Amine catalysis in the vinylic substitution of α-methylthio-α-arylmethylene Meldrum's acids and its absence in the substitution of methyl β-iodo-α-nitrocinnamate by amines

Article abstract of DOI:10.1039/b103486n

Substitution of the iodine of (E)- and (Z)-methyl β-iodo-α-nitrocinnamates (5) by amines gives identical (Z)-enamines with aniline (Ani) and piperidine (Pip). No amine catalysis was observed with Pip, Ani, morpholine (Mor), orp-MeOC₆H₄NHMe (MMA) in MeCN nor with Pip or Mor in EtOH: k_Pip/k_Mor = 115-138 (MeCN), 3.3-6.9 (EtOH); k_MeCN/k_EtOH = 25.5 ± 2.2 (Pip), 0.79-1.16 (Mor); k_(Z)-5/k_(E)-5 = 1.3-2.9 (13.5 with MMA in MeCN). Replacement of the MeS group in six α-amethylthio-α-arylmethylene Meldrum's acid (6-X) by Pip resulted in amine catalysis in MeCN and EtOH. In EtOH, the p-anisyl derivative (6-MeO) and in MeCN 6-MeO, 6-Me and 6-H displayed second order catalysis in Pip. Other 6-X compounds show orders between one and two in Pip with amine catalyzed (k_3B)/non-catalyzed (k₂) rate coefficient ratios of 281-731 (EtOH) and 504-635 (MeCN) at 30°C. k_MeCN/k_EtOH = 3.0-4.9. In MeCN ΔH? = -0.8 to -5.9 kcal mol^-1 and ΔS? = -50 to -72 e.u. An intermediate zwitterion, 3a, is formed in all cases. For system 5 the rate of I^- expulsion from 3a exceeds its deprotonation rate, and the observed rate coefficient is composite: k_obs = k₁k₂/k_-1 in MeCN (k₁ = rate coefficient of nucleophilic attack) but k_obs = k₁ in EtOH. In MeCN the deprotonation is faster than the expulsion rate of MeS^-, and more so for 6-X with X = p-Br, p-CF₃, m,m′-(CF₃)₂. Different electrophilicities of 6-X, different extents of hydrogen bonding, steric and electronic effects account for the kinetic differences.

Full text of DOI:10.1039/b103486n

View Article Online / Journal Homepage / Table of Contents for this issue
Amine catalysis in the vinylic substitution of ꢀ-methylthio-ꢀ-
arylmethylene Meldrum’s acids† and its absence in the substitution
of methyl ꢁ-iodo-ꢀ-nitrocinnamate by amines‡§
2
Michal Beit-Yannai, Xin Chen and Zvi Rappoport*
Department of Organic Chemistry, The Hebrew University, Jerusalem 91904, Israel
Received (in Cambridge, UK) 18th April 2001, Accepted 10th July 2001
First published as an Advance Article on the web 16th August 2001
Substitution of the iodine of (E)- and (Z)-methyl β-iodo-α-nitrocinnamates (5) by amines gives identical
(Z)-enamines with aniline (Ani) and piperidine (Pip). No amine catalysis was observed with Pip, Ani, morpholine
(Mor), or p-MeOC6H4NHMe (MMA) in MeCN nor with Pip or Mor in EtOH: kPip/kMor = 115–138 (MeCN),
3.3–6.9 (EtOH); kMeCN/kEtOH = 25.5 2.2 (Pip), 0.79–1.16 (Mor); k(Z)-5/k(E)-5 = 1.3–2.9 (13.5 with MMA in MeCN).
Replacement of the MeS group in six α-methylthio-α-arylmethylene Meldrum’s acid (6-X) by Pip resulted in amine
catalysis in MeCN and EtOH. In EtOH, the p-anisyl derivative (6-MeO) and in MeCN 6-MeO, 6-Me and 6-H
displayed second order catalysis in Pip. Other 6-X compounds show orders between one and two in Pip with
amine catalyzed (k3B)/non-catalyzed (k2) rate coefficient ratios of 281–731 (EtOH) and 504–635 (MeCN) at 30 ЊC.
kMeCN/kEtOH = 3.0–4.9. In MeCN ∆H‡ = Ϫ0.8 to Ϫ5.9 kcal molϪ1 and ∆S‡ = Ϫ50 to Ϫ72 e.u. An intermediate
zwitterion, 3a, is formed in all cases. For system 5 the rate of IϪ expulsion from 3a exceeds its deprotonation rate,
and the observed rate coefficient is composite: kobs = k1k2/kϪ1 in MeCN (k1 = rate coefficient of nucleophilic
attack) but kobs = k1 in EtOH. In MeCN the deprotonation is faster than the expulsion rate of MeSϪ, and
more so for 6-X with X = p-Br, p-CF3, m,mЈ-(CF3)2. Different electrophilicities of 6-X, different extents
of hydrogen bonding, steric and electronic effects account for the kinetic differences.
A major question concerning the mechanism of nucleophilic
(Z)-precursor;1 (b) the element effect, i.e., when X = halogen
vinylic substitution of an electrophilic alkene [eqn. (1), XϪ =
the concerted route would show kF/kCl, kCl/kBr << 1 and the
multi-step route would give kF/kCl >> 1, kBr/kCl ≈ 1);1 (c) the
kinetics, since a deviation from an overall second order reaction
may indicate the formation of an intermediate;1 (d) calculations
which compare the energies, and hence the feasibility of both
routes;1g,2 and (e) attempts or success in direct observation of
the intermediate 3.3 The main conclusion from the study of
many systems is that the transition state is variable1d,f and that
the reaction may be concerted when X is a very good nucleo-
fuge and the alkene is only slightly electrophilic, whereas highly
electrophilic alkenes, especially those carrying a poor or a
moderate nucleofuge, react via the multi-step route.
R1C (X)᎐C R2R3 ϩ NuϪ
R1C (Nu)᎐C R2R3 ϩ XϪ (1)
᎐
᎐
α
β
α
β
nucleofuge (leaving group), NuϪ = anionic nucleophile]1 is
whether the reaction proceeds via a single step (concerted) route
via transition state 2 [eqn. (2)] or whether it is a multi-step route
(2)
A variation of eqn. (2) is when the nucleophile is neutral,
mostly an amine. The first-formed intermediate is then the zwit-
terion 3a rather than a carbanion3b [eqn. (2a)]. In moderately
electrophilic alkenes, i.e., when only one of the groups R2 or R3
is a strongly electron-withdrawing group (EWG) the kinetics
are of an overall second order, i.e., first order in the amine.
However, with highly electrophilic alkenes when both R2 and
R3 are strongly EWGs the reaction will proceed by two multi-
step routes, a second order 1 3a 4 process and a third
order 1 3a 3b 4 process. This is because the expulsion rate
coefficient of XϪ (k2) in carbanions 3 is usually very high com-
pared with the nucleophilic attack step k1 which becomes rate
determining, leading to an overall second order. In the zwitter-
ion 3a the expulsion rate of XϪ is significantly reduced due to
the strong electron withdrawal by the positively charged
ammonio moiety. The longer life-time and the presence of the
acidic proton in 3a enables a rate limiting proton transfer from
3a to another amine molecule (k3B), forming carbanion 3b,
which expels XϪ rapidly to form 4 [eqn. (2a)].
proceeding via formation of intermediate 3 which expels XϪ to
form the product 4. Several major probes which were applied
to answer this question1g are (a) the stereochemistry of
the reaction, since an intermediate carbanion will give stereo-
convergence of the product starting from the pure (E)- or
† The IUPAC name for Meldrum’s acid is 2,2-dimethyl-1,3-dioxane-
4,6-dione.
‡ Dedicated to the memory of Lennart Eberson, a friend and a great
chemist.
§ Electronic supplementary information (ESI) available: stereoviews of
compounds 7a and 7b and experimental details of X-ray crystal-
lography. See http://www.rsc.org/suppdata/p2/b1/b103486n/
A steady state treatment of eqn. (2a) gives rate eqn. (3)
and the observed second order rate coefficient kobs is given by
eqn. (4). The observed kinetics will depend on the relative
1534
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b103486n

