Mapping the mechanism of nickel-ferrophite catalysed methylation of baylis-hillman-derived SN2′ electrophiles

Article abstract of DOI:10.1002/ejoc.200801029

Enantioselective Ni-catalysed methylation of Baylis-Hillman-derived allylic electrophiles in the presence of ferrophite ligands has been investigated computationally and experimentally. The sense and degree of enantioselectivity attained is independent of both the leaving group and the isomeric structure of the initial allylic halide. DFT studies support the selective formation of a limited number of energetically favoured anti and syn π-allyl intermediates. The observed regio- and enantioselectivity can be rationalised based on the energetics of these structures.

Full text of DOI:10.1002/ejoc.200801029

FULL PAPER
DOI: 10.1002/ejoc.200801029
Mapping the Mechanism of Nickel-Ferrophite Catalysed Methylation of
Baylis–Hillman-Derived SN2Ј Electrophiles
Andrew Novak,[a] Maria José Calhorda,[b] Paulo Jorge Costa,[c] and Simon Woodward*[a]
Keywords: Nickel / Aluminum / P ligands / Asymmetric catalysis / Alkylation / Density functional calculations
Enantioselective Ni-catalysed methylation of Baylis–Hill-
man-derived allylic electrophiles in the presence of ferro-
phite ligands has been investigated computationally and ex-
perimentally. The sense and degree of enantioselectivity at-
tained is independent of both the leaving group and the iso-
meric structure of the initial allylic halide. DFT studies sup-
port the selective formation of a limited number of energeti-
cally favoured anti and syn π-allyl intermediates. The ob-
served regio- and enantioselectivity can be rationalised
based on the energetics of these structures.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
Introduction
We briefly reported an enantioselective (49–94% ee)
methylation of 1–2[1] in the presence of the (Rp) planar chi-
ral ferrophite ligands LA-LB (Scheme 1).[2] The allylic chlo-
ride 1 and acetate 2 were chosen as these substrates are
readily available through Baylis–Hillman chemistry and
lead to useful α-chiral methylene carboxylate products.[3] In
comparison to our previously investigated copper chemis-
try, that afforded only γ-alkylated products,[3,4] the Ni-ferro-
phite system delivered mixtures of γ-methylated 3 and α-
methylated 4 products (with [4]/[3] down to 0.16). Moderate
to good levels of stereoinduction were also achieved (49–
94% ee). Attempts to rationalise the observed trends in re-
gio and stereochemistry of these reactions were hindered by
lack of basic information on the behaviour of the nickel π-
allyl species. Therefore theoretical calculations on the ferro-
phite bearing species that can be supposed to be key inter-
mediates in these catalytic transformations were carried out.
Compared to their PdII analogues, NiII(η3-allyl)(X)(L)
species are much more stereochemically rigid.[5] Typically,
the onset of η3-η1 allyl interconversion does not begin until
ca. 80 °C and this is seldomly observed due to decomposi-
tion before this temperature is attained. Rotation of NiII(η3-
allyl) species typically demonstrate barriers of ca. 16–
Scheme 1. Preliminary methylation studies of allylic electrophiles
1–2.
17 kcalmol–1 leading to minimal interconversion in these
square-planar complexes below ambient temperature. If ra-
pid exchange between the diastereomeric nickel π-allyl in-
termediates in an asymmetric process is not achieved then
the relative populations of these species in asymmetric pro-
cess could become highly important. We proposed to
undertake further experiments supported by DFT calcula-
tions as means of attaining pointers towards improved re-
gio- and stereoselectivity in the chemistry of Scheme 1.
[a] School of Chemistry, University of Nottingham,
Nottingham, NG7 2RD, UK
Fax: +44-115-9513564
E-mail: simon.woodward@nottingham.ac.uk
[b] Departamento de Química e Bioquímica, CQB, Faculdade de
Ciências, Universidade de Lisboa,
Results and Discussion
1749-016 Lisboa, Portugal
Structural Assignments
Fax: +351-21-7500088
E-mail: mjc@fc.ul.pt
The Ni(acac)2-LA process of Scheme 1 provides a mix-
ture of 3 and 4 whose chromatographic properties on silica
gel are identical. To allow accurate quantification of their
separate yields by quantitative GC access to pure (Ϯ)-3 free
[c] Departamento de Química, CICECO, Universidade de Aveiro,
3810-193 Aveiro, Portugal
E-mail: pjcosta@ua.pt
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
898
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 898–903

Methylation of Baylis–Hillman-Derived Electrophiles
of 4 (and vice versa) was required. The former was attained for the enantiopurity of the two samples). Thus, it was con-
by γ-specific ZnMe2-based methylation, using CuCN/ firmed that (+)-(S)-3 elutes first upon chiral GC on oc-
NBu4Br.[4] We found pure 4 was attained by AlMe3 methyl- takis(2,6-di-O-methyl-3-O-pentyl)-γ-cyclodextrin (see Sup-
ation of 1 in the presence of un-ligated Ni(acac)2. The porting Information).
