Substitution of chlorocyclobutanones with xanthate salts. The remarkable effect of added base

Article abstract of DOI:10.1016/j.tetlet.2009.03.133

The substitution reaction of chlorocyclobutanones with dithiocarbonates (xanthates) is greatly accelerated by the addition of a mild base such as DABCO or DBU.

Full text of DOI:10.1016/j.tetlet.2009.03.133

Tetrahedron Letters 50 (2009) 3613–3616
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Substitution of chlorocyclobutanones with xanthate salts. The remarkable
effect of added base
*
Rama Heng, Béatrice Quiclet-Sire, Samir Z. Zard
Laboratoire de Synthèse Organique associé au CNRS, Ecole Polytechnique, F-91128 Palaiseau, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The substitution reaction of chlorocyclobutanones with dithiocarbonates (xanthates) is greatly acceler-
ated by the addition of a mild base such as DABCO or DBU.
Received 21 January 2009
Accepted 10 March 2009
Available online 24 March 2009
Ó 2009 Elsevier Ltd. All rights reserved.
This Letter is dedicated with respect to the
memory of our friend, Dr. Christian
Marazano
As part of our ongoing work on the chemistry of xanthates
(dithiocarbonates),1 we required a convenient access to
-xanthyl
cyclobutanones. These could not normally be obtained from
In the same way, except for the use of DMF as the reaction sol-
vent, dixanthates 6b–f were obtained from the corresponding
a
a
-
chlorocyclobutanones by the nucleophilic substitution of the chlo-
ride, and a circuitous route via the more reactive bromide had to be
used.2 Even then, the substitution was sometimes capricious, giv-
ing unexpected ring-expanded products and forcing us to replace
the usual O-ethyl xanthate with the more robust O-neopentyl xan-
thate. We have now found that by addition of an organic base, the
substitution process is accelerated considerably, allowing us to
Cl
Cl
C
O
O
O
Et3N
Cl
1
R
Cl
R
Cl
R
R'
R'
Cl
R'
Cl
2
3
start with the more readily accessible
a-chlorocyclobutanones
and to access derivatives which would be extremely difficult to ob-
tain otherwise.
–C l
Cl
O
Cl
O
a,a-Dichlorocyclobutanones such as 2 can be produced by
cycloaddition of in situ generated dichloroketene 1 with various
electron-rich alkenes or dienes.3 They undergo a number of unu-
sual transformations, but the triethylamine induced cine isomeri-
R
R
5
4
Cl
Cl
R'
R'
Scheme 1. Cine rearrangement of a,a-dichlorocyclobutanones.
sation into the corresponding
a,
a0-dichlorocyclobutanones 5 was
especially intriguing (Scheme 1).4 Only scant mechanistic studies
for this transformation are available, but the ionic chain process
displayed in Scheme 1 and involving SN20 substitution on enolate
3 appears reasonable (vide infra). We therefore contemplated the
possibility of replacing the triethylamine with a xanthate salt,
which would act both as a mild base and as a nucleophile.5,6
Indeed, treatment of dichlorocyclobutanone 2a with potassium
O-ethyl xanthate or sodium O-neopentyl xanthate in acetonitrile
resulted in the smooth formation of rearranged dixanthates 6a1
and 6a2 in 84% and 73% yields, respectively, presumably through
S
O
S
O
S
EtOC(=S)SK
or
neoPnOC(=S)SNa
R"O
S
R
Cl
R
DMF or MeCN
R'
Cl
rt
R'
2
OR"
6
6a1, R = H, R' = 3-butenyl, R" = Et (84%)
6a2, R = H, R' = 3-butenyl, R" = neoPn (50%)
6b1, R = H, R' = 4-t-BuC6H4-, R" = Et (84%)
6b2, R = H, R' = 4-t-BuC6H4-, R" = neoPn (73%)
6c, R = R' = Et, R" = neoPn (71%)
S
S
O
S
further reaction of an intermediate a
-chloro-a0-xanthate (Scheme
neoPnO
6d, R = R' = Ph, R" = neoPn (80%)
2; neoPn = neopentyl).
