Effect of ionic liquid organizing ability and amine structure on the rate and mechanism of base induced elimination of 1,1,1-tribromo-2,2-bis(phenyl- substituted)ethanes

Article abstract of DOI:10.1016/j.tet.2005.11.061

The kinetics of the elimination reaction of 1,1,1-tribromo-2,2-bis(phenyl- substituted)ethanes into the corresponding 1,1-dibromo-2,2-bis(phenyl- substituted)ethenes induced by amines were studied in three room temperature ionic liquids ([BMIM][BF₄

Full text of DOI:10.1016/j.tet.2005.11.061

Tetrahedron 62 (2006) 1690–1698
Effect of ionic liquid organizing ability and amine structure
on the rate and mechanism of base induced elimination
of 1,1,1-tribromo-2,2-bis(phenyl-substituted)ethanes
*
Francesca D’Anna, Vincenzo Frenna, Vitalba Pace and Renato Noto
*
` `
Dipartimento di Chimica Organica “E. Paterno”, Universita degli Studi di Palermo,
Viale delle Scienze-Parco d’Orleans II, 90128 Palermo, Italy
Received 16 September 2005; revised 4 November 2005; accepted 24 November 2005
Available online 20 December 2005
Abstract—The kinetics of the elimination reaction of 1,1,1-tribromo-2,2-bis(phenyl-substituted)ethanes into the corresponding 1,1-
dibromo-2,2-bis(phenyl-substituted)ethenes induced by amines were studied in three room temperature ionic liquids ([BMIM][BF4],
[BMIM][PF6], [BdMIM][BF4]). In order to have information about reagent–ionic liquid interactions, the reaction was carried out over the
temperature range (293.1–313.1 K). To study the effect of the amine on the rate and occurrence of the elimination reaction, several primary,
secondary and tertiary amines with different structure (cyclic and acyclic), basicity and steric requirements were used. The data collected
show that the reaction occurs faster in ionic liquids than in other conventional solvents. Furthermore, ionic liquids seem to be able to induce,
for the studied reaction, a shift of mechanism from E1cb (in MeOH) versus E2 (in ionic liquid).
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
b-substituent (generally an aryl-substituted ring) are able to
affect the reactivity and, in some cases, the mechanism of
base promoted elimination. We chose as substrates the
1,1,1-tribromo-2,2-bis(phenyl-substituted)ethanes (1) as the
alkoxide-induced b-elimination has been studied by some of
us (Fig. 1).4
There is growing interest in the utility and application of the
room temperature ionic liquids (RTILs). Ionic liquids have
been proposed as environmentally benign solvents for their
non-detectable vapor pressure, their non-flammability and
for easy recyclability.1 ILs promise several advantages in
organic synthesis both making the process more efficient
and reducing use of raw materials.2 Several organic
reactions have been performed with success and for some
of these the quantitative aspects in ILs have also been much
investigated.3 The ILs, with respect to conventional organic
solvents, can provide a very different microenvironment and
this can significantly influence the outcome of a reaction. At
least two factors must be considered in determining how IL
and solute influence one another. The former concerns the
fact that ILs are systems having some degree of organization
and this could be strongly affected by guest molecules. The
latter is related to organizing ability of ILs, these, owing to
p–p interactions, can help the substrate to maximize the
stabilizing interactions. Therefore, we believed it interesting
to investigate kinetically the effect that ILs have on classical
organic reactions such as b-elimination. This is one of the
most studied reactions in organic chemistry. It is well
known that both the electronic and steric properties of
CBr3
CBr2
MeO
MeO
OMe
OMe
MeO
MeO
OMe
OMe
Amine
IL
1a, b
2a, b
a = 3,4-dimethoxy
b = 2,5-dimethoxy
Figure 1.
The data collected allowed us to hypothesize that the
reaction of 1 to the corresponding 1,1-dibromo-2,2-
bis(phenyl-substituted)ethenes (2) occurs through an irre-
versible E1cb mechanism. The E2 mechanism seemed
unlikely. The analyzed substrates differ for steric require-
ments (1a,b) of aryl groups. Their reactivity changes
drastically with the conformation. This seemed stimulating
for quantitative studies because the IL, owing to p–p
interactions, could constrain 1 to assume a planar
conformation, that should maximize the electronic effects
exerted by aromatic rings on the route of reaction. To avoid
strong interactions between charged bases, as alkoxide ions,
Keywords: Ionic liquid; Kinetic; Elimination mechanism.
*
Corresponding authors. Tel.: C39 091596919; fax: C39 091596825;
e-mail addresses: fdanna@unipa.it; rnoto@unipa.it
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.11.061

F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
1691
C
and imidazolium cation, we chose to use neutral amine
bases. These are, in conventional organic solvents, largely
used while, in ILs, they have been less investigated both as
nucleophiles and bases. Several primary, secondary and
tertiary amines with different structure (cyclic or acyclic),
basicity and steric requirements were chosen as base for
inducing elimination of 1a. The pseudo first-order kinetics
of amine-promoted elimination reaction were studied
following the appearance of 2 spectrophotometrically. The
reaction was performed in solution of 1-butyl-3-methylimi-
dazolium tetrafluoroborate ([BMIM][BF4]) at various
concentrations (from 0.008 up to 0.0304 M) of amine over
the temperature range (293.1–313.1 K). Several recent
studies demonstrated that changes in the nature of cation
part or the counterion could bring about notable variations
in the reaction mechanism;5 for this reason the piperidine-
promoted elimination reaction was also studied in 1-butyl-
3-methylimidazolium hexafluorophosphate ([BMIM][PF6])
and in 1-butyl-2,3-dimethylimidazolium tetrafluoroborate
([BdMIM][BF4]). The [BMIM][PF6] was chosen to inves-
tigate the effect of the anion. As a matter of fact, the
presumably unlike cation–anion of different size inter-
actions between BMIM and two anions (a larger PFK6 and a
smaller BFK4 ) could in turn affect the interaction between
the activated complex and solvent. The [BdMIM][BF4] was
chosen to evaluate the effect of hydrogen bond donor ability
of imidazolium cation.
