Biheteroaromatic diphosphine oxides-catalyzed stereoselective direct aldol reactions

Article abstract of DOI:10.1016/j.tet.2010.11.009

A highly stereoselective direct aldol condensation of ketones to aromatic aldehydes was realized; the trichlorosilyl enolether generated in situ in the presence of tetrachlorosilane is activated by catalytic amounts of an enantiomerically pure biheteroaromatic phosphine oxide to react with aldehydes, coordinated as well as activated by the chiral cationic hypervalent silicon species. This Lewis acid-mediated Lewis base-catalyzed transformation allowed, starting from two carbonyl compounds, to directly synthesize β-hydroxy ketones generally with high anti stereoselectivity and up to 93% ee for the anti isomer.

Full text of DOI:10.1016/j.tet.2010.11.009

Tetrahedron 67 (2011) 158e166
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Biheteroaromatic diphosphine oxides-catalyzed stereoselective direct aldol
reactions
,
a
b
c
Sergio Rossi a, Maurizio Benaglia a , Andrea Genoni , Tiziana Benincori , Giuseppe Celentano
*
a
ꢀ
Dipartimento di Chimica Organica e Industriale, Universita degli Studi di Milano, Via Golgi 19, 24100 Milano, Italy
Dipartimento di Scienze Chimiche ed Ambientali dell’Universita dell’Insubria, Via Valleggio, 11, 22100 Como, Italy
Dipartimento di Dipartimento di Scienze Molecolari Applicate ai Biosistemi, Universita degli Studi di Milano, Via Golgi 19, 20133 Milano, Italy
b
ꢀ
c
ꢀ
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 August 2010
Received in revised form 14 October 2010
Accepted 1 November 2010
Available online 5 November 2010
A highly stereoselective direct aldol condensation of ketones to aromatic aldehydes was realized; the
trichlorosilyl enolether generated in situ in the presence of tetrachlorosilane is activated by catalytic
amounts of an enantiomerically pure biheteroaromatic phosphine oxide to react with aldehydes, co-
ordinated as well as activated by the chiral cationic hypervalent silicon species. This Lewis acid-mediated
Lewis base-catalyzed transformation allowed, starting from two carbonyl compounds, to directly syn-
thesize
b
-hydroxy ketones generally with high anti stereoselectivity and up to 93% ee for the anti isomer.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Phosphine oxides
Organocatalysis
Aldol reaction
Stereoselection
Chirality
1. Introduction
develop a series of catalysts displaying different electronic prop-
erties, where the influence of both the electronic availability of the
Small organic Lewis bases are protagonists in an incredibly rich
amount of chemistry and are efficient catalysts for a wide variety of
synthetic transformations.1 They include different classes of com-
pounds, such as naturally-occurring alkaloids and aminoacids,
synthetic amines-based catalysts and phosphines. While N-oxides
derived from both heteroaromatic systems2 and aliphatic amines3
have found widespread application in organocatalysis, quite sur-
prisingly phosphine oxides have been scarcely used.4
heterocyclic system and of the position of the phosphorus atoms on
the latter may be investigated10 (Scheme 1).
S
S
O
PPh2
O
PTol2
PPh2
O
PTol2
O
The first truly organocatalyzed reaction with chiral phosphine
oxides was reported by Nakajima, who employed substoichiometric
amounts of bis-(diphenylphosphanyl)-binaphthyl dioxide (BINAPO)
to promote the enantioselective addition of allyltrichlorosilane to
aldehydes.5 Based on the pioneering work by Denmark,6 later it was
shown that chiral phosphine oxides could also catalyze other nu-
cleophilic additions to carbonyl compounds,7 including the reaction
of trichlorosilyl enolethers8 and silyl keteneacetals9 with aldehydes.
However relatively few classes of phosphine oxides have been in-
vestigated and only bis-(diphenylphosphanyl)-binaphthyl dioxide
(BINAPO) has been used with some success.
(S) -tetra-Me-BITIOPO
(S) -Tol-BINAPO
SiCl3
OH
Ph-CHO
Ph
cat. (10% mol),
DIPEA (3 eq.), 0 °C
cat = Tol-BINAPO
cat = tetra-Me-BITIOPO
55% yield, 51% e.e.
85% yield, 93% e.e.
Our group has recently started to explore the activity of bihe-
teroaryldiphosphine oxides, which may offer the possibility to
Scheme 1. Organocatalytic enantioselective allylation of aromatic aldehydes.
It was found that electron-rich compounds, like (S)-tetramethyl-
bithiophene phosphine oxide, (S)-tetra-Me-BITIOPO, showed signif-
icant catalytic activity, promoting the addition of allyltrichlorosilane
* Corresponding author. E-mail address: maurizio.benaglia@unimi.it (M. Benaglia).
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.11.009

S. Rossi et al. / Tetrahedron 67 (2011) 158e166
159
to benzaldehyde at 0 ꢀC in higher yield (85%) and enantioselectivity
(93%) than the less electron-rich diphosphine oxide Tol-BINAPO.
Based on these promising results we decided to extend the use
of (S)-tetra-Me-BITIOPO to other reactions. In this field a real
breakthrough was achieved by Denmark, who developed phos-
phoroamide-catalyzed and SiCl4-mediated stereoselective re-
actions.11 The coordination of a Lewis base to silicon tetrachloride
makes the new hypervalent trichlorosilyl cationic species more
electrophilic; this new adduct of increased Lewis acidity was shown
to be able to promote the addition of different nucleophiles to
carbonyl compounds.1 However, until last year, basically only chiral
phosphoroamide-based catalysts have been successfully used.
Table 1
Tetra-Me-BITIOPO-catalyzed condensation of cyclohexanone with benzaldehyde
at 0 ꢀC
Entry Solvent R. time (h) Yielda (%) anti/syn Ratiob ee anti-(ee syn)c (%)
1d
2e
3
4
5
6
7
8
9
CH3CN
CH3CN
CH3CN
DCM
Toluene 15
THF
CHCl3
DCM
CH3CN
15
15
15
15
53
67
91
63
50
70
40
51
65
92/8
92/8
60 (33)
67 (15)
69 (37)
75 (42)
31 (7)
61 (17)
67 (41)
70 (31)
53 (15)
90/10
88/12
25/75
84/16
88/12
89/11
90/10
15
15
2
2
a
Reactions were run with 3 mol equiv of SiCl4, 2 mol equiv of ketone, 1 mol equiv
of aldehyde and 10 mol % amount of catalyst; yields were determined after chro-
matographic purification.
2. Results and discussion
b
Diastereoisomeric ratio was determined by 1H NMR and confirmed by HPLC.
Enantiomeric excess was determined by HPLC on chiral column (see
c
The possibility to promote reactions through a chiral Lewis acid
generated by coordination to SiCl4 of catalytic amounts of a chiral
phosphine oxide was obviously very attractive. The direct aldol
reaction12 of a ketone to aromatic aldehydes was selected for
a preliminary study in view of its importance in synthetic organic
chemistry and of its simple procedure that does not involve the
preparation of enolethers or keteneacetals. The trichlorosilyl eno-
lether is generated in situ in the presence of tetrachlorosilane and it
is activated by phosphine oxide to react with an aldehyde, that is,
also coordinated by the chiral cationic hypervalent silicon species.
Prompted by a recent work in the field by Nakajima where BINAPO
was employed as organocatalyst,13 we wish to report here our
studies on a direct aldol condensation of a ketone to aromatic al-
dehydes promoted by the chiral biheteroaryldiphosphine oxide
tetra-Me-BITIOPO.
Supplementary data).
d
Ketone (1 mol equiv) was used.
SiCl4 (1 mol equiv) was used.
e
the diastereoselectivity was observed when the reaction was con-
ducted in the non-coordinating solvent toluene, where a syn se-
lectivity was observed (entry 5). Shorter reaction times did not
bring appreciable variations of the stereochemical outcome of the
reaction (entries 8e9).
