by Curtius rearrangement of carboxylic acids in the presence of thiol on silica
gel. The solid supported thiocarbamates were found to be stable isocyanate equivalents, which upon treatment with amines effi-
ciently afforded di- and tri-substituted ureas. The urea products released from the catch and release support were, in the majority
of cases, greater than 95% pure and required no further work up.
Ó 2007 Elsevier Ltd. All rights reserved.
The urea functionality is common to many biologically
active compounds.1 The prevalence of this group in
medicinal chemistry may be attributed to the metabolic
stability of the N–C(O)–N linkage and to the large num-
ber of variations possible. Moreover, the urea scaffold is
an attractive target for combinatorial and parallel syn-
thesis approaches since up to four different substituents
are possible on this low molecular weight core.
report an efficient route to ureas that is not reliant on
commercially available isocyanates and one in which
the products released from diversity-building cleavage
reaction do not require further purification.
The Curtius rearrangement reaction is a well-known
method for the synthesis of ureas that is not dependent
on commercially available isocyanates. A carboxylic
acid is converted into the corresponding acyl azide that
undergoes a thermal rearrangement to the isocyanate in
one pot. The isocyanate is then treated with an amine to
afford the desired urea product. As there are over 40,000
commercially available carboxylic acids, an impressive
list of diverse ureas is possible.3 Practically, however,
the Curtius approach involves a purification step, typi-
cally chromatography, of either the intermediate isocya-
nate or the final urea product that detracts from the
potential of preparing analogues quickly. Furthermore,
the instability of isocyanates, especially alkyl substi-
tuted, makes the isolation of the intermediate isocya-
nates a difficult task. It was clear to us that the
Curtius route would be much more practical if the inter-
mediary isocyanates were captured as resin-bound iso-
cyanate equivalents, that upon treatment with amines
afforded ureas.4,5 We are pleased to report that thiol re-
sin, an odourless solid-support, can be used as catch and
release resin to trap isocyanate intermediates as thio-
carbamates and then to release the urea products by
treating the resin-bound thiocarbamates with amines
(Scheme 1).6,7
Among the reported routes to ureas, those based on
solid phase chemistry that include a diversity-building
resin cleavage reaction show great promise for the quick
and efficient synthesis of diverse sets of ureas.2 In these
approaches, the polymer-bound carbamoyl or urea
synthons are easily handled and can be purified from
the excess reagents used to drive reactions to completion
by using simple filtration and washing protocols. Whilst,
a diversity-building cleavage reaction allows for yet fur-
ther diversification of the urea synthon concomitant
with releasing the product from the resin. However,
the structural diversity of ureas that can be prepared
by some of these methods is limited by the number of
commercially available isocyanates.2a–c Many of the
other reported routes are not compatible with the syn-
thesis of basic ureas since strong cationic exchange res-
ins are used in the final workup to scavenge excess
starting materials or reagents.2d–f Herein, we wish to
Keywords: Urea; Solid phase synthesis; Catch and release; Thiocarba-
mate; Curtius rearrangement.
*
In our method development, 3-bromobenzoic acid was
subjected to a Curtius rearrangement with diphenyl
Corresponding author. Fax: +1 514 832 3232; e-mail:
v.santhakumar@astrazeneca.com
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.05.050

4926
Y. Bolshan et al. / Tetrahedron Letters 48 (2007) 4925–4927
Table 2. Variation of acids
O
O
Thiol resin
HNR1R2
R1
RCO2H
N
NHR
S
NHR
Acetonitrile
reflux, 17 h
O
DPPA, NET3
Toluene
HNR1R2
R2
Thiol resin
RCO2H
R
r.t., 2 h and
R2R1N
N
5eq. DPPA
5eq. NEt3
°
CH3CN, 70 °C
100 C, 2 h
H
(5eq.)
S
Toluene, 100 °C
= Resin
NH
R
O
= Resin
Scheme 1.