enantiomers of 3 were initially assigned[1] by assuming the
second eluting species under GC octakis(2,6-di-O-methyl-
Regioisomer Effects
3-O-pentyl)-γ-cyclodextrin analysis corresponded to the (R)
enantiomer as it had in the analogous ethyl addition prod-
uct (–)-(R)-5 (Scheme 2).[4] The assignment of the ethyl de-
rivative is secured on the basis of ester hydrolysis and an
X-ray structure of the insoluble (R)-PhCHMeNH2 amine
salt (R,R)-6.[4] To ensure the correctness of this assignment
racemic 3 was hydrolysed and the derived acids exposed to
(R)-phenylethylamine. In identical behaviour to that of (R)-
5 the less soluble amine salt afford (–)-3 upon re-esterifica-
tion with TMSCHN2. The assignment of R stereochemistry
to (–)-3 is also in accord with the only other literature re-
port on this compound: (–) equivalent to R, 69% ee, at-
tained by oxidative degradation of asymmetric “Heck”
products.[6] Our more soluble (R)-phenylethylamine salt
yielded (+)-3 with an equal and opposite rotation (allowing
The reactivity of regioisomeric allylic halides starting
materials was compared to the “linear” species 1–2
(Table 1). Ligand LA was used throughout to allow direct
comparison to be made. In all cases, where there is a reac-
tion, only the 4/3 ratio is affected by the substrate choice
with the enantioselectivity of the derived (S)-3 being con-
stant within the measurement error.
One explanation for these observations is that a mixture
of non converting diastereoisomeric Ni(η3-allyl) species is
present in the reaction mixture leading to 3 and 4. If the
relative populations of these are defined at the moment of
oxidative addition then the observed regiochemical behav-
iour might be accounted for. Direct observation of the ac-
tual catalytic reaction mixture described in Schemes 1 and 2
by 31P NMR shows the presence of only four ferrophite
coordinated species, one of significantly higher concentra-
tion, when substrate 1 is used (δP = 206.9, 205.1, 195.1,
194.0; ratio 1.6:1.0:0.9:7.7). As nothing was known about
the structures of Ni-coordinated ferrophite complexes we
turned our attention to a computational study of possible
Ni(π-allyl) intermediates in this reaction.
Theoretical Studies
η3-Coordination
of
the
substituted
allyl
“CH2=C(CO2Me)CHPh” to a “NiIIMe(LA) unit is expected
to generate 16 possible intermediates due to: (i), the enanti-
oface of the allyl selected; (ii), the presence of two rotamers
per complex; and (iii), the placement of the phenyl substitu-
ent either anti or syn to the central CO2Me. The factors
that affect the relative energetics of these 16 possible combi-
nations are not expected to be obvious as only small energy
differences between the allyls are expected. This expectation
Scheme 2. Partial resolution and stereochemical assignment of 3.
All optical rotations in CHCl3: (R)-5 c = 1.00; (R)-3 c = 0.62; (S)-
3 c = 0.75.
Table 1. Effect of varying the electrophile structure on selectivity.[a]
Entry
Allyl electrophile
% Conversion
% Yield of 3+4
4:3 ratio
% ee
1
2
3
4
5
1
2
7
8
9
82
50
73
26
24
71
–
0.52
0.30
0.85
0.49
–
71, S
71, S
71, S
70, S
–
31[b]
80[b]
Ͻ5[b]
[a] Reactions carried out on 0.25 mmol substrate with 0.02/0.04/2 equiv. Ni(acac)2/LA/AlMe3 in THF (3.0 mL), 10 °C, 3 d. Conversions,
1
yields regio- and enantioselectivities by GC. [b] Determined by H NMR spectroscopy.
Eur. J. Org. Chem. 2009, 898–903
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
899

A. Novak, M. J. Calhorda, P. J. Costa, S. Woodward
FULL PAPER
is not always apparently the case. In early work,
Ni(C6Cl5)(η3-CH2CHCHMe)(PMe2Ph) was shown to exist
as a single syn species by 1H NMR spectroscopy.[7] Our own
investigations of the NiII/LA system were carried out by
DFT methods[8] using a B3LYP/sdd,6-311G** approach.
Because the pre-existing anti arrangement in the initial al-
lylic electrophiles 1–2 and 7 is expected to favour formation
of anti π-allyl species these were investigated first. In the
nomenclature of Scheme 3 the “V” and “Λ” subscripts refer
to the conformation of the allyl rotamer with respect to the
ferrocenyl unit (always placed topmost in Scheme 3).
The pre-existing anti arrangement of the leaving group
and phenyl function in our allylic electrophiles means that
the equivalent syn allyls to Scheme 3 can only be accessed
by a anti-syn exchange process; predicted to be inefficient
under our reaction conditions (10 °C). Nevertheless, for
completness the equivalent syn-I–IV structures were calcu-
lated (Scheme 4).
