A general route to fully terminally tert-butylated linear polyenes

Article abstract of DOI:10.1002/chem.201000019

Starting from the readily available α,β-unsaturated ketone, 3-tert-butyl4,4-dimethyl-2-pentenal, higher vinylogues, and fully terminally tert-butylated polyolefins with up to 13 consecutive conjugated double bonds have been prepared by either McMurry dimerization or Wittig chain-elongation routes. The highly unsaturated conjugated π systems, which show a remarkable stability, have been characterized by spectroscopic methods and, in many cases, by X-ray structural analysis. The yields are high enough to allow for thorough chemical reactivity studies.

Full text of DOI:10.1002/chem.201000019

FULL PAPER
DOI: 10.1002/chem.201000019
A General Route to Fully Terminally tert-Butylated Linear Polyenes**
Dagmar Klein,[a] Pinar KiliÅkiran,[a] Cornelia Mlynek,[a] Henning Hopf,*[a] Ina Dix,[a] and
Peter G. Jones[b]
Dedicated to Professor Pelayo Camps on the occasion of his 65th birthday
Abstract: Starting from the readily available a,b-unsaturated ketone, 3-tert-butyl-
4,4-dimethyl-2-pentenal, higher vinylogues, and fully terminally tert-butylated poly-
Keywords: McMurry coupling
carotenoids polyolefins
·
·
olefins with up to 13 consecutive conjugated double bonds have been prepared by
either McMurry dimerization or Wittig chain-elongation routes. The highly unsatu-
rated conjugated p systems, which show a remarkable stability, have been charac-
terized by spectroscopic methods and, in many cases, by X-ray structural analysis.
The yields are high enough to allow for thorough chemical reactivity studies.
·
tert-butyl stabilization
·
X-ray
diffraction
Introduction
In spite of this century-old interest, much information is
still lacking on oligo- and polymeric polyenes. This concerns
both their structures—in the gas phase (for example, as de-
termined by electron diffraction), solution (NMR spectros-
copy), and the solid state (X-ray structural analysis)—and
chemical behavior. It is, for example, surprising how rarely
the reactions so typical of simple alkenes and alkadienes
(electrophilic additions, cycloadditions, hydrogenation, poly-
merization, etc.) have been performed on the higher vinylo-
gues of these parent systems. The reason that our ignorance
grows so rapidly with the chain length is very simple: the
unsubstituted polyolefins are a class of compounds very dif-
ficult to handle under normal laboratory conditions. They
are thermally unstable and react readily with air. They poly-
merize and cross-link easily and often undergo cycloaddi-
tions with each other under mild conditions. Additionally,
polyolefins are configurationally labile, which makes the iso-
lation of pure diastereomers difficult, even when the most
modern chromatographic methods are applied. These prob-
lems are already noticeable for structurally simple mole-
cules, such as 1,3,5-hexatriene,[9] and quickly become insur-
mountable with its higher vinylogues. Consequently, the
Stabilization of polyolefins by substituent groups: Linearly
conjugated olefins constitute one of the central classes of or-
ganic chemical compounds and they have been studied ever
since organic chemistry became an exact natural science.[2,3]
Their importance ranges from 1,3-butadiene as a large-scale
starting material in industrial organic chemistry[4] to vitami-
n A, b-carotene, and numerous other retinoids[5] and carote-
noids,[6] which play a crucial role in many biochemical pro-
cesses, notably those in which light energy is converted into
chemical energy and information (vision).[7]
Furthermore, polyolefins have always played a pivotal
role in the development of theoretical concepts of organic
chemistry and in recent times have gained prominence as
the parent compounds of organic metals; (doped) polyacety-
lene is the most frequently investigated representative.[8]
[a] D. Klein, P. KiliÅkiran, C. Mlynek, Prof. H. Hopf, I. Dix
Institut fꢀr Organische Chemie, Technische Universitꢁt Braunschweig
Hagenring 30, 38106 Braunschweig (Germany)
Fax : (+49)531-391-5388
classical early studies on the HACHTNURGTNEUNG(CH=CH)nH hydrocarbons
1a performed by Sondheimer and co-workers[10] no longer
fulfill present-day quality standards.
E-mail: H.Hopf@tu-bs.de
[b] Prof. P. G. Jones
Institut fꢀr Anorganische und Analytische Chemie
Technische Universitꢁt Braunschweig, Postfach 3329d-38023
Braunschweig (Germany)
One often applied route to more stable polyolefins in-
volves the introduction of substituents. This approach was
first applied comprehensively by Nayler and Whiting[11] and
Bohlmann and Mannhardt[12] who prepared numerous poly-
olefins 1b with methyl groups at the terminal positions
(Scheme 1; n=longest chain length prepared).
[**] Highly hindered olefins and polyolefins, Part XIV; for Part XIII see
reference [1].
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000019.
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10507

Scheme 1. A selection of classical polyolefinic hydrocarbons.
Although these compounds show a pronounced increase
in stability relative to 1a, they are still reactive hydrocar-
bons, especially the higher members of the series. For exam-
ple, Bohlmannꢃs higher dimethyl derivatives 1b (n>3) all
showed decomposition and polymerization on heating,
which prevented the measurement of clearly defined melting
points.[12]
Another step toward the synthesis of longer polyenes
(molecular wires) was developed by Kuhn, who incorporat-
ed two phenyl substituents at the a- and w-positions and de-
scribed a general route to polyolefins 1c with up to 15 con-
secutive double bonds (Scheme 1).[13] In this case, the termi-
nal substituents are unsaturated and as a result they influ-
ence the electronic properties of the polyolefin chain (see
discussion on electronic spectra and Table 1 below), which is
a disadvantage if one wishes to study the effect of increasing
chain length/unsaturation on the spectroscopic and chemical
properties. However, incorporation of substituted-aryl
groups provides a convenient way to create polarization of
the unsaturated chain, if required.
cule 1,3-butadiene, but that the potential-energy curves of
the two compounds are rather similar; the transoid confor-
mation is the most stable. On the other hand, if the tert-
butyl moiety is introduced into the 2-position of 1,3-buta-
diene (3) a gauche conformation becomes the most stable, a
result borne out by experiment (dihedral angle=328, deter-
mined by electron diffraction of 3).[16] For the dienes 4 and
5, which have two tert-butyl groups in the 1,1- and 2,3-posi-
tions, respectively, the results are even more pronounced.
