Novel tetrahydroisoquinoline based organocatalysts for asymmetric Diels-Alder reactions: Insight into the catalytic mode using ROESY NMR and DFT studies

Article abstract of DOI:10.1016/j.tetasy.2010.11.010

For the first time an organocatalyst bearing a secondary nitrogen within a cyclohexane ring has been evaluated in the asymmetric Diels-Alder reaction. This organocatalyst is also the first of its kind based on a (1R,3S)-6,7-dimethoxy- 1-phenyl-1,2,3,4-tetrahydroisoquinoline backbone. These catalysts were tested over a range of dienes and dienophiles and displayed promising chemical conversions of up to 100% with up to 64% ee with triflic acid as the cocatalyst. Density functional theory computational studies and 2D NMR spectroscopy were used to determine the structure of the intermediate iminium ion formed between the most efficient catalyst and cinnamaldehyde. The reaction profile for each of the four possibilities in this reaction were calculated and it was found that the iminium intermediate leading to the major product is higher in energy but kinetically preferred. The activation energies of all possible reaction paths were calculated and the results correlated with the observed products. These experiments revealed that the presence of both (E)- and (Z)-isomers of the cinnamaldehyde were contributing factors for the low enantioselectivity of the reaction products.

Full text of DOI:10.1016/j.tetasy.2010.11.010

Tetrahedron: Asymmetry 21 (2010) 2859–2867
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Novel tetrahydroisoquinoline based organocatalysts for asymmetric
Diels–Alder reactions: insight into the catalytic mode using ROESY NMR
and DFT studies
Tricia Naicker a, Katja Petzold b, Thishana Singh c, Per I. Arvidsson a,d,e, Hendrik G. Kruger b,
Glenn E. M. Maguire b, Thavendran Govender a,
⇑
a School of Pharmacy and Pharmacology, University of KwaZulu-Natal, Durban Private Bag 4000, South Africa
b School of Chemistry, University of KwaZulu-Natal, Durban Private Bag 4000, South Africa
c Department of Chemistry, Durban University of Technology, Durban Private Bag 4000, South Africa
d Organic Pharmaceutical Chemistry, Department of Medicinal Chemistry, Uppsala Biomedical Centre, Uppsala University, Box 574, SE-751 23 Uppsala, Sweden
e Discovery CNS & Pain Control, AstraZeneca R&D Södertälje, S-151 85 Södertälje, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
For the first time an organocatalyst bearing a secondary nitrogen within a cyclohexane ring has been
evaluated in the asymmetric Diels–Alder reaction. This organocatalyst is also the first of its kind based
on a (1R,3S)-6,7-dimethoxy-1-phenyl-1,2,3,4-tetrahydroisoquinoline backbone. These catalysts were
tested over a range of dienes and dienophiles and displayed promising chemical conversions of up to
100% with up to 64% ee with triflic acid as the cocatalyst. Density functional theory computational studies
and 2D NMR spectroscopy were used to determine the structure of the intermediate iminium ion formed
between the most efficient catalyst and cinnamaldehyde. The reaction profile for each of the four possi-
bilities in this reaction were calculated and it was found that the iminium intermediate leading to the
major product is higher in energy but kinetically preferred. The activation energies of all possible reaction
paths were calculated and the results correlated with the observed products. These experiments revealed
that the presence of both (E)- and (Z)-isomers of the cinnamaldehyde were contributing factors for the
low enantioselectivity of the reaction products.
Received 28 October 2010
Accepted 5 November 2010
Available online 21 December 2010
Ó 2010 Elsevier Ltd. All rights reserved.
O
1. Introduction
Me
N
Ph
Me
Me
Ph
In the last decade there has been an explosive growth in the field
of organocatalysis and it has emerged as a powerful method for
accelerating various asymmetric transformations.1–3 Within this
topic, the in situ generation of iminium and enamine intermediates
using chiral amines (aminocatalysis) to facilitate the catalysis of
carbonyl transformations has received a tremendous amount of
interest from several research groups.4–7 Significant contributions
from the groups of List, MacMillan and JØrgensen have reported
the use of proline (i), imidazolinone (ii) and diaryl prolinols (iii)
as successful chiral organocatalysts, respectively (see Fig. 1).8–10
All of these organocatalysts consist of five membered hetero-
atom rings and were evaluated for numerous important enantio-
selective reactions such as Diels–Alder cycloadditions, Michael
additions, Mannich and Henry reactions. The iminium activation
of carbonyl compounds using secondary amines (the organocata-
lyst) allows for lowering of the lowest unoccupied molecular orbital
COOH
N
H
N
H
N
H
Ph
OTMS
i
iii
ii
Figure 1. Some examples of successful secondary amine organocatalysts.
(LUMO), thus emulating classical Lewis acid catalysts.8 The principle
behind iminium activation is based on the reversible condensation
between a secondary amine and an unsaturated aldehyde or ketone
substrate to form a positively charged iminium intermediate.
This results in a redistribution of the
double bond of the substrate towards the iminium cation. This low-
ers the energy of the LUMO on the unsaturated -system and the
p-electron density from the
p
iminium ion intermediate facilitates nucleophilic attack on the
substrate.5
The tetrahydroisoquinoline (TIQ) molecule and its derivatives
have been widely investigated due to their biological and pharma-
ceutical properties.11–14 Due to our ongoing pursuit to establish
⇑
Corresponding author.
E-mail address: govenderthav@ukzn.ac.za (T. Govender).