6e, R = Me, R' = Ph, R" = neoPn (81%)
6f (75%; 35:65)
Scheme 2. Synthesis of a,
a0-bis(xanthyl)-cyclobutanones.
neoPnO
S
* Corresponding author. Tel.: +33 1693 34872; fax: +33 1693 33851.
E-mail address: zard@poly.polytechnique.fr (S.Z. Zard).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.03.133

3614
R. Heng et al. / Tetrahedron Letters 50 (2009) 3613–3616
dichlorocyclobutanones 2b–f (same definition for substituents R
and R’ for 2a–f and 6a–f). We found DMF to be a generally better
solvent than acetonitrile, and the use of the neopentyl group sim-
plified the interpretation of the NMR spectra.
weaker base, DABCO, as it was easier to monitor the consumption
of the starting chloride and avoid the formation of side products
due to overexposure to the base. As depicted in the lower half of
Scheme 3, treatment of chlorides 7b–f with sodium O-neopentyl
xanthate in DMF in the presence of substoichiometric amounts of
DABCO gave rise to the corresponding xanthates 8b–f in good yield
and short-reaction times. Again, no substitution was observed in
the absence of DABCO. It is worth pointing out that the chlorine
in 7a–g is neopentylic and, in the case of 7d, also tertiary, and
therefore almost impossible to displace in normal circumstances.
That it can be smoothly substituted under such mild conditions
is thus quite remarkable. 2-Chloro-3,3,4,4-tetramethyl cyclobuta-
none, 7g, which cannot produce the desired enolate (vide infra), re-
acted sluggishly to give xanthate 8g in the presence of DABCO, and
not at all in its absence.
The replacement of both chlorines in
a,a-dichlorocyclobuta-
nones by a nucleophile does not seem to have been reported, and
the present reaction provides a convenient access to useful dix-
anthates. Surprisingly, however, the smooth reaction of the
dichlorocyclobutanones contrasted sharply with the total lack of
reactivity of the monochlorocyclobutanones towards substitution
by the xanthate.7 We conjectured that, in the case of the less acidic
monochlorocyclobutanones, the xantate salt was perhaps too weak
a base to induce the formation of useful concentrations of the cor-
responding enolates. Using 2-chloro-3,3-diethyl-cyclobutanone 7a
as a test substrate, we repeated the substitution in the presence of
added base. Indeed, incorporation of 1.7 equiv of triethylamine re-
sulted in a reasonably smooth reaction to give 62% (by NMR) of the
desired xanthate 8a (Scheme 3).
Diisopropylethylamine (DIPEA) was less effective, presumably
because of steric hindrance. To obtain the same yield, 6 equiv of
the base and a 24 h reaction time were required. In contrast, DMAP
and especially DABCO and DBU proved significantly more efficient:
with only 0.2 equiv of the DABCO or DBU, an essentially complete
conversion was observed after 1 hour for the former and 7 min for
the latter, the isolated yield of 8a being 80% and 84%, respectively.
Even though addition of DBU resulted in the fastest reaction,
subsequent experiments were conducted in the presence of the
Various hypotheses can be cast to rationalize the extreme ease
of substitution of the chlorine atom in the presence of added base,
but we believe the most coherent to be the one detailed in Scheme
4 for the case of 7b. The geometry of the corresponding enolate 9
allows an efficient overlap between the high lying filled
(HOMO of the enolate) with the empty
p orbital
r
* of the C–Cl bond (better
seen in 90). The energy level of the HOMO of the enolate is, in addi-
tion, increased by the ring strain inherent in cyclobutenes,8 bring-
ing it closer to the level of the
r
* and further improving the extent
of the interaction. By partially populating the antibonding molecu-
lar orbital, this interaction causes a significant weakening, and
hence lengthening, of the C–Cl bond. This stereoelectronic configu-
ration is qualitatively similar to that prevailing at the anomeric po-
sition in carbohydrates, where one of the lone pairs of the ring
S
oxygen is ideally disposed to interact with the
r
* of the axial ano-
O
O
meric C–X bond causing its lengthening and increased reactivity
towards substitution (the ‘anomeric effect’).