Lowering the reaction temperature generally had a positive ef-
fect on the stereocontrol of the aldol reaction, but it was detri-
mental for the chemical efficiency (Table 2). When the reaction
temperature was decreased to ꢁ45 ꢀC longer reaction times and
20 mol % of tetra-Me-BITIOPO catalyst were necessary in order to
obtain satisfactory chemical yields. Once again DCM secured better
performances than acetonitrile: after 36 h at ꢁ25 ꢀC in DCM, the
product was obtained in 70% yield, 93/7 anti/syn ratio and 83%
enantioselectivity for the anti isomer. Enantioselectivity was fur-
ther improved by running the reaction at ꢁ45 ꢀC, when the major
diastereoisomer was produced with 86% ee (entries 2e3).
Bases other than DIPEA were investigated in the attempt of fur-
ther increasing the chemical yield; however DABCO or NMM did not
bring any significant change. A comparison between the perfor-
mances of BITIOPO and BINAPO was made; by performing the re-
action in the reported literature conditions13 (CHCN, 0 ꢀC, 2 h) in our
hands BINAPO promoted the reaction in 64% yield, 89/11 anti/syn
ratio and 40% ee for the anti isomer, lower than enantioselection
observed with BITIOPO (53% ee, entry 6 Table 2 vs entry 9 Table 1). At
lower temperature bithiophene-based phosphine oxide confirmed
to perform better than binaphthyl-derived diphenylphosphine,
affording the aldol adduct in 93/7 anti/syn ratio and 80% ee, while
First the reaction between cyclohexanone and benzaldehyde
was investigated in the presence of stoichiometric amounts of SiCl4
and a catalytic amount of enantiomerically pure tetra-Me-BITIOPO
(Scheme 2).
O
OH
O
OH
O
O
cat (0.1-0.2 eq.)
H
+
+
SiCl4 (3 eq),
DIPEA
R
R
R
anti
syn
S
O
cat
(S) - tetra-Me-BITIOPO
PPh2
PPh2
O
S
Table 2
Scheme 2. Enantioselective direct aldol condensation of cyclohexanone with aromatic
Stereoselective aldol condensation of cyclohexanone with benzaldehyde
aldehydes.
Entry Solvent Temp (ꢀC) CAT
Yielda (%) anti/syn ee anti-
Different experimental conditions of solvent, temperature and
additives were screened in order to optimize the performance of
the Lewis base; a few selected results are shown in Table 1. The
ratiob
(ee syn)c (%)
1
2
CH3CN
DCM
DCM
DCM
DCM
ꢁ25
ꢁ25
ꢁ45
ꢁ25
ꢁ25
0
BITIOPO 30
BITIOPO 70
BITIOPO 44
BITIOPO 42
BITIOPO 40
81/19
93/7
79/21
76/24
91/9
65 (27)
83 (49)
86 (11)
57 (10)
65 (18)
40 (5)
3d
4e
5f
6g
7
direct condensation of
1 mol equiv of cyclohexanone with
1 mol equiv of benzaldehyde in the presence of 3 mol equiv of
silicon tetrachloride and 10 mol equiv of DIPEA afforded the cor-
CH3CN
DCM
BINAPO
BINAPO
64
40
89/11
79/21
responding
b
-hydroxy ketone 1 (Scheme 2, R¼H), after 15 h at 0 ꢀC
ꢁ25
60 (3)
in CH3CN in 53% yield, as mixture of anti/syn isomers (92/8) and
60% enantioselectivity for the major anti isomer (entry 1, Table 1).
By looking for best experimental conditions the reaction was found
to proceed better in dichloromethane, by employing 2 mol equiv of
ketone for 1 mol equiv of aldehyde; in that case the anti di-
astereoisomer was isolated as major product (88/12 anti/syn) in 75%
ee (entry 4, Table 1).
a
Reactions were run for 30 h with 3 mol equiv of SiCl4, 2 mol equiv of ketone,
1 mol equiv of aldehyde and 10 mol % amount of catalyst in the presence of DIPEA;
yields were determined after chromatographic purification.
b
Diastereoisomeric ratio was determined by 1H NMR and confirmed by HPLC.
Enantiomeric excess was determined by HPLC on chiral column (see
c
Supplementary data).
d
Reaction run in the presence of 20 mol % cat.
Reaction run in the presence of DABCO.
Reaction run in the presence of N-methylmorpholine.
Reaction time 2 h.
e
f
The use of other solvents was detrimental to the enantiose-
lectivity of the process, even if an interesting switch of the sense of
g

160
S. Rossi et al. / Tetrahedron 67 (2011) 158e166
OH
O
OH O
BINAPO led to the anti isomer in the same conditions with 60%
enantioselectivity and loweranti selectivity (entries 2 and 7 Table 2).
The general applicability of such a methodology to different al-
dehydes was then investigated; when cyclohexanone was reacted
at ꢁ25 ꢀC with 4-nitrobenzaldehyde in the presence of a catalytic
amount of Lewis base the product was isolated in 87/13
diastereoisomeric ratio and 93% ee for the major isomer (entry 3,
Table 3). A further decrease of the reaction temperature did not
improve the stereoselectivity (92% ee foranti isomer entry 5, Table 3).
O
cat (0.1-0.2 eq.)
O
R
R
+
+
R
H
SiCl4 (3 eq),
DIPEA
anti
syn
S
O
PPh2
cat
(S) - tetra-Me-BITIOPO
PPh2
O
S
Table 3
Tetra-Me-BITIOPO-catalyzed aldol condensation of cyclohexanone with different
aldehydes
Scheme 3. Reaction of cyclohexanone with different aldehydes.
Entry
R
Temp (ꢀC) Product Yielda (%) anti/syn ee anti-
ratiob
(ee syn)c (%)
1
2
3
Ph
Ph
ꢁ25
ꢁ45
1
1
2
2
2
3
3
4
5
6
7
8
70
34
71
51
50
41
56
55
57
65
56
55
42
40
55
36
41
d
93/7
79/21
87/13
90/10
86/14
81/19
98/2
90/10
92/8
94/6
94/6
98/2
92/8
93/7
98/2
98/2
90/10
d
81 (49)
86 (11)
93 (65)
85 (71)
92 (67)
88 (11)
71 (n.d.)
87 (27)
77 (33)
83 (n.d.)
73 (31.)
77 (n.d.)
87 (22)
77 (n.d.)
53 (n.d.)
51 (n.d.)
37 (n.d.)
d
Also in the aldol condensation of cyclopentanone with benzal-
dehyde tetra-Me-BITIOPO catalyzed the transformation with
higher stereoselectivity than BINAPO (83% ee vs 53% ee13) and with
very high anti selectivity (94%). Lower selectivity was observed in
the reaction of cycloheptanone, where no diastereoselectivity and
a modest enantioselectivity were observed.
Also heterocyclic ketones as well as 4,4-disubstituted cyclo-
hexanone may be employed as aldol donors in this Lewis base-
catalyzed SiCl4-mediated direct aldol condensation (see Scheme 4).
The products were generally obtained in high anti/syn ratios (>80%
dr) and enantioselectivity up to 86% ee Best result was obtained
with ketone 21 that afforded after condensation with benzaldehyde
the anti isomer of compound 22 in 93/7 dr and 86% ee.
The methodology could be applied to ketones containing oxy-
gen, nitrogen or sulfur atoms. Usually a good anti selectivity was
obtained with all substrates. For example ketone 24 was employed
in the condensation with benzaldehyde and afforded the aldol
product with good anti selectivity and 83% ee for the major isomer.
A tentative model of the stereoselectivity observed in the re-
actions promoted by tetra-Me-BITIOPO must take in account
mechanistic considerations. It is well known that the coordination
of a Lewis base to a Lewis acid makes it more electrophilic;1 the
chiral cationic hypervalent silicon species15 coordinates and acti-
vate the aldehyde towards the attack of the in situ generated tri-
chlorosilyl enolether.13
A chairlike six-membered cyclic transition state may be hy-
pothesized and would explain the anti selectivity generally ob-
served for these reactions16 (Fig. 1).
The steric hindrance of the diphenyl-phosphinoyl group of the
Lewis base with cyclohexanone-derived enolether would be re-
sponsible for the favourable attack of the enolether onto the Re face
of the aldehyde, to afford the experimentally observed major
(2S,10R)-anti stereoisomer.