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.11.010

2860
T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
novel chiral catalysts,15–17 and
L-DOPA being commercially avail-
2. Results and discussion
2.1. Catalyst synthesis
able, this class of compounds represented an attractive skeleton
for a source of chirality. There are only a few reports that utilise
the TIQ backbone as a catalyst precursor.18–21 From these reports,
only studies performed by Stingl et al.21 and Basavaiah et al.18 made
use of a TIQ derivative as an organocatalyst in borane-mediated
hydrogenation reactions. Given our recent success with TIQ based
ligands for catalytic asymmetric transfer hydrogenation of prochi-
ral ketones,22,23 Henry reactions24 and hydrogenation of olefins,25
we decided to expand the potential of TIQ derivatives as organocat-
alysts. On route to the synthesis of a novel organocatalyst bearing
the TIQ framework that would potentially behave as a bifunctional
organocatalyst we discovered that one of its precursors was able to
form an iminium ion. This was unexpected; secondary amines that
are part of five-, rather than six-, membered ring systems are
known to be more efficient catalysts for enamine catalysis,26 and
to the best of our knowledge no previous report has shown
that six-membered ring amines have been successful for activating
Compounds 5, 8 and 9 (Fig. 2) are novel, whereas the syntheses
of the remaining compounds have been reported in the literature
for other applications. However, this is the first report of these
derivatives as organocatalysts. Based on the simplicity of the struc-
ture, TIQ catalysts 1 and 2 were the first to be synthesized accord-
ing to a literature procedure from (S)-phenylalanine.27 Thereafter
the more complex TIQ derivatives 3–5 were derived from L-DOPA
10 (Scheme 1). We recently reported a modification to the litera-
ture procedure for compound 11.22 L-DOPA 10 was treated with
benzaldehyde in the presence of K2CO3 and aqueous ethanol to af-
ford the trans-substituted derivative 11, and the relative configura-
tion was confirmed by comparison to the proton NMR spectral data
reported by Aubry et al. who elucidated both diastereomers of
11.28 This compound was N-protected with benzyl chloroformate
(Cbz) and then methylated at the phenolic and carboxylic acid
positions to yield 12. This was achieved by refluxing the compound
in acetone in the presence of Me2SO4 and KHCO3.29 Deprotection of
the Cbz group furnished catalyst 4.
a,b-unsaturated aldehydes or ketones through iminium ion
formation. This sparked our interest to further investigate this com-
pound and its derivatives for application in reactions known to
proceed via iminium activation. Herein, we report the evaluation
of novel organocatalysts 1–9 in the asymmetric Diels–Alder cyclo-
addition between a,b-unsaturated aldehydes and cyclopentadiene.
This is the first report of a chiral organocatalyst with the tetrahy-
Hydrolysis of the ester group in 4 afforded the acid derivative 3
and catalyst 5 was then obtained by simple esterification of 3 with
thionyl chloride in isopropanol. Notably, catalysts 3–5 and 7–9 pos-
sess a second stereogenic centre and could not be synthesized from
droisoquinoline backbone that contains two stereogenic centres.
O
Ph Ph
OR
O
O
O
O
O
O
O
OR
OR
O
NH
NH
NH
NH
O
Ph
Ph
1 R = H
3 R = H
6
7 R = H
2 R = Me
4 R = Me
8 R = Me
9 R = TMS
5 R = iPro
Figure 2. Catalysts evaluated for the Diels–Alder reaction.
O
O
O
1. CbzCl, KHCO3
HO
HO
HO
O
O
Dioxane/H2O
PhCHO, K2CO3
OH
O
OH
NH2
N
NH
EtOH/H2O, 0 oC-rt
Cbz
HO
2. KHCO3, Me2SO4
Acetone, reflux,
overnight;
10
12
11
Pd/C, H2
MeOH
O
O
O
O
O
O
O
O
O
OH
HCl (aq)
O
O
SOCl2
NH
NH
NH
(CH3)2CHOH
4
5
3
Scheme 1. Synthetic route to catalysts 3–5.

T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
2861
phenylalanine as it was essential to employ the activated aromatic
group of L-DOPA to facilitate the cyclisation. Hence, methylation of
the free phenolic hydroxyl groups had to be done after cyclization
in order to simplify the synthesis. In order to test the effect of these
methoxy groups on the reactivity of the catalysts, derivative 6 was
synthesized by a literature procedure (scheme not shown) from
L-DOPA and formaldehyde followed by our modified methylation
procedure for the hydroxyl and acid positions.30 Given the success
of diaryl proline derivatives as organocatalysts31 we synthesized
the six membered ring TIQ analogues 7–9. In order to introduce
the phenyl groups, the secondary amine 4 was first benzyl
protected to give 13, after which a Grignard reaction with phenyl
magnesium bromide afforded 14. Deprotection of 14 (Scheme 2) re-
sulted in the formation of catalyst 7.
Performing the reaction with our catalysts in methanol did show
a slightly higher conversion than acetonitrile, nevertheless, aceto-
nitrile was chosen as the solvent since it avoids hydrolysis of the
dimethyl-acetal formed when methanol is used and thus, greatly
simplifies the workup.
From the TIQ catalysts 1 and 2 the ester derivative 2 showed a
higher conversion than the acid derivative 1 (entries 1 and 2).
Thereafter, it was decided to test the TIQ derivatives that contained
a second stereogenic centre (entries 3 and 4). At that point, com-
pound 4 emerged as the most reactive catalyst amongst 1–4 (entry
4). It appeared logical to substitute the methyl ester of compound 4
with a more bulky group in the hope of increasing the enantio-
meric excess of the reaction products; in this case an isopropyl es-
ter 5 was used (entry 5). However, this had an insignificant effect
on the enantioselectivity. In order to evaluate if the higher conver-
sion rate (cf. the acid derivatives 1 and 3 (53% and 67% conversion
after 24 h, respectively) and the methyl esters 2 and 4 (67% and
73% conversion, respectively) was due to the introduction of the
phenyl ring at the carbon adjacent to the nitrogen or the introduc-
tion of the methoxy groups, catalyst 6 was synthesized and tested
(entry 6). It was clear that the methoxy groups were not responsi-
ble for the increased conversion. The activity and selectivity of cat-
alysts 3–5 compared to that of catalyst 1–2 were then attributed to
the electron withdrawing nature of the phenyl ring, which in-
creases the acidity of the nitrogen hence favoring iminium forma-
tion. To further establish if this was the case, we attempted the
reaction with pipecolic acid as the catalyst as well, but obtained
Derivative 14 was then treated with NaH followed by MeI to
yield the diphenyl methoxy N-benzyl protected compound 15,
which underwent debenzylation to give catalyst 8. Catalyst 9 was
obtained by the hydroxyl protection of 7 using trimethylsilyl
trifluoromethanesulfonate.32
2.2. Reactions of catalysts
As a model, we investigated the reaction between cinnamalde-
hyde 16 and cyclopentadiene 17 in the presence of various tetrahy-
droisoquinoline derivatives 1–9 as potential catalysts (Table 1).
