Metal-free activation of C-C multiple bonds through halide ion pairs: Diels-Alder reactions with subsequent aromatization

Article abstract of DOI:10.1016/j.tetlet.2008.05.027

Imidazolium and ammonium halides, both in homogeneous phase or immobilized on mesoporous silica, proved able to promote the aromatization reaction of Diels-Alder adducts under ethylene and H₂ elimination. These observations point to the increased nucleophilicity of such halide ions. They also open the possibility to employ such ion pairs as nucleophilic organocatalysts, for example, as substitutes for phosphines or amines.

Full text of DOI:10.1016/j.tetlet.2008.05.027

Tetrahedron Letters 49 (2008) 4546–4549
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Metal-free activation of C–C multiple bonds through halide ion pairs:
Diels–Alder reactions with subsequent aromatization
Helena Kaper a, Markus Antonietti a, Frédéric Goettmann b,
*
a Colloid Chemistry Department, Max-Planck Institute of Colloids and Interfaces, Scientific Campus Golm, 14476 Potsdam, Germany
b Institut de Chimie Separative de Marcoule, UMR 5257, ICSM site de Marcoule, BP 17171, 30207 Bagnols Sur Ceze, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Imidazolium and ammonium halides, both in homogeneous phase or immobilized on mesoporous silica,
proved able to promote the aromatization reaction of Diels–Alder adducts under ethylene and H2
elimination. These observations point to the increased nucleophilicity of such halide ions. They also open
the possibility to employ such ion pairs as nucleophilic organocatalysts, for example, as substitutes for
phosphines or amines.
Received 28 February 2008
Revised 23 April 2008
Accepted 7 May 2008
Available online 10 May 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Organocatalysis
Ionic Liquids
C–C bond forming reactions
Lewis basic catalysis
Ionic liquids (ILs) have attracted much interest in recent years
as they turned out to be extremely useful for a whole range of cut-
ting edge applications.1 Indeed, due to their very low volatility,
they are presented as sustainable solvents for industrially relevant
reactions,2 allowing to isolate the products by evaporation or
extraction, while possible (organo-metallic) catalysts remaining
in the IL phase can be directly recycled.3–5 Additionally, ILs can sta-
bilize polar or charged species and reaction intermediates and are,
thus, able to accelerate the rate of organic reactions through pure
solvation effects.6–8
Chloride and bromide are very common counter ions of cationic
surfactants but in ionic liquids more complex anions are often
employed (like BFÀ4 ; PF6À or bis(trifluoromethylsulfonyl)imide,
NTfÀ2 ) to lower their melting points and/or tailor their hydro-
philic/hydrophobic balance. However, since halide ions are both
typical counter ions for surfactants and known to be much more
nucleophilic (and were thus shown to alter the observed reactivi-
ties in halogenated ILs19), we preferred to choose halides as coun-
ter ions for this study. We therefore evaluated the capacity of
C16mimCl immobilized in MCM-48 (prepared as previously
Nevertheless, when the products of the reaction cannot be
recovered by distillation, their extraction with standard organic
solvents can be difficult, which makes the sustainability of the
whole process at least questionable. Therefore, many works con-
centrated on the development of ‘heterogenised’ ILs by their
anchoring on high surface solids, usually silica,9–12 or their trap-
ping in mesoporous systems as for ionogels.13 MCM type silica
has also been made by directly using long chain imidazolium-
based ILs (typically 1-hexadecyl-3-methylimidazolium chloride,
described14,15) type silica (C16mimCl@MCM-48, Fig. 1) to promote
Diels–Alder20–22 type [4 + 2] cyclo additions, as they were already
shown to be strongly accelerated by ILs.8 Other reports, however,
state that chloride containing ionic liquids lower the yield and
selectivity of Diels–Alder reactions due to their ability to act as
strong hydrogen bond acceptor.23
C
16mimCl) as the templating agent.14,15 The latter material seemed
to be very convenient to test whether immobilized ILs would fea-
ture catalytic properties similar to that of the bulk equivalents.
This point is not obvious, as both confinement (within meso-
pores)16 and site isolation (of the catalytic species)17,18 are known
to strongly affect the catalytic activity of many systems.