Confirmation of the proposed cyclic transition state comes from
an additional experiment, where the aldol condensation between
benzaldehyde and cyclohexanone has been performed at ꢁ25 ꢀC in
the presence of a stoichiometric amount of BF3, a Lewis acid able to
coordinate the aldehyde. In that case the product was isolated in
65% yield as 60/40 syn/anti selectivity, both isomers basically as
racemic compounds, clearly suggesting the importance of the al-
dehydes coordination to the chiral cationic silicon species.
Preliminary NMR investigations17 seem to support the proposed
mechanism: 31P NMR analysis shows that the coordination of tetra-
Me-BITIOPO with SiCl4 at 0 ꢀC in CDCl3 caused a shift of the P signal
from 24.3 to 34.8 ppm (neutral SiCl4/phosphine oxide complex)
and, after addition of DIPEA, to 27.7 ppm. The same cationic silicon
species was observed when the ketone was treated with DIPEA in
the presence of SiCl4 and tetra-Me-BITIOPO (P signal at 28.0 ppm).18
Finally we decided to test our methodology in the cross-aldol
condensation between two aldehydes.19
4-NO2C6H4 ꢁ25
4-NO2C6H4 ꢁ25
4-NO2C6H4 ꢁ45
4d
5e
6e
7
4-ClC6H4
4-ClC6H4
2-ClC6H4
4-CF3C6H4
ꢁ45
ꢁ25
ꢁ25
ꢁ25
8
9
10
11
12
13e
14
15
16
17
18
19
3,5CF3C6H3 ꢁ25
2-MeC6H4
4-MeC6H4
ꢁ25
ꢁ25
4-OMeC6H4 ꢁ45
9
1-Naph
ꢁ25
ꢁ25
ꢁ25
ꢁ25
0
10
11
12
13
14
15
2-Thioph
2-Furyl
PhCH]CH
Ph(CH2)2
C(CH3)3
0
d
d
d
a
Reactions were run with 3 mol equiv of SiCl4, 2 mol equiv of ketone, 1 mol equiv
of aldehyde and 10 mol % amount of catalyst; yields were determined after chro-
matographic purification.
b
Diastereoisomeric ratio was determined by 1H NMR and confirmed by HPLC.
Enantiomeric excess was determined by HPLC on chiral column (see
c
Supplementary data).
d
Reaction was run in CH3CN.
With 20 mol % of BITIOPO.
e
Generally aromatic aldehydes bearing electron-withdrawing
groups reacted at ꢁ25 ꢀC with high anti selectivity in fair yields and
enantioselectivity often higher than 80% (entries 6e10 in Table 3)
(see Scheme 3).
The same high levelof stereoselectivity was maintained alsowith
other aromatic aldehydes with electron-donating residues. For ex-
ample, the reactionof cyclohexanonewith 4-methoxybenzaldehyde
led to the corresponding hydroxy ketone 9 in comparable 87%
enantioselectivity for the anti isomer (entry 13, Table 3) Also ortho-
substituted aromatic aldehydes may be suitable substrates for the
reaction, as shown by the case of 2-chlorobenzaldehyde, where the
product4 wasformedin 90/10 anti/synratioand 87%eeforthe major
isomer. The condensation of 2-methyl-benzaldehyde afforded the
corresponding aldol 7 in high diastereoselectivity (94/6) but lower
enantioselectivity (73% ee, entry 11, Table 3).
Heteroaromatic aldehydes were also investigated and proofed
to be suitable substrates for the present methodology; satisfactory
levels of anti selectivity were reached, while only modest enan-
tioselectivity was observed both with 2-thiophene carbox-
yaldehyde and 2-furfurol (entries 15e16).
Finally the aldol condensation was attempted with non aromatic
aldehydes; as expected aliphatic aldehydes did not react, even at
0 ꢀC,14 while cinnamic aldehyde reacted with cyclohexanone in 41%
yield 90/10 anti/syn ratio but only gave 37% enantioselectivity of the
major isomer.
The reaction of different ketones with aromatic aldehydes was
then investigated (Scheme 4).

S. Rossi et al. / Tetrahedron 67 (2011) 158e166
161
Cl-
Cl
+
LB
LB
Si
R
O
O
Cl
Ph
Cl
O P
Ph
Ph
H
O
O
P
Cl
H
Si
Cl
P
P
O
O
O
O
O
H
Ph
Si
Cl
H
Unfavored
Favored
Fig. 1. A proposed model of stereoselectivity.
0 ꢀC. More studies are needed in order to fully investigate the scope
of the methodology but the feasibility of the synthetic approach
was surely demonstrated (Scheme 5).
O
1) BITIOPO (0.1 eq.)
OH
O
SiCl4 (3 eq), DIPEA
H
OH
H
+
2) NaBH4
28
0°C, 55% yield, 48% e.e.
-45°C, 35% yield, 70% e.e.
O
1) BITIOPO (0.1 eq.)
O
OH
SiCl4 (3 eq), DIPEA
H
OH
H
+
2) NaBH4
29
0°C, 57% yield, 47% e.e.
-45°C, 45% yield, 55% e.e.
CHO
Cl
1) BITIOPO (0.1 eq.)
SiCl4 (3 eq), DIPEA
OH
O
O
OH
H
+
2) NaBH4
Cl
30
0°C, 55% yield, 48% e.e.
CHO
NO2
1) BITIOPO (0.1 eq.)
SiCl4 (3 eq), DIPEA
OH
OH
H
+
Scheme 4. Tetra-Me-BITIOPO-catalyzed aldol reaction of different ketones with aro-
matic aldehydes.
2) NaBH4
O2N
31
0°C, 55% yield, 93% e.e.
Scheme 5. Lewis base-catalyzed cross-aldol reaction.
In very early studies the reaction of isobutyraldehyde with
benzaldehyde was attempted in DCM in the presence of 3 equiv of
SiCl4 and 0.1 equiv of tetra-Me-BITIOPO, followed by reduction of
the carbonyl group by treatment with NaBH4. By working at 0 ꢀC 1-
phenyl, 2,2-dimethyl propanediol 28 was isolated in 55% yield and
48% ee, that was increased up to 70% by running the reaction
at ꢁ45 ꢀC.
3. Conclusion
In conclusion, we have realized a direct aldol condensation cat-
alyzed by tetra-Me-BITIOPO, a chiral bithiophene-based phosphine
oxide. This Lewis base was shown to perform better than carboxylic
The cross condensation between benzaldehyde and cyclohex-
ane carboxyaldehyde afforded the 1,3-diol 29 with lower enantio-
selectivity. By exploring the use of other aromatic aldehydes in the
reaction with isobutyraldehyde different results were observed;
while 4-chloro benzaldehyde afforded the product 30 in 48% ee
only, 4-nitrobenzaldehyde led to the formation of 31 in 93% ee at
aromatic derivatives, such as BINAPO, affording b-hydroxy ketones
with good anti stereoselectivity, and enantioselectivity often higher
than 85%and upto95%. The methodology was successfullyextended
to structurally different ketones, including compounds containing

162
S. Rossi et al. / Tetrahedron 67 (2011) 158e166
heteroatoms, such as nitrogen, oxygen and sulfur. Finally the
methodology was showed to be feasible for the development of
cross-aldol condensation between an aromatic and an aliphatic
aldehyde.
The major drawback of the proposed catalytic system is the
sometimes low chemical efficiency. However, even if many issues
still need to be tackled, the results clearly show that a still un-
explored novel class of chiral biheteroaryl-based phosphine ox-
ides are now available as suitable catalysts, whose electronic and
steric properties could be modulated by a proper choice of
substituents.
513
m
l) tetrachlorosilane (1.5 equiv, 0.45 mmol, 52
ml); benzalde-
hyde (1 equiv, 0.30 mmol, 30
m
l).
