1,1′-Binaphthylazepine-based ligands for asymmetric catalysis. Part 2: New aminoalcohols as chiral ligands in the enantioselective addition of ZnEt2 to aromatic aldehydes

Article abstract of DOI:10.1016/S0957-4166(01)00200-2

The new enantiopure 1,2-aminoalcohols 1b-1h having 1,1′-binaphthylazepine skeleton have been tested as catalytic precursors in the enantioselective addition of ZnEt₂ to benzaldehyde. The best results were seen with ligand 1d, which owes its chi

Full text of DOI:10.1016/S0957-4166(01)00200-2

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 1235–1239
1,1%-Binaphthylazepine-based ligands for asymmetric catalysis.
Part 2: New aminoalcohols as chiral ligands in the
enantioselective addition of ZnEt2 to aromatic aldehydes
Stefano Superchi, Tommaso Mecca, Egidio Giorgio and Carlo Rosini*
Dipartimento di Chimica, Universita` della Basilicata, Via Nazario Sauro 85, 85100 Potenza, Italy
Received 6 April 2001; accepted 2 May 2001
Abstract—The new enantiopure 1,2-aminoalcohols 1b–1h having 1,1%-binaphthylazepine skeleton have been tested as catalytic
precursors in the enantioselective addition of ZnEt2 to benzaldehyde. The best results were seen with ligand 1d, which owes its
chirality only to the atropisomerism of the binaphthyl nucleus and does not have any stereogenic carbon atom. In the presence
of 1d benzaldehyde was quickly and cleanly transformed to (S)-1-phenylpropanol in 99% yield and 87% e.e. The same ligand was
also used in the asymmetric ZnEt2 addition to other aryl aldehydes giving rise to (S)-1-arylpropanols in almost quantitative yields
and e.e.s up to 90%. © 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
in asymmetric catalysis,6 we decided to evaluate the
scope and limitations, as chiral catalysts, of 1,2-
aminoalcohols having the general formula reported in
Fig. 1.
Enantiopure natural1 and synthetic2 aminoalcohols
have found widespread application in asymmetric syn-
thesis. In general, these systems owe their chirality to
the presence of stereogenic carbon atoms, and only
recently aminoalcohols having either planar3,4 or
atropisomeric5,6d,m chirality have been employed. Tak-
ing into account the high efficiency shown by atropiso-
meric ligands having a 1,1%-binaphthylazepine skeleton
To test the efficiency of these ligands we evaluated their
behavior in the enantioselective addition of ZnEt2 to
aldehydes,7 taken as a benchmark reaction. As a matter
of fact, only a single report by Noyori et al.6d concerns
the use of the parent aminoalcohol of this family,
(S)-1a, in the ZnEt2 addition to benzaldehyde, leading
to (S)-1-phenylpropanol in 49% e.e. In the commonly
accepted mechanism7 for this reaction (Scheme 1) the
reaction between the aminoalcohol and a molecule of
ZnEt2 firstly gives rise to the formation of the chelated
ethylzinc–alkoxide intermediate (A), which then coordi-
nates with both the aldehyde and a second molecule of
ZnEt2, allowing the stereoselective addition of the ethyl
group to the aldehyde carbonyl. Recently, Goldfuss
Figure 1.
Scheme 1.
* Corresponding author. Tel.: +39-0971-202241; fax: +39-0971-202223; e-mail: rosini@unibas.it
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00200-2

1236
S. Superchi et al. / Tetrahedron: Asymmetry 12 (2001) 1235–1239
and Houk8 attempted a theoretical analysis aimed at
clarifying the relationship between the stereoselectivity
of this reaction and the structural features of several
aminoalcohols used as chiral ligands.9 They calculated
the relative energies of the possible transition states for
each ligand and explained the moderate enantioselectiv-
ity obtained with (S)-1a, pointing out that in this ligand
‘…the binaphthyl substituent at N does not efficiently
distinguish the faces of the five-membered Zn–chelate
ring. The lack of a substituent at C(O) eliminates
significant repulsive interactions with the bulky ZnEt2
moieties’.
