Trimethylsilyl bis(trifluoromethanesulfonyl)imide as a tolerant and environmentally benign Lewis acid catalyst of the Diels-Alder reaction

Article abstract of DOI:10.1016/S0040-4020(02)00971-7

N-trimethylsilyl bis(trifluoromethanesulfonyl)imide (TMSNTf₂) was readily prepared in situ by protodesilylation of trimethylsilane, allyl- or phenyltrimethylsilane with bis(trifluoromethylsulfonyl)imide. NMR studies showed that TMSNTf₂ was much more effective than TMSOTf in complexing the carbonyl group of trans-methylcrotonate. As a result, TMSNTf₂ was found to be superior to TMSOTf as a catalyst for the Diels-Alder reaction of α,β-unsaturated esters with a wide variety of dienes. In contrast to many metal derived Lewis acids, TMSNTf₂ was found tolerant of many sensitive functional groups present in the diene partner.

Full text of DOI:10.1016/S0040-4020(02)00971-7

Tetrahedron 58 (2002) 8219–8226
Trimethylsilyl bis(trifluoromethanesulfonyl)imide as a tolerant
and environmentally benign Lewis acid catalyst of the
Diels–Alder reaction
a
Benoit Mathieu and Leon Ghosez
a,b
,
*
´
aDepartment of Chemistry, University of Louvain, 1 Place Louis Pasteur, B-1348 Louvain-la-Neuve, Belgium
bEuropean Institute of Chemistry and Biology (IECB-ENSCPB), 16 Avenue Pey-Berland, F-330607 Pessac, France
Received 2 May 2002; accepted 15 May 2002
Abstract—N-trimethylsilyl bis(trifluoromethanesulfonyl)imide (TMSNTf2) was readily prepared in situ by protodesilylation of
trimethylsilane, allyl- or phenyltrimethylsilane with bis(trifluoromethylsulfonyl)imide. NMR studies showed that TMSNTf2 was much
more effective than TMSOTf in complexing the carbonyl group of trans-methylcrotonate. As a result, TMSNTf2 was found to be superior to
TMSOTf as a catalyst for the Diels–Alder reaction of a,b-unsaturated esters with a wide variety of dienes. In contrast to many metal derived
Lewis acids, TMSNTf2 was found tolerant of many sensitive functional groups present in the diene partner. q 2002 Elsevier Science Ltd. All
rights reserved.
1. Introduction
Lewis acids are among the most useful catalysts of organic
reactions.1 They are often used to activate a carbonyl group
of a reactant and consequently increase its electrophilicity.
However, problems can be encountered when the reactants
contain several functional groups which can act as Lewis
bases. This can lead to the deactivation of the catalyst and,
often, to undesired side reactions. We ran into this problem
while studying Diels–Alder reactions of 1 and 2-azadienes
Scheme 2.
which were found to decompose when exposed to several
commonly used Lewis acids (Scheme 1).2
No degradation of the sensitive 2-azadiene was observed
under these conditions. However, trialkylsilyltriflates did
not catalyze the reaction of methyl acrylate, a weaker Lewis
base. A more powerful silylating agent was thus needed.3
These studies also showed that trialkylsilyltriflates acted as
efficient Lewis acid catalysts for the reaction of a,b-
unsaturated amides with 2-azadienes (Scheme 2).3
In recent years, the interest of many research group for
genuine silyl cations4 has stimulated the search for highly
electrophilic silylating agents. Table 1 shows some very
electrophilic silylating agents along with their 29Si chemical
shift.
Table 1. 29Si chemical shift of silylating agents
Entry
Silylating agent
d
29Si
Reference
a
b
c
d
e
f
Me3SiCB11F11
Me3SiB(C6F5)4
Me3SiNTf2
Me3SiN(SO2F)2
Me3SiOTf
Me3SiClO4
120
84.8
55.9
44.9
43.5
43.4
32.5
5
6
7
8
9
10
10
Scheme 1.
Keywords: 2-azadienes; Diels–Alder reactions; amides.
*
Corresponding author. Fax: þ32-10-47-29-44; þ33-5-57-96-22-22;
g
Me3SiCl
e-mail: ghosez@chim.ucl.ac.be
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00971-7

8220
B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
Table 2. Comparative properties of HOTf and HNTf2
As a result of our previous experience with bis(trifluoro-
methylsulfonyl)imide salts which are stable and safe
substitutes for the corresponding perchlorates,11 we decided
to study the corresponding silylated derivatives. The TMS
derivative shows much lower 29Si chemical shift than the
TMS–carboranes and tetra(pentafluoro-phenyl)borate
(entries a and b) but it is cheaper and easier to prepare. In
1997, we reported our preliminary results on the use of
TMSNTf2 as an efficient catalyst for the cycloaddition of
methylacrylate to various dienes and for representative ene
reactions.12 Shortly after, the group of Mikami13 reported
that TMSNTf2 was an effective catalyst for the Friedel–
Crafts alkylation reaction and the group of Robertson14
reported the use of TMSNTf2 for the b-allylation of enones.
More recently, we have described the first chiral silicon
Lewis acid which led to significant facial selectivity in a
model reaction.15 Also Yamamoto’s group recently demon-
strated the efficiency of TMSNTf2 as a catalyst for
Mukayama–Aldol and Sakurai–Hosomi allylation reac-
tion.16 We now report the full detail of our study on the
preparation, physical properties and catalytic efficiency of
trimethylsilyl bis(trifluoromethansulfonyl)-imide.
HA/A2
pKa in acetic acid20
Donor number21
HNTf2/2NTf2
HOTf/2OTf
7.8
4.2
5.4
16.7
contaminated by some impurities. A purification by
distillation was needed.
No reaction was observed with TMSCl or tetrametylsilane
which had been previously shown to undergo protodesilyla-
tion in the presence of trifluoromethanesulfonic acid.18
protodesilylation reaction depends on both the strength of
the acid which activates the substrate by protonation and the
nucleophilicity of the conjugated base to displace the
trimethylsilyl substituent.19 As can be seen from Table 2,
both pKa and donor number should favor HOTf over HNTf2.