basic morpholine (pKBH Z8.33). We are aware of the fact
that the amines basicity, measured in water, could not be
adequate to describe the effective strength in IL. Indeed, the
occurrence of peculiar interactions IL–amine should
determine a different basicity order in IL and water. In
this light, dramatically different behaviors can be
surely related to deeply unlike IL–amine interactions.
Nevertheless, structurally similar amines (i.e., secondary
cyclic) should interact with IL in a comparable manner and
C
the aqueous pKBH values constitute a good starting point
for analyzing the reactivity trend.
For reactive amines a quantitative study was undertaken. To
make a comparison with conventional organic solvents, a
kinetic measurement of the elimination reaction of 1a in
dioxane (the cosolvent used for reaction in IL solution, see
Section 4) in the presence of pyrrolidine (7.5!10K3 M) as a
model amine was carried out. The elimination reaction was
very slow (kobs!5.5!10K7 sK1). In IL solution, the above
reaction was faster and the complete course of the
absorbance as a function of the time is shown in Figure 2.
0.65
0.6
0.55
0.5
2. Results and discussion
0.45
0.4
First, we verified that 1a did not spontaneously give the
elimination product 2a in IL within 10 days (i.e., if a
reaction occurs kobsz10K8 sK1). Then, we analyzed the
behavior of 1a, in [BMIM][BF4], in the presence of amines.
The results are reported in Table 1.
0.35
0.3
Table 1
0.25
pKBH in H2Oa
nr/erb
0
1 104 2 104 3 104 4 104 5 104 6 104 7 104
C
Amine
Primary amines
Butylamine
Cyclohexylamine
Secondary amines
N-Butyl-N-methylamine
Pyrrolidine
Time (s)
10.68
10.66
nr
nr
Figure 2. Experimental plot of Abs versus time for the elimination reaction
of 1a in the presence of pyrrolidine (0.0304 M) in [BMIM][BF4] at 298.1 K
and lZ280 nm.
c
nr
er
er
er
er
nr
nr
11.27
11.12
10.89
10.78
8.33
Piperidine
It can be seen that the curve does not show a typical trend of
a simple kinetic process. It seems that at least two
kinetically relevant steps are responsible for the observed
trend. In fact, first- or higher-order kinetic equations do not
fit the experimental trace. From a careful analysis of the
system we found that the absorbance of the IL–pyrrolidine
mixture, for each of studied ILs, changed as function of the
time. Furthermore a slow variation in the UV–vis spectrum
of IL, induced by some of the used amines, was observed.
For example, in Fig. 3, the spectra of IL in the presence of
heptamethyleneimine, collected over a time range (48 h),
are reported.
Hexamethyleneimine
Heptamethyleneimine
Morpholine
2,2,6,6-Tetramethylpiperidine
Tertiary amines
Triethylamine
11.07
10.75
10.46
10.08
nr
nr
nr
N-Methylpyrrolidine
N-Methylpiperidine
a See Ref. 6.
b nrZno elimination reaction was detected; erZelimination reaction was
detected.
c The pKBH value should be in the range 10.64 (N,N-dimethylamine)–11.
25 (N,N-dibutylamine).
C
As can be seen from Table 1, the primary and tertiary
amines, independent of their basicity and alkyl groups
structure, were unable to induce the elimination reaction.
The secondary amines were more variable. The acyclic
N-butyl-N-methylamine was unreactive as was the highly
hindered 2,2,6,6-tetramethylpiperidine and the scarcely
Data previously obtained by us seem to exclude that the
observed variation could be a consequence of an acid–
base equilibrium between the acidic imidazolium ion and
the amine. Indeed, in this case, the observed variation of
UV–vis spectrum is fast.7

1692
F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
3
2.5
2
0 min
45 min
86 min
130 min
202 min
1430 min
1700 min
2069 min
2894 min
1.5
1
0.5
0
250
300
350
400
450
500
λ (nm)
Figure 3. UV–vis spectra of [BMIM][BF4] in the presence of
heptamethyleneimine 0.0304 M collected in 48 h.
The variation in UV–vis spectrum cannot be accounted as
well for the occurrence of an amino demethylation of
imidazolium cation. Indeed, this reaction occurs with scarce
yields to 398 K.8 To confirm of this, the IL–amine mixture,
after some days from mixing, was extracted with
diethylether. The extract, analyzed by HPLC and GC–MS,
did not show the presence of methylamine or other reaction
products.
1
The IL–amine interaction was also studied by H NMR
spectroscopy recording the spectra of a mixture IL–
heptamethyleneimine during a time range. Firstly, we
analyzed the effect of dioxane addition to IL and a
significant variation of the NMR spectrum was observed.