Organocatalysts tested for this reaction have been shown to
react well in either methanol or acetonitrile/water mixtures.8,33
O
O
Ph
N
Ph
O
O
O
O
BnBr, K2CO3
O
O
OH
PhMgBr
THF
N
NH
CH3CN
O
O
Ph
Ph
14
13
4
NaH, MeI
THF
Pd/C, H2
MeOH
Ph
N
Ph
O
Ph
Ph
O
Ph
Ph
O
O
O
O
O
Pd/C, H2
MeOH
OH
NH
NH
O
Ph
8
15
7
TMSOTf, Et3N
CH2Cl2
Ph
Ph
O
O
OTMS
NH
9
Scheme 2. Synthetic route to catalysts 7–9.

2862
T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
Table 1
Organocatalysed [4+2] cycloaddition between trans-2-cinnamaldehdye 16 and cyclopentadiene 17 to give cycloaddition products exo- and endo-18a
10 mol %
Catalyst 1-9
Ph
+
OHC
Ph
+
O
CH3CN-H2O
CHO
(2R)-endo-18
Ph
(2R)-exo-18
17
16
Entry
Catalyst
Conv. (%) 24 h
Conv. (%) 48 h
ee (%) exo-18b
ee (%) endo-18b
exo:endoc
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
53
67
67
73
73
58
<5
<5
<5
78
94
90
94
94
62
—
47 (2R)
43 (2R)
44 (2R)
45(2R)
46 (2R)
36 (2R)
—
48 (2R)
50 (2R)
50 (2R)
51 (2R)
57 (2R)
36(2R)
—
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
—
—
—
—
—
—
a
Conditions: 3.0 equiv of diene, 1.0 equiv of dienophile, 0.1 equiv of catalyst and 0.1 equiv of HCl (37%) in 475
duplicate.
ll of CH3CN and 25 ll of H2O, all reactions were performed in
b
Determined by GC analysis using a chiral capillary column and absolute and relative configurations were determined by correlation of GC retention times. The absolute
and relative configurations were initially determined by correlation to known compounds.
c
The product ratios were determined by 1H NMR recorded at ambient temperature.
only low (<10%) conversions, even in the presence of an acidic
co-catalyst (vide supra).
A range of aliphatic dienophiles including electron-withdrawing
and -donating group substituents on the cinnamaldehyde aryl ring
were tested with cyclopentadiene (entries 1–6). All substrates
gave excellent conversion with the exception of 4-methoxy
cinnamaldehyde. The TIQ catalyst also proved to be efficient when
varying the diene (entries 7 and 8) producing excellent conversion
and facial selectivity (entry 8) but unfortunately poor enantio-
selectivity.
Given the success of the diaryl proline analogues as organocat-
alysts31 we synthesized the TIQ derivatives 7–9 to be tested on the
model reaction. The molecules showed very low conversion rates
(entries 7–9). Evidence for iminium formation from catalysts 7–9
with cinnamaldehyde was confirmed with proton NMR spectros-
copy. The low conversion could be due to one of two reasons, first,
the diene could not attack the dienophile or second the reaction
product is not released from the catalyst.
As with other organocatalysts tested for this reaction, an acid
co-catalyst proved to be necessary. Therefore, we investigated
the effect of varying the type of acid with catalysts 2 and 4
(Table 2). The trend observed was that the conversion is propor-
tional to the pKa of the acids. Using a sterically large acid did not
influence the enantioselectivity (entry 2). The triflic counterion
gave optimal conversion and enantioselectivity (entry 5). The pos-
sibility of such a strong acid catalyzing this Diels–Alder reaction
and compromising the enantioselectivity by a competing achiral
process has been thoroughly investigated by Lemay et al.34 From
their study it was concluded that triflic acid was not detrimental
to the selectivity of the reaction.
2.3. Structural elucidation of the iminium ion intermediate by
NMR spectroscopy and computational chemistry
Seebach et al. have thoroughly investigated the structural char-
acterization of reactive iminium ion intermediates derived from
proline, diphenyl prolinol and imidazolidinones with cinnamalde-
hyde, employing X-ray diffraction, NMR spectroscopy and density
functional theory (DFT) computational studies.35–38 They have con-
cluded that (E)- and (Z)-isomers are possible for the iminium ion
intermediates as illustrated in Scheme 3.
We then decided to follow a similar methodology to elucidate
the structure of the iminium ion formed between cinnamaldehyde
and our catalyst 4. We recently reported the X-ray crystal struc-
tures of precursors to catalyst 4 and 7, which revealed that the
N-containing six membered ring could exist either as a half boat
or half chair form, respectively.39,40 Based on these forms of the
Having optimized the conditions for the system, the scope of
these new TIQ based catalysts using various other dienophiles
and dienes was investigated (Table 3).