View Article Online
NO2 and CO2Me groups, where the stereoconvergence in the
substitution of (E)-5 and (Z)-5 by a thio-nucleophile5 indicates
a multi-step route although iodine is a good nucleofuge. Sys-
tems activated by a Meldrum’s acid residue which are capable
of an extensive negative charge dispersal are also expected to
show amine catalysis. Since we were unable to prepare such
systems which carry good nucleofuges we studied such systems
(6) which carry the moderate nucleofuge MeS. Vinylic substi-
tution in such systems was previously investigated with anionic
nucleophiles.3i,j,m We expected to observe amine catalysis and to
compare the observed ratios of rate coefficients with those of
other activated systems.
(2a)
Rate = k1[1][Amine]{k2 ϩ k3B[Amine]}/
{kϪ1 ϩ k2 ϩ k3B[Amine]} (3)
Results
Precursors
kobs = Rate/[1][Amine] =
A mixture of (E)-5 and (Z)-5 was prepared by addition of
nitrogen tetraoxide and iodine to methyl phenylpropiolate
[eqn. (8)]. The preparation and separation to isomers5 will be
discussed elsewhere.
k1{k2 ϩ k3B[Amine]}/{kϪ1 ϩ k2 ϩ k3B[Amine]} (4)
importance of the competing non-catalyzed (k2) and the base
(B) catalyzed (k3B) terms. If kϪ1 < k2 ϩ k3B[Amine] eqn. (5)
(8)
kobs = k1
(5)
applies and k1 becomes rate determining. This is expected to
apply in most cases when X is a good nucleofuge. However,
when X is a poor nucleofuge k2 is smaller than in the previous
case, and the inequality kϪ1 > k2 ϩ k3B[Amine] and hence eqn.
(6) may apply, i.e., both routes compete with one another.
The six Meldrum’s acid derivatives 6-X (X = p-MeO, p-Me,
H, p-Br, m-CF3, m,mЈ-(CF3)2) were prepared in two consecutive
steps. (1) Reaction of Meldrum’s acid with CS2 and Et3N and
alkylation of the formed dianion with methyl iodide in DMSO
led to the 5-bis(methylthio)vinylidene derivative. (2) Vinylic
substitution of a MeS group of the vinylidene derivative by an
aryl group using arylmagnesium bromide [eqn. (9)].6
kobs = k1k2/kϪ1 ϩ k1k3[Amine]/kϪ1
(6)
Three kinetic situations could therefore be observed: (a) no
dependence of kobs on the amine [eqn. (5)]; (b) a linear kobs vs.
[Amine] plot [eqn. (6)] whose slope is k1k3B/kϪ1 and intercept
k1k2/kϪ1 give the k3B/k2 ratios; (c) a non-linear kobs vs. [Amine]
plot [eqn. (4)] which leads to an inverted 1/kobs vs. 1/[Amine]
linear plot [eqn. (7)] whose intercept and slope are 1/k1 and kϪ1
/
k1k3B, respectively.
1/kobs = 1/k1 ϩ (kϪ1/k1k3)(1/[Amine])
(7)
The three situations were experimentally observed: (a) amine
catalysis was not observed in the major part of vinylic substi-
tutions which include moderately activated systems carrying
good nucleofuges such as Cl, Br;1 (b) linear kobs vs. [Amine]
plots were observed in the substitution of systems carrying
poor nucleofuges such as F, OR, CN and the reactions were of
a second order or of an order between 1 and 2 in the amine. In
aprotic solvents the k3B/k2 ratios were usually higher than in
protic solvents.4
(9)
Since amine catalysis serves as evidence for the multi-step
route, a search of such catalysis for systems carrying a good
nucleofuge, where unequivocal evidence for this route is meagre,
was conducted. The prediction that in such systems the k3B/k2
ratios would be very low was fulfilled, but catalysis was
observed only for the highly activated β,β-dicyano activated
system 4 (X = Cl, Br)4a,e and for (Z)-α,β-dinitrostilbene.4g In all
cases the k3B/k2 ratios were indeed low.
Substitution
ArC(X)᎐C(CN)
4, Ar = p-Me2NC6H4, p-O2NC6H4; X = Cl, Br
᎐
2
A single geometrical isomer of each of the substitution
products 7a–d was obtained starting from either (E)-5 or (Z)-5
with aniline (Ani), piperidine (Pip), morpholine (Mor) or
p-methoxy-N-methylaniline (MMA). Enamines 7a–d were
synthesised according to eqn. (10) on a preparative scale and
identified. The substitution products 8-X of 6-X were prepared
according to eqn. (11) and identified by microanalysis when
sufficient material was available and by spectroscopic data
(including HRMS) when available amounts were very small.
(c) Curved kobs vs. [Amine] plots which give inverted l/kobs vs.
1/[Amine] plots were found for activated systems carrying poor
nucleofuges such as OEt or OCH2CF3.4c,f
In the present work we extended these studies to two add-
itional systems. We looked for amine catalysis in the reaction of
β-iodo-α-nitrocinnamate 5, a system highly activated by both
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545
1535

View Article Online
Table 1 Rate coefficients for the reaction of (E)-5 and (Z)-5 with amines at 30 ЊCa
Substrate
Amine
103[Amine]/M
Solvent
Order in amine
103kobs/MϪ1 sϪ1
(Z)-5
(E)-5
(Z)-5
(E)-5
(Z)-5
(E)-5
(Z)-5
(E)-5
(Z)-5
(E)-5
(Z)-5
(E)-5
(Z)-5
(E)-5
Piperidine
Piperidine
Morpholine
Morpholine
Aniline
1.1–1.9
72–95
0.75–10
0.75–5.0
9.1–35
5.0–24
49–80
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
EtOH
EtOH
EtOH
EtOH
THFb
MeCN
1
1
1
1
1
1
1
1
1
1
1
1
1770
725
12.8
6.3
0.8
0.6
3.1
0.2
76.0
26.2
11.0
8.0
Aniline
p-MeOC6H4NHMe
p-MeOC6H4NHMe
Piperidine
49–80
1.2–9.7
5.0–25
5.0–29
5.0–20
6.0–290
10
Piperidine
Morpholine
Morpholine
Aniline
1
0.16
c
1,3,5-(t-Bu)3C6H2NH2
a [Substrate] = 10Ϫ4 M. b Reaction in THF is very slow and no further experiments were conducted. c No reaction during 48 hours.
Table 2 Relative rates of reactions of nucleophiles with (E)- and (Z)-5
(10)
at 30 ЊCa
Substrate
Solvent
kPip/kMor
kPip/kAni
kPip/kMMA
(Z)-5
(E)-5
(Z)-5
(E)-5
MeCN
MeCN
EtOH
EtOH
138
115
6.9
3.3
2212
1208
571
3625
a Pip = Piperidine. Mor = Morpholine. Ani = Aniline. MMA = p-
Methoxy-N-methylaniline.
(11)
Table 3 Solvent effect on the relative reactivities of (E)-5 and (Z)-5 at
30 ЊC
System
kMeCN/kEtOH
kMeCN/kTHF
Piperidine ϩ (Z)-5
Morpholine ϩ (Z)-5
Piperidine ϩ (E)-5
Morpholine ϩ (E)-5
Aniline ϩ (Z)-5
23.3
1.16
27.7
0.79
Kinetics
5.0
The kinetics of the reaction of (E)-5 and (Z)-5 with piperidine,
morpholine, aniline and p-methoxy-N-methylaniline in MeCN
and of piperidine and morpholine in EtOH were followed.
No reaction of (E)-5 with the bulky amine 1,3,5-tri-tert-
butylaniline in MeCN was observed after 48 hours at 30 ЊC.
Table 4 Reactivity differences of (E)-5 and (Z)-5 at 30 ЊC
Amine
Solvent
k(Z)-5/k(E)-5
Piperidine
Piperidine
Morpholine
Morpholine
Aniline
MeCN
EtOH
MeCN
EtOH
MeCN
MeCN
2.4
2.9
2.0
1.4
1.3
Also no (E)-5
(Z)-5 isomerization was observed during the
substitution. All the reactions were of an overall second order,
being first order each in the amine and the substrate. No amine
catalysis was observed in spite of the use of amines within an
appreciable pKa range. The results are given in Table 1.
p-MeOC6H4NHMe
15.5
Three reactivity ratios of pairs of nucleophiles, i.e., kPip/kMor
,
kPip/kAni and kPip/kMMA are given in Table 2. Piperidine was
always the most reactive nucleophile, e.g., with (Z)-5 in MeCN
the relative reactivities were 1 (PhNH2) < 4.1 (p-MeOC6H4-
NHMe) < 16 (morpholine) < 2212 (piperidine). However, the
relative reactivities were much smaller in the protic EtOH, i.e.,
the kPip/kMor reactivity ratios were 20–35 fold higher in MeCN.
Solvent effects kMeCN/kEtOH were strongly dependent on the
amine. Aniline reacted five times faster with (Z)-5 in MeCN
than in THF (Table 3).
The relative reactivities of (Z)-5 to (E)-5 for piperidine,
morpholine and aniline are similar, with (Z)-5 being 1.3–2.9
fold more reactive in MeCN and EtOH. Only with p-methoxy-
N-methylaniline does this ratio increase significantly in MeCN
to 15.5 (Table 4).
m-F in the aromatic ring (6-X). These are a priori better
candidates to show amine catalysis than system 5 since MeSϪ
is a much poorer nucleofuge than IϪ and the non-catalytic
expulsion rate coefficient k2 becomes less competitive with the
catalyzed route (rate constant k3B[Pip]). System 6 also enables
a study of the substituent effect on the competition between
the two routes.
Tables 5–7 give the rate coefficients for the reaction of 6-X
with piperidine in MeCN at 30 and 40 ЊC and in EtOH at 30 ЊC.
D∞ values were measured at ≥10 half-lives. The amine concen-
trations were much higher than those of 6-X, and a pseudo first
order equation was used. Several measurements were taken and
the estimated errors in the rate coefficients and in the plots used
to calculate them are R2 ≥ 0.99. Isosbestic points were detected
for each of the reactions. The plots of the second order kobs vs.
[Amine] were usually linear [eqn. (12)]. Their slopes and inter-
cepts are the rate coefficients kЈ and kЉ. Assuming that eqn.
(2a) and hence eqn. (4) describe the mechanism, comparison of
Reaction of piperidine with 6-X
The substitution by amines was extended to system 6 carrying
the SMe nucleofuge, and five single substituents and one pair of
1536
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545