For example, the signal due to the H-2 proton was split in a
couple of signals of different intensity, the more shielded
signal had the lesser intensity and was broad. The same
splitting, but with opposite relative intensity and shape of
signals, was observed for adding of a solution of amine in
dioxane (see Fig. 4)
However, as can be observed, every signal is affected by
amine solution addition and after 5 min, all the signals result
enlarged and lose the multiplicity.
The signals splitting can be easy explained considering that
two different ion-pairs are present in the IL–amine mixture.
This hypothesis perfectly agrees with both the formation of
ion-pairs in imidazolium-based ionic liquids,9 and with the
observed variations induced by trace amounts of water in
the H NMR spectrum of [BMIM][BF4].10 However, the
1
Figure 4. 1H NMR spectra of: (A) neat [BMIM][BF4]; (B) [BMIM][BF4] in
the presence of 50 mL of 1,4-dioxane; (C) [BMIM][BF4] in the presence of
a dioxane solution of heptamethyleneimine (0.0304 M) at tZ0 min.
shape of signals changes slowly in the time. For example,
the signals at 7.34 and 7.29 ppm due to H-4 and H-5,
respectively, (see Fig. 5 spectrum A) change. In fact, in a
first time (60 min, see Fig. 5 spectrum B) they seem to
collapse, later (180 min, see Fig. 5 spectrum C) they return
as two distinct signals.
The above picture could be a consequence of a little and
slow reorganization of molecules of IL; the different and
variable interactions between two imidazolium rings could
explain the variation in NMR and UV–vis spectra. To
verify if the reorganization process is operative only for
The spectrum acquired after 8 days show distinct signals for
the heterocyclic hydrogen atoms but broad signals for
aliphatic ones (see Fig. 5, spectrum D).

F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
1693
Figure 5. 1H NMR spectra of: (A) [BMIM][BF4] in the presence of a dioxane solution of heptamethyleneimine (0.0304 M) at tZ0 min; (B) at tZ1 h; (C) at tZ
3 h; (D) at tZ8 days.
0.5
amines for which a variation in UV–vis spectrum was
observed, we collected NMR spectra of an IL–piperidine
mixture in a range of time. A very similar behavior was
observed. However, it must be remarked that no signal
due to transformation products of IL and amines was
0.45
detected. Furthermore, the ratios among integrals relative
to different signals were constant in the time. It is
0.4
0.35
0.3
1
noteworthy that both variations in UV–vis and H NMR
spectra can not be related to a deficiency in homogeneity
of IL–amine mixtures. Indeed, these were perfectly clear.
2.1. Kinetic data
0.25
1 104
1.5 104
Time (s)
2 104
2.5 104
In order to avoid invalidating the kinetic data by variations
in UV–vis spectrum of the IL–amine mixture, the kinetic
runs were registered using a sample of IL containing the
same amine concentration of the kinetic run as reference. In
this manner, excellent pseudo first-order curves were
obtained (Fig. 6).
0
5000
Figure 6. Experimental plot of Abs versus time for the elimination reaction
of 1a in the presence of pyrrolidine (0.0304 M) in [BMIM][BF4] at 298.1 K
and lZ280 nm recorded by using the [BMIM][BF4]–amine mixture as
reference.

1694
F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
To verify that the effect of the IL–amine mixture on the
studied reaction was not a function of IL reorganization
time, a kinetic run was carried out adding the substrate
solution to an equilibrate (after 3 h from mixing) of IL–
amine mixture. The observed kinetic constant was equal to
that collected with a typical methodology (see Section 4).
Changes in reaction mechanism going from conventional
organic solvents to IL are well documented. For example,
in a nucleophilic displacement reaction, a change in
mechanism was proposed on the grounds of a different
nucleophilicity order observed going from conventional
solvents to IL.13
For reactive amines, an excellent linear dependence of kobs
(collected values are reported in Supplementary data:
Table 4) on amine concentration was found according to
Eq. 1.
The kinetic data for the elimination reaction of 1a to 2a
induced by different IL–amine mixtures are reported in
Table 2 (data collected at different amine concentration are
available in Supplementary data: Table 4).
kobs Z a CðkIIÞamine½Amineꢀ
(1)
The secondary cyclic amines namely pyrrolidine, piperi-
dine, hexamethyleneimine and heptamethyleneimine (five,
six, seven and eight membered rings, respectively) were
able to induce the elimination reaction. Similar results were
previously reported by Yadav et al.11 The geometry of
amine seems to be one of the determinant factors for their
reactivity. Indeed, for example, the opposite behavior of
butylamine (unable to induce the elimination in 1a) and
Significant ‘negative’ intercept values were calculated for
[BMIM][BF4]. They were related to an acid–base inter-
action between the acidic imidazolium ion and the amine as
previously reported.7
The kII values show that reactivity of 1a decreases on going
from pyrrolidine to piperidine to hexamethyleneimine to
heptamethyleneimine in the order 22.7:3.1:2.7:1. The
ability of amines to induce the elimination seems to be a
function of their flexibility (i.e., higher flexibility,
lower reactivity), due to the ring dimension, rather than of
nitrogen basicity. Indeed, this changes in the order
3.1:2.2:1.3:1 on going from pyrrolidine to piperidine to
heptamethyleneimine.