Table 2
Organocatalysed [4+2] cycloaddition between trans-2-cinnamaldehdye 16 and cyclopentadiene 17 using catalyst 2 and 4 in CH3CN with different acidsa
Entry
Catalyst
Acids
Conv. (%) 24 h
Conv. (%) 48 h
ee (%) exo-18b
ee (%) endo-18b
exo:endoc
1
2
3
4
5
6
7
8
9
2
2
2
2
2
4
4
4
4
4
4
HCl
p-TsOH
TFA
CH3SO2H
TfOH
HCl
p-TsOH
TFA
CH3SO2H
TfOH
71
83
75
77
100
73
72
65
70
95
93
98
92
91
—
90
90
90
90
—
47 (2R)
43 (2R)
43 (2R)
46 (2R)
43 (2R)
45 (2R)
48 (2R)
49 (2R)
47 (2R)
47 (2R)
47 (2R)
48 (2R)
50 (2R)
50 (2R)
46 (2R)
51 (2R)
51 (2R)
47 (2R)
55 (2R)
42 (2R)
57 (2R)
57 (2R)
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
10
11
TfOH
85 (12 h)
—
a
Conditions: 3.0 equiv of diene, 1.0 equiv of dienophile, 0.1 equiv of catalyst and 0.1 equiv of acid in 475
duplicate.
ll of CH3CN and 25 ll of H2O, all reactions were performed in
b
Determined by GC analysis using a chiral capillary column and the absolute and relative configuration were determined by correlation of GC retention times. The absolute
and relative configuration were initially determined by correlation to known compounds.
c
The product ratios were determined by 1H NMR recorded at ambient temperature.

T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
2863
Table 3
Organocatalysed Diels–Alder cycloadditions between various dienophiles and dienes utilising catalyst 4 and TfOH in CH3CNa
Entry
Dienophile
Diene
Conv. (%) 12 h
85 (55d)
ee (%) exob
ee (%) endob
exo:endoc
1
47 (64d)
57 (59)
2:1
O
O
nPr
2
3
4
100
100
100
22
rac
5
rac
4
1:1
1.2:1
3.5:1
O
O
10
O2N
5
6
100
40
35
52
42
4
2:1
O
MeO
1.6:1
O
O
O
7
8
100
100
30
—
—
rac
38
7:1
a
b
c
Conditions: 3.0 equiv of diene, 1.0 equiv of dienophile, 0.1 equiv of catalyst and 0.1 equiv of acid in 475
Determined by GC analysis using a chiral capillary column.
ll of CH3CN and 25 ll of H2O.
The product ratios were determined by 1H NMR recorded at ambient temperature.
Reaction carried out at 0 °C and conversion tested after 36 h.
d
N
N
Z-configuration
E-configuration
Scheme 3. Possible isomers of the iminium ion formed between trans-2-cinna-
maldehyde and tetrahydroisoquinoline.
catalyst together with the possibility of the (E)- and (Z)-isomers of
the reacting aldehyde, the structure of the iminium ion formed be-
tween cinnamaldehyde and catalyst 4 (Fig. 3) were examined. The
four possibilities A–D were computationally studied utilising DFT
calculations and are presented in Figure 3.
The calculations indicated that intermediate B had the lowest
energy conformation. NMR spectra of cinnamaldehyde in the pres-
ence of catalyst 4 were obtained to determine the geometry of cin-
namaldehyde in the resulting complex. Specific features from the
ROESY spectrum revealed that forms A and B coexist in the solu-
tion phase which corresponds to the (E)- and (Z)-isomers around
the C@N bond, respectively. Formation of imines is reversible at
room temperature,8 therefore, it is possible for A and B to exist
in equilibrium. The presence of these structures was inferred from
the ROESY correlations of protons H1 and H9 to the HB protons on
the substrate (see Fig. 4).
There was no indication of intermediate structures C or D from
the correlations in the ROESY spectrum. Peak integration of the H1
and H9 to the HB correlations showed a ratio of 2:1 between inter-
mediates A and B. However, the computational results reveal form A
to be 1.45 kcal molꢀ1 higher in energy. The ratio of A:B will depend
Figure 3. Optimized structures and relative energies (kcal molꢀ1) of the iminium
ion formed between cinnamaldehyde and catalyst 4 at the B3LYP/6-31+G(d) level of
theory.
on the energy barrier leading to imine formation, suggesting that the
product ratio (A:B) at room temperature is kinetically determined.
Therefore, based on the NMR evidence and the results of the
iminium intermediate computations, it was concluded that both
Integration of the correlation dots on the ROESY spectrum was performed.

2864
T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
We concluded, that the iminium proton was in the intermediate
exchange regime ( s to ms timescale) of a two-site exchange at
l
room temperature and, therefore, only observable at a lower tem-
perature (in this case ꢀ38 °C). This confirmed our initial room tem-
perature result of the interconversion between structures A and B.
From the imine intermediates A and B (see Fig. 3), the course of
the [2+4]-cycloaddition reaction was then studied computationally
following the method reported by Houk et al.41 Four possible
modes of attack for the incoming cyclopentadiene on each imine
intermediate exists. The first two products arise from ‘top side’ at-
tack of cyclopentadiene on the imine (see Fig. 3) where one CH2
group of cyclopentadiene is pointing out of the plane of the page
(indicated with the symbol I) and the other one with the CH2
pointing into the plane of the page (indicated as II). The second pair
of products arise from the corresponding ‘bottom side’ attack. The
eight transitions states were calculated and the energy profile for
each intermediate is presented in Figure 6. Similar to the model
proposed by the Houk group, our transition structures followed a
‘concerted but very asynchronous pathway’.41
The experimentally observed products are in the following or-
der: exo-(R), endo-(R), exo-(S) and endo-(S) (see Table 1 for the cor-
responding structures). It is clear from the activation energies that
the reaction preferably proceeded through intermediate A. The
transition state with the lowest energy barrier with respect to
the imine intermediate A (TS-top-II) led to the major experimen-
tally observed product. This observation agreed with our NMR
study which confirmed the dominant presence of intermediate A.
The competing reaction product was the endo-(R)-adduct. The
transition state leading to that (TS-top-I) has the second lowest
activation energy (15.5 kcal molꢀ1—see Fig. 6). The presence of
both (E)- and (Z)-isomers of the cinnamaldehyde/iminium ion
complex is a contributing factor for the low enantioselectivity of
the reaction products.
Figure 4. Expanded ROESY spectrum of catalyst 4 and cinnamaldehyde in CD3CN at
room temperature with characteristic cross-peaks marked.