Entry
1
Amine (HNR1R2)
Acid (RCO2H)
Yielda (%)
77
Cl
Br
CO2H
NH2
Table 1. Optimization of reaction conditions
O
Br
CO2H
Step 2
Step 1
Cl
2
3
65
99
NH2
N
H
Cl
COOH
CO2H
Cl
Thiol resin
S
HN
HN
NH2
5eq. DPPA
5eq. NEt3
Toluene, 100 C
°
Br
NH
O
CH3CN, 70 C
Cl
O
(5eq.)
°
O
CO2H
4
5
6
7
79
80
87
45
Br
N
H
Br
= Resin
Cl
Cl
MeO
MeO
NC
CO2H
CO2H
Entry
Additives (step 2)
Equivalent
Yielda (%)
NH2
1
2
3
4
None
AgOTf
NEt3
—
2
3
44
11
49
90b
O
NEt3
3
N
H
a Isolated yield.
Cl
CO2H
b The mixture was stirred at room temperature for 2 h before heating at
NH2
100 °C in step 1.
Cl
CO2H
NH2
8
9
55
74
O
phosphoryl azide (DPPA) in the presence of thiol
support (1 equiv) (Table 1).8,9 The resulting solid bound
thiocarbamate was treated with 3-chlorobenzyl amine
(0.9 equiv) to give the corresponding urea in 44% yield.
Attempts to boost the yield by using silver triflate as an
activator in step 2 were unsuccessful and a considerable
amount of the symmetrical urea (dibenzyl) by-product
was detected. However, an excellent conversion to the
desired urea was obtained by stirring the mixture of step
1 at room temperature prior to heating and by adding
triethyl amine in step 2. Therefore, in our optimized con-
ditions, the carboxylic acid (3–5 equiv) was treated with
DPPA (3–5 equiv) and triethylamine (3–5 equiv) in tolu-
ene at room temperature for 2 h followed by the addi-
tion of thiol resin (1 equiv) and heating overnight at
100 °C. After washing and drying, the resin was treated
with an amine (0.9 equiv) in the presence of triethyl-
amine (1 equiv) in acetonitrile and stirred overnight.
After simple evaporation, the urea products were iso-
lated with >95% purity in most cases.
CO2H
N
H
a Isolated yield after chromatography, purities >95% by 1H NMR.
using benzoic acids substituted with strong electron
withdrawing groups such as cyano (e.g., 7, 45% yield).
Next, we investigated the scope of amines tolerated in
the diversity-building resin cleavage reaction (Table 3).
Primary amines, including the hindered cyclohexyl
amine, gave excellent yields of di-substituted ureas. In
addition, secondary amines, both cyclic and acyclic,
gave products in high yields and purities. Importantly,
no purification was necessary following cleavage from
the resins as the products were >95% pure. Anilines,
on the other hand, under these conditions gave only
low yields, probably because of the poorer nucleophilic-
ity of these amines. The reactions with anilines were not
aided by the use of a stronger base such as sodium
hydride and in this case, the products were contami-
nated with the symmetrical ureas (bis anilino).
In order to explore the scope of carboxylic acids toler-
ated by this route, 5 different carboxylic acids were trea-
ted with 2 different amines: 3-chlorobenzyl amine, a
representative primary amine and morpholine, a repre-
sentative secondary amine (Table 2). Reactions worked
well for both electron rich and most electron deficient
benzoic acids with an average yield of 78% for 3-chloro-
benzyl amine (e.g., 1, 3, 5 and 8) and 76% for mor-
pholine (e.g., 2, 4, 6 and 9). Poorer yields were realized
In summary, we have demonstrated that silica gel thiol
resin can be used as a catch and release resin for the syn-
thesis of ureas starting from carboxylic acids via the
Curtius rearrangement reaction. Urea products were
obtained with good yields and excellent purities from a

Y. Bolshan et al. / Tetrahedron Letters 48 (2007) 4925–4927
4927
Table 3. Reactions with amines
well as alkylation of the thiocarbamates to allow for the
synthesis of tetra-substituted ureas.
COOH
S
Acknowledgements
HNR1R2
Thiol resin
O
O
NH
5eq. DPPA
5eq. NEt3
Toluene, 100 C
R2R1N
N
°
CH3CN, 70 C
We thank Dr. Yun-Jin Hu and Carmen Leung for their
comments.