C
heptamethyleneimine, bases with comparable pKBH values
(10.68 and 10.78, respectively), could be attributed to their
different conformational freedom. This, higher for the
butylamine than for the hexamethyleneimine, should disturb
the ionic liquid organization with the consequent loss of its
activating effect. However, for amines having comparable
geometry and flexibility, large differences in basicity, as for
C
piperidine and morpholine (pKBH Z11.12 and 8.33,
respectively) are, of course, responsible of the unlike
behavior. Also, the steric hindrance can change the
reactivity as in the case of piperidine and 2,2,6,6-
tetramethylpiperidine that have barely different basicity
It is noteworthy that pyrrolidine induced elimination of 1a
(kIIZ9.16!10K3 MK1 sK1) is faster than the methoxide/
methanol reaction (kIIZ8.3!10K3 MK1 sK1) despite the
difference in basicity between the two considered bases, and
that [BMIM][BF4] and methanol have comparable polarity.
Indeed, the ENR values for [BMIM][BF4] and methanol are
217.2 and 217.7, respectively.14 This is surely indicative
that the IL has an activating effect on the reaction that only
the polarity is not able to explain. Probably, this effect could
be a consequence both of the electrostatic interactions
(bromine–imidazolium ion) and p–p interactions. The
positive effect of electrostatic interactions can be explained
by the E2 mechanism where an assistance to bromine
departure by IL cation owing to its donor hydrogen bond
ability is important. Also, Chiappe et al. studying the
bromination of alkynes in ILs, accounted the possibility that
C
(pKBH Z11.12 and 11.07, respectively), the same geometry
and flexibility but very different reactivity.
The above data show that the elimination reaction, in IL
solution, is very sensitive to the amine structure (i.e.,
flexibility and steric hindrance). In contrast, the base
strength plays a minor part. Then, it is possible to think,
on the grounds of literature reports,12 that the mechanism
proceeding by an E1cb (in MeOH)4 will be moved in the E2
direction in IL solution. Indeed, the E2 mechanism
involving a highly crowded activated complex should be
more affected than the E1cb mechanism by steric effects.
Table 2. Calculated second order rate constants and third order rate constants at 298.1 K for the elimination reaction of 1a and 1b, in ionic liquids solution, in
the presence of aminesa,b
IL
Substrate
Amine
kII (MK1 sK1
)
kIII (MK2 sK1
)
i
n
R
[BMIM][BF4]
1a
Pyrrolidine
0.00916
(0.00038)
0.00124
(0.00010)
0.00109
K4.88!10K5
(8.48!10K6
K3.33!10K6
(2.28!10K6
K5.21!10K6
(4.87!10K7
1.42!10K6
(6.00!10K7
2.21!10K6
(3.54!10K7
7
0.996
)
)
)
)
)
Piperidine
7
7
7
7
6
7
0.983
0.999
0.989
0.999
0.996
0.987
Hexamethyleneimine
Heptamethyleneimine
Piperidine
(2.20!10K5
0.000404
)
(2.71!10K5
0.000851
)
[BMIM][PF6]
[BdMIM][BF4]
[BMIM][BF4]
1a
1a
1b
(1.65!10K5
)
Piperidine
0.0633
(0.014)
Piperidine
0.000372
(3.02!10K5
K1.11!10K6
(6.31!10K7
)
)
a The values of kobs, from which the k and kIII values were obtained, were reproducible within G3%.
II
b Standard deviations are given in parenthesis.

F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
1695
the imidazolium cation could assist the bromine–bromine
bond breaking, in the 1:1 p complex, to give a bromirenium
bromide intermediate.15 On the other hand, p–p inter-
actions are able to increase the coplanarity between
aromatic rings then favoring the conjugation and stabilizing
the transition state. The relevance of p–p interactions in IL
media has been previously reported by Atwood et al.
according to the hypothesis that the high solubility of
aromatic compounds in ILs could be related to some kind of
clathrate formation.16
methyl acrylate and cyclopentadiene, in IL, was influenced by
hydrogen bond ability of the cations.20 At last, conversions
and reaction rates of Tsuji–Trost allylic substitution, in BMIM
and BdMIM, were explained considering the different nature
of cations used.21
Previously, it was reported that the elimination of 1,1,1-
trihalo-2,2-bis(phenyl-substituted)ethanes was affected by
steric effects. In fact, the ortho-substituted derivatives were
found to be less reactive than the corresponding unsub-
stituted derivatives as a consequence of the hindrance to
conjugation between aromatic ring and p-electrons of the
forming double bond.4b As ionic liquids are organizing
media, we verified if they were able to decrease the entity of
steric effects. So we carried out the piperidine induced
elimination of 1b in [BMIM][BF4]. Effectively, the kinetic
results show that the 1b reactivity in IL is less affected by
steric effects and prevalently determined by electronic
effects. In fact, the reactivity ratio (kII)1a/(kII)1b is only 3.3 in
IL compared to 600 calculated for methanol solution;22 this
high value was attributed to the fact that steric effects were,
in determining the different reactivities, more important
than electronic ones. Indeed, the latter should equally act in
both substrates. So the result obtained for reaction of 1b in
IL–piperidine mixture can be interpreted as due to a
decrease in steric effects. The IL, by means of p–p
interactions, constrains the aryl rings in 1b to a coplanar
conformation despite the ortho-methoxy groups.
It is noteworthy that the imidazolium cation acidity, as well
as, its hydrogen bond ability could negatively affect the
reaction owing to a decrease in free amine concentration.
However, the data collected show that the activating
assistance to bromine departure seems to be predominant.
In order to study the effect of a different structure of IL
cation or anion, the piperidine induced elimination of 1a
was also carried out in [BMIM][PF6] and in [BdMIM][BF4].