The theoretical observation that attack of the cyclopentadiene is
from the top (A1-top-II) of the substrate, which corresponds to the
lowest activation energy, enabled us to rationalize the results ob-
served with catalysts 7–9 (Table 1, entries 7–9) as well. Although
the iminium intermediate was forming with these catalysts (seen
from 1H NMR), the two phenyl rings at the C9 position will prevent
attack of the diene from the top face, hence leading to the poor
conversion.
iminium structures A (E)-isomer and B (Z)-isomer were present in
solution at room temperature with the majority being the kinetically
preferred structure A.
Another interesting observation from the NMR experiments
was that the iminium proton HA was not seen at room tempera-
ture. However, proton signals HB and HC were clearly visible and
confirmed using 2D NMR experiments which included HMBC,
HSQC and COSY. These signals were clearly distinguished from
the free form of cinnamaldehyde. Performing the 1H NMR experi-
ment at ꢀ38 °C showed the appearance of two broad peaks in
the expected iminium proton region see Figure 5.
3. Conclusion
For the first time an organocatalyst bearing
a secondary
nitrogen within a cyclohexane ring has been evaluated in the
Figure 5. Expanded 1H NMR spectrum of catalyst 4 and cinnamaldehyde in CD3CN.

T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
2865
Figure 6. Calculated energy profiles for the catalysed Diels–Alder reaction for the pathways corresponding to imine intermediate A (top) and B (bottom), respectively.
asymmetric Diels–Alder reaction, thus leading to a new class of TIQ
based organocatalysts. Catalyst 4 afforded good to excellent chem-
ical conversion but poor selectivity with the addition of TfOH as
the cocatalyst. Furthermore, detailed computational and NMR
spectroscopic studies on the reaction intermediates and the reac-
tion mechanism allowed us to rationalize the modest level of ste-
reocontrol due to the presence of both (E)- and (Z)-isomers of the
cinnamaldehyde and suggested how to design second generation
catalysts. Studies into this class of organocatalysts are ongoing in
our laboratory.
at ambient temperatures. All melting points are uncorrected. The
enantiomeric excess of the chiral Diels–Alder products was deter-
mined by gas chromatography Agilent 6890 GC-Ms with a Agilent
7683 auto injector system equipped with an Astec Chiraldex gam-
ma-TA column (30 m ꢁ 0.25 mm), with helium gas as carrier gas
and electron impact ionization (EI, 70 eV) or a Agilent 6820 capil-
lary gas chromatograph with
a CP-Chirasil-b-Dex column
(25 m ꢁ 0.25 mm), nitrogen as carrier gas and a flame ionization
detector.
4.2. Synthesis of (S)-1,2,3,4-tetrahydroisoquinoline-3-carboxylic
acid 1
4. Experimental
4.1. General
This compound was prepared by following the literature proce-
dure from (S)-phenylalanine.27
Reagents and solvents were purchased from Aldrich, Merck and
Fluka. All NMR spectra were recorded on Bruker AVANCE III
400 MHz or 600 MHz instrument at room temperature unless
otherwise stated. Chemical shifts are expressed in ppm relative
to TMS unless otherwise stated and coupling constants are re-
ported in Hertz. Thin layer chromatography (TLC) was performed
using Merck Kieselgel 60 F254 plates. Crude compounds were puri-
fied with column chromatography using silica gel (60–200 mesh).
All solvents were dried using standard procedures. All IR spectra
were recorded on a Perkin Elmer spectrum 100 instrument with
a universal ATR attachment. Optical rotations were recorded on a
Perkin Elmer Polarimeter. High resolution mass spectrometric data
was obtained using a Bruker microTOF-Q II instrument operating
4.3. Synthesis of (S)-methyl 1,2,3,4-tetrahydroisoquinoline-3-
carboxylate 2
This compound was prepared by following the literature proce-
dure from 1 which was in turn derived from (S)-phenylalanine.27
4.4. Synthesis of (1R,3S)-6,7-dimethoxy-1-phenyl-1,2,3,4-tetra-
hydroisoquinoline-3-carboxylic acid 3
This compound was prepared by following the literature proce-
dure28 from L-DOPA with a slight modification which we have re-
cently reported.22

2866
T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
4.5. Synthesis of (1R,3S)-methyl-6,7-dimethoxy-1-phenyl-1,2,3,-
4-tetrahydroisoquinoline-3-carboxylate 4
126.73, 126.34, 126.31, 126.06, 125.53, 112.10, 111.59, 79.43,
64.65, 56.78, 55.84, 55.78, 51.60, 23.49.
The benzyl group was then removed following a procedure we
have recently reported for the analogous compound 7 to yield
compound 8 (0.2 g, 60%) as a white solid. Melting point 190–
This compound was prepared by following the literature proce-
dure from compound 3.28
192 °C. ½a 2D0
ꢂ
¼ ꢀ10:0 (c 0.11, CHCl3). IR (neat)
mmax: 2934, 1514,
4.6. Synthesis of (1R,3S)-isopropyl-6,7-dimethoxy-1-phenyl-
1,2,3,4-tetrahydroisoquinoline-3-carboxylate 5
1448, 1244, 1224, 1063, 698 cmꢀ1. HRMS calcd for C31H31NO3
(M+H+) 466.2377, found 466.2369. 1H NMR (400 MHz, CDCl3)
d = 7.47–7.12 (m, 12H), 7.08 (t, J = 7.6 Hz, 2H), 6.92 (d, J = 7.6 Hz,
2H), 6.65 (s, 1H), 6.40 (s, 1H), 5.23 (s, 1H), 3.95 (dd, J = 11.5,
3.6 Hz, 1H), 3.86 (s, 3H), 3.70 (s, 3H), 2.92–2.75 (m, 4H), 2.52 (dd,
J = 16.2, 11.5 Hz, 3H). 13C NMR (101 MHz, CDCl3) d = 147.83,
147.16, 143.92, 141.62, 140.86, 129.68, 129.40, 129.04, 128.40,
128.09, 127.41, 127.39, 127.30, 127.19, 127.18, 111.60, 110.51,
84.68, 59.96, 55.91, 55.81, 51.16, 49.48, 29.52.