neoPnO
SNa
S
Et
Et
Base, DMF, rt
Cl
The substitution reaction by the xanthate salt on enolate 9 takes
place by an SN2 fashion to give ultimately compound 8b in high
yield following the protonation of enolate 10. The regioisomeric
chloride, 7h, easily prepared by reaction of in situ generated meth-
ylchloroketene with 2-ethyl-butene 12, leads to enolate 11, where
the SN2 reaction mode is hampered by the methyl group. The sub-
stitution now takes place in an SN20 fashion (shown on 110) to give
xanthate 8b in 52% overall yield from alkene 12, via same enolate
10. Thus, the reaction leads to identical products starting from
Et
S
Et
OneoPn
7a
Base (equiv.) Reaction time
8a
8a Yield
None
6 hrs
0%
TEA
4 hrs
62% (by NMR)
DIPEA
DMAP
DMAP
DABCO
DBU
24 hrs
6 hrs
62% (incomplete conversion)
78%
18 hrs
1 hr
100% (by NMR)
80%
84%
7 minutes
Me
O
Me
O
Et
Me
BH
BH
Cl
Me
Et
O
Me
Et
O
X
X
Et
Me
Me
Et
O
DABCO (B)
7b, X = Cl
7c, X = Cl
8b, X = SCSOneoPn (cis: trans 1:9)
8c, X = SCSOneoPn (cis: trans 3:7)
TEA (3 equiv.; 48 hrs) , 63%
9'
DABCO (0.2 equiv.; 0.5 hr) , 94%
Cl
Cl
Et
Et
DABCO (0.2 equiv.; 1.5 hrs) , 65%
Et
9
7b
S
Me
O
O
X
BH
neoPnO
S
Me
Et
O
Me
O
Me
Me
X
Me
Et
SCSOneoPn
SCSOneoPn
7d, X = Cl
7e, X = Cl
Et
Et
8d, X = SCSOneoPn (cis: trans 2:3)
8e, X = SCSOneoPn (cis: trans 1:2)
DABCO (0.2 equiv.; 0.5 hrs) , 65%
8b
7h
10
DABCO (0.6 equiv.; 3 hrs) , 65%
O
Cl
BH
Me
O
BH
O
Cl
Cl
Me
Et
O
Me
Et
Me
Et
O
DABCO (B)
Me
Me
Me
X
X
11'
Et
Et 11
Et
t-Bu
S
MeCHClCOCl
TEA
7g, X = Cl
Et
Et
8g, X = SCSOneoPn
DABCO (1.2 equiv.; 1 hr)
15% (by NMR)
12
7f, X = Cl
hexane
neoPnO
S
8f, X = SCSOneoPn (cis: trans 1:7)
DABCO (0.6 equiv.; 0.3 hrs) , 71%
Scheme 3. Effect of added base.
Scheme 4. Mechanistic rationale for the effect of base.

R. Heng et al. / Tetrahedron Letters 50 (2009) 3613–3616
3615
neoPnO
neoPnOC(=S)SNa
either chlorides 7b or 7h. In the case of 7g, the required enolate
cannot be formed, and the small yield of product probably arises
Me
Me
R
O
S
O
S
by an alternative route involving a carbene, generated by
nation of the chloride.
a-elimi-
H
Cl
R
DMF, rt
Cl
R
Cl
2g,h
R
In the case of slightly strained, fused cyclobutanones, the reac-
tion furnished two regioisomeric products, in contrast to the pre-
ceding examples. Thus, reaction of chlorocyclobutanone 7i under
the usual conditions furnished, in addition to the expected xan-
thate 8i, a significant amount of regioisomer 13i. Clearly, the SN2
and SN20 occur with comparable rates (Scheme 5). A similar behav-
iour was exhibited by dihydropyrane fused chlorocyclobutanone
7j, which afforded 8j and 13j in 61% and 14% yields, respectively.
In further support of the postulated mechanism, we found that
it is possible to obtain mono-xanthates from dichlorocyclobuta-
nones by blocking the enolisation after the introduction of the first
xanthate group and therefore preventing the second substitution
from taking place, as illustrated by the examples displayed in
Scheme 6.