Scheme 3. Calculated (B3LYP/sdd,6-311G**) relative energies of
anti-NiMe(η3-allyl)LA complexes.
Rewardingly, the calculations indicate that the lowest en-
ergy π-allyl species anti-IIΛ correlates to the formation of a
preponderance of (S)-3 in the actual catalytic reaction. Not
discounting the clear importance of transition state ef-
Scheme 4. Calculated (B3LYP/sdd,6-311G**) relative energies of
fects,[9] the energy differences between the calculated lowest syn-NiMe(η3-allyl)LA complexes.
energy (S)-3 vs. (R)-3 producing π complexes (anti-IIΛ vs.
The lowest energy specie syn-IΛ does not correlate with
anti-IIIV) corresponds well with the enantioselectivity range
observed for LA,B [∆Go(283 K) = 1.30 kcalmol–1, K =
0.0985 equiv. to 82% eecalcd.]. The origin of the poorer re-
gioselectivity in Scheme 1 is also clear. The presence of near
iso-energetic anti-IIIΛ leads to significant production of 4
upon reductive elimination.
We predict that to down weight the contribution of the
anti-IIIΛ component the steric profile of one or both Cp
ferrocenyl ring should be further increased. However, such
ligand modifications are non trivial synthetic exercises and
were not pursued. It is also clear that in the absence of
added LA,B an anti-IIIΛ related species becomes the major
reaction manifold and this might be used for highly selec-
tive formation of the α homologation product 4.
the observed stereochemistry of the final product (S)-3 (see
Scheme 1). Interestingly, however, the computational be-
haviour of the syn-I–IV group mirrors the behaviour of
Ni(C6Cl5)(η3-CH2CHCHMe)(PMe2Ph) in that a single
dominant syn isomer is predicted. In our case access to
these species is prevented by the anti arrangement in the
starting electrophiles and the nonconversion of the anti-syn
allyl forms under the reaction conditions.
Generalised α-Homologation of Allyl Chlorides
In the absence of any added ligand, 4 was formed as
essentially the sole AlMe3 SN2 displacement product (α/γ
900
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 898–903

Methylation of Baylis–Hillman-Derived Electrophiles
attack Ͼ50:1). Prompted by the low energy of the anti-IIIΛ possibility of accessing a syn-IΛ-based route to (R)-3 should
π-allyl further investigation into this selective one carbon be viable. Overall, the study clearly points to three potential
homologation reaction was undertaken. A range of aryl- routes for improving the selectivity of the process: (i) ad-
substituted Baylis–Hillman-derived allylic chlorides 10–19, ditional groups placed on one or both cyclopentadienyl
attained by standard methods were trialled (Table 2). In all rings of LA,B might be useful in destabilising anti-IIIΛ im-
cases full conversion of the starting chloride was achieved. proving the regiochemistry; (ii) using (E)-1, rather than (Z)-
Moderate to good yields of the α-products were isolated 1, should allow the syn (rather than the anti) π-allyl mani-
with high to excellent α/γ selectivity being attained. Unfor- fold to be populated resulting in dramatically improved re-
tunately, alkyl-substituted 19 gave an unclean reaction gio- and enantioselectivity due to the lower relative energy
which contained only trace amounts of the desired product of syn-IΛ; (iii) identification of a different ligand type that
1
by H NMR spectroscopy (Entry 11). Overall though this allowed fast anti-syn interconversion and simultaneous sta-
Ni chemistry is particularly useful as a complimentary bilisation of syn-IΛ would allow equivalent use of (Z)-1 to
method to the γ-selective copper-catalysed chemistry our good effect. In the cases of (ii) and (iii) formation of (R)-3
group reported earlier.[3]
is predicted. Investigation of these approaches is underway
in our laboratory.