To a stirred solution of compound 3 (1.00 g, 3.2 mmol) in dry
isopropanol (100 ml) at 0 °C, thionyl chloride (4.6 ml, 64.8 mmol)
was added dropwise. The mixture was then allowed to warm up
to room temperature and stirred overnight. The solution was then
concentrated in vacuo and the residue washed with sodium
bicarbonate solution (50 ml) and extracted with ethyl acetate
(2 ꢁ 25 ml). The organic extracts were combined and dried over
anhydrous Na2SO4 and the solvent was removed in vacuo. The
resulting residue was purified by column chromatography (50:50
EtOAc/Hexane, Rf 0.6) to afford the isopropyl TIQ ester 5 (0.95 g,
4.10. Synthesis of (1R,3S)-3-(diphenyl(trimethylsilyl)methyl)-
6,7-dimethoxy-1-phenyl-1,2,3,4-tetrahydroisoquinoline 932
84%) as a solid. Melting point 73–75 °C. ½a D20
ꢂ
¼ ꢀ70:0 (c 0.12 in
To a stirred solution of compound 7 (0.50 g, 1.1 mmol) and tri-
ethylamine (0.18 ml, 1.3 mmol) in dry dichloromethane (20 ml)
trimethylsilyl triflate (0.24 ml, 1.33 mmol) was added dropwise
at 0 °C under an inert atmosphere. The mixture was then allowed
to warm up to room temperature and stirred overnight. The
mixture was washed with water and the organic extracts were
combined, dried over anhydrous Na2SO4 and the solvent was
removed in vacuo. The resulting residue was purified by column
chromatography (20:80 EtOAc/Hexane, Rf 0.55) to afford the
diphenyl trimethylsilyl derivative 9 (0.52 g, 86%) as a white solid.
CHCl3). IR (neat)
mmax: 2938, 1724, 1512, 1246, 1218, 1103,
702 cmꢀ1. HRMS calcd for C21H25NO4 (M+H+) 356.1856, found
356.1872. 1H NMR (400 MHz, CDCl3) d = 7.39–7.11 (m, 5H), 6.67
(s, 1H), 6.34 (s, 1H), 5.26 (s, 1H), 5.02 (dq, J = 12.5, 6.3 Hz, 1H),
3.87 (s, 3H), 3.75 (dd, J = 8.6, 5.0 Hz, 1H), 3.68 (s, 3H), 3.14 (dd,
J = 16.0, 5.0 Hz, 1H), 2.98 (dd, J = 16.0, 8.7 Hz, 1H), 1.37–1.09 (m,
6H). 13C NMR (101 MHz, CDCl3) d = 172.90, 147.88, 147.40,
144.65, 128.71, 128.38, 128.05, 127.32, 125.87, 111.14, 110.83,
68.40, 58.89, 55.82, 51.42, 31.16, 21.82, 21.77.
Melting point 79–81 °C. ½a D20
¼ ꢀ30:0 (c 0.10, CHCl3). IR (neat)
ꢂ
4.7. Synthesis of (S)-methyl 6,7-dimethoxy-1,2,3,4-tetrahydro-
m .
max: 2952, 1513, 1446, 1245, 1225, 1067, 834, 752, 698 cmꢀ1
isoquinoline-3-carboxylate 6
HRMS calcd for C33H37NO3Si (M+H+) 524.2615, found 524.2595.
NMR chemical shifts are expressed in ppm relative to the CHCl3
peak. 1H NMR (400 MHz, CDCl3) d = 7.63–7.44 (m, 3H), 7.42–7.18
(m, 11H), 7.07 (d, J = 7.4 Hz, 2H), 6.73 (s, 1H), 6.58 (s, 1H), 5.40
(s, 1H), 4.08–3.92 (m, 4H), 3.85 (s, 3H), 2.71 (dd, J = 14.2, 11.1 Hz,
2H), -0.01 (s, 9H). 13C NMR (101 MHz, CDCl3) d = 145.61, 144.90,
143.05, 142.41, 142.03, 127.19, 126.47, 125.94, 125.89, 125.39,
125.24, 125.07, 124.70, 124.67, 109.41, 108.34, 80.66, 57.98,
53.71, 53.61, 49.68, 27.11, ꢀ0.00.
This compound was prepared by following the literature proce-
dure from L-DOPA.30
4.8. Synthesis of ((1R,3S)-6,7-dimethoxy-1-phenyl-1,2,3,4-tetra-
hydroisoquinolin-3-yl)diphenylmethanol 7
We recently reported the synthesis of this compound which is
derived from compound 3.22
4.11. General procedure for the Diels–Alder reaction
4.9. Synthesis of (1R,3S)-6,7-dimethoxy-3-(methoxydiphenyl-
methyl)-1-phenyl-1,2,3,4-tetrahydroisoquinoline 832
To a vial containing the catalyst (0.1 mmol) and the acid
(0.1 mmol) in 457
ll of CH3CN and 25 ll of H2O was added the
To a stirred solution of the N-benzyl protected derivative of
compound 7 (0.50 g, 0.92 mmol) in dry THF (20 ml) was added so-
dium hydride (0.07 g, 3.0 mmol) at 0 °C under an inert atmosphere.
The reaction mixture was then stirred for three hours at room tem-
perature and then MeI (0.2 ml, 2.4 mmol) was added. The mixture
was heated under reflux overnight. The excess NaH was hydroly-
sed with aqueous NH4Cl solution. The organic layer separated
and the aqueous layer extracted with ethyl acetate (2 ꢁ 10 ml).