After a proper time (see tables) the reaction was quenched by
the addition of a satd aqueous solution of NaHCO3 (3 mL). The
mixture was allowed to warm up to room temperature and stirred
for 30 min, then water (5 mL) and ethyl acetate (15 mL) were
added. The two-layers mixture was separated and the aqueous
layer was extracted with ethyl acetate (15 mL). The combined
organic layers were washed with 10% HCl (20 mL), sat NaHCO3
(20 mL) and brine (20 mL), dried over Na2SO4, filtered and con-
centrated under vacuum at room temperature. The crude product
was purified by column chromatography with different hexane/
ethyl acetate mixture as eluent (see below) to afford the pure aldol
adducts. Yields and enantiomeric excess for each reaction are in-
dicated in the tables. The syn/anti ratio was calculated by 1H NMR
spectroscopy; the enantiomeric excess was determined by HPLC
(S)-tetra-Me-BITIOPO was quantitatively recovered by further
elution with 10% MeOH in CH2Cl2 without any loss of optical pu-
rity (as determined by HPLC analysis on chiral column, recovered
4. Experimental
4.1. General methods
All reactions were carried out in oven-dried glassware with
magnetic stirring under nitrogen atmosphere, unless otherwise
stated. Dry solvents were purchased and stored under nitrogen
over molecular sieves (bottles with crown cap). Reactions were
monitored by analytical thin-layer chromatography (TLC) using
silica gel 60 F254 pre-coated glass plates (0.25 mm thickness) and
visualized using UV light or phosphomolibdic acid. Proton NMR
spectra were recorded on spectrometers operating at 200, 300 or
500 MHz, respectively. Proton chemical shifts are reported in parts
25
25
catalyst [
a
]
ꢁ65.3 (c 0.2, C6H6), known value [
a
]
ꢁ68 (c 0.5,
D
D
C6H6)10).
4.2.1. Characterization of products 2-(hydroxyphenylmethyl)cyclo-
hexan-1-one (1). This product was purified by flash column chro-
matography on silica gel with a 9:1 hexane/ethyl acetate mixture as
eluent. The purification afforded a mixture of anti and syn aldol
adducts. Its 1H NMR data were in agreement with those reported in
the literature.20,21
per million (d) with the solvent reference relative to tetrame-
thylsilane (TMS) employed as the internal standard (CDCl3
d
¼7.26 ppm). 13C NMR spectra were recorded on 300 or 500 MHz
spectrometers operating at 75 and 125 MHz, respectively, with
complete proton decoupling. Carbon chemical shifts are reported in
Data for anti: Rf¼0.21 (Hex/EtOAc 9:1 stained blue with phos-
phomolibdic acid) 1H NMR (200 MHz, CDCl3):
d 7.31e7.24 (m, 5H),
parts per million (
resonance as the internal standard (CDCl3,
d
) relative to TMS with the respective solvent
¼77.0 ppm). Optical
4.79 (d, 1H, J¼8.6 Hz), 3.95 (br s, 1H), 2.72e2.35 (m, 3H), 2.12e2.05
d
(m, 1H), 1.71e1.52 (m, 4H), 1.31e1.26 (m, 1H).
rotations were obtained on a polarimeter at 589 nm using a 5 mL
cell with a length of 1 dm. HPLC for enantiomeric excess de-
termination was performed under the conditions reported below.
Mass spectra (MS) were performed at CIGA (Centro Inter-
dipartimentale Grandi Apparecchiature), with Mass Spectrometer
APEX II & Xmass software (Bruker Daltonics).
Data for syn: Rf¼0.32 (Hex/EtOAc 9:1 stained blue with phos-
phomolibdic acid) 1H NMR (200 MHz, CDCl3):
d 7.38e7.25 (m, 5H),
5.39 (d, 1H), 2.60 (m, 1H), 2.60e2.32 (m, 2H), 2.08e2.01 (m, 1H),
1.87e1.29 (m, 5H).
The enantiomeric excess was determined by chiral HPLC with
Daicel Chiralpak OD-H column [eluent: 98:2 Hex/IPA; 0.8 mL/min
flow rate, detection: 210 nm; tR 19.7 min (syn-minor), tR 21.8 min
(syn-major), tR 27.8 min (anti-major (10R, 2S)), tR 44.9 min (anti-
minor (10S, 2R)). The absolute configuration of aldol products was
determined by comparison of the literature data].
Starting materials: 2,20,5,50-Tetramethyl-4,40-bis-(diphenyl-phos-
hino)-3,30-bithiophene oxide (tetra-Me-BITIOPO) was prepared
by literature method;1 ketones (cyclohexanone, cyclopentanone,
1,4-cycloexandion-monoethylenacetal, tetra-hydro-4H-pyran-4-one,
dihydro-2H-thiopyran-4(3H)-one, 1-Z-4-piperidone,) and aldehydes
(benzaldehyde, p-Cl-benzaldehyde, o-Cl-benzaldehyde, p-NO2-
benzaldehyde, p-OMe-benzaldehyde, p-Me-benzaldehyde, o-Me-
benzaldehyde, thiophene-2-carbaldehyde, furan-2-carbaldehyde,
p-CF3-benzaldehyde, 1-naphthaldehyde 3,5-bis(trifluoromethyl)-
benzaldehyde, cyclohexane carboxyaldehyde, isobutyraldehyde)
were obtained commercially and used after immediately distillation.
SiCl4 was also freshly distilled before use.
4.2.2. 2-(Hydroxy-(4-nitrophenyl)methyl)cyclohexan-1-one
(2). This product was purified by flash column chromatography on
silica gel with a 7:3 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the
literature.21
Data for anti: Rf¼0.11 (Hex/EtOAc 7:3, stained blue with phos-
25
phomolibdic acid) [
a
]
D
25 þ12.3 (c 0.26, CHCl3), 95% ee, [
a
d
]
Dmax þ12.8
4.2. Typical procedure of enantioselective direct aldol-type
reactions between ketones and aldehydes
(c 0.26, CHCl3), >99% ee.21 1H NMR (200 MHz, CDCl3):
8.20 (d, 2H,
J¼8.5 Hz), 7.51 (d, 2H, J¼8.5 Hz), 4.90 (d, 1H), 2.65e2.55 (m, 1H),
2.55e2.25 (m, 1H), 2.20e1.95 (m, 1H), 1.95e1.70 (m, 1H), 1.65e1.15
(m, 5H).
To a stirred solution of (S)-tetra-Me-BITIOPO (0.1 or 0.2 equiv) in
the chosen solvent (2 mL), the ketone (2 equiv) and diisopropyle-
thylanime (10 equiv) were added. The mixture was then cooled to
the chosen temperature and freshly distilled tetrachlorosilane
(1.5 equiv) was added dropwise via syringe. After 15 min, freshly
distilled aldehyde (1 equiv) was added. The mixture was stirred for
5 h (if the operating temperature is 0 ꢀC) or 12 h (if the operating
temperature is ꢁ25 ꢀC), then the same amount of tetrachlorosilane
(1.5 equiv) was added.
Data for syn: Rf¼0.15 (Hex/EtOAc 7:3, stained blue with phos-
phomolibdic acid) 1H NMR (200 MHz, CDCl3):
d 8.23 (d, 2H,
J¼8.8 Hz), 7.51 (d, 2H, J¼8.8 Hz), 5.48 (s, 1H), 2.70e2.65 (m, 1H),
2.65e2.30 (m, 2H), 2.20e2.05 (m, 1H), 1.95e1.35 (m, 4H),
1.30e1.20 (m, 1H). The enantiomeric excess was determined by
chiral HPLC with Daicel Chiralpak OJ-H column [eluent: 7:3 Hex/
IPA; 0.8 mL/min flow rate, detection: 210 nm; tR 14.3 min (syn-
major), tR 23.4 min (syn-minor), tR 12.9 min (anti-major (10R, 2S)),
tR 15.6 min (anti-minor (10S, 2R)). The absolute configuration of
aldol products was confirmed by comparison of the optical rota-
tion values.