The reaction was faster (1.45 times) in [BMIM][BF4] than
in [BMIM][PF6]. A similar reactivity trend in these ionic
liquids was recently observed. Chi attributed this trend to a
lower solubility of nucleophile in hexafluorophosphate than
in tetrafluoroborate.17 Welton claimed that changing the
anion affected the halide nucleophilicity.18 Recently we,
studying an heterocyclic rearrangement, explained the
higher reactivity in [BMIM][BF4] than in [BMIM][PF6],
as a consequence of a different ‘packing’ of two ionic
liquids. This determines a different catalytic effect.7
Probably the same explanation could be used for the data
reported herein, as the present elimination reaction seems
more influenced by steric or stereoelectronic factors rather
than by basicity of amines.
2.2. Activation parameters
It is well known that, for kinetics carried out in ILs, in some
cases, a significant curvature in Arrhenius or Eyring plots
can be observed as a consequence of structural changes in
the solvent.23 So for a careful analysis of the temperature
effect, the elimination reaction was carried out at five
temperatures going from 293.1 K up to 313.1 K. For each
amine and IL, an excellent linear correlation of log (kobs/T)
versus 1/T was obtained. This indicates that, in the analyzed
range, the above upsetting effect is not operative. So the
calculated activation parameters are only dependent on the
elimination process. The activation parameter values are
reported in Table 3, for an useful comparison the values for
methoxide/methanol induced elimination are also reported
(data collected at different temperatures are available in
Supplementary data: Table 5).
A more interesting behavior was found in [BdMIM][BF4]
solution; in fact, in this case the observed kinetic rate constant
shows a dependence on second order amine concentration,
that is, kobsZ(kIII)amine[Amine].2 The observed trend could be
a further support to the occurrence of the E2 mechanism.
Indeed, the lesser donor hydrogen bond ability of the
[BdMIM][BF4] that hampers the assistance to bromine
departure, causes the intervention of a second amine molecule
to favor product formation. The effect of imidazolium cation
in BMIM and BdMIM due to different hydrogen bond donor
ability is well documented. Welton et al. reported a decrease
in nucleophilicity of chloride anion, going from BdMIM to
BMIM, owing to its stabilization via hydrogen bonding.19
Also the endo-selectivity for Diels–Alder reaction between
Theenthalpyvaluesrangefrom47.4 kJ/molupto64.7 kJ/mol,
whereas the entropy values range from K180 J/K mol up
toK119 J/K mol. Thecollectedvaluesshowthat, withrespect
Table 3. Activation parameters for the elimination reaction of 1a and 1b, in ionic liquids solution, in the presence of aminesa
IL
Substrate
Base
DHs(kJ/mol)
DSs(kJ/mol)
MeOH
[BMIM][BF4]
1a
1a
MeOK
Pyrrolidine
Piperidine
Hexamethyleneimine
Heptamethyleneimine
Piperidine
71.9b
K72b
54.9 (3.2)
47.4 (2.2)
59.0 (2.8)
52.1 (4.9)
60.5 (2.4)
64.7 (4.6)
36.7 (2.8)
K135 (10)
K180 (7)
K138 (9)
K165 (16)
K132 (8)
K119 (15)
K220 (9)
[BMIM][PF6]
[BdMIM][BF4]
[BMIM][BF4]
1a
1a
1b
Piperidine
Piperidine
a Standard deviations are given in parenthesis.
b See Ref.4a.

1696
F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
to that induced by methoxide, the elimination induced by
amines is enthalpy favored but entropy disfavored. These
differencescould be due tothe fact that in methanol, compared
to the IL, strongerinteractions and more extensive solvation of
initial state with respect to the transition state are operative.
This causes a larger enthalpic contribution but a less
unfavorable entropic one.
The same factor could cause the highly negative entropic
contribution as a forcedly ordered activated complex,
entropically hampered, is needed.
3. Conclusions
The data collected herein confirm the idea that ionic liquids
represent an intriguing solvent system that can not be only
described by means of the usual solvent parameters. Indeed,
most of the obtained results seem to be a consequence of the
order as well as the organizing ability of these systems. For
example, the addition of a small quantity of amine solution
induces a reorganization of ionic liquid structure as detected
by NMR experiments. In addition, the structure rather than
amine basicity determines the occurrence of the reaction.
This seems to indicate that ionic liquids induce, for the
studied reaction, a shift of mechanism from E1cb (in MeOH)
versus E2 (in ionic liquid). However, it is probable that in
ionic liquid deep changes in amine structure could
correspond to significant variations in amine basicity, so
the observed reactivity could reflect the basicity in ionic
liquids. Unfortunately the lack of basicity data in ionic
liquid does not allow to verify this hypothesis. The
activating effect of ionic liquid on elimination of 1a could
be a consequence both of the electrostatic interactions
(bromine–imidazolium ion) and p–p interactions.
In IL, the particularly unfavorable entropic contribution is
according to both the E2 mechanism and not relevant
differences in solvation between initial state and transition
state.
The activation parameter values trend for elimination
induced by piperidine in [BMIM][BF4], [BdMIM][BF4]
and [BMIM][PF6] can be explained considering that
different cation–anion interactions are operative. Welton
claimed that a charge-separated activated complex from
neutral starting materials can induce a disruption of the ionic
liquid structure leading to a less negative entropy value.24
The main contributions to this effect could be due to the
degree of cation–anion interaction and to hydrogen-bond
acceptor and donor ability of ionic liquid. It is noteworthy
that the outcome of the reaction depends on the balance of
the above factors that can also act in opposite directions.
