Development of cyclic hydrazine and hydrazide type organocatalyst-mechanistic aspects of cyclic hydrazine/hydrazide-catalyzed diels-alder reactions

Article abstract of DOI:10.3987/COM-08-S(D)54

Some hydrazines and hydrazides were prepared and screened for their catalytic efficiencies in Diels-Alder reactions. ¹H-NMR studies and ab initio calculations revealed that catalytic efficiencies of these catalysts are greatly dependent on the release of the catalysts from the Diels-Alder adducts.

Full text of DOI:10.3987/COM-08-S(D)54

HETEROCYCLES, Vol. 79, 2009
851
HETEROCYCLES, Vol. 79, 2009, pp. 851 - 863. © The Japan Institute of Heterocyclic Chemistry
Received, 30th September, 2008, Accepted, 28th November, 2008, Published online, 3rd December, 2008.
DOI: 10.3987/COM-08-S(D)54
DEVELOPMENT OF CYCLIC HYDRAZINE AND HYDRAZIDE TYPE
ORGANOCATALYSTMECHANISTIC ASPECTS OF CYCLIC
HYDRAZINE/HYDRAZIDE –CATALYZED DIELS-ALDER REACTIONS
Ichiro Suzuki,* Ai Hirata, and Kei Takeda
Department of Synthetic Organic Chemistry, Graduate School of Biomedical
Sciences, Hiroshima University, 1-2-3, Kasumi, Minami-ku, Hiroshima, 734-8553
Abstract – Some hydrazines and hydrazides were prepared and screened for their
catalytic efficiencies in Diels-Alder reactions. 1H-NMR studies and ab initio
calculations revealed that catalytic efficiencies of these catalysts are greatly
dependent on the release of the catalysts from the Diels-Alder adducts.
INTRODUCTION
In the early 1970s, two industrial research groups reported the first examples of (S)-proline-catalyzed
enantioselective intramolecular aldol reactions. These reactions, so-called
Hajos–Parrish–Eder–Sauer–Wiechert reactions, attracted little attention for about 30 years despite their
potential values.1 In the early 2000s, List and coworkers reported (S)-proline-catalyzed intermolecular
asymmetric aldol reactions.2 Thereafter, as well as proline, related five-membered cyclic secondary
amine derivatives have also been applied to many reactions that proceed through enamine or iminium ion
intermediates.3 It is well established that a nitrogen atom that is directly bonded to an atom with one or
more lone pairs, such as H2N-NH2 and H2N-OH, tends to be a stronger nucleophile than would otherwise
be expected. The nucleophic enhancement is called “α-heteroatom effect”,4 and such α-nucleophiles
seem to be more advantageous as nitrogen-based heterocyclic organocatalysts than simple secondary
amine catalysts, but there are not many reported examples. This type of organocatalyst was first
reported to promote Diels-Alder reactions efficiently via iminium ions in 2003,5 and camphor-based
chiral hydrazide catalysts were also reported.6 Recently, we have investigated Biginelli reaction
catalyzed by cyclic hydrazine-type organocatalysts as shown in Figure 1, and we have reported that
pyrazolidine dihydrochloride 1a could catalyze Biginelli reactions very efficiently under mild conditions.7
These results promoted us to screen several cyclic hydrazines and hydrazides for their catalytic activities

852
HETEROCYCLES, Vol. 79, 2009
in other reactions, including Diels-Alder reactions; however, during these studies, we became aware of
ambiguities in the catalytic activities of the hydrazine-type organocatalysts.
O
NH•HCl
NH•HCl
NCbz
NAc
NH•HCl
NH•HCl
NH
NH•HCl
NH•HCl
NH•HCl
H3C
1b
1c
1d
1e
1a
Figure 1. List of hydrazines and hydrazides employed in this study.
For example, N-Cbz pyrazolidine hydrochloride 1c accelerated Diels-Alder reactions more efficiently
than did catalyst 1a, which showed a significantly higher level of catalytic activity than did 1c in Biginelli
reactions as shown in Scheme 1.
OMe
EtO2C
CHO
O
catalysts
MeOH
O
O
Me
1a: 98% (rt, 4 h);
1c: NR
+
+
(a)
H2N
NH2
EtO
Me
HN
NH
MeO
Ph
O
Ph
CH(OMe)2
catalysts
MeOH
1a: 28% (rt, 24 h);
1c: 70% (rt, 11 h)
CHO
+
(b)
Scheme 1. Biginelli reactions (a) and Diels-Alder reactions (b) catalyzed by 1a and 1c.
In order to develop chiral hydrazine-type organocatalysts superior to conventional pyrrolidine-based
organocatalysts, it is important to elucidate these ambiguities, and we have therefore carried out the
mechanistic studies on hydrazine/hydrazide-catalyzed Biginelli and Diels-Alder reactions. In this paper,
we present the results of experiments and ab initio calculations on hydrazine- and hydrazide-catalyzed
Diels-Alder reactions. These results showed that not the formation of reactive hydrazonium ions but the
regeneration of hydrazine/hydrazide catalysts was more important in the catalyzed Diels-Alder reactions.
RESULTS AND DISCUSSION
All hydrazines 1a, 1b and hydrazides 1c – 1e were prepared by methods reported in the literature8-11 as
depicted in Scheme 2. For catalysts 1a and 1b, we first tried to isolate them in acid-free forms, but all
attempts failed because they are prone to immediately suffer from air oxidation under neutral or basic
conditions to give only polymeric materials. In contrast to 1a and 1b, hydrazides 1c – 1e were relatively
stable in acid-free forms when stocked in a refrigerator for a few weeks.

