Facile preparation of protected benzylic and heteroarylmethyl amines via room temperature Curtius rearrangement

Article abstract of DOI:10.1016/j.tetlet.2010.03.101

A step-wise, room temperature procedure for acyl azide formation and the subsequent Curtius rearrangement of phenyl and heteroaryl acetic acids is described. We have developed a protocol for room temperature Curtius rearrangement in MeOH or CHCl₃ that provides an improvement over standard conditions, avoiding the use of additives or heat. This room temperature optimization of the Curtius rearrangement prevents the formation of side products often observed with benzylic acids, allowing access to a variety of benzylic and heteroarylmethyl amines.

Full text of DOI:10.1016/j.tetlet.2010.03.101

Tetrahedron Letters 51 (2010) 2888–2891
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Facile preparation of protected benzylic and heteroarylmethyl amines via
room temperature Curtius rearrangement
Matthew L. Leathen a, Emily A. Peterson b,
*
a University of Michigan, Dept. of Chemistry, 930 N. University Ave. Ann Arbor, MI 48109, United States
b Amgen, 360 Binney St., Cambridge, MA 02142, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A step-wise, room temperature procedure for acyl azide formation and the subsequent Curtius rearrange-
ment of phenyl and heteroaryl acetic acids is described. We have developed a protocol for room temper-
ature Curtius rearrangement in MeOH or CHCl3 that provides an improvement over standard conditions,
avoiding the use of additives or heat. This room temperature optimization of the Curtius rearrangement
prevents the formation of side products often observed with benzylic acids, allowing access to a variety of
benzylic and heteroarylmethyl amines.
Received 22 February 2010
Revised 18 March 2010
Accepted 23 March 2010
Available online 27 March 2010
Ó 2010 Elsevier Ltd. All rights reserved.
The Curtius rearrangement is a powerful and widely used trans-
formation in the field of organic synthesis. The rearrangement con-
verts an acyl azide, which can be derived from the corresponding
carboxylic acid, to an isocyanate, excising a single carbon atom.1
The isocyanate is commonly trapped by an alcohol nucleophile to
generate a carbamate-protected amine. The most commonly used
reagent for this transformation is diphenylphosphorylazide
(DPPA), which allows the direct transformation of carboxylic acids
into carbamates.2 The use of this reagent is potentially complicated
by the high temperatures required to achieve conversion to the de-
sired carbamate, which could compromise the stability of sensitive
functionalities. Furthermore, substrates that contain acidic protons
adjacent to the carboxylic acid undergoing rearrangement, such as
benzylic or malonic acid, are problematic for the Curtius rearrange-
ment using DPPA, often leading to ester byproducts.3 This could
potentially occur through elimination of hydrazoic acid, formation
of the ketene, and subsequent trapping with the alcohol nucleo-
phile (Scheme 1, Eq. 1).4 We observed this side reaction when
attempting to convert acid 4 to carbamate 6 by heating 4 with
DPPA and triethylamine in t-BuOH (Scheme 1, Eq. 2).5 To avoid
the side reaction, we found that sequential formation of the acyl
azide 7 and treatment with BCl3 at rt in CH2Cl2 followed by addi-
tion of t-BuOH led to the desired product in moderate yield
(Scheme 1, Eq. 3).
carboxylic acids would prove particularly useful, especially when
considering the powerful asymmetric hydrogenation methods that
have been developed for accessing a
-chiral benzylic acids.7 Herein
we describe a step-wise procedure for the conversion of benzylic
and heteroarylmethyl acids to carbamate-protected amines via a
room temperature Curtius rearrangement.
Our first task was to screen a variety of Lewis acids and solvents
with the aim of determining the optimal conditions for Curtius
rearrangement of benzylic acyl azides. For this initial study, the
acyl azide 9, derived from 4-methoxybenzoic acid, was employed
The widespread abundance of benzylic amines in biologically
active molecules as well as the use of benzylic amines bearing ste-
reogenic centers as ligands in asymmetric synthesis prompted us
to further explore the optimization of this procedure.6 A general
method for the formation of benzylic amines from readily available
* Corresponding author. Tel.: +1 617 444 5027; fax: +1 617 621 3908.