For example, in Table 3, entry 1 (typical procedure), the quan-
tities are: (S)-tetra-Me-BITIOPO (0.1, 0.03 mmol, 18.7 mg); cyclo-
hexanone (2 equiv, 0.60 mmol, 62 ml); DIPEA (10 equiv, 3 mmol,

S. Rossi et al. / Tetrahedron 67 (2011) 158e166
163
4.2.3. 2-(Hydroxy-(4-chlorophenyl)methyl)cyclohexan-1-one
(3). This product was purified by flash column chromatography on
silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the
literature.3,7
silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the
literature.10
Data of a mixture syn/anti: Rf¼0.22 (Hex/EtOAc 9:1 stained blue
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d 7.42e7.15
Data of a mixture syn/anti: Rf¼0.25 (Hex/EtOAc 9:1, stained blue
(m, 4H), 5.60 (m, 1H, syn) 5.15 (d, J¼8.6 Hz, anti), 3.93 (br s, 1H),
2.60e2.55 (m, 1H), 2.55e1.95 (m, 3H), 1.90e1.00 (m, 8H). The en-
antiomeric excess was determined by chiral HPLC with Daicel
Chiralcell IB column [eluent: 99:1 Hex/IPA; 0.8 mL/min flow rate,
detection: 210 nm; tR 11.4 min (syn-minor), tR 12.5 min (syn-major),
tR 18.1 min (anti-major), tR 20.6 min (anti-minor).
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d
7.33e7.22 (m, 4H), 5.35 (s, 1H syn), 4.78 (d, 1H, J¼8.7 Hz, anti),
3.04e2.20 (m, 3H), 2.20e1.95 (m, 1H), 1.95e1.40 (m, 3H), 1.40e1.05
(m, 1H), 1.05e0.85 (m, 1H). The enantiomeric excess was de-
termined by chiral HPLC with Daicel Chiralcell OD column [eluent:
95:5 Hex/IPA; 0.8 mL/min flow rate, detection: 210 nm; tR 11.06 min
(syn-major), tR 11.82 min (syn-minor), tR 15.7 min (anti-major), tR
25.5 min (anti-minor)].
4.2.8. 2-(Hydroxy-(4-methylphenyl)methyl)-cyclohexan-1-one
(8). This product was purified by flash column chromatography on
silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the
literature.10
4.2.4. 2-(Hydroxy-(2-chlorophenyl)methyl)cyclohexan-1-one
(4). This product was purified by flash column chromatography on
silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the literature.7
Data of a mixture syn/anti: Rf¼0.10 (Hex/EtOAc¼9:1, stained blue
Data of a mixture syn/anti: Rf¼0.18 (Hex/EtOAc 9:1 stained blue
with phosphomolibdic acid). 1H NMR (300 MHz, CDCl3):
d 7.13e7.21
(m, 4H), 5.38 (m, 1H, syn) 4.73e4.75 (d, J¼8.6 Hz, anti), 3.87 (br s,
1H), 2.59e2.48 (m, 1H), 2.44e2.20 (m, 2H), 2.18e2.09 (m, 1H),
1.78e1.42 (m, 4H), 1.30e1.12 (m, 4H). The enantiomeric excess was
determined by chiral HPLC with Daicel Chiralcell IB column [eluent:
99:1 Hex/IPA; 0.8 mL/min flow rate, detection: 213 nm; tR 19.0 min
(anti-major), tR 21.2 min (anti-minor).
with phosphomolibdic acid). 1H NMR (300 MHz, CDCl3):
d 7.55 (d,
1H, J¼8.4 Hz), 7.35e7.20 (m, 3H), 5.75 (s, 1H syn), 5.35 (d, 1H,
J¼8.0 Hz, anti), 4.10e3.85 (br s, 1H), 2.71e2.65 (m, 1H), 2.59e2.46
(m, 1H), 2.45e2.25 (m, 1H), 2.15e2.00 (m, 1H), 1.90e1.50 (m,
5H).The enantiomeric excess was determined by chiral HPLC with
Daicel Chiralcell AD column [eluent: 8:2 Hex/IPA; flow rate: 0.8 mL/
min; detection: 210 nm; tR: 6.4 min (syn-major), tR: 6.9 (syn-minor),
tR: 9.0 min (anti-major), tR: 10.2 min (anti-minor)].
4.2.9. 2-(Hydroxy-(4-nitrophenyl)methyl)cyclohexan-1-one
(9). This product was purified by flash column chromatography on
silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the literature.3
Data of a mixture syn/anti Rf¼0.10 (Hex/EtOAc 9:1, stained blue
4.2.5. 2-(Hydroxy-(4-trifluoromethylphenyl)methyl)-cyclohexan-1-
one (5). This product was purified by flash column chromatogra-
phy on silica gel with a 9:1 hexane/ethyl acetate mixture as eluent.
The purification afforded a mixture of anti and syn aldol adducts. Its
1H NMR data were in agreement with those reported in the
literature.9
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d 7.25 (dd,
2H, J¼8.4 Hz, J¼14.6 Hz), 6.89 (d, 2H, J¼8.4 Hz, J¼14.6 Hz), 5.30 (s,
1H, syn), 4.76 (d, 1H, J¼8.8 Hz, anti), 3.82 (s, 3H), 2.65e2.45 (m, 1H),
2.45e2.25 (m, 2H), 2.20e2.00 (m, 1H), 1.85e1.65 (m, 2H), 1.65e1.45
(m, 2H), 1.35e1.10 (m, 1H). The enantiomeric excess was de-
termined by chiral HPLC with Daicel Chiralcell AD column [eluent:
97:3 Hex/IPA; 0.8 mL/min flow rate, detection: 225 nm; tR 37.9 min
(syn-major), tR 46.4 min (syn-minor), tR 74.0 min (anti-minor), tR
80.1 min (anti-major)].
Data of a mixture syn/anti: Rf¼0.13 (Hex/EtOAc 9:1 stained blue
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d
7.70e7.60 (m, 2H), 7.57e7.33 (m, 2H), 5.44 (m, 1H, syn), 4.86e4.88
(d, 1H, J¼8.55 Hz, anti), 4.04 (s, 1H), 2.70e2.20 (m, 3H), 2.15e2.05
(m, 1H), 1.85e1.25 (m, 5H). The enantiomeric excess was de-
termined by chiral HPLC with Daicel Chiralcell IB column [eluent:
99:1 Hex/IPA; 0.8 mL/min flow rate, detection: 213 nm; tR 13.5 min
(syn-major), tR 14.5 min (syn-minor), tR 18.7 min (anti-major), tR
21.2 min (anti-minor).
4.2.10. 2-(Hydroxyl-1-napthylmethyl)cyclohexan-1-one (10). This
product was purified by flash column chromatography on silica gel
with a 9:1 hexane/ethyl acetate mixture as eluent. The purification
afforded a mixture of anti and syn aldol adducts. Its 1H NMR data
were in agreement with those reported in the literature.8
4.2.6. 2-(Hydroxy-(3,5–bis(trifluoromethyl)phenyl)methyl)-cyclo-
hexan-1-one (6). This product was purified by flash column chro-
matography on silica gel with a 95:5 hexane/ethyl acetate mixture
as eluent afforded a mixture of anti and syn aldol adducts.
Data of a mixture syn/anti: Rf¼0.14 (Hex/EtOAc 95:5, stained
Data of a mixture syn/anti: Rf¼0.18 (Hex/EtOAc 9:1 stained blue
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d
8.24e8.27 (m, 1H), 7.80e7.89 (m, 2H), 7.45e7.56 (m, 4H), 6.25 (br
blue with phosphomolibdic acid). 1H NMR (300 MHz, CDCl3):
d
7.80
s, 1H, syn), 5.60 (d, J¼9.0 Hz, anti), 4.16 (br, 1H), 3.05 (m, 1H),
2.35e2.51 (m, 2H), 2.06e2.12 (m, 2H), 1.65e1.75 (m, 2H), 1.36e1.52
(m, 2H). The enantiomeric excess was determined by chiral HPLC
with Daicel Chiralcell AD column [eluent: 9:1 Hex/IPA; 0.8 mL/min
flow rate, detection: 210 nm; tR 23.9 min (anti-minor), tR 28.8 min
(anti-major).