A
3. Evaluation of the Lewis acidity of TMSNTf2
To assess the Lewis acidity of TMSNTf2, we have selected
the much used22 though sometimes debated23 method of
Childs et al. It involves the measurement of the variation of
chemical shift of the proton at position 3 of an a,b-
unsaturated carbonyl compound upon complexation with a
Lewis acid.
2. Synthesis of TMSNTf2
TMSNTf2 had been first prepared by Foropoulos and
Desmarteau by protodesilylation of trimethylsilane with
HNTf2 (Scheme 3).17 This method requires the use of a
closed vessel to avoid the loss of the volatile
trimethylsilane.
A prerequisite for this study is the determination of the
stoechiometry of the complex. We thus followed the change
of the 1H NMR spectrum of a solution of trans-
methylcrotonate in CDCl3 upon successive addition of
small amounts of TMSNTf2 (Fig. 1). This spectrometric
titration showed an homogeneous chemical shift variation
of all the protons up to a saturation level slightly above
1 equiv. of TMSNTf2. This meant that the complex has a 1:1
stoechiometry.
TMSNTf2 has also been prepared by metathesis of
trimethylsilyl chloride with AgN(SO2CF3)2 (Scheme 3).7
This method required the use of an expensive and light
sensitive reagent. Also it is not well adapted to an in situ
preparation of TMSNTf2 since it would involve a filtration
under a strictly dry atmosphere. Scheme 3 shows the yields
obtained by these methods after distillation of the crude
products.
Table 3 shows that TMSNTf2 is very effective in complex-
ing various types of carbonyl groups. In contrast, TMSOTf
did not give any variation of the chemical shift. It is thus
clear that TMSNTf2 behaved as a much stronger Lewis acid
than TMSOTf (Scheme 4). This was in line with the
predictions based on the 29Si chemical shifts of the two
silylating agents. However it did not follow the order of
Bro¨nsted acidities which is HOTf.HNTf2.20 This was
A more practical method involved the protodesilylation of
allyltrimethylsilane. The reaction was fast at room tem-
perature and yielded very pure TMSNTf2 in essentially
quantitative yields. It is particularly well suited for the in
situ generation of TMSNTf2. Phenyltrimethylsilane reacted
equally well but vinyltrimethylsilane gave TMSNTf2
Scheme 3.
Figure 1. Spectrometric titration of trans-methylcrotonate with TMSNTf2.

B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
8221
3
¨
Table 3. Effect of Lewis and Bronsted acids on the chemical shifts of H
protons of a,b-unsaturated carbonyl compounds
further illustrated by the variation of chemical shifts of
protons at the position 3 of trans-methylcrotonate, trans-
crotonaldehyde and cyclohexenone upon addition of
1.5 equiv. of HOTf or HNTf2 (Table 3): the larger variations
of chemical shift were observed with HOTf.
LA/HA
Dd H3
Methylcrotonate
Crotonaldehyde
Cyclohexanone
TMSNTf2
TMSOTf
HNTf2
0.89
,0
0.11
0.46
1.74
,0
0.46
1.28
1.43
,0
This surprising reversal of acidity sequence in going from
the protic acid to the trimethylsilyl derivative could result
from the size difference of the two anions. The bis(tri-
fluoromethylsulfonyl)imide anion is much larger than the
triflate anion. Therefore, the higher I-strain in TMSNTf2
would thermodynamically favor the formation of the
complex with a smaller ligand such as an ester group.
This difference in I-strain does not exist in the protic acid.
0.2
HOTf
1.18
Table 3 also showed that the chemical shift variations
induced by HNTf2 are one order of magnitude lower than
those brought about by a similar amount of TMSNTf2. This
meant that a contamination of TMSNTf2 by small amounts
of HNTf2 could only lead to a negligible underestimation of
the Lewis acidity of TMSNTf2.
Scheme 4.
Table 4. Diels–Alder reactions of a,b-unsaturated esters in toluene with 10 mol% silylating agent
Entry
Diene
Dienophile
Conditions
Product (major)
Yield (%)
,5
83
Endo/exo
a
b
08C, 20 min, TMSOTf
08C, 20 min, TMSNTf2
–
24:1
c
d
e
f
g
h
rt, TMSOTf
rt, TMSNTf2
nr
Degradation
nr
nr
nr
94
–
–
–
–
rt, TMSNTf2þ10 mol% iPrNEt2
rt, TMSNTf2þ10 mol% pyridine
rt, TMSNTf2þ10 mol% collidine
rt, TMSNTf2þ10 mol% BTBMP
–
11.5:1
i
rt, 15 min, TMSNTf2
93
1.03:1
j
k
08C, 1 h, TMSOTf
08C, 1 h, TMSNTf2
nr
92
–
49:1
l
m
08C!rt, 1 h, TMSOTf
08C!rt, 1 h, TMSNTf2
nr
79
–
1.5:1
n
o
08C!rt, 1 h, TMSOTf
08C!rt, 1 h, TMSNTf2
nr
76
–
24:1
p
q
2508C!rt, 1 h, TMSOTf
2508C!rt, 1 h, TMSNTf2
nr
91
–
.99:1
r
2508C!rt, 1 h, TMSNTf2
79
9:1
s
t
08C, 90 min, TMSOTf
08C, 90 min, TMSNTf2
,5
74
–
2:1

8222
B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
4. Diels–Alder reactions catalyzed by TMSNTf2
The difference in Lewis acidity between TMSNTf2 and
TMSOTf is dramatically illustrated by their effect on the
rate of cycloaddition of dienes with a,b-unsaturated esters
(Table 4).
TMSOTf was unable to catalyze the Diels–Alder reactions
of methyl acrylate or trans-methylcrotonate whereas
TMSNTf2 always acted as an excellent Lewis acid catalyst.
Another important conclusion of Table 4 is that TMSNTf2
did not cause the degradation of the reactants even when
very sensitive dienes like 2-azadienes were used. The
reaction of methylcrotonate with cyclopentadiene is quite
illustrative of this quality of the catalyst. We had observed
degradation reactions in the presence of TMSNTf2. It was
believed that it was not due to TMSNTf2 itself, but rather to
traces of HNTf2 contaminating the catalyst. The reaction of
cyclopentadiene with methylcrotonate being slower than
that of methylacrylate, degradation of the reactants
(probably cyclopentadiene) by the residual acid could
become more significant. Addition of various bases (entries
e–g) did indeed suppress the degradation of the reactants
but also deactivated the catalyst: no reaction was observed.