H
(5eq.)
°
= Resin
Supplementary data
Entry
1
Amine (HNR1R2)
Yielda (%)
89
Purityb (%)
>95
Cl
General procedure and NMR and LC/MS data for all
the compounds in the tables are available. Supplemen-
tary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2007.05.050.
NH2
MeO
NH2
2
3
92
87
>95
>95
NH2
NH2
References and notes
1. 12,355 compounds containing urea functional group cited
in the MDL Drug Data Report (MDDR) database Ver.
2004.2.
4
5
98
88
>95
>95
NH2
2. Kenner safety catch linker: (a) Fattori, D.; D’Andrea, P.;
Porcelloni, M. Tetrahedron Lett. 2003, 44, 811; Thiohydr-
oxamic acid linker: (b) Whitehead, D. M.; Jackson, T.;
McKeown, S. C.; Routledge, A. React. Funct. Polym.
2002, 52, 81; Phoxime resin: (c) Scialdone, M. A.
Tetrahedron Lett. 1996, 37, 8141; Carbamoyl linker: (d)
Lee, S.-H.; Matsushita, H.; Koch, G.; Zimmermann, J.;
Clapham, B.; Janda, K. D. J. Comb. Chem. 2004, 6, 822;
Benzotriazole linker: (e) Paio, A.; Crespo, R. F.; Seneci,
P.; Ciraco, M. J. Comb. Chem. 2001, 3, 354; Phoxime
resin: (f) Hamuro, Y.; Marshall, W. J.; Scialdone, M. A. J.
Comb. Chem. 1999, 1, 163; Thiophenoxy carbonyl linker:
(g) Dressman, B. A.; Singh, U.; Kaldor, S. W. Tetrahedron
Lett. 1998, 39, 3631.
NH
6
7
73
98
>95
>95
N
H
O
8
9
85
45
>95
>95
N
H
NH2
3. ACDfind: a comprehensive database of commercially
available compounds from MDL. Ver. 2004.2.
MeO
4. Isocyanates trapped by amine resin: Migawa, M. T.;
Swayze, E. E. Org. Lett. 2000, 2, 3309.
NH2
10
52
65
5. Isocyanates trapped by alcohol resin: Sunami, S.; Sagara,
T.; Ohkubo, M.; Morishima, H. Tetrahedron Lett. 1999,
40, 1721.
a Isolated yield.
b Determined by LC/MS and 1H NMR after removal of solvent.
6. For use of odourless resin bound thiols in combinatorial
chemistry applications, see: (a) Gayo, L. M.; Suto, M. J.
Tetrahedron Lett. 1997, 38, 211; (b) Kaljuste, K.; Tam, J.
P. Tetrahedron Lett. 1998, 39, 9327; (c) Vlattas, I.;
Dellureficio, J.; Dunn, R.; Sytwu, I. I.; Stanton, J.
Tetrahedron Lett. 1997, 38, 7321; (d) Dodd, D. S.;
Wallace, O. B. Tetrahedron Lett. 1998, 39, 5701.
7. Solution phase conversion of thiocarbamates to ureas, see:
Anbazhagan, M.; Rakeeb, A.; Deshmuk, A. S.; Rajappaa,
S. Tetrahedron Lett. 1998, 39, 3609.
variety of acids and amines making this a general meth-
od for the synthesis of di- and tri-substituted ureas that
is amenable to combinatorial chemistry. It is worthy to
note that our thiocarbamate resin can be stored at room
temperature in a closed container without any decompo-
sition and as such is a stable-canned version of
isocyanates.10
8. Kim, D.; Weinreb, S. M. J. Org. Chem. 1978, 43, 125.
9. Thiol-3 resin, a functionalized silica gel resin purchased
from Silicycle was used.
10. No loss in yield was observed in step 2 after storing the
resin at room temperature for over 2 months.
Currently we are in the process of evaluating the possi-
bility of expanding the chemistry of our polymer-bound
thiocarbamates. In particular we are interested in evalu-
ating the stability of this group to other reaction types as