H. Thomas et al. / Tetrahedron Letters 50 (2009) 7374–7378
7375
thalene using bromine14 and successive alkylation with didodecyl
OC12H25
bromide was expected to give isomerically pure 2. Attempted
deprotonation of 1 with aqueous base resulted in extensive decom-
position presumably because this type of bromohydroxynaphtha-
lene behaved as a reactive bromoketone. When deprotonation
was performed with sodium hydride in dry NMP smooth alkylation
with dodecyl bromide afforded pure 2. Double Br/Li-ex-
change13,15,16 was accomplished by treating 2 with n-butyllithium
followed by quenching with DMF to afford 3 (Scheme 1). Heating
the reaction mixture from ꢁ78 °C to room temperature during
the metalation was necessary in order to obtain complete Br/Li-ex-
change. Keeping the reaction mixture at ꢁ78 °C gave almost solely
the mono-formylated product 6-bromo-1,5-bis(dodecyloxy)-2-
naphthaldehyde after quenching with DMF according to NMR,
since the mono-lithiated intermediate precipitated without react-
ing further with n-butyllithium.
Ti(0), 1,4-dioxane
W, 20min, 130 ºC
3
µ
OC12H25
2,6-PNV
n
Scheme 2.
130 °C for 30 min and 1 h, respectively. Extended reaction times
seemed to have no influence on the yield and polymer chain
length. Isolation of 2,6-PNV was performed by Soxhlet extraction
due to its limited solubility in all the solvents examined. However,
even upon prolonged extraction, non-extracted material remained
in the Soxhlet filter. The results from GPC-analysis of the Soxhlet
extracts versus a polystyrene standard of extracts of products ob-
tained after extended heating were similar and showed a polymer
with an average molecular weight (Mw) of approximately 65,000 g/
mol and with a polydispersity of Mw/Mn ꢀ 3.
3. Microwave-assisted McMurry polymerization
Initial polymerization experiments involved heating 3 in a
microwave oven at 110 °C for 10 min in dioxane. After this period,
3 was completely consumed and the reaction mixture was poured
into water. The polymer was filtered off and Soxhlet-extracted
with isopropanol in order to remove low weight material. Upon
Soxhlet extraction with CHCl3 a polymer with an average molecu-
lar weight of 16,000 g/mol was obtained. In order to obtain mate-
rial with higher molecular weight, the polymerization temperature
was increased and the reaction time extended. It was very impor-
tant that the dioxane was absolutely dry otherwise large amounts
of decomposed organic material would settle on the walls of the
reaction flask. Interestingly, when the low-valent titanium reagent
was prepared in dry media the reaction reached a maximum tem-
perature of approximately 130 °C, even when aiming for a much
higher maximum reaction temperature by subjecting the reaction
to microwave pulses of up to several hundred Watts. This temper-
ature was sufficient in all cases (Scheme 2).
The model compound, 1,5-bis(dodecyloxy)-2,6-distyrylnaph-
thalene (5) prepared for optical studies of an isolated 2,6-naphth-
ylene vinylene segment, was obtained by a Horner–Wadsworth–
Emmons (HWE)16,17 reaction between
3 and diethyl ben-
zylphosphonate (4) (Scheme 3).
The crystal structures of 3 and 5 were determined by single
crystal X-ray diffraction.18 The molecular inversion symmetry of
both molecules was utilized in their crystal packing. The packing
motifs of both compounds are comparable as they consist of layers
of mainly the aromatic moieties in between layers of alkoxy chains
(see Fig. 1). Despite this similarity, there are minor differences in
the crystal packing. The interactions between the aromatic system
in compound 3 consist purely of
ring systems), whereas the interactions between the aromatic sys-
tems in 5 are comprised of as well as perpendicular edge-to-
ꢂ ꢂ ꢂ
plane interactions (C–Hꢂ ꢂ ꢂ ). This introduces a small difference in
pꢂ ꢂ ꢂ
p interactions (overlapping
p
p
p
It should be noted that the McMurry reagent should preferably
be too diluted rather than too concentrated in order to avoid
decomposition of material. Essentially, the heat-sink effect of diox-
ane was a key element allowing for metal reactions by keeping the
temperature low, and it was necessary to have a uniform metal
slurry with metal particles as small as possible suspended in the
maximum amount of solvent in order to ensure optimal cooling.
Visual verification of polymerization in dilution experiments facil-
itated selection of the optimal concentrations of reactants for poly-
merization. The reaction mixture must become a green-coloured
finely dispersed slurry; if too concentrated, decomposition oc-
curred due to unfavourable settling of the metal-containing parti-
cles. In the aggregated solid phase the local temperature was
presumably very high and therefore caused extensive decomposi-
tion. A number of polymerization experiments with different reac-
tion times were performed at the highest possible practical
temperature of 130 °C. A maximum yield of 67% was obtained by
heating for 20 min at this temperature. The importance of the poly-
merization time-span was also investigated by polymerizing at
the direction of the alkoxy groups with respect to the plane of
the aromatic ring system. In compound 5 the alkoxy groups extend
parallel to the plane, and in 3 they extend about 25° out of the
plane, see Figure 2. Overall the packing of both compounds is very
efficient, as reflected by the similar crystal densities of 1.14 and
1.12 g cmꢁ3 for 3 and 5, respectively. The conjugated aromatic sys-
tem of 5 is slightly distorted, with an angle between the least-
OC12H25
PhCH2P(O)(OEt)2 4
3
-BuOK, THF
t
OC12H25
5 (77%)
Scheme 3.
OH
OC12H25
Br
OC12H25
Br
CHO
NaH, C12H25Br
NMP, rt
i. n-BuLi, THF, -78 ºC to rt
ii. DMF
Br
Br
OHC
iii. H2O
OH
OC12H25
OC12H25
1
2 (84%)
3 (37%)
Scheme 1.