HETEROCYCLES, Vol. 79, 2009
853
(iii)
(i)
(ii)
n
n
BocHN NHBoc
H2NNH2•H2O
BocN NBoc
HN NH•2HCl
2
4a (n=1): 85%
4b (n=2): 76%
1a (n=1): 98%
1b (n=2): 94%
(iv)
76%
(vii)
56%
(iii)
(ii)
CbzHN NHBoc
95%
85%
CbzN NBoc
CbzN NH•HCl
O
3
5
1c
NH
NH
(v)
82%
H3C
(vi), (iii)
73%
1e
AcN NH•HCl
HN NBoc
6
1d
Reagents and conditions: (i) (Boc)2O (2.1 equiv.)/MeOH, rt; (ii) NaH (2.2 equiv.), Br(CH2)3Br (1.2
equiv.) for 4a and 5, Br(CH2)4Br (1.2 equiv.) for 4b; (iii) 4M HCl/1,4-dioxane, rt; (iv) (Boc)2O (0.2
equiv.)/IPA, rt, then CbzCl (1.2 equiv.), 4-Methylmorpholine (1.5 equiv.)/THF, 0 ˚C to rt; (v) H2 baloon,
10% Pd-C/MeOH; (vi)AcCl (1.2 equiv.), Et3N (1.5 equiv.)/DCM, 0 ˚C to rt; (vii) crotonic acid (1.2
equiv.), Al2O3/n-BuOH-toluene.
Scheme 2. Preparation of hydrazine and hydrazide catalysts 1a – 1e.
Diels-Alder reactions of cinnamaldehyde and cyclopentadiene were performed in the presence of 10
mol% of catalysts 1a – 1e at room temperature, and the results are summarized in Table 1. When the
reaction was carried out in the presence of 1a, the reaction was very sluggish, affording an inseparable
mixture of endo/exo isomers of dimethylacetals (endo/exo)-712 in 25% yield after 24 h (entry 1). The
yield was slightly improved when the reaction time was extended to 48 h (entry 2). Catalyst 1b showed
a level of catalytic activity similar to that of catalyst 1a (entries 3 and 4). On the other hand, N-Cbz
pyrazolidine hydrochloride 1c showed a higher level of catalytic activity and the reaction was completed
within 11 h, giving acetals (endo/exo)-7 in 70% yield with 6% aldehydes (endo/exo)-813 (entry 5). The
reaction proceeded in an endo selective manner in the presence of catalyst 1c, though catalysts 1a and 1b
accelerated the reactions exo-selectively. These reactions were constituted by three reversible processes:
(1) hydrazonium ion formation, (2) Diels-Alder reaction, and (3) solvolysis of Diels-Alder adducts.
This mechanistic complexity make it difficult to clearly explain the obtained endo/exo ratios by simple
kinetic and/or thermodynamic reasons. Interestingly, N-acetyl pyrazolidine hydrochloride 1d is much
less effective than 1c, and the reaction did not complete even after 46 h, with acetals 7 and aldehydes 8
being obtained in 63% and 6% yields, respectively accompanied by 29% recovery of cinnamaldehyde
(entry 6). Hydrazides 1e could also catalyze the reaction, but the catalytic efficiency of 1e was
unexpectedly low, giving Diels-Alder adducts in a total 32% yield even after 48 h (entries 7 and 8).