Table 2. Highly selective α-methylation of allylic chlorides under
nickel catalysis.[a]
Experimental Section
General: Procedures were performed under atmospheres of argon
or nitrogen using standard Schlenk techniques. Trimethylalumin-
ium (2  in heptane) and ZnMe2 (2  in toluene) were commercial
products (Sigma–Aldrich) used as supplied. Column chromatog-
raphy was performed using Fluorochem (35–70 micron) silica gel
1
and TLC analysis using Merck Kieselgel 60 F245+366. H and 13C
NMR spectra were recorded on either a Bruker AV400, DRX500
or JEOL EX270 spectrometers, using tetramethylsilane as an in-
ternal standard. Chemical shifts are quoted in ppm and coupling
constants (J) are given in Hz. Chiral GC analysis was performed
on a Varian 3380 gas chromatograph using suitable cyclodextrin
based stationary phases as indicated using low molecular weight
alkanes, such as pentadecane, as internal standards. Allylic halides,
acetates and methoxides were prepare by literature procedures: 1,
2, 7, 8, 9–19.[10]
Entry R1
R2
α-Product % Yield
α/γ[b]
1
2
Ph
Ph
Me
Et
4
49
68
77
71
69
76
77
75
73
43
Ͻ5
Ͼ50
5.26
Ͼ50
Ͼ50
Ͼ50
6.67
7.70
9.09
25
20
21
22
23
24
25
26
27
28
29
3
4
5
6
7
8
9
4-(OMe)C6H4 Me
4-FC6H4
4-ClC6H4
4-MeC6H4
4-tBuC6H4
2-BrC6H4
1-C10H8
Me
Me
Me
Me
Me
Me
Me
Me
(؎)-Methyl 2-Methylene-3-phenylbutanoate (3): Under an argon at-
mosphere a 2  solution of ZnMe2 in dry toluene (0.5 mL;
1.0 mmol) was added to a cooled (–20 °C) stirred solution of allylic
chloride 1 (0.50 mmol), containing CuCN (9.0 mg; 0.10 mmol) and
NBu4Br (32.2 mg; 0.10 mmol) in dry THF (2 mL). The reaction
mixtures were stirred for 1 h at –20 °C and then warmed to room
temp. over 20 h. The reaction mixture was quenched by the slow
addition of 2  HCl (2 mL). The product was extracted with Et2O
(10 mL), the combined organic layers were dried with MgSO4, con-
centrated and purified by flash column chromatography using mix-
10
11
2-C10H8
n-C7H15
Ͼ50
–
[a] Reactions carried out on 0.50 mmol substrate with 0.02/2 equiv.
Ni(acac)2/AlMe3 in THF (5.0 mL), 10 °C, 24 h. Isolated yields. [b]
Determined by H NMR spectroscopy.
1
Conclusions
All experimental and computational data are consistent tures of Et2O/hexane as eluent to afford (Ϯ)-3 as a colourless oil
1
(81 mg; 85%); Rf = 0.32 (hexane/Et2O, 9:1). H NMR (400 MHz,
CDCl3): δ = 7.30–7.16 (m, 5 H, Ar), 6.28 (s, 1 H, =CH2α), 5.61 (s,
1 H, =CH2β), 4.03 (q, J = 7.2 Hz, 1 H, CHMe), 3.67 (s, 3 H, OMe),
1.42 (d, J = 7.2 Hz, 3 H, Me) ppm. 13C NMR (100 MHz, CDCl3):
δ = 167.6, 145.0, 144.5, 128.6, 127.6, 126.5, 124.0, 52.0, 40.7, 20.9
with Nickel-ferrophite methylation of allylic electrophiles 1
proceeding via formation of a somewhat favoured anti-allyl
species, anti-IIΛ. This intermediate is formed irrespective of
the leaving group (starting materials 2 and 7) or the re-
gioisomer of the electrophile employed (2 vs. 8). The origin
of the poorer regioselectivity has been traced to the pres-
ence of a near iso-energetic anti-allyl (IIIΛ) in which the Ni-
Me and α-allyl termini are mutually cis. Rapid reductive
elimination leads to the α-homologation product. In the ab-
ppm. IR (thin film): ν = 2969, 1721 (C=O), 1627, 1437, 1149, 947
˜
cm–1. EIMS: m/z (%) = 190 (92) [M+], 158 (42), 130 (100). HR-
EIMS: m/z 190.0998 calcd. for C12H14O2; found 190.0998; elemen-
tal analysis: C 75.76, H 7.42 calcd. for C12H14O2; found C 75.75,
H 7.37. The GC assay: 25 m (2,6-O-dimethyl,3-O-pentyl)-γ-cyclo-
sence of added ligand a related process dominates leading dextrin silica column, 120 °C (30 min): 10 °C/min: 200 °C (20 min),
He carrier gas, 12 psi, (S): 14.8 min. (R): 15.2 min.
to the development of a synthetically useful process for α-
methylation. Based on their calculated energies syn allyl
species are not important in this chemistry. However, in-
triguingly one of syn-allyls (IΛ) is predicted to be of con-
(E)-Methyl 2-Benzylidenebutanoate (4): A flame-dried Schlenk tube
under argon was charged with Ni(acac)2 (2.4 mg, 4 mol-%) and an-
hydrous THF (5 mL) was stirred at –10 °C for 5 min. To this the
siderably lower energy than its seven partners. Thus, the neat allylic chloride 1 (0.50 mmol) was added and the mixture
Eur. J. Org. Chem. 2009, 898–903
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
901

A. Novak, M. J. Calhorda, P. J. Costa, S. Woodward
FULL PAPER
stirred at –10 °C for 10 min. A solution of AlMe3 (2  solution in
heptane, 0.5 mL; 1.0 mmol) was added dropwise to the reaction
mixture, which was then warmed to 10 °C. After 24 h at 10 °C the
black reaction mixture was quenched with 2  HCl (2 mL). The
product was extracted into Et2O (2ϫ5 mL), the combined organic
layers were dried with MgSO4. After removal of the solvent the
crude product was purified by flash column chromatography using
mixtures of Et2O/hexane to yield 4[11] as a olourless oil (47 mg;
49%) (63% by GC); Rf = 0.32 (hexane/Et2O, 9:1). 1H NMR
(400 MHz, CDCl3): δ = 7.66 (s, 1 H, =CH), 7.40–7.35 (m, 5 H,
Ar), 3.83 (s, 3 H, OMe), 2.57 (q, J = 7.6 Hz, 2 H, CH2Me), 1.18
(t, J = 7.6 Hz, 3 H, Me) ppm. 13C NMR (100 MHz, CDCl3): δ =
168.8, 138.6, 135.8, 134.8, 129.2, 128.3, 128.0, 51.9, 20.8, 13.9 ppm.