The calculated potential-energy curve for 4 favors the anti
conformation, which is more stable by approximately
8 kJmolꢀ1 relative to the gauche conformation. For 5 this
difference increases to approximately 55 kJmolꢀ1 and makes
an orthogonal structure much more favorable than a planar
one. In fact, the experimentally determined dihedral angle
amounts to 101.58 by gas-phase electron diffraction[16] and
96.68 by single-crystal X-ray analysis.[17] Extending these ob-
servations to the entire series of conjugated polyenes, we
can state that terminally tert-butylated polyenes should
prefer a planar structure, whereas introduction of an inter-
nal tert-butyl group should produce twisted polyolefins.
Eventually completely orthogonal structures can be ac-
cessed, in which the double bonds are oriented in alternat-
ing, perpendicular planes, thus inhibiting any overlap be-
tween the respective p electrons.
The most popular group used for the stabilization of reac-
tive organic molecules and intermediates is the tert-butyl
substituent, as demonstrated by the two valence isomers tet-
rakis(tert-butyl)tetrahedrane and -cyclobutadiene studied by
Maier et al.[14] When bonded to a polyolefinic backbone, the
tert-butyl moiety can exhibit vastly different effects, both
from a reactivity and structural perspective.
As far as chemical behavior is concerned, tert-butyl
groups will sterically protect the polyolefin chain from any
attacking reagent, regardless of where they are positioned.
They function as a “molecular fence” around the highly re-
active p-conjugated backbone. This effect, which depends
on the number of these bulky moieties, is particularly pro-
nounced for terminally bonded tert-butyl groups because the
a- and the w-positions are the most reactive centers of the
unsaturated chain.
The polyenes can adopt different conformations because
the molecules 1a–c also contain s-bonds around which rota-
tion is possible. The extent of rotation is controlled by the
placement of the tert-butyl substituents (terminal or inter-
nal) as discussed for the representative diene derivatives 2–
5.
Besides our own investigations on the preparation and
structural properties of various tert-butylated conjugated
dienes, extended oligo- and polyenes blocked by a single
tert-butyl substituent at each end were described, more or
less simultaneously, when we initiated our studies. These
have largely been prepared by the two strategies discussed
below.[18]
Schrock and Knoll[19] developed a route to the end-
capped polyenes 10 by subjecting triene 6 to a catalytic ring-
opening oligomerization with the tungsten catalyst 7. Cleav-
age of the more strained double bond of 6 led to metal-con-
taining oligomers 8 (Scheme 2).
Addition of aldehydes, such as pivaldehyde, induced
Wittig-type coupling with 8 to form 9, from which the 1,2-
bis(trifluoromethyl)benzene “protecting” group was cleaved
by short-time pyrolysis. Although this approach could be ex-
tended as far as the decapentaenes (in 10: n=13), it yielded
polydisperse mixtures that had to be separated by extensive
chromatography. As a result, the overall yield of the differ-
We have shown, by molecular-mechanics methods
(MM3)[15] and ab initio Hartree–Fock (HF) calculations, that
a tert-butyl group introduced into the 1-trans position (2) in-
creases the steric-strain energy relative to the parent mole-
10508
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
Scheme 2. Preparation of polyolefins by metathesis reactions.
ent polyenes was low and only limited structural data (NMR
spectra of representative hydrocarbons)[20] and scarcely any
chemical properties of the hydrocarbons 10 could be report-
ed. Nevertheless, the amounts of various end-capped poly-
olefins prepared by this route were sufficient to carry out
spectroscopic studies of radical cations generated radiolyti-
cally in Freon matrices, for example.[21]
scribed by Mꢀllen and co-workers.[23] The routes to the hex-
aene 16 and octaene 18 presented in Scheme 3 are typical
examples.
Stille coupling of vinyl iodide 11 with chlorodiene 12
yielded the extended polyunsaturated chloride 13, which
could easily be converted into stannane 14, then iodide 15.
Stille coupling of 14 and 15 provided 16 directly, whereas
the Pd-catalyzed cross-coupling of 15 with the bis-stannane
17 led to 18, with the most stable all-trans isomer being
formed in a final photoisomerization step.
Although this method gave the desired polyolefins in
pure form and in sufficient amounts to collect all of the
spectroscopic data and, in the case of 16, the X-ray struc-
ture, it must be mentioned that the starting material for
both 12 and 17, 1,4-dichloro-1,3-butadiene, is not a readily
available substrate. Like compound 6, it has to be prepared
from cyclooctatetraene, a hydrocarbon that has become ex-
pensive, although it is still available commercially.
In
a related ring-opening metathesis polymerization
(ROMP) route, which used a Schrock catalyst system,
Grubbs and Swager employed benzvalene as the starting
material. After the polymerization step, the polymer ob-
tained, which contained bicyclobutane units, was isomerized
by treatment with heavy-metal ions (Hg2+, Ag+) to a mix-
ture of polyenes (in the limiting case, polyacetylene). Com-
pared with the Schrock and Knoll route, this process avoids
the extrusion of a molecular fragment but, again, it is not a
suitable method for the preparation of distinct polyenes in
larger amounts.[22]
Stepwise protocols that combine high efficiency with com-
plete control of every step are preferable and were de-
The route to fully terminally tert-butylated polyolefins de-
veloped by us, and described herein in detail, avoids these
Scheme 3. Preparation of polyolefins bearing one terminal tert-butyl moiety.
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10509

H. Hopf et al.
pitfalls. It uses simple substrates, nontoxic reagents and in-
termediates, does not require sophisticated chromatograph-
ic-separation techniques, and allows the synthesis of a com-
prehensive series of fully conjugated polyolefins (n=even
or odd).
The Wittig reaction was performed at room temperature
in most cases and the yield was excellent up to eight consec-
utive C=C double bonds (27 f, 90%). From this compound
onwards the solubility of the aldehydes decreased dramati-
cally and it was no longer possible to perform the reaction
at room temperature. For the last two compounds in the
series (27g and h) reflux conditions were necessary, despite
which significant amounts of the substrate aldehyde re-
mained insoluble in the reaction mixture and the yields de-
creased. Workup was performed, in all cases, by simple
column chromatography on silica gel and the aldehydes
were stable as long as they were kept cold and under a ni-
trogen atmosphere. Nevertheless, the longest aldehydes pre-
pared (27g and h) displayed an increased tendency to form
insoluble products (by polymer-
Results and Discussion
Synthesis of fully terminally tert-butylated polyolefins 30
(n=odd): The starting point of our polyene synthesis is 3-
tert-butyl-4,4-dimethyl-2-pentenal (23), prepared in a four-
step synthesis from pivaloyl chloride (19), as shown in
Scheme 4.
ization) on standing. Clearly,
the iterative approach to these
aldehydes reached its limits at
this point.
