A simple and scalable procedure for TiCl4-promoted aldol reaction

Article abstract of DOI:10.1016/j.tetlet.2010.06.037

TiCl₄-promoted aldol reaction was carried out by adding TiCl₄ to a solution of the aldol reaction substrates and (i-Pr) ₂NEt (DIPEA) in CH₂Cl₂. Compared to the conventional order of addition (sequentially adding TiCl₄, DIPEA, and piperonal to the lactone 2 in CH₂Cl₂), this simplified procedure, gave a much cleaner reaction that could be executed on large scale and without cryogenic cooling. However this procedure provided no stereoselectivity.

Full text of DOI:10.1016/j.tetlet.2010.06.037

Tetrahedron Letters 51 (2010) 4237–4239
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A simple and scalable procedure for TiCl4-promoted aldol reaction
*
Qiaogong Su , Jeffery L. Wood
Synthetic Chemistry Department, GlaxoSmithKline, 709 Swedeland Rd, King of Prussia, PA 19406, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
TiCl4-promoted aldol reaction was carried out by adding TiCl4 to a solution of the aldol reaction sub-
strates and (i-Pr)2NEt (DIPEA) in CH2Cl2. Compared to the conventional order of addition (sequentially
adding TiCl4, DIPEA, and piperonal to the lactone 2 in CH2Cl2), this simplified procedure, gave a much
cleaner reaction that could be executed on large scale and without cryogenic cooling. However this pro-
cedure provided no stereoselectivity.
Received 22 April 2010
Revised 28 May 2010
Accepted 7 June 2010
Available online 11 June 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Since Evans reported the procedure for the direct generation of
Ti-enolates,1 the TiCl4-promoted aldol reaction has emerged as an
extremely powerful method for the formation of carbon–carbon
bonds, and has been applied widely in natural product and phar-
maceutical synthesis because of its high stereoselectivity.2 SB-
235000 (1), a key intermediate in a concise total synthesis of endo-
thelin receptor antagonists,3 was prepared from the aldol reaction
of lactone SB-235619 (2)4 with piperonal. Our initial study re-
vealed that when the reaction was performed via Li-enolate, trans-
metalation with zirconocene dichloride (Cp2ZrCl2) or magnesium
chloride (MgCl2) was required to stabilize the aldol product 1.
Alternatively, the aldol reaction could be carried out via the Ti-eno-
late according to the conventional Evans’ procedure (sequentially
adding TiCl4, DIPEA, and piperonal to the lactone 2 in CH2Cl2),
but we found the reaction needed to be kept below ꢀ72 °C to give
a clean reaction (96% HPLC solution yield) on a gram-scale. How-
ever, as the reaction was scaled up it became less clean and the
yield suffered even at ꢀ72 °C or lower, the solution yield for 30 g
and 100 g dropped to 92% and 82%, respectively, which was not
acceptable from a process chemistry view. The drop in yield may
be attributed to problems mixing the thick slurry of the Ti-enolate
with the aldehyde. Enolate stability was also problematic: decom-
position was observed especially at temperatures above ꢀ72 °C
and worsened upon scale up. After a series of experiments we were
pleased to find that by changing the order of addition, adding TiCl4
last to a solution of 2, piperonal, and DIPEA in CH2Cl2, the reaction
proceeded cleanly at temperatures as high as ꢀ10 °C (Eq. 1). Pre-
sumably, in situ generation of the Ti-enolate and immediate reac-
tion with pre-charged aldehyde avoided the thick enolate slurry
and prevented its decomposition. This simplified addition se-
quence produced a much cleaner reaction than the conventional
order of addition at higher temperature, and most importantly,
was reproducible on large scale. For example (Table 1, entry 1),
when the reaction was carried out at ꢀ10 to ꢀ25 °C, the new pro-
cedure (B) provided 97% HPLC solution yield of 1 on a 280 g scale.
For comparison, the conventional order of addition was also per-
formed at ꢀ10 to ꢀ25 °C (procedure A),5 but only gave 36% yield
even on a 1.0 g small scale.
OMe
OMe
1) (i-Pr)2NEt
CH2Cl2, rt
O
R
RCHO
O
H
+
H
n-PrO
2) TiCl4,
-10 to -25 °C
n-PrO
O
ð1Þ
O
1
HO
SB-235000 (1) R =
SB-235619 (2) (enantiomerically pure)
SB-223885 (3) (racemic mixture)
O
O
* Corresponding author. Tel.: +1 610 270 6908; fax: +1 610 270 4829.
E-mail address: qiaogong.su@gsk.com (Q. Su).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.06.037

