Highly Stereoselective Synthesis of syn-1,3-Diols through a Sequential Titanium-Mediated Aldol Reaction and LiBH4 Reduction

Article abstract of DOI:10.1002/ejoc.201000293

A sequential transformation based on a titanium-mediated aldol reaction from chiral α-tert-butyldimethylsilyloxy ketones followed by reduction of the resultant aldolates with LiBH₄ provides a straightforward, access to syn-1,3-diols. These diols, containing up to three new stereocentres, can be easily isolated in high yields after a simple work-up without any additional oxidative treatment, which confers to this procedure an appealing position to prepare stereoselectively such sort of structures.

Full text of DOI:10.1002/ejoc.201000293

FULL PAPER
DOI: 10.1002/ejoc.201000293
Highly Stereoselective Synthesis of syn-1,3-Diols through a Sequential
Titanium-Mediated Aldol Reaction and LiBH4 Reduction
Judit Esteve,[a] Sònia Matas,[a] Miquel Pellicena,[a] Javier Velasco,[a] Pedro Romea,*[a]
Fèlix Urpí,*[a] and Mercè Font-Bardia[b]
Dedicated to Professor Pelayo Camps on the occasion of his 65th birthday
Keywords: Stereoselective synthesis / Sequential transformation / Aldol reactions / Reduction / Chirality / Ketones
A sequential transformation based on a titanium-mediated
aldol reaction from chiral α-tert-butyldimethylsilyloxy
ketones followed by reduction of the resultant aldolates with
LiBH4 provides a straightforward access to syn-1,3-diols.
These diols, containing up to three new stereocentres, can
be easily isolated in high yields after a simple work-up with-
out any additional oxidative treatment, which confers to this
procedure an appealing position to prepare stereoselectively
such sort of structures.
the use of boron reagents requires a final oxidative treat-
ment that occasionally turns out to be problematic.[12]
Thus, other metals have been surveyed to act as boron sur-
rogates.[13,14] Titanium(IV) has been one of them. DiMare
first uncovered the diastereoselective reduction of TiCl4-
chelates from chiral β-alkoxy[15a] and β-hydroxy[15b] ketones
with several borane complexes. Later, Bartoli studied in de-
tail this transformation with BH3·py[16a,16b] and the ad-
dition of other boron reducing agents to β-hydroxy ketone
titanium alcoholates prepared by transmetallation of the
corresponding lithium alcoholates with TiCl4.[16c] Remark-
ably, this procedure did not afford the desired 1,3-diols but
the related cyclic boronates, which had to be treated with
H2O2 in a basic medium to obtain the syn-1,3-diols. Thus,
considering these precedents and taking advantage of our
own experience on titanium-mediated aldol reactions from
chiral α-silyloxy ketones,[17] we envisaged that the resultant
titanium aldolates might be reduced in situ to provide ster-
eoselectively syn-1,3-diols in a straightforward manner.[18,19]
Herein, we document our studies on this sequential tita-
nium-mediated aldol and reduction reactions from chiral α-
tert-butyldimethylsilyloxy ketones 1–3 represented in
Scheme 1.
Introduction
The increased demand for more efficient syntheses has
stimulated the development of new concepts and more
stringent strategies during the last years. Thus, synthetic de-
sign in our days has to address not only good yields and
high levels of chemo-, regio-, and stereoselectivity, but also
different economy requirements.[1] In this context, it was
early recognized that sequential transformations[2,3] could
meet such challenges, since the benefits of carrying out
several transformations in a single operation are consider-
able.[4] The asymmetric synthesis of 1,3-diols is a nice exam-
ple of these endeavours.[5,6] Indeed, Narasaka and Prasad
reported that the intermolecular hydride delivery to a bo-
ron-chelated intermediate from a β-hydroxy ketone fur-
nishes syn-1,3-diols with a remarkable stereocontrol.[7] Inas-
much as this approach relies on the formation of an struc-
turally rigid boron intermediate, Paterson exploited some
highly diastereoselective boron-mediated aldol reactions
from chiral ketones to obtain a boron-chelated aldolate,
which could then participate in the subsequent stereoselec-
tive reduction that finally renders syn-1,3-diols.[8–10] Re-
gardless of the success of this sequential transformation,[11]
[a] Departament de Química Orgànica, Universitat de Barcelona,
Carrer Martí i Franqués 1–11, 08028 Barcelona, Catalonia,
Spain
Fax: +34-93 3397878
E-mail: pedro.romea@ub.edu
felix.urpi@ub.edu
[b] Departament de Cristal.lografia, Mineralogia i Dipòsits
Minerals, Universitat de Barcelona,
Carrer Martí i Franqués s/n, 08028 Barcelona, Catalonia, Spain
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000293.