In the studied reaction the above effects could be a
consequence of the interaction between the anion and the
ammonium acid proton in the activated complex. A careful
analysis of entropy values seems to indicate that only the b
solvent parameter is unable to explain the observed trend.
The [BMIM][BF4] having the higher b value shows the more
negative DSs value. Thus the strong interaction cation–
anion should be responsible of the scarce disruption of ionic
liquid structure. This can only be a partial explanation, as
[BdMIM][BF4], having similar b value, shows a less
negative entropy value. Furthermore, [BMIM][PF6] having
the lowest b value shows an intermediate entropy value. Also
the other solvent parameters do not explain the variation of
activation parameters as a function of IL. This could be a
consequence of the fact that the solvent parameters for ILs
are probably inadequate. Indeed, it is well known that it is
questionable whether the empirically derived measurements
of solvent properties could be exclusively referred to room
temperature ionic liquids or whether they are also affected by
the nature of the tested compounds.25 However, the entropy
increase, going from an IL to another, could be attributed to
an easier disruption of ionic liquid structure owing to weaker
electrostatic interactions between cation–anion pairs,
according to that previously reported by Welton et al.17
The data collected show that the reaction rate is influenced
by dimension and charge distribution in ionic liquid anion as
well as by the hydrogen bond ability of cation that could
assist the bromine departure. Finally, in the case of 1b, the
organizing ability of ionic liquids is able to minimize the
unfavorable steric effects, operative in conventional organic
solvents.
4. Experimental
4.1. Materials
1,1,1-Tribromo-2,2-bis(3,4-dimethoxyphenyl)ethane (1a)
and the corresponding 1,1-dibromo-2,2-bis(3,4-dimethox-
yphenyl)ethene (2a) were prepared according to a procedure
reported.26
4.1.1. 1,1,1-Tribromo-2,2-bis(2,5-dimethoxyphenyl)-
ethane (1b). To a stirred solution of 1,4-dimethoxybenzene
(4.25 g, 0.03 mol) and bromal (1.68 g, 0.006 mol) in glacial
acetic acid (20 mL), 98% sulfuric acid (7.5 mL) was added
dropwise, while the temperature was maintained below
30 8C. After standing at room temperature overnight, the
mixture was poured on to crushed ice and the precipitate
was filtered, neutralized and dried. The product was purified
by chromatography over silica gel employing a mixture of
light petroleum–ethyl acetate (10/1) and crystallized from
ethanol (yield 1.93 g). White crystals, mp: 118–120 8C.
The above discussion is also supported by enthalpy values,
that are higher for [BMIM][PF6] and [BdMIM][BF4] than
[BMIM][BF4].
Activation parameter values for 1b show that the piperidine
induced elimination is enthalpy but not entropy favored with
respect to 1a. The decrease in enthalpy could be related to a
gain in energy due to an increase, going from initial state to
transition state, in conjugation between the aromatic rings
and p-electrons of the forming double bond, which favors
the reaction of 1b.
IR (Nujol) nmax 1059, 1284 cmK1. 1H NMR dH (250 MHz;
CDCl3): 3.76 (s, 6H, 2OCH3); 3.83 (s, 6H, 2OCH3); 5.40 (s,
1H); 6.70–6.83 (m, 4H, Ar); 7.10 (d, 2H, JZ2.7 Hz, Ar);
13C NMR dC (250 MHz; CDCl3): 48.3; 52.9; 55.6; 56.2;
112.1; 112.3; 115.9; 130.3; 151.3; 153.3. Anal. Calcd for

F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
1697
C18H19Br3O4: C, 40.10; H, 3.55; Br, 44.47%. Found: C,
40.15; H, 3.48; Br, 44.70%.
To evaluate the possibility of reusing ILs, we tried a fast and
simple treatment of the solvent used. Thus, 5 mL of the used
[BMIM][BF4] was extracted four times with 3 mL of Et2O.
The IL layer was kept under vacuum at 60 8C for 2 h and
reused. The apparent first-order rate constants then obtained
were reproducible within G15% with respect to values
determined in fresh IL.
4.1.2. 1,1-Dibromo-2,2-bis(2,5-dimethoxyphenyl)ethene
(2b). 1,1,1-Tribromo-2,2-bis(2,5-dimethoxyphenyl)ethane
(2.14 g, 0.004 mol) (1b) was dehydrobrominated by heating
under reflux with a solution of CH3ONa (0.43 g, 0.008 mol)
in dry CH3OH (10 mL). The crude dehydrohalogenated
was purified by chromathography over silica gel employing
a mixture of light petroleum–ethyl acetate (15/1) and
crystallized from ethanol (yield 1.0 g). White crystals, mp:
115–116 8C.
All kinetic data were analyzed by means of the KALEIDA-
GRAPH 3.0.1 software.
Acknowledgements
IR (Nujol) nmax 1053, 1309, 1583 cmK1 1H NMR dH
.
We thank MIUR (PRIN 2004): ‘Non-aromatic heterocycles
in stereo-controlled processes’ and University of Palermo
for financial support.
(250 MHz; CDCl3): 3.72 (s, 6H, 2OCH3); 3.77 (s, 6H,
2OCH3); 6.77–6.87 (m, 4H); 7.05 (s, 2H); 13C NMR dC
(250 MHz; CDCl3): 55.7; 56.5; 112.7; 112.9; 116.7; 130.1;
140.3; 151.3; 153.4. Anal. Calcd for C18H18Br2O4: C, 47.19;
H, 3.96; Br, 34.88%. Found: C, 47.40; H, 3.85; Br, 34.79%.