4238
Q. Su, J. L. Wood / Tetrahedron Letters 51 (2010) 4237–4239
Table 1
ꢀ25 °C and produced 1:1 ratio of two diastereomers at C1 of SB-
Aldol reaction of aldehydes with 2 or 3
235000, while the conventional order of addition performed below
ꢀ72 °C provided a diastereomer ratio of 2:1. Fortunately, the dia-
stereoselectivity was not a concern for our synthesis because the
stereo center at C1 was destroyed in the following step and both
isomers led to the desired product.3
Entry
1
RCHO
Procedurea
Yield (%)
A
B
36b
97c
OHC
O
O
To study the advantages of this new order of addition, we com-
pared the aldol reactions of racemic lactone SB-223885 (3) with a
(piperonal)
A
B
20b
92b
OHC
OHC
series of aromatic,
a
,b-unsaturated, and aliphatic aldehydes at ꢀ10
2
3
to ꢀ25 °C using both procedures (Table 1). We found that the new
procedure B gave dramatically (>60%) higher yields of the aldol
products than procedure A (entries 2–5), except when the alde-
Br
OH
Ae
Be
26b
87b, 94d
hyde contained an enolizable a-H (entry 6). In this case, both pro-
cedures gave low yields (<20%) of the aldol products.
OMe
CHO
A
B
37b
We also compared the two procedures on the aldol reactions of
piperonal with several other carbonyl compounds (Table 2). Phenol
4, another potential precursor used in our total synthesis, produced
81% yield of the aldol product using procedure B, while only 32%
yield was obtained when employing procedure A (entry 1). We
then compared the two procedures using compounds 5, 6 and 7,
chosen as representatives for five-membered-ring lactones, ke-
tones, and esters, respectively (entries 2–4). While because of
decomposition of the Ti-enolate along with self-condensation of
the carbonyl substrates, no aldol products were detected by LC–
MS when applying procedure A, procedure B produced 34%, 58%,
and 90%, respectively, of aldol products (plus 19% yield of the dehy-
dration product in the case of ketone 6). It is worth noting that, 1H
NMR analysis of the crude reaction mixture of methyl phenylace-
tate (7, entry 4) with piperonal under the new procedure (B) gave
57:43 ratio, almost no stereoselectivity between the syn and anti
isomers.6
100b
4
5
6
(CH3)3CCHO
A
B
A
B
22b
84b, 94d
<20
(CH3)2CHCHO
<20
a
Procedure A: Sequentially adding TiCl4, DIPEA and a solution of aldehyde in
CH2Cl2 to 2 (or 3) in CH2Cl2 at ꢀ10 to ꢀ25 °C. Procedure B: Adding TiCl4 to the
CH2Cl2 solution of 2 (or 3), aldehyde and DIPEA at ꢀ10 to ꢀ25 °C.
b
Isolated yield after chromatography.
Solution yield for a 280 g scale reaction of 2 by HPLC.
Conversion yield based on recovered lactone.
2 equiv of TiCl4 and 2 equiv of DIPEA were used.
c
d
e
Table 2
Aldol reaction of piperonal with carbonyl compounds
Entry
Carbonyl compounds
Procedurea
Yield (%)
In conclusion, we have developed a simple and scalable process
for the TiCl4-promoted aldol reaction by changing the order of
reagent addition. This new procedure gives clean reaction at tem-
peratures as high as ꢀ10 °C (vs ꢀ72 °C), needs only one tempera-
ture-sensitive addition (vs three temperature-sensitive additions),
and is reproducible on large scale comparing to the conventional
order of addition. However this new procedure provided no
stereoselectivity.
Ac
Bc
32b
81b
OMe
1
O
HO
O
(4)
A
B
0d
O
34b
O
2
References and notes
(5)
1. (a) Evans, D. A.; Urpí, F.; Somers, T. C.; Clark, J. S.; Bilodeau, M. T. J. Am. Chem. Soc.
1990, 112, 8215–8216; (b) Evans, D. A.; Rieger, D. L.; Bilodeau, M. T.; Urpí, F. J.
Am. Chem. Soc. 1991, 113, 1047–1049; (c) Evans, D. A.; Clark, J. S.; Metternich, R.;
Novack, V. J.; Sheppard, G. S. J. Am. Chem. Soc. 1990, 112, 866–868.
2. (a) Schetter, B.; Mahrwald, R. Angew. Chem., Int. Ed. 2006, 45, 7506–7525; (b)
Crimmins, M. T.; She, J. Synlett 2004, 1371–1374; (c) Perkins, M. V.; Sampson, R.
A. Org. Lett. 2001, 3, 123–126; (d) Abbineni, C.; Sasmal, P. K.; Kukkanti, K.; Iqbal,
J. Tetrahedron Lett. 2007, 48, 4259–4262.
3. Wood, J. L.; McGuire, M. A.; Mills, R. J.; Pridgen, L. N.; Yu, M. S.; Su, Q. U. S. Patent
6,114,549, 2000; U.S. Patent 6,441,173, 2002.
4. McGuire, M. A.; Shilcrat, S. C.; Sorenson, E. Tetrahedron Lett. 1999, 40, 3293–
3296.
A
B
0d
O
58e
O
3
4
O
(6)
A
B
0d
COOMe
(7)
90b
a
b
c
Same as footnote a in Table 1.
Isolated yield after chromatography.
5. See Refs. 1c and 2b for performing the aldol reactions by the conventional order
of addition at temperatures as high as 0 °C.
2 equiv of TiCl4 and 2 equiv of DIPEA were used.
Messy reaction with self-condensation products detected by LC–MS. Piperonal
6. Examples of typical procedures for TiCl4-promoted aldol reactions:
Procedure A. TiCl4 (0.39 mL, 1.2 equiv) was added dropwise to a solution of
methyl phenylacetate (7) (0.47 g, 1.05 equiv) in CH2Cl2 (2.5 mL) at ꢀ10 to
ꢀ25 °C under N2. After 2 min, DIPEA (0.63 mL, 1.2 equiv) was added dropwise.
The resulting dark slurry was stirred at ꢀ10 to ꢀ25 °C for 15 min. After the
dropwise addition of a CH2Cl2 (2 mL) solution of piperonal (0.45 g, 3.0 mmol,
1.0 equiv), stirring was continued at ꢀ10 to ꢀ25 °C for 15 min. The reaction was
quenched with H2O (8 mL) and stirred to 15 °C. The aqueous layer was extracted
with CH2Cl2 (2 mL). LC–MS analysis of the CH2Cl2 layer indicated no desired
aldol products, some self-condensation products were detected.
d
was unreacted.
e
Plus 19% yield of the dehydration product from the aldol adduct.
The results were surprising since initially we were concerned
about formation of TiCl4 and DIPEA complex in view of the litera-
ture report on the order of reagent addition: that it is critical for
substrate-TiCl4 complexation to precede the introduction of base
because the uncomplexed TiCl4 would otherwise irreversibly com-
plex with DIPEA resulting in no enolization.1a It is important to
point out that, procedure B gave no stereoselectivity at ꢀ10 to
Procedure B.
A mixture of methyl phenylacetate 7 (0.47 g, 1.05 equiv),
piperonal (0.45 g, 3.0 mmol, 1.0 equiv) and DIPEA (0.63 mL, 1.2 equiv) in
CH2Cl2 (4.5 mL) was stirred at rt under N2 for 5 min and then cooled to ꢀ25 °C.
TiCl4 (0.39 mL, 1.2 equiv) was then added dropwise. The reaction was kept at
ꢀ10 to ꢀ25 °C during the addition and stirred at this temperature for another
15 min after the addition. The reaction was quenched with H2O (8 mL) and

