1,2-H shift in copper-chlorocarbenoid intermediate during CuCl/bpy-promoted stereoselective dechlorination of 2,2,2-trichloroethyl alkyl ethers to (Z)-1-alkoxy-2-chloroethenes

Article abstract of DOI:10.1021/ol8006524

(Chemical Equation Presented) Reaction of 2,2,2-trichloroethyl alkyl ethers with 2 molar equiv of CuCl/bpy in refluxing DCE yielded (Z)-1-alkoxy-2- chloroethenes stereoselectively as the major product via 1,2-H shift in copper-chlorocarbenoid intermediate. 2,2,2-Trichloroethyl carboxylates undergo a radical 1,2-acyloxy shift under similar conditions.

Full text of DOI:10.1021/ol8006524

ORGANIC
LETTERS
2008
Vol. 10, No. 11
2243-2246
1,2-H Shift in Copper-Chlorocarbenoid
Intermediate during CuCl/bpy-Promoted
Stereoselective Dechlorination of
2,2,2-Trichloroethyl Alkyl Ethers to
(Z)-1-Alkoxy-2-chloroethenes
Ram N. Ram* and T. P. Manoj
Department of Chemistry, Indian Institute of Technology, Delhi, Hauz Khas,
New Delhi-110016, India
rnram@chemistry.iitd.ernet.in
Received March 21, 2008
ABSTRACT
Reaction of 2,2,2-trichloroethyl alkyl ethers with 2 molar equiv of CuCl/bpy in refluxing DCE yielded (Z)-1-alkoxy-2-chloroethenes stereoselectively
as the major product via 1,2-H shift in copper-chlorocarbenoid intermediate. 2,2,2-Trichloroethyl carboxylates undergo a radical 1,2-acyloxy
shift under similar conditions.
Recently, we observed a facile dechlorinative rearrangement
of 2,2,2-trichloroethyl carboxylates with 2 equiv of CuCl/
bpy to give 1-acyloxy-1-chloroalkenes.1 The reaction was
proposed to involve a radical 1,2-acyloxy shift in conformity
with the well-known propensity of trichloromethyl com-
pounds to generate radicals under these conditions2 and that
of radicals to undergo this rearrangement, which is known
as Surzur-Tanner rearrangement.3 Subsequently, Falck4a and
earlier Takai et al.5 reported the same transformation with 3
and 4 equiv of CrCl2 and proposed a mechanism involving
1,2-acyloxy shift in a chromium-carbenoid intermediate.
Falck et al. also reported stereospecific 1,2-H shift in the
carbenoid intermediate generated from 2,2,2-trichloroethyl
alkyl ethers4a,b and trichloromethylalkanes4b,c under the same
conditions to form (Z)-1-alkoxy-2-chloroalkenes and (Z)-
chloroalkenes, respectively. A rationale based on the steric
effect of chromium has been provided for the Z-specificity.
In the case of trichloroethyl carboxylates5 and trichloro-
methylalkanes,4c the chromium carbenoid intermediate could
be intercepted by nucleophilic addition to aldehydes. Al-
though formation of a carbene or copper-carbenoid inter-
mediate in the reaction of compounds containing a trichlo-
romethyl group with CuCl/bpy has not been reported despite
a large number of reported experiments on CuCl/bpy-
catalyzed reactions of such compounds,2 there are a few
reports on the formation of copper-carbenoids during the
reaction of some gem-dichloro and trichloromethyl com-
pounds with metallic copper and Cu2O.6 Therefore, it was
(1) Ram, R. N.; Meher, N. K. Org. Lett. 2003, 5, 145–147.
(2) (a) Clark, A. J. Chem. Soc. ReV. 2002, 31, 1–11. (b) Ram, R. N.;
Kumar, N. Tetrahedron Lett. 2008, 49, 799–802. (c) Ram, R. N.; Charles,
I. Chem. Commun. 1999, 2267–2268. (d) Destarac, M.; Matyjaszewski, K.;
Boutevin, B. Macromol. Chem. Phys. 2000, 201, 265–272.
(3) (a) Beckwith, A. L. J.; Crich, D.; Duggan, P. J.; Yao, Q. Chem.
ReV. 1997, 97, 3273–3312. (b) Lacote, E.; Renaud, P. Angew. Chem., Int.