View Article Online
Table 5 Rate coefficients for the reactions of 6-X with piperidine in EtOH at 30 ЊC
Substrate
103[Amine]/M
Order in amine
kЈ/MϪ1 sϪ1
kЉ/MϪ2 sϪ1
kЉ/kЈ = k3B/k2/MϪ1
kٞ/M s
kЉЉ/M2 s
kٞ/kЉЉ = k3B/k1/MϪ1
6-MeO
6-Me
6-H
6.2–11
7.4–12
5.0–9.9
3.7–8.7
3.7–6.2
0.62–3.1
2
0.0022
0.018
0.014
0.074
0.079
6.82
7.12
10.24
20.58
30.00
3100
395
731
281
380
1–2
1–2
1–2
1–2
1–2
6-Br
6-CF3
6-(CF3)2
1.44
0.0021
686
Table 6 Rate coefficients for the reactions of 6-X with piperidine in MeCN at 30 ЊCa
Substrate
103 [Amine]/M
Order in amine
kЈ/MϪ1 sϪ1
kЉ/MϪ2 sϪ1
kЉ/kЈ/MϪ1
6-OMe
6-Me
6-H
2.5–7.4
1.2–3.7
3.7–6.2
1.2–3.7
1.2–3.7
0.6–1.8
2
2
2
1–2
1–2
1–2
0.004
0.007
0.036
0.12
0.29
0.58
20.4
27.7
34.8
76.2
146.0
341.0
5100
3960
9670
635
504
589
6-Br
6-CF3
6-(CF3)2
a [Substrate] = 5.8–6.0 × 10Ϫ5 M.
Table 7 Rate coefficients for the reactions of 6-X with piperidine in MeCN at 40 ЊCa
Substrate
103[Amine]/M
Order in amine
kЈ/MϪ1 sϪ1
kЉ/MϪ2 sϪ1
kЉ/kЈ/MϪ1
6-OMe
6-Me
6-H
2.2–3.7
2.5–5.0
3.7–6.2
1.2–3.7
1.2–2.4
0.6–1.8
2
2
2
1–2
1–2
1–2
0.0011
0.0017
0.0013
0.093
0.044
0.61
17.5
20.9
27.3
61.0
140.8
336.2
15910
12290
21000
656
3200
551
6-Br
6-CF3
6-(CF3)2
a [Substrate] = 5.8–6.0 × 10Ϫ5 M.
eqn. (4) and (12) give the k3B/k2 and k3B/k1 ratios in terms of kЈ
and kЉ [eqn. (13) and Tables 5–7].
kobs = kЈ ϩ kЉ[Amine]
kЈ = k1k2/kϪ1 kЉ = k1k3B/kϪ1
(12)
(13)
;
Compound 6-(CF3)2 reacted slowly both in EtOH or in EtOH
containing Et3N in the absence of piperidine. Substitution by
EtOϪ (and less likely by EtOH) may be responsible. The prod-
uct was not isolated. The half-life of the reaction in EtOH was
ca. 24 h at 30 ЊC (kobs = 1.16 × 10Ϫ5 Ϫ1), whereas with added
s
piperidine the half-lives were 80–730 min, depending on the
amine concentration.
The order of the reaction in the amine was substrate- and
solvent-dependent. In EtOH only 6-OMe having the least
electron-withdrawing aryl group displayed a second order reac-
tion in the amine. For all other substrates the order in the amine
was between 1 and 2 and the catalysis was extensive, kЉ/kЈ ratios
being 281–731. The reaction of 6-(CF3)2 with piperidine was the
only one which did not give a linear kobs vs. [Amine] plot [Fig.
1a, eqn. (14)]. However, a plot of 1/kobs vs. 1/[Amine] was linear
(Fig. 1b) and the derived kٞ and kЉЉ terms and their mechanistic
equivalents [eqn. (15)] are given in Table 5.
1/kobs = kٞ ϩ kЉЉ/[Amine]
kٞ = 1/k1; kЉЉ = kϪ1/k2k3B
(14)
(15)
The substitution rate increased on increasing the electron-
withdrawing ability of the substituent except that 6-H was
slower than 6-Me. In MeCN the dependence of the reaction
rate on the substituents followed that in EtOH. However, the
order in the amine was two for 6-OMe, 6-Me and 6-H and
only systems with more electron withdrawing substituents X
displayed an order between 1 and 2 in the amine. Fig. 2 is a plot
Fig. 1 (a) A kobs vs. [Amine] plot for the reaction of piperidine with 6-
(CF3)2 inEtOHat30 ЊC. (b)A1/kobs vs. 1/[Amine] plotforthesamereaction.
of the second order kobs vs. [piperidine] for 6-OMe in EtOH at
30 ЊC, and its linearity (and the small intercept) demonstrate
the overall third order of the reaction.
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545
1537

View Article Online
Table 9 kMeCN/kEtOH ratios in the reactions of 6-X with piperidine at
30 ЊC
Table 8 Activation parameters for the reactions of 6-X in MeCNa
∆S‡/cal molϪ1
Substrate
Ea/kcal molϪ1
∆H‡/kcal molϪ1
KϪ1
Substrate
kMeCN/kEtOH
6-OMe
6-Me
6-H
Ϫ3.0
Ϫ3.6
Ϫ64.4
Ϫ71.6
Ϫ57.5
Ϫ55.7
Ϫ69.5
Ϫ49.9
6-OMe
6-Me
6-H
3.0
3.9
3.4
3.7
4.9
Ϫ5.3
Ϫ1.1
Ϫ1.1
Ϫ0.2
Ϫ0.3
Ϫ5.9
Ϫ1.8
Ϫ1.7
Ϫ0.8
Ϫ0.9
6-Br
6-Br
6-CF3
6-(CF3)2
6-CF3
a Based on kЉ values.
Fig. 2 A plot of kobs of 6-OMe vs. [piperidine] in EtOH at 30 ЊC.
Fig. 4 An ORTEP drawing of 7a.
were determined by X-ray diffractions of single crystals. The
formation of the same product from both isomers is ascribed to
post-isomerization, i.e., a rapid isomerization of the least stable
to the most stable enamine after substitution, a known phe-
nomenon in the vinylic substitution by amines.1a,7
The structure around the double bond is of interest, since it
was suggested for β-nitroenamines having an amino hydrogen
that hydrogen bonding between the NH and the NO2 groups
determines the configuration4g,8 However, in our system
there are two EWGs, an NO2 and a CO2Me, and the question
is which one will be the preferred acceptor of hydrogen
bonding in 7a. Moreover, since no N–H proton is present
in 7b a related question is what will be its structure in the
absence of hydrogen bonding. It was previously suggested
for similar β-nitroenamines that the NO2 and amino groups
will still be cis to one another,4g and the X-ray structure
Fig. 3 A log kobs vs. σ plot for 6-X at 30 ЊC in MeCN.
Hammett plots of log kЉ vs. σ values for 6-X with piperidine
in EtOH at 30 ЊC and in MeCN at 30 (Fig. 3) and 40 ЊC are
linear except that in MeCN 6-H shows slightly negative devi-
ations at both temperatures, giving ρ values of 1.06 (30 ЊC) and
1.14 (40 ЊC) with correlation coefficients R2 > 0.98. In EtOH
the linearity is only approximate with R2 = 0.96 and ρ = 0.85
(30 ЊC).
With the caution required when using two close temperatures
(30 and 40 ЊC) for calculating the activation parameters, all the
Ea and ∆H‡ values in MeCN are slightly negative (Ea = Ϫ0.2–
Ϫ3.0 and ∆H‡ = Ϫ0.8–Ϫ5.9 kcal molϪ1) and the activation
entropies are highly negative (Ϫ50–Ϫ72 e.u.) (Table 8).
The aprotic (MeCN) vs. protic (EtOH) reactivity ratios do
not differ significantly with the substituent: kMeCN/kEtOH = 3.0–
4.9, with the lowest value for 6-OMe and the highest value for
6-(CF3)2 but with no observed trend for the whole set (Table 9).
of PhC(NO )᎐C(Ph)NC H O (NC H O = morpholino) shows
᎐
2
4
8
4
8
a (Z)-configuration.9
The ORTEP structure of 7a is shown in Fig. 4 and bond
lengths and angles are in Table 10. Analogous data for 7b are
given in Fig. 5 and Table 11. Additional data, including the
stereoviews of 7a and 7b are available as supplementary data.§
In both 7a and 7b the amino and the nitro groups are cis to one
another in a (Z)-configuration, suggesting a preferred hydrogen
bonding to the nitro group. Although an NO2 ؒ ؒ ؒ HN hydrogen
bonding may be important in the structure of 7a, the similar
structure of 7b where such a bond is impossible may indicate
that hydrogen bonding is not necessarily the major structure
determining factor.
Interesting features are: (a) the long C(1)–C(2) bonds of
1.378 and 1.408 Å, resulting from the partial single bond char-
acter of the double bond due to the contribution of valence
hybrid 9; (b) the wider angles around the double bond are N(1)–
C(2)–C(1) and C(2)–C(1)–C(15), i.e., between pairs of trans-
substituents and the double bond; (c) the double bond is
twisted, by 4.2Њ for 7a and by 32.6Њ for 7b; (d) the dihedral angle
of the Ph group and the C(3)–C(2)–N(1) plane is 24.9Њ for 7a
and 53Њ for 7b.
Crystallographic data for the substituted enamines formed from 5
with piperidine and aniline
The solid state structures of the aniline and the piperidine
enamines (7a and 7b) formed by the substitution of either (E)-5
or (Z)-5 in MeCN with aniline and piperidine, respectively,
1538
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545

