A Baker-Venkataraman retro-Claisen cascade delivers a novel alkyl migration process for the synthesis of amides

Article abstract of DOI:10.1016/j.tetlet.2015.02.077

A simple extension of the carbamoyl Baker-Venkataraman rearrangement has been developed. If residual water in the reaction is not strictly excluded a Baker-Venkataraman retro-Claisen cascade takes place, giving amide products, in which an alkyl group apparently migrates between two functionalities of the substrate.

Full text of DOI:10.1016/j.tetlet.2015.02.077

Accepted Manuscript
A Baker-Venkataraman retro-Claisen cascade delivers a novel alkyl migration
process for the synthesis of amides
Dana Ameen, Timothy J. Snape
PII:
S0040-4039(15)00358-5
DOI:
http://dx.doi.org/10.1016/j.tetlet.2015.02.077
TETL 45947
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
7 January 2015
2 February 2015
17 February 2015
Please cite this article as: Ameen, D., Snape, T.J., A Baker-Venkataraman retro-Claisen cascade delivers a novel
alkyl migration process for the synthesis of amides, Tetrahedron Letters (2015), doi: http://dx.doi.org/10.1016/
j.tetlet.2015.02.077
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A Baker-Venkataraman retro-Claisen cascade delivers a novel alkyl migration
process for the synthesis of amides
Dana Ameen1 and Timothy J. Snape2,*
1 Department of Pharmaceutical Chemistry, College of Pharmacy, Hawler Medical University, Erbil 44001, Kurdistan
region, Iraq.
2
School of Pharmacy and Biomedical Sciences, University of Central Lancashire, Preston, PR1 2HE, UK. Tel: +44
(0)1772 895805; E-mail: tjsnape@uclan.ac.uk
Key words Baker-Venkataraman; retro-Claisen; rearrangement; amide
Abstract A simple extension of the carbamoyl Baker-Venkataraman rearrangement has been
developed. If residual water in the reaction is not strictly excluded a Baker-Venkataraman retro-
Claisen cascade takes place, giving amide products, in which an alkyl group apparently migrates
between two functionalities of the substrate.
Acyl migrations permeate the chemical literature, yet most involve heteroatom→heteroatom
rearrangements.1 However, the specific rearrangement which sees the O→C migration of phenyl
esters—the so-called Baker-Venkataraman rearrangement—has also received considerable citations
in organic chemistry, especially since it can be used for the regioselective formation of carbon–
carbon bonds.1 The majority of such reactions cover examples which describe the migration of a
phenyl ester from oxygen to carbon, but the first major variant of the rearrangement was reported
by Snieckus, in which ortho-acyl arylcarbamates 1 were rapidly and efficiently converted into
substituted 4-hydroxycoumarins 3 in 79-95% overall yields, via the rearrangement product 2
(Scheme 1).2
Scheme 1. Snieckus’s carbamoyl variant of the Baker-Venkataraman rearrangement.
Recently we sought to develop a new series of asymmetric reactions by expanding the substrate
scope of the Baker-Venkataraman reaction along with its mechanistically related O→C aryl
migration counterpart, the Truce-Smiles reaction.3 During the course of preliminary development
work, attention turned to Snieckus’s carbamoyl Baker-Venkataraman variant as a way of exploring

an asymmetric variant, but utilising a series of substrates where the 1,3-dicarbonyl products (see 2,
Scheme 1) are less prone to enolisation, and thus less likely to racemise any induced asymmetry.
During the initial stages of screening suitable reaction conditions for the Baker-Venkataraman
rearrangement, a new cascade reaction was identified which saw an alkyl group apparently migrate
from the ketone motif of the starting material to the carbamoyl group of the same molecule; this
communication describes the nature of the proposed Baker-Venkataraman retro-Claisen cascade
mechanism that gives rise to the observed products.
Whilst initially attempting to develop suitable rearrangement conditions on achiral substrates,
carbamates, such as 4 (Scheme 2), were studied and familiarisation of the reaction began by
employing Snieckus’s conditions2a on substrate 4a. Unfortunately, in our hands, this reaction gave a
mixture of products which contained, amongst other things, the desired rearrangement product 5a,
along with significant quantities of salicylic acid (6). Repeating the reaction with carbamoyl
substrates 4b-d again resulted in a mixture of the desired rearrangement products 5b-d and salicylic
acid (6). However, on these repeat runs the unwanted hydrolysis products (7a,b) as well as the
amides (8b-d) were also obtained (Scheme 2). These results were in contrast to those of Snieckus
where the 1,3-dicarbonyl products 2 were obtained in good to excellent yields (Scheme 1, 78-
97%).2a Conversely, in our hands, when using strictly anhydrous conditions, results comparable to
those of Snieckus (Table 1 column (I)) were eventually successful. Based on these conflicting, yet
intriguing results, attempts were made to study the “wet” reaction to see if the unexpected process
could be repeated and made more general through establishing some preliminary substrate scope
and limitations.
Scheme 2. Initial attempts at the carbamoyl Baker-Venkataraman rearrangement.