Ed. 1998, 37, 2259–2262.
(4) (a) Bejot, R.; Tisserand, S.; Reddy, L. M.; Barma, D. K.; Baati, R.;
Falck, J. R.; Mioskowski, C. Angew. Chem., Int. Ed. 2005, 44, 2008–2011.
(b) Baati, R.; Barma, D. K.; Krishna, U. M.; Mioskowski, C.; Falck, J. R.
Tetrahedron Lett. 2002, 43, 959–961. (c) Baati, R.; Barma, D. K.; Falck,
J. R.; Mioskowski, C. J. Am. Chem. Soc. 2001, 123, 9196–9197.
(5) Takai, K.; Kokumai, R.; Nobunaka, T. Chem. Commun. 2001, 1128–
1129.
(6) (a) Tezuka, Y.; Hashimoto, A.; Ushizaka, K.; Imai, K. J. Org. Chem.
1990, 55, 329–333. (b) Leonel, E.; Lejaye, M.; Oudeyer, S.; Paugam, J. P.;
Nedelec, J.-Y. Tetrahedron Lett. 2004, 45, 2635–2638.
10.1021/ol8006524 CCC: $40.75
Published on Web 05/06/2008
2008 American Chemical Society

considered necessary to re-examine the mechanism of the
CuCl/bpy-promoted rearrangement of 2,2,2-trichloroethyl
carboxylates and also to investigate the reactions of 2,2,2-
trichloroethyl ethers and trichloromethylalkanes under these
conditions in order to discern more information about the
similarities and differences between CuCl/bpy- and CrCl2-
promoted reactions. The possibility of a rare7 radical 1,2-H
shift in trichloroethyl ethers under these seemingly favorable
conditions as well as the advantage of lower amounts of the
milder reducing agent CuCl/bpy that might be required in
these reactions were also envisioned.
Table 1. Copper(I)-Promoted Dechlorinative 1,2-H Shift in
2,2,2-Trichloroethyl Ethers 1a
isolated yield
2:3
entry ether 1 solvent time (h)
(%) 2
(%) 3 ratiob
1
2
3
4
5
6
7
8
a
b
c
d
e
f
g
h
a
a
a
a
a
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
CHCl3
PhH
THF
DCE
DCE
3
3
3
70
60
70
71
6
17
6
86:14
78:22
87:13
88:12
87:13
85:15
85:15
100:0
62:38
80:20
88:12
61:39
63:37
3
6
15
14
14
14
40
24
16
3
71c
82c
83c
Thus, the reaction of the trichloroethyl ethers 1 (Scheme
1) with CuCl/bpy (equimolar mixture) in DCE at reflux under
72c
9
16:84d
67:33d
89:11d
60c, e
10
11
12
13
Scheme 1
.
Reaction of 2,2,2-Trichloroethyl Ethers 1 and DDT
4
51:49d, f
with 2 molar equiv of CuCl/bpy
a All of the experiments were performed in refluxing solvent under a
nitrogen atmosphere. b Calculated by 1H NMR. c Total isolated yield of (2
d
+ 3). (2 + 3):1 as calculated by H NMR. e With 2 molar equiv of Cu
1
f
powder/bpy (1:1 mixture) in place of CuCl/bpy. With N1,N1,N2,N3,N3-
Pentamethyldiethylenetriamine ligand in place of bpy.
less stereoselective and gave the products in lower yields
(entry 12). The reaction with N1,N1,N2,N3,N3-pentamethyldi-
ethylenetriamine, an efficient ligand in CuCl-catalyzed
halogen atom transfer radical cyclization (HATRC),2a in
place of bpy was found to be slower and less stereoselective
(entry 13). The reaction of the dideuteriated ether 1a(d2) did
not show any exchange of deuterium with proton in the
rearranged enol ethers, as observed by Falck et al.4a As
expected, CuCl/bpy was found to be milder than CrCl2
because, contrary to CrCl2-promoted reactions,4c a hydro-
carbon analogue, 1,1,1-trichlorononane, did not form the
expected alkenes under similar conditions and the trisubsti-
tuted ethenes were formed with significantly slower rate than
a nitrogen atmosphere afforded an isomeric mixture of (Z)-
and (E)-1-alkoxy-2-chloroethenes 2 and 3, respectively, by
Z-selective dechlorinative 1,2-H shift. Although the yields
of the products were comparable to that of the CrCl2-
promoted reactions,4a,b the stereoselectivity was observed to
be lower. The reaction required 2 molar equiv of CuCl/bpy
for completion. The disubstituted ethenes were separated by
column chromatography, but the trisubstituted ethenes were
relatively less stable8on the column and were isolated as a
mixture of Z and E isomers. The structures of the products
were determined by IR and 1H and 13C NMR spectroscopy.