854
HETEROCYCLES, Vol. 79, 2009
Table 1. Diels-Alder reactions in the presence of cataltsts 1a – 1e.
Ph
catalyst 1a - 1e
CHO
+
+
X
Ph
X
Ph
X = CH(OMe)2: endo-7;
X = CHO: endo-8
X = CH(OMe)2: exo-7;
X = CHO: exo-8
yield (%)
recovery
(%)
66
56
63
64
-
entry catalyst
solvent
time (h)
endo:exo
7
25
30
28
30
70
63
30
36
14
35
60
28
-
8
1
2
1a
1a
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
24
48
24
48
11
46
28
47
trace
81:19a
78:22a
79:21a
77:23a
33:67a
40:60a
63:37a
57:43a
70:30a
29:71a
34:66a
41:59a
52:48b
34:66b
52:48b
63:37a
66:34a
55:45b
2
trace
2
3
1b
4
1b
5
1c
6
6
1d
6
29
52
35
47
36
15
55
44
-
7
1e
2
8
1e
2
9
1a
MeOH-H2O (9:1)
MeOH-H2O (9:1)
MeOH-H2O (9:1)
MeOH-H2O (9:1)
H2O
48
9
trace
5
10
11
12
13
14
15
16
17
18
1c
1c
49
48
45
48
63
24
24
24
6
1d
2
1a
18
73
12
-
1c
H2O
-
1d
H2O
-
63
64
62
68
none
none
none
MeOH
6
MeOH-H2O (9:1)
H2O
7
-
-
2
a) Ratios were determined for adduct 7 by 1H-NMR. b) Ratios were determined for adduct 8 by 1H-NMR.
Since aqueous MeOH and water were employed for the reaction solvent in reported examples,5, 6 we also
examined the reactions in these solvents. When the reactions were carried out in MeOH containing 10
v/v% of water, the reactions became inhomogeneous and very sluggish and acetals 7 were also obtained
(entries 9 – 12). For instance, after 49 h, the starting aldehyde did not disappear even in the reaction

HETEROCYCLES, Vol. 79, 2009
855
using the most effective catalyst 1c (entry 11). Furthermore, when water was used as a reaction solvent,
catalysts worked less efficiently as in MeOH, giving miserable results except for 1c (entries 13 – 15).
In addition, Diels-Alder reactions of cinnamaldehyde and cyclopentadiene in MeOH, MeOH-H2O (9:1)
and water were carried out in the absence of catalyst. In these reactions, Diels-Alder adducts were
obtained in only low yields (entries 16 – 18). From these results, it became clear that catalytic efficiency
of 1a - 1e is in the following order in MeOH;
O
NH•HCl
NH•HCl
>
>
!
n
CbzN NH•HCl
AcN NH•HCl
ClH•HN NH
1c
1d
1e
1a or 1b
These catalysts would be in equilibria with acid-free forms in solutions and the catalysts should work as
acid-free forms. In considering catalytic activities of these catalysts, it is reasonable to assume that
pyrazolidine is more nucleophilic than N-acylpyrazolidines and pyrazolidinones and catalyst 1c therefore
shows a higher level of catalytic activities than do other catalysts; however, the experimentally obtained
order did not reflect this expected order.14 Then, we investigated the reasons why more nucleophilic
pyrazolidine showed only limited catalytic activity in Diels-Alder reactions by using ab initio calculations
and 1H-NMR techniques.
Diels-Alder reaction is thought to include three processes as shown in Figure 2: (1) hydrazonium ion
formation from a catalyst hydrazine and an aldehyde, (2) cycloaddition of diene to the hydrazonium ion,
and (3) hydrolysis of the resulting hydrazonium ion to release hydrazine.
Y
X
N
2nd step
1st step
R
CHO
R
Ph
N
X
X = NH, NAc, NCbz;
Y = CH2, C=O
NH
Y
X
Ph
CHO
Y
3rd step
Figure 2. Catalytic cycle of hydrazine/hydrazide-catalyzed Diels-Alder reactions.

856
HETEROCYCLES, Vol. 79, 2009
In a cycloadditon step, the second step, Diels-Alder reaction is generally considered to be an
orbital-controlled reaction, and therefore the facility of the reaction must be dependent on LUMO levels
of the hydrazonium ions. We therefore calculated LUMO levels of cinnamaldehyde, iminium ion 9, and
hydrazonium ions 10a – 10e.15 To avoid conformational ambiguity, we employed 10c and 10e as
computational models. For hydrazonium ions 10a – 10e, both E and Z isomers were calculated. All
geometries were optimized at the B3LYP/6-31+G(d) level of theory, followed by calculations of LUMO
levels and energies at the B3LYP/6-311+G(d, p) level16, and the results are shown in Table 2.
Table 2. Ab initio calculations of LUMO levels by B3LYP/6-311+G(d, p)//B3LYP/6-31+G(d).
X
N
X
N
Y
Y
9
(Z)-10a
(Z)-10b
(Z)-10c
(Z)-10d
(Z)-10e
X = Y = CH2
9
(E)-10a
(E)-10b
(E)-10c
(E)-10d
(E)-10e
X = Y = CH2
X = NH•HCl, Y = CH2
X = NH•HCl, Y =(CH2)2
X = NCO2Me, Y = CH2
X = NAc, Y = CH2
X = NH•HCl, Y = CH2
X = NH•HCl, Y =(CH2)2
X = NCO2Me, Y = CH2
X = NAc, Y = CH2
X = NH, Y = C=O
X = NH, Y = C=O
entry
1
Dienophile
cinnamaldehyde
iminium ion 9
(E)-10a
Energy (Hartrees)
-423.0925709
LUMO (eV)
-2.55
2
-559.687863
-6.55
3
-575.697198
-6.55
4
(Z)-10a
-575.698419 (-0.77)a
-6.48
5
(E)-10b
-615.025357
-6.53
6
(Z)-10b
-615.02777 (-1.5)a
-803.643674
-6.49
7
(E)-10c
-6.39
8
(Z)-10c
-803.641975 (+1.1)a
-6.40
9
(E)-10d
-728.394822
-6.39
10
11
12
(Z)-10d
-728.392926 (+1.2)a
-649.737450
-6.42
(E)-10e
-6.80
(Z)-10e
-649.739326 (-1.2)a
-6.81
a) Values in parentheses indicate the ΔE values (EZ-isomer – EE-isomer) in kcal/mol
For hydrazonium ions 10a, 10b and 10e, Z isomers were more stable than E isomers by 0.77 kcal/mol, 1.5
kcal/mol and 1.2 kcal/mol, respectively (entries 3 – 6, 11 and 12). On the other hand, for 10c and 10d, E
isomers were more stable than Z isomers by 1.1 kcal/mol and 1.2 kcal/mol, respectively (entries 7 – 10).
Since LUMO energy levels of E and Z isomers for 10a – 10e were essentially the same, 17 the preference