128.2, 113.9, 55.2, 51.8, 20.7, 13.7 ppm. IR (CDCl solution): ν =
˜
3
2952, 2839, 1694 (C=O), 1608, 1514, 1435, 1310, 1258, 1181, 1131,
1035, 833 cm–1. EIMS: m/z (%) = 220 (100) [M+], 189 (18), 160
(53), 145 (46), 115 (11). HR-EIMS: m/z: calcd. for C13H16O3:
220.1099; found 220.1097; elemental analysis: C 70.89, H 7.32
calcd. for C13H16O3; found C 70.73, H 7.37. The HPLC assay: Chi-
ralcel OD-H, isocratic (n-hexane/iPrOH, 99.5:0.5, flow
0.5 mLmin–1), λ = 254 nm, 24.9 min.
(E)-Methyl 2-(4-Fluorobenzylidene)butanoate (22): Prepared analo-
gously to 4 using 0.50 mmol of 12. Colourless oil 74 mg; 71%; Rf
= 0.70 (pentane/Et2O, 9:1). 1H NMR (400 MHz, CDCl3): δ = 7.60
(s, 1 H, =CH), 7.37–7.33 (m, 2 H, Ar), 7.10–7.06 (m, 2 H, Ar), 3.82
(s, 3 H, OMe), 2.52 (q, J = 7.4 Hz, 2 H, CH2Me), 1.16 (t, J =
7.4 Hz, 3 H, CH2Me) ppm. 13C NMR (100 MHz, CDCl3): δ =
IR (CHCl3 solution): ν = 3027, 2952, 2877, 1698 (C=O), 1628,
˜
1495, 1435, 1312, 1284, 1238, 1204, 1130, 1045, 810 cm–1. EIMS:
m/z (%) = 213 (26) [MNa+], 191 (15) [MH+]. HR-EIMS: m/z: calcd.
for C12H14O2Na: 213.0897; found 213.0898. The GC assay: 25 m
(2,6-O-dimethyl,3-O-pentyl)-γ-cyclodextrin silica column, 120 °C
(30 min): 10 °C/min: 200 °C (20 min), 24.9 min.
1
4
168.6, 162.5 (d, JCF = 249.0 Hz), 137.4, 134.5, 131.8 (d, JCF
=
3
2
3.3 Hz), 131.0 (d, JCF = 8.0 Hz), 115.5 (d, JCF = 21.7 Hz), 51.9,
20.7, 13.7 ppm. IR (CHCl solution): ν = 2952, 1609 (C=O), 1600,
˜
3
1511, 1435, 1306, 1233, 1129, 1045, 836 cm–1. EIMS: m/z (%) =
208 (100) [M+], 176 (27), 148 (86), 133 (38). HR-EIMS; m/z: calcd.
for C12H13FO2: 208.0900; found 208.0899.
Partial Resolution of (؎)-Methyl 2-Methylene-3-phenylbutanoate
(3): A sample of (Ϯ)-3 (192 g, 0.92 mmol) was dissolved in THF/
water (3:1, 20 mL), treated with LiOH·H2O (138 mg, 3.29 mmol)
and stirred at ambient temperature (24 h) after which time the reac-
tion mixture was diluted with water (10 mL) and THF was removed
in vacuo. The aqueous phase was extracted with EtOAc in order
to remove any residual starting material and neutralised with 2 
HCl (50 mL). The product was extracted with EtOAc, the com-
bined organic layers were dried with Na2SO4 and solvent removed
in vacuo to give a quantitative yield of the derived acid (162.0 mg,
(E)-Methyl 2-(4-Chlorobenzylidene)butanoate (23): Prepared analo-
gously to 4 using 0.50 mmol of 13. Colourless oil (77 mg; 69%); Rf
1
= 0.42 (pentane/EtOAc, 9:1). H NMR (400 MHz, CDCl3): δH
=
7.58 (s, 1 H, =CH), 7.36 (d, J = 8.5 Hz, 2 H, Ar), 7.29 (d, J =
8.5 Hz, 2 H, Ar), 3.82 (s, 3 H, OMe), 2.52 (q, J = 7.5 Hz, 2 H,
CH2Me), 1.16 (t, J = 7.5 Hz, 3 H, CH2Me) ppm. 13C NMR
(100 MHz, CDCl3): δ = 168.5, 137.2, 135.3, 134.2, 134.2, 130.4,
128.7, 52.0, 20.8, 13.7 ppm. IR (CHCl solution): ν = 2902, 1708
˜
3
(C=O), 1243, 1130, 1014 cm–1. EIMS: m/z (%) = 226 (27) [M+,
37Cl], 224 (95) [M+, 35Cl], 193 (32), 165 (56), 139 (100). HRMS:
m/z: calcd. for C12H1335ClO2: 224.0604; found 224.0613.