Upon examining the reaction in more detail, it was observed (Table 1, column II) that heating the
substrates (4a-d) in tetrahydrofuran (0.2 M) at reflux, with sodium hydride (2.5 equivalents), for an
extended period of 24 hours, offered a more reliable way of obtaining a mixture of salicylic acid
(6), the hydrolysis products (7), and amides (8), but with none of the associated rearranged products
(5). Presumably, when the longer reaction time was employed (i.e. 24 h rather than 1-3 h), the
Baker-Venkataraman rearranged products (5a-d) do form in situ but subsequently undergo a second
reaction, reminiscent of an intramolecular retro-Claisen condensation, to give salicylic acid (6) and
amides 8a-d (Scheme 3).
Table 1. Substrates undergoing the carbamoyl Baker-Venkataraman rearrangement (I) and the
carbamoyl Baker-Venkataraman retro-Claisen cascade (II).
(I) “anhydrous”
5 (%)b
74
(II) “wet”
Entry
R1-R3
4a (%)b
27
6 (%)b
45
7 (%)
8 (%)b
tracee
tracee
41
d
1
2
3
4
-
a
b
c
d
32
N.A.c
48-92f
N.A.c
50
-
d
76
42
-
d
94
27
-
tracee
d
Reaction conditions: “anhydrous”: sodium hydride (2.5 eq.), reflux, 1-3 hours, anhydrous THF (0.2
M), 4Å molecular sieves; “wet”: sodium hydride (2.5 eq.), reflux, 24 hours, THF (0.2 M). a
Prepared in one-step from the corresponding phenol (1.0 eq.), carbamoyl chloride (1.5 eq.),
triethylamine (1.5 eq.) and 4-dimethylaminopyridine (0.1 eq.) in CH2Cl2 (0.1 M) at 0 °C; b Isolated
yields; c N.A. = not attempted; d Product not isolated; e In these cases the trace product observed
may be the result of having lost the amide product during reaction work-up, not necessarily that it
was not formed during the reaction; f On certain occasions this compound was observed to cyclise
to give a mixture of products, see Scheme 4.
When using substrates 4a,b,d the reaction did not yield any significant quantities of isolated amide
products (8a,b,d), whilst carbamate 4c gave amide 8c in 41% yield. These results do not necessarily
mean that amides 8a,b,d did not form during the reaction, simply that they are too volatile or water
soluble, to be isolated after aqueous workup. Nevertheless, upon reflection of the proposed
mechanism (Scheme 3), evidence that the amide must have formed at the same level of conversion
as salicylic acid during the reaction is provided by the formation, isolation and characterisation of
salicylic acid itself (6, 27-50%) in each of those reactions (see Scheme 2 and Table 1 (column II)).

Scheme 3. Proposed mechanism of the Baker-Venkataraman retro-Claisen cascade.
Scheme 3 shows that under the conditions employed, a proton would be abstracted by sodium
hydride (or sodium hydroxide formed in situ) to form enolate 9 which in turn attacks the carbonyl
of the carbamate group to form a six-membered cyclic intermediate which would collapse to
generate the Baker-Venkataraman rearrangement product 5. Nucleophilic addition of trace
hydroxide to the rearranged product at the more reactive ketone carbonyl group and consequently a
retro-Claisen carbon-carbon bond cleavage would occur to yield salicylic acid (6) and the amide
derivatives (8). A process which would see the overall transformation (4→8) as the apparent
migration of –CH2R1 from the ketone in the starting material to the carbamoyl group.
As expected,2a when the carbamoyl Baker-Venkataraman rearrangement substrate 4c was heated at
reflux for the shorter time of 1-3 hours, in anhydrous tetrahydrofuran containing sodium hydride
and 4Å molecular sieves, only the rearrangement product 5c (48%) was obtained (Scheme 4, in
addition to product 10 (44%), which is the result of the phenolic –OH in 5c intramolecularly
attacking the amide carbonyl), but no retro-Claisen products (6 and 8c) were observed.
Scheme 4. The products obtained under anhydrous conditions.