The structures of 26 and
27a–h were determined from
their spectroscopic data and, in
the case of 27a, c, and d, by X-
ray crystal-structure analysis.
The structural data are dis-
cussed below and summarized
in the Experimental Section.
To prepare fully protected
oligo- and polyenes with an odd
number of double bonds from
polyunsaturated aldehydes 23,
Scheme 4. Preparation of the building block, tert-butyl-4,4-dimethyl-pent-2-enal (23).
Reaction of 19 with tert-butyl magnesium chloride yielded
di-tert-butyl ketone 20 in 68% yield. When 20 was reacted
with vinyl magnesium bromide, the expected allyl alcohol 21
was obtained in very good yield (90%), and subsequent
treatment with thionyl chloride in pyridine at 08C led to the
isolation of chloride 22 in 58% yield. At higher reaction
temperatures increasing amounts of the isomeric tertiary al-
lylic chloride were produced. To convert 22 into aldehyde
23, we employed the method of Hass and Bender (KOH,
isopropanol, 2-nitropropane)[24] and obtained 23 in 66%
yield. The pentenal can easily be prepared in 20 g batches
by this route, enough for the chain-extension reactions de-
scribed below. Surprisingly, compound 23 had not been re-
ported at the beginning of our studies; we have since de-
scribed its gas-phase conformation as determined by elec-
tron diffraction.[25] As expected, compound 23 exists exclu-
sively in the transoid conformation (shown in Scheme 4, full
spectroscopic data are given in the Experimental Section).
The p system of 23 was first elongated by Wittig reaction
with triphenyl phosphonium bromide 24 (Scheme 5). The vi-
nylogous aldehyde 26, obtained after acid-catalyzed hydroly-
sis of the initially formed acetal 25, was then subjected to
further chain-extension cycles with the same reagent until
this protocol approached its limits with the aldehyde 27 f.
Scheme 5. Homologation (vinylation) of aldehyde 23.
26, and 27a–d, the McMurry coupling reaction was used.
Various protocols have been proposed in the literature for
this reductive dimerization.[26–28] We first tested these on the
simple aldehyde 23 to find the optimum conditions for this
class of compounds. The titanium(0) species generated by
10510
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
zinc reduction of TiCl4 in pyridine (Lenoir conditions) gave
the best results. When the dimerization was carried out for
shorter reaction times an intermediate product could be iso-
lated, which was identified as a mixture of diastereomers of
the pinacol 28 (Scheme 6). When 28 was reintroduced to the
reduction conditions it was converted into the trienes (E)-
and (Z)-29. A 2:1 E/Z ratio of 29 was isolated directly when
a solution of 23 in THF was heated at reflux temperature
for longer reaction times.
The two diastereomers could be separated by gradient-
temperature high-vacuum sublimation (at 0.05 mbar the Z
and E isomers sublime at 90 and 1158C, respectively). Struc-
tural assignment followed from the spectroscopic data and
the single-crystal X-ray analyses (discussed below).
Under comparable conditions, the aldehydes 26 and 27a–
d were reductively dimerized to the hydrocarbons 30a–e.
The longest system, 30e, contains 13 consecutive double
bonds; however, the yield drops rapidly (increased lability
and poor solubility) and the purification of the material be-
comes increasingly difficult (separation of by-products be-
comes a problem) across the series. In the examples with a
shorter chain length (30a–d) the yields are acceptable to
good. The stability of these polyolefins will be addressed
below.
Synthesis of fully terminally tert-butylated polyolefins 30
(n=even): The route discussed above does not allow the
preparation of linear polyolefins with an even number of
double bonds. To prepare these, we decided to synthesize an
appropriate Wittig reagent and react it with the vinylogous
aldehydes 26 and 27.
Scheme 6. Fully terminally tert-butylated polyolefins by McMurry dimeri-
zation.
The required Wittig reagent
was obtained from the phos-
phonium salt 35, itself prepared
by reaction of allylic bromide
31 with triphenylphosphine. In-
termediate 31 was accessed by
reaction of alcohol 21 with
phosphorus tribromide in the
presence
of
pyridine
(Scheme 7).
Accompanying the bromina-
tion of 21, we noted the forma-
tion of small amounts of a qua-
ternary hydrocarbon, to which
we assigned structure 32 based
on spectroscopic analysis of the
material (see the Experimental
Section). To rationalize the gen-
eration of 32, we propose that
21 is first dehydrated to the ter-
tiary cation 33, followed by
methyl migration to give the
isomeric cation 34, then proton
loss to furnish diene 32. The
presence of two geminal tert-
butyl moieties in 33 is thought Scheme 7. Fully terminally tert-butylated polyolefins by Wittig reaction.
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10511

H. Hopf et al.
to induce migration to relieve steric strain.
The Wittig reaction with aldehydes 26, 27b, and 27d pro-
ceeded without difficulties, although the yields of polyenes
30 f–h were unsatisfactory. This is attributed, in part, to the
expected formation of Z isomers during the chain-elonga-
tion process and to loss of material during the isolation and
separation processes. In the case of 30h, only the 3-Z
isomer could be obtained in analytically pure form. Detailed
structures of these compounds are discussed below.
Stability of tert-butylated polyolefins 29 and 30: The poly-
enes described in this study are considerably more stable
than any alkylated derivatives described to date in the liter-
ature. In pure form, no change is observed up to tempera-
tures of approximately 508C, either in solution or in the
solid state, and when working with the polyenes no protec-
tion by an inert gas is necessary. From the triene (Z)-29
(m.p.=1368C) to the undecaene 30d (m.p.=2578C) melting
takes place without any visible change. Only for the tride-
caene 30e did we note decomposition (by polymerization),
which began at 2908C. The shorter homologues, with up to
five consecutive double bonds, can be sublimed at
0.05 mbar—a property not shared by Kuhnꢃs phenylpo-
lyenes.[29] When metal salts (FeCl3) or metal oxides (Fe2O3)
are added to solutions of the polyenes in THF or dichloro-
methane, an immediate color change to black is observed on
mild heating (358C), or at room temperature (for 7 or more
double bonds; nꢁ5).