HETEROCYCLES, Vol. 79, 2009
857
for E or Z isomer should not affect the net reactivity of each type of hydrazonium ions. The LUMO
level of cinnamaldehyde was -2.55 eV and the levels of iminium ion 9 and hydrazonium ions 10a – 10e
were in the range of -6.81 to -6.39 eV. These results indicate that hydrazonium ions 10a – 10e are more
active for Diels-Alder reactions than is the parent aldehyde, as well as iminium ion 9; however, we could
not find out any remarkable differences in LUMO energy levels that are consistent with catalytic
activities of 10a – 10e.
Since we could not rationalize the tendency of the reactivity of catalysts 1a – 1e only by LUMO levels of
intermediate hydrazonium ions, we next investigated the formation of hydrazonium ions (the first step in
Figure 2). Cinnamaldehyde was treated with 1 equivalent of catalyst 1a – 1e in CD3OD, and the
1
reactions were monitored by H-NMR, and the results are summarized in Table 3. All reactions
proceeded to give hydrazonium ions 10a – 10e and/or acetal 11 without any by-products.
Table 3. 1H-NMR study for firmation of hydrazonium ions 10a – 10e.
HZ
HE
H
OCD3
Z
X
N
X
N
Z
OCD3
6.14 ppm
Y
H
Y
11
(Z)-10a
(Z)-10b
(Z)-10c
(Z)-10d
(Z)-10e
X = NH•HCl, Y = CH2, Z = CH2
X = NH•HCl, Y =(CH2)2, Z = CH2
X = NCO2Bn, Y = CH2, Z = CH2
X = NAc, Y = CH2, Z = CH2
(E)-10a
(E)-10b
(E)-10c
(E)-10d
(E)-10e
X = NH•HCl, Y = CH2, Z = CH2
X = NH•HCl, Y =(CH2)2, Z = CH2
X = NCO2Bn, Y = CH2, Z = CH2
X = NAc, Y = CH2, Z = CH2
X = NH, Y = C=O, Z = CHCH3
X = NH, Y = C=O, Z = CHCH3
HZ/HE
ratioa (%)
entry Hydrazonium ion
10 (E/Z)b
11
0
Cinnamaldehyde
(ppm)
1
2
3
4
5
10a
10b
10c
10d
10e
8.22/8.36
8.47/8.50
8.97/9.13
8.90/9.18
100 (24:76)
73 (15:85)
59 (67:33)
21 (71:29)
61 (31:69)
-
4
23
33
68
35
8
11
4
8.58/9.44
1
a) ratios were determined by H-NMR. b) Stereochemistry of hydrazonium ions were assigned on the
basis of ab initio calculations.
When catalyst 1a was reacted with cinnamaldehyde, exclusive formation of hydrazonium ion 10a was
observed within 10 min at 25 ˚C, giving a mixture of E- and Z-isomers (24:76). On the other hand, the
reaction of catalyst 1b and 1c with cinnamaldehyde gave equilibrium mixtures containing hydrazonium
ion 10b (73%), 10c (59%) and acetal 11 within 10 min. Additionally, in the reaction of catalyst 1d and