View Article Online
Table 10 Selected bond lengths and angles for 7a
Bond
Length/Å
Bonds
Angle/Њ
C(1)–C(2)
C(1)–C(15)
C(2)–C(3)
C–C(Ar)
C–C(Ani)
O(1)–C(15)
O(2)–C(15)
O(2)–O(14)
N(2)–O
1.378(5)
1.487(6)
1.491(6)
1.355(8)–1.389(6)
1.368(7)–1.380(8)
1.187(5)
1.315(5)
1.447(6)
C(2)–C(1)–C(15)
N(1)–C(2)–C(1)
N(1)–C(2)–C(3)
C(1)–C(2)–C(3)
C(2)–C(3)–C(4)
C(2)–C(3)–C(8)
C–C–C(Ar)
N(1)–C(9)–C(10)
N(1)–C(9)–C(14)
C–C–C(Ani)
O(1)–C(15)–O(2)
O(1)–C(15)–C(1)
O(2)–C(15)–C(1)
C(15)–O(2)–C(16)
C(2)–N(1)–C(9)
123.8(4)
124.3(4)
119.2(4)
116.5(4)
119.7(4)
120.8(4)
119.5(4)–120.8(5)
118.6(4)
121.6(4)
119.5(4)–120.8(5)
123.1(4)
125.8(5)
111.1(4)
115.9(4)
128.5(4)
121.1(4)
118.5(4)–120.5(4)
122.7(4)
113.4(4)
175.76
1.243(4)–1.244(4)
1.330(5)
1.429(5)
N(1)–C(2)
N(1)–C(9)
O(3)–N(2)–O(4)
2N–O–C
N(2)–C(1)–C(2)
N(2)–C(1)–C(15)
N(2)–C(1)–C(15)–N(1)–C(2)–C(3)a
C(15)–C(1)–N(2)–O(3)–O(4)a
O(1)–O(2)–C(15)–C(1)–N(2)a
Ph–C(2)–N(1)a
178.27
80.21
114.87
48.57
Ani–C(2)–C(3)a
a Dihedral angle.
further with a second amine molecule. The difference between
an anionic and a neutral amine nucleophile is that the inter-
mediate in the multi-step route is the zwitterion 3a [eqn. (2a)],
rather than the carbanion 3b [eqn. (2a)]. Electron-withdrawal
by the ammonio moiety of 3a reduces the expulsion rate k2
of the nucleofuge, compared with the related value from the
corresponding anion 3b when the neutral amino group assists
the nucleofuge expulsion by the resonative electron-donation of
its non-bonded electron pairs.
Reactions of the ꢀ-nitro activated system
In the study of system 5 we intended to find out if increased
activation in a system which carries simultaneously the two
strong electron-withdrawing groups CO2Me and NO2 but also a
good iodo nucleofuge would show amine catalysis. β-Chloro- or
β-iodo-α-nitrostilbene4g or β-chloro-α-nitrostyrene10 do not
show amine catalysis on substitution by amines. We assume that
the activation is sufficient to give a multi-step route via zwitter-
ion 3a judged by (a) previous studies of system 5 with a thio
nucleophile,3i,j,m (b) the pKa(CH2YYЈ)Јs in 1 : 1 DMSO–H2O of
CH2(NO2)CO2Me (5.95)3m and CH2(CN)2 (10.21)3m and the
correlations between pKa(CH2YYЈ) and the equilibrium
constant for intermediate formation or the life-time of the
intermediate and (c) the mild catalysis observed for 4, X = Cl,
Br.4a,f However, higher electron withdrawal also means a higher
acidity of the ammonio hydrogen, and if the rate of this proton
expulsion is roughly correlated with its acidity, the k3B term
should also increase for system 5, unless k3B is diffusion
controlled and insensitive to substitution. Consequently, it is
difficult to predict how the k3B/k2 ratio will be affected for 5.
The basicity and nucleophilicity of the amine should also
affect rate constants k3B and k1 and, since the catalysis for
systems 4 was observed with weak anilino nucleophiles, we
used both weakly basic primary and secondary aniline bases,
i.e., aniline [pKa(H2O) = 4.6]11a and p-MeOC6H4NHMe [pKa-
(H2O) = 5.36],11b and more basic amines such as morpholine
and piperidine [pKa(H2O) = 8.33 and 11.12, respectively].11a
Whereas the latter amines gave convenient rates the former were
rather slow.
Fig. 5 An ORTEP drawing of 7b.
Discussion
The study of amine catalysis in the substitution of aromatic and
vinylic systems carrying a poor nucleofuge has been used to
investigate whether the substitution is a single-step or multi-
step process. For a first order reaction in the amine the kinetics
are of little help in answering the question since both the single
step and the multi-step routes are first order in the amine. How-
ever, if the reaction displays amine catalysis, i.e., a kinetic order
in the amine greater than one, it is clear that an intermediate
which is formed in the reaction with one amine molecule reacts
The lack of observed amine catalysis, regardless of the amine
used, could be due to two different reasons: either to the
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545
1539

View Article Online
Table 11 Selected bond lengths and angles for 7b
Bond
Length/Å
Bonds
Angle/Њ
O(1)–C(14)
O(2)–C(14)
O(2)–C(15)
O(3)–N(2)
O(4)–N(2)
N(1)–C(1)
N(1)–C(9)
N(1)–C(13)
4C–C(Pip)
N(2)–C(1)
C(1)–C(2)
C(1)–C(14)
C(2)–C(3)
C–C(Ar)
1.208(4)
1.350(4)
1.440(5)
1.232(4)
1.243(4)
1.331(4)
1.480(4)
1.466(4)
1.511(5)–1.518(5)
1.415(4)
1.408(4)
1.456(5)
1.495(4)
C(14)–O(2)–C(15)
C(2)–N(1)–C(9)
C(2)–N(1)–C(13)
C(9)–N(1)–C(13)
O(3)–N(2)–O(4)
O(3)–N(2)–C(1)
O(4)–N(2)–C(1)
N(2)–C(1)–C(2)
N(2)–C(1)–C(14)
C(2)–C(1)–C(14)
N(1)–C(2)–C(1)
N(1)–C(2)–C(3)
C(1)–C(2)–C(3)
C(2)–C(3)–C(4)
C(2)–C(3)–C(8)
C–C–C(Ar)
117.1(3)
123.8(3)
122.8(3)
113.2(3)
121.9(3)
119.7(3)
118.3(3)
120.9(3)
116.8(3)
122.2(3)
124.2(3)
117.8(3)
118.0(3)
119.8(3)
120.2(3)
119.6(3)–120.2(3)
109.6(3)–111.5(3)
110.5(3)–112.3(3)
121.7(3)
127.4(3)
110.9(3)
147.36
110.86
53.00
1.370(5)–1.384(5)
N–C–C(Pip)
C–C–C(Pip)
O(1)–C(14)–O(2)
O(1)–C(14)–C(1)
O(2)–C(14)–C(1)
N(2)–C(1)–C(14)–N(1)–C(2)–C(3)a
Ph–C(14)–C(1)–N(2)a
Ph–C(2)–N(1)a
a Dihedral angle.
inequality k2 ϩ k3B[Amine] >> kϪ1 [kobs = k1, eqn. (5)] or to
The ratios for the nucleophilic attack step (analogous to k1) in
protic media (H2O, DMSO–H2O) are relatively small (1.5–9.7),
with the value of 23.6 for α-cyano-4-nitrostilbene16 being an
exception. The values of the ratios in the aprotic MeCN (5.8)
and CHCl3 (12) for the addition to benzylidene Meldrum’s acid
are larger than those (1.54–2.4) in the protic media. The ratios
in EtOH resemble the values of the ratios (3.3–9.5) observed in
the substitution of related nitro-activated systems in both
EtOH and MeCN.4g Dissection of the overall kPip/kMor ratios to
their k1, kϪ1 and k2 components was achieved only in the single
case of the nitro-activated compound 12 in DMSO–H2O (1 : 1),
and were determined as 3.7, 0.028 and 55.3, respectively.3c
k
Ϫ1 > k2 >> k3B[Amine] [kobs = k1k2/kϪ1, eqn. (6)].
Three factors that should contribute to the different behavior
of systems 4 (X = Cl, Br) and 5 are a difference in the extent of
transition state imbalance, as observed for addition of amines
to benzylidenemalononitriles and to nitro-activated alkenes,12
and steric and hydrogen bonding effects. System 5 is more
crowded than system 4 and the same applies for the derived
zwitterions 10 and 11. This will make k1 smaller and kϪ1, k2 and
k3B higher for the reaction of 5 via 10 than for the reaction of 4
via 11. The lack of reactivity of 2,4,6-tri-tert-butylaniline with
(E)-5 is probably due to a very low k1/kϪ1 equilibrium combined
with the relatively low amine basicity.
Mor
Mor
Based on the relatively low k1Pip/k1
ratios and the k2Pip/k2
ratios which are only ca. 2.5 times lower than those found by us
for 10 we conclude that the relatively high observed rate coef-
ficients are composite, i.e., kobs = k1k2/kϪ1 and their ratio is given
by eqn. (16). Indeed, the kPip/kMor ratio for 12 calculated from
the right hand side of eqn. (16) is 7500. Although the nucleo-
fuge, solvent and the intermediate (3b for 12, 3a for 10) are
different, it is clear that a relatively high ratio is expected for a
composite kobs. Consequently, the rate of iodide expulsion from
10 becomes sufficiently slow to make it part of the rate deter-
mining step and the ratio reflects the combination of an
RRЈHNϩC(Ph)(I)–CϪ(NO2)CO2Me
10
RRЈNHϩC(Ar)(Br)–CϪ(CN)2
11
Hydrogen bonding of the ammonio hydrogen in 10 with
either the NO2 or the CO2Me group will give a six membered
ring whereas the hydrogen bond acceptor cyano nitrogens in
zwitterion 11 are too remote from the ammonio proton to form
an intramolecular hydrogen bond. This should increase k1 and
reduce k3B for 10 compared with 11 whereas k3B will be less
affected, thus decreasing the k3B/k2 ratio for 10. We note that the
(Z)-configuration of 7a resembles the X-ray determined (Z)-
Pip
Mor
Mor
increase in both the k1Pip/kϪ1 vs. k1Mor/kϪ1
(= K1Pip/K1
when K = k1/kϪ1) and the k2Pip/k2Mor terms.
PhC(OMe)᎐C(NO )Ph
᎐
2
configuration of solid MeNHCH᎐C(NO )CO Me,13 and that
12
᎐
2
2
the (Z)-configuration is mostly preferred over the (E)-
configuration in the solid,14 whereas both isomers prevail in
solution.14b The (Z)-configuration for 7b is reminiscent of that
in α-morpholino-β-nitrostilbene.9
Although the zwitterion 10 derived from weakly basic amines
such as aniline should be more acidic (and hence display larger
k3B) compared with those derived from piperidine and morpho-
line, catalysis was not observed even in reactions of the former
amines.
Pip
Mor
Pip
kPip/kMor = k1Pipk2
k
Mor/k1Mork2
k
(16)
Ϫ1
Ϫ1
In conclusion, the kPip/kMor ratio can be much higher when
kobs is a product of the rate coefficients for the single steps,
rather than when it is the rate coefficient for the first, nucleo-
philic attack step. A similar conclusion was reached for the
reactions of α-halo-β-nitrostilbenes (halogen = Cl, I) with
amines based on the kI/kCl relative reactivity ratios.4g
The much larger kPip/kMor ratios in MeCN (115–138) than in
EtOH (3.3–6.9) have a mechanistic significance. Many kPip/kMor
ratios which were determined for the addition of amines to
electrophilic alkenes are summarized in Bernasconi’s review.15
Based on this analysis, the lower ratios in EtOH reflect a rate
determining nucleophilic attack (k1). We ascribe this both to the
presence of hydrogen bonds between the solvent and the amines
which apparently reduce k1 more than they affect kϪ1 and k2,
1540
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545