Supplementary data
All other products were commercial. [BMIM][BF4],
[BdMIM][BF4] and [BMIM][PF6] were purchased from
Solvent innovation, were dried on a vacuum line at 60 8C at
least for 2 h and stored in a dryer under argon and over
calcium chloride. 1,4-Dioxane (for fluorescence) was
purchased from Fluka and was used without further
purification. Amines (Aldrich) were freshly distilled before
use. UV–vis spectra and kinetic measurements were carried
out by using a Beckman DU 800 spectrophotometer
equipped with a peltier temperature controller, able to
keep the temperature within 0.1 K. NMR spectra were
collected on a Bruker AC-E Series 250 spectrometer.
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.tet.2005.11.061.
References and notes
1. (a) Welton, T. Chem. Rev. 1999, 99, 2071–2083. (b) Holbrey,
J. D.; Seddon, K. R. J. Chem. Soc., Dalton Trans. 1999,
2133–2139. (c) Earle, M. J.; Seddon, K. R. Pure Appl. Chem.
2000, 72, 1391–1398. (d) Wasserscheid, P.; Keim, M. Angew.
Chem., Int. Ed. 2000, 39, 3772–3789. (e) Rogers, R. D.;
Seddon, K. R. Ionic Liquids: Industrial Applications to Green
Chemistry; ACS Symposium Series 818; American Chemical
Society: Washington, DC, 2002. (f) Ionic Liquids in Synthesis;
Wasserscheid, P., Welton, T., Eds.; Wiley-VCH: Weinheim,
2003. (g) Rogers, R. D.; Seddon, K. R. Ionic Liquids as Green
Solvents. Progress and Prospects; ACS Symposium Series
855; American Chemical Society: Washington, DC, 2003.
(h) Wilkes, J. S. J. Mol. Catal. A: Chem. 2004, 214, 11–17.
(i) Picquet, M.; Poinsot, D.; Stutzmann, S.; Tkatchenko, I.;
Tommasi, I.; Wasserscheid, P.; Zimmermann, J. Top. Catal.
2004, 29, 139–143.
4.2. Measurements and calculations
In NMR measurements 500 mL of IL were added to a 5 mm
NMR tube, under argon. 75 mL of 1,4-dioxane or 75 mL of
amine solution were added to IL, by a syringe. A stem
coaxial capillary tube, loaded with DMSO-d6, was inserted
into the 5 mm NMR tube and this solvent was used as
external lock.
All UV–vis spectra of IL–amine solutions were recorded
against air.
Kinetic runs were carried out over the temperature range
293.1–313.1 K. The sample for a typical kinetic run was
prepared by injecting into a quartz cuvette (optical
path 0.2 cm) 500 mL of IL, 50 mL of a solution of 1 in
1,4-dioxane, and then 25 mL of a concentrated solution of
amine in 1,4-dioxane, previously thermostated. The con-
centration of 1 was constant and equal to 0.00019 M, and
the amine concentration ranging from 0.008 up to 0.0304 M.
To avoid that the reorganization process of IL, induced by
amine solution, affected the kinetic run, all measurements
were carried out by using as reference a sample prepared
injecting into a quarz cuvette 500 mL of IL, 50 mL of
1,4-dioxane and then 25 mL of a concentrated solution of
amine in 1,4-dioxane. In this manner, the net absorbance at
lZ280 nm for 1a and at lZ270 nm for 1b was plotted
versus time and showed a simple exponential dependence.
The reactions were all studied over 6 half-lives or more. In
all cases the correlation coefficient was higher than 0.9998.
2. (a) Laali, K. K.; Borodkin, G. I. J. Chem. Soc., Perkin Trans. 2
´
2002, 953–957. (b) Lagrost, C.; Carrie, D.; Vaultier, M.;
Hapiot, P. J. Phys. Chem. A 2003, 107, 745–752. (c)
Wishart, J. F.; Neta, P. J. Phys. Chem. A 2003, 107,
745–752. (d) Debreul, J. F.; Bazureau, J. P. Tetrahedron
Lett. 2004, 41, 7351–7355. (e) Conte, V.; Floris, B.; Galloni, P.
Green Chem. 2005, 7, 262–266. (f) Klein, H.; Jackstell, R.;
Beller, M. Chem. Commun. 2005, 17, 2283–2285. (g)
Xiao, J. C.; Shreeve, J. M. J. Org. Chem. 2005, 70, 3072–3078.
(h) Siddiqui, S. A.; Narkhede, U. C.; Palimkar, S. S. Tetrahedron
2005, 61, 3539–3546.
3. (a) McLean, A. J.; Muldoon, M. J.; Gordon, C. M.; Dunkin, I. R.
Chem. Commun. 2002, 1880–1881. (b) Kim, D. W.; Song, C. E.;
Chi, D. Y. J. Am. Chem. Soc. 2002, 124, 10278–10279. (c)
Skzypczak, A.; Neta, P. J. Phys. Chem. A 2003, 107,
7800–7803. (d) Kim, D. W.; Hong, H. K.; Seo, J. W.;
Kim, H. S.; Kim, H. K.; Song, C. E.; Chi, D. Y. J. Org.
Chem. 2004, 69, 3186–3189. (e) Chiappe, C.; Pieraccini, D.