858
HETEROCYCLES, Vol. 79, 2009
the aldehyde, hydrazonium ions 10d and acetal 11 were formed similarly; however, the equilibrium
between hydrazonium ions and the acetal was shifted to the acetal formation and the yield of
hydrazonium ions 10d was reduced to 21% with increasing the formation of acetal 11 to 68% yield.
While we could not observe noticeable differences in the rates of hydrazonium ion formations in our
1H-NMR experiments, it is noteworthy that the ratios of hydrazonium ions to the acetal 11 changed from
64:36 (for 10c) to 24:76 (for 10d). These results can rationalize the observed catalytic activities of
catalysts 10c and 10d. In the catalyzed reactions, the formation of hydrazonium ions, which are reactive
intermediates, is essential, and catalyst 1c yielded much more hydrazonium ion 10c than did 1d, being
reflected in the higher level of catalytic activity of 1c than that of 1d. On the other hand, although the
reaction of cinnamaldehyde and catalyst 1a afforded reactive intermediary hydrazonium ions 10a
exclusively, 1a catalyzed the Diels-Alder reactions less efficiently than did catalyst 1c. This result
implies that the formation of hydrazonium ions is not necessarily a rate-determining step. In this case,
we must additionally consider release of a catalyst 1a from Diels-Alder adducts (Figure 2, the third step).
In this step, hydrazine/hydrazide moieties work as leaving groups, and hence pKa values of their
conjugate acids should be taken into account.14, 18 Since acyl hydrazines, such as catalyst 1c, are
considered to be less basic than hydrazines 1a and 1b by 4~5 pka units, it is reasonable that catalysts 1c
worked as a leaving group more efficiently than did 1a and 1b, leading to enhanced net catalytic
efficiency. Although catalyst 1e gave hydrazonium ion 10e and acetal 11 in a ratio similar to that of 10c
and their acidities are considered to be almost same, catalytic efficiency of 1e was considerably inferior to
catalyst 1c. It is plausible that the steric repulsion between the alkene part and the acyl moiety in
hydrazonium ion 10c might be more severe than the steric repulsion in hydrazonium ion 10e as shown in
Figure 3. This severe steric repulsion would facilitate the hydrolytic release of 1c from hydrazonium ion
10c.
Ph
Ph
Ph
Ph
H
H
N
N
CH3
H3C
MeOH
MeOH
N
N
HN
O
NH
O
CO2Bn
N
BnO2C N
10e
10c
Figure 3. Steric repulsion between substituents in hydrazonium ion 10c and 10e.
In conclusion, we investigated the mechanistic features of hydrazine/hydrazide-catalyzed Diels-Alder
reactions. 1H-NMR studies and ab initio calculations revealed that catalytic efficiencies of these
catalysts are greatly dependent on the release of the catalysts from the Diels-Alder adducts.

HETEROCYCLES, Vol. 79, 2009
859
EXPERIMENTAL
General Remarks
1H-NMR spectra were measured in CDCl3, DMSO-d6 or CD3OD solutions and referenced to CHCl3 (7.26
ppm), DMSO-d6 (2.54 ppm) and CD3OD (3.30 ppm) using JEOL Delta-500 (500 MHz) spectrometer.
13C-NMR spectra were measured in CDCl3, DMSO-d6 or CD3OD solutions and referenced to CDCl3 (77.0
ppm), DMSO-d6 (39.7 ppm) and CD3OD (49.0 ppm) using JEOL Delta-500 (125 MHz) spectrometers.
Column chromatography was performed on silicagel, KANTO KAGAKU N-60.
Thin-layer
chromatography was performed on precoated plates (0.25 mm, silicagel Merck Kieselgel 60 F254). All
solvents were distilled prior to use. All reactions were performed in oven-dried glassware under positive
pressure of nitrogen, unless otherwise noted.
Pyrazolidine dihydrochloride (1a).8
1H-NMR (500 MHz, DMSO-d6) δ 3.06 (4H, t, J = 7.4 Hz), 1.98 (2H, quint, J = 7.4 Hz); C-NMR (125
13
MHz, DMSO-d6) δ 46.5, 26.1.
1,2-Piperazine dihydrochloride (1b).9
1H-NMR (500 MHz, DMSO-d6) δ 2.99 (4H, m), 1.67 (4H, m); C-NMR (125 MHz, DMSO-d6) δ 45.2,
13
21.7.
Benzyl pyrazolidine-1-carboxylate hydrochloride (1c).10
1H-NMR (500 MHz, CD3OD) δ 7.50-7.30 (5H, m), 5.28 (2H, s), 3.76 (2H, t, J = 6.9 Hz), 3.59 (2H, t, J =
13
6.9 Hz), 2.36 (2H, quint, J = 6.9 Hz); C-NMR (125 MHz, CD3OD) δ 153.6, 135.6 and 135.2, 128.5,
128.4, 128.3, 128.2, 68.9 and 68.0, 46.4 and 46.2, 24.3.
N-Acetyl Pyrazolidine hydrochloride (1d).10
1H-NMR (500 MHz, CD3OD) δ 3.90 (2H, t, J = 6.5 Hz), 3.57 (2H, t, J = 6.5 Hz), 2.50-2.40 (2H, m); 2.19
(3H, s); 13C-NMR (125 MHz, CD3OD) δ 168.6, 46.6 and 46.3, 45.4, 25.6 and 25.0, 19.9.
5-Methylpyrazolidin-3-one (1e).11
1H-NMR (500 MHz, DMSO-d6) δ 4.11 (1H, ddd, J = 6.9, 8.0 and 8.7 Hz), 3.59 (2H, brs), 2.70 (1H, dd, J
13
= 8.0 and 16.7 Hz), 2.34 (1H, dd, J = 8.7 and 16.7 Hz), 1.38 (3H, d, J = 6.9 Hz); C-NMR (125 MHz,
CD3OD) δ 173.8, 55.3, 35.3, 15.6.