2g; 14g, R = Me
2h; 14h, R = Et
14g, 91% (cis : trans 3:1)
14h, 94% (cis : trans 7:3)
Cl
O
Me
O
Me
Cl
Cl
S
Me
OneoPn
Me
H
Me
Me
S
S
3g
14'g
neoPnO
S
H
O
O
Cl
Cl
Cl
2i
14i, 60% (cis : trans 5:3)
Treatment of 2g with sodium O-neopentyl xanthate in DMF
gave a high yield of distal chloroxanthate 14g, arising by an SN2’
reaction on the corresponding enolate 3g. Clearly, further enolisa-
tion in 14g, in a manner allowing substitution of the second chlo-
rine is no longer possible and only the mono-xanthate is obtained.
Furthermore, none of the isomeric products 140g, possibly derived
from a prior cine-rearrangement of the starting geminal dichloride
2g followed by SN2 substitution by the xanthate salt, was ob-
served.9 In the same manner, dichloroketones 2h and 2i furnished
xanthates 14h and 14i in 94% and 60% yield, respectively. No added
base is required in these cases, in line with the greater acidity of
the dichlorocyclobutanones, as discussed above (Schemes 1 and 2).
In summary, this study has provided a simple solution to the
Scheme 6. Synthesis of mono-xanthates from
a,a-dichlorocyclobutanones.
the chloride by a normal SN2 process is already reasonably
efficient.
Acknowledgement
We thank Ecole Polytechnique for generous financial support to
one of us (R.H.).
Supplementary data
synthesis of
ther processed through the powerful xanthate transfer technol-
ogy.1 Extension of this approach to other
-chloroketones is
currently being investigated. Preliminary results with -chlorocy-
clopentanones and -chlorocyclohexanones seem to indicate a
a
-xanthyl-cyclobutanones,10 which can now be fur-
Experimental procedures, characterization data and copies of
1H, 13C and NMR spectra are presented. Supplementary data asso-
ciated with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2009.03.133.
a
a
a
much less pronounced effect of added base. It must be pointed
out, however, that in these derivatives the direct substitution of
References and notes
1. For general reviews see: Zard, S. Z. Angew. Chem., Int. Ed. Engl. 1997, 36, 672;
Zard, S. Z. Xanthates and Related Derivatives as Radical Precursors. In Radicals
in Organic Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim, 2001;
Vol. 1, p 90; Quiclet-Sire, B.; Zard, S. Z. Top. Curr. Chem. 2006, 264, 201; Quiclet-
Sire, B.; Zard, S. Z. Chem. Eur. J. 2006, 12, 6002.
2. Binot, G.; Zard, S. Z. Tetrahedron Lett. 2003, 44, 7703.
3. Brady, W. T. Synthesis 1971, 415; Brady, W. T. Tetrahedron 1981, 37, 2949;
Bellus, D.; Ernst, B. Angew. Chem., Int. Ed. Engl. 1988, 27, 797.
O
O
neoPnOC(=S)SNa
DABCO (0.2 eq.)
DMF, rt
Cl
S
7i
4. Conia, J. M.; Robson, M. J. Angew. Chem., Int. Ed. Engl. 1975, 14, 473; Martin, P.;
Greuter, H.; Bellus, D. J. Am. Chem. Soc. 1979, 101, 5853; Martin, P.; Greuter, H.;
Rihs, G.; Winkler, T.; Bellus, D. Helv. Chim. Acta 1981, 64, 2571; Zengin, M.;
Dastan, A.; Balci, M. Synth. Commun. 2001, 31, 1993.
8i, 56%
neoPnO
S
S
5. For rare examples of cine substitutions of chloro- and dichloro-cyclobutanones,
see: (a) Harmata, M.; Huang, C.; Rooshenas, P.; Schreiner, P. R. Angew. Chem.,
Int. Ed. Engl. 2008, 47, 8696; (b) Hassner, A.; Naidorf-Meir, S.; Frimer, A. A. J. Org.
Chem. 1996, 61, 4051; (c) Hassner, A.; Naidorf-Meir, S.; Gottlieb, H. E.;
Goldberg, I. J. Org. Chem. 1993, 58, 5202; (d) Hassner, A.; Naidorf-Meir, S. J. Org.