The configurations of the disubstituted alkenes 2a-d and
3a-d were established by 1H NMR and that of the
trisubstituted alkene 2e, 3e, 2f, 3f, and 2h by NOE. The
reaction gave better results in DCE in terms of product yield
and reaction time than the other solvents investigated such
as CHCl3, benzene, and THF (Table 1, entries 9-11). The
reaction of 1a with copper powder in place of CuCl was
Scheme 2. Mechanism of Dechlorinative 1,2-H Shift
(7) For the rarity of radical 1,2-H shift in solution, see: (a) Freidlina,
R. K.; Kost, V. N.; Khorlina; M. Y., Russ. Chem. ReV. 1962, 31, 1–18. For
some other examples, see: (b) Lendvay, B. V. G.; Kortvelyesi, T.; Seres,
L. J. Am. Chem. Soc. 1996, 118, 3006–3009. (c) Kunz, M.; Retey, J. Bioorg.
Chem. 2000, 28, 134–139.
the disubstituted ones as against nearly equal rates with
CrCl2.4a Consequently, easily reducible groups, such as nitro,
could be safely tolerated (entry 2). Understandably,9 this
(8) The compounds tend to decompose to the corresponding aracyl
chlorides on standing in open atmosphere.
2244
Org. Lett., Vol. 10, No. 11, 2008

aspect of selectivity was not examined for CrCl2-promoted
reactions. However, 1,1,1-trichloro-2,2-bis-(4-chlorophe-
nyl)ethane 4 (DDT) (Scheme 1), with two aromatic rings at
the migration origin, gave the expected product 1-chloro-
2,2-bis-(4-chlorophenyl)ethene 5 in 81% yield. 2,2-Dichlo-
roethyl benzyl ether was found to be refractory under similar
reaction conditions, as observed by Falck et al.4a
A mechanism involving 1,2-H shift in the copper-carbenoid
intermediate 7 (Scheme 2) formed by successive abstraction
of two chlorine atoms by CuCl/bpy through a free or, more
preferably a copper-associated, radical 6 (vide infra) has been
proposed for the reaction. The intervention of a free or
copper-associated radical 6 was inferred by the fact that the
reaction of 1a was inhibited by excess of TEMPO (2 molar
equiv). Further, 2,2,2-trichloroethylcinnamyl ether 82c (Figure
1), the vinylogous analogue of 1a, and other trichloroethyl
The evidence for the involvement of a carbene, or more
preferably, the copper-carbenoid intermediate 7 (vide infra)
was obtained when the reaction of 1a was performed in the
presence of a protic quencher with a view to ascertain
whether this association is organometallic type or coordina-
tion through a heteroatom. Thus, the reaction of 1a in the
presence of 20 molar equiv of methanol yielded the dimethyl
acetal 12 (Scheme 2) as the major product, about 5 times of
the rearranged products 2a and 3a, along with small amounts
of the unreacted 1a and some other minor unidentified
products (Table 2, entry 1). The structure of the acetal 12
was confirmed by comparison with an authentic sample. It
did not arise from the rearranged products 2a and/or 3a
because the latter were found to be stable under the
conditions as well as toward the action of CuCl2/bpy or its
equimolar mixture with CuCl/bpy, even in a 1:1 v/v mixture
of DCE-MeOH. The reaction of 1a in i-PrOH DCE (1:1
v/v) (Scheme 2) yielded the corresponding acetal 13 (Table
2, entry 2). However, the reaction of 1a in the presence of
a lower amount (2 molar equiv) of methanol gave only 2a
and 3a (85:15) in 72% combined isolated yield. No acetal
could be detected. The formation of the acetals 12 and 13
from 1a indicated that 1,2-H shift in the ethers might be
takingplacethroughachlorocarbeneorcopper-chlorocarbenoid
intermediate which, in the presence of an alcohol, underwent
insertion into the O-H bond of the alcohol followed by
solvolysis of the R-chloro ether thus formed to give the
acetal.12 In contrast, only benzoyloxy shift occurred in the