Figure 1. Electronic spectra of the aldehydes 23 (c), 26 (a), and 27a
(d) (in acetonitrile).
Hydrocarbons of medium-chain length (hexaene 30g and
heptaene 30b) tend to include solvent molecules (dichloro-
methane, chloroform) in the solid state, shown by NMR
spectroscopy and X-ray crystallography (see below). The
solvent molecules, which are enclosed between the layers of
the double-bond chains, are held very tightly in the crystal
lattice and cannot be removed under high vacuum or by
heating. From eight double bonds onwards, reproducible el-
emental analyses could not be obtained for the polyolefins,
possibly because of incomplete combustion. A similar obser-
vation has been made by Mꢀllen et al. for bis-tert-butylpo-
lyenes.[23]
Figure 2. Electronic spectra of the polyenaldehydes 27b (c), c (a),
and d (b) (in acetonitrile).
The electronic spectra of selected tert-butyl-substituted alde-
hydes and polyolefins
UV/Vis spectra of polyenaldehydes 23, 26, and 27a–h: UV/
Vis spectra were determined for the series beginning with
one (23) and ending with ten consecutive C=C double
bonds (27h); representative examples are displayed in Fig-
ures 1 and 2.
As expected, the absorption maxima (lmax) shift to longer
wavelengths with increasing chain length; the color of the
aldehydes changes from colorless to deep red. The absorp-
tion bands are very broad and show no vibrational fine
structure. Bands from the n!p* transitions are hidden
under the p!p* transition bands. Figure 3, summarizing all
of the UV/Vis spectra measurements, shows that after ap-
proximately six double bonds the redshift trend begins to
Figure 3. Bathochromic shift of the absorption maximum of the polyenal-
dehydes 27 as a function of chain length.
decrease—a well-known phenomenon for polyenes—and al-
dehydes 27g and h (n=8 and 9, respectively) absorb at
nearly the same maximum. Very similar trends have been
reported for analogous polyenals that contain a methyl[30] or
phenyl substituent[31] at the w-position, and also for various
homologues of vitamin A aldehyde.[32] The lmax of the
phenyl derivatives are shifted to longer wavelengths by 10–
40 nm, depending on the chain length, whereas the methyl
compounds shift to shorter wavelengths.
10512
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
UV/Vis spectra of polyenes 29 and 30a–h: The UV/Vis spec-
tra of polyenes have been discussed extensively in the chem-
ical literature because of the importance of polyenes in bio-
logical chemistry, material science, and as model compounds
for the theoretical interpretation of the relationship between
color and constitution.[33] Unsurprisingly, the absorption
spectra of the all-trans compounds described herein are very
similar to those of conjugated polyenes with other terminal
substituents (see Table 1 below). The spectra display a pro-
file characterized by three to four bands, with the maxima at
the highest wavelengths being of greatest intensity. As in the
case of the polyenals, the lmax are shifted bathochromically
with increasing chain length. The distance between the
maxima amounts to (1500ꢂ150) cmꢀ1, independent of the
stabilizing end groups, as is typical for such polyolefinic mol-
ecules.[34] Figures 4 and 5 show typical absorption curves for
a selection of polyenes prepared in this work and Table 1
contains the data from the groups of Kuhn (a phenyl sub-
stituent at either end),[13,34] Sondheimer (unsubstituted),[10]
Bohlmann (a methyl substituent at each end),[12] and our tet-
rakis(tert-butyl) derivatives.
Figure 5. Electronic spectra of 30b (d, acetonitrile), 30d (c, di-
chloromethane), and 30e (a, dichloromethane).
cal feature apparently has only a weak influence on its elec-
tronic spectrum; qualitatively its UV spectrum correlates
with lower and higher vinylogues. In a thorough study on
the influence of solvents on the electronic spectra of several
of these polyenes, it has been shown that 30c is an excellent
probe for the empirical determination of the polarizability
of a wide variety of media.[35]
X-ray structural analysis of selected aldehydes (27a, c, d)
and polyolefins (29 and 30a–c, f, g, and cis-30h): The molec-
ular structures of the aldehydes 27a, c, and d (Figures 6a,
7a, and 8a, respectively) have several common features.
Torsion angles in the chains are as expected for all-trans sys-
tems, generally within (180ꢂ)108. Chain lengths, expressed
by the distance from the aldehyde carbon atom to the
carbon atom on which the tert-butyl groups are located, are
7.40, 12.26, 14.74, and 14.80 ꢄ for 27a, 27c, and the two in-
dependent forms of 27d, respectively. The disposition of the
tert-butyl groups is such that 27a and 27c have approximate
mirror symmetry (root mean square deviation (RMSD)=
0.10 and 0.13 ꢄ, respectively). The two forms of 27d depart
Figure 4. Electronic spectra of 29 (c), 30 f (a), and 30b (d) (in
acetonitrile).
Table 1. Absorption maxima of polyenes with different terminal substituents.
As can be seen in Table 1,
the mesomeric effect of the
phenyl substituent is considera-
bly stronger than the hypercon-
jugative influence of the alkyl
groups, even in the case of the
fully tert-butylated polyenes.
Kuhnꢃs compounds show fluo-
rescence in solution, whereas
this effect is not observed for
our tert-butyl-substituted hydro-
carbons. Although hydrocarbon
30h has one Z-configured
double bond, this stereochemi-
n
(in benzene) [nm]
(in isooctane) [nm]
(in hexane) [nm]
(in acetonitrile) [nm]
2
3
4
5
6
7
8
9
334
358
384
403
420
435
–
–
–
–
217
257
290
334
364
390
410
–
227
275
310
341
380
396
–
–
–
–
248
308
342
370
398
420
442
460
11
13
–
–
502 (in dichloromethane)
526 (in dichloromethane)
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10513

H. Hopf et al.
Figure 8. a) ORTEP plots of the two independent forms of 27d. Ellip-
soids represent 50% probability levels. b) Packing diagram of 27d. Hy-
drogen atoms are omitted. Selected stacks of molecules 1 and 2 are indi-
cated by the digits to the left of the stacks.
whereas all other angles are appreciably greater than 1208
Figure 6. a) ORTEP plot of 27a. Ellipsoids represent 30% probability
ꢀ
1
(122–1268). The final angle at (tBu)2C=CH CH is larger
levels. b) Packing diagram of 27a in the region zꢃ = . Hydrogen bonds of
4
ꢀ
again (131–1328), presumably for steric reasons. Steric
the form C H···O are indicated by dashed lines. Hydrogen atoms not in-
volved in hydrogen bonds and methyl carbon atoms are omitted.