1
0.92 mmol). H NMR (400 MHz, CDCl3): δ = 7.35–7.30 (m, 2 H,
Ar), 7.26–7.22 (m, 3 H, Ar), 6.48 (s, 1 H, =CH2α), 5.76 (s, 1 H,
=CH2β), 4.05 (q, J = 7.1 Hz, 1 H, CHMe), 1.47 (d, J = 7.1 Hz, 3
H, CHMe) ppm. The latter was dissolved in Et2O (5 mL) and neat
(R)-PhCHMeNH2 (120 µL, 0.95 mmol) added. The resulting white
amine salt was fractionated with CH2Cl2/Et2O. After acidification
(HCl, 2 ), re-extraction (Et2O), drying (Na2SO4) and re-esterifica-
tion (four fold excess TMSCHN2, Et2O, 30 min) the less soluble
amine salt afforded (R)-3 (36.5 mg, 0.19 mmol, 20.7% yield) with
[α]2D1 = –65.6 (c = 0.62, CHCl3) for an 86% ee sample (as deter-
mined by GC as above); ref.[6] [α]2D0 = –47.7 (c = 0.65, CHCl3) for
a 69% ee (R)-3 sample. Similar treatment of the more soluble amine
salt gave (S)-3 (52.2 mg, 0.27 mmol, 30% yield) with [α]2D1 = +51.5
(c = 0.75, CHCl3) for an 66% ee sample (as determined by GC as
above).
(E)-Methyl 2-(4-Methylbenzylidene)butanoate (24): Prepared analo-
gously to 4 using 0.50 mmol of 14. Colourless oil (78 mg; 76%); Rf
= 0.70 (pentane/EtOAc, 9:1). 1H NMR (400 MHz, CDCl3): δ =
7.79 (s, 1 H, =CH), 7.45 (d, J = 8.0 Hz, 2 H, Ar), 7.36 (d, J =
8.0 Hz, 2 H, Ar), 3.98 (s, 3 H, OMe), 2.73 (q, J = 7.4 Hz, 2 H,
CH2), 2.53 (s, 3 H, ArMe), 1.34 (t, J = 7.4 Hz, 3 H, CH2Me) ppm.
13C NMR (100 MHz, CDCl3): δ = 168.9, 138.6, 138.4, 133.8, 132.8,
129.2, 129.1, 51.8, 21.3, 20.8, 13.8 ppm. IR (CHCl solution): ν =
˜
3
2951, 1694 (C=O), 1511, 1435, 1312, 1133, 1045 cm–1. EIMS: m/z
(%) = 204 (100) [M+], 173 (21), 144 (57), 129 (50). HR-EIMS; m/z:
calcd. for C13H16O2: 204.1150; found 204.1147.
(E)-Methyl 2-(4-tert-Butylbenzylidene)butanoate (25): Prepared
analogously to 4 using 0.50 mmol of 15. Colourless oil (93 mg;
76%); Rf = 0.72 (pentane/EtOAc, 9:1). 1H NMR (400 MHz,
CDCl3): δ = 7.64 (s, 1 H, =CH), 7.43 (d, J = 8.4 Hz, 2 H, Ar), 7.35
(d, J = 8.4 Hz, 2 H, Ar), 3.82 (s, 3 H, OMe), 2.59 (q, J = 7.4 Hz,
2 H, CH2Me), 1.34 (s, 9 H, tBu), 1.20 (t, J = 7.4 Hz, 3 H, CH2Me)
ppm. 13C NMR (100 MHz, CDCl3): δ = 168.9, 151.6, 138.5, 133.9,
132.8, 129.2, 125.4, 51.8, 34.7, 31.2, 20.9, 13.8 ppm. IR (CHCl3
(E)-Ethyl 2-Benzylidenebutanoate[12] (20): Prepared analogously to
4 using 0.50 mmol of 10. Colourless oil (69 mg; 68%); Rf = 0.50
(pentane/EtOAc, 9:1). H NMR (400 MHz, CDCl3): δ = 7.65 (s, 1
H, =CH), 7.39–7.37 (m, 5 H, Ar), 4.28 (q, J = 7.1 Hz, 2 H,
CH2Me), 2.55 (q, J = 7.4 Hz, 2 H, OCH2Me), 1.36 (t, J = 7.1 Hz,
3 H, CH2Me), 1.18 (t, J = 7.4 Hz, 3 H, OCH2Me) ppm. 13C NMR
(100 MHz, CDCl3): δ = 168.4, 138.3, 135.9, 135.1, 129.2, 128.5,
1
128.3, 60.7, 20.8, 14.4, 13.9 ppm. IR (CHCl solution): ν = 2970,
˜
solution): ν = 2951, 1694 (C=O), 1626, 1510, 1435, 1314, 1269,
˜
3
1698 (C=O), 1455, 1369, 1311, 1128, 1046 cm–1. EIMS: m/z (%) =
204 (87) [M+], 158 (52), 131 (100), 129 (42), 115 (37), 91 (42). HR-
EIMS: m/z: calcd. for C13H16O: 204.1150; found 204.1163.
