Selective electrochemical deprotection of cinnamyl ethers, esters, and carbamates

Article abstract of DOI:10.1016/S0040-4039(03)00009-1

Electrochemical deprotection of the cinnamyl moiety from ethers, esters, and carbamates was studied with the focus on O- versus N- selectivity as well as selectivity over allyl or benzyl systems.

Full text of DOI:10.1016/S0040-4039(03)00009-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1575–1578
Selective electrochemical deprotection of cinnamyl ethers, esters,
and carbamates
Jeff Hansen,a Stanley Freemanb,* and Tomas Hudlickyb
aDepartment of Chemistry, DePauw University, Greencastle, IN 46135, USA
bDepartment of Chemistry, University of Florida, Gainesville, FL 32611, USA
Received 23 November 2002; revised 20 December 2002; accepted 30 December 2002
Abstract—Electrochemical deprotection of the cinnamyl moiety from ethers, esters, and carbamates was studied with the focus on
O- versus N- selectivity as well as selectivity over allyl or benzyl systems. © 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
Extending our work on selective electrochemical depro-
tection of cinnamyl ethers, we have investigated the
selectivity of cinnamyl ethers in the presence of benzyl
ethers (cinnamyl versus benzyl), cinnamyl ethers in the
presence of cinnamoyl esters (cinnamyl versus cin-
namoyl), cinnamyl ethers in the presence of cinnamyl
amine (O-cinnamyl versus N-cinnamyl), and cinnamyl
esters in the presence of cinnamyl amides (cinnamate
versus cinnamide).
Although the use of the cinnamyl ether as an alcohol
protecting group (with electrochemical cleavage) was
initially reported in 1968, it has been used only infre-
quently.1 It has also been used as a carboxylic acid
protecting group.2 The allyl group has been used more
often, and many methods have been developed for its
deprotection.3 Among these are several that employ a
palladium catalyst and a nucleophile4 to trap the allyl
group from the p allyl palladium intermediate. Another
common approach to the deprotection of allylic pro-
tecting groups is isomerization of the double bond to a
vinyl ether followed by acid hydrolysis.5 Allyl and
cinnamyl groups have also been removed electrochemi-
cally using a Pd catalyst.6 We have previously reported
the selective electrochemical deprotection of cinnamyl
ethers in the presence of allyl ethers in conduritol-like
systems.7 The electrochemical method is remarkably
selective in removing exocyclic cinnamyl groups over
those in which the cinnamyl double double bond is
contained in the ring. Allyl ethers are inert under such
conditions.
We studied the electrochemical deprotection of the
cinnamyloxycarbonyl group with focus on cinnamyl
versus allyl selectivity, and O- versus N- selectivity, as
an alternative to chemical methods for COC deprotec-
tion, which require stoichiometric amounts of reagents.
Finally, we achieved selective reduction of a vinyl bro-
mide in the presence of a cinnamyl ether. Herein we
report the results of these studies.
2. Results and discussion
2.1. Cinnamyl ethers and esters
The allyloxycarbonyl (Alloc) and cinnamyloxycarbonyl
(COC) moieties are used as protecting groups for
amines. In the case of the COC group, the protecting
group is introduced via cinnamyloxycarbonylbenzotria-
zole (COC-OBT)8 or cinnamyloxycarbonyl imidazole
(COC-Im).9 Removal of COC9 or Alloc3,10 groups is
similar to removal of the ethers. To our knowledge no
methods to deprotect selectively either the cinnamyl
group or the allyl group in the presence of the other
have been reported, save our earlier publication.7
The selectivity of electrochemical cinnamyl deprotec-
tion in the presence of allyl ethers7 prompted us to
investigate cinnamyl versus benzyl selectivity (Table 1).
Cyclic voltammetry (CV) of benzyl ether 2 (for details
of CV conditions see the Experimental section) exhib-
ited a weak reduction wave at −3.1 V. CV of 1,
containing both benzyl and cinnamyl ethers, exhibited
reduction waves at −2.8 and −3.1 V. We initially
assumed that the wave at −2.8 V was due to the
cinnamyl ether and the wave at −3.1 V was due to the
benzyl ether. However, the CV of 3, with only the
* Corresponding author.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(03)00009-1

