Asymmetric alkylation of diarylmethane derivatives. Improved results using methoxyethoxy substituent

Article abstract of DOI:10.1016/j.tetlet.2004.05.058

Alkylation of 2-methoxyethoxyphenyl phenyl methane using sec-BuLi and (-)-sparteine has been carried out in excellent yields and up to 94% ee. The best results were obtained in allylation reactions but methylation, ethylation, benzylation and trimethylsilylation have all been carried out with acceptable ee.

Full text of DOI:10.1016/j.tetlet.2004.05.058

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 5481–5483
Asymmetric alkylation of diarylmethane derivatives.
Improved results using methoxyethoxy substituent
James A. Wilkinson,a,* Steven B. Rossington,a John Leonardb and Nigel Hussainc
aCentre for Drug Design, Biosciences Research Institute, Cockcroft Building, University of Salford, Salford M5 4WT, UK
bAstraZeneca Process R+D, Silk Road, Cheshire, Macclesfield SK10 2NG, UK
cGlaxoSmithKline Chemical Development, Old Powder Mills, Kent, Tonbridge TN11 9AN, UK
Received 24 March 2004; revised 28 April 2004; accepted 7 May 2004
Abstract—Alkylation of 2-methoxyethoxyphenyl phenyl methane using sec-BuLi and ())-sparteine has been carried out in excellent
yields and up to 94% ee. The best results were obtained in allylation reactions but methylation, ethylation, benzylation and tri-
methylsilylation have all been carried out with acceptable ee.
Ó 2004Elsevier Ltd. All rights reserved.
In a previous paper we reported the asymmetric alkyl-
O
O
ation of prochiral diphenylmethane derivatives includ-
ing 1 using sec-BuLi and ())-sparteine.1 The best result
is shown (Scheme 1). When the 2-substituent was an
unprotected hydroxyl group the selectivity was reversed
giving a 46% ee of the S-product.
Li
Ph
Figure 1.
(Scheme 2).3 Initially, allylation of 4 was investigated
since this had been a profitable area for reactions of 1.
The results are shown in Table 1. Allylation under
conditions A (entry 1, substrate, base and ())-sparteine
stirred together at )78 °C for 6 h before introduction of
electrophile) gave 5 in a 44% ee, lower than that ob-
tained in the methoxy series. However, use of a warm–
cool cycle according to the procedure of Beak (entry 2,
conditions B: substrate, base and ())-sparteine mixed at
)78 °C, warmed to )20 °C for 1 h and recooled) resulted
in a 50% ee.4 When the electrophile was added in two
batches of 0.5 equiv with a warm–cool cycle in between
the additions the ee rose again (entry 3, conditions C).5
Finally when the electrophile was allyl tosylate, under
the same conditions (entry 4) the reaction gave a very
pleasing 94% ee with no significant loss of yield.
Here we report asymmetric alkylations of a similar
substrate, compound 4, substituted with a methoxyeth-
oxy group. The rationale behind the choice of the
methoxyethoxy group was to provide an extra oxygen to
coordinate the lithium in the reactive species (Fig. 1).
This coordination effect has been used to promote ortho-
lithiation of phenol derivatives.2
The synthesis of 4 was carried out, trivially, by O-
alkylation of the commercially available phenol 3
OMe
OMe
i.sec-BuLi, (-)-sparteine, -78 °C
ii. allyl bromide
Ph
Ph
1
2
95% yield, 60% ee
Scheme 1.
O
O
OH
Ph
BrCH2CH2OMe
NaH, THF, rt
80%
Ph
Keywords: Asymmetric; Alkylation; Diarylmethane; Sparteine.
* Corresponding author. Tel.: +44-161-295-4046; fax: +44-161-295-
5111; e-mail: j.a.wilkinson@salford.ac.uk
3
4
Scheme 2.
0040-4039/$ - see front matter Ó 2004Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.05.058