E-mail address: epeterso@amgen.com (E.A. Peterson).
Scheme 1. Formation of undesired ester byproducts.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.03.101

M. L. Leathen, E. A. Peterson / Tetrahedron Letters 51 (2010) 2888–2891
2889
as our standard substrate, with MeOH as the nucleophile to trap
the isocyanate.8 A wide variety of Lewis acids and solvents were
tested, with no improvement over the initial result. We made a
key observation during these studies that in some instances, acyl
azide 9 was not undergoing rearrangement to the isocyanate until
after the addition of methanol. This observation led us to the dis-
covery that in the presence of MeOH, 9 was cleanly converted to
the desired methyl carbamate 10a after 18 h at room temperature,
in the absence of any additive (Fig. 1).9
There are few reports of low temperature Curtius rearrange-
ments; one notable exception is the work by Lebel and co-workers
where they report a useful one-pot Curtius rearrangement using
TMSN3, Boc2O, and Zn(OTf)2 at 40 °C.10 The authors reveal that a
benzylic acid substrate was problematic, leading to a mixture of
desired product and the anticipated ester byproduct, thus high-
lighting the utility of our discovery. This simple solution to the
problematic Curtius rearrangement of benzylic and heteroaryl-
methyl acyl azides is complementary to existing methods and pro-
vides an alternative procedure for more sensitive compounds that
are unstable to heat or Lewis Acids. For example, the acyl azide de-
rived from 2-(benzofuran-3-yl)acetic acid forms extensive side
products in the presence of Lewis acids at room temperature or
when heated with DPPA, but converts cleanly to carbamate 10i
at room temperature in methanol (Table 1, entry 9).
Figure 1. Room temperature Curtius rearrangement in MeOH.
To determine the generality of the step-wise acyl azide forma-
tion and room temperature Curtius rearrangement in methanol,
the scope of the transformation was investigated (Table 1). A wide
variety of benzylic and heterocyclic acids convert to the corre-
sponding methyl carbamates in good yield (Table 1). Chiral, non-
racemic acids can also be converted to the corresponding protected
amines without racemization of the adjacent stereogenic center
(Table 1, entries 6 and 7). The reaction can be run under an air
atmosphere without strict moisture exclusion, and purification of
the resulting products is simplified by the absence of additives.
In many applications, simply concentrating the reaction mixture
provides product of sufficient purity for subsequent chemical
transformations. Aromatic acids also undergo rearrangement
cleanly in high yield (Table 1, entry 12), however aliphatic acids
provide the corresponding carbamate in poor yields due to com-
petitive hydrolysis of the acyl chloride during acyl azide formation.
Substrates bearing a quaternary carbon adjacent to the acyl azide
undergo the rearrangement at a slower rate, achieving 86% yield
after 12 d. (Table 1, entry 11). This limitation is not especially trou-
bling, since substrates bearing quaternary carbons will not be sus-
ceptible to ester byproduct formation using established
procedures.11 It should also be noted that increasing the reaction
temperature (40–60 °C) can shorten the reaction time considerably
for substrates listed in Table 1, but often leads to increased forma-
tion of undesired byproducts.