Scheme 1. Aldol-reduction transformation from chiral α-tert-butyl-
dimethylsilyloxy ketones.
3146
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 3146–3151

Highly Stereoselective Synthesis of syn-1,3-Diols
experimental conditions to other aliphatic, aromatic, and
α,β-unsaturated aldehydes. The results summarized in
Table 2 prove that the TiCl4-mediated aldol reactions of chi-
ral α-tert-butyldimethylsilyloxy ketone 1 followed by the re-
duction of the resultant aldolates with LiBH4 reliably afford
a single diastereomer of the desired syn-1,3-diols 4 in high
yields (see entries 1–5 in Table 2). Interestingly, these condi-
tions could be also used for the (iPrO)TiCl3-mediated aldol
reaction from 1,[17c] although the yields were in this oc-
casion slightly lower than for TiCl4 (compare entries 1–5
and 6–10 in Table 2).
Results and Discussion
Our concern on the feasibility of the abovementioned se-
quence led us to assess initially the reduction of the tita-
nium aldolate from the TiCl4-mediated aldol reaction of
lactate-derived ketone 1[20] with isobutyraldehyde (a). Thus,
different reducing agents were added to the reaction mix-
ture containing the titanium aldolate prepared according
to the previously optimized experimental conditions, which
involve the addition of the titanium enolate from 1 to an
aldehyde (1.5 equiv.) at –78 °C for 30 min.[17c] Preliminary
experiments with three and four equivalents of LiBH4 af-
forded complex reaction mixtures, but we were glad to ob-
serve that their simple purification through column
chromatography provided the desired syn-1,3-diol 4a as a
single diastereomer in high yield (see entries 1 and 2 in
Table 1). Further optimization of the reductive step showed
that two equivalents of LiBH4 were enough to achieve the
same result after one hour at –78 °C (compare entries 1–
4 in Table 1). These conditions were next applied to other
reducing agents as NaBH4 and -selectride. The former
turned out to be less active than LiBH4, probably because
of its poor solubility in CH2Cl2, and afforded diol 4a in
24% yield together with non-reduced 2,4-syn-4,5-syn aldol
adduct in 55% yield (see entry 5 in Table 1). However, -
selectride proceeded in a different way and afforded the cor-
responding cyclic boronate in 53% yield and the 2,4-syn-
4,5-syn aldol adduct in 32% yield (see entry 6 in Table 1).
Therefore, these preliminary experiments revealed that both
NaBH4 and -selectride were unsuitable reducing agents,
while it was possible to obtain a diastereomerically pure
syn-1,3-diol by simple addition of LiBH4 to the titanium
aldolate from ketone 1 without any further oxidative step.
Table 2. Aldol-reduction transformation from ketone 1.
Entry TiL4
Aldehyde
R
Yield 4 [%][a,b]
1
2
3
TiCl4
TiCl4
TiCl4
TiCl4
a
b
c
d
e
a
b
c
d
e
iPr
iBu
Pr
Ph
81
87
71
87
80
68
67
60
70
70
4[c]
5[c]
6
TiCl4
(E)-CH3CH=CH
iPr
iBu
Pr
(iPrO)TiCl3
(iPrO)TiCl3
(iPrO)TiCl3
(iPrO)TiCl3
(iPrO)TiCl3
7
8
9[c]
10[c]
Ph
(E)-CH3CH=CH
[a] A single diastereomer (dr Ͼ 97:3) was obtained after column
chromatography. [b] Isolated yields of syn diols 4. [c] 1.2 equiv. of
the aldehyde was used.
The successful development of this methodology on ethyl
ketone 1 encouraged us to tackle the more sensitive (S) 1-
bromo-3-(tert-butyldimethylsilyloxy)-2-pentanone (2, see
Scheme 1).[21] Indeed, we have recently disclosed that the
aldol reaction of this lactate-derived bromo ketone requires
a milder titanium Lewis acid as (iPrO)2TiCl2 in order to
achieve high diastereoselectivities, and a careful handling of
the resultant bromo aldol adducts to avoid undesired de-
compositions.[17d] Hence, bromo ketone 2 was a suitable
platform to challenge such sequential transformation and,
additionally, it offered the opportunity of setting up four
contiguous and functionalized stereocentres in a single syn-
thetic operation. Keeping in mind these ideas, we carried
out stereoselective (iPrO)2TiCl2-mediated aldol additions of
2 to a representative set of aldehydes and the resultant ald-
olates were reduced with two equivalents of LiBH4 at
–78 °C for 1 h. As shown in Table 3, we were pleased to
confirm that the corresponding syn-1,3-diols 5 could be
also isolated as a single diastereomer in good yields.[22]
Eventually, we centered our attention on the α-tert-butyl-
dimethylsilyloxy methyl ketone 3.[17e] Since the stereoselec-
Table 1. Influence of the reducing agent on the aldol-reduction
transformation from ketone 1 and isobutyraldehyde.