5482
J. A. Wilkinson et al. / Tetrahedron Letters 45 (2004) 5481–5483
Table 1. Asymmetric allylations of 4
OCH2CH2OMe
OCH2CH2OMe
Ph
conditions
X
Ph
5
4
Entry
R–X
Conditions
Yield (%)
Config.
Ee
1
Allyl–Br
Allyl–Br
Allyl–Br
C
A
B
84
85
96
R
R
R
R
44
50
71
94
2
3
4Allyl–OTs
C
93
Conditions: (A) Mixture of 4, ())-sparteine (1.1 equiv) and sec-BuLi (1.1 equiv) in ether stirred at )78 °C for 6 h followed by addition of alkylating
agent (1.2 equiv) and warming to rt. (B) Mixture of 4, ())-sparteine (1.1 equiv) and sec-BuLi (1.1 equiv) in ether mixed at )78 °C, then warmed to
)20 °C for 1 h and recooled to )78 °C before addition of electrophile (1.2 equiv) and warming to rt. (C) Mixture of 4, ())-sparteine (1.1 equiv) and
sec-BuLi (1.1 equiv) in ether mixed at )78 °C, then warmed to )20 °C for 1 h and recooled to )78 °C before addition of electrophile (0.5 equiv).
Mixture warmed to )20 °C for a further 1 h before recooling to )78 °C and addition of electrophile (0.5 equiv) and warming to rt. Ee values were
calculated by conversion to 6.
These results are in accordance with those obtained by
Beak for reactions of N-o-pivaloyl ethylaniline but dis-
agree with our results for the methoxy compound 1
(Scheme 1), where the best ee was obtained using con-
ditions A. Our assumption is that the pair of diaste-
reomeric organolithium/sparteine complexes derived
from 1 were interconverting rapidly at )78 °C. In the
case of 4 it would seem that the organolithium species is
less configurationally labile (as would be expected with
extra stabilisation from the two oxygens) and needs to
be warmed to allow equilibration.
compound has been synthesised by Meyers who as-
signed the absolute configuration as shown. Enantio-
meric excesses could also be determined by use of the
chiral shift reagent 2,2,2-trifluoro-1-(9-anthryl)ethanol
(TFAE), which cleanly split the methoxy singlets of the
two enantiomers in the 1H NMR spectrum.8 The results
correlated well with those obtained by comparison of
optical rotation.
We were then able to extend the methodology to a range
of other electrophiles as shown in Table 2. Methylations
(entries 1–3) were achieved with marked improvements
over similar reactions performed on 1 (the best ee for
methylation of 1 was 6%). Using a warm–cool cycle and
batch addition of electrophile (conditions C from Table
1) the ee was still very modest for methyl iodide but
could be increased considerably by using methyl triflate
or methyl tosylate as the alkylating agent. This increase
in enantioselectivity from halide to a tosylate is prece-
dented in the work of Beak and has been rationalised as
a consequence of slower reaction and possible coordi-
nation of the sulfonyl oxygen to lithium.9 Ethylation,
benzylation and reaction with trimethylsilyl chloride
(entries 4–6) were also carried out in good yields and at
least respectable ee. Enantiomeric excesses were deter-
mined using TFAE analysis on the methoxy singlet
(compounds 7–9) or the trimethylsilyl singlet (com-
pound 10).
The ee and predominant enantiomer (R) were confirmed
by conversion of product 5 into the known lactone 6 and
comparison of optical rotation (Scheme 3).6 This was
achieved in two steps and a 70% overall yield by oxi-
dative cleavage of the double bond with potassium
permanganate in the presence of a crown ether and
treatment with boron tribromide to remove the meth-
oxyethyl group with concomitant cyclisation.7 This
O
O
OCH2CH2OMe
Ph
i. KMnO4, 18-crown-6, CH2Cl2/H2O
ii. BBr3, 0 °C, CH2Cl2
70%
Ph
6
5
Scheme 3.
Table 2. Asymmetric alkylations/silylation of 4
OCH2CH2OMe
OCH2CH2OMe
conditions C (Table 1)
R
R-X
Ph
Ph
7-10
4
Entry
Product
R–X
Conditions
Yield
Config.
Ee
1
2
3
4
5
6
7
MeI
C
C
C
C
C
C
86
90
95
65
90
91
R
12
36
58
70
68
82
7
MeOTf
MeOTs
EtI
R
7
R
8
Rꢀ
Rꢀ
Rꢀ
9
10
BnBr
TMS–Cl
* Indicates major enantiomer assigned by analogy with other reactions.

J. A. Wilkinson et al. / Tetrahedron Letters 45 (2004) 5481–5483
5483
Acknowledgements
OH
Me
OCH2CH2OMe
AlI3, CH3CN, reflux
12 hrs
Me
Ph
75%
We thank the EPSRC and GlaxoSmithKline for a
CASE award (S.B.R.).
Ph
7
11
Scheme 4.
References and notes
Removal of the methoxyethyl group from compounds
lacking the assistance of an internal carboxylic acid re-
quired rather vigorous conditions as exemplified by the
conversion of 7 to the phenol 11 (Scheme 4).10 The
reaction proceeded in acceptable yield allowing us to
assign the stereochemistry of the major product as R
by comparison of optical rotation. Products 8–10 are
assigned as having the same configuration as the methyl
and allyl products.
1. Wilkinson, J. A.; Rossington, S. B.; Leonard, J.; Hussein,
N. Tetrahedron Lett. 2004, 45, 1191–1193.
2. Wada, A.; Kanamoto, S.; Nagai, S. Chem. Pharm. Bull.
1985, 33, 1016–1022.
3. All new compounds gave satisfactory spectral data
including HRMS.
4. Basu, A.; Beak, P. J. Am. Chem. Soc. 1996, 118, 1575–
1576; see also: Basu, A.; Thayumanavan, S. Angew.
Chem., Int. Ed. 2002, 41, 717–738, and references cited
therein.
5. Park, Y. S.; Boys, M. L.; Beak, P. J. Am. Chem. Soc. 1996,
118, 3757–3758.
6. Meyers, A. I.; Smith, R. K.; Whitten, C. E. J. Org. Chem.
To conclude, we have increased the stereoselectivity of
diarylmethane alkylation considerably by employing an
extra coordinating oxygen in the stabilising group. In
addition to recording excellent yields and some high
enantiomeric excesses we have demonstrated the versa-
tility of the reaction with regard to different electrophiles
and shown that the methoxyethyl group can be removed
efficiently.
1979, 44, 2250–2256, ½aꢁ of optically pure S-isomer is
D
+43.7 (c 5.6, benzene).
7. Oxidative cleavage: Lee, D.; Chang, V. J. Org. Chem.
1978, 43, 1532–1536.
8. Between 1 and 2 equiv of TFAE were required to effect full
resolution. The methoxy peak for the major, R-enantio-
mer was found at 3.296 ppm, that for the minor at
3.285 ppm.
Future work will include investigations into the origin of
the enantioselectivity, screening other potential coordi-
nating groups and the application of the methodology in
synthesis.
9. Thayumanavan, S.; Basu, A.; Beak, P. J. Am. Chem. Soc.
1997, 119, 8209–8216.
10. Bhatt, M. J.; Babu, R. J. Tetrahedron Lett. 1984, 25, 3497–
3500.