860
HETEROCYCLES, Vol. 79, 2009
General procedure for catalyzed Diels-Alder reaction
To a solution of (E)-cinnamaldehyde (126 µL, 1.0 mmol) and cyclopentadiene (247 µL, 3.0 mmol),
which was distilled by cracking dicyclopentadiene at 180 ˚C prior to use, in distilled MeOH (333 µL) was
added N-Cbz pyrazolidine hydrochloride 1c (24 mg, 0.1 mmol), and the resulting mixture was stirred at rt
until the reaction was judged to be completed by TLC analysis. The reaction was quenched with 1N
HCl and the mixture was extracted twice with Et2O. The combined organic layers were washed
successively with saturated aqueous solution of NaHCO3 and brine, and dried over MgSO4. After
evaporation, the residue was purified with silica gel chromatography (5% AcOEt in hexane was used as
an eluent) to give the desired adducts as a mixture of acetals 7 (colorless oil, 171 mg, 70%, endo/exo =
33:67) and aldehydes 8 (colorless oil, 12 mg, 6%).
5-(Dimethoxymethyl)-6-phenylbicyclo[2.2.1]hept-2-ene (endo-7).12
1H-NMR (500 MHz, CDCl3) δ 7.37-7.12 (5H, m), 6.34 (1H, m), 6.15 ( 1H, dd, J = 2.8 and 5.7 Hz), 3.93
(1H, d, J = 9.2 Hz), 3.37 (3H, s), 3.13 (3H, s), 2.96 (1H, brs), 2.87 (1H, brs), 2.54 (1H, m), 2.40 (1H, dd, J
= 1.5 and 4.8 Hz), 1.77 (1H, d, J = 8.7 Hz), 1.55 (1H, m);13C-NMR (125 MHZ, CDCl3) δ 145.1, 138.5,
135.1, 128.3, 127.9, 125.7, 108.3, 52.7, 52.4, 49.4, 46.7, 47.3, 44.5.
5-(Dimethoxymethyl)-6-phenylbicyclo[2.2.1]hept-2-ene (exo-7).12
1H-NMR (500 MHz, CDCl3) δ 7.37-7.12 (5H, m), 6.35 (1H, m), 5.94 (1H, dd, J = 2.8 and 5.7 Hz), 4.36
(1H, d, J = 8.3 Hz), 3.38 (3H, s), 3.12 (1H, dd, J = 3.4 and 5.1 Hz), 3.07 (3H, s), 3.00 (1H, brs), 2.90 (1H,
dd, J = 1.4 and 4.9 Hz), 2.03 (1H, ddd, J = 1.6, 5.1 and 8.2 Hz), 1.68 (1H, d, J = 8.6 Hz), 1.49 (1H, m);
13C-NMR (125 MHZ, CDCl3) δ 144.3, 137.4, 135.3, 128.1, 127.9, 125.9, 108.0, 53.5, 52.4, 49.4, 49.1,
47.6, 45.3.
3-Phenylbicyclo[2.2.1]hept-5-ene-2-carboxaldehyde (endo-8).13
1H-NMR (500 MHz, CDCl3) δ 9.60 (1H, d, J = 2.3 Hz), 7.13–7.34 (5H, m), 6.35 (1H, dd, J = 3.2 and 5.5
Hz), 6.10 (1H, dd, J = 3.2 and 5.8 Hz), 3.26 (1H, brs), 3.06 (1H, brs), 3.02 (1H, dd, J = 1.6 and 4.9 Hz),
13
2.91 (1H, ddd, J = 2.3, 3.5 and 4.9 Hz), 1.74 (1H, m), 1.62 (1H, m); C-NMR (125 MHZ, CDCl3) δ
203.6, 143.7, 139.3, 133.9, 128.7, 127.5, 126.3, 60.9, 48.5, 47.2, 45.8, 45.3.
3-Phenylbicyclo[2.2.1]hept-5-ene-2-carboxaldehyde (exo-8).13
1H-NMR (500 MHz, CDCl3) δ 9.85 (1H, d, J = 2.1 Hz), 7.13–7.34 (5H, m), 6.27 (1H, dd, J = 3.5 and 5.8
Hz), 6.00 (1H, dd, J = 3.5 and 5.8 Hz), 3.65 (1H, dd, J = 3.5 and 4.9 Hz), 3.15 (2H, m), 2.52 (1H, ddd, J =
2.3, 3.5 and 4.9 Hz), 1.53–1.59 (2H, m);13C-NMR (125 MHZ, CDCl3) δ 202.9, 143.7, 136.6, 136.4, 128.2,
128.0, 126.4, 59.5, 48.5, 47.7, 45.59, 45.56.