(m, 3H), 5.50 (s, 1H, syn), 4.92 (d, 1H, J¼8.4 Hz, anti,), 4.16 (s, 1H),
2.75e2.25 (m, 2H), 2.25e2.00 (m, 1H), 2.00e1.80 (m, 1H), 1.75e1.40
(m, 3H), 1.40e1.05 (m, 2H). 13C NMR (300 MHz, CDCl3):
d 214.53 (s),
143.79 (s), 131.56 (q), 127.24 (s), 123.14 (q), 73.97 (s), 42.64 (s), 30.67
(s), 27.58 (s), 24.66 (s). Mass (ESIþ): m/z¼[MþNa]þ calcd for
C15H14O2F6Na 363.07902, found 363.07858 [MþNa]þ. The enan-
tiomeric excess was determined by chiral HPLC with Daicel Chir-
alcell OD-H column [eluent: 95:5 Hex/IPA; 0.8 mL/min flow rate,
detection: 254 nm; tR 9.0 min (anti-major), tR 10.2 min (anti-
minor)].
4.2.11. 2-(Hydroxyl-(thiophen-2-yl)methyl)cyclohexan-1-one
(11). This product was purified by flash column chromatography
on silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the literature9.
Data of a mixture syn/anti: Rf¼0.16 (Hex/EtOAc 9:1 stained blue
4.2.7. 2-(Hydroxy-(2-methylphenyl)methyl)-cyclohexan-1-one
(7). This product was purified by flash column chromatography on
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d 7.30 (m,

164
S. Rossi et al. / Tetrahedron 67 (2011) 158e166
1H), 6.93e6.95 (m, 2H), 5.52 (m, 1H, syn), 5.05e5.08 (d, J¼8.37 Hz,
anti), 4.05 (br,1H), 2.80e2.60 (m,1H), 2.60e2.20 (m, 2H), 2.20e2.05
(m, 1H), 1.90e1.50 (m, 4H), 1.40e1.20 (m, 1H). The enantiomeric
excess was determined by chiral HPLC with Daicel Chiralcell OD-H
column [eluent: 9:1 Hex/IPA; 0.8 mL/min flow rate, detection:
230 nm; tR 8.93 min (anti-major), tR 12.7 min (anti-minor).
4.2.16. Tetrahydro-3-(hydroxyphenylmethyl)-4H-pyran-4-one
(20). This product was purified by flash column chromatography
on silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the literature.2
Data of a mixture syn/anti: Rf¼0.12 (Hex/EtOAc¼9:1, stained
stained blue with phosphomolibdic acid).1H NMR (300 MHz, CDCl3):
4.2.12. 2-(Hydroxyl-(furan-2-yl)methyl)cyclohexan-1-one (12). This
product was purified by flash column chromatography on silica gel
with a 9:1 hexane/ethyl acetate mixture as eluent. The purification
afforded a mixture of anti aldol adduct. Its 1H NMR data were in
agreement with those reported in the literature.9
d
7.52e7.47 (m, 5H), 5.75 (s, 1H, syn), 4.91 (m, 1H, J¼8.2 Hz, anti),
4.20e3.75(m, 4H), 3.05(m,1H), 2.70e2.45(m, 2H). Theenantiomeric
excess was determined by chiral HPLC with Daicel Chiralcell AD
column [eluent: 9:1 Hex/IPA; flow rate: 0.8 mL/min; detection:
270 nm; tR: 19.5 min (anti-major), tR 23.5 min (anti-minor)].
Data of anti: Rf¼0.21 (Hex/EtOAc 9:1 stained blue with phos-
phomolibdic acid). 1H NMR (200 MHz, CDCl3):
d
7.40 (s, 1H),
4.2.17. 4,4-(Ethylenedioxy)-2-(hydroxyphenylmethyl)cyclohexan-1-
one (22). Thisproduct waspurifiedbyflashcolumnchromatography
on silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the literature.2
Data of a mixture syn/anti: Rf¼0.10 (Hex/EtOAc¼8:2, stained
stained blue with phosphomolibdic acid).1H NMR (300 MHz, CDCl3):
6.27e6.33 (m, 2H), 4.81e4.84 (dd, J¼8.22 Hz), 3.85 (d, 1H,
J¼3.68 Hz), 2.97e2.83 (m, 1H), 2.50e2.30 (m, 2H), 2.20e2.05 (m,
1H), 1.95e1.45 (m, 4H), 1.40e1.25 (m, 1H). The enantiomeric excess
was determined by chiral HPLC with Daicel Chiralcell IB column
[eluent: 99:1 Hex/IPA; 0.8 mL/min flow rate, detection: 213 nm; tR
23.2 min (anti-major), tR 26.3 min (anti-minor).
d
7.32e7.26 (m, 5H), 5.43 (m, 1H, syn), 4.83 (d, 1H, J¼8.5 Hz, anti),
4.2.13. 2-(1-Hydroxy-3-phenyl-2-propenyl)cyclohexanone (13). This
product was purified by flash column chromatography on silica gel
with a 9:1 hexane/ethyl acetate mixture as eluent. Its 1H NMR data
were in agreement with those reported in the literature.11
3.88e3.82 (m, 5H), 3.05e2.95 (m, 1H), 2.80e2.65 (m,1H), 2.55e2.40
(m, 1H), 2.10e1.90 (m, 2H), 1.60e1.45 (m, 2H). The enantiomeric ex-
cess was determined by chiral HPLC with Daicel Chiralcell OD-H
column [eluent: 95:5 Hex/IPA; 0.8 mL/min flow rate; detection:
210 nm; tR: 20.8 min (anti-major), tR 29.0 min (anti-minor)].
TLC: Rf 0.27 (Hex/EtOAc¼9:1, stained blue with phosphomo-
lybdic acid). 1H NMR (300 MHz, CDCl3):
d 7.40e7.10 (m, 10H),
6.70e6.50 (m, 2H), 4.77 (br, 1H, syn), 4.43 (t, 1H, anti), 3.60 (br, 1H),
2.60e2.30 (m, 6H), 2.20e2.00 (m, 2H), 1.90e1.80 (m, 2H), 1.80e1.30
(m, 8H). The enantiomeric excess was determined by chiral HPLC
with Daicel Chiralcell AD column [eluent 97:3 Hex/IPA; flow rate:
0.5 mL/min; detection: 254 nm; tR: 47.7 min (syn-major), tR:
52.2 min (syn-minor), tR: 56.7 min (anti-major), tR: 66.9 min (anti-
minor)].
4.2.18. Tetrahydro-3-(hydroxyphenylmethyl)-4H-thiopyran-4-one
(25). This product was purified by flash column chromatography
on silica gel with a 9:1 hexane/ethyl acetate mixture as eluent. The
purification afforded a mixture of anti and syn aldol adducts. Its 1H
NMR data were in agreement with those reported in the literature.2
Rf¼0.21 (Hex/EtOAc 9:1 stained blue with phosphomolibdic
acid).1HNMR (300 MHz, CDCl3):
d7.36e7.24(m, 5H), 5.38 (s,1H syn),
4.97 (dd,1H, J¼3.2, 8.7 Hz, anti), 3.37 (d,1H, J¼3.2 Hz), 3.01e2.85 (m,
3H), 2.85e2.65 (m, 2H), 2.65e2.55 (m, 2H). The enantiomeric excess
was determined by chiral HPLC with Daicel Chiralcell OD column
[eluent: 97:3 hex/IPA; 1 mL/min flow rate, detection: 210 nm; tR:
28.3 min (anti-major), tR: 39.5 min (anti-minor)].
4.2.14. 2-(Hydroxyphenylmethyl)cyclopentan-1-one(16). This product
was purified by flash column chromatography on silica gel with
a 9:1 hexane/ethyl acetate mixture as eluent. The purification affor-
ded a mixture of anti and syn aldol adducts. 1H NMR data were in
agreement with those reported in the literature.2,4e6
Data of a mixture syn/anti: Rf¼0.27 (Hex/EtOAc 8:2, stained blue
4.2.19. 3-(Hydroxy(phenyl)methyl)-4-oxopiperidine-1-benzylcarbox-
ylate (27). This product was purified by flash column chromatog-
raphy on silica gel with a 9:1 hexane/ethyl acetate mixture as
eluent afforded a mixture of anti and syn aldol adducts.
with phosphomolibdic acid). 1H NMR (200 MHz, CDCl3):
d
7.25e7.35 (m, 5H), 5.29 (br s, 1H syn), 4.72 (d, 1H, J¼9.1 Hz, anti),
2.49e2.25 (m, 3H), 2.25e1.85 (m, 1H), 1.85e1.30 (m, 3H). The en-
antiomeric excess was determined by chiral HPLC with Daicel
Chiralcell OD-H column [eluent: 9:1 Hex/IPA; 0.5 mL/min flow rate,
detection: 210 nm; tR 14.1 min (syn-minor), tR 16.6 min (syn-major)
tR 21.1 (anti-major (2S, 10R)), tR 25.7 min (anti-minor (2R, 10S)). The
absolute configuration of aldol products was determined by com-
parison of the literature data].