1576
J. Hansen et al. / Tetrahedron Letters 44 (2003) 1575–1578
Table 1. Electrochemical cleavage of cinnamyl ethers
cinnamyl ether, was identical to that of 1. Electrochem-
ical reduction of 1 (for details of electrolysis conditions
see the Experimental section) at −2.85 V provided 2
cleanly with no evidence of reduction of the benzyl
ether (Table 1). Indeed, the reduction could be per-
formed even at −3.1 V with the benzyl ether unaffected.
With the ability to deprotect cinnamyl ethers selectively
to allyl and benzyl ethers, we turned our attention to
nitrogen versus oxygen reactivity.
cinnamyl group with waves at −1.84, −2.47, −2.99, and
−3.13 V. Electrochemical reduction at −1.8 V produced
only acid 8. These initial findings demonstrate the
selective removal of the cinnamyl group from oxygen
over nitrogen in ethers and esters over amines and
amides, as well as the selectivity over O-benzyl groups.
Previous experiments7demonstrated that no selectivity
can be expected when traditional methods such as
dissolved metal reductions are used.
Electrochemical reduction of amine 4 provided only
recovered starting material, which boded well for selec-
tive O-deprotection. Consequently, we investigated the
reductive deprotection of ether–amine 5 and ester–
amide 7. The CV of 5 exhibited reduction waves at
−2.95 and −3.15 V, similar to other cinnamyl ethers. As
expected, 5 was selectively deprotected to afford alco-
hol 6. The CV of 7 showed reduction waves for each
Previously we reported that electrolysis of compounds
containing both cinnamyl ether and vinyl bromide moi-
eties afforded products with both cinnamyl group

J. Hansen et al. / Tetrahedron Letters 44 (2003) 1575–1578
1577
removal and debromination.7 Noting that the cinnamyl
group is removed very sluggishly using a carbon elec-
trode, we sought to selectively debrominate systems
containing both functionalities with the cinnamyl ether
intact. Thus, electrolysis of 9 at −2.6 V with a reticu-
lated vitreous carbon (RVC) electrode provided 10 in
66% yield. Electrolysis of 9 with a mercury pool elec-
trode resulted in both debromination and cinnamyl
group removal as previously reported.7 This is useful in
reduction of vinyl bromides in the presence of other
olefins or hydride sensitive groups.
(16) exhibited similar CV’s with peaks at −2.48 V, −2.72
V and −3.08 V for 14 and −2.57 and −3.15 V for 16.
Electrochemical reduction at −2.45 V produced the
corresponding amines in 80% and 54% isolated yields
respectively (Table 2). The lower yield of morpholine is
most likely related to the isolation of a volatile product
rather than the total conversion. In both cases, TLC
and NMR of the crude reaction mixture indicated the
absence of starting material and little or no other
products. Interestingly, when the reduction was per-
formed in the absence of a proton source, products that
appear to arise from the addition of the electrochemi-
cally generated nitrogen anion to acetonitrile were
obtained (Scheme 1). This finding suggests that it might
be possible to couple the electrochemical COC depro-
tection with subsequent trapping of the nitrogen anion
by electrophiles in a base-free alkylation procedure.
Cyclic voltammetry of 12, containing both a cinnamyl
ether and a cinnamoyl ester, exhibited peaks at −2.18,
−2.82 (broad), and −3.11 V. Based on the CV we
expected that the ester might be selectively reduced
preferentially to the ether. However, reduction of 12 at
−2.15 V produced a small amount of cyclohexane diol
(<10%) and a 1:1 mixture of 12 and 13. This result
indicates a selective hydrogenation of cinnamyol ester
versus the cinnamyl ether.
COC versus Alloc selectivity was tested with carbamate
18. As with the cinnamyl versus allyl ethers,7 only two
reduction waves (−2.54 and −3.08 V), both attributable
to the COC group, were observed in the CV. Electro-
chemical reduction of 18 produced 19 as the only
isolated product in 79% yield (Table 2).
2.2. Cinnamyloxycarbonyldeprotection
Finally, the question of N- versus O- selectivity was
examined with 20. Cyclic voltammetry showed a very
broad reduction wave at −2.3 V and two more waves at
−2.75 and −2.93 V. Electrochemical reduction at −2.3 V
did not selectively remove the O-cinnamyloxycarbonyl
group. Both cinnamyloxycarbonyl groups were
removed, indicating that the driving force for the
departure of the heteroatom is controlled at the site of
cinnamyl connection, not at the carbamate or carbon-
The cinnamyl–oxycarbonyl group has been used to
protect amines.9 Typical deprotection conditions
involve a Pd catalyst and ammonium formate. These
conditions are also used for deprotection of the Alloc
group making the selective deprotection of COC in the
presence of Alloc unlikely. We wished to determine if
the COC group could be selectively removed electro-
chemically in the presence of an Alloc group as well as
to determine if N- versus O- selectivity could be
achieved as with the cinnamyl group.
Substrates were prepared by reaction of the appropriate
amine with cinnamyloxycarbonyl imidazole.9 COC–
protected tetrahydroisoquinoline (14) and morpholine
Scheme 1.
Table 2. Electrochemical cleavage of the cinnamyloxycarbonyl group