Chem. 1992, 57, 5102; (e) Hassner, A.; Naidorf-Meir, S.; Gottlieb, H. E.;
Goldberg, I. J. Org. Chem. 1992, 57, 2442; (f) Butenschön, H. J. Chem. Soc., Perkin
Trans. 1 1991, 483; (g) Hassner, A.; Naidorf-Meir, S.; Gottlieb, H. E. Tetrahedron
Lett. 1990, 31, 2181; (h) Snider, B. B.; Walner, M. Tetrahedron 1989, 45, 3171; (i)
Hassner, A.; Naidorf-Meir, S. J. Org. Chem. 1989, 54, 4954; (j) Hassner, A.; Dillon,
J. J. Org. Chem. 1986, 51, 4954; (k) Hassner, A.; Dillon, J.; Onan, K. O. J. Org. Chem.
1986, 51, 3315; (l) Tsunetsugu, J.; Asai, M.; Hiruma, S.; Kurata, Y.; Mori, A.; Ono,
K.; Uchiyama, H.; Sato, M.; Ebine, S. J. Chem. Soc., Perkin Trans. 1 1983, 285; (m)
Hassner, A.; Dillon, J. L.; Krepski, R. L.; Onan, K. D. Tetrahedron Lett. 1983, 24,
2181; (n) Harding, K. E.; Trotter, J. W.; May, L. M. J. Org. Chem. 1977, 42, 2715.
SN2'
+
neoPnO
S
S
O
SN2
neoPnO
S
O
S
Cl
neoPnO
S
13i, 26%
O
S
neoPnO
O
Cl
S
O
O
7j
6. Only one example involving the reaction of an
a-un-substituted-a-chloro-
cyclobutanone with silver trifluoroacetate has been reported, see: Schmidt, A.
H.; Kircher, G.; Bräu, E. J. Org. Chem. 1998, 63, 1954.
neoPnOC(=S)SNa
DABCO (0.2 eq.)
DMF, rt
+
7. For relevant studies in the context of the Favorskii rearrangement, see: (a)
Föhlisch, B.; Radl, A.; Schweitzer-Raschke, R.; Henkel, S. Eur. J. Org. Chem. 2001,
4357; (b) Brady, W. T.; Patel, A. D. J. Org. Chem. 1974, 39, 1949; (c) Brady, W. T.;
Patel, A. D. J. Org. Chem. 1973, 38, 4106; (d) Brady, W. T.; Hieble, J. P. J. Org.
Chem. 1971, 36, 2033; (e) Bordwell, F. G.; Carlson, M. W. J. Am. Chem. Soc. 1970,
92, 3370. 3377; (f) Bordwell, F. G.; Carlson, M. W. J. Am. Chem. Soc. 1969, 91,
O
O
S
8j, 61%
13j, 14%
neoPnO
S
Scheme 5. Cases leading to regioisomers.

3616
R. Heng et al. / Tetrahedron Letters 50 (2009) 3613–3616
3951; (g) Bordwell, F. G.; Carlson, M. W.; Knipe, A. C. J. Am. Chem. Soc. 1969, 91,
3949.
8. The ionization potential of cyclobutene is 9.59 eV, while that of ethylene is
10.51 eV. See: Wiberg, K. B.; Ellison, C. B.; Wendoloski, J. J.; Brundle, C. R.;
Kuebler, N. A. J. Am. Chem. Soc. 1976, 98, 7179.
10. Typical experimental procedure: To a stirred solution of 2-chlorocyclobutanones
2 (1.0 mmol) in DMF (1 mL) under nitrogen, were added sodium O-neopentyl
xanthate (279 mg, 1.5 mmol), and DABCO (24 mg, 0.20 mmol). Stirring was
continued until reaction was complete (TLC). Brine and diethyl ether were then
added, and the aqueous layer was extracted twice with diethyl ether. The
organic layer was then washed repeatedly with water then with saturated
ammonium chloride, dried over anhydrous MgSO4, and the solvent removed in
9. The spectroscopic distinction between isomers 14g and 140g is not obvious. To
resolve this ambiguity, the xanthate was subjected to ozonolysis, which cleanly
converted the thiocarbonyl group into a carbonyl (Quiclet-Sire and Zard, S. Z.,
unpublished work). In the case of 140g, this should have caused a significant
shift (ca 0.5 ppm) in the 1H NMR signal of the ring proton (indicated in bold
type), which was not observed.
vacuo. In most cases, the crude residue was essentially pure
a-
xanthylcyclobutanone 8; otherwise, it was further purified by
chromatography.