case of the trichloroethyl benzoate 14 to give 15 (Figure 1)
even in a (1:1 v/v) mixture of DCE-MeOH (Table 2, entry
3). 2,2,2-Trichloroethylcinnamyl ether 8 also afforded only
the HATRC product 9 (Figure 1) on reaction with CuCl/
bpy in refluxing DCE-MeOH (1:1 v/v) (Table 2, entry 4);
no acetal was detected in the reaction mixture. The insertion
of free chlorocarbenes into O-H bond of methanol is known
to be a fast diffusion-controlled reaction.10,11 The fact that
this was not observed in the case of 2,2,2-trichloroethyl
benzoate 14 and 2,2,2-trichloroethylcinnamyl ether 8 sug-
gested that probably the acyloxy shift and HATRC involve
a nonorganometallic type copper-associated/free-radical in-
termediate as proposed earlier.1,2 The fact that no acetal was
detected in the reaction of 1a at lower concentration of
methanol (2 mol equiv) indicated that the 1,2-H shift was
also very fast probably due to the known powerful rate
accelerating effect of the bystander alkoxy group on the
1,2-H shifts in carbenes12 Consequently, intramolecular C-H
insertion in the case of 1c and 1d was not observed and
attempts to intercept the carbenenoid intermediate in the
reaction of 1a by cyclopropanation with excess of cyclo-
hexene or methyl acrylate failed (Table 2, entries 5 and 6).
Intra- or intermolecular cyclopropanation with alkenes was
complicated by a preceding radical reaction, such as
Figure 1. Structures of some relevant compounds.
allylic ethers2b have been shown to undergo HATRC under
similar conditions to give the tetrahydrofurans of the type
9. Therefore, it was tempting to believe that the products
arose by the anticipated radical 1,2-H shift. However, the
extreme rarity of 1,2-H shift in alkyl radicals in solution7
demanded more careful scrutiny. Thus, when the reaction
of 1a with Bu3SnH/AIBN in refluxing benzene was per-
formed with a view to generate a radical intermediate through
a proven and trusted method of free radical generation, no
rearrangement products 2a or 3a could be detected even
under the conditions favorable to 1,2-H shift (slow addition
of Bu3SnH to a solution of 1a at sufficiently high dilution).
Only the elimination product 10 (Figure 1) (at higher
concentration) and the reduction product 11 were formed.
In a competition experiment between CuCl/bpy and Bu3SnH,
the reaction of 1a with Bu3SnH/CuCl/bpy even under the
conditions more conducive to reduction (more concentrated
solution of 1awith all the other reactants taken together)
yielded the rearranged products 2a and 3a (79:21) in nearly
double the amount in moles of the reduction product 11.
These results indicated that either a radical intermediate was
not involved in the 1,2-H shift or it had a different reactivity
from the free radical generated in Bu3SnH reactions. Similar
results were obtained by us1 during the 1,2-acyloxy shift as
mentioned earlier, and it was thought that the difference in
the reactivity of the two types of radicals might be due to
some sort of association of the radical intermediate with
copper in the CuCl/bpy-promoted 1,2-acyloxy shift.1
(10) Griller, D.; Liu, M. T. H.; Scaiano, J. C. J. C. J. Am. Chem. Soc.
1982, 104, 5549–5551
.
(11) For a review, see: (a) Miller, D. J.; Moody, C. J. Tetrahedron 1995,
51, 10811–10843. For some other examples, see: (b) Jiang, N.; Wang, J.;
Chan, A. S. C. Tetrahedron Lett. 2001, 42, 8511–8513. (c) Motschiedler,
K.; Gudmundsdottir, A.; Toscano, J. P.; Platz, M.; Garcia-Garibay, M. A.
(9) For reviews on CrCl2-promoted reductions, see: (a) Ho, T.-L.
Synthesis 1979, 1–20. (b) Hanson, J. R. Synthesis 19, 74, 1–8.