Entry Reducing agent
Equiv.
tred [h]
Yield 4a [%][a,b]
1
2
3
4
5
6
LiBH4
LiBH4
LiBH4
LiBH4
NaBH4
-selectride
4
3
2
2
2
2
0.5
0.5
0.5
1
1
1
83
83
65
81
24[c]
(53)[d]
[a] A single diastereomer (dr Ͼ 97:3) was obtained after column
chromatography. [b] Isolated yields of syn diols (in parentheses iso-
lated yields of cyclic boronates). [c] 2,4-syn-4,5-syn Aldol was also
isolated in 55% yield. [d] 2,4-syn-4,5-syn Aldol was also isolated in
32% yield.
Having established the ability of LiBH4 to reduce the ti- tive acetate aldol reaction still represents a challenge,[23] we
tanium aldolate from isobutyraldehyde to 1,3-diol 4a with considered that the application of the combined aldol and
an outstanding stereocontrol, we next applied the optimized reduction reactions on that chiral methyl ketone would defi-
Eur. J. Org. Chem. 2010, 3146–3151
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3147

P. Romea, F. Urpí et al.
Table 4. Aldol-reduction transformation from ketone 3.
FULL PAPER
Table 3. Aldol-reduction transformation from ketone 2.
En-
try
Aldehyde
R
dr[a]
Yield 6 [%][b]
Entry
Aldehyde
R
Yield 5 [%][a,b]
1
a
c
d
f
iPr
Pr
Ph
94:6
92
91
92
91
1
a
c
d
e
iPr
Pr
Ph
65
74
52
60
2
93:7
2
3[c]
80:20[d]
3[c]
4[c]
4[c]
CH2=C(CH3) 88:12[d]
(E)-CH3CH=CH
[a] Established by 1H NMR of the reaction mixtures. [b] Overall
isolated yields. [c] 1.2 equiv. of the aldehyde was used. [d] Tiny
amounts (Ͻ5%) of a third diastereomer were also observed.
[a] A single diastereomer (dr Ͼ 97:3) was obtained after column
chromatography. [b] Isolated yields of syn diols 5. [c] 1.2 equiv. of
the aldehyde was used.
nitely prove the synthetic efficiency of the abovementioned
transformation. Therefore, we initially examined the reac-
tion of 3 with isobutyraldehyde under the optimized experi-
mental conditions.[24] Preliminary experiments were bother-
some as an equimolar mixture of two diols were always ob-
tained. However, we recognized after a careful analysis of
the resulting mixtures that they contained the desired syn-
1,3-diol and an isomer produced by the migration of the
TBS protecting group, which presumably occurs as a conse-
quence of the steric hindrance of TBS-protected alcohol.[25]
Hence, subsequent treatment of the reaction mixtures with
HF afforded triol 6a in 92% yield and dr = 94:6. Other
aldehydes as butanal, benzaldehyde and methacrolein also
furnished excellent yields but rather different diastereoselec-
tivities (see Table 4). Indeed, the sequential transformation
on an aromatic aldehyde as benzaldehyde furnishes an out-
standing 92% yield but a modest diastereoselectivty (dr =
80:20, see entry 3 in Table 4), similar to that observed in
the corresponding aldol reaction (dr = 83:17).[17e] Other-
wise, the stereochemical control becomes excellent on ali-
Scheme 2. Synthesis of derivatives from diols 4–6.
Importantly, the absolute configuration was secured
phatic aldehydes as isobutyraldehyde and butanal (dr Ͼ through X-ray analysis of diol 4d shown in Figure 1.[27]
93:7), close again to the diastereoselectivity obtained in the
aldol step (dr = 96:4).[17e] Therefore, these results prove that
the reported methodology can be applied to a broad scope
of chiral α-tert-butyldimethylsilyloxy ketones to create a
carbon–carbon bond and up to three new stereocentres in
such a way that the stereochemical control mostly relies on
the aldol step of the sequence.