MS and when complete conversion of the aldehyde was
detected the mixture was quenched with 10% aqueous
HCl. After extraction with Et2O, drying and evapora-
tion of solvent, the product ratio was directly deter-
mined on the crude mixture by GC–MS and the e.e. of
the 1-phenyl-1-propanol product was measured by
HPLC on a chiral stationary phase.
The results collected in Table 1 show that, in the
presence of catalytic amounts of compounds 1b–1h,
diethylzinc adds cleanly to benzaldehyde providing 1-
phenyl-1-propanol (with complete conversion and with-
out any trace of benzyl alcohol, a common byproduct
of this reaction). Only in run 8, where ligand (aS,R)-1f
was used, the reaction was slower and a significant
amount of benzyl alcohol (10%) was detected. Reduc-
ing the reaction temperature (from 20 to 0°C) slowed
down the reaction but did not increase the enantioselec-
tivity (compare runs 2 and 3), while reducing the
amount of ligand from 8 to 3% led to only a small
decrease in the enantioselectivity, giving product with
e.e.s of 81 and 75%, respectively (compare runs 2 and
4). With all the only-atropisomerically chiral ligands
1b–1e, the (S)-configuration of the binaphthyl moiety
always induced predominant formation of the (S)-alco-
hol product (runs 1–7). Interestingly, higher e.e.s were
obtained with increasing bulk of the R groups, the
lowest selectivity being seen with the dimethyl deriva-
tive 1b. This result is clearly in keeping with the conclu-
sions of Houk and Goldfuss, concerning the need of
substituents on the C(O). The highest e.e. and a shorter
reaction time were achieved with the diphenyl-substi-
tuted ligand (S)-1d. However, the presence of the
bulkier tert-butyl groups in 1e (run 7) led to a signifi-
cant reduction in product e.e. and a longer reaction
time. This result can be interpreted considering that the
presence of the tert-butyl groups prevents efficient
coordination of the second ZnEt2 molecule and/or the
aldehyde to the alkoxide (A) (Scheme 1), thus reducing
the influence of the catalyst on the overall process.
This conclusion prompted us to carry out an investiga-
tion aimed at verifying whether the introduction of
substituents of increasing size at the C(O) center of the
ligand and the resultant increases in steric demand
would lead to higher enantioselectivity.
2. Results and discussion
We prepared10 aminoalcohols 1b–1e having, respec-
tively, two methyl groups, a cyclohexyl ring, two
phenyl, and two tert-butyl groups on the C(O). Addi-
tionally we prepared the aminoalcohols 1f–1h, where
the C(O) atom is stereogenic, in order to evaluate the
role on the enantioselectivity of ligands bearing two
different types of chirality. The b-aminoalcohols 1b–1h
were tested in the enantioselective addition of diethyl-
zinc to benzaldehyde under standard conditions (Scheme
2) in which the reaction was carried out in dry toluene
at 20°C in the presence of 8 mol% of the chiral ligand.
The reaction mixture was monitored by TLC and GC–
Scheme 2.
Table 1. Enantioselective addition of ZnEt2 to benzaldehyde mediated by ligands 1b–1h
Run
Liganda
Time (h)
Temp. (°C)
Yield (%)b
E.e. (%)c (ac)d
1
2
3
4
5
6
7
8
9
(S)-1b
(S)-1c
14
5
16
16
0.5
2
14
20
12
3
20
20
0
20
20
20
20
20
20
20
97e
99
99
99
99
99
99
90h
99
99
64 (S)
81 (S)
78 (S)
75 (S)
87 (S)
86 (S)
64 (S)
38 (R)
79 (S)
84 (S)
(S)-1c
(S)-1cf
(S)-1d
(S)-1dg
(S)-1e
(S,R)-1f
(S,S)-1g
(S,S)-1h
10
a 8 mol% of ligand was used.
b Chromatographic (GLC) yield. No traces of benzyl alcohol were detected.
c Determined by HPLC on Chiralcel OD.
d Determined by elution order on Chiralcel OD.7d
e About 1 mol% of benzyl alcohol was detected.
f 3 mol% of ligand was used.
g Lithium salt of the ligand was used.
h 10 mol% of benzyl alcohol present.