J. Org. Chem. 1999, 64, 5139–5147
.
(12) Nickon, A. Acc. Chem. Res. 1993, 26, 84–89
.
Org. Lett., Vol. 10, No. 11, 2008
2245

Table 2. Experiments toward Mechanistic Investigationa
entry
reactants (molar equiv)/solvent
time (h)
products (% isolated yield)
1
2
3
4
5
6
1a/CuCl/bpy/MeOH (1:2:2:20)/DCE
3
3
3
3
6
3
b
2a/3a/12/1a (11:5:79:5)b
13 (45)
15 (88)
1a/CuCl/bpy (1:2:2)/DCE/i-PrOH (1:1 v/v)
14/CuCl/bpy (1:2:2)/DCE/MeOH (1:1 v/v)
8/CuCl/bpy (1:2:2)/DCE/MeOH (1:1 v/v)
1a:CuCl: bpy: cyclohexene (1: 3: 3: 40)/DCE
1a/CuCl/bpy/methyl acrylate (1:2:2:50)/DCE
9 (73)
2a/3a/9/1a (38:10:13:39)b
2a (19), 3a (3), telomers (major)
a All the experiments were performed in refluxing solvent under a nitrogen atmosphere. Product ratio as calculated by H NMR.
1
cyclization2a–c (Table 2, entry 4), reduction (Table 2, entry
5), or telomerization2d (Table 2, entry 6).
In conclusion, the CuCl/bpy-promoted reaction of 2,2,2-
trichloroethyl ethers is advantageous over the CrCl2-promoted
reaction for stereoselective synthesis of (Z)-2-chloroenol
ethers14 in terms of the amount and chemoselectivity of the
metal promoter used, though the Z-selectivity is lower. The
CuCl/bpy-promoted 1,2-H and acyloxy shifts probably occur
in a carbenoid and radical intermediate, respectively, as
against the reported involvement of carbenoids in both the
reactions promoted by CrCl2. We could not find a literature
precedent for 1,2-H shift under the present reaction condi-
tions.15
Thus, it appears that the formation of the carbene or
copper-carbenoid intermediate occurs stepwise succeeding
the formation of a free radical or copper-associated radical
only in the absence of availability of an easier route for a
radical reaction, such as HATRC or 1,2-acyloxy shift. The
direct formation of chromium-carbenoid intermediate in the
CrCl2-promoted reactions might be due to higher reactivity
of CrCl2 than CuCl/bpy which would give little time to the
radical intermediate, if formed,13 to undergo 1,2-acyloxy
shift. Though the involvement of a free carbene has not been
ruled out, the intermediacy of a copper carbenoid is preferred.
It is less likely for a free carbene to exist in the presence of
CuCl/bpy in the medium. It also seems likely that the
abstraction of the second chlorine atom by CuCl/bpy occurs
from a radical stabilized by association with copper rather
than a more reactive unassociated free radical.
Acknowledgment. We thank the Indian Institute of
Technology, Delhi, for a Teaching Assistantship to M.T.P.
and the Council of Scientific & Industrial Research, New
Delhi, for financial assistance.
Supporting Information Available: Experimental pro-
cedures, spectral data, and NMR spectra of all new and
relevant compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
Unlike chromium-carbenoids, these copper-carbenoids
appear to be less nucleophilic, as attempted trapping of the
copper-carbenoid generated from 1a with 4-nitrobenzalde-
hyde led only to 1,2-H shift and recovery of the aldehyde.
It could also be due to faster 1,2-H shift which precluded its
reaction with the aldehyde.
OL8006524
(14) For their potential as AD prodrugs, see: Yoshimatsu, M.; Sakai,
M.; Moriura, E. Eur. J. Org. Chem. 2007, 498–507.
(13) For CrCl2-promoted radical cyclization of trihaloethyl allyl/prop-
argyl ethers, see: Nakagawa, M.; Saito, A.; Soga, A.; Yamamoto, N.;
Taguchi, T. Tetrahedron Lett. 2005, 46, 5257-5261.
(15) The copper-carbenoid was generated by the reaction of methyl
2,2-dichloropropanoate with copper powder in DMSO; the only reported
example with such a possibility yielded dimerization product (ref 6a).
2246
Org. Lett., Vol. 10, No. 11, 2008