The organic extracts were combined and dried over anhydrous
Na2SO4 and the solvent was removed in vacuo. The resulting resi-
due was purified by column chromatography (30:70 EtOAc/Hex-
ane, Rf 0.65) to afford the N-benzyl diphenyl methoxy derivative
a
,b-unsaturated aldehyde (1.0 mmol) followed by the diene
(3.0 mmol). In the case of cyclopentadiene, it was freshly distiled
before use. The reaction mixture was stirred for the time specified
in the text. It was then diluted with Et2O and washed successively
with H2O and brine. The organic layer was dried over anhydrous
Na2SO4, and concentrated in vacuo. The resulting residue was puri-
fied by silica gel chromatography and analysed as described
according to the following references Table 3, entry 1,8 entry 2,8
entry 3,8 entry 4,42 entry 5,43 entry 6,44 entry 7,8 entry8.8 Chroma-
tographs and retention times for all chiral products were compara-
ble to those reported.
of compound 8 (0.40 g, 78%) as a yellow oil. ½a D20
ꢂ
¼ ꢀ80:0 (c 0.10,
4.12. NMR spectroscopy details
CHCl3). IR (neat)
mmax: 2939, 1509, 1446, 1241, 1093, 1079,
695 cmꢀ1. HRMS calcd for C38H37NO3 (M+H+) 556.2846, found
556.2855. 1H NMR (400 MHz, CDCl3) d = 7.28–7.00 (m, 18H),
6.96–6.89 (m, 2H), 6.77 (s, 1H), 6.31 (s, 1H), 4.68 (s, 1H), 4.22
(dd, J = 12.5, 3.4 Hz, 1H), 4.11–4.02 (m, 1H), 3.92 (s, 3H), 3.67 (s,
3H), 3.29–3.14 (m, 2H), 3.03–2.89 (m, 4H). 13C NMR (101 MHz,
CDCl3) d = 147.71, 147.48, 146.14, 145.51, 143.47, 140.22, 129.80,
129.06, 128.20, 128.12, 127.95, 127.70, 127.62, 127.04, 126.78,
Study of the intermediate iminium ion structure. To a solution
of 4 (5.0 mg, 0.015 mmol) and TfOH (1.4
l
l, 0.015 mmol) in
475 of CD3CN and 25
l
l
l
l H2O, (E)-cinnamaldehyde (9.6 l,
l
0.075 mmol) and was added. 1D 1H and 13C experiments were re-
corded according to the standard Bruker library, using 16 and 1024
scans, respectively. 2D homonuclear COSY experiments and heter-
onuclear HSQC and HMBC experiments were recorded according

T. Naicker et al. / Tetrahedron: Asymmetry 21 (2010) 2859–2867
2867
13. Tarver, J. E.; Pfizenmayer, A. J.; Joullié, M. M. J. Org. Chem. 2001, 66, 7575–7587.
14. Zarranz De Ysern, M. E.; Ordoñez, L. A. Prog. Neuro-Psychopharmacol. 1981, 5,
343–355.
15. Arvidsson, P. I.; Govender, T.; Kruger, H. G.; Maguire, G. E. M.; Naicker, T. S. Afr.
J. Chem. 2009, 62, 60–66.
16. Boyle, G. A.; Govender, T.; Kruger, H. G.; Maguire, G. E. M. Tetrahedron:
Asymmetry 2004, 15, 3775–3781.
17. Boyle, G. A.; Govender, T.; Kruger, H. G.; Maguire, G. E. M. Tetrahedron:
Asymmetry 2004, 15, 2661–2666.
the standard Bruker library with 8 and 512; 8 and 256; and 16 and
512 scans and number of complex points in f1 dimension, respec-
tively. 2D homonuclear. ROESY experiments were recorded accord-
ing to Thiele et al., with 40 scans and 512 complex points.45
A
mixing time of 250 ms was applied to achieve proper transfer
and a relaxation delay of 2 s was applied, when distances where
extracted. ROE distances were used as a range from 2.5 to 5 Å for
calculations to restrain.
18. Basavaiah, D.; Das, U.; Roy, S. J. Chem. Sci. 2009, 121, 1003–1010.
19. Blanc, C.; Hannedouche, J.; Agbossou-Niedercorn, F. Tetrahedron Lett. 2003, 44,
6469–6473.
4.13. Computational details
20. Hari, Y.; Sakuma, M.; Miyakawa, A.; Hatano, K.; Aoyama, T. Heteroycles 2008,
305–311.
21. Stingl, K.; Martens, J.; Wallbaum, S. Tetrahedron: Asymmetry 1992, 3, 223–226.
22. Chakka, S. K.; Andersson, P. G.; Maguire, G. E. M.; Kruger, H. G.; Govender, T.
Eur. J. Org. Chem. 2010, 972–980.
23. Peters, B. K.; Chakka, S. K.; Naicker, T.; Maguire, G. E. M.; Kruger, H. G.;
Andersson, P. G.; Govender, T. Tetrahedron: Asymmetry 2010, 21, 679–687.
24. Kawthekar, R. B.; Chakka, S. K.; Francis, V.; Andersson, P. G.; Kruger, H. G.;
Maguire, G. E. M.; Govender, T. Tetrahedron: Asymmetry 2010, 21, 846–852.
25. Chakka, S. K.; Peters, B. K.; Maguire, G. E. M.; Kruger, H. G.; Andersson, P. G.;
Govender, T. Tetrahedron: Asymmetry 2010.
Complexes A–D and transitions states were optimized in the
gas phase using GAUSSIAN 0946 at the the density functional theory
(DFT) level employing the B3LYP (Becke’s three-parameter non-lo-
cal exchange function47–49) with the correlation functional of Lee,
Yang and Par50 in conjunction with the 6-31+G(d) basis set. set.
Diffuse functions are typically used for a more accurate description
where lone pair electrons are involved, while polarization func-
tions remove some limitations of the basis set by expansion of
the virtual space. Solvation effects were not considered in order
to simplify the model. Geometry optimizations were performed
without restrictions in order to locate extrema presented herein.