Control experiments by 1H NMR showed that the base was
silylated by TMSNTf2 under these conditions. The use of a
very hindered base was anticipated to suppress the reaction
with TMSNTf2 while still allowing for a proton abstraction
from HNTf2. Rewardingly, it was found that, in the presence
of 2,6-bis-tert-butyl-4-methylpyridine (BTBMP), no
degradation occurred and the catalytic activity of TMSNTf2
was restored (entry k). This observation led us to
recommend the use of a small amount of BTBMP while
running reactions in the presence of TMSNTf2: this will
suppress any reaction due to the presence of traces of HNTf2
which are very difficult to avoid. A similar situation can be
encountered with reactions catalyzed by TMSOTf: here also
we recommend to add small amount of BTBMP to
neutralize the residual HOTf.
Scheme 6.
yielded trans adducts which were separated by preparative
HPLC. The cycloaddition with dimethylmaleate gave three
products, which were separated by HPLC. Control experi-
ments showed that under the reaction conditions (1) no
isomerization of dimethylmaleate to dimethylfumarate had
occurred, (2) no isomerisation of the primary adducts was
observed as shown in Scheme 6. The endo-cis adduct was
silylated to give a mixture of enol ethers, which did not
epimerize under the conditions used for the cycloaddition or
upon treatment with HCl–EtOH.
The results described in Schemes 5 and 6 are of mechanistic
significance. The formation of a mixture of cis and trans-
adducts from dimethylmaleate suggest a two step
mechanism for the cycloaddition reaction (Scheme 7).24
Reaction of the silylenol ether function of the diene on the
activated form of dimethylmaleate would lead to an
intermediate adduct which should undergo several con-
formationnal changes before undergoing the Mukaiyama-
type cyclisation. These conformational changes of the
intermediate adduct easily account for the formation of
mixtures of cis and trans-adducts. However, conformational
equilibrium has not been attained since we observed only
trans-adducts from dimethylfumarate.
We have also examined the stereochemistry of the catalyzed
reaction (Scheme 5). The reaction of dimethylfumarate only
We believe that these results conclusively show that in situ
generated TMSNTf2 is an efficient Lewis acid catalyst for
the Diels–Alder reaction of a,b-unsaturated esters with a
wide variety of dienes, including some which have shown to
Scheme 5. Reaction conditions: 10 mol% TMSNTf2, 10 mol% BTBMP,
toluene, 08C.
Scheme 7.

B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
8223
decompose in the presence of many conventional Lewis
5.2. Complexation experiments
acids. They also show the superiority of TMSNTf2 as a
Lewis acid catalyst over TMSOTf. We are confident that
this strong, non-metallic Lewis acid will find many
applications as a tolerant and ‘green’ catalyst of reactions
triggered by carbonyl groups.
The experiments were performed under argon atmosphere
in a 5 mm oven-dried NMR tube closed by a septum. The
carbonyl compound (0.4 mmol) and 0.6 ml of CDCl3 were
1
syringed in and a H spectrum was recorded at 08C. The
¨
sample was cooled at 2408C and the Lewis or Bronsted acid
were added to the mixture which was vortexed. The H
1
NMR spectra were recorded at 08C.
5. Experimental
1H NMR spectra were recorded in CDCl3 on Varian
Gemini-200, 300 or on Bruker DRX-500 spectrometers at
200, 300 or 500 MHz. 13C NMR spectra were recorded at
50, 75 or 125 MHz. Chemical shifts are given in ppm
5.3. General procedure for the catalyzed Diels–Alder
reactions
A solution of the dienophile in toluene was placed in a two
necked flask isolated from moisture with a silicagel trap.
The solution was brought up to the selected temperature and
10 mol% of TMSNTf2 followed by the diene were added to
the solution of the dienophile. The reaction mixture was
quenched by addition of 5 ml of a 1 M aqueous solution of
NaHCO3 and the resulting mixture was stirred for 1 h. The
aqueous phase was extracted twice with 15 ml of CH2Cl2.
The combined organic phases were dried over MgSO4.
Removal of the solvent in vacuo left a residue which was
purified by flash chromatography on silica gel.
1
relative to Me4Si (0 ppm, H) or CDCl3 (77.0 ppm, 13C).
Mass spectra were recorded on a Finnigan MAT-TSQ 700
spectrometer. IR spectra were recorded on a BIO-RAD TFS
135 FT-IR spectrometer. All absorption values are
expressed in wavenumbers (cm21). Melting point were
obtained on a Bu¨chi Melting Point B-545 or a Microthermal
8103 apparatus and are uncorrected. HPLC were run
on Waters apparatus including a Waters 600 Controller
and a Waters 486 detector fitted with a 6 mm silica
column. TLC were run on silicagel 60F254. Column
chromatographies were performed on silicagel 40
(230–400 mm, Merck). All solvents were distilled before
use. HN(SO2CF3)2 was graciously provided by Rhodia
(Lyon, France) and was purified by distillation from 98%
H2SO4. The various silanes were purchased from Aldrich
and used without further purification. All reactions were run
in flame dried glassware under strictly dry atmosphere of
argon.
5.3.1. Cycloaddition of methylacrylate with cyclopenta-
diene. Methylacrylate (200 ml, 191 mg, 2.22 mmol,
1 equiv.), cyclopentadiene (366 ml, 293 mg, 4.44 mmol,
2 equiv.), TMSNTf2 (78 mg, 222 mmol, 0.1 equiv.), toluene
(3 ml); 08C, 20 min; cyclohexane/AcOEt 95:5; yield:
0.282 g (1.85 mmol, 83%); endo/exo: 96:4.
endo Cycloadduct; RN: 2903-75-5; 1H NMR (CDCl3,
300 MHz): 1.27 (1H, d, J¼8.0 Hz), 1.36–1.45 (2H, m),
1.84–1.97 (1H, m), 2.85–2.99 (2H, m), 3.19 (1H, s large),
3.54 (3H, s), 5.93 (1H, dd, J¼5.6, 2.9 Hz), 6.19 (1H, dd,
J¼5.6, 3.0 Hz); 13C NMR (CDCl3, 50 MHz): 29.33, 42.56,
43.23, 45.68, 49.61, 51.29, 132.41, 137.65, 175.09.