crowding is indicated by intramolecular H···H contacts as
ꢀ
short as 1.85 ꢄ between the tert-butyl groups. The Me3C C
bonds are somewhat elongated at 1.55 ꢄ (relative to the
[37]
ꢀ
standard value of 1.52 ꢄ for R3C C), although the angles
Me3C-C-CMe3 are not especially wide (122–1238). Taking
27a as an example, the carbon atoms C9 (endo) and C13
(exo) lie in the mirror plane and, thus, short contacts (ap-
proximately 2 ꢄ) must exist between H6 and two of the hy-
drogen atoms at C9, and also between H5 and one hydrogen
ꢀ
atom at each of C14 and C15. The wide (tBu)2C=CH CH
angle may serve to prevent very short contacts from the tert-
butyl hydrogen atoms to H5. A Cambridge Database search
gave only three hits for this moiety and all had relevant
angles of 131–1338 (reference codes ADAHUJ,
LEWSUB10, and ZIFPAF).
The molecular packing of 27a (Figure 6b) is quite differ-
ent from that of the other two aldehydes (Figures 7b and
8b). As would be expected in space group Pbca, there is no
preferred chain direction; instead, the molecules associate
Figure 7. a) ORTEP plot of 27c. Ellipsoids represent 50% probability
levels. b) Packing diagram of 27c. Hydrogen atoms are omitted.
1
3
to form broad layers at zꢃ = , = , which involve weak hy-
4
4 ꢀ
from this symmetry because of a bowing of the chain (C1-
C7-C13=171 and 1728) and a nonideal torsion angle of 1538
drogen bonds of the form C H···O (non-normalized
H2···O=2.74, H4···O=2.60, H5···O=2.73 ꢄ; the first two in-
teractions constitute a bifurcate system). For 27b (Fig-
ure 7b), parallel packing of the longer chains is clearly more
ꢀ
for C10’-C11’-C12’-C13’ in form 2. The C C bond lengths al-
ternate in the polyene chains, between formal double and
single bonds of approximately 1.35 and 1.44 ꢄ, respectively
(scatter approximately 0.01 ꢄ; for individual values of these,
and other dimensions, see the Supporting Information). The
ꢀ
important than C H···O interactions because there are no
H···O distances shorter than 2.89 ꢄ. Each chain is surround-
ed by four others generated via various inversion centers,
with the shortest C···C distances between chains at approxi-
mately 3.55 ꢄ. Perhaps unexpectedly, the packing in 27c,
with longer chains still, is a hybrid of the other two types
(Figure 7b). The overall packing corresponds to the paral-
ꢀ
bond angles C C=O are 125–1268; a search of the Cam-
bridge Database[36] gave an average of 125.98 for 37 occur-
ꢀ
rences of the acyclic fragment trans-CH=CH CH=O.
Moving along the chain, away from the aldehyde function,
ꢀ
the next bond angle is approximately ideal (120–1228),
lel-chain type but, in addition, there are two weak C H···O
10514
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
interactions H17···O=2.74 and H7···O’ 2.68 ꢄ (not shown in
Figure 8b).
The triene (E)-29 (Figure 9a) displays crystallographic in-
version symmetry, although the actual symmetry is close to
2/m (RMSD=0.09 ꢄ); the isomer (Z)-29 (Figure 10a) has
Figure 10. a) ORTEP plot of (Z)-29. Ellipsoids represent 50% probability
1
levels b) Packing diagram of (Z)-29; layer at yꢃ =. Hydrogen atoms are
4
Figure 9. a) ORTEP plot of (E)-29. Ellipsoids represent 50% probability
omitted.
1
levels. b) Packing diagram of (E)-29; layer at xꢃ = . Hydrogen atoms are
4
omitted.
no crystallographic symmetry, but has approximate mm2
symmetry (RMSD=0.19 ꢄ). General features of molecular
dimensions, including distortions and H···H contacts, are
much as those described above. The triene chain of the Z
isomer has particularly wide angles at C3 and C4 (1268), but
there is little rotation about the central bond (torsion
angle=38). The packing of (E)-29 involves hexagonal layers
(Figure 9b), whereas the corrugated layers in (Z)-29 involve
a herringbone-type arrangement (Figure 10b).
The remaining all-trans polyenes 30a–c (n=odd), f, and g
(n=even) all display crystallographic inversion symmetry
(Figures 11a, 12a, 13a, 14, and 15a, respectively), although
the true symmetry is, in most cases, close to 2/m (RMSD=
0.13, 0.05, 0.05, and 0.09 ꢄ for 30a, b, f, and g, respectively).
The exception is the longest-chain vinylogue 30c, which is
distorted to a flattened-“S” shape (Figure 13b). The general
features of the molecular geometries of the polyene chains
are similar to those described above for the aldehydes,
except that the chain angles for 30c are marginally lower
(C4-C5-C6=1228).
Figure 11. a) ORTEP plot of 30a. Ellipsoids represent 30% probability
levels. b) Packing diagram of 30a, hydrogen atoms are omitted.
tures at first glance (see Figure 15b for 30g), but the mole-
cules in the vertical rows are not parallel. Instead, they have
angles of approximately 31 and 378 (for 30b and g, respec-
tively; calculated from the vectors from C1 to its symmetry
equivalent in the same molecule). This can clearly be seen
in the view parallel to the c axis for 30b (Figure 12b), which
also shows the voids where the disordered solvent molecules
are located.
The packing patterns of 30a–c, f, and g fall into two
groups, depending on the space group. The solvent-free
¯
structures 30a, c, and f all crystallize in P1 with Z=1; thus,
the chains are necessarily all parallel and the molecules
form layers (Figure 11b). The dichloromethane solvates 30b
and g crystallize in C2/c with Z=4; the packing, as viewed
¯
along the short b axis, appears to be similar to the P1 struc-
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10515

H. Hopf et al.
Figure 12. a) ORTEP plot of 30b. Ellipsoids represent 50% probability
levels. b) Packing diagram of 30b, hydrogen atoms are omitted.
Figure 15. a) ORTEP plot of 30g. Ellipsoids represent 50% probability
levels. b) Packing diagram of 30g, hydrogen atoms are omitted.
Figure 13. a) ORTEP plot of 30c. Ellipsoids represent 50% probability
levels. b: Side view of 30c (arbitrary radii, hydrogen atoms are omitted).