HETEROCYCLES, Vol. 79, 2009
861
General procedure for 1H-NMR experiments
A solution of cinnamaldehyde (15 µL, 0.12 mmol) in CD3OD (0.4 mL) was mixed with pyrazolidine
dihydrochloride 1a (17 mg, 0.12 mmol) in an NMR tube at rt. The reaction was monitor by 1H-NMR .
Hydrazonium ion 10a.
Major isomer: 1H-NMR (500 MHz, CD3OD) δ 8.22 (1H, d, J = 10.3 Hz), 7.80-7.70 (2H, m), 7.65 (1H, d,
J = 15.6 Hz), 7.50-7.40 (3H, m), 7.37 (1H, dd, J = 10.3 and 15.6 Hz), 4.38 (2H, m), 3.60 (2H, m),
1
2.50-2.30 (2H, m); minor isomer: H-NMR (500 MHz, CD3OD) δ 8.36 (1H, d, J = 10.6 Hz), 7.80-7.70
(2H, m), 7.60 (1H, d, J = 15.4 Hz), 7.50-7.40 (3H, m), 7.24 (1H, dd, J = 10.6 and 15.4 Hz), 4.38 (2H, m),
3.51 (2H, m), 2.50-2.30 (2H, m).
Hydrazonium ion 10b.
1
Major isomer: H-NMR (500 MHz, CD3OD) δ 8.47 (1H, d, J = 10.1 Hz), 8.00-7.20 (7H, m), 4.12 (2H,
1
m), 3.29 (2H, m), 2.05 (2H, m), 1.91 (2H, m); minor isomer: H-NMR (500 MHz, CD3OD) δ 8.50 (1H, d,
J = 10.8 Hz), 8.00-7.20 (7H, m), 4.26 (2H, m), 3.15 (2H, m), 1.80 (4H, m).
Hydrazonium ion 10c.
1
Major isomer: H-NMR (500 MHz, CD3OD) δ 9.13 (1H, d, J = 10.3 Hz), 8.00-7.20 (6H, m), 7.24 (1H,
dd, J = 10.3 and 17.9 Hz), 5.24 (2H, s), 4.53 (2H, m), 4.07 (2H, m), 2.70-2.50 (2H, m); minor isomer:
1H-NMR (500 MHz, CD3OD) δ 8.97 (1H, d, J = 10.8 Hz), 8.40-7.20 (7H, m), 7.08 (1H, dd, J = 10.8 Hz,
and 17.9 Hz), 5.30 (2H, s), 4.39 (2H, m), 4.07 (2H, m), 2.70-2.50 (2H, m).
Hydrazonium ion 10d.
1
Major isomer: H-NMR (500 MHz, CD3OD) δ 9.18 (1H, d, J = 10.8 Hz), 8.40-7.20 (7H, m), 4.51 (2H,
1
m), 4.18 (2H, m), 2.60-2.30 (2H, m), 2.34 (3H, s); minor isomer: H-NMR (500 MHz, CD3OD) δ 8.90
(1H, d, J = 10.3 Hz), 8.40-7.20 (7H, m), 4.37 (2H, m), 4.33 (2H, m), 2.60-2.30 (2H, m), 2.37 (3H, s).
Hydrazonium ion 10e.
1
Major isomer: H-NMR (500 MHz, CD3OD) δ 9.44 (1H, d, J = 10.1 Hz), 7.90-7.20 (7H, m), 5.00 (1H,
ddd, J = 5.0, 6.9 and 9.0 Hz), 3.43 (1H, dd, J = 9.0 and 17.5 Hz), 2.89 (1H, dd, J = 5.0 and 17.5 Hz), 1.69
1
(3H, d, J = 6.9 Hz); minor isomer: H-NMR (500 MHz, CD3OD) δ 8.58 (1H, d, J = 11.0 Hz), 7.90-7.20
(7H, m), 5.38 (1H, ddd, J = 2.1, 6.7 and 9.0 Hz), 3.53 (1H, dd, J = 9.0 and 17.2 Hz), 2.84 (1H, dd, J = 2.1
and 17.2 Hz), 1.61 (3H, d, J = 6.7 Hz)