S. Superchi et al. / Tetrahedron: Asymmetry 12 (2001) 1235–1239
1237
It is interesting to note that by simply changing the
nature of the R groups, without changing the chiral
backbone of the molecule, we were able to markedly
increase the enantioselectivity from the 49% e.e. of
Noyori’s ligand (S)-1a, to 64% with (S)-1b, 81% by the
use of (S)-1c and 87% with (S)-1d. We then showed
that making suitable structural modifications to these
catalysts, which have only atropisomeric chirality and
do not possess any stereogenic carbon atom, high enan-
tioselectivities can be obtained. Previous experimental7
and theoretical8,9 investigations showed that the abso-
lute configuration of C(O) primarily determines the
stereochemical outcome of the b-aminoalcohol cata-
lyzed reaction. In the present case no stereogenic center
is present on the carbon bearing the hydroxy group, but
the chiral environment created by the 1,1%-binaphthyl
nucleus is transmitted through the molecule and affects
events occurring at the NꢀZnꢀO moiety. The possibility
of this long range control of chirality depends on the
nature of the substituents at the C(O) atom: it is low for
R=H, higher for R=Me, and reaches a maximum
value for R=Ph. In the latter case the presence of the
diphenylhydroxymethyl group allows the maximum
enantioselectivity, thus confirming the efficiency of this
achiral moiety in asymmetric catalysis.11
the second reaction a matched pair effect is seen. These
experiments also indicate that the stereochemical out-
come of the reaction is mainly determined by the
absolute configuration of the carbon atom bearing the
OH.
Interestingly, (aS,R)-1f induced predominant formation
of product with (R)-configuration, although we have
shown above that the (S)-configuration of the binaph-
thyl moiety induces an (S)-configured alcohol (runs 1–7
in Table 1). This result is in keeping with Noyori’s
finding6d that (R)-1-phenyl-2-N,N-dimethylamino-
ethanol (structurally comparable with the aliphatic moi-
ety of (aS,R)-1f) induced the formation of (R)-1-
phenyl-1-propanol in 59% e.e. in the ZnEt2 addition to
benzaldehyde.
In order to better investigate the reactivity of such
ligands the most efficient, (S)-1d, was used in the
enantioselective addition of ZnEt2 to different aryl alde-
hydes (Scheme 3) under standard conditions and the
results are summarized in Table 2.
The reactivity of the substrates was clearly related to
the electrophilic character12 of the carbonyl group.
Benzaldehyde (run 1) was fully reacted after 30 min-
utes. With electron donating ring substituents the reac-
tion rate was lowered whilst electron withdrawing
groups served to enhance the rate of reaction. Thus, the
slowest reaction (90 minutes) was observed with 4-
methoxybenzaldehyde (run 2), while a very rapid reac-
tion of less than 10 minutes was observed with
4-cyanobenzaldehyde (run 3).
The introduction of a stereogenic center on the C(O) in
the monosubstituted ligands (aS,R)-1f, (aS,S)-1g and
(aS,S)-1h led to some interesting observations. In all
cases the enantioselectivities were lower with respect to
the simply atropisomeric counterparts 1a–1e (in Table 1
compare run 5 with runs 8 and 9). The ligand (aS,R)-1f
induced the formation of the alcohol with opposite
(R)-configuration and with low 38% e.e. while its
epimer, (aS,S)-1g, led to the expected (S)-configured
alcohol in 79% e.e. Clearly, in the former the stereo-
genic elements constitute a mismatched pair, while in
In contrast, the e.e. of the product was found to be
little dependent on the nature of the substrate. This
result is in agreement with many investigations7 report-
ing that enantioselectivity is mainly determined by
steric factors, but is in contrast with a recent report5c
where enantioselectivity was observed to increase with
substrate reactivity. The e.e.s obtained in the case of
para-substituted benzaldehydes ranged from 80% for
4-methoxybenzaldehyde to 90% for 4-(trifluoro-
Scheme 3.