Q. Su, J. L. Wood / Tetrahedron Letters 51 (2010) 4237–4239
4239
stirred to 15 °C. The aqueous layer was extracted with CH2Cl2 (2 mL). The
combined CH2Cl2 layer was washed with H2O (2 mL) and concentrated.
Purification by silica gel column chromatography (10–50% EtOAc in Hexanes)
provided totally 0.79 g (90% yield) of the syn and anti aldol products. Isomer 1:
1H NMR (300 MHz, CDCl3) d 3.56 (s, 3H), 3.85 (d, J = 9.0 Hz, 1H), 5.21 (d,
J = 9.0 Hz, 1H), 5.95 (s, 2H), 6.73–6.86 (m, 3H), 7.31–7.42 (m, 5H). 13C NMR
(75 MHz, CDCl3) d 52.5, 60.2, 75.2, 101.4, 107.5, 108.3, 120.7, 128.4, 129.1,
129.5, 135.2, 135.4, 147.6, 148.0, 173.1; MS (m/z): 283.3 [M+HꢀH2O]+, 323.3
[M+Na]+. Isomer 2: 1H NMR (300 MHz, CDCl3) d 3.74 (s, 3H), 3.86 (d, J = 9.0 Hz,
1H), 5.13 (d, J = 9.0 Hz, 1H), 5.90 (dd, J = 6.0, 3.0 Hz, 2H), 6.47–6.50 (m, 1H),
6.59 (d, J = 9.0 Hz, 1H), 6.72 (d, J = 3.0 Hz, 1H), 7.10–7.13 (m, 2H), 7.20–7.22 (m,
3H). 13C NMR (75 MHz, CDCl3) d 52.7, 60.3, 76.8, 101.3, 107.3, 108.1, 120.8,
128.0, 128.9 (2C), 135.1, 135.6, 147.4, 147.9, 174.3; MS (m/z): 283.3
[M+HꢀH2O]+, 323.3 [M+Na]+.