862
HETEROCYCLES, Vol. 79, 2009
Cinnamaldehyde dimethylacetal-d6 (11).
1H-NMR (500 MHz, CD3OD) δ 7.60-7.20 (5H, m), 6.71 (1H, d, J = 16.1 Hz), 6.14 (1H, dd, J = 5.1 and
16.1 Hz), 4.92 (1H, d, J = 5.1 Hz).
REFERENCES AND NOTES
1. E. Eder, G. Sauer, and R. Wiechert, Angew. Chem., Int. Ed. Engl., 1971, 10, 496; Z. G. Hajos and D.
R. Parrish, J. Org. Chem., 1974, 39, 1615.
2. B. List, R. A. Lerner, and C. F. Barbas, III, J. Am. Chem. Soc., 2000, 122, 2395; W. Notz and B. List,
J. Am. Chem. Soc., 2000, 122, 7386; B. List, P. Pojarliev, and C. Castello, Org. Lett., 2001, 3, 573.
3. A. Dondoni and A. Massi, Angew. Chem. Int. Ed., 2008, 47, 4638; B. List, Chem. Commun., 2006,
819; K. A. Ahrendt, C. J. Borths, and D. W. C. MacMillan, J. Am. Chem. Soc., 2000, 122, 4243; .
4. N. J. Fina and J. O. Edwards, Int. J. Chem. Kinet., 1973, 5, 1; A. P. Grekov and V. Y. Veselov, Russ.
Chem. Rev., 1978, 47, 631; S. Hoz and E. Buncel, Israel J. Chem., 1985, 26, 313.
5. J. L. Cavill, J.–U. Peters, and N. C. O. Tomkinson, Chem. Commun., 2003, 728.
6. M. Lemay and W. W. Ogilvie, Org. Lett., 2005, 7, 4141; M. Lemay and W. W. Ogilvie, J. Org.
Chem., 2006, 71, 4664; H. He, B.–J. Pei, H.–H. Chou, T. Tian, W.–H. Chan, and A. W. M. Lee, Org.
Lett., 2008, 10, 2421.
7. I. Suzuki, Y. Iwata, and K. Takeda, Tetrahedron Lett., 2008, 49, 3238.
8. E. E. Boros, F. Bouvier, S. Randhawa, and M. H. Rabinowitz, J. Heterocycl. Chem., 2001, 38, 613.
9. D. E. Wilkinson, B. E. Thomas, D. C. Limburg, A. Holmes, H. Sauer, D. T. Ross, R. Soni, Y. Chen,
H. Guo, P. Howorth, H. Valentine, D. Spicer, M. Fuller, J. P. Steiner, G. S. Hamilton, and Y.–Q. Wu,
Bioorg. Med. Chem., 2003, 11, 4815.
10. J. H. Ahn, J. A. Kim, H.–M. Kim, H.–M. Kwon, S.–C. Huh, S. D. Rhee, K. R. Kim, S.–D. Yang,
S.–D. Park, J. M. Lee, S. S. Kim, and H. G. Cheon, Bioorg. Med. Chem. Lett., 2005, 15, 1337; W.–J.
Zhang, A. Berglund, J. L.–F. Kao, J.–P. Couty, M. G. Gershengorn, and G. R. Marshall, J. Am.
Chem. Soc., 2003, 125, 1221; R. E. Melendez and W. D. Lubell, J. Am. Chem. Soc., 2004, 126,
6759.
11. S. T. Perri, S. C. Slater, S. G. Toske, and J. D. White, J. Org. Chem., 1990, 55, 6037.
12. B. F. Bonini, E. Capitò, M. Comes-Franchini, M. Fochi, A. Ricci, and B. Zwanenburg, Tetrahedron:
Asymmetry, 2006, 17, 3135.
13. K. Ishihara, H. Kurihara, M. Matsumoto, and H. Yamamoto, J. Am. Chem. Soc., 1998, 120, 6920.
14. C. K. M. Heo and J. W. Bunting, J. Chem. Soc., Perkin Trans. 2, 1994, 2279.
15. All calculations were performed using Spartan ’04 Macintosh, ver 1.03.

HETEROCYCLES, Vol. 79, 2009
863
16. C. Lee, W. Yang, and R. G. Parr, Phys. Rev. B, 1988, 37, 785; A. D. Becke, Phys. Rev. A, 1988, 38,
3098; A. D. Becke, J. Chem. Phys., 1992, 96, 2155; A. D. Becke, J. Chem. Phys., 1992, 97, 9173; A.
D. Becke, J. Chem. Phys., 1993, 98, 5648.
17. From calculations of the geometries of hydrazonium ions 10c and 10d, it was suggested that nitrogen
atom that connected to an acyl group is considerably pyramidalized to avoid the steric repulsion
between the alkene and the acyl group. By this structural feature, an N-acyl group would not work
well as an electron-withdrawing group.
18. R. L. Hinman, J. Org. Chem., 1958, 23, 1587; C. R. Lindegren and C. Niemann, J. Am. Chem. Soc.,
1949, 71, 1504.