Table 2. Addition of ZnEt2 to aryl aldehydes mediated by ligand (S)-1da
Run
Ar
Time
Temp. (°C)
Yield (%)b
E.e. (%)c (ac)d
1
2
3
4
5
6
7
8
C6H5
30 min
90 min
10 min
25 min
75 min
2 h
20
20
20
20
20
20
20
20
99
99
99
99
98
99
92
91
87 (S)
80 (S)
84e (S)f
90e (S)f
86e (S)f
87 (S)
77 (S)g
68 (S)
4-CH3OC6H4
4-CNC6H4
4-CF3C6H4
2-Naphtyl
1-Naphtyl
9-Anthryl
C6H5-CHꢁCH
16 h
45 min
a 8% of aminoalcohol was used.
b Chromatographic (GLC) yield. No traces of benzyl alcohol were detected.
c Determined by HPLC on Chiralcel OD.
d Determined by elution order on Chiralcel OD.7d
e Determined by HPLC on Chiralcel OJ.
f Determined by comparison of [h]D with literature values.13
g Tentatively assigned by comparison of [h]D sign with literature value for (S)-1-(9-anthryl)-1-ethanol.14

1238
S. Superchi et al. / Tetrahedron: Asymmetry 12 (2001) 1235–1239
methyl)benzaldehyde. In the case of 2-naphthaldeyde,
an e.e. of 86%, similar to that of the sterically compara-
ble benzaldehyde, was obtained. The more hindered
1-naphthaldehyde was also alkylated with a similar e.e.
of 87% (run 6). These results seem to indicate that
(S)-1d allows alkylation of aromatic aldehydes with
little dependence on the steric effects in the substrate.
Also a very rigid substrate such as 9-anthraldehyde (run
7) afforded a good 77% e.e., while lower 68% e.e. was
obtained with trans-cinnamaldehyde (run 8) where the
aromatic ring is far from the reaction center.
and 9-anthraldehyde were used as purchased. Analyti-
cal TLC was performed on 0.2 mm silica gel plates
(Merck 60 F-254). Mixture composition was deter-
mined by GLC-MS on a Hewlett Packard 6890 chro-
matograph equipped with an HP-5973 mass detector.
4.2. Typical procedure for the diethylzinc addition to
aldehydes
4.2.1. (S)-(−)-1-(9-Anthryl)-1-propanol. To a solution of
(S)-1d (24.5 mg, 0.05 mmol) in dry toluene (3 mL),
under a nitrogen atmosphere at rt, was added a solu-
tion of ZnEt2 in hexane (1.0 M, 1.25 mL, 1.25 mmol).
The mixture was stirred for 30 min. A solution of
9-anthraldehyde (129 mg, 0.625 mmol) in dry toluene
(1.3 mL) was added. The reaction was monitored by
GC–MS and when no more traces of the aldehyde were
detected (16 h) the reaction was quenched by addition
of 10% aqueous HCl. The mixture was extracted with
Et2O and the organic phase was washed with brine, and
dried over anhydrous Na2SO4. After evaporation of the
solvent the solid residue was directly analyzed by GC–
MS and HPLC. GC–MS yield 92%; e.e. 77% by HPLC
on Chiralcel OD (hexane/propan-2-ol 9:1, flow 0.5 mL/
min, tr1 18.7 min, tr2 34.9 min). The crude product was
purified by column chromatography (silica gel, CHCl3)
to afford pure (−)-1-(9-anthryl)-1-propanol (128 mg,
87%). The sample was assigned (S)-configuration by
comparison of the [h]D sign with the literature value14
for (S)-(−)-1-(9-anthryl)-1-ethanol. [h]2D0=−19.2 (c 0.53,
3. Conclusions
We have clearly demonstrated that enantiopure
atropisomeric aminoalcohols having the structure of
1b–1h can act as efficient promoters of the enantioselec-
tive addition of ZnEt2 to aryl aldehydes. This result is
important from a practical point of view, because a new
class of efficient ligands has been made available, but
also from a wider view, because in the most efficient
ligands of this type, compounds 1b–1e, their chirality
results only from the atropisomerism of the binaphthyl
nucleus and they do not have any stereogenic carbon
atom.