Further evidence for the proposed mechanism (Scheme 3) was obtained when the reaction was
performed with 4c, in a THF:MeOH solution and the corresponding methyl salicylate was obtained,
rather than the usual salicylic acid obtained when water is present.
In order to demonstrate the generality of this Baker-Venkataraman retro-Claisen cascade, five more
2′-hydroxypropiophenone derivatives 11a-e were acylated with N-methyl-N-phenylcarbamoyl
chloride to give the corresponding carbamoyl-based Baker-Venkataraman rearrangement substrates
12a-e, compounds from which it is expected that non-volatile, non-water-soluble, and isolable
amide products would form. Repeatedly, it was found that four of the carbamoyl substrates
attempted (12a,b,d,e) underwent the Baker-Venkataraman retro-Claisen cascade to give the
corresponding salicylic acid and amide derivatives (14a,b,d,e and 15a,b,d,e respectively) when
subjected to the “wet” reaction conditions, Scheme 5 and Table 2, column II.
Scheme 5. Reaction conditions: a) “anhydrous” conditions: sodium hydride (2.5 eq.), reflux, 1-3
hours, anhydrous THF (0.2 M), 4Å molecular sieves; b) “wet” conditions: sodium hydride (2.5 eq.),
reflux, 24 hours, THF (0.2 M). TEA = triethylamine; DMAP = 4-dimethylaminopyridine.

Table 2. Carbamoyl Baker-Venkataraman rearrangement substrates undergoing the Baker-
Venkataraman rearrangement only (I), and a Baker-Venkataraman retro-Claisen cascade (II).
(I) “anhydrous”
(II) “wet”
15
Entry
1
R1-R2
12 (%)a,b
13 (%)a
14 (%)a, d
(%)a, d
63
22
25
43
85
60
a
2
3
4
5
96
50
91
99
N.A.c
N.A.c
0f
b
c
d
d
-
-
0
0e
d
e
58
42
42
Reaction conditions: “anhydrous”: sodium hydride (2.5 eq.), reflux, 1-3 hours, anhydrous THF (0.2
M), 4Å molecular sieves; “wet”: sodium hydride (2.5 eq.), reflux, 24 hours, THF (0.2 M). a Isolated
yields; b For the synthesis of 12 see Table 1; c N.A. = not attempted; d In all cases the remaining
yield is made up of the corresponding hydrolysis products (11). e In the case of substrate 12d, a
different rearrangement product 22 was formed instead of the expected amide 15d (for a detailed
explanation see Scheme 7). f The rearrangement product 22 was isolated in 33% yield.
As can be seen from Table 2 (column II), only substrate 12c (R1 = COCH3, R2 = H) failed to
undergo the Baker-Venkataraman retro-Claisen cascade. Instead the hydrolysed product 11c was
detected as the sole product. This result can be rationalised by the fact that the 1,3-dicarbonyl
system in 12c can form a stabilised enolate (as shown in Scheme 6, 16↔17), therefore, although it
might undergo a slow rearrangement to 18 it would no doubt suffer from a faster hydrolysis step to
19 in the conditions used with residual water (Scheme 6).
O
O
O
H3C
H3C
H3C
stabilised
enolate
N
O
O
N
O
O
N
O
O
OH
CH3
O
O
NaH
Ph
Ph
Ph
CH3
CH3
12c
16
17
fast
hydrolysis
slow rearrangement
OH
O
O
N
OH
O
OH
CH3
Ph
CH3
19
O
CH3
18