Table 1
Acyl azide formation and rt Curtius rearrangement
Entry
1
Time
18 h
Product
Yield (%)a
NHCO2Me
10a
10b
10c
81
MeO
Cl
NHCO2Me
2
3
24 h
48 h
80
78
NHCO2Me
Cl
NHCO2Me
4
5
48 h
24 h
10d
10e
87
85
OMe
NHCO2Me
NHCO2Me
The procedure described above provides a convenient and mild
method for accessing benzylic and heteroarylmethyl amines with-
out forming ester byproducts, however, removal of methyl carba-
6
18 h
10f
70 (>99% ee)
Cl
Table 2
Solvent screen
NHCO2Me
7
8
18 h
24 h
10g
10h
85 (>99% ee)
61
NHCO2 Me
NHCO 2Me
O
9
24 h
48 h
10i
10j
78
75
Entry
Solvent (condition)
Product
% Conv.a
O
1
2
3
4
5
6
7
8
9
THF, MeOH (5 equiv)
p-Dioxane, MeOH (5 equiv)
MeCN, MeOH (5 equiv)
Benzene, MeOH (5 equiv)
CH2Cl2, MeOH (5 equiv)
CHCl3, MeOH (5 equiv)
CHCl3, MeOH (3 equiv)
CHCl3, MeOH (1 equiv)
iPrOH
10a, R = Me
10a, R = Me
10a, R = Me
10a, R = Me
10a, R = Me
10a, R = Me
10a, R = Me
10a, R = Me
10m, R = iPr
10n, R = tBu
10n, R = tBu
37
21
46
68
NHCO2Me
Cl
N
10
O
74
87 (81)b
75
NHCO2Me
11
12
12 d
24 h
10k
10l
86
92
47
48b
28b
Trace
NHCO2Me
Cl
10
11
t-BuOH
CHCl3, t-BuOH (5 equiv)
Br
a
Conversion to product 10 measured by HPLC–LRMS.
Isolated yield.
a
b
Isolated yield starting from corresponding carboxylic acid.

2890
M. L. Leathen, E. A. Peterson / Tetrahedron Letters 51 (2010) 2888–2891
mates can often require harsh basic conditions. Thus, we sought to
further expand this protocol so that alternative alcohol nucleo-
philes could be utilized to form more readily deprotected carba-
mates. A simple solution would be to perform the rearrangement
in t-BuOH to form the Boc carbamate, however, we found that both
iPrOH and t-BuOH provided slow conversion to products 10m and
10n, revealing that the steric nature of the alcohol nucleophile
greatly effects the rate of product formation (Table 2, entries 9–11).
An important consideration was to determine whether the rear-
rangement would take place without using the desired alcohol as
the solvent. This would allow the preparation of alternative carba-
mates that could be deprotected under more mild conditions with-
out using excessive amounts of potentially expensive alcohol
nucleophiles. A variety of solvents were screened using 5 equiv
of MeOH as an additive. Chloroform was uniquely effective in pro-
viding the fastest rearrangement.12 Comparison of entries 6–8 (Ta-
ble 2) reveals that the 5 equiv of the alcohol nucleophile is the
optimal condition to achieve a reasonable yield in 24 h.
Again, the transformation appears to be general for both benzylic
and heteroaryl substrates with less sterically hindered nucleo-
philes.13 Carbamate derivatives bearing a variety of more easily
deprotected side chains were formed in good yield.
In conclusion, a simple step-wise procedure to achieve a room
temperature Curtius rearrangement of benzylic and heteroarylm-
ethyl carboxylic acids is reported. The conditions described herein
provide an alternative to the already described one-pot procedures
that require heating or the use of Lewis acids. It is our hope that
reporting this simple result will enable others to easily prepare
benzylic and heteroarylmethyl carbamates without the formation
of unwanted ester byproducts or other deleterious side reactions.
The mild nature of these conditions and the ease of purification
will hopefully encourage their use in more complicated synthetic
applications.
Experimental section
Using the optimal conditions from Table 2 (entry 6) a variety of
substrates and different alcohol nucleophiles were tested (Table 3).