View Article Online
and to an electrophilic solvent assistance to the expulsion of the
nucleofuge which increases k2. The overall result is that the kPip
/
kMor ratio is similar to the k1Pip/k1Mor ratio.
Hydrogen bonding and a change in the rate determining step
are also responsible for the solvent reactivity ratios kMeCN/kEtOH
which are more than an order of magnitude higher for piper-
idine (23.3–27.7) than for morpholine (0.79–1.16). The trends
of the values are reminiscent of those (Pip: 8.2–16.7; Mor: 2.4–
6.2) for the reaction of amines with (E)-PhC(Cl)᎐C(Ph)NO .4g
An a posteriori explanation is that EtOH ؒ ؒ ؒ HNR hydrogen
bonds (stronger for Pip) reduce the nucleophilicity whereas
intramolecular hydrogen formed in the transition state leading
to 10 (stronger for the morpholinium ion) which increase the
nucleophilicity account qualitatively for the results.
Electron-withdrawing substituents are expected to increase
the electrophilicity of Cα with a consequent increase in k1 and to
increase the acidity of the zwitterion, thus decreasing k3B and
increase the difficulty in expelling the nucleofuge, thus decreas-
ing k2. The result is an increase in the k3B/k2 ratio, until k2
becomes rate determining. The observed increase in kЉ values is
therefore ascribed to a combination of an increase in k3B and k1
and a decrease in the kϪ1 values.
᎐
2
The single kMeCN/kTHF ratio with aniline is 5.0 which
This analysis also suggests a regular increase of kЈ values on
increasing the electron-withdrawal by X. This was observed in
most cases and the few deviations are ascribed to a high error in
kЈ which is obtained as a small value from the intercept of the
kobs vs. [Amine] plot.
The Hammett correlations of log kЉ vs. σX in MeCN at 30
and 40 ЊC are linear. The slopes ρ are positive (1.06–1.14) and
relatively low. They are comparable to the values summarized
for many nucleophilic reactions with electrophilic alkenes.21
Since kЉ is a composite value, the linearity of the plots indicates
also a linearity in the Hammett plots of the logarithms of the
individual rate constants k1, k2 and k3B vs. σX. We noted above
that K = k1/kϪ1 increases for 13 with electron withdrawing sub-
stituents.17b,c,20 Similar behavior in our system together with a
parallel increase in k3B due to increased acidity of the ammonio
group of 3a will lead to a positive ρ.
In EtOH the ρ value for log kЉ vs. σX is lower (0.85) and the
linearity is poorer than in MeCN. This is mainly due to very
similar reaction rates of 6-OMe and 6-Me which lead to a
slightly positive deviation of 6-OMe from the plot. A similar
deviation was also observed in the log k1 vs. σX for the substitu-
tion of 6-X by HOCH2CH2SϪ in 1 : 1 DMSO–H2O,22 whereas
in the reaction of 13 the p-MeO derivative behaved normally.
The difference was ascribed by Bernasconi22 to a difference in
the ground state resonative stabilization by p-MeO in both
cases. In 13 the only stabilization is by the p-MeO group (cf.
the dipolar hybrid 14) whereas in 6-MeO the SMe plays this
role more effectively (cf. 15) so that the demand for resonance
resembles the ratio for the reaction of ArC(X)᎐C(CO Et) ,
᎐
2
2
X = the very good nucleofuges OTf, OMs; Ar = Ph, p-O2NC6H4
with piperidine and morpholine.18 This ratio for an amine
which forms stronger intramolecular hydrogen bonds in 10 than
piperidine or morpholine is ascribed to higher stabilization of
the transition state leading to 10 in the more polar solvent.
The higher reactivity of the (Z)-isomer compared with the
(E)-isomer reflects a lower crowding in (Z)-5 than in (E)-5
which enables a less hindered approach of the nucleophile
to the double bond. This is corroborated by the 5–10 times
higher ratio for the most hindered nucleophile studied, i.e.,
p-MeOC6H4NHMe where the combined higher steric conges-
tion in the transition state make (Z)-5 much more reactive than
(E)-5 compared with less bulky amines. Steric effects in the
nucleophile were reflected in the substitution of 12 by the
nucleophiles MeONH2 and MeONHMe which have similar
pKa’s but differ in the steric bulk around the nitrogen. The
reactivity of the latter is reduced significantly, mainly due to a
decrease in kϪ1 by two orders of magnitude.3g
Reactions of 6-X
Since the highly reactive 6-X system had not hitherto been stud-
ied with amines, variables in this reaction were investigated. It is
not surprising that in contrast with 5, a strong amine catalysis
was observed for all systems 6-X under a variety of experi-
mental conditions since, in contrast with IϪ, MeSϪ is regarded
at most as a moderate nucleofuge judged by the “rank” (nucleo-
fugality) of PhSϪ.19 Consequently, high k3B/k2 ratios with
piperidine indicate a strong contribution of the catalytic route,
which reach a maximum when k2 is negligible and the reaction
order in the amine is two.
It should be mentioned that an alternative to the last step
of the substitution ought also to be considered. This is the
specific base general acid catalyzed route where following a fast
deprotonation of 3a by the amine, the ammonium ion formed
electrophilically catalyzed the nucleofuge expulsion in a slow
step. This route was suggested for the expulsion of the very
poor nucleofuge CNϪ4d but was excluded for a better nucleo-
fuge like FϪ by the observation of an added base (amine) rather
than ammonium ion catalysis for 4, X = F.4a Hence, only eqn.
(3) will be discussed here.
The change in the substituents of 6-X covers the whole range
from the resonative electron-donating MeO to the strong elec-
tron withdrawing two m-CF3 groups. This change is accom-
panied by a change in the reaction order in both MeCN and
EtOH from a second order in the amine for the most electron
donating substituent to an order between first and second in the
amine as the aryl group becomes more electron withdrawing.
Electron withdrawal increases the rate of nucleophilic attack
(i.e., of k1) in nucleophilic reactions on electrophilic alkenes. A
relevant reaction is the nucleophilic addition reaction to
13,17b,c,20 the analog of 6 carrying an α-H rather than α-SMe: k1
increases and kϪ1 decreases with the increased σX, and hence
K = k1/kϪ1 also increases with σX.
contribution by p-MeO decreases. These effects also operate in
our case, but since the effect is small and the rate constant is
composite a detailed analysis is unwarranted.
An important mechanistic point is that in EtOH all the reac-
tions, except that of 6-MeO are between first and second order
in the amine, whereas in MeCN only the reactions of 6-X,
where X = EWG (p-Br, p-CF3, m,mЈ-(CF3)2), have a mixed
order, and those reactions for X = H and electron-donating
substituents are second order in the amine. This is ascribed to
the higher basicity of EtOH which introduces an additional
deprotonation step of the zwitterion with EtOH, acting as a
base. This effect is diminished for 6-MeO since electron-
donation by MeO reduces the acidity of the zwitterion, and the
deprotonation is then carried out exclusively by the more basic
amine. We note that EtOH, in contrast with MeCN, can
increase k2 by electrophilic catalysis via hydrogen bonding to
the MeSϪ expulsion and that lower k3B/k2 ratios in MeCN com-
pared with alcohols were previously observed.4a
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545
1541