Figure 16. a) ORTEP plot of cis-30h. Ellipsoids represent 50% probabili-
ty levels. b) Packing diagram of cis-30h, hydrogen atoms are omitted.
1
1
Molecules with thick or thin bonds occupy the areas at yꢃ = or = , re-
2
4
spectively.
Figure 14. ORTEP plot of 30 f. Ellipsoids represent 50% probability
levels.
pret these data in view of the unresolved disorder. The
1
packing involves layers of molecules at intervals of = along
4
Interestingly, in the case of 30h we also isolated a cis
isomer from the Wittig reaction and obtained single crystals
suitable for X-ray analysis. Octaene cis-30h (Figure 16a),
again, shows broadly similar features to its all-trans conge-
ners (indeed, the chain angles are even wider; 1318 at C4
and 1278 at C6 and C8), but it would be unwise to overinter-
the b axis (Figure 16b).
10516
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
were neutralized with saturated aqueous sodium bicarbonate solution,
dried over anhydrous MgSO4, and the solvent was removed by rotary
evaporation. The residue was purified by column chromatography on
silica gel. Note: When the reaction time was reduced to 2 h varying
amounts of the diastereomer with the Z configuration at the C2=C3
double bond could be isolated. To ensure the formation of the most
stable, all-trans configuration of the polyenaldehydes the reaction was
always carried out for 6 h.
Conclusion
Starting from the readily available building block 23, we
have developed a general method for the preparation of oli-
goenes with up to 13 consecutive conjugated double bonds
stabilized at their terminal positions by tert-butyl groups.
Oligoenes with an odd number of double bonds can be pre-
pared by McMurry coupling of the appropriate polyenals;
the even-numbered double-bond derivatives by a Wittig ap-
proach. All of these compounds could be obtained in prepa-
rative amounts and, as a result, we can now begin to study
their chemical properties. Structurally, these compounds
prefer a planar arrangement, which allows optimal p-orbital
overlap. All of these compounds are achiral. We emphasize
this point because the introduction of the bulky tert-butyl
moiety to the interior of these hydrocarbons prevents full
conjugation and causes the formation of orthogonal poly-
enes. The resulting conformations possess sufficiently high
rotational barriers to make these inner-substituted oligoenes
chiral; these observations will be reported in a later publica-
tion of this series.
Compound 26: According to the general procedure above, reaction of 23
(3.0 g, 18 mmol) in THF (30 mL), 24 (19.15 g, 44 mmol) in THF (90 mL),
tBuOK (4.81 g, 43 mmol), and oxalic acid (15.76 g, 125 mmol) in water
(125 mL), followed by silica-gel chromatography (ethyl acetate/hexane=
1:10), afforded 26 as
a
colorless solid (3.00 g, 86%). M.p. ꢃ58C;
1H NMR (400 MHz, CDCl3): d=1.27 (s, 9H; 7-H), 1.44 (s, 9H; 9-H),
5.99–6.05 (dd, J=7.6, 14.7 Hz, 1H; 2-H), 6.27 (d, J=11.8 Hz, 1H; 4-H),
7.80–7.86 (dd, J=11.8, 14.7 Hz, 1H; 3-H), 9.62 ppm (d, J=8.1 Hz, 1H; 1-
H); 13C NMR (100 MHz, CDCl3): d=193.80 (d, C1), 169.88 (C5), 151.23
(d, C3), 131.09 (d, C2), 122.21 (d, C4), 39.82 and 38.63 (C6 and C8),
34.17 (q, C9), 31.31 ppm (q, C7); IR (film): n˜ =2960 (s), 1683 (vs), 1609
(s), 1394 and 1396 (m), 1145 (s), 978 cmꢀ1 (m); UV/Vis (acetonitrile): see
Figure 1; MS (EI, 70 eV): m/z (%): 194 (2) [M]+, 138 (34), 123 (42), 57
(100); elemental analysis calcd (%) for C13H22O (194.31): C 80.36, H
11.41; found: C 79.97, H 11.23.
Compound 27a: According to the general procedure above, reaction of
26 (2.46 g, 12.7 mmol) in THF (25 mL), 24 (13.65 g, 31.8 mmol) in THF
(60 mL), tBuOK (3.42 g, 30.5 mmol), and oxalic acid (11.34 g, 90.0 mmol)
in water (90 mL), followed by silica-gel chromatography (ethyl acetate/
hexane=1:10), afforded 27a as a yellow solid (2.45 g, 88%). M.p. 30–
1
318C; H NMR (400 MHz, CDCl3): d=1.25 (s, 9H; 9-H), 1.41 (s, 9H; 11-
Experimental Section
H), 6.11–6.17 (m, 1H; 2-H), 6.16 (d, J=11.7 Hz, 1H; 6-H), 6.28–6.35 (dd,
J=11.2, 14.4 Hz, 1H; 4-H), 7.19–7.26 (dd, J=11.2, 14.9 Hz, 1H; 3-H),
7.31–7.38 (dd, J=11.8, 14.4 Hz, 1H; 5-H), 9.56 ppm (d, J=7.9 Hz, 1H; 1-
H); 13C NMR (100 MHz, CDCl3): d=193.43 (d, C1), 164.93 (C7), 152.62
(d, C3), 142.12 (d, C5), 130.30 (d, C2), 129.18 (d, C4), 123.39 (d, C6),
39.44 and 38.37 (C8 and C10), 34.00 (q, C11), 31.49 ppm (q, C9); IR
(film): n˜ =2956 (m), 1680 (vs), 1668 (vs), 1594 (vs), 1390 and 1368 (m),
1120 cmꢀ1 (s); UV/Vis (acetonitrile): see Figure 1; MS (EI, 70 eV): m/z
(%): 220 (2) [M]+, 164 (33), 136 (17), 69 (13), 57 (100); elemental analy-
sis calcd (%) for C13H22O (220.34): C 81.83, H 10.98; found: C 82.09, H
10.79.
General remarks: TLC: Commercial TLC-plates “Polygram Sil G/UV254
”
by Macherey, Nagel & Co. (Dꢀren). Column chromatography: Kieselgel
60 (70–230 mesh) by Merck (Darmstadt). M.p.: Bꢀchi 530 melting-point
apparatus, uncorrected. 1H and 13C NMR spectroscopy: referenced to in-
ternal tetramethylsilane (dH =0.00 ppm) or CDCl3 (dC =77.05 ppm), per-
formed on Bruker AC-200 or WM-400 spectrometers. IR spectroscopy:
recorded on a Nicolet 320 FTIR spectrometer. UV/Vis spectroscopy: re-
corded on Beckman UV 5230 and HP 8452A Diode Array spectrometers.