The configuration of C4 and C5 stereocentres of diols
4 and 5 had been firmly established in our studies on the
stereoselective titanium-mediated aldol reactions from chi-
ral α-silyloxy ketones 1 and 2.[17a–17d] The all syn configura-
tion of C3–C5 stereocentres was next secured by compari-
son of NMR spectroscopic data of diols 4 and 5 with those
obtained from the corresponding aldol adducts using the
Narasaka–Prasad[7] methodology (see equation 1 in
Scheme 2). Furthermore, some of these diols were protected
Figure 1. X-ray of diol 4d.
as isopropylidene acetals (see equation 2 in Scheme 2) and
analyzed by 1D and 2D NMR techniques.[26] Parallel stud-
Although we have not carried out any detailed analysis
ies on triol 6a also supported the syn relative configuration on the mechanism of the aforementioned sequence, its
(see equation 3 in Scheme 2).[17e,26]
stereochemical outcome can be understood by means of
3148
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 3146–3151

Highly Stereoselective Synthesis of syn-1,3-Diols
tion mixture was stirred for 1 h at –78 °C. Finally, the reaction was
quenched by a slow addition of AcOH (1 mL) followed by satd.
NH4Cl (5 mL). The mixture was partitioned with Et2O and H2O,
and the organic layer was washed with satd. NaHCO3 and brine.
The combined organic extracts were dried (MgSO4) and concen-
trated. The resulting oil was purified by column chromatography
(hexanes/EtOAc, 9:1) to afford 235 mg (81% yield) of
(2S,3S,4R,5S)-2-(tert-butyldimethylsilyloxy)-4,6-dimethyl-3,5-hep-
tanediol (4a) as a single diastereomer. Colorless oil. Rf = 0.15 (hex-
well established models for the reduction of six-membered
chelated ketones (see Scheme 3). Indeed, the titanium-medi-
ated aldol reactions from ketones 1–3 afford syn-chelated
aldolates (I in Scheme 3),[28] which must adopt a conforma-
tion with the most bulky groups placed at pseudoequatorial
positions (see II in Scheme 3). Then, the nucleophilic atack
of LiBH4 proceeds through the face of the C=O bond that
provides chair-like transition states to yield the syn-1,3-diols
4–6. Interestingly and in accordance to the examples re-
anes/EtOAc, 9:1). [α]2D0 = +9.2 (c 1.2, CHCl ). IR (film): ν = 3473,
˜
3
ported by DiMare,[15] we have not observed the formation 2957, 1460, 1394, 1256, 1086 cm–1. H NMR (500 MHz, CDCl3):
1
δ = 3.76 (dq, J = 7.7, J = 6.0 Hz, 1 H, CHOTBS), 3.45 (dd, J =
7.7, J = 2.3 Hz, 1 H, TBSOCHCHOH), 3.34 [dd, J = 8.7, J =
2.0 Hz, 1 H, CHOHCH(CH3)2], 1.80–1.68 [m, 2 H, CHCH3 and
CH(CH3)2], 1.12 (d, J = 6.0 Hz, 3 H, CH3CHOTBS), 1.02 (d, J =
6.4 Hz, 3 H, CHCH3), 0.91 [s, 9 H, SiC(CH3)3], 0.88 (d, J = 7.0 Hz,
3 H, CH3CHCH3), 0.83 (d, J = 7.0 Hz, 3 H, CH3CHCH3), 0.11 (s,
3 H, SiCH3), 0.10 (s, 3 H, SiCH3) ppm. 13C NMR (75.4 MHz,
CDCl3): δ = 82.1, 81.5, 70.2, 34.7, 30.8, 25.8, 19.8, 19.5, 19.1, 18.0,
5.2, –4.0, –4.8 ppm. HRMS (FAB): calcd. for C15H35O3Si [M +
H]+ 291.2355; found 291.2353.
of stable boronates, which constitutes a remarkable feature
of the present methodology since no further steps are re-
quired to obtain diastereomerically pure diols.
Typical Procedure for the (iPrO)TiCl3-Mediated Transformation
from Ketone 1: Freshly distilled Ti(OiPr)4 (83 µL, 0.28 mmol) was
added to a solution of TiCl4 (92 µL, 0.84 mmol) in CH2Cl2 (1 mL)
at 0 °C under N2. The mixture was stirred for 10 min at 0 °C and
10 min at room temperature, diluted with CH2Cl2 (1 mL), and the
resulting solution was added via cannula (2ϫ 0.5 mL) to a solution
of 1 (216 mg, 1 mmol) in CH2Cl2 (2 mL) at –78 °C. The pale yellow
solution was stirred for 3–4 min and iPr2NEt (0.19 mL, 1.1 mmol)
was added dropwise. From this point, the TiCl4-based experimental
protocol was followed, which afforded 4a in 68% yield as a single
diastereomer.