1698
F. D’Anna et al. / Tetrahedron 62 (2006) 1690–1698
J. Org. Chem. 2004, 69, 6059–6064. (f) Dal, E.; Lancaster, N. L.
Org. Biomol. Chem. 2005, 3, 682–686.
14. Carmichael, A. J.; Seddon, K. R. J. Phys. Org. Chem. 2000, 13,
591–595.
15. Chiappe, C.; Conte, V.; Pieraccini, D. Eur. J. Org. Chem.
2002, 2831–2837.
4. (a) Fontana, G.; Frenna, V.; Gruttadauria, M.; Natoli, M. C.;
Noto, R. J. Phys. Org. Chem. 1998, 11, 54–58. (b) Fontana, G.;
Frenna, V.; Lamartina, L.; Natoli, M. C.; Noto, R. J. Phys.
Org. Chem. 2002, 15, 108–114.
16. (a) Atwood, J. L. In Atwood, J. L., Davies, J. E. D., MacNicol,
D. D., Eds.; Liquid Clathrates, Inclusion Compounds;
Academic: London, 1984; Vol. 1. (b) Holbrey, J. D.;
Reichert, W. M.; Nieuwenhuyzen, M.; Sheppard, O.;
Hardacre, C.; Rogers, R. D. Chem. Commun. 2003, 476–477.
(c) Deetlefs, M.; Hardacre, C.; Nieuwenhuyzen, M.; Sheppard, O.;
Soper, A. K. J. Phys. Chem. B 2005, 109, 1593–1598.
17. Kim, D. W.; Song, C. E.; Chi, D. Y. J. Org. Chem. 2003, 68,
4281–4285.
5. (a) Earle, M. J.; Katdare, S. P.; Seddon, K. R. Org. Lett. 2004,
6, 707–710. (b) Chiappe, C.; Pieraccini, D. Arkivoc 2002, xi,
249–255. (c) Shen, H. Y.; Yudeh, Z. M. A.; Ching, C. B.;
Xia, Q. H. J. Mol. Catal. A: Chem. 2004, 212, 301–308.
6. (a) Frenna, V.; Vivona, N.; Consiglio, G.; Spinelli, D. J. Chem.
Soc., Perkin Trans. 2 1985, 1865–1868. (b) Hall, H. K., Jr. J.
Am. Chem. Soc. 1957, 79, 5441–5444. (c) Lide, D. R. CRC
Handbook of Chemistry and Physics, 80th ed.; CRC: London,
1999–2000.
18. Lancaster, N. L.; Welton, T. J. Org. Chem. 2004, 69,
5986–5992.
19. Lancaster, N. L.; Salter, P. A.; Welton, T.; Young, G. B.
J. Org. Chem. 2002, 67, 8855–8861.
7. D’Anna, F.; Frenna, V.; Noto, R.; Pace, V.; Spinelli, D. J. Org.
Chem. 2005, 70, 2828–2831.
20. Aggarwal, A.; Lancaster, N. L.; Sethi, A. R.; Welton, T. Green
Chem. 2002, 4, 517–520.
8. Glenn, A. G.; Jones, P. B. Tetrahedron Lett. 2004, 45,
6967–6969.
21. (a) Chen, W. P.; Xu, L. J.; Chatterton, C.; Xiao, J. L. Chem.
Commun. 1999, 1247–1248. (b) Ross, J.; Chen, W. P.;
Xu, L. J.; Xiao, J. L. Organometallics 2001, 20, 138–142.
22. The (kII)1b value (4.51!10K5 MK1 sK1) in methanol solution,
at 313.1 K, was determined in this work.
9. (a) Avent, A. G.; Chaloner, A. P.; Day, M. P.; Seddon, K. R.;
Welton, T. J. Chem. Soc., Dalton Trans. 1994, 3405–3413.
ˆ
(b) Bonhote, P.; Dias, A.-P.; Papageorgiou, N.;
¨
Kalymnasundraram, K.; Gratzel, M. Inorg. Chem. 1996, 35,
1168–1178. (c) Tubbs, J. D.; Hoffmann, M. M. J. Sol. Chem.
2004, 33, 381–394.
23. Gordon, C. M.; McLean, A. J.; Muldoon, M. J.; Dunkin, I. R.
In Ionic Liquids as Green Solvents. Progress and Prospects;
Rogers, R. D., Seddon, K. R., Eds.; ACS Symposium Series
855; American Chemical Society: Washington, DC, 2003;
pp 357–369.
10. Mele, A.; Tran, C. D.; De Paoli Lacerda, S. H. Angew. Chem.,
Int. Ed. 2003, 42, 4364–4366.
11. Yadav, J. S.; Reddy, B. V. S.; Basak, A. K.; Navsaiah, A. V.
Tetrahedron Lett. 2003, 44, 2217–2220.
24. Crowhurst, L.; Lancaster, N. L.; Arlandis, J. M.; Welton, T.
J. Am. Chem. Soc. 2004, 126, 11549–11555.
12. Smith, M. B.; March, J. March’s Advanced Organic
Chemistry: Reactions, Mechanisms and Structure 5th ed.;
Wiley-Interscience, Wiley: New York, 2001; pp 1317–1320.
13. Chiappe, C.; Pieraccini, D.; Sullo, P. J. Org. Chem. 2003, 68,
6710–6715.
25. Armstrong, D. W.; He, L.; Liu, Y.-S. Anal. Chem. 1999, 71,
3873–3876.
26. Arcoleo, A.; Paternostro, M. P. Ann. Chim. (Rome) 1968, 58,
290–297.