5.1. General procedure for the reaction of silanes with
HN(SO2CF3)2
In a one necked flask equipped with a condenser, the acid
was introduced and covered with solvent. The silane was
added at 08C under stirring. Reaction was completed in the
given conditions then the volatile compounds were
evaporated under strictly dry vacuum. The residue was
cannulated in a Kugelrohr distillation set and distillated
under vacuum.
5.3.2. Cycloaddition of methylcrotonate with cyclo-
pentadiene. Methylcrotonate (200 ml, 188 mg, 1.88 mmol,
1 equiv.), cyclopentadiene (466 ml, 373 mg, 5.65 mmol,
3 equiv.), TMSNTf2 (45 ml, 66 mg, 188 mmol, 0.1 equiv.),
toluene (4 ml); rt, 15 min; petroleum ether/Et2O 96:4;
endo/exo: 92:8.
From allyltrimethylsilane. HNTf2: 5 g (17.78 mmol,
1 equiv.),
35.56 mmol, 2 equiv.); No solvent, reaction at 208C until
end of gas evolution; yield: 5.7 g (16.13 mmol, 91%).
allyltrimethylsilane:
5.65 ml
(4.06 g,
1
endo Cycloadduct; RN: 155325-46-5; H NMR (CDCl3,
500 MHz): 1.18 (3H, d, J¼6.9 Hz), 1.43 (1H, dd, J¼8.6,
1.9 Hz), 1.54 (1H, d, J¼8.6 Hz), 1.82 (1H, ddq J¼4.4, 6.9,
1.9 Hz), 2.37 (1H, dd, J¼3.9, 4.4 Hz), 2.46 (1H, s large), 3.1
(1H, s large), 3.61 (3H, s), 5.98 (1H, dd, J¼5.8, 3.2 Hz),
6.26 (1H, dd, J¼5.8, 3.2 Hz); 13C NMR (CDCl3, 125 MHz):
45.72, 52.24, 37.7, 48.65, 138.48, 133.12, 45.78, 20.74,
174.95, 51.19.
From vinyltrimethylsilane. HNTf2: 1.15 g (4.09 mmol,
1 equiv.), vinyltrimethylsilane: 3.15 ml (2.05 g,
20.45 mmol, 5 equiv.); no solvent, reaction at reflux for
18 h; yield: 1.06 g (3.01 mmol, 74%).
From phenyltrimethylsilane. HNTf2: 98 mg (348 mmol,
1.2 equiv.), Me3SiPh: 50 ml (43 mg, 290 mmol, 1 equiv.);
CDCl3: 0.6 ml, 1 h at 208C; yield: .95% (determined by
NMR).
exo Cycloadduct; RN: 23217-30-3; 1H NMR (CDCl3,
500 MHz): 0.91 (3H, d, J¼6.9 Hz), 1.44 (1H, dq, J¼8.6,
1.7 Hz), 1.64 (1H, dd, J¼5.2, 1.8 Hz), 1.67 (1H, d,
J¼8.6 Hz), 2.35 (1H, m), 2.7 (1H, s large), 2.94 (1H, s
large), 3.69 (3H, s), 6.11 (1H, dd, J¼5.8, 3.2 Hz), 6.2 (1H,
dd, J¼5.8, 3.2 Hz); 13C NMR (CDCl3, 125 MHz): 47.19,
51.06, 38.94, 47.08, 135.14, 136.43, 47.97, 18.87, 176.52,
51.46.
1
RN: 82113-66-4; colourless liquid; bp: 1158C (2 Torr); H
NMR (200 MHz): 0.58 (s, 3H); 19F NMR (282 MHz):
277.94 (s, CF3); 13C NMR (50 MHz): 0.29 (Me), 118.72 (q,
JC–F¼320 Hz); 29Si NMR (99 MHz): 54.3.

8224
B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
5.3.3. Cycloaddition of methylmethacrylate with cyclo-
pentadiene. Methylmetacrylate (200 ml, 187 mg,
1.86 mmol, 1 equiv.), cyclopentadiene (462 ml, 370 mg,
5.60 mmol, 3 equiv.), TMSNTf2 (45 ml, 66 mg, 186 mmol,
0.1 equiv.), toluene (4 ml); rt, 15 min; petroleum ether/Et2O
96:4; endo/exo: 57:43; yield: 0.289 g (1.74 mmol, 93%).
1735, 2935s, 2955; MS (EI, 70 eV): 45 (59%), 55 (90%), 73
(100%), 75 (56%), 89 (30%), 151 (11%), 210 (5%), 239
(11%), 296 (1%); Elemental analysis calcd (%) for
C16H28O3Si: C 64.81, H 9.51; found: C 64.91, H 9.53.
5.3.6. Cycloaddition of methylacrylate with 1,3-bis(tert-
butyldimethylsiloxy)cyclohexa-1,3-diene.
1,3-Bis(tert-
endo Cycloadduct; RN: 7167-29-5; 1H NMR (CDCl3,
500 MHz): 1.35–1.45 (2H, m), 1.49 (1H, d, J¼8.6 Hz),
1.87 (3H, s), 1.87 (1H, dd, J¼11.9, 2.6 Hz), 2.71 (1H, s
large), 2.76 (1H, s large), 3.54 (3H, s), 5.95 (1H, dd, J¼5.6,
2.9 Hz), 6.06 (1H, dd, J¼5.6, 3.2 Hz); 13C NMR (CDCl3,
125 MHz): 50.77, 49.72, 37.75, 42.35, 137.42, 135.05,
46.55, 26.11, 177.47, 51.07.
butyldimethylsiloxy)cyclohexa-1,3-diene (450 mg, 1.32
mmol, 1.23 equiv.), methylacrylate (96 ml, 91 mg,
1.07 mmol, 1 equiv.), TMSNTf2 (37 mg, 106 mmol,
0.1 equiv.), toluene (3 ml); addition of reagents at 08C
then reaction at rt for 1 h; cyclohexane/AcOEt 9:1; yield:
0.255 g (0.82 mmol, 76.5%); endo/exo: 96:4.