Data of a mixture syn/anti: Rf¼0.14 (Hex/EtOAc 9:1, stained blue
with phosphomolibdic acid).1H NMR (300 MHz, CDCl3):
d 7.40e7.20
(m, 10H), 5.35 (m, 1H, syn), 5.20e5.00 (m, 2H), 4.85 (d, 1H, J¼4 Hz,
anti), 4.25e4.15 (m, 1H, anti), 4.05e3.95 (m, 1H, syn), 3.85e3.65 (m,
2H), 3.60e3.30 (m, 2H), 3.10e2.40 (m, 1H), 2.60e2.40 (m, 2H),
1.75e1.50 (m, 1H). 13C NMR (300 MHz, CDCl3):
d 213.2 (s), 155.0 (s),
141.4 (s), 128.6 (m), 126.8 (s), 72.7 (s), 67.6 (s), 56.6 (s), 45.9 (s), 43.8
(s), 41.3 (s), 40.8 (s). Mass (ESIþ): m/z¼[MþNa]þ calcd for
C20H21O4NNa 362.13628, found 362.13599 [MþNa]þ. The enantio-
meric excess was determined by chiral HPLC with Daicel Chiralpak
OD column [eluent: 97:3 Hex/IPA; 1 mL/min flow rate, detection:
210 nm; tR 37.4 min (anti-minor), tR 41.0 min (anti-major)].
4.2.15. 2-(Hydroxyl-(4-nitrophenyl)methyl)cyclopentan-1-one
(17). This product was purified by flash column chromatography
on silica gel with a 9:1 then 8:2 hexane/ethyl acetate mixture as
eluent. The purification afforded a mixture of anti and syn aldol
adducts. Its 1H NMR data were in agreement with those reported in
the literature.3
Data of a mixture syn/anti: Rf¼0.12 (Hex/EtOAc 9:1 then 8:2,
4.3. Typical procedure of enantioselective direct aldol-type
reactions between aldehydes
stained blue with phosphomolibdic acid). 1H NMR (300 MHz,
CDCl3):
d
8.22 (d, 2H, J¼8.6), 7.54 (d, 2H, J¼8.6), 5.41 (d, 1H, J¼2.8,
syn), 4.86 (d,1H, J¼9 Hz, anti), 2.50e2.10 (m, 2H), 2.10e1.85 (m, 2H),
1.85e1.60 (m, 2H), 1.60e1.40 (m, 1H). The enantiomeric excess was
determined by chiral HPLC with Daicel Chiralcell AD column [elu-
ent: 95:5 Hex/IPA; 0.8 flow rate, detection: 254 nm; tR 24.6 min
(syn-minor), tR 31.5 min (syn-major) tR 160.8 min (anti-minor (10R,
2S)), tR 171.3 min (anti-major (10S, 2R))].
Tetrachlorosilane (1.5 equiv, 0.45 mmol, 52
lution of isobutyraldehyde (1.5 equiv, 0.45 mmol, 42
hyde (1.0 equiv, 0.30 mmol, 30 l), diisopropylethylamine (5 equiv,
1.50 mmol, 185 l) and (S)-tetra-Me-BITIOPO (0.1 equiv) in CH2Cl2
ml) was added to a so-
ml), benzalde-
m
m
(2 mL) at the chosen temperature. After being stirred for 12 h, the
reaction was quenched with satd NaHCO3 (3 mL) and the slurry was

S. Rossi et al. / Tetrahedron 67 (2011) 158e166
165
stirred for 0.5 h at room temperature. The mixture was extracted with
EtOAc (15 mL). The combined organic layers were washed with brine
(2ꢂ15 mL), dried over Na2SO4, filtered and concentrated. The crude
product was used for the next step without further purification. Tet-
rachlorosilane(1.5equiv)wasaddedtoasolutionofisobutyraldehyde
(1.5 equiv), benzaldehyde (1.0 equiv), diisopropylethylamine
(10 equiv) and (S)-tetra-Me-BITIOPO (0.1 equiv) in CH2Cl2 (2 mL) at
the chosen temperature. After being stirred for 12 h, the reaction was
quenchedwithsatd NaHCO3 (3mL)andtheslurrywasstirred for0.5h
at room temperature. The mixture was extracted with EtOAc (15 mL).
The combined organic layers were washed with brine (2ꢂ15 mL),
dried over Na2SO4, filtered and concentrated. The crude product was
used for the next step without further purification.
(s, 1H), 3.45 (q, 2H), 0.85 (s, 3H), 0.80 (s, 3H). The enantiomeric
excess was determined by chiral HPLC with Daicel Chiralpak AD
column [eluent 98:2 Hex/IPA; flow rate: 0.8 mL/min; detection:
210 nm; tR: 46.2 min (minor), tR: 48.2 min (major)].
4.3.5. 2,2-Dimethyl-1-(4-nitrophenyl)-1,3-propanediol (31). This
product was purified by flash column chromatography on silica
gel with a 7:3 hexane/ethyl acetate mixture as eluent. Its 1H NMR
data were in agreement with those reported in the literature.22
TLC: Rf 0.19 (Hex/EtOAc¼7:3, stained blue with phosphomolyb-
dic acid). 1H NMR (300 MHz, CDCl3):
d 8.25 (d, 2H), 7.55 (d, 2H), 4.78
(m, 1H), 3.59 (br, 1H), 3.57 (d, 2H), 2.40 (br, 1H), 0.86 (s, 6H). The
enantiomeric excess was determined by chiral HPLC with Daicel
Chiralpak AD column [eluent 98:2 Hex/IPA; flow rate: 0.8 mL/min;
detection: 270 nm; tR: 53.1 min (MINOR), tR: 70.3 min (major)].
NaBH4 (1.5 equiv, 0.45 mmol, 38 mg) was added to the solution
of the obtained crude product in MeOH (5 mL), and the reaction
mixture was stirred for 30 min at room temperature. The reaction
was quenched with satd NH4Cl (5 mL) and extracted by CH2Cl2
(3ꢂ10 mL). The combined organic layer was washed with brine
(10 mL) and dried over Na2SO4. Evaporation of the solvent fur-
nished the crude product, which was purified by column chroma-
tography (Hex/EtOAc¼8:2) to give the desired diol.
Acknowledgements
Financial support from MIUR (Rome) within the national project
‘Nuovi metodi catalitici stereoselettivi e sintesi stereoselettiva di
molecole funzionali’ is gratefully acknowledged by M.B. The au-
thors thank Chemi S.p.A for generous gift of (S)-tetramethyl-
BITIOPO. T.B. thanks the Dipartimento di Chimica Organica e
Industriale for hospitality.
(S)-Tetra-Me-BITIOPO was quantitatively recovered by further
elution with 10% EtOH in CH2Cl2 without any loss of optical purity.
4.3.1. 2,2-Dimethyl-1-phenyl-1,3-propanediol (28). This product
was purified by flash column chromatography on silica gel with
a 9:1 hexane/ethyl acetate mixture as eluent. The purification
afforded a mixture of anti and syn aldol adducts. Its 1H NMR data
were in agreement with those reported in the literature.21
Supplementary data
Experimental details of the direct aldol condensation. Charac-
terization details for reaction products. 1H NMR spectra and HPLC
chromatograms on chiral stationary phase. Supplementary data
associated with this article can be found in online version, at
doi:10.1016/j.tet.2010.11.009. These data include MOL files and
InChIKeys of the most important compounds described in this
article.