Frequency calculations were performed for all structures. Transi-
tions states were characterized by a single imaginary frequency,
which corresponds to the movement of atoms consistent with
the expected reaction. To ensure that the lowest energy transition
state for the first step (bond formation between atoms 1 and 2 in
Figure 6) was found, a relaxed scan (using a semi-empirical calcu-
lation with Parameterized Model number 6)51 was performed with
the atom distance for atoms 1 and 2 kept fixed at about 1.89 Å. The
scan entailed a 360° rotation of the cyclopentadiene molecules in
15° steps. The structure corresponding to the lowest energy struc-
ture on the energy profile was used for a normal unconstrained
transition state for the DFT calculation.
26. List, B. Synth. Lett. 2001, 1675–1686.
27. Grunewald, G. L.; Sall, D. J.; Monn, J. A. J. Med. Chem. 1988, 31, 824–830.
28. Aubry, S.; Pellet-Rostaing, S.; Faure, R.; Lemaire, M. J. Heterocycl. Chem. 2006,
43, 139–148.
29. Cai, T. B.; Zou, A.; Thomas, J. B.; Brieaddy, L.; Navarro, H. A.; Carrol, F. I. J. Med.
Chem. 2008, 1849–1860.
30. Tang, Y. F.; Liu, Z. Z.; Chen, S. Z. Tetrahedron Lett. 2003, 44, 7091–7094.
31. Palomo, C.; Mielgo, A. Angew. Chem., Int. Ed. 2006, 45, 7876–7880.
32. Aggarwal, V. K.; Sandrinelli, F.; Charmant, J. P. H. Tetrahedron: Asymmetry 2002,
13, 87–93.
33. Hartikka, A.; Hojabri, L.; Bose, P. P.; Arvidsson, P. I. Tetrahedron: Asymmetry
2009, 20, 1871–1876.
34. Lemay, M.; Ogilvie, W. W. Org. Lett. 2005, 7, 4141–4144.
35. Groselj, U.; Schweizer, W. B.; Ebert, M.-O.; Seebach, D. Helv. Chim. Acta 2009,
92, 1–13.
36. Groselj, U.; Seebach, D.; Badine, D. M.; Schweizer, W. B.; Beck, A. K.;
Krossing, I.; Klose, P.; Hayashi, Y.; Uchimaru, T. Helv. Chim. Acta 2009, 92,
1225–1259.
37. Seebach, D.; Gilmour, R.; Groselj, U.; Deniau, G.; Sparr, C.; Ebert, M.-O.; Beck, A.
K.; McCusker, L. B.; Sisak, D.; Uchimaru, T. Helv. Chim. Acta 2010, 93, 603–634.
38. Seebach, D.; Groselj, U.; Badine, D. M.; Schweizer, W. B.; Beck, A. K. Helv. Chim.
Acta 2008, 91, 1999–2034.
39. Naicker, T.; Govender, T.; Kruger, H. G.; Maguire, G. E. M. Acta Crystallogr., Sect.
E: Struct. Rep. 2010, 66, 0638.
40. Naicker, T.; McKay, M.; Govender, T.; Kruger, H. G.; Maguire, G. E. M. Acta
Crystallogr., Sect. E: Struct. Rep. 2009, 65, o3278.
41. Gordillo, R.; Houk, K. N. J. Am. Chem. Soc. 2006, 128, 3543–3553.
42. Kano, T.; Tanaka, Y.; Maruoka, K. Org. Lett. 2006, 8, 2687–2689.
43. Gotoh, H.; Hayashi, Y. Org. Lett. 2007, 9, 2859–2862.
44. He, H.; Pei, B. J.; Chou, H.; Tian, T.; Cha, W. H.; Lee, A. W. M. Org. Lett. 2008, 10,
2421–2424.
Acknowledgements
The authors wish to thank the National Research Foundation
(South Africa, GUN 65387) The SA/Swedish Research Links Pro-
gramme grant and Aspen Pharmacare, South Africa.
References
45. Thiele, C. M.; Petzold, K.; Schleucher, J. Chem. Eur. J. 2009, 15, 585–588.
46. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven,
T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas,
Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. GAUSSIAN 09 Revision A. 1;
Gaussian, Inc.: Wallingford CT, 2009.
1. Enantioselective Organocatalysis; Dalko, P. I., Ed.; Wiley-VCH Verlag Gmb & Co
KGaA: Weinheim, 2007.
2. Dondoni, A.; Massi, A. Angew. Chem., Int. Ed. 2008, 47, 4638–4660.
3. Melchiorre, P.; Marigo, M.; Carlone, A.; Bartoli, G. Angew. Chem., Int. Ed. 2008,
47, 6138–6171.
4. Erkkila, A.; Majander, I.; Pihko, P. M. Chem. Rev. 2007, 107, 5416–5470.
5. Lelais, G.; MacMillan, D. W. C. Aldrichim. Acta 2006, 39, 79–87.
6. List, B. Chem. Commun. 2006, 819–824.
7. Mukherjee, S.; Yang, J. W.; Hoffmann, S.; List, B. Chem. Rev. 2007, 107, 5471–
5569.
8. Ahrendt, K. A.; Borths, C. J.; MacMillan, D. W. C. J. Am. Chem. Soc. 2000, 122,
4243–4244.
9. Franzen, J.; Marigo, M.; Fielenbach, D.; Wabnitz, T. C.; Kjaersgaard, A.;
Jorgensen, K. A. J. Am. Chem. Soc. 2005, 127, 18296–18304.
10. List, B.; Lerner, R. A.; Barbas, C. F. J. Am. Chem. Soc. 2000, 122, 2395–2396.
11. Liu, Z.-Z.; Wang, Y.; Tang, Y.-F.; Chen, S.-Z.; Chen, X.-G.; Li, H.-Y. Bioorg. Med.
Chem. Lett. 2006, 16, 1282–1285.
47. Becke, A. D. J. Chem. Phys. 1992, 97, 9173–9177.
48. Becke, A. D. J. Chem. Phys. 1992, 96, 2155–2160.
49. Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
50. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
51. Stewart, J. J. P. J. Mol. Model. 2007, 13, 1173–1213.
12. Scott, J. D.; Williams, R. M. Chem. Rev. 2002, 102, 1669–1730.