endo Cycloadduct; 1H NMR (CDCl3, 200 MHz): 0.07 (3H,
s), 0.11 (3H, s), 0.83 (9H, s), 1.74–1.87 (4H, m), 1.98–2.05
(2H, m), 2.2–2.35 (1H, m), 2.23 (1H, d, J¼19 Hz), 2.9 (1H,
dd, J¼9.2, 7.9 Hz), 3.22 (1H, d, J¼18.6 Hz), 3.67 (3H, s);
13C NMR (CDCl3, 50 MHz): 211.75, 174.78, 74.31, 51.6,
47.76, 47.45, 41.35, 34.82, 27.43, 25.41, 22.25, 17.69,
22.21; IR (neat): 860, 1140, 1170, 1205, 1260, 1325, 1740,
2860, 2940, 2960; MS (EI, 70 eV, þQ1MS): 55 (43%), 75
(35%), 89 (30%), 195 (12%), 227 (26%), 255 (100%), 312
(1%); Elemental analysis calcd (%) for C16H28O4Si: C
61.49, H 9.03; found: C 61.50, H 8.90.
exo Cycloadduct; RN: 7167-28-4; 1H NMR (CDCl3,
500 MHz): 0.77 (1H, dd, J¼11.9, 2.2 Hz), 1.02 (3H, s), 1.29
(1H, d, J¼8.6 Hz), 1.36 (1H, dd, J¼8.6, 2.2 Hz), 2.36 (1H, dd,
J¼11.9, 3.9 Hz), 2.74 (1H, s large), 2.95 (1H, s large), 3.62
(3H, s), 6.00 (1H, dd, J¼5.6, 3.1 Hz), 6.14 (1H, dd, J¼5.6,
3.0 Hz); 13C NMR (CDCl3, 125 MHz): 50.22, 49.33, 37.48,
42.62, 138.38, 133.31, 48.82, 23.97, 178.84, 51.57.
5.3.4. Cycloaddition of methylacrylate with 1,3-cyclo-
hexadiene. Methylacrylate (200 ml, 191 mg, 2.22 mmol,
1 equiv.), 1,3-cyclohexadiene (413 ml, 355 mg, 4.44 mmol,
2 equiv.), TMSNTf2 (78 mg, 222 mmol, 0.1 equiv.), toluene
(6 ml); 08C, 60 min; cyclohexane/AcOEt 95:5; yield:
0.338 g (2.03 mmol, 92%); endo/exo: .95:5.
exo Cycloadduct; 1H NMR (CDCl3, 200 MHz): 0.07 (3H, s),
0.11 (3H, s), 0.84 (9H, s), 1.57 (1H, tdd, J¼11.9, 4.4,
2.1 Hz), 1.77 (1H, tm, J¼13.0 Hz), 1.85–2.13 (3H, m),
2.27–2.54 (4H, m), 2.77 (1H, ddd, J¼10.8, 6.4, 2.0 Hz),
3.69 (3H, s); 13C NMR (CDCl3, 50 MHz): 212.02, 174.71,
74.16, 53.3, 51.29, 48.27, 41.46, 28.03, 26.63, 25.39, 22.27,
17.77, 22.25; IR (neat): 870, 1135, 1170, 1205, 1260, 1335,
1370, 1735, 2860, 2940, 2960; MS (EI, 70 eV): 55 (16%),
75 (100%), 89 (30%), 227 (6%), 255 (16%), 312 (1%);
Elemental analysis calcd (%) for C16H28O4Si: C 61.49, H
9.03; found: C 61.60, H 9.00.
endo Cycloadduct; RN: 25578-17-0; 1H NMR (CDCl3,
500 MHz): 1.00–1.9 (6H, m), 2.34–3.00 (3H, m), 3.54 (3H,
s), 5.59–6.40 (2H, m).
5.3.5. Cycloaddition of methylacrylate with 1-(tert-
butyldimethylsilyloxy)cyclohexa-1,3-diene. Methylacryl-
ate (100 ml, 95 mg, 1.11 mmol, 1 equiv.), 1-(tert-butyl-
dimethylsilyloxy)cyclohexa-1,3-diene (303 mg, 1.44 mmol,
1.3 equiv.), TMSNTf2 (39 mg, 111 mmol, 0.1 equiv.),
toluene (3 ml); addition of reagents at 08C then reaction at
rt for 1 h; cyclohexane/AcOEt 96:4; yield: 0.257 g
(0.23 mmol, 79%); endo/exo: 60:40.
5.3.7. Cycloaddition of methylacrylate with 1-phenyl-3-
(trimethylsilyloxy)buta-1,3-diene. 1-Phenyl-3-(trimethyl-
silyloxy)buta-1,3-diene (315 mg, 1.44 mmol, 1.3 equiv.),
methylacrylate (100 ml, 95 mg, 1.11 mmol, 1 equiv.),
TMSNTf2 (39 mg, 111 mmol, 0.1 equiv.), toluene (3 ml);
addition of reagents at 2508C then reaction at rt for 1 h;
cyclohexane/AcOEt 8:2; yield: 0.232 g (1 mmol, 91%);
endo/exo: .95:5 (single isomer observed).
endo Cycloadduct; 1H NMR (CDCl3, 500 MHz): 0.11 (6H,
s), 0.87 (9H, s), 1.35–1.46 (2H, m), 1.5–1.51 (1H, m),
1.56–1.66 (2H, m), 1.87 (1H, ddd, J¼12.5, 9.9, 2.8 Hz),
2.5–2.6 (1H, m), 2.74 (1H, dd, J¼9.9, 5.9 Hz), 3.61 (3H, s),
6.16 (1H, d, J¼8.6 Hz), 6.21 (1H, dd, J¼8.6, 6.5 Hz); 13C
NMR (CDCl3, 50 MHz): 175.35, 136.35, 131.17, 76.40,
51.26, 49.93, 35.33, 33.22, 29.41, 25.78, 25.61, 17.98,
22.24, 22.44; IR (neat): 880, 1130, 1170, 1205, 1265,
1360, 1435, 1740, 2940, 2960; MS (EI, 70 eV): 73 (20%),
89 (14%), 151 (55%), 210 (46%), 239 (100%), 296 (5%);
Elemental analysis calcd (%) for C16H28O3Si: C 64.81, H
9.51; found: C 64.90, H 9.75.