It is interesting to note that this investigation fully
confirms the theoretical analysis of Goldfuss and
Houk8 about the need of substituents at C(O) in order
to achieve higher enantioselectivity. We are convinced
that the present experimental results may stimulate
further theoretical and experimental investigation
aimed at clarifying the correlation between structure
and catalytic activity of these aminoalcohol ligands.
Work is now in progress to also study the efficiency of
compounds 1b–1h in other asymmetric reactions.
1
CHCl3). H NMR (300 MHz, CDCl3): l 1.00 (t, 3H,
J=7.4 Hz); 2.1–2.3 (m, 1H); 2.32 (br s, 1H); 2.3–2.5 (m,
1H); 6.13 (t, 1H, J=7.3 Hz); 7.4–7.5 (m, 4H); 8.0 (m,
2H); 8.36 (s, 1H); 8.5–8.8 (br s, 2H). 13C NMR (75
MHz, CDCl3): l 135.0; 131.7; 129.4; 129.3; 127.9;
125.4; 125.0; 124.7; 72.6; 30.8; 11.3.
Acknowledgements
4. Experimental
Financial support from MURST (COFIN2000) and
Universita` della Basilicata (Potenza) is gratefully
acknowledged.
4.1. General procedures
All the 1-aryl-1-propanols obtained by the diethylzinc
addition to aryl aldehydes showed NMR spectra fully
in agreement with literature data. H NMR (300 MHz)
1
References
and 13C NMR (75 MHz) spectra were recorded in
CDCl3 on a Bruker Aspect 300 spectrometer. Optical
rotations were measured with a JASCO DIP-370 digital
polarimeter. E.e.s of the optically active 1-aryl-1-
propanols were determined by HPLC analysis per-
formed on a JASCO PU-1580 pump with a Varian
2550 UV detector and Daicel Chiralcel OD or OJ
columns. Toluene was freshly distilled from sodium
benzophenone ketyl under nitrogen atmosphere prior to
use. Diethylzinc was used as a 1.0 M solution in hexane
(Aldrich) and was used as purchased. Commercially
available (Aldrich) benzaldehyde, 4-anisaldehyde, 4-
trifluoromethylbenzaldehyde, 1-naphthaldehyde and
trans-cinnamaldehyde were distilled prior their use and
stored under nitrogen atmosphere. Commercially avail-
able (Aldrich) 4-cyanobenzaldehyde, 2-naphthaldehyde
1. Wynberg, H. Topics Stereochem. 1986, 16, 87.
2. (a) Ager, D. J.; Prakash, I.; Schaad, D. R. Chem. Rev.
1996, 96, 835; (b) Fache, F.; Schulz, E.; Tommasino, M.
L.; Lemaire, M. Chem. Rev. 2000, 100, 2159.
3. (a) Bolm, C.; Muniz-Fernandez, K.; Seger, A.; Raabe, G.
Synlett 1997, 1051; (b) Bolm, C.; Muniz-Fernandez, K.;
Seger, A.; Raabe, G.; Gunther, K. J. Org. Chem. 1998,
63, 7860; (c) Bolm, C.; Muniz, K. Chem. Commun. 1999,
1295; (d) Patti, A.; Nicolosi, G.; Howell, J. A. S.;
Humphries, K. Tetrahedron: Asymmetry 1998, 9, 4381; (e)
Dosa, P. I.; Ruble, J. C.; Fu, G. C. J. Org. Chem. 1997,
62, 444; (f) Watanabe, M. Synlett 1995, 1050.
4. (a) Uemura, M.; Miyake, R.; Nakayama, K.; Shiro, M.;
Hayashi, Y. J. Org. Chem. 1993, 58, 1238; (b) Jones, G.
B.; Heaton, S. B. Tetrahedron: Asymmetry 1993, 4, 261;

S. Superchi et al. / Tetrahedron: Asymmetry 12 (2001) 1235–1239
1239
(c) Jones, G. B.; Huber, R. S.; Chapman, B. J. Tetra-
hedron: Asymmetry 1997, 8, 1797.
