McMurry intermolecular cross-coupling between an ester and a ketone: Scope and limitations

Article abstract of DOI:10.1016/S0040-4039(02)00617-2

In the course of our studies towards the synthesis of a dioxetane bearing chemiluminescent probe for the detection of thiols, we were faced with the synthesis of sterically hindered benzylic enol ethers. This issue was solved via the use of an intermolecular McMurry cross-coupling between an ester and a ketone. In this article, together with the synthesis of the chemiluminescent probe, the scope and limitations of this low valent titanium based carbon-carbon double bond formation are discussed.

Full text of DOI:10.1016/S0040-4039(02)00617-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3645–3648
McMurry intermolecular cross-coupling between an ester and a
ketone: scope and limitations
Ste´phane Sabelle, Je´roˆme Hydrio, Eric Leclerc, Charles Mioskowski and Pierre-Yves Renard*
Service de Marquage Mole´culaire et de Chimie Bioorganique, Baˆt. 547, DBJC, CEA Saclay, 91191 Gif sur Yvette Cedex,
France
Received 15 March 2002; revised 27 March 2002; accepted 28 March 2002
Abstract—In the course of our studies towards the synthesis of a dioxetane bearing chemiluminescent probe for the detection of
thiols, we were faced with the synthesis of sterically hindered benzylic enol ethers. This issue was solved via the use of an
intermolecular McMurry cross-coupling between an ester and a ketone. In this article, together with the synthesis of the
chemiluminescent probe, the scope and limitations of this low valent titanium based carbonꢀcarbon double bond formation are
discussed. © 2002 Elsevier Science Ltd. All rights reserved.
In the course of our studies towards the synthesis of
chemiluminescent probes for the detection of thiols, our
efforts were led on 1,2-dioxetane disulfides (Scheme 1).1,2
These probes were conceived by analogy with the chemi-
luminescent 1,2-dioxetane substrates used for the detec-
tion of phosphatase activities.3–5 The strategy we chose
is depicted on Scheme 1. Nucleophilic addition of a
thiolate on an unsymmetrical disulfide compound should
cleave the sulfurꢀsulfur bond, yielding in the formation
of an unstable thiophenolate. Cleavage of the 1,2-dioxe-
tane ring, through chemically induced electron exchange
luminescence (CIEEL), should then lead to an excited
m-mercapto benzoate anion, whose decay to the ground
state should provide light, as for the corresponding
m-oxobenzoate anion. In order to synthesize first gener-
ation chemiluminescent probes 1 and 2, we were faced
with the need to build a sterically hindered benzylic enol
ether, as depicted on Scheme 2.
It is no longer necessary to demonstrate the remarkable
efficiency of low-valent titanium compounds for car-
bonꢀcarbon bond forming.6 McMurry coupling reaction
is for instance a very powerful tool to achieve reductive
coupling of aldehydes or ketone to alkenes.7 Yet, due to
the particularly high reactivity of the titanium intermedi-
ate, its scope is drastically limited to the homo-dimerisa-
tion of ketones and aldehyde, or to intramolecular
cross-coupling reactions. It must also be acknowledged
that this reaction suffers from a serious lack of reproduci-
bility, having a bad reputation as being tricky and highly
‘co-workers and suppliers dependent’. Mixed coupling
reactions usually lead to a statistical mixture of all
possible alkene, thus reasonable yield of a single cross-
coupling product requires the use of an excess of one of
the two substrates. Few examples of preparatively useful
results have been obtained if the substrates in question
exhibit sufficiently different redox potentials.8,9
X
X
X
*
X
O
O
O O
O O
O
O
O
Electron
transfer
Trigger
O
O
R
Z
Z
Z
S
hν
O
R
S
S
S
S
S
S
S
Z
EWG
EWG
Scheme 1. Proposed mechanism for light emission wijth the designed chemiluminescent reagents for the chemiluminescent probes.
* Corresponding author. Tel.: +33-169-085-497; fax: +33-169-087-991; e-mail: pierre-yves.renard@cea.fr
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00617-2

3646
S. Sabelle et al. / Tetrahedron Letters 43 (2002) 3645–3648
The following experimental conclusions were drawn:
O
O
O
O
ꢀ
Oxygen and moisture traces have to be cautiously
precluded, thence the reaction was best performed in
a glove box, under nitrogen.
Reagents had to be freshly purchased and kept
under nitrogen atmosphere. Noteworthy, lower
yields were obtained with TiCl3 in solution or from
other commercial sources than Fluka. Use of 1 M
solution of LiAlH4 in diethyl ether also gave supe-
rior yields.