1
endo Cycloadduct; RN: 136145-71-6; H NMR (CDCl3,
300 MHz): 2.1 (1H, dddd, J¼14.1, 4.9, 5.5, 5.9 Hz), 2.24
(1H, dddd, J¼14.1, 4.9, 5.5, 10.3 Hz), 2.4 (1H, ddd, J¼15.5,
5.5, 5.5 Hz), 2.6 (1H, dd, J¼14.6, 5.6 Hz), 2.8 (1H, ddd,
J¼15.5, 5.9, 10.3 Hz), 3.12 (1H, q, J¼4.9 Hz), 3.27 (1H, dd,
J¼14.6, 10.4 Hz), 3.5 (1H, ddd, J¼5.6, 10.4, 4.9 Hz), 3.55
(3H, s), 7.1–7.35 (5H, m); 13C NMR (CDCl3, 50 MHz):
26.14, 38, 43.36, 44.53, 45.08, 51.21, 127.08, 127.29,
128.38, 140.73, 173.23, 210.08.
exo Cycloadduct; 1H NMR (CDCl3, 200 MHz): 0.07 (3H, s),
0.12 (3H, s), 0.89 (9H, s), 1.2–1.8 (5H, m), 2.24 (1H, td,
J¼11.3, 3.5 Hz), 2.5–2.6 (2H, m), 3.66 (3H, s), 6.14 (1H,
dd, J¼8.7, 6.3 Hz), 6.26 (1H, d, J¼8.4 Hz); 13C NMR
(CDCl3, 50 MHz): 175.5, 139.31, 132.08, 77.19, 51.26,
49.11, 30.75, 29.52, 28.65, 26.13, 25.62, 18.4, 22.33,
22.56; IR (neat): 880, 1150, 1180, 1195, 1210, 1260, 1360,
5.3.8. Cycloaddition of methylcrotonate with 1-phenyl-3-
(trimethylsilyloxy)buta-1,3-diene. 1-Phenyl-3-(trimethyl-
silyloxy)buta-1,3-diene: (315 mg, 1.44 mmol, 1.3 equiv.),
methylcrotonate (117 ml, 111 mg, 1.11 mmol, 1 equiv.),
TMSNTf2 (39 mg, 111 mmol, 0.1 equiv.), toluene (3 ml);
addition of reagents at 2508C then reaction at rt for 1 h;

B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
8225
cyclohexane/AcOEt 8:2; endo/exo: 9:1; yield: 0.215 g
(0.87 mmol, 79%).
m), 7.72 (3H, s); 13C NMR (C6D6, 125 MHz): 204.55,
42.53, 46.02, 51.77, 46.66, 47.83, 173.42, 173.13, 51.87,
52.38, 141.84, 127.15 (CDCl3), 128.8 (CDCl3), 127.63
(CDCl3); IR (neat): 702, 772, 1011, 1168, 1434, 1712, 1727,
2952, 3035; MS (EI, 70 eV): 59 (12%), 104.1 (14%), 129.1
(17%), 171.1 (100%), 230.1 (84%), 259.1 (28%), 290.2
(32%); HRMS (EI, 70 eV) calcd (%) for C16H18O5
290.1154; found 290.1160.
endo Cycloadduct; RN: 83194-78-9; 1H NMR (CDCl3,
500 MHz): 1.13 (3H, d, J¼6.8 Hz), 2.17 (1H, dd, J¼14.8,
6.8 Hz), 2.53 (1H, m, J¼5.5, 6.8 Hz), 2.58 (1H, dd, J¼14.6,
5.03 Hz), 2.85 (1H, t, J¼5.5 Hz), 2.89 (1H, dd, J¼14.8,
5.5 Hz), 3.2 (1H, dd, J¼14.6, 9.86 Hz), 3.46 (3H, s),
3.61 (1H, ddd, J¼5.03, 9.86, 5.5 Hz), 7.1–7.4 (5H, m); 13C
NMR (CDCl3, 125 MHz): 210.45, 43.02, 41.24, 52,
31.16, 45.32, 173.25, 51.09, 140.4, 127.31, 128.24,
126.96, 20.43.
endo-trans Cycloadduct; mp: 90.58C; HPLC (6 mm silica
column, eluent: hexane (90%)/iPrOH, flow: 0.5 ml/min,
1
l¼254 nm): 19.26 min; H NMR (CDCl3, 500 MHz): 2.6
(1H, dd, J¼15.7, 9.3 Hz), 2.72 (1H, dd, J¼15.7, 5.9 Hz), 2.9
(1H, dd, J¼15.7, 5.3 Hz), 2.99 (1H, dd, J¼15.7, 5.9 Hz),
3.24 (1H, ddd, J¼8.6, 9.3, 5.3 Hz), 3.49 (1H, dd, J¼5.1,
8.6 Hz), 3.54 (3H, s), 3.71 (3H, s), 3.79 (1H, q, J¼5.9 Hz),
7.03–7.32 (5H, m); 13C NMR (CDCl3, 125 MHz): 207.42,
43.57, 41.58, 47.33, 40.68, 40.82, 173.32, 172.03, 51.62,
52.31, 139.62, 127.44, 128.48, 127.38; IR (neat): 702, 908,
1018, 1164, 1276, 1435, 1738, 2953, 3003, 3030; HPLC-MS
(APCI): 129.3 (16%), 143.2 (21%), 157.2 (24%), 171.2
(61%), 181.2 (35%), 199.1 (100%), 227.1 (22%), 231.1
(17%), 259.1 (45%), 290.9 (3%); HRMS (EI, 70 eV) calcd
(%) for C16H18O5 290.1154; found 290.1169.