Scheme 3. Mechanistic hypothesis for the overall transformation.
Typical Procedure for the (iPrO)2TiCl2-Mediated Transformation
from Ketone 2: Freshly distilled Ti(OiPr)4 (163 µL, 0.55 mmol) was
added to a solution of TiCl4 (60 µL, 0.55 mmol) in CH2Cl2 (1 mL)
at 0 °C under N2. The mixture was stirred for 10 min at 0 °C and
10 min at room temperature, diluted with CH2Cl2 (1 mL), and the
resulting solution was added via cannula (2ϫ 0.5 mL) to a solution
of 2 (281 mg, 1 mmol) in CH2Cl2 (2 mL) at –78 °C, which devel-
oped a yellow-orange color. It was stirred for 3–4 min and iPr2NEt
(0.19 mL, 1.1 mmol) was added dropwise. The resultant mixture
was stirred for 20 min at –78 °C and 1 h at –20 °C. Eventually, the
color of the reaction mixture became dark red. Then, it was cooled
to –78 °C and freshly distilled isobutyraldehyde (0.14 mL,
1.5 mmol) was added dropwise, and the stirring was continued for
2 h at –78 °C and 1 h at –20 °C. The reaction mixture was cooled
to –78 °C, 2  LiBH4 in THF (1 mL, 2 mmol) was added carefully
and stirring was continued for 1 h at –78 °C. Finally, the reaction
was quenched by slow addition of AcOH (1 mL) followed by satd.
NH4Cl (5 mL). The mixture was partitioned with Et2O and H2O,
and the organic layer was washed with satd. NaHCO3 and brine.
The combined organic extracts were dried (MgSO4) and concen-
trated. The resulting oil was purified by column chromatography
(hexanes/EtOAc, 9:1) to afford 232 mg (65% yield) of
(2S,3R,4R,5S)-4-bromo-2-(tert-butyldimethylsilyloxy)-6-methyl-
3,5-heptanediol (5a) as a single diastereomer as a colorless oil. Rf
= 0.15 (hexanes/EtOAc, 9:1). [α]2D0 = +3.8 (c = 1.05, CHCl3). IR
Conclusions
The titanium-mediated aldol reactions from α-tert-butyl-
dimethylsilyloxy ketones with aliphatic, aromatic, and α,β-
unsaturated aldehydes, followed by the reduction of the re-
sultant aldolates with LiBH4, proceed with an excellent ste-
reochemical control to afford the corresponding syn-1,3-di-
ols. This sequential transformation is compatible with a
broad scope of titanium(IV) Lewis acids, from the stronger
TiCl4 to the milder (iPrO)2TiCl2, and can be applied to sen-
sitive substrates as bromo ketone 2 or methyl ketone 3.
Noteworthy, no additional oxidative steps are required and
the desired syn-1,3-diols can be easily isolated by simple
column chromatography in high yields, which can be useful
to design more efficient synthesis of natural products.
Experimental Section
Typical Procedure for the TiCl4-Mediated Transformation from
Ketone 1: Neat TiCl4 (0.12 mL, 1.1 mmol) was added slowly to a
solution of ketone 1 (216 mg, 1 mmol) in CH2Cl2 (5 mL) at –78 °C
under N2. The resulting yellow suspension was stirred for 3–4 min
and iPr2NEt (0.19 mL, 1.1 mmol) was added dropwise. The re-
sulting dark red solution was stirred for 30 min at –78 °C, freshly
(film): ν = 3516, 2957, 1472, 1389, 1256, 1136, 1086 cm–1. 1H NMR
˜
(400 MHz, CDCl3): δ = 4.16 (t, J = 2.9 Hz, 1 H, CHBr), 4.01
(quint, J = 6.2 Hz, 1 H, CHOTBS), 3.54–3.49 (m, 1 H, TBSOCH-
distilled isobutyraldehyde (0.14 mL, 1.5 mmol) was added drop- CHOH), 3.40–3.34 [m, 1 H, CHOHCH(CH3)2], 1.98–1.88 [m, 1 H,
wise, and stirring was continued for 30 min at –78 °C. Then, 2 
LiBH4 in THF (1 mL, 2 mmol) was added carefully and the reac-
CH(CH3)2], 1.18 (d, J = 6.1 Hz, 3 H, CH3CHOTBS), 1.04 (d, J =
6.6 Hz, 3 H, CH3CHCH3), 0.92 (d, J = 6.6 Hz, 3 H, CH3CHCH3),
Eur. J. Org. Chem. 2010, 3146–3151
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3149

P. Romea, F. Urpí et al.
FULL PAPER
recently, Baker and Bazan have stated that “Sequential reac-
tions involve coupling of transformations that may operate in-
dependently and often require additional reagents or changes
in reaction conditions. In a practical sense, they allow for reac-
tions to be carried out in a single reaction vessel without purifi-
cation between steps” (see ref.[2c]).