The reaction had to be led in THF via a three steps
procedure. First activation of TiCl3 with a careful
addition of hydride (0°C), followed by refluxing with
the amine, and finally very slow addition (over a 3 h
period of time) of a mixture of ketone and benzoate,
still in refluxing THF.
If both carbonyl compounds were added together
and very slowly, use of a large excess of ketone was
NOT necessary. Best yields were obtained with a
simple 1.2 excess of 2-adamantanone 3 over ben-
zoate 4. On the other hand, a 10 times excess of
TiCl3; a 5 times excess of LiAlH4 and a 6 times
excess of triethylamine were required.
O
+
ꢀ
O
S
S
R
O
+
SP
Cl
O2N
S
O2N
SP
P : protective group
1 : R = H
2 : R = NO2
R
ꢀ
Scheme 2. Structure and retrosynthetic scheme for chemilu-
minescent probes 1 and 2.
Keto-esters cross-coupling cyclisations via intramolecu-
lar alkylidenation reactions on treatment with low-
valent titanium have been scarcely described.10,11
Highest yields have been obtained with titanium species
formed from TiCl3/LiAlH4 in the presence of triethyl-
amine. Yet, aqueous work-up of the reaction mixtures
hydrolyzes the enol ethers initially formed, and leads to
the respective cyclic ketones as the final products.12
Apart from descriptions in patents,2,13,14 intermolecular
McMurry cross-coupling are described as inefficient,
ꢀ
ꢀ
Best treatment was obtained by partitioning the
crude reaction mixture with same amount of diethyl
ether and water, and extracting thrice the aqueous
phase with diethyl ether. Crude reaction mixtures
yielded enol ether 5 together with low amounts of
bis(adamantylene), which were easily separated
through silicagel chromatography. Enol ether proved
moderately stable to such conditions. Addition of ca.
0.1% triethylamine in the mobile phase was required.
yielding
mainly
the
homodimerized
product
bis(adamantylene). Hindered ketones such as adaman-
tanone or ditert-butyl ketone are required. Low func-
tional groups compatibility, and both availability and
efficiency of the low valent titanium reagents are con-
sidered as a major issue.15 In a few cases, this issue has
been circumvented by replacing the McMurry conden-
sation by a Wadsworth–Emmons reaction between a-
alkoxyphosphonates and hindered ketones.15 Yet,
formation of the phosphonates required use of a strong
Lewis acid (TiCl4), followed by basic reaction condi-
tions (LDA), dramatically lowering the functional
groups compatibility of this alternative route. We thus
decided to study the scope and the limitations of the
intermolecular McMurry condensation between a ben-
zylic ester and a hindered ketone.
With those optimized reaction conditions in hand, the
McMurry condensation was assayed by a wide range of
co-workers, from undergraduate student to post-doc-
toral fellow. According to the experimental skill of the
‘co-worker’, purified yields in enol ether 5 varied
between 41 and 59%, thus showing an acceptable
reproducibility.
If deprotection of the silyl ether could be performed in
high yields (nBu4NF/THF), activation of the phenol
group into trifluoromethyl sulfonate, mesylate or tosyl-
ate ended up in hydrolysis of the fragile enol ether
moiety. With the goal in mind to introduce the required
sulfur atom via substitution of aryl halides, and in
order to widen the scope and study the limitations of
the McMurry intermolecular cross coupling reaction,
the condensation was then undertaken on 2-adaman-
tanone and 3-halogenated benzoates (Scheme 4).
In order to clear the low reproducibility and efficiency
issues up, a first screening of the reaction conditions
with 2-adamantanone 3 and benzoate 4 was undertaken
(Scheme 3). For this purpose, the already described6,10
low valent titanium source TiCl3/LiAlH4 in the pres-
ence of triethylamine was used.
The reaction gave satisfactory yields on aryl fluoride 6
and aryl chloride 9. Byproduct Bis(adamantylene) 8
was isolated in, respectively, 22 and 19% yield. On the
contrary, aryl bromide 11 was totally reduced under the
McMurry reaction conditions, yielding 38% enol ether
12 together with low amounts (24%) of hydrolyzed
ketone 13.
O
O
O
O
i
+
41-59%
3
4
5
O
O
Si
Si
Scheme 3. McMurry condensation of adamantanone 3 on
3-tert-butyldimethylsilyloxy methylbenzoate 4. (i) TiCl3 (10
equiv.), LiAlH4 (5 equiv.), Et3N (6 equiv.), THF, 70°C, 20 h.