View Article Online
Another interesting phenomenon in EtOH is that the only
reaction which does not give a linear kobs vs. [Amine] plot is
that of the most reactive substrate 6-(CF3)2. Since the inverted
1/kobs vs. 1/[Amine] plot according to eqn. (7) is linear,
regardless of whether the rate coefficient for the reaction is k1,
k1k3B/kϪ1 or k1k2/kϪ1
.
Experimental
k
k
Ϫ1 ∼ k2 ϩ k3B[Amine] in contrast with the other cases where
Ϫ1 >> k2 ϩ k3B[Amine]. In 6-(CF3)2 the acidity of the
General
ammonio group is the highest of all our systems. This increases
k3B[Amine] (unless k3b is diffusion controlled) and in parallel
decreases both kϪ1 and k2 until, in spite of the lower basicity of
the deprotonating base the condition above is apparently ful-
filled. The value of k1 was calculated from eqn. (15) but we have
no value in a related system for comparison.
As for 5, the reactions in MeCN are faster than those in
EtOH, but only by a moderate factor (kMeCN/kEtOH (30 ЊC) =
3.0–4.9). Again the explanation is reduced reactivity of the
piperidine by hydrogen bonding to the EtOH.
Melting points were determined with a Thomas-Hoover appar-
atus and are uncorrected. 1H and 13C NMR spectra were
measured with Bruker DRX-400 and AMX-300 spectrometers.
IR spectra were recorded with a Nicolet Impact 400 spec-
trometer. X-Ray diffraction was conducted with a Philips PW
100 diffractometer. The UV spectra were obtained and the
kinetic study conducted using a Contron UNIKON 930
spectrophotometer.
The activation parameters for the catalyzed process (Table 8)
are characterized by negative activation enthalpies (and ener-
gies) and high negative activation entropies. Negative or very
small activation enthalpies and very highly negative activation
entropies were previously observed in similar vinylic substi-
tutions of poor leaving groups by amines.4a,b,d–f,23a Whereas the
negative activation energies seem unusual, the accumulating
evidence and examples indicate that such behavior is common
rather than exceptional for this type of reaction in aprotic
media. Examples are the displacement of CF3CH2OϪ from
compound 4, X = CF3CH2O, by amines in MeCN (∆H‡ = 11–
Ϫ16 kcal molϪ1 for the catalyzed reaction), the displacement of
EtOϪ from 4, X = EtOϪ, with piperidine in MeCN (∆H‡ = 0.6
kcal molϪ1 for the catalyzed reaction), of FϪ by morpholine
from 4, X = F (∆H‡ = Ϫ3.2 kcal molϪ1),4g or the displacement
of CNϪ from tricyanovinyl chloride by aliphatic amines in
CHCl3 (∆H‡ = 1.4–2.6 kcal molϪ1).23a Other examples, which
include addition of amines to electrophilic alkenes, show
similar low ∆H‡ (2–2.4 kcal molϪ1) and high negative ∆S‡
values23b,c and are collected in ref. 4g.
(E )-Methyl ꢀ-nitro-ꢁ-anilinocinnamate (7a)
A mixture of (Z)-5 (0.2 g, 0.6 mmol) and aniline (0.27 ml, 3.0
mmol) in MeCN (11 ml) was stirred for 3 days at room temper-
ature under argon. The solvent was evaporated and the residue
was recrystallized from EtOH giving 7a as yellow needles (0.1 g,
0.33 mmol, 56%), mp 168 ЊC. The spectral data are given in
Table 12.
Anal. Calcd. for C16H14N2O4: C, 64.42; H, 4.73; N, 9.39.
Found: C, 64.21; H, 4.88; N, 9.19%.
Crystallographic data: space group P21/c; a = 11.452(2),
b = 8.899(1), c = 15.330(3) Å, V = 1502.2(6) Å3, Z = 4,
ρcalc = 1.32 g cmϪ3; µ(Mo-Kα) = 0.90 cmϪ1; no. of unique reflec-
tions 2255, no. of reflections with I ≥ 2σI 1380, R = 0.062,
RW = 0.070. CCDC reference number 168692.
(E )-Methyl ꢀ-nitro-ꢁ-piperidinocinnamate (7b)
A CaCl2-protected solution containing (E)-5 (0.12 g, 0.36
mmol) and piperidine (0.1 ml, 1.1 mmol) in MeCN (7 ml) was
stirred at room temperature for 24 h. The solvent was evapor-
ated and the remainder was recrystallized from EtOH, giving
yellow crystals of 7b (47 mg, 0.16 mmol, 45%), mp 130 ЊC. The
spectral data are given in Table 12.
The low and especially the negative ∆H‡ values serve as
strong evidence against a single-step substitution but they fit a
multi-step reaction where ∆H‡ and ∆S‡ are composite, and are
the sums of the corresponding values for the individual steps
[eqn. (17) and (18)]. In eqn. (17) and (18) ∆HЊ and ∆SЊ are the
Anal. Calcd. for C15H18N2O4: C, 62.06; H, 6.25; N, 9.65.
Found: C, 61.77; H, 5.98; N, 9.66%.
Crystallographic data: space group P21/c; a = 8.344(2),
b = 10.309(2), c = 17.769(2) Å, V = 1496.2 Å3, Z = 4, ρcalc = 1.29
g cmϪ3; µ(Mo-Kα) = 0.88 cmϪ1; no. of unique reflections 2821,
no. of reflections with I ≥ 3σI 1789, R = 0.050, RW = 0.077.
CCDC reference number 168693.
‡
∆H‡ = ∆H1‡ Ϫ ∆HϪ1‡ ϩ ∆H3B‡ (or ∆H2 ) =
∆HЊ ϩ ∆H3B‡ (or ∆H2 ) (17)
‡
‡
∆S‡ = ∆S1‡ Ϫ ∆SϪ1‡ ϩ ∆S3B‡ (or ∆S2 ) =
∆SЊ ϩ ∆S3B‡ (or ∆S2 ) (18)
‡
(E )-Methyl ꢀ-nitro-ꢁ-morpholinocinnamate (7c)
corresponding enthalpies for the equilibrium of the first nucleo-
A solution containing (E)-5 (0.15 g, 0.45 mmol) and morph-
oline (0.12 ml, 1.35 mmol) in MeCN (10 ml) was stirred under
CaCl2 at room temperature for 165 min. The solvent was evap-
orated and the remainder was washed with water and recrys-
tallized from ethanol giving 7c (0.1 g, 0.34 mmol, 76%), mp
188 ЊC. The spectral data are given in Table 12.
‡
philic attack step, ∆H3B‡, ∆S3B are the terms for the cata-
‡
‡
lyzed process and ∆H2 , ∆S2 are those for the uncatalyzed
process.The low ∆H‡ terms arise mainly from the ∆HЊ terms.
This conclusion is based on reactions where no leaving group is
expelled, especially on the ∆HЊ values of Ϫ13–Ϫ21 kcal molϪ1
(∆SЊ = Ϫ27–Ϫ51 e.u.) for the nucleophilic addition of tri-n-
Anal. Calcd. for C14H16N2O5: C, 57.53; H, 5.52; N, 9.53.
Found: C, 57.37; H, 5.59; N, 9.47%.
butylphosphine to ArCH᎐C(CN) which gives the zwitterion
᎐
2
16.24
(E )-Methyl ꢀ-nitro-ꢁ-N-methyl-p-anilinocinnamate (7d)
ArCH(PϩBu3)–CϪ(CN)2
A CaCl2-protected solution containing (Z)-5 (0.1 g, 0.3 mmol)
and p-methoxy-N-methylaniline (0.08 g, 0.6 mmol) in MeCN
(10 ml) was stirred at room temperature for 22 h and then
poured into water and extracted with ether (3 × 20 ml). The
combined organic fractions were dried (CaCl2) and the solvent
was evaporated. The mixture was separated on a preparative
TLC plate using 3 : 7 ether–petroleum ether (40–60 ЊC) as the
eluent. The second fraction was recrystallized from EtOH
giving 7d (22 mg, 0.064 mmol, 21%), mp 100 ЊC. The spectral
data are given in Table 12.
16
If similar ∆HЊ values apply in our system, the addition of a
not too high ∆H3B or ∆H2 term will give an overall negative
‡
‡
∆H‡.
The high negative entropies of activation are ascribed to the
assembly of two or three molecules in the rate determining step
of the substitution reaction, coupled with the formation of a
zwitterion intermediate. The transition state, being much more
organized than the reactants, leads to a high negative ∆S‡
1542
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545

Table 12 Spectral data for 7a–d and 8-X
Cmpd.
UV (MeCN) λmax/nm (ε)a 1H NMR δ(CDCl3)/ppm
13C NMR δ(CDCl3)/ppm
MS m/z (relative %, assignment)
7a
361 (5800)
3.48 (3H, s, MeO), 6.74–7.38 (10H, m, Ani, Ph), 52.69 (OMe), 120.82, 124.59, 125.87, 126.72, 128.30,
298 (18, M), 252 (33, M Ϫ NO2), 251 (19, M Ϫ HNO2), 220
10.81 (NH)
128.65, 128.77, 129.12, 129.43, 130.16, 130.74 (C᎐C, Ani, (100, M Ϫ NO2 Ϫ MeOH), 193 (43, M Ϫ NO2 Ϫ CO2Me),
᎐
Ph)b
180
M Ϫ NO2 Ϫ CO2Me Ϫ C2H4), 105 (32, [PhNHCH]ϩ)
290 (4, M), 244 (100, M Ϫ NO2), 212 (25, M Ϫ
(43,
M Ϫ NO2 Ϫ CO2Me Ϫ CH),
165
(18,
7b
7c
7d
274 (20900), 388 (15900)
275 (11560), 386 (8300)
401 (10900)
1.79 (6H, s, (CH2)3), 3.28 (4H, s, N(CH2)2), 3.51 (s,
3H, OMe), 7.39–7.55 (5H, m, Ph)
23.34 (CH2CH2CH2), 26.70 (N(CH2)2), 51.70 (CH2N),
53.38 (OMe), 120.72, 128.99, 129.45, 131.87, 134.23, 163.97 NO2 Ϫ MeOH), 184 (6, M Ϫ CO2Me Ϫ HNO2), 117 (15,
ϩ
(Ph, 2C᎐C), 166.90 (C᎐O)
C(NO2)CO2Me), 105 (14, PhCNH2
)
᎐
᎐
3.34 (4H, t, N(CH2)2), 3.54 (3H, s, Me), 3.87 (4H, t,
51.99, 52.02 (CH2N, OCH3), 66.80 (OCH2), 121.89, 129.24, 292 (100, M), 262 (2, M Ϫ OCH2), 206 (5, M Ϫ C4H8NO),
O(CH2)2), 7.48–7.58 (5H, m, Ph)
129.66, 132.12, 133.28, 163.62 (Ph, 2C᎐C), 165.04 (C᎐O)
191 (15, M Ϫ C4H8NO Ϫ Me), 171 (55, M Ϫ C4H8O Ϫ
OMe)
342 (11, M), 296 (74, M Ϫ NO2), 264 (100,
᎐
᎐
3.39 (3H, s, NMe), 3.49 (s, 3H, COOMe), 3.72 (s, 3H,
C6H4OMe), 6.90 (4H, q, Ar, J = 8.0 Hz), 7.26–7.40
(m, 5H, Ph)
45.11 (NMe), 51.98 (COOMe), 55.41 (C6H4OMe), 114.47,
126.83, 128.64, 130.42, 131.18, 134.23, 138.12, 158.20 (C᎐C, M Ϫ NO2 Ϫ MeOH),
249
(7,
M Ϫ Me Ϫ OMe Ϫ
᎐
An, Ph), 163.66 (C᎐O)
NO2 Ϫ H), 237 (56, M Ϫ CO2Me Ϫ NO2), 222 (7,
᎐
M Ϫ CO2Me Ϫ NO2 Ϫ Me),
210
(15),
118
(80,
[PhCNMe]ϩ), 117 (19, [C6H4N(Me)C]ϩ), 105 (42,
[C6H4NMe]ϩ)
8-MeO 311 (9500)
1.71 (6H, s, 2Me), 1.83–1.84 (6H, m, (CH2)3), 3.52
(2H, t, J = 5.8 Hz, CH2N), 3.78 (2H, t, J = 5.6 Hz,
CH2N), 3.86 (3H, s, OMe), 7.18 (4H, q, Ar, J = 6.9
Hz)
345 (100, M), 277 (4, M Ϫ C5H8)
8-Me
8-H
335 (9800)
338 (9300)
1.71 (6H, s, 2Me), 1.82–1.86 (6H, m, (CH2)3), 2.42
(3H, s, Me), 3.48 (2H, t, J = 4.8 Hz, CH2N), 3.71
(2H, t, J = 4.6 Hz, CH2N), 7.27 (4H, q, Ar, J = 8.0
Hz)
1.71 (6H, s, 2Me), 1.81–1.85 (6H, m, (CH2)3), 3.46
(2H, t, J = 5.6 Hz, CH2N), 3.73 (2H, t, J = 5.7 Hz,
CH2N), 7.52 (5H, m, Ph)
329 (34, M), 271 (25, M Ϫ OCMe2), 243 (7,
M Ϫ OCMe2CO), 227 (100, M Ϫ COOCMe2O), 198 (34),
170 (22)
315 (100, M), 257 (38, M Ϫ OCMe2), 228 (26,
M Ϫ OCMe2 Ϫ CO Ϫ H), 213 (75, M Ϫ OCMe2 Ϫ CO2),
184 (73, M Ϫ CMe2 Ϫ 2CO2 Ϫ H), 156 (29, M Ϫ CMe2 Ϫ
2CO2 Ϫ H Ϫ MeCH2), 129 (31, M Ϫ CMe2 Ϫ 2CO2 Ϫ
4CH2), 102 (35, [PhC2H]ϩ)
8-Br
341 (9200)
338 (3700)
1.70 (6H, s, 2Me), 1.89–1.91 (6H, m, (CH2)3), 3.48
(2H, t, J = 5.6 Hz, CH2N), 3.68 (2H, t, J = 5.7 Hz,
CH2N), 7.40 (4H, q, Ar, J = 9.0 Hz)
395, 393 (84, 77, M), 337, 335 (59, 51, M Ϫ OCMe2), 293,
291 (100, 99, M Ϫ OCMe2 Ϫ CO2), 264, 262 (47, 40,
M Ϫ OCMe2 Ϫ CO Ϫ H),
209,
207
(24,
23,
M Ϫ CMe2 Ϫ 2CO2 Ϫ 4CH2), 182, 180 (42, 36, M Ϫ
CMe2 Ϫ 2CO2 Ϫ 5CH2 Ϫ CH), 162 (12, M Ϫ Br Ϫ
NC5H10 Ϫ CMe2 Ϫ 2CH), 128 (16, [C6H4C2NCH2]ϩ)
383 (13, M), 366 (8, M Ϫ OH), 325 (20, M Ϫ OCMe2), 299
(21, M Ϫ Me Ϫ CF3), 282 (47, M Ϫ CF3 Ϫ 2CH2), 254
(100, M Ϫ NC5H10 Ϫ CO2H), 253 (16, M Ϫ CMe2 Ϫ
2CO2), 197 (27, M Ϫ CMe2 Ϫ 2CO2 Ϫ 4CH2)
8-CF3
1.72 (6H, s, 2Me), 1.83–1.89 (6H, m, (CH2)3), 3.36
(2H, t, J = 5.4 Hz, CH2N), 3.67 (2H, t, J = 5.6 Hz),
7.45 (4H, q, Ar, J = 8.5 Hz)
8-(CF3)2 343 (8700)
1.71 (6H, s, 2Me), 1.82–1.89 (6H, m, (CH2)3), 3.35
(2H, t, J = 5.4 Hz, CH2N), 3.70 (2H, t, J = 5.8 Hz,
CH2N), 7.85 (1H, s, Ar), 8.03 (2H, s, Ar)
451 (10, M), 422 (19, M Ϫ Me Ϫ CH2), 394 (100,
MH Ϫ OCMe2), 374 (27, M Ϫ C(O)Me2 Ϫ F), 349 (9,
M Ϫ C(O)Me2 Ϫ CO2), 326 (16, M Ϫ CMe2 Ϫ C5H10 Ϫ
CH), 243 (12)
a Measured in dm3 molϪ1 cmϪ1 b Too weak to be observed.
.