Haslinger, U.; Mereiter, K.; Widhalm, M. Tetrahedron:
Asymmetry 2000, 11, 4207.
5. (a) Bringmann, G.; Breuning, M. Tetrahedron: Asymme-
try 1998, 9, 667; (b) Vyskocil, S.; Jaracz, S.; Smrcina, M.;
Sticha, M.; Hanus, V.; Polasek, M.; Kocovsky, P. J. Org.
Chem. 1998, 63, 7727; (c) Zhang, H.; Xue, F.; Mak, T. C.
W.; Chan, K. S. J. Org. Chem. 1996, 61, 8002.
7. (a) Noyori, R.; Kitamura, M. Angew. Chem., Int. Ed.
Engl. 1991, 103, 34; (b) Soai, K.; Niwa, S. Chem. Rev.
1992, 92, 833; (c) Pu, L.; Yu, H.-B. Chem. Rev. 2001, 101,
757; (d) Dai, W.-M.; Zhu, H.-J.; Hao, X.-J. Tetrahedron:
Asymmetry 2000, 11, 2315 and references cited therein.
8. Goldfuss, B.; Houk, K. N. J. Org. Chem. 1998, 63, 8998.
9. For other recent computational studies on asymmetric
diethylzinc addition to aldehydes, see: (a) Yamakawa,
M.; Noyori, R. Organometallics 1999, 18, 128; (b) Gold-
fuss, B.; Steigelmann, M.; Khan, S. I.; Houk, K. N. J.
Org. Chem. 2000, 65, 77 ; (c) Vazquez, J.; Pericas, M. A.;
Maseras, F.; Lledos, A. J. Org. Chem. 2000, 65, 7303; (d)
Goldfuss, B.; Steigelmann, M.; Rominger, F. Eur. J. Org.
Chem. 2000, 1785.
6. (a) Mazaleyrat, J. P.; Cram, D. J. J. Am. Chem. Soc.
1981, 103, 4585; (b) Hawkins, J. M.; Fu, G. C. J. Org.
Chem. 1986, 51, 2820; (c) Kanth, J. V. B.; Periasamy, M.
J. Chem. Soc., Chem. Commun. 1990, 1145; (d) Noyori,
R.; Suga, S.; Okada, S.; Kawai, K.; Kitamura, M.;
Oguni, N.; Hayashi, M.; Kaneko, T.; Masuda, Y. J.
Organomet. Chem. 1990, 382, 19; (e) Hawkins, J. M.;
Lewis, T. A. J. Org. Chem. 1994, 59, 649; (f) Kubota, H.;
Koga, K. Tetrahedron Lett. 1994, 35, 6689; (g) Wimmer,
P.; Widhalm, M. Tetrahedron: Asymmetry 1995, 6, 657;
(h) Kubota, H.; Koga, K. Heterocycles 1996, 42, 543; (i)
Rosini, C.; Tanturli, R.; Pertici, P.; Salvadori, P. Tetra-
hedron: Asymmetry 1996, 7, 2971; (j) Aggarwal, V. K.;
Wang, M. F. Chem. Commun. 1996, 191; (k) Rych-
novsky, S. D.; McLernon, T. L.; Rajapakse, H. J. Org.
Chem. 1996, 61, 1194; (l) Bourghida, M.; Widhalm, M.
Tetrahedron: Asymmetry 1998, 9, 1073; (m) Arroyo, N.;
10. Mecca, T.; Superchi, S.; Giorgio, E.; Rosini, C. Tetra-
hedron: Asymmetry 2001, 12, 1225.
11. Braun, M. Angew. Chem., Int. Ed. Engl. 1996, 35, 519.
12. Corey, E. J.; Yuen, P.; Hannon, F. J.; Wierda, D. A. J.
Org. Chem. 1996, 55, 784.
13. Williams, D. R.; Fromhold, M. G. Synlett 1997, 523.
14. Reiners, I.; Martens, J. Tetrahedron: Asymmetry 1997, 8,
27.
.