0.91 [s, 9 H, SiC(CH3)3], 0.13 (s, 3 H, SiCH3), 0.12 (s, 3 H, SiCH3)
ppm. 13C NMR (100.6 MHz, CDCl3): δ = 79.0, 78.6, 71.1, 60.3,
32.1, 25.8, 19.5, 19.0, 18.1, 18.0, –4.1, –4.7 ppm. HRMS (FAB):
calcd. for C14H3279BrO3Si [M + H]+ 355.1299; found 355.1289;
calcd. for C14H3281BrO3Si [M + H]+ 357.1280; found 357.1271.
[4]
[5]
[6]
Typical Procedure for the TiCl4-Mediated Transformation from
Ketone 3: Neat TiCl4 (0.12 mL, 1.1 mmol) was added slowly to a
solution of ketone 3 (230 mg, 1 mmol) in CH2Cl2 (5 mL) at –94 °C
under N2. The resulting yellow suspension was stirred for 3–4 min
and iPr2NEt (0.19 mL, 1.1 mmol) was added dropwise. The re-
sulting dark red solution was stirred for 30 min at –94 °C, freshly
distilled isobutyraldehyde (0.14 mL, 1.5 mmol) was added drop-
wise, and stirring was continued for 30 min at –78 °C. Then, 2 
LiBH4 in THF (1 mL, 2 mmol) was added carefully and the reac-
tion mixture was stirred for 1 h at –78 °C. Finally, the reaction was
quenched by slow addition of AcOH (1 mL) followed by satd.
NH4Cl (5 mL). The mixture was partitioned with Et2O and H2O,
and the organic layer was washed with satd. NaHCO3 and brine,
dried (MgSO4) and concentrated. A solution of the residue in
CH3CN (10 mL) was treated with 48% HF (0.33 mL) for 20 min
at room temperature. Then, it was partitioned in CH2Cl2 and satd.
NaHCO3, and the organic layer was washed with satd. NaHCO3.
The aqueous layers were extracted with EtOAc and the combined
organic extracts were dried (MgSO4) and concentrated. The re-
sulting oil was purified by column chromatography (hexanes/
EtOAc, 1:1) to afford 174 mg of (3S,4S,6R)-2,7-dimethyl-3,4,6-oc-
K. C. Nicolaou, D. J. Edmonds, P. G. Bulger, Angew. Chem.
Int. Ed. 2006, 45, 7134–7186.
For a review on the stereoselective synthesis of 1,3-diols, see:
S. E. Bode, M. Wolberg, M. Müller, Synthesis 2006, 557–588.
For a recent example on a sequential aldol transformation
leading to 1,3-diols, see: M. B. Boxer, M. Akakura, H. Yamam-
oto, J. Am. Chem. Soc. 2008, 130, 1580–1582.
a) K. Narasaka, F.-C. Pai, Tetrahedron 1984, 40, 2233–2238;
b) K.-M. Chen, G. E. Hardtmann, K. Prasad, O. Repic, M. J.
Shapiro, Tetrahedron Lett. 1987, 28, 155–158.
a) I. Paterson, M. V. Perkins, Tetrahedron Lett. 1992, 33, 801–
804; b) I. Paterson, M. V. Perkins, Tetrahedron 1996, 52, 1811–
1834.
For parallel processes involving chiral boron Lewis acids, see:
a) C. Bonini, R. Racioppi, G. Righi, L. Rossi, Tetrahedron:
Asymmetry 1994, 5, 173–176; b) Y. Kaneko, T. Matsuo, S.-i.
Kiyooka, Tetrahedron Lett. 1997, 38, 4107–4110.
Otherwise, anti-1,3-diols can be prepared through the Tishch-
enko aldol procedure, a sequential transformation based on an
aldol reaction followed by the intramolecular hydride delivery.
For a review and a recent example, see: a) J. Mlynarski, Eur. J.
Org. Chem. 2006, 4779–4786; b) K. Rohr, R. Herre, R.
Mahrwald, J. Org. Chem. 2009, 74, 3744–3749.
For examples of this sequential transformation in total synthe-
sis, see: a) I. Paterson, M. D. McLeod, Tetrahedron Lett. 1997,
38, 4183–4186; b) K. C. Nicolaou, F. Murphy, S. Barluenga, T.