View Article Online
Anal. Calcd. for C18H18N2O5: C, 63.15; H, 5.30; N, 8.18.
Found: C, 63.08; H, 5.33; N, 8.01%.
IR (Nujol) 834, 1089, 1294, 1493, 1712, 1742 cmϪ1. Anal. Calcd
for C15H13O4F3S: C, 52.02; H, 3.78. Found: C, 52.28; H, 4.02%.
Synthesis of 6-X
2,2-Dimethyl-5-[piperidino(3,5-trifluoromethylphenyl)-
methylene]-1,3-dioxane-4,6-dione (8-(CF3)2)
Compound 6-H was prepared following literature methods.6
The other derivatives were prepared by a similar procedure and
purified by chromatography or recrystallization from EtOAc–
petroleum ether. The detailed procedure is given for 6-(CF3)2
followed by specific data for the other derivatives.
A CaCl2-protected solution containing 6-(CF3)2 (0.14 g, 0.34
mmol) and piperidine (0.08 ml, 0.84 mmol) in MeCN (5 ml) was
stirred for 28 h. The solvent was evaporated and the remainder
was recrystallized twice from ethanol, giving 8-(CF3)2 (70 mg,
0.15 mmol, 46%), mp 184–185 ЊC. The spectral data are in
Table 12. Anal. Calcd. for C20H19NF6O4: C, 53.22; H, 4.24; N,
3.10. Found: C, 53.26; H, 4.28; N, 3.14%.
2,2-Dimethyl-5-[3,5-bis(trifluoromethyl)-ꢀ-
thiomethoxybenzylidene]-1,3-dioxane-4,6-dione [6-(CF3)2]
A solution of the Grignard reagent prepared from 3,5-
bis(trifluoromethyl)bromobenzene (0.41 g, 1.4 mmol) and Mg
turnings (35 mg, 1.4 mmol) in dry THF (2.0 ml) was added
dropwise to a solution of isopropylidene bis(methylthio)meth-
ylenemalonate (100 mg, 0.4 mmol)25 in dry THF (3.0 ml), and
stirred at room temperature for 2 h. A solution of 5% aq. HCl
(2 ml) was added dropwise to the mixture, which was then
extracted with dichloromethane (2 × 5 ml), washed with water
(3 × 10 ml) and dried (Na2SO4). After removal of the solvent,
the crude product was purified by silica gel chromatography
using 1 : 4 EtOAc–petroleum ether as an eluent to give 144 mg
2,2-Dimethyl-5-[piperidino( p-methoxyphenyl)methylene-1,3-
dioxane-4,6-dione (8-OMe)
A CaCl2-protected solution containing 6-OMe (30 mg, 0.097
mmol) and piperidine (0.024 ml, 0.25 mmol) was stirred at
room temperature for 48 h. The solvent was evaporated and
recrystallization from EtOH gave 8-OMe (27 mg, 0.078 mmol,
81%), mp 204 ЊC. The spectral data are given in Table 12. Anal.
Calcd. for C19H23NO5: C, 66.07; H, 6.71; N, 4.06. Found: C,
65.89; H, 6.67; N, 3.90%.
1
(86.3%) of a white solid mp 151–152 ЊC; H NMR (CDCl3) δ:
2,2-Dimethyl-5-[piperidino( p-bromophenyl)[or phenyl, p-tolyl,
p-trifluoromethylphenyl]methylene]-1,3-dioxane-4,6-dione
(8-Br or 8-H, 8-Me, 8-CF3)
1.78 (s, 6H), 1.87 (s, 3H), 7.53 (s, 2H), 7.96 (s, 1H); IR (Nujol)
985, 1291, 1376, 1719, 1746 cmϪ1. Anal. Calcd for C16H12O4F6S:
C, 46.38; H, 2.92. Found: C, 46.33; H, 3.10%.
A CaCl2-protected solution containing piperidine (0.01 ml, 0.1
mmol) and 6-Br (4.5 mg, 0.013 mmol) or 6-H (2.5 mg, 0.009
mmol) or 6-Me (5 mg, 0.015 mmol) or 6-CF3 (5 mg, 0.013
mmol) in MeCN (5 ml) was stirred at room temperature for 24
h and the solvent was then evaporated giving 8-Br, 8-H, 8-Me
and 8-CF3, respectively as determined by HRMS. 8-Br,
Calcd. for C18H2079BrNO4: 393.0576. Found: 393.0583. 8-H,
Calcd. for C18H21NO4: 315.1470. Found: 315.1451. 8-Me,
Calcd. for C18H23NO4: 329.1627. Found: 329.1649. 8-CF3,
Calcd. for C19H20NF3O4: 383.1344. Found: 383.1383. The
spectral data of the four compounds are given in Table 12.
5-( p-Methoxy-ꢀ-thiomethoxybenzylidene)-2,2-dimethyl-1,3-
dioxane-4,6-dione (6-MeO)
Starting with 0.45 g of isopropylidene bis(methylthio)methyl-
enemalonate (1.8 mmol), 0.25 g (44.7%) of a yellow solid, mp
1
171–173 ЊC was obtained. H NMR (CDCl3) δ: 1.78 (s, 6H),
1.95 (s, 3H), 3.86 (s, 3H), 7.00 (m, 4H); IR (Nujol) 834, 1025,
1295, 1493, 1611, 1723, 1749 cmϪ1; MS (CI) m/z (%) 308 (MϪ,
100), 293 (9.4). Anal. Calcd for C15H16O5S: C, 58.43; H, 5.23; S,
10.40. Found: C, 58.56; H, 5.24; S, 9.96%.
5-( p-Methyl-ꢀ-thiomethoxybenzylidene)-2,2-dimethyl-1,3-
dioxane-4,6-dione (6-Me)
Acknowledgements
Starting with 0.45 g of isopropylidene bis(methylthio)methyl-
enemalonate (1.8 mmol), 0.41 g (77.4%) of a very slightly
yellow–white solid, mp 169–171 ЊC was obtained. 1H NMR
(CDCl3) δ: 1.77 (s, 6H), 1.91 (s, 3H), 2.41 (s, 3H), 6.95–6.97 (m,
2H), 7.28–7.30 (m, 2H); IR (Nujol) 670, 815, 1012, 1209, 1295,
1460, 1723, 1749 cmϪ1; MS (CI) m/z (%) 292 (MϪ, 100.0), 277
(10.4). Anal. Calcd for C15H16O4S: C, 61.62; H, 5.52; S, 10.97.
Found: C, 61.55; H, 5.44; S, 10.86%.
We are indebted to Dr Shmuel Cohen for the crystallo-
graphic determination and to Professor C. F. Bernasconi for
discussions. This research was supported by the Israel–USA
Binational Science Foundation (BSF).
References
1 For reviews see (a) Z. Rappoport, Adv. Phys. Org. Chem., 1969, 7, 1;
(b) G. Modena, Acc. Chem. Res., 1971, 4, 73; (c) S. I. Miller,
Tetrahedron, 1977, 33, 1211; (d ) Z. Rappoport, Acc. Chem. Res.,
1981, 14, 7; (e) Z. Rappoport, Recl. Trav. Chim. Pays-Bas, 1985, 104,
3092; ( f ) B. A. Shainyan, Usp. Chim., 1986, 55, 942; (g) Z.
Rappoport, Acc. Chem. Res., 1992, 25, 474.
2 (a) Y. Apeloig and Z. Rappoport, J. Am. Chem. Soc., 1989, 101,
5095; (b) D. Cohen, R. Bar and S. Shaik, J. Am. Chem. Soc., 1986,
108, 231; (c) M. N. Glukhovtsev, A. Pross and L. Radom, J. Am.
Chem. Soc., 1994, 116, 5961; (d ) C. K. Kim, K. H. Hyun, C. K. Kim
and I. Lee, J. Am. Chem. Soc., 2000, 122, 2294.
3 (a) C. F. Bernasconi, R. B. Killion, Jr., J. Fassberg and Z.
Rappoport, J. Am. Chem. Soc., 1989, 111, 6862; (b) C. F.
Bernasconi, J. Fassberg, R. B. Killion, Jr. and Z. Rappoport, J. Am.
Chem. Soc., 1990, 112, 3169; (c) C. F. Bernasconi, J. Fassberg, R. B.
Killion, Jr. and Z. Rappoport, J. Org. Chem., 1990, 55, 4568; (d )
C. F. Bernasconi, J. Fassberg, R. B. Killion, Jr. and Z. Rappoport,
J. Org. Chem., 1991, 113, 4937; (e) C. F. Bernasconi, A. E. Leyes,
Z. Rappoport and I. Eventova, J. Am. Chem. Soc., 1993, 115, 7513;
( f ) C. F. Bernasconi, D. F. Schuck, R. D. Ketner, M. Weiss and
Z. Rappoport, J. Am. Chem. Soc., 1994, 116, 11764; (g) C. F.
Bernasconi, A. E. Leyes, Z. Rappoport and I. Eventova, J. Am.
Chem. Soc., 1995, 117, 1703; (h) C. F. Bernasconi, D. F. Schuck, R. J.
5-( p-Bromo-ꢀ-thiomethoxybenzylidene)-2,2-dimethyl-1,3-
dioxane-4,6-dione (6-Br)
Starting with 0.45 g of isopropylidene bis(methylthio)methyl-
enemalonate (1.8 mmol), 0.32 g (39.0%) of a very slightly
yellow–white solid, mp 168–170 ЊC was obtained. 1H NMR
(CDCl3) δ: 1.76 (s, 6H), 1.92 (s, 3H), 6.95–6.97 (m, 2H), 7.62–
7.64 (m, 2H); IR (Nujol) 802, 894, 1209, 1282, 1497, 1519, 1703,
1743 cmϪ1; MS (CI) m/z (%) 358 (MϪ ϩ 2, 100), 356 (MϪ, 90.6),
343 (34.4), 341 (25.5), 278 (2.8). Anal. Calcd for C14H13O4BrS:
C, 47.07; H, 3.67; Br, 22.37; S, 8.98. Found: C, 47.00; H, 3.77;
Br, 22.70; S, 8.81%.
5-( p-Trifluoromethyl-ꢀ-thiomethoxybenzylidene)-2,2-dimethyl-
1,3-dioxane-4,6-dione (6-CF3)
Starting with 100 mg of isopropylidene bis(methylthio)methyl-
enemalonate (0.4 mmol), 46.0 mg (33.0%) of a white solid, mp
1
162–164 ЊC was obtained. H NMR (CDCl3) δ: 1.77 (s, 6H),
1.88 (s, 3H), 7.21 (d, J = 8.10 Hz, 2H), 7.76 (d, J = 8.10 Hz, 2H);
1544
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545