Our attempts to perform arylic substitution of the
chlorine or fluorine atoms of enol ether 7 and 10 by a
sulfur bearing nucleophile or by a more reactive halo-

S. Sabelle et al. / Tetrahedron Letters 43 (2002) 3645–3648
3647
i
O
SH
S
SnBu3
+
95%
O
O
O
EtO
iii
O
O
i
i
+
+
O
O
75%
18
S
ii
EtO
EtO
6
9
7: 38%
8 : 19%
8 : 22%
F
F
v
92%
17
83%
Br
I
O
O
iv
37%
EtO
10: 40%
O
Cl
Cl
20
EtO
S
O
O
O
2
19
O
i
SH
8 : 11%
+
+
11
12: 38%
13: 24%
Scheme 6. Synthesis of 3-mercapto substituted ethyl ben-
zoates. (i) Bu3SnCl, Et3N, CCl4, 18 h; (ii) KI (15 equiv.), CuI
(5 equiv.), HPMA, 155°C, 3 days; (iii) Pd(PPh3)4, toluene,
110°C, 48 h; (iv) thiourea, NiBr2, NaBH3CN, DMF; (v) HBr,
AcOH, DMSO, 3 h.
Br
Scheme 4. McMurry coupling of adamantanone on 3-
halobenzoates. (i) TiCl3 (10 equiv.), LiAlH4 (5 equiv.), Et3N
(6 equiv.), 3 (1.2 equiv.), THF, 70°C, 20 h.
OEt
gen group were unsuccessful, or ended up in hydrolysis
of the fragile enol ether function. Use of an activated
aryl halide using a p-substituted nitro function could
not be achieved, since during the McMurry condensa-
tion of 2-adamantanone on 2-nitro-5-chlorobenzoate
14, the nitroaryl function was reduced in aniline, and
enol ether 15 proved too unstable to be purified in
reasonable yields (Scheme 5).
OEt
O
i
20
SH
12: 52%
OEt
O
i
12: 57%
19
S
2
NO2
NH2
O
O
O
i
8 : 35%
+
OEt
OEt
O
i
14
15 : 17%
Cl
Cl
18
21: 60%
S
S
Scheme 5. McMurry condensation with 2-nitrobenzoate. (i)
TiCl3 (10 equiv.), LiAlH4 (5 equiv.), Et3N (6 equiv.), 3 (1.2
equiv.), THF, 70°C, 20 h.
Scheme 7. McMurry coupling of m-mercapto benzoates. (i)
TiCl3 (10 equiv.), LiAlH4 (5 equiv.), Et3N (6 equiv.), 3 (1.2
equiv.), THF, 70°C, 20 h.
Although a thiophenol would appear poorly stable
under the harsh reaction conditions of the McMurry
condensation, we decided to test the intermolecular
McMurry cross coupling on arylic mercapto com-
pounds 18–20 (Scheme 6).
hand, the tert-butyl protective group on 18 allowed
isolation of enol ether 21 in 60% yield.
These benzoates were synthesized using 3-iodo ethyl-
benzoate 17 as starting material. Compound 17 was
obtained in high yields through an halogen exchange
which could be performed using a combination of KI
and CuI (3/1) at 155°C in HPMA.16 Thiophenol 19 was
synthesized in one step, although in moderate yield
from 17, using thiourea as the sulfur source, via a
nickel based catalytic process described by Takagi.17
S-tBu protected thiophenol 18 was synthesized in high
yields via a Stille-like coupling18 of tert butylsulfanyl
tributyltin 16 on iodide 17. Disulfide 20 was synthesized
from 18 using the Dickman procedure,19 using DMSO
as the oxidative specie.
Chemiluminescent probes 1 and 2 were then obtained
in acceptable yields from 21 in a two-step procedure
(Scheme 8). The first step was a mild and controlled
addition of activated sulfenyl chlorides on the protected
sulfur atom. Contrary to the described reaction proce-
dure,19 use of an external source of acid (HBr/AcOH)
proved not necessary. Enol ether function remains par-
ticularly fragile, and once partially hydrolyzed, purifica-
tion from corresponding ketone could not be achieved
using conventional chromatographic techniques. In
order to obtain acceptable yields in unsymmetrically
substituted disulfides 22 and 23, we found that the
reaction had to be carefully monitored, and stopped
once ca. one third of the starting material was con-
sumed. Recovered starting material could be recycled
McMurry coupling performed on 19 and 20 expectedly
yielded reduced enol ether 12 (Scheme 7). On the other

3648
S. Sabelle et al. / Tetrahedron Letters 43 (2002) 3645–3648
Acknowledgements
O O
OEt
OEt
S.S. was financially supported by SPI Bio, P.-Y. R.
by De´le´gation Ge´ne´rale pour l’Armement.
i
ii
S
S
21
S
R
S
R
NO2
NO2
References
1 : R = H, 64%
2 : R = NO2, 66%
22 : R = H, 75%
23 : R = NO2, 57%
1. Sabelle, S.; Renard, P. Y.; Pecorella, K.; de Suzoni-
De´zard, S.; Cre´minon, C.; Grassi, J.; Mioskowski, C. J.
Am. Chem. Soc., in press.
2. Grassi, J.; Sabelle, S.; Renard, P. Y. Fr. Patent, WO
0044719, 2000; Chem. Abstr. 2000, 133, 150547
3. Bronstein, I.; Edwards, B.; Voyta, J. C. J. Biolumin.
Chemilumin. 1988, 2, 186.
Scheme 8. Final steps for the synthesis of chemiluminescent
probes 1and 2. (i) R=H, 2-nitrophenyl sulfuryl chloride,
THF, 2 h; R=NO2: 2,4-dinitrophenyl sulfuryl chloride, THF,
2 h; (ii) (PhO)3P, O3, −78 to 20°C, CH2Cl2. Yields are given
once starting material was recycled (see text).
4. Bronstein, I.; Voyta, J. C.; Thorpe, G. H. G.; Kricka, L.
J.; Amstrong, G. Clin. Chem. 1989, 35, 1146–1441.
5. Edwards, B.; Sparks, A.; Voyta, J. C.; Strong, R.; Mur-
phy, O.; Bronstein, I. J. Org. Chem. 1990, 55, 6225–6229.
6. Fu¨rstner, A.; Bogdanovic, B. Angew. Chem., Int. Ed.
1996, 35, 2442–2469.
7. McMurry, J. E. Chem. Rev. 1989, 1513–1524.
8. McMurry, J. E.; Krepski, L. R. J. Org. Chem. 1976, 41,
3929.
ending up in a global 75% yield in 22 and 57% yield
in 23 after three reaction batches.
The final step of the synthesis of 1,2-dioxetanes 1 and
2 was the careful [2+2] cycloaddition of singlet oxy-
1
gen O2 on enol ether. In view of the great sensitivity
of the disulfide bond to highly oxidative oxygenated
species, oxygen bubbling under UV light irradiation
seemed unacceptable. In order to monitor precisely
the amount of 1O2 delivered, we decided to use
9. Paquette, L. A.; Yan, T.-H.; Wells, G. J. J. Org. Chem.
1984, 49, 3610.
10. Schaap, P.; Handley, R. S.; Giri, B. P. Tetrahedron Lett.
1987, 28, 935–938.
11. Fu¨rstner, A.; Huppert, A. J. Am. Chem. Soc. 1995, 117,
4468.
12. McMurry, J. E.; Miller, D. D. J. Am. Chem. Soc. 1983,
105, 1660.
13. Arghavani, Z.; Akhaven-Tafti, H.; Desilva, R.; Thakur,
K. US Patent US 6036892, 2000; Chem. Abstr. 2000, 132,
207841.
14. Schaap, A. P.; Akhavan-Tafti, H. US Patent WO9007511,
1990; Chem. Abstr. 1991, 114, 181726.
15. Roeschlaub, C. A.; Sammes, P. G. Perkin Trans. 1 2000,
2243–2248.
1
triphenylphosphite ozonide, which slowly delivers O2
upon heating from –78°C to room temperature.20,21
Using this singlet oxygen source, fragile 1,2-dioxetane
chemiluminescent probes 1 and 2 were synthesized in
25 and 22% yield, respectively, after careful silica gel
chromatography. No trace of by-products displaying
oxidation at any of the two sulfur atoms was
detected. Recovered starting material could be recy-
cled. Practical yields in 1 and 2 could then be esti-
mated at 64 and 66%, respectively.
In conclusion, for the synthesis of chemiluminescent
probes 1 and 2, we experimented the scope and limi-
tations of an unprecedented intermolecular cross-cou-
pling McMurry condensation of hindered ketones on
3-substituted benzoates. The reaction conditions were
assayed in order to ensure a good reproducibility.
Acceptable yields were obtained with arylfluorides,
chlorides, protected phenols and thiophenols. Other
attempts ended up in the reduction of the aromatic
substituting function. Luminescence properties of 1
and 2 will be reported in due time.1
16. Suzuki, H.; Kondo, A.; Ogawa, T. Chem. Lett. 1985,
411–412.
17. Takagi, K. Chem. Lett. 1985, 1307–1308.
18. Terada, M.; Migita, T.; Sano, H.; Ogata, T.; Kosagi, M.
Bull. Chem. Soc. Jpn. 1985, 58, 3657.
19. Dickman, D. A.; Chemburkar, S.; Konopacki, D. B.;
Elisseou, E. M. Synthesis 1993, 573–574.
20. Murray, R. W.; Kaplan, M. L. J. Am. Chem. Soc. 1969,
91, 5358–5364.
21. Stephenson, L. M.; Zielinski, M. B. J. Am. Chem. Soc.
1982, 104, 5819–5820.