Representative synthesis of acyl azide intermediates
To a flask under nitrogen atmosphere containing 4-methoxy-
phenylacetic acid (11a, 166.2 mg, 1.0 mmol) was added CH2Cl2
(5 mL, 0.2 M). The resulting solution was cooled to 0 °C and oxalyl
Table 3
Formation of easily deprotected carbamates
chloride (175 lL, 2.0 mmol) was added followed by DMF (39 lL,
0.5 mmol). The solution was allowed to warm to rt and maintained
until gas evolution ceased (ꢀ1 h). The solution was concentrated in
vacuo and the resulting residue was taken up in acetone (5 mL,
0.2 M) and transferred dropwise to a vigorously stirring aqueous
solution (0.4 M) of sodium azide (130.0 mg, 2.0 mmol) at 0 °C.
The resulting solution was maintained at 0 °C for 15 min, at which
time it was partitioned between EtOAc (10 mL) and H2O (5 mL).
CAUTION: the aqueous layer may contain hydrazoic acid (HN3) as
a byproduct.14 The layers were separated and the aqueous layer
was extracted with EtOAc (2 Â 5 mL) and the combined organic
layers were washed with deionized water (10 mL), brine (5 mL),
then dried (Na2SO4) and concentrated in vacuo to provide acyl
azide 9, which could be stored at rt for several hours, or at –
20 °C for several days without decomposition or rearrangement.15
Entry Time R3OH
Product
Yield (%)b
O
O
N
H
OMe
OSEM
OBn
1
2
3
4
18 h
18 h
18 h
18 h
MeOH
SEMOHa
BnOH
10a 77
MeO
MeO
MeO
MeO
N
H
10o 80
10p 82
10q 88
O
N
Preparation of methyl carbamates (Table 2)
H
To a flask charged with acyl azide 9 (1.0 mmol, prepared as de-
scribed above) was added 5 mL of MeOH; the solution was main-
tained at rt for 18 h, at which time it was concentrated in vacuo.
The residue was purified by silica gel chromatography using a gra-
dient of 2–80% EtOAc in hexanes to yield 158.0 mg (81% yield) of
methyl 4-methoxybenzylcarbamate (10a).
O
Allyl-
OH
N
O
O
H
N
H
OSEM
5
6
24 h
72 h
SEMOH
SEMOH
10r
90
Preparation of alternative carbamates (Table 3)
O
MeO
To a flask charged with 2-(6-methoxyubenzofuran-3-yl-)acetyl
azide (0.84 mmol, prepared from 2-(6-methoxybenzofuran-3-
yl)acetic acid (11r) analogously to azide 9) was added CHCl3
(5 mL, anhydrous, stabilized with amylenes)13 followed by 2-(tri-
methylsilyl)ethanol (0.60 mL, 4.2 mmol). The resulting solution
was maintained at rt for 24 h, at which time it was concentrated
in vacuo. The residue was purified by silica gel chromatography
using a gradient of 2% EtOAc in hexanes –100% EtOAc to yield
242 mg (90% yield) of (2-(trimethylsilyl)ethoxy)methyl (6-meth-
oxybenzofuran-3-yl)methylcarbamate (10r).
O
N
OSEM
10s 81
H
N
O
O
7
8
24 h
24 h
SEMOH
SEMOH
10t
76
89
N
H
OSEM
OSEM
S
O
10u 99%
N
H
Supplementary data
ee
a
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.03.101.
SEMOH = 2-(Trimethylsilyl)ethanol.
Isolated yields.
b

M. L. Leathen, E. A. Peterson / Tetrahedron Letters 51 (2010) 2888–2891
2891
and trifluoroethanol led in all cases to ester byproducts in addition to
product 10a.
References and notes
10. (a) Lebel, H.; Leogane, O. Org. Lett. 2005, 7, 4107–4110; (b) Yukawa, Y.; Tsuno,
Y. J. Am. Chem. Soc. 1959, 81, 2007–2012; (c) Newman, M.; Gildenhorn, H. J. Am.
Chem. Soc. 1948, 70, 317–319; (d) Coleman, R. A.; Newman, M. S.; Garrett, A. B.