View Article Online
12 (a) C. F. Bernasconi and R. A. Renfrow, J. Org. Chem., 1987, 52,
3035; (b) C. F. Bernasconi and R. B. Killion, Jr., J. Org. Chem., 1989,
54, 2878; (c) C. F. Bernasconi, Acc. Chem. Res., 1992, 25, 9.
13 V. G. Adrianov, Y. T. Struchkov and K. K. Babievskii, Cryst. Struct.
Commun., 1981, 11, 35.
14 (a) O. S. Wolfbeis, Chem. Ber., 1977, 110, 2480; (b) J.-L. Chiara, A.
Gomez-Sanchez, F. J. Hidalgo and J. Bellanto, J. Chem. Soc., Perkin
Trans. 2, 1988, 1691.
Ketner, I. Eventova and Z. Rappoport, J. Am. Chem. Soc., 1995,
117, 2719; (i) C. F. Bernasconi, R. J. Ketner, X. Chen and Z.
Rappoport, J. Am. Chem. Soc., 1998, 120, 7461; (j) C. F.
Bernasconi, R. J. Ketner, X. Chen and Z. Rappoport, Can. J. Chem.,
1999, 77, 584; (k) C. F. Bernasconi, A. E. Leyes and Z. Rappoport,
J. Org. Chem., 1999, 64, 2897; (l ) C. F. Bernasconi, A. E. Leyes,
I. Eventova and Z. Rappoport, J. Org. Chem., 1999, 64, 8829; (m)
C. F. Bernasconi, R. D. Ketner, M. L. Ragains, X. Chen and
Z. Rappoport, J. Am. Chem. Soc., 2001, 123, 2155.
15 C. F. Bernasconi, Tetrahedron, 1989, 45, 4017.
4 (a) Z. Rappoport and R. Ta-Shma, J. Chem. Soc (B), 1971, 871;
Z. Rappoport and R. Ta-Shma, J. Chem. Soc (B), 1971, 1461; (b)
Z. Rappoport and N. Ronen, J. Chem. Soc., Perkin Trans. 2, 1972,
955; (c) Z. Rappoport and P. Peled, J. Chem. Soc., Perkin Trans. 2,
1973, 616; (d ) Z. Rappoport and D. Ladkani, J. Chem. Soc., Perkin
Trans. 2, 1973, 1045; (e) Z. Rappoport and A. Topol, J. Chem. Soc.,
Perkin Trans. 2, 1975, 863; ( f ) Z. Rappoport and P. Peled, J. Am.
Chem. Soc., 1979, 101, 2682; (g) Z. Rappoport and A. Topol, J. Org.
Chem., 1989, 54, 5967.
16 C. F. Bernasconi, C. J. Murray, J. P. Fox and D. J. Carre, J. Am.
Chem. Soc., 1983, 105, 4349.
17 (a) C. F. Bernasconi and C. J. Murray, J. Am. Chem. Soc., 1986, 108,
5251; (b) C. F. Bernasconi and M. Panda, J. Org. Chem., 1987, 52,
3042; (c) B. Schreiber, H. Martinek, P. Wolschann and P. Schuster,
J. Am. Chem. Soc., 1979, 101, 4708.
18 E. Z. Schottland and Z. Rappoport, J. Org. Chem., 1996, 61, 8543.
19 (a) D. R. Marshall, P. J. Thomas and C. J. M. Stirling, J. Chem. Soc.,
Perkin Trans. 2, 1977, 1898; D. R. Marshall, P. J. Thomas and
C. J. M. Stirling, J. Chem. Soc., Perkin Trans. 2, 1977, 1914; (b) P. G.
Thomas and C. J. M. Stirling, J. Chem. Soc., Perkin Trans. 2, 1977,
1909; P. G. Thomas and C. J. M. Stirling, J. Chem. Soc., Perkin
Trans. 2, 1998, 1130; (c) C. J. M. Stirling, Acc. Chem. Res., 1979, 12,
198.
5 I. Eventova and Z. Rappoport, unpublished results.
6 X. Huang and B.-C. Chen, Synthesis, 1987, 480.
7 Z. Rappoport, J. Chem. Soc., Perkin Trans. 2, 1977, 1000.
8 I. Allade, P. Dubois, P. Levillain and P. Viel, Bull. Soc. Chim. France,
1983, 14, 83.
9 K. P. Park and J. H. Jeong, Acta Crystallogr., Sect. C, 1990, 46, 1659;
Z. V. Todres, A. Y. Kosnikov, K. I. Dyusergaliev, D. S. Ermekov, S. V.
Lindman and Y.-T. Struchkov, Izv. Akad. Nauk USSR, Ser. Khim.,
1990, 791.
20 C. F. Bernasconi and S. Fornarini, J. Am. Chem. Soc., 1980, 102,
5329.
21 Z. Rappoport and D. Ladkani, Chem. Scr., 1974, 5, 124.
22 C. F. Bernasconi, personal communication.
10 Z. Rappoport and S. Hoz, J. Chem. Soc., Perkin Trans. 2, 1975, 272.
11 (a) Z. Rappoport, Handbook of Tables for Organic Compounds
Identification, 3rd Edn., CRC Press, Boca Raton, Florida, 1985,
p. 436; (b) Dictionary of Organic Compounds, ed. J. I. G. Cadogan,
S. L. Ley and G. Pattenden, Chapmann and Hall, 6th Edn., Vol. 5,
1985, p. 4190.
23 (a) Z. Rappoport, P. Greenzaid and A. Horowitz, J. Chem. Soc.,
1964, 1334; (b) F. M. Menger and J. H. Smith, J. Am. Chem. Soc.,
1969, 91, 4211; (c) H. Shenhav, Z. Rappoport and S. Patai, J. Chem.
Soc. (B), 1970, 469.
24 Z. Rappoport and S. Gertler, J. Chem. Soc., 1964, 1360.
25 X. Huang and B.-C. Chen, Synthesis, 1986, 967.
J. Chem. Soc., Perkin Trans. 2, 2001, 1534–1545
1545