Scheme 6. Failure of 12c to undergo the Baker-Venkataraman retro-Claisen cascade.
In the case of substrate 12d, a different rearrangement product was formed (see 22, Scheme 7)
instead of the expected amide 15d. The rationale for the formation of 22 is that the hydroxide ion
may attack the arylketone via route A or route B. Based upon the structure of the isolated product,
the hydroxide attacks more electrophilic benzoyl carbon (route B) during the retro-Claisen step to
give product 22 (33%) and benzoic acid (21). This result was observed only in "anhydrous"
conditions, whereas in our normal "wet" conditions only 8% of 22 was detected, the reduced yield
presumably due to a faster competing hydrolysis of 12d, which gave 11d in 92% yield; a similar
level of reactivity was seen with 12c.
Ph
CH3
O
O
OH
O
N
H3C
N
O
O
OH
Ph
20
Ph
O
12d
O
HO
OH
O
route B
21
+
route A
OH
O
CH3
N
OH
O
O
O
Ph
22
CH3
+
OH
Ph
15d
N
Ph
6
- not detected
Scheme 7. Proposed mechanism for the formation of compound 22.
Whilst compound 15e (Table 2) is a known chymotrypsin inhibitor,4 it was believed that a worthy
test of the method would be to attempt to synthesise 26 (Scheme 8), a known alanyl aminopeptidase
inhibitor,5 that would otherwise require a multistep protocol. Therefore, 2′-hydroxyacetophenone
(7a) was subject to a Truce-Smiles rearrangement,3c which resulted in the α-arylated product 23 in
quantitative yield, after which, carbamoylation gave the Baker-Venkataraman precursor 24 (46%).
Subjecting 24 to the standard conditions (2.5 eq. NaH, “wet” THF 0.2 M, reflux, 24 h) then
afforded a separable mixture of the Baker-Venkataraman rearranged product 25 (47%) and the
desired retro-Claisen cascade product 26 (25%), the low yield presumably due to incomplete retro-
Claisen of 25 during the reaction.

Scheme 8. Synthesis of a known alanyl aminopeptidase inhibitor (26). Reaction conditions: a) 7a (1
eq.), 1-fluoro-4-nitrobenzene (1 eq.), K2CO3 (2.5 eq.), DMSO (0.1 M), 60 ○C, 24 hours; b) 23 (1
eq.), 4-morpholinecarbonyl chloride (1.5 eq.), triethylamine (1.5 eq.) and 4-dimethylaminopyridine
(0.1 eq.), dichloromethane (0.1 M), room temp., 16 hours; c) 24 (1 eq.), sodium hydride (2.5 eq.),
reagent grade tetrahydrofuran (0.2 M), reflux, 24 hours.
Conclusion
In conclusion, it has been shown that the carbamoyl variant of the Baker-Venkataraman
rearrangement gives retro-Claisen cascade products if the anhydrous conditions of the reaction are
not carefully controlled. Rather than the expected products resulting from carbamate hydrolysis that
are obtained with the analogous ester and carbonate compounds,6 when residual water is present in
the carbamoyl variant, a Baker-Venkataraman retro-Claisen cascade takes place giving amide
products in which an alkyl group apparently migrates from the ketone motif of the starting material
to the carbamoyl group of the same molecule.
Acknowledgements
The authors would like to thank Hawler Medical University, Kurdistan, Iraq and the EPSRC UK
National Mass Spectrometry Facility at Swansea University, UK for performing the accurate mass
analysis.
References
(1)
(2)
Ameen. D.; Snape, T. J. Synthesis, 2014, In Press (Accepted).
(a) Kalinin, A. V.; da Silva, A. J. M.; Lopes, C. C.; Lopes, R. S. C.; Snieckus, V.
Tetrahedron Lett. 1998, 39, 4995; (b) Kalinin, A. V.; Snieckus, V. Tetrahedron Lett. 1998,
39, 4999; (c) Chauder, B. A.; Kalinin, A. V.; Taylor, N. J.; Snieckus, V. Angew. Chem. Int.
Ed. 1999, 38, 1435.

(3)
(a) Ameen, D.; Snape, T. J. Eur. J. Org. Chem. 2014, 1925; (b) Snape, T. J. Synlett 2008,
2689; (c) Snape, T. J. Chem. Soc. Rev. 2008, 37, 2452.
(4)
(5)
Baker, B. R.; Hurlbut, J. A. J. Med. Chem. 1967, 10, 1129.
Ansorge, S.; Bank, U.; Nordhoff, K.; Striggow, F.; Taeger, M. Novel alanyl-amino
peptidase inhibitors for functionally influencing different cells and treating immunological,
inflammatory, neuronal, and other diseases, WO 2005037257 A2, 2005.
Ameen, D. M. H., Ph.D. Thesis, University of Central Lancashire 2014.
(6)

alkyl migration
one-step
O
H3C
N
Ph
O
O
OH O
O
NaH, THF
trace H2O
R
R
CH3
+
N
Ph
OH
Baker-
Venkataraman
rearrangement
OH O
O
NaH,
THF
retro-Claisen
H2O
CH3
N
Ph
R
isolable intermediate
GRAPHICAL ABSTRACT