Ohshima, H. Wei, J. Xu, D. L. F. Gray, O. Baudoin, J. Am.
Chem. Soc. 2000, 122, 3830–3838; c) M. Carda, S. Rodriguez,
B. Segovia, J. A. Marco, J. Org. Chem. 2002, 67, 6560–6563; d)
S. A. Burova, F. E. McDonald, J. Am. Chem. Soc. 2004, 126,
2495–2500; e) I. Paterson, I. Lyothier, J. Org. Chem. 2005, 70,
5494–5507; f) I. Paterson, K. Ashton, R. Britton, G. Cecere,
G. Chouraqui, G. J. Florence, H. Knust, J. Stafford, Chem.
Asian J. 2008, 3, 367–387.
[7]
[8]
[9]
[10]
[11]
1
tanetriol (6a) as a diastereomeric mixture (dr = 94:6 by H NMR
analysis) in 92% overall yield.[29] White solid; m.p. 71–72 °C. Rf =
0.30 (hexanes/EtOAc, 1:1). [α]2D0 = +11.9 (c = 1.1, CHCl3). IR
(film): ν = 3371, 2957, 2929, 1468, 1448, 1430, 1387 cm–1. 1H NMR
˜
(400 MHz, CDCl3): δ = 3.90 (ddd, J = 9.6, J = 3.8, J = 3.2 Hz, 1
H, HOCHCHOHCH2), 3.72 [ddd, J = 10.0, J = 5.0, J = 2.3 Hz,
1 H, CH2CHOHCH(CH3)2], 3.09 (dd, J = 6.3, J = 3.8 Hz, 1 H,
HOCHCHOHCH2), 1.85–1.55 [m, 4 H, CH2 and 2ϫ CH(CH3)2],
0.97 (d, J = 6.8 Hz, 3 H, CHCH3), 0.97 (d, J = 6.7 Hz, 3 H,
CHCH3), 0.94 (d, J = 6.9 Hz, 3 H, CHCH3), 0.93 (d, J = 6.8 Hz,
3 H, CHCH3) ppm. 13C NMR (100.6 MHz, CDCl3): δ = 79.6, 77.2,
72.6, 36.5, 34.2, 30.4, 19.6, 18.3, 17.6, 17.4 ppm. HRMS (ESI):
calcd. for C10H23O3 [M + H]+ 191.1642; found 191.1643.
[12]
Boron-mediated aldol reaction/LiBH4 reduction sequential
transformations on chiral TBS-protected ketone
1 (see
Scheme 1) delivered the corresponding cyclic boronates as a
single diastereomer in high yields. Unfortunately, the treatment
of these boronates with H2O2/NaOH triggered the partial mi-
gration of the TBS protecting group. Although this drawback
was sorted out by using milder oxidative conditions, the purifi-
cation of the resultant 1,3-diols was hampered by the presence
of non-identified impurities. See, M. Galobardes, M. Mena, P.
Romea, F. Urpí, J. Vilarrasa, Tetrahedron Lett. 2002, 43, 6145–
6148.
Supporting Information (see also the footnote on the first page of
this article): Characterization of 4–6 and proof of the stereochemis-
try.
Acknowledgments
Financial support from the Spanish Ministerio de Ciencia e Innov-
ación (MICINN) and Fondos FEDER (Grant CTQ2009-09692),
and the Generalitat de Catalunya (2009SGR825), as well as a doc-
torate studentship (FPU, Ministerio de Educación) to M. P., are
acknowledged.
[13]
[14]
[15]
For an Al-based procedure, see: S.-i. Kiyooka, H. Kuroda, Y.
Shimasaki, Tetrahedron Lett. 1986, 27, 3009–3012.
For a Zn-based procedure, see: T. Nakata, Y. Tani, M. Hato-
zaki, T. Oishi, Chem. Pharm. Bull. 1984, 32, 1411–1415.
a) C. R. Sarko, I. C. Guch, M. DiMare, J. Org. Chem. 1994,
59, 705–706; b) C. R. Sarko, S. E. Collibee, A. L. Knorr, M.
DiMare, J. Org. Chem. 1996, 61, 868–873.
a) G. Bartoli, M. Bosco, E. Marcantoni, M. Massaccesi, S.
Rinaldi, L. Sambri, Eur. J. Org. Chem. 2001, 4679–4684; b) G.
Bartoli, M. Bosco, M. C. Bellucci, R. Dalpozzo, E. Marcan-
toni, L. Sambri, Org. Lett. 2000, 2, 45–47; c) G. Bartoli, M. C.