1192, 1132, 1045 cm–1. EIMS: m/z (%) = 246 (39) [M+], 231 (100).
HR-EIMS: m/z: for C16H22O2: 246.1620; found 246.1618.
(E)-Methyl 2-(2-Bromobenzylidene)butanoate (26): Prepared analo-
gously to 4 using 0.50 mmol of 16. Colourless oil determined to be
a 9:1 mix of α/γ products (101 mg; 75%); Rf = 0.76 (pentane/
EtOAc, 9:1). 1H NMR (270 MHz, CDCl3): δ = 7.83 (s, 1 H, =CH),
(E)-Methyl 2-(4-Methoxybenzylidene)butanoate (21): Prepared
analogously to 4 using 0.50 mmol of 11. Colourless oil (85 mg;
77%); Rf = 0.35 (pentane/EtOAc, 9:1). 1H NMR (400 MHz,
CDCl3): δ = 7.60 (s, 1 H, =CH), 7.37–7.34 (m, 2 H, Ar), 6.94–6.90 7.82–7.79 (m, 3 H, Ar), 3.85 (s, 3 H, OMe), 2.54 (q, J = 8.1 Hz, 2
(m, 2 H, Ar), 3.82 [s, 3 H, C(O)OMe], 3.80 (s, 3 H, ArOMe), 2.57 H, CH2Me), 1.09 (t, J = 8.1 Hz, 3 H, CH2Me) ppm. 13C NMR
(q, J = 7.4 Hz, 2 H, CH2Me), 1.18 (t, J = 7.4 Hz, 3 H, Me) ppm. (68 MHz, CDCl3): δ = 168.1, 137.9, 136.4, 136.2, 132.7, 129.9,
13C NMR (100 MHz, CDCl3): δ = 169.0, 159.7, 138.2, 132.5, 131.0, 129.5, 127.1, 123.9, 52.0, 20.9, 13.8 ppm. EIMS: m/z (%) = 270 (41)
902
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 898–903

Methylation of Baylis–Hillman-Derived Electrophiles
[M+, 81Br], 268 (38) [M+, 79Br], 238 (100). HR-EIMS: m/z: calcd.
[2]
[3]
[4]
V. E. Albrow, A. J. Blake, R. Fryatt, C. Wilson, S. Woodward,
Eur. J. Org. Chem. 2006, 2549.
K. Biswas, C. Börner, J. Gimeno, P. J. Goldsmith, D. Ramaz-
zotti, A. L. K. So, S. Woodward, Tetrahedron 2005, 61, 1433.
P. J. Goldsmith, S. Woodward, Angew. Chem. Int. Ed. 2005, 44,
2235.
for C12H1379BrO2: 268.0099; found 268.0115.
(E)-Methyl 2-(Naphthalen-1-ylmethylene)butanoate (27): Prepared
analogously to 4 using 0.50 mmol of 17. Colourless oil (88 mg;
73%); Rf = 0.68 (pentane/EtOAc, 5:1). 1H NMR (400 MHz,
CDCl3): δ = 8.15 (s, 1 H, =CH), 7.95–7.86 (m, 3 H, Ar), 7.54–7.47
(m, 4 H, Ar), 3.90 (s, 3 H, OMe), 2.43 (q, J = 7.4 Hz, 2 H, CH2Me),
1.10 (t, J = 7.4 Hz, 3 H, CH2Me) ppm. 13C NMR (100 MHz,
CDCl3): δ = 168.4, 137.3, 136.9, 133.4, 133.3, 131.5, 128.4, 126.3,
126.1, 125.9, 125.8, 125.2, 124.7, 51.9, 21.3, 14.2 ppm. IR (CHCl3
[5] Review: a) H. Kurosawa, S. Ogoshi, Bull. Chem. Soc. Jpn. 1998,
71, 973. Recent developments: b) M. Aresta, A. Dibenetto, E.
Quaranta, M. Lanfranchi, A. Tiripicchio, Organometallics
2000, 19, 4199; c) L. C. Silva, P. T. Gomes, L. F. Veiros, S. I.
Pascu, M. T. Duarte, S. Namorando, J. R. Ascenso, A. R. Dias,
Organometallics 2006, 25, 4391; d) A. Bottoni, G. P. Miscione,
M. A. Carvajal, J. J. Novoa, J. Organomet. Chem. 2006, 691,
4498. Histrorical cases: e) B. Bogdanovic, H. Bönnemann, G.
Wilke, Angew. Chem. Int. Ed. Engl. 1966, 5, 582; f) D. Walter,
G. Wilke, Angew. Chem. Int. Ed. Engl. 1966, 5, 897.
solution): ν = 2952, 2877, 1698 (C=O), 1435, 1315, 1132, 1048
˜
cm–1. EIMS: m/z (%) = 240 (76) [M+], 181 (100), 165 (92), 155 (70).