1578
J. Hansen et al. / Tetrahedron Letters 44 (2003) 1575–1578
ate site and that lower basicity (i.e. alkoxide versus
amide) controls the cleavage.
Acknowledgements
The authors are grateful to the following agencies for
support of this work: National Science Foundation
(CHE-9910412), the donors of the Petroleum Research
Fund administered by the American Chemical society
(PRF-38075-AC), TDC Research, Inc., and Merck
Research Laboratories Doctoral Fellowship (funding
for Stanley Freeman).
References
3. Conclusion
1. Veinberg, A. Ta.; Mairanovskii, V. G.; Samokhvalov, G.
I. Zh. Obshch. Khim. 1968, 38, 667.
2. Corey, E. J.; Tius, M. A. Tetrahedron Lett. 1977, 24,
2081.
We have shown that cinnamyl and cinnamyloxycar-
bonyl groups are versatile protecting groups. Each
group can be selectively removed in the presence of the
corresponding allyl7 or allyloxycarbonyl group. The
cinnamyl group can also be removed in the presence of
a benzyl ether. In both cases (cinnamyl and cinnamy-
loxycarbonyl), oxygen deprotection occurs selectively
over nitrogen deprotection except in the case of carbon-
ate/carbamate competition. Reaction setup up is simple
and requires no special equipment other than a poten-
tiostat. The only reagents that are used are phenol (as a
proton source) and Et4NBr (electrolyte), both of which
can be recovered.11
3. Guibe´, F. Tetrahedron 1998, 54, 2967.
4. (a) Chandrasekhar, S.; Reddy, C. R.; Rao, R. J. Tetra-
hedron 2001, 57, 3435; (b) Honda, M.; Morita, H.;
Nagakura, I. J. Org. Chem. 1997, 62, 8932; (c) Lemaire-
Audoire, S.; Savignac, M.; Geneˆt, J. P. Tetrahedron Lett.
1995, 36, 1267; (d) Beugelmans, R.; Bourdet, S.; Bigot,
A.; Zhu, J. P. Tetrahedron Lett. 1994, 35, 4349; (e)
Merzouk, A.; Guibe´, F. Tetrahedron Lett. 1992, 33, 477;
(f) Dangles, O.; Guibe´, F.; Balavoine, G. J. Org. Chem.
1987, 52, 4984; (g) Deziel, R. Tetrahedron Lett. 1987, 28,
4371.
5. For a recent example of this type of allyl group deprotec-
tion see: Bertrand, M. P.; Escoubet, S.; Gastaldi, S.;
Timokhin, V. I. Chemical Comm. 2002, 216.
6. Franco, D.; Panyella, D.; Rocamora, M.; Gomez, M.;
Clinet, J. C.; Muller, G.; Dun˜ach, E. Tetrahedron Lett.
1999, 40, 5685.
4. Experimental
Cyclic voltammetry experiments were performed with
25–30 mg of sample in 10 mL of 0.15 M Et4NBr in
CH3CN using a carbon cathode and platinum anode.
An Ag/Ag+ reference electrode was used (0.5785 V
versus NHE). Sweep rates were typically 50 mV/s.
Reported potentials are for peak maxima. Bulk elec-
trolysis was performed in a 150-mL beaker with a Hg
pool cathode and Ag/Ag+ reference electrode. A Pt
anode was place in a separate chamber divided from
the rest of the cell by a fine glass frit. Sample (1.0
mmol) and phenol (2.0 mmol) were dissolved in 50 mL
of 0.15 M Et4NBr in acetonitrile electrolyte solution,
placed in the beaker, and the solution was purged of
oxygen bubbling argon through the solution for 10
minutes. Electrolyte solution (10 mL) was added to the
anode chamber. Electrolysis was performed at the
appropriate potential. The reaction was monitored by
observing the drop in current, the Coulombs delivered,
and by TLC. The reaction was judged complete when
the current leveled off, the number of Coulombs
exceeded 2 F per mole, and/or TLC indicated little or
no remaining starting material. The solution was
decanted into ether (150 mL) and filtered to remove the
charge carrier. The filtrate was evaporated, and the
product was isolated.
7. Solis-Oba, A.; Hudlicky, T.; Koroniak, L.; Frey, D.
Tetrahedron Lett. 2001, 42, 1241.
8. Kinoshita, H.; Inomata, K.; Kameda, T.; Kotake, H.
Chem. Lett. 1985, 515.
9. D’Addona, D.; Bochet, C. Tetrahedron Lett. 2001, 42,
5227.
10. (a) Fernadez-Forner, D.; Casals, G.; Navarro, E.; Ryder,
H.; Albericio, F. Tetrahedron Lett. 2001, 42, 4471; (b)
Sigismondi, S.; Sinou, D. J. Chem. Res. 1996, 46; (c)
Beugelmans, R.; Neuville, L.; Boischoussy, M.; Chas-
tanet, J.; Zhu, J. P. Tetrahedron Lett. 1995, 36, 3129; (d)
Geneˆt, J. P.; Blart, E.; Savignac, M.; Lemeune, S.;
Lemaire-Audoire, S.; Paris, J. M.; Bernard, J. M. Tetra-
hedron 1994, 50, 497.
11. When the reduction was conducted in the presence of 1.1
equiv. of cyclohexene, the yield improved to 64%
although the conversion was slower (73%, 27% starting
material). Apparently, removing bromine generated at
the anode by scavenging it improves the yield. The
authors are very grateful to the referee who suggested this
experiment.