Bellucci, M. Bosco, R. Dalpozzo, E. Marcantoni, L. Sambri,
Chem. Eur. J. 2000, 6, 2590–2598.
a) S. Figueras, R. Martín, P. Romea, F. Urpí, J. Vilarrasa, Tet-
rahedron Lett. 1997, 38, 1637–1640; b) I. Larrosa, P. Romea,
F. Urpí, Org. Lett. 2006, 8, 527–530; c) J. Nebot, S. Figueras,
P. Romea, F. Urpí, Y. Ji, Tetrahedron 2006, 62, 11090–11099;
d) J. Nebot, P. Romea, F. Urpí, J. Org. Chem. 2009, 74, 7518–
[1] a) B. M. Trost, Science 1991, 254, 1471–1477; b) B. M. Trost,
Angew. Chem. Int. Ed. Engl. 1995, 34, 259–281; c) P. A. Wender,
M. P. Croatt, B. Witulski, Tetrahedron 2006, 62, 7505–7511; d)
N. Z. Burns, P. S. Baran, R. W. Hoffmann, Angew. Chem. Int.
Ed. 2009, 48, 2854–2867.
[2] a) L. F. Tietze, U. Beifuss, Angew. Chem. Int. Ed. Engl. 1993,
32, 131–163; b) L. F. Tietze, Chem. Rev. 1996, 96, 115–136; c)
J.-C. Wasilke, S. J. Obrey, R. T. Baker, G. C. Bazan, Chem. Rev.
2005, 105, 1001–1020.
[16]
[17]
[3] Tietze’s concept of sequential transformation relies on “a series
of reaction steps in which several bonds are formed or broken,
without the isolation of any intermediates” (see ref.[2a]). More
3150
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 3146–3151

Highly Stereoselective Synthesis of syn-1,3-Diols
7521; e) A. Lorente, M. Pellicena, P. Romea, F. Urpí, Tetrahe-
dron Lett. 2010, 51, 942–945.
[25]
We speculate that the known ability of TBS protecting group
to migrate can be enhanced in this occasion because C3-hy-
droxy is sterically more hindered than C4-hydroxy.
[18] For a parallel boron-mediated approach, see ref.[12]
.
[19] Interestingly, Mahrwald described a single example of such sort
of transformation, which involves the TiCl4-mediated aldol ad-
dition of propiophenone to propanal followed by reduction
with NaBH4 to afford the racemic 1,2-syn-1,3-syn diol in 62%
yield and ds = 91%. R. Mahrwald, B. Ziemer, Tetrahedron
1999, 55, 14005–14012.
[26]
[27]
For the proof of the stereochemistry, see Supporting Infor-
mation.
CCDC-766003 contains the supplementary crystallographic
data for this paper. The data can be obtained free of charge
from The Cambridge Crystallographic Data Center via
www.ccdc.cam.ac.uk/data_request/cif.
[20] a) R. Martín, O. Pascual, P. Romea, R. Rovira, F. Urpí, J. Vil-
arrasa, Tetrahedron Lett. 1997, 38, 1633–1636; b) M. Ferreró,
M. Galobardes, R. Martín, T. Montes, P. Romea, R. Rovira,
F. Urpí, J. Vilarrasa, Synthesis 2000, 1608–1614.
[28] For seminal contributions on stereoselective aldol reactions
based on chiral α-hydroxy ketones, see: a) S. Masamune, W.
Choy, F. A. J. Kerdesky, B. Imperiali, J. Am. Chem. Soc. 1981,
103, 1566–1568; b) N. A. Van Draanen, S. Arseniyadis, M. T.
Crimmins, C. H. Heathcock, J. Org. Chem. 1991, 56, 2499–
2506; c) I. Paterson, D. J. Wallace, S. M. Velazquez, Tetrahe-
dron Lett. 1994, 33, 9083–9086.
[21] For the preparation of bromo ketone 2, see ref.[17d]
[22] It is worth mentioning that the aldol reaction from bromo
ketone 2 usually affords slightly lower yields than the related
process from ketone 1.
[29] A small amount of diastereomerically pure triol 6a was isolated
and was used for a better identification.
[23] Modern Aldol Reactions (Ed.: R. Mahrwald), Wiley-VCH,
Weinheim, 2004.
[24] Valine-derived methyl ketone 3 was chosen because it offers
more diastereoselective aldol reactions than the related lactate-
derived one; see ref.[17e]
Received: March 3, 2010
Published Online: April 28, 2010
Eur. J. Org. Chem. 2010, 3146–3151
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3151