EIMS: recorded on a Finnigan MAT 8430 spectrometer (70 eV). The fol-
lowing compounds were prepared according to literature methods: 20,[38]
21,[39] 22,[39] and 24.[40]
Compound 27b: According to the general procedure above, reaction of
27a (1.15 g, 5.23 mmol) in THF (10 mL), 24 (5.60 g, 13.0 mmol) in THF
(25 mL), tBuOK (1.40 g, 12.5 mmol), and oxalic acid (4.61 g, 36.0 mmol)
in water (36 mL), followed by silica-gel chromatography (ethyl acetate/
hexane=1:10), afforded 27b as a yellow solid (1.17 g, 91%). M.p. 43–
458C; 1H NMR (400 MHz, CDCl3): d=1.24 (s, 9H; 11-H), 1.39 (s, 9H;
13-H), 6.11–6.17 (m, 2H; 2-H, 8-H), 6.17–6.23 (dd, J=10.8, 14.0 Hz, 1H;
6-H), 6.39–6.45 (dd, J=11.4, 14.8 Hz, 1H; 4-H), 6.75–6.82 (dd, J=10.8,
14.5 Hz, 1H; 5-H), 7.09–7.15 (dd, J=11.8, 14.3 Hz, 1H; 7-H), 7.11–7.27
(dd, J=11.2, 15.1 Hz, 1H; 3-H), 9.55 ppm (d, J=8.0 Hz, 1H; 1-H);
13C NMR (100 MHz, CDCl3): d=193.40 (d, C1), 162.25 (d, C9), 152.08
and 143.48 (C3, C5, C7), 130.86 (d, C6), 130.50 and 123.73 (2ꢅd, C2, C8),
128.78 (d, C4), 39.33 and 38.27 (C10, C12), 33.96 (q, C13), 31.65 ppm (q,
C11); IR (KBr): n˜ =2958 (m), 1677 (vs), 1580 (vs), 1192 cmꢀ1 (vs); UV/
Vis (acetonitrile): see Figure 2; MS (EI, 70 eV): m/z (%): 246 (32) [M]+,
189 (100), 133 (28), 57 (76); HRMS: m/z: calcd for C17H26O: 246.198
[M]+; found: 246.198.
Compound 23: Freshly distilled 2-nitropropane (72.7 g, 73 mL, 0.82 mol)
was added to a solution of potassium hydroxide (30.86 g, 0.55 mol) in
water (70 mL) and isopropanol (230 mL) at RT.[24] The mixture was
stirred for 30 min and allyl chloride 22 (38.00 g, 0.20 mol) in isopropanol
(80 mL) was added dropwise at RT. The reaction mixture was heated at
reflux temperature for 6.5 h, cooled to RT, and hydrolyzed with ice
water. The product mixture was extracted with diethyl ether, then the or-
ganic phases were combined and dried over anhydrous Na2SO4. The sol-
vent was removed in vacuo and the resultant oil was distilled at 658C/
0.5 mbar to afford 23 as a colorless liquid (22.50 g, 66%). 1H NMR
(400 MHz, CDCl3): d=1.28 (s, 9H; tBu), 1.49 (s, 9H; tBu), 5.96 (d, J=
7.4 Hz, 1H; 2-H), 10.50 ppm (d, J=7.4 Hz, 1H; 1-H); 13C NMR
(100 MHz, CDCl3): d=193.76 (d, C1), 177.97 (C3), 126.62 (d, C2), 39.69
and 39.23 (C4, C6), 35.31, 30.96 ppm (q, CH3); IR (film): n˜ =2963 (vs),
1733 (vs), 1393 and 1368 (m), 1152 cmꢀ1 (m); UV/Vis (acetonitrile): see
Figure 1; MS (EI, 70 eV): m/z (%): 167 (38) [M+ꢀH], 127 (20), 57 (100);
elemental analysis calcd (%) for C11H20O (168.27): C 78.51, H 11.98;
found: C 78.12, H 12.03.
Compound 27c: According to the general procedure above, reaction of
27b (0.75 g, 2.85 mmol) in THF (5 mL), 24 (3.05 g, 7.11 mmol) in THF
(15 mL), tBuOK (0.77 g, 6.83 mmol), and oxalic acid (2.52 g, 20.0 mmol)
in water (20 mL), followed by silica-gel chromatography (ether/petrole-
um ether=3:10), afforded 27c as a yellow solid (0.75 g, 96%). M.p.
898C; 1H NMR (400 MHz, CDCl3): d=1.24 (s, 9H; 13-H), 1.39 (s, 9H;
15-H), 6.11 (d, J=11.8 Hz, 1H; 10-H), 6.11–6.21 (m, 2H; 2-H, 8-H),
6.29–6.35 (dd, J=11.2, 14.7 Hz, 1H; 6-H), 6.41–6.47 (dd, J=11.3,
14.7 Hz, 1H; 4-H), 6.57–6.64 (dd, J=11.1, 14.7 Hz, 1H; 7-H), 6.69–6.75
(dd, J=11.2, 14.6 Hz, 1H; 5-H), 6.98–7.04 (dd, J=11.8, 14.3 Hz, 1H; 9-
H), 7.11–7.18 (dd, J=11.3, 15.1 Hz, 1H; 3-H), 9.56 ppm (d, J=8.0 Hz,
General procedure for the preparation of aldehydes 26 and 27: Under a
nitrogen atmosphere, tBuOK (2.6 equiv) was added to a stirred suspen-
sion of phosphonium salt 24 (2.5 equiv) in anhydrous THF. An immedi-
ate color change of the mixture from colorless to yellow was observed.
After stirring for 30 min, a solution of the respective aldehyde (1 equiv)
in THF was added dropwise and the stirring was continued at RT for 6 h.