5.3.9. Cycloaddition of methylacrylate with 1-phenyl-3-
trimethylsilyloxy-2-azapenta-1,3-diene. 1-Phenyl-3-tri-
methylsilyloxy-2-azapenta-1,3-diene (310 mg, 1.33 mmol,
1.2 equiv.), methylacrylate: (100 ml, 95 mg, 1.11 mmol,
1 equiv.), TMSNTf2 (39 mg, 111 mmol, 0.1 equiv.), toluene
(3 ml); reaction at 08C for 90 min; AcOEt/cyclohexane
75:25; yield: 0.205 g (0.83 mmol, 74%); endo/exo: 67:33.
endo Cycloadduct; 1H NMR (CDCl3, 200 MHz): 1.31 (3H,
d, J¼6.9 Hz), 1.76 (1H, q, J¼13 Hz), 1.94 (1H, ddd,
J¼14.0, 6.9, 3.7 Hz), 2.4 (1H, hept, J¼6.9, 12.3 Hz), 3.22
(1H, ddd, J¼4.6, 3.7, 12.8 Hz), 3.56 (3H, s), 5.01 (1H, t,
J¼4.6 Hz), 6.54 (1H, s large), 7.1–7.35 (5H, m); 13C NMR
(CDCl3, 50 MHz): 16.91, 25.89, 35.28, 44.69, 51.57, 57.11,
128.24, 128.29, 127.21, 138.42, 171.04, 174.72; IR (neat):
801, 1261, 1662, 1741, 2950, 3033, 3209; MS (EI, 70 eV):
55.1 (41%), 77 (34%), 106.1 (95%), 133.1 (21%), 161.1
(100%), 177 (97%), 188.2 (20%), 232.1 (4%), 247 (64%);
HRMS (EI, 70 eV) calcd (%) for C16H28O4Si 247.1208;
found 247.1213.
5.3.11. Cycloaddition of dimethylmaleate with 1-phenyl-
3-(trimethylsilyloxy)buta-1,3-diene.
(200 ml, 230 mg, 1.60 mmol, 1 equiv.), 1-phenyl-3-(tri-
methylsilyloxy)buta-1,3-diene (453 mg, 2.08 mmol,
Dimethylmaleate
1.3 equiv.), 2,6-bis(tert-butyl)-4-methylpyridine (32 mg,
159 mmol, 0.1 equiv.), TMSNTf2 (38 ml, 56 mg,
159 mmol, 0.1 equiv.), toluene (4 ml); reaction at 08C for
1 h; petroleum ether/AcOEt 75:25; yield: 0.390 g
(1.34 mmol,
84%);
34:17:49 (determined by HPLC).
endo-cis/exo-trans/endo-trans:
exo Cycloadduct; 1H NMR (CDCl3, 200 MHz): 1.21 (3H, d,
J¼7.2 Hz), 1.72 (1H, ddd, J¼10.8, 5.6, 3.8 Hz), 2–2.4 (2H,
m), 2.81 (1H, ddd, J¼7.4, 4, 1.6 Hz), 3.62 (3H, s), 4.85 (1H,
dd, J¼1.5, 7.4 Hz), 6.51 (1H, s large), 7.2–7.4 (5H, m); 13C
NMR (CDCl3, 75 MHz): 17.81, 29.68, 33.42, 44.68, 52.04,
58.3, 126.56, 128.3, 128.82, 140.58, 172.79, 175.24; IR
(neat): 1193, 1355, 1455, 1661, 1734, 2954, 3063, 3219; MS
(EI, 70 eV): 119 (4%), 161.1 (22%), 177.1 (100%), 188.2
(18%), 232.6 (1%), 247.3 (10%); HRMS (EI, 70 eV) calcd
(%) for C16H28O4Si 247.1208; found 247.1204.
endo-cis Cycloadduct; mp: 84.58C; HPLC (6 mm silica
column, eluent: hexane (90%)/iPrOH, flow: 0.5 ml/min,
1
l¼254 nm): 16.27 min; H NMR (CDCl3, 500 MHz): 2.52
(1H, dd, J¼14.4, 4.6 Hz), 2.68 (1H, dd, J¼14.4, 4.6 Hz),
3.13 (1H, dt, J¼13.7, 4.6 Hz), 3.28 (1H, dd, J¼14.4,
13.7 Hz), 3.32 (1H, dt, J¼13.7, 4.6 Hz), 3.4 (1H, dd,
J¼14.4, 13.7 Hz), 3.41 (1H, t, J¼4.6 Hz), 3.43 (3H, s), 3.7
(3H, s), 7.1–7.4 (5H, m); 13C NMR (CDCl3, 125 MHz):
208.17, 41.42, 44.01, 46.95, 44.46, 39.15, 171.78, 171.85,
51.29, 52.26, 139.79, 126.78, 128.53, 127.44; IR (neat):
700, 992, 1174, 1220, 1272, 1438, 1712, 1720, 1736, 2952,
3029, 3062; HPLC-MS (APCI): 129.3 (11%), 143.2 (22%),
157.2 (16%), 171.2 (56%), 181.3 (48%), 199.1 (100%), 227
(28%), 259.1 (20%); HRMS (EI, 70 eV) calcd (%) for
C16H18O5 290.1154; found 290.1150.
5.3.10. Cycloaddition of dimethylfumarate with 1-
phenyl-3-(trimethylsilyloxy)buta-1,3-diene. Dimethyl-
fumarate (200 mg, 1.39 mmol, 1 equiv.); 1-phenyl-3-(tri-
methylsilyloxy)buta-1,3-diene
(393 mg,
1.80 mmol,
1.3 equiv.), 2,6-bis(tert-butyl)-4-methylpyridine: (28 mg,
138 mmol, 0.1 equiv.), TMSNTf2 (33 ml, 49 mg,
138 mmol, 0.1 equiv.), toluene (4 ml); reaction at 08C for
1 h; petroleum ether/AcOEt 75:25, HPLC separation of
isomers: hexane/iPrOH 80:20, 258C, 0.3 ml/min; yield:
0.293 g (1.01 mmol, 73%); endo/exo: 72:28.
Acknowledgements
exo-trans Cycloadduct; mp: 107.58C; HPLC (6 mm silica
column, eluent: hexane (90%)/iPrOH, flow: 0.5 ml/min,
l¼254 nm): 17.373 min;1H NMR (C6D6, 500 MHz): 2.53
(1H, dd, J¼15.0, 13.8 Hz), 2.76 (1H, dd, J¼15.0, 13.0 Hz),
2.8 (1H, ddd, J¼15.0, 4.0, 1.8 Hz), 3.11 (1H, ddd, J¼15.0,
4.4, 1.8 Hz), 3.33 (1H, ddd, J¼11.1, 13.8, 4.0 Hz), 3.47 (1H,
t, J¼11.1 Hz), 3.5–3.55 (1H, m), 3.54 (3H, s), 7.3–7.7 (5H,
Financial support was generously provided by the
‘Ministere de l’Education et de la Recherche Scientifique
´
de la Communaute franc¸aise de Belgique’ (action de
recherche concertee 96/01-197), F.R.I.A., the University
´
of Louvain (fellowships to B. M.) and Rhodia. We thank Dr
R. Touillaux for NMR analyses and Pr E. de Hoffman for
the mass spectra.