HR-EIMS: m/z: calcd. for C16H16O2: 240.1150; found 240.1143.
(E)-Methyl 2-(Naphthalen-2-ylmethylene)butanoate (28): Prepared
analogously to 4 using 0.50 mmol of 18. Colourless oil (52 mg;
43%); Rf = 0.36 (pentane/EtOAc, 9:1). 1H NMR (400 MHz,
CDCl3): δ = 7.86–7.78 (m, 5 H, Ar and = CH), 7.53–7.43 (m, 3 H,
Ar), 3.85 (s, 3 H, OMe), 2.64 (q, J = 7.4 Hz, 2 H, CH2Me), 1.23
(t, J = 7.4 Hz, 3 H, Me) ppm. 13C NMR (100 MHz, CDCl3): δ =
168.8, 138.6, 135.0, 133.3, 133.1, 132.9, 128.8, 128.3, 128.0, 127.8,
[6]
K. S. Yoo, C. P. Park, C. H. Yoon, S. Sakaguchi, J. O’Nieill,
K. W. Jung, Org. Lett. 2007, 9, 3933.
[7]
[8]
M. Wada, T. Wakabayashi, J. Organomet. Chem. 1975, 96, 301.
R. G. Parr, W. Yang, Density Functional Theory of Atoms and
Molecules, Oxford University Press, New York, 1989.
The calculation of transition states featuring these heavy bime-
tallic Ni-ferrophite ligands proved highly challenging.
See Supporting Information for preparation and data.
P. Dauben, Nouv. J. Chim. 1977, 1, 243.
[9]
127.6, 126.6, 126.4, 52.0, 20.9, 13.9 ppm. IR (CHCl solution): ν =
˜
3
[10]
[11]
[12]
2962, 1705 (C=O), 1600, 1436, 1285, 1242, 1125 cm–1. EIMS: m/z
(%) = 240 (84) [M+], 209 (8), 180 (48), 166 (24), 155 (100). HR-
EIMS: m/z: calcd. for C16H16O2: 240.1150; found 240.1141.
J. Castells, F. Lopez-Calahorra, Z. Yu, Tetrahedron 1994, 50,
13765.
[13]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morok-
uma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzew-
ski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V.
Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Re-
vision C.02, Gaussian, Inc., Wallingford CT, 2004.
W. J. Hehre, L. Radom, P. von R. Schleyer, J. A. Pople, Ab initio
Molecular Orbital Theory, Wiley, New York, 1986.
A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
Computational Studies: All DFT calculations were performed by
using the Gaussian03 software package,[13] with the B3LYP func-
tional. This functional includes a mixture of Hartree–Fock[14] ex-
change with DFT[8] exchange-correlation, given by Becke’s three-
parameter functional[15] with the Lee, Yang, and Parr correlation
functional, which includes both local and nonlocal terms.[16,17] The
geometries were optimized without any symmetry constraints using
the Stuttgart/Dresden ECP (sdd) basis set and associated ECP[18]
for Fe and Ni while a standard 6-311G** basis set was used for
the other elements. The molecules with all the substituents were
considered in the calculations.
Supporting Information (see also the footnote on the first page of
this article): Summary data on the DFT calculated syn and anti-I–
IV allyl complexes. Experimental procedures for the preparation of
the allylic halide electrophiles.
Acknowledgments
[14]
Financial support from European Union (project FP6-NMP3-C-
2003-505267 associated with Ligbank® and the COST D40 pro-
gramme) is acknowledged. One of us (A. N.) is grateful to GlaxoS-
mithKline for the provision of a studentship. M. J. C. and P. J. C.
thank Fundação do Ministério de Ciência e Tecnologia (FCT),
Programa Operacional Ciência e Inovaçao (POCI) and Fundo
Europeu de Desenvolvimento Regional (project POCI/QUI/58925/
2004).
[15]
[16]
B. Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys. Lett.
1989, 157, 200.
[17]
[18]
C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
a) U. Haeusermann, M. Dolg, H. Stoll, H. Preuss, Mol. Phys.
1993, 78, 1211; b) W. Kuechle, M. Dolg, H. Stoll, H. Preuss, J.
Chem. Phys. 1994, 100, 7535; c) T. Leininger, A. Nicklass, H.
Stoll, M. Dolg, P. Schwerdtfeger, J. Chem. Phys. 1996, 105,
1052.
Received: October 21, 2008
Published Online: January 12, 2009
[1] A. Novak, R. Fryatt, S. Woodward, C. R. Chim. 2007, 10, 206.
Eur. J. Org. Chem. 2009, 898–903
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
903