TLC: Rf 0.25 (Hex/EtOAc¼2:1, stained blue with phosphomo-
lybdic acid/EtOH). 1H NMR (300 MHz, CDCl3):
d 7.34e7.31 (m, 5H),
4.66 (m, 1H), 3.56 (d, 1H, J¼10.5 Hz), 3.48 (d, 1H, J¼10.5 Hz), 2.70 (s,
2H), 0.88 (s, 3H), 0.84 (s, 3H). The enantiomeric excess was de-
termined by chiral HPLC with Daicel Chiralpak AD column [eluent
98:2 Hex/IPA; flow rate: 0.8 mL/min; detection: 220 nm; tR:
54.6 min (S-major), tR: 60.0 min (R-minor)].
References and notes
4.3.2. 2-Cyclohexyl-1-(4-phenyl)-1,3-propanediol (29). This product
was purified by flash column chromatography on silica gel with
a 7:3 hexane/ethyl acetate mixture as eluent. Its 1H NMR data were
in agreement with those reported in the literature.14
1. Review: Denmark, S. E.; Beutner, G. L. Angew. Chem., Int. Ed. 2008, 47, 1560.
2. Malkov, A. V.; Kocovsky, P. Eur. J. Org. Chem. 2007, 29.
ꢁ
ꢂ
3. For the few examples of organocatalytic reactions with aliphatic amines-
N-oxides see: (a) Traverse, J. F.; Zhao, Y.; Hoveyda, A. H.; Snapper, M. L. Org. Lett.
2005, 7, 3151; (b) Qin, B.; Liu, X.; Shi, J.; Zheng, K.; Zhao, H.; Feng, X. J. Org. Chem.
2007, 72, 2374 and references cited; (c) Simonini, V.; Benaglia, M.; Guizzetti, S.;
Pignataro, L.; Celentano, G. Synlett 2008, 1061.
TLC: Rf 0.22 (Hex/EtOAc¼7:3, stained blue with phosphomo-
lybdic acid). 1H NMR (300 MHz, CDCl3):
d 7.35e7.25 (m, 5H), 4.73 (s,
4. For a report where a stoichiometric amount of chiral phosphine oxide was
employed see: Ogawa, C.; Sugiura, M.; Kobayashi, S. Angew. Chem., Int. Ed. 2004,
43, 6491.
5. Nakajima, M.; Kotani, S.; Ishizuka, T.; Hashimoto, S. Tetrahedron Lett. 2005,
46, 157.
1H), 3.80e3.60 (dd, 2H), 2.80 (br, 1H), 1.95e1.80 (m, 1H), 1.65e1.50
(m, 5H), 1.40e1.00 (m, 4H). The enantiomeric excess was de-
termined by chiral HPLC with Daicel Chiralcell AD column [eluent
98:2 Hex/IPA; flow rate: 0.8 mL/min; detection: 210 nm; tR:
61.6 min (minor), tR: 80.2 min (major)].
6. See Ref. 1; for recent works in the field see: Denmark, S. E.; Chung, W.-J. Angew.
Chem., Int. Ed. 2008, 47, 1890 and references cited.
7. (a) Sugiura, M.; Sato, N.; Kotani, S.; Nakajima, M. Chem. Commun. 2008, 4309;
(b) Sugiura, M.; Kumahara, M.; Nakajima, M. Chem. Commun. 2009, 3585.
8. Kotani, S.; Hashimoto, S.; Nakajima, M. Tetrahedron 2007, 63, 3122.
9. Shimoda, Y.; Tando, T.; Kotani, S.; Sugiura, S.; Nakajima, M. Tetrahedron:
Asymmetry 2009, 20, 1369.
10. Simonini, V.; Benaglia, M.; Benincori, T. Adv. Synth. Catal. 2008, 350, 561.
11. For mechanistic aspects in aldol reactions see: see Ref. 1 and Denmark, S. E.;
Wynn, T.; Beutner, G. L. J. Am. Chem. Soc. 2002, 124, 13405; Denmark, S. E.;
Pham, S. M.; Stavenger, R. M.; Su, X.; Wong, K.-T.; Nishigaichi, T. J. Org. Chem.
2006, 71, 3904.
12. Modern Aldol Reactions; Mahrwald, R., Ed.; Wiley-VCH: Weinheim, 2004; Trost,
B. M.; Brindle, C. S. Chem. Soc. Rev. 2010, 39, 1600; Guillena, G.; Najera, C.;
Ramon, D. Tetrahedron: Asymmetry 2007, 18, 2249.
13. Kotani, S.; Shimoda, Y.; Sugiura, M.; Nakajima, M. Tetrahedron Lett. 2009, 50,
4602.
4.3.3. 2,2-Dimethyl-1-(4-chlorophenyl)-1,3-propanediol (30). This
product was purified by flash column chromatography on silica gel
with a 7:3 hexane/ethyl acetate mixture as eluent. Its 1H NMR data
were in agreement with those reported in the literature.13
TLC: Rf 0.25 (Hex/EtOAc¼7:3 stained blue with phosphomolyb-
dic acid/EtOH). 1H NMR (300 MHz, CDCl3):
d 7.40e7.20 (m, 4H), 4.60
(s, 1H), 3.45 (q, 2H), 0.85 (s, 3H), 0.80 (s, 3H). The enantiomeric
excess was determined by chiral HPLC with Daicel Chiralcell AD
column [eluent 98:2 Hex/IPA; flow rate: 0.8 mL/min; detection:
210 nm; tR: 46.2 min (minor), tR: 48.2 min (major)].
14. The reaction with aliphatic aldehydes in the present conditions did not afford
the expected aldol product; it is known that for those substrates the formation
of chlorohydrin may occur (see discussion in Ref. 1).
15. Benaglia, M.; Guizzetti, S.; Pignataro, L. Coord. Chem. Rev. 2008, 252, 492 and
Ref. 1.
16. The observed syn stereoselectivity for the reaction in toluene might be ex-
plained with the preference for a boat-like transition state instead of a chairlike
transition state (see Fig. 1 for a proposed model of stereoselection).
4.3.4. 2,2-Dimethyl-1-(4-chlorophenyl)-1,3-propanediol (30). This
product was purified by flash column chromatography on silica gel
with a 7:3 hexane/ethyl acetate mixture as eluent. Its 1H NMR data
were in agreement with those reported in the literature.23
TLC: Rf 0.25 (Hex/EtOAc¼7:3 stained blue with phosphomolyb-
dic acid/EtOH). 1H NMR (300 MHz, CDCl3):
d 7.40e7.20 (m, 4H), 4.60

166
S. Rossi et al. / Tetrahedron 67 (2011) 158e166
17. For recent 29Si and 31P NMR studies in the field see: Denmark, S. E.; Eklov, B. M.
Chem.dEur. J. 2008, 14, 234.
18. For similar observations see Without Lewis base the enolization process did not
2420; (c) Hayashi, Y.; Aratake, S.; Okano, T.; Takahashi, J.; Sumiya, T.; Shoji, M.
Angew. Chem., Int. Ed. 2006, 45, 5527; (d) Kano, T.; Yamaguchi, Y.; Tanaka, Y.;
Maruoka, K. Angew. Chem., Int. Ed. 2007, 46, 1738.
occurr. Ref. 13, 8.
20. For products characterization see Ref. 13 and references cited there.
21. Tzeng, Z.-H.; Chen, H.-Y.; Reddy, R. J.; Huang, C.-T.; Chen, K. Tetrahedron 2009,65, 2879.
22. Kawano, Y.; Fujisawa, H.; Mukaiyama, T. Chem. Lett. 2005, 34, 614.
23. Hoeve, V.; Wynberg, H. J. Org. Chem. 1985, 50, 4508.
19. Selected recent examples of organocatalyzed cross-aldol reactions between
two aldehydes: (a) Northrup, A. B.; MacMillan, D. W. C. J. Am. Chem. Soc. 2002,
124, 6798; (b) Mase, N.; Tanaka, F.; Barbas, C. F. Angew. Chem., Int. Ed. 2004, 43,