J. Am. Chem. Soc. 1954, 76, 4534–4538; (e) Fahr, E.; Neumann, L. Angew. Chem.
1965, 77, 591.
1. (a) Harwood, L. M. Polar Rearrangements; Oxford University Press: Oxford,
1992. p. 49; (b) Shioiri, T. Degradation Reactions. In Comprehensive Organic
Synthesis – Selectivity; Trost, B. M., Fleming, I., Eds.; Strategy & Efficiency in
Modern Organic Chemistry; Pergamon Press: New York, 1991; Vols. 6, 4.4, pp
795–828.
2. Ninomiya, K.; Sioiri, T.; Yamada, S. Tetrahedron 1974, 30, 2151–2157.
3. (a) Yamada, S.; Ninomiya, K.; Shioiri, T. Tetrahedron Lett. 1973, 26, 2343–2346;
(b) Ninomiya, K.; Shioiri, T.; Yamada, S. Chem. Pharm. Bull. 1974, 22, 1398–
1404; For an example of a benzylic substrate successfully used with DPPA see:
(c) Spino, C.; Tremblay, M.-C.; Gobdout, C. Org. Lett. 2004, 6, 2801–2804.
4. Lebel, H.; Leogane, O.; Huard, K.; Lectard, S. Pure Appl. Chem. 2006, 78, 363–375.
5. This byproduct was observed by LC–MS.
6. (a) Cimarelli, D.; Fratoni, D.; Palmieri, G. Synth. Commun. 2009, 39, 3184–3190;
(b) Kündig, E. P.; Meier, P. Helv. Chim. Acta 1999, 82, 1360–1370; (c) Corey, E. J.;
Imwinkelried, R.; Pikul, S.; Xiang, Y. B. J. Am. Chem. Soc. 1989, 111, 5493–5495.
7. (a) Ohta, T.; Takaya, H.; Kitamura, M.; Nagai, K.; Noyori, R. J. Org. Chem. 1987,
52, 3174–3176; (b) Kagan, H. B.; Dang, T.-P. J. Am. Chem. Soc. 1972, 94, 6429–
6433; (c) Manimaran, T.; Wu, T.-C.; Klobucar, W. D.; Kolich, C. H.; Stahly, G. P.;
Fronczek, F. R.; Watkins, S. R. Organometallics 1993, 12, 1467–1470.
11. Lebel, H.; Leogane, O. Synthesis 2009, 1935–1940.
12. Although it is possible that rearrangement could be promoted by trace
amounts of HCl present in CHCl3, we found that the product was formed with
equal efficiency in base-washed CHCl3. The hydrogen-bonding capability of
CHCl3 could also be responsible, see: Huang, Y.; Rawal, V. H. J. Am. Chem. Soc.
2002, 124, 9662–9663.
13. Note that the majority of commercially purchased chloroform is stabilized
with added ethanol, even if unspecified. Ethanol can compete with other
alcohol nucleophiles to trap the isocyanate intermediate. We recommend
using anhydrous chloroform stabilized with amylenes to avoid this
complication.
14. See: Wiss, J.; Fleury, C.; Onken, U. Org. Process Res. Dev. 2006, 10, 349–353. and
references therein.
15. Although acyl azides are a widely used intermediate in organic synthesis, the
authors caution that low molecular weight acyl azides can be potentially
explosive if evaporated to dryness. See: Overman, L. E.; Jessup, P. J.; Petty, C. B.;
Roos, J. Org. Synth. 1980, 59, 1. On larger scale, the authors recommend partial
concentration of the acyl azide intermediate, followed by introduction of
MeOH.
8. As
a
control, we tested alternate Curtius conditions (refluxing 4-
methoxybenzoic acid, DPPA, Et3N, and MeOH or t-BuOH), which led to
numerous side products instead of rearrangement product 10a.
9. Attempts to increase the rate of conversion to 10a by addition of catalytic
HCl, and other more acidic alchohol solvents such as hexafluoroisopropanol