* Corresponding author. Tel.: +33 466 905 758; fax: +33 466 797 611.
E-mail address: Frederic.goettmann@cea.fr (F. Goettmann).
Figure 1. Transmission electron micrograph and schematic representation of
C16mimCl@MCM-48, one of the catalysts we employed.
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.05.027

H. Kaper et al. / Tetrahedron Letters 49 (2008) 4546–4549
4547
We tested a variety of possible dienes and dienophiles. In most
cases, our solid had no influence on the outcome of the reaction.
Moreover, under the chosen moderate conditions mostly no reac-
tion was observed at all. However, our interest was attracted by
the reaction of naphthoquinone and benzoquinone with 1,3-cyclo-
hexadiene (Scheme 1).
In the first case, a 100% conversion of the reactants was
observed. However, instead of the expected cyclo addition product
(1,2,3,4,4a,9a-hexahydro-1,4-etheno-anthraquinone, 1, which is
the only product obtained when no salt is added under the same
reaction conditions), anthraquinone was observed as the major
product. The formation of anthraquinone is also observed with
benzoquinone (in even higher yields). The diaddition product is
not detected in the non catalyzed reaction, and no elimination of
the 1,4-etheno-bridge was observed.
Obviously, a formal elimination of ethene and hydrogen takes
place under our reaction conditions. In the case of benzoquinone,
the second addition presumably takes place after the first aromati-
zation step toward naphthoquinone. Indeed no double aliphatic
addition product (i.e., no compound with two bicyclic groups)
is detected. Moreover, reference tests with non-immobilized
C16mimCl also showed the occurrence of dealkylation, while only
the immobilization within SiO2 enabled diaddition.
In order to test the versatility of this reaction, various combina-
tions of dienes and dienophiles (all featuring carbonyl functions as
they seem essential to observe the aromatisation product) have
O
O
O
C16mimCl@MCM-48
+
+
+
Tol., 65°, 72h.
100%
2
1
O
O
45%
O
55%
O
O
O
C16mimCl@MCM-48
2
+
Tol., 65°, 72 h.
85%
2
1
O
O
O
75%
25%
Scheme 1. Diels–Alder cyclo addition of naphthoquinone on cyclohexadiene catalyzed by C16mimCl@MCM-48.
Table 1
Other Diels–Alder type reactions in presence of C16mimCl@MCM-48a
Entry
Dienophile
Diene
Extend of conversionb (%)
Product distributionb (%)
O
O
1
66
O
O
O
O
100%
2
3
100
100
O
O
100%
OMe
OMe
O
O
O
O
MeO
O
O
OMe
OMe
OMe
8%
92%
OMe
O
O
O
MeO
O
4
100
OMe
OMe
100%
O
O
O
O
5
40
O
O
91%
9%
a
Reaction conditions: 2 mmol of substrate, 0.1 g of C16mimCl@MCM-48 (0.15 mmmol C16mimCl), in toluene (2 ml) at 65 °C for 72 h.
b
Conversion rates and product distribution were determined by 1H NMR.

4548
H. Kaper et al. / Tetrahedron Letters 49 (2008) 4546–4549
been tested under similar conditions. The obtained results are
summarized in Table 1. As can be seen, it turned out that a couple
of Diels–Alder addition products were able to undergo either dehy-
drogenation (when butadiene or dimethylbutadiene is used) or a
dealkylation to yield aromatic species even if some yields
remained low.
The reaction was analyzed in more details to shed some light on
the actual mechanism. In order to decouple the various factors
impacting the observed reactions, we presynthesized the Diels–
Alder product 1 and submitted it to various possible catalysts or
for the enhanced activity of this solid. In some tests with supported
ion pairs, minor leaching could be detected (either in the GC–MS
chromatograms or in the NMR spectra, see supporting informa-
tion), but evaluating the actual impact of this leaching was difficult
and would require more detailed investigations.