Aqueous oxalic acid (1 m, 7 equiv) was added and the mixture was left to
stir overnight. After the phases had separated, the aqueous layer was ex-
tracted with diethyl ether several times. The combined organic phases
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10517

H. Hopf et al.
1H; 1-H); 13C NMR (100 MHz, CDCl3): d=193.47 (d, C1), 160.75 (C11),
152.01 (d, C3), 143.01 (d, C5), 139.68 (d, C7), 136.35 (d, C9), 131.56 and
130.59 (d, C2, C6, C8), 129.29 (d, C4), 123.91 (d, C10), 39.26 and 38.19
(C12, C14), 33.92 and 31.73 (2ꢅq, C13, C15); IR (KBr): n˜ =2956 (m),
1673 (vs), 1568 (vs), 1390 and 1369 (m), 1151 (vs), 1012 cmꢀ1 (vs); UV/
Vis (acetonitrile): see Figure 2; MS (EI, 70 eV): m/z (%): 272 (40) [M]+,
215 (100), 188 (24), 159 (30), 57 (38); elemental analysis calcd (%) for
C19H28O (272.43): C 82.04, H 10.73; found: C 82.50, H 10.47.
Compound 27g: Under
a
nitrogen atmosphere, tBuOK (0.140 g,
1.23 mmol) was added to a stirred suspension of phosphonium salt 24
(0.550 g, 1.29 mmol) in anhydrous THF (30 mL). An immediate color
change of the mixture from colorless to yellow was observed. After stir-
ring for 30 min, aldehyde 27 f (0.180 g, 0.514 mmol) in THF (20 mL) was
added dropwise and the reaction mixture was heated at reflux tempera-
ture for 6 h. After cooling to RT, aqueous oxalic acid solution (1 m,
7 equiv) was added and the mixture was stirred overnight. After the
phases had separated the aqueous layer was extracted with diethyl ether.
The organic phases were combined and neutralized with saturated aque-
ous sodium bicarbonate solution. After removal of the solvent by rotary
evaporation, the residue was purified by column chromatography on
silica gel (ethyl acetate) to afford 27g as dark-red crystals (0.116 g,
60%), poorly soluble in most common solvents. 1H NMR (400 MHz,
CDCl3): d=1.22 and 1.32 (2ꢅs, 2ꢅ9H; 21-H, 23-H), 6.05–6.51 (m, 14H;
2-H, 4-H, 6-H, 7-H, 8-H, 9-H, 10-H, 11-H, 12-H, 13-H, 14-H, 15-H, 16-H,
18-H), 6.66–6.95 (m, 2H; 5-H, 17-H), 7.13–7.26 (dd, J=15.0, 11.1 Hz,
1H; 3-H), 9.58 ppm (d, J=7.9 Hz, 1H; 1-H); IR (KBr): n˜ =2956 (m),
1669 (s), 1612 (w), 1523 (s), 1463 (w), 1392 (w), 1132 (m), 1010 cmꢀ1 (vs);
UV/Vis (chloroform): lmax (loge)=456 (4.87), 472 nm (4.84, sh). The al-
dehyde could not be characterized fully; in the NMR solvent (CDCl3) a
fast color change from deep red to green–brown was noted, accompanied
by a rapid loss of the aldehyde proton signal. A mass spectrum could not
be obtained because of polymerization of the compound. The aldehyde
group is at least partially responsible for the high reactivity of 27g,
shown by the reaction of freshly prepared material with malononitrile,
which led to the corresponding dinitrile (characterized spectroscopical-
ly[42]).
Compound 27d: According to the general procedure above, reaction of
27c (0.53 g, 1.95 mmol) in THF (4 mL), 24 (2.10 g, 4.87 mmol) in THF
(10 mL), tBuOK (0.52 g, 4.67 mmol), and oxalic acid (1.72 g, 13.6 mmol)
in water (14 mL), followed by silica-gel chromatography (ethyl acetate/
hexane=1:6), afforded 27d as an orange-red solid (0.54 g, 93%). M.p.
1278C; 1H NMR (400 MHz, CDCl3): d=1.23 (s, 9H; 15-H), 1.38 (s, 9H;
17-H), 6.10 (d, J=11.9 Hz, 1H; 12-H), 6.13–6.19 (m, 2H; 2-H, 10-H),
6.27–6.33 (dd, J=11.2, 14.6 Hz, 1H; 8-H), 6.37 (dd, J=11.3, 14.6 Hz, 1H;
6-H), 6.40–6.47 (dd, J=11.3, 14.6 Hz, 1H; 4-H), 6.56 (m, 2H; 7-H, 9-H),
6.69–6.75 (dd, J=11.2, 14.6 Hz, 1H; 5-H), 6.92–6.98 (dd, J=11.9,
14.2 Hz, 1H; 11-H), 7.10–7.16 (dd, J=11.3, 15.1 Hz, 1H; 3-H), 9.56 ppm
(d, J=7.9 Hz, 1H; 1-H); 13C NMR (100 MHz, CDCl3): d=193.26 (C1),
159.70 (C15), 151.72 (d, C3), 142.81 (d, C5), 139.11 and 137.39 (2ꢅd, C7,
C9), 135.09 (d, C11), 131.82 and 130.96 (2ꢅd, C2, C10), 131.25 (d, C8),
130.96 (d, C6), 128.38 (d, C4), 132.92 (d, C12), 39.10 and 38.04 (C14,
C16), 33.79 (q, C17), 31.68 ppm (q, C15); IR (KBr): n˜ =2957 (m), 1668
(vs), 1554 (vs), 1391 and 1367 (s), 1150 (vs), 1012 (vs), 1002 cmꢀ1 (vs);
UV/Vis (acetonitrile): see Figure 2; MS (EI, 70 eV): m/z (%): 298 (60)
[M]+, 241 (100), 214 (20), 57 (54); HRMS: m/z: calcd for C21H30O:
298.230 [M]+; found: 298.229.
Compound 27h: According to the procedure described for 27g, reaction
of phosphonium salt 24 (0.199 g, 0.465 mmol) in THF (20 mL), 27g
(0.070 g, 0.186 mmol) in THF (35 mL), and tBuOK (0.050 g, 0.464 mmol)
afforded the aldehyde 27h as a dark-red solid (0.70 g, 53%). Compound
27h is extremely insoluble in most organic solvents and during its (in-
complete) spectroscopic characterization we noted extensive polymeri-
Compound 27e: According to the general procedure above, reaction of
27d (0.29 g, 0.97 mmol) in THF (7 mL), 24 (1.04 g, 2.43 mmol) in THF
(15 mL), tBuOK (0.26 g, 2.34 mmol), and oxalic acid (0.86 g, 6.81 mmol)
in water (7 mL), followed by silica-gel chromatography (dichlorome-
thane), afforded 27e as a red solid (0.30 g, 95%). M.p. 140–1