8226
B. Mathieu, L. Ghosez / Tetrahedron 58 (2002) 8219–8226
1995, 565–567. (b) Tamion, R.; Mineur, C.; Ghosez, L.
References
Tetrahedron Lett. 1995, 36, 8977–8980.
12. (a) Mathieu, B.; Ghosez, L. Tetrahedron Lett. 1997, 38,
5497–5500. (b) Mathieu, B.; Desmurs, J. R.; Ghosez,
L. French Patent 99 05195, 1999, World Patent 0064908 2000.
13. Ishii, A.; Kotera, O.; Saeki, T.; Mikami, K. Synlett 1997, 10,
1145–1146.
1. Reviews: (a) Oppolzer, W. Angew. Chem., Int. Ed. Engl. 1984,
23, 876. (b) Paquette, L. A. Assymetric Synthesis; Morisson,
J. P., Ed.; Academic: New York, 1984; Vol. 3. (c) Kagan,
H. B.; Riant, O. Chem. Rev. 1992, 92, 1007. (d) Pindur, U.;
Lutz, G.; Otto, C. Chem. Rev 1993, 93, 741. (e) Ghosez, L. In
Stereocontrolled Organic Synthesis; Trost, B., Ed.; Blackwell
Scientific Publication: Oxford, 1994; p 193. (f) Yamamoto, H.;
Maruoka, K.; Ishihara, K.; Yanagisawa, A.; Oguni, N.;
Mikami, K.; Terada, M.; Kobayashi, S.; Hosomi, A.; Miura,
K.; Sawamura, M.; Suzuki, K.; Fukuzawa, S.; Nakai, T.;
Tomooka, K.; Nishiyama, H.; Motoyama, Y.; Aida, T.;
Takeuchi, D. Lewis Acid Reagents, A Practical Approach;
Oxford University: Oxford, 1999.
14. (a) Kuhnert, N.; Peverley, J.; Robertson, J. Tetrahedron Lett.
1998, 39, 3215–3216. For the silylation of functional groups,
see also: (b) Simchen, G.; Jonas, S. J. Prakt. Chem. Chem. Ztg.
1998, 340, 506–512.
15. Mathieu, B.; de Fays, L.; Ghosez, L. Tetrahedron Lett. 2000,
41, 9561–9564.
16. Ishihara, K.; Hiraiwa, Y.; Yamamoto, H. Synlett 2001, 12,
1851–1854.
2. Mbiya, K. PhD Dissertation, University of Louvain, 1994.
3. (a) Lamy-Schelkens, H.; Giomi, D.; Ghosez, L. Tetrahedron
Lett. 1989, 30, 5887–5890. (b) Lamy-Schelkens, H.; Ghosez,
L. Tetrahedron Lett. 1989, 30, 5891–5894.
17. Foropoulos, J.; DesMarteau, D. D. J. Am. Chem. Soc. 1982,
104, 4260–4261.
18. (TMSClþHOTf) see Ref. 9 (Me4SiþHOTf) Demuth, M.;
Mikhail, G. Tetrahedron 1983, 39, 991–997.
4. (a) Lambert, J.; Zhao, Y.; Zhang, S. J. Phys. Org. Chem. 2001,
14, 370–379. (b) Lambert, J.; Kania, L.; Zhang, S. Chem. Rev.
1995, 95, 1191–1201. (c) Olsson, L.; Ottosson, C.-H.; Cremer,
D. J. Am. Chem. Soc. 1995, 117, 7460–7479.
19. (a) Hagen, G.; Mayr, H. J. Am. Chem. Soc. 1991, 113,
4954–4961. (b) Bassindale, A. R.; Taylor, P. G. In The
Chemistry of Organic Silicon Compounds—Chapter 14:
Activating and Directive Effects of Silicon; Patai, S.,
Rappoport, Z., Eds.; Wiley: New York, 1989; p 893.
20. Foropoulos, J.; DesMarteau, D. D. Inorg. Chem. 1984, 23,
3720–3723.
5. Ivanov, S. V.; Rockwell, J. J.; Polyakov, O. G.; Gaudinski,
C. M.; Anderson, O. P.; Solntsev, K. A.; Strauss, S. H. J. Am.
Chem. Soc. 1998, 120, 4224–4225.
6. Lambert, J. B.; Zhang, S.; Ciro, S. M. Organometallics 1994,
13, 2430–2443.
21. Michot, C. (Rhodia-Lyon) Unpublished results. Donnor
numbers measured from dissociation of [Cu(acac)(tmen)X]
(X¼OTf, NTf2) in 1,2-dichloroethane.
7. Vij, A.; Zheng, Y. Y.; Kirchmeier, R. L.; Shreeve, J. M. Inorg.
Chem. 1994, 33, 3281–3288.
22. Childs, R. F.; Mulholland, D. L.; Nixon, A. Can. J. Chem.
1982, 60, 801–808.
8. Trehan, A.; Vij, A.; Walia, M.; Kaur, G.; Verma, R. D.;
Trehan, S. Tetrahedron Lett. 1993, 34, 7335–7338.
9. Olah, G. A.; Laali, K.; Farooq, O. Organometallics 1984, 3,
1337–1340.
23. Laszlo, P.; Teston, M. J. Am. Chem. Soc. 1990, 112,
8750–8754.
24. A similar study was realized with trityl perchlorate, see:
Kobayashi, S.; Sagawa, Y.; Akamatsu, H.; Mukaiyama, T.
Chem. Lett. 1988, 1777–1780.
10. Marsmann, H. C.; Horn, H. G. Z. Naturforsch. 1972, 27b,
1448–1451.
11. (a) Handy, S. T.; Grieco, P. A.; Mineur, C.; Ghosez, L. Synlett