As can be seen, in the case of non-immobilized ILs, a reasonable
amount of dealkylation compound was observed as soon as the IL
is miscible with the solvent and its anion is a halide. For example,
C4mimCl yielded 10% of anthraquinone in acetonitrile while
C4mimBF4 (also soluble in acetonitrile) failed to provide any
detectable reaction. This suggests that, solubility provided (NaCl,
for example was not soluble in either solvent and only yields traces
of dealkylation product (Table 2, entry 12)), dealkylation is primar-
ily driven by the halide counter ion, while the nature of the cationic
amine (either an imidazolium (Table 2, entries 8, 10, and 11) or an
ammonium (Table 2, entry 9)) only plays a secondary role. Inter-
estingly, when the dealkylation reaction of 1 was run in pure
C4mimCl as both the solvent and the catalyst under otherwise
identical reaction conditions only 1% of dealkylated product could
be extracted.
The physical properties of ILs with halide ions as counterions
already give first indications to understand their special role in this
reaction. Indeed, Armstrong et al. analyzed the solvatation proper-
ties of various ILs using a multiple parameter approach and found
imidazolium halide based ILs to be among the most polar and
hydrogen-bond basic ILs.19 This is in agreement with the
increased nucleophilicity of halide ion containing ILs, as also stated
by Aggarwal et al.23 While the strong nucleophilicity of ILs can
lower their catalytic activity due to a strong interaction with the
cation of the IL, it is known that strong nucleophiles can add to
activated C–C multiple bonds to form anionic or zwitterionic
species. Dimethyl acetylene dicarboxylate (DMAD) has, for
example, been reported to undergo attacks from DABCO or even
pyridine.24,25
catalyst components (such as
C16mimCl@MCM-48, calcinated
mesoporous silica without IL, and non-immobilized ILs with differ-
ent counterions, etc.) in toluene at 65 °C for 72 h. Under the term
catalyst, we mean here a compound which can promote the aro-
matisation reaction, being understood, that our salts have only a
minor effect on the first Diels–Alder step. Since certain ILs are
not soluble in toluene, reactions were also carried out in acetoni-
trile as the solvent. Table 2 summarizes the obtained results.
Immobilized ions pairs all proved able to yield the aromatization
product in moderate yields. The obtained conversions were rela-
tively modest as compared to those reported in Scheme 1. This
can, on the one hand, be explained by diffusion limitations. Indeed,
access to the active pores of C16mimCl@MCM-48 is more difficult
for the relatively bulky compound 1 than for naphthoquinone or
cyclohexadiene. But, on the other hand, the fact that non supported
ion pairs proved even less active than immobilized ones also points
out the synergetic role of silica (e.g., supported CTAC (Table 2, en-
try 27) yielded 27% of anthraquinone in acetonitrile while the same
salt without silica only yielded 9% of product under otherwise sim-
ilar reaction conditions). The origin of this effect of silica still has to
be studied in detail, which would add a stone to the current work
on the often observed confinement effects in porous catalysts.16
Silica alone (without ion pairs) (Table 2, entry 6), however, was
not able to promote the expected reaction. In the case in which
the washed C16mimCl@MCM-48 (obtained by a 12 h Soxhlet
extraction with acetone as a solvent) was used (Table 2, entry 7),
a thermogravimetric analysis of the resulting powder showed that
only about 50% of the salt had been removed. This possibly facili-
tated the diffusion of the reactants through the solid, accounting
A possible mechanism for the dealkylation of [2,2,2] bicyclo-
octene derivatives could thus involve the attack of the chloride
on the C@C double bond, followed by a pericyclic shift of C–C
bonds, resulting in a formal retro-Diels–Alder reaction (See Scheme
2) .
The possibility for a bicyclic molecule to undergo such an
aromatization would then directly depend on the ease with which
the chloride anion can attack its double bond, that is on the polar-
izability thereof. And actually, the polarizability of such C@C
Table 2
Conversion of 1 into anthraquinone by various saltsa
Entry Salt
Amount of
Amount of anthraquinone
in acetonitrileb (%)
anthraquinone in
tolueneb (%)
1
2
—
0
0
20
C16mimCl@MCM- 13.5
48
Cl
Cl-
3
C16mimBr@MCM- 31
48
33
Cl-
4
5
6
CTAC@MCM-48
CTAB@MCM-48
SBA-15 type,
calcined
25
21
0
27
19
0
Scheme 2. Putative mechanism for the chloride catalysed dealkylation of a formal
[2,2,2,] bicyclooctene.
7
Washed
33
40
C
48
16mimCl@MCM-
OMe
OMe
a)
b)
8
9
C16mimCl
2
4
1
0
19
9
10
3
Traces
0
1
O
O
O
O
CTAC
10
11
12
13
14
C4mimCl
C4mimBr
NaCl
C4mimBF4
C4mimPF6
OMe
OMe
+ Cl-
Traces
0
0
40%
Cl - Cl-
a
Reaction conditions: 0.2 mmol of 1, 0.05 g of possible catalyst, in 2 ml solvent,
Cl
at 65 °C for 72 h.
b
The amount of formed anthraquinone was determined by 1 H NMR. (See
supplementary information.)
Scheme 3. Other C16mimCl catalyzed bicycle openings.

H. Kaper et al. / Tetrahedron Letters 49 (2008) 4546–4549
4549
double bonds can be affected by various factors such as the pres-
Acknowledgement
ence of carbonyl groups, but also the presence of other C@C double
bonds in a close neighborhood or the tension within the cycle. This
is in good agreement with the results reported in Table 1 as well as
with the following observations: dimethyl-[2,2,2]-bicyclooctadi-
ene dicarboxylate could be quantitatively converted into dimethyl-
phthalate when treated with C16mimCl in methanol for 24 h
(Scheme 3a) and norbornadiene ([2,2,1]-bicyclooheptadiene) was
converted into toluene when submitted to a similar reaction envi-
ronment (Scheme 3b). In the latter case, no carbonyl group seems
to be required to observe a transformation. However, the com-
pound is already a diene and the methylene bridge imposes a
strong constrain on the cycle, which is eased by the isomerisation.
Furthermore, an isomerisation was observed in state of a dealkyl-
ation because the release of a formal methylene group would be
very unfavorable. It is worth being noticed here that for commod-
ity reasons these additional tests were run without the silica sup-
port. Indeed, the use of MCM-48 strongly increases the complexity
of the system, adding considerations of diffusion limitations, con-
finement, etc. As this report focuses on the first observations of
such an aromatization process, simplicity was required. We thus
focused on solubilized ions pairs. All these observations, however,
are in good agreement with our proposed bond attack of the chlo-
ride. Nevertheless, further theoretical investigations are required
to assess a more detailed reaction mechanism and additional
experiments would be required to address the actual role of porous
silica.
This vision of an anion induced retro-Diels–Alder reaction adds
to a longer discussion on the usually admitted concerted orbital
control of this type of reactivity. The classical mechanistic view
on Diels–Alder reactions was nicely reviewed by Sauer and Sust-
mann.26 Schleyer et al., however, pointed out that for polarisable
starting products a two-step pathway essentially involving a zwit-
terionic intermediate was energetically more favourable than the
Woodward-Hoffmann route.27 In 1993, Reetz and Gansauer found
that a heterogenous dispersion on Lithium perchlorate was able to
catalyse Diels–Alder reactions and regioselective 1,3-Claisen rear-
rangements,28 a system which was already quite closely related
to the present observations. Meanwhile, there is collected theoret-
ical and experimental evidence that at least Diels–Alder reactions
with more polarizable reactants progress along a two-step mecha-
nism involving a charged or zwitterionic intermediate (e.g., see
Refs. 29–32). On the other hand, as only a limited number of systems
seem to undergo zwitterionic activation, it appears that this mech-
anism is more an additional reaction channel, i.e., an alternative to
the usual case of concerted orbital controlled cyclo addition. Nev-
ertheless, these observations open some promising opportunities
for the metal-free functionalization of C@C double bonds, as the
anionic intermediate could react for instance in a Knoevenagel-like
fashion (in which it would deprotonate the nucleophile and then
act as an electrophile) or in Morita-Baylis-Hillman reactions.
The Max-Planck Society and the CEA are gratefully acknow-
ledged for financially supporting this work.
Supplementary data
Detailed catalytic procedures as well as a typical example of 1H
NMR spectrum are presented. Supplementary data associated with
this article can be found, in the online version, at doi:10.1016/
j.tetlet.2008.05.027.
References and notes
1. Endres, F.; El Abedin, S. Z. Phys. Chem. Chem. Phys. 2006, 8, 2101–2116.
2. Zhang, Z. C. Adv. Catal. 2006, 49, 153–237.
3. Cole-Hamilton, D. J. Science 2003, 299, 1702–1706.
4. Dupont, J.; de Souza, R. F.; Suarez, P. A. Z. Chem. Rev. 2002, 102, 3667–3691.
5. Wasserscheid, P.; Keim, W. Angew. Chem., Int. Ed. 2000, 39, 3773–3789.
6. Zhao, D. B.; Wu, M.; Kou, Y.; Min, E. Catal. Today 2002, 74, 157–189.
7. Welton, T. Coord. Chem. Rev. 2004, 248, 2459–2477.
8. Olivier-Bourbigou, H.; Magna, L. J. Mol. Catal. A: Chem. 2002, 182, 419–437.
9. Gu, Y. L.; Ogawa, C.; Kobayashi, J.; Mori, Y.; Kobayashi, S. Angew. Chem., Int. Ed.
2006, 45, 7217–7220.
10. Gu, Y. L.; Karam, A.; Jerome, F.; Barrault, J. Org. Lett. 2007, 9, 3145–3148.
11. Riisager, A.; Fehrmann, R.; Haumann, M.; Wasserscheid, P. Top. Catal. 2006, 40,
91–102.
12. Mehnert, C. P. Chem. Eur. J. 2004, 11, 50–56.
13. Neouze, M. A.; Le Bideau, J.; Gaveau, P.; Bellayer, S.; Vioux, A. Chem. Mater.
2006, 18, 3931–3936.
14. Kaper, H.; Smarsly, B. Z. Phys. Chemie-Int. J. Res. Phys. Chem. Chem. Phys. 2006,
220, 1455–1471.
15. Wang, T. W.; Kaper, H.; Antonietti, M.; Smarsly, B. Langmuir 2007, 23, 1489–
1495.
16. Goettmann, F.; Sanchez, C. J. Mater. Chem. 2007, 17, 24–30.
17. Hecht, S.; Frechet, J. M. J. Angew. Chem., Int. Ed. 2001, 40, 74–91.
18. Notestein, J. M.; Katz, A. Chem. Eur. J. 2006, 12, 3954–3965.
19. Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc. 2002,
124, 14247–14254.
20. Diels, O.; Alder, K. Justus Liebigs Ann. Chem. 1928, 460, 98–122.
21. Diels, O.; Alder, K. Justus Liebigs Ann. Chem. 1929, 62, 2337–2372.
22. Diels, O.; Alder, K. Justus Liebigs Ann. Chem. 1930, 478, 137–154.
23. Aggarwal, A.; Lancaster, N. L.; Sethi, A. R.; Welton, T. Green Chem. 2002, 4, 517–
520.
24. Fan, M. J.; Li, G. Q.; Liang, Y. M. Tetrahedron 2006, 62, 6782–6791.
25. Nair, V.; Menon, R. S.; Sreekanth, A. R.; Abhilash, N.; Biju, A. T. Acc. Chem. Res.
2006, 39, 520–530.
26. Sauer, J.; Sustmann, R. Angew. Chem. Int. Ed. Engl. 1980, 19, 779–807.
27. Neumann, F.; Lambert, C.; Schleyer, P. V. J. Am. Chem. Soc. 1998, 120, 3357–
3370.
28. Reetz, M. T.; Gansauer, A. Tetrahedron 1993, 49, 6025–6030.
29. Mallakpour, S. E.; Hajipour, A. R.; Mahdavian, A. R.; Khoee, S. J. Polym. Sci.
Polym. Chem. 1999, 37, 1211–1219.
30. Assfeld, X.; Ruizlopez, M. F.; Gonzalez, J.; Lopez, R.; Sordo, J. A.; Sordo, T. L. J.
Comput. Chem. 1994, 15, 479–487.
31. Roberson, M.; Jepsen, A. S.; Jorgensen, K. A. Tetrahedron 2001, 57, 907–913.
32. Griffith, G. A.; Hillier, I. H.; Moralee, A. C.; Percy, J. M.; Roig, R.; Vincent, M. A. J.
Am. Chem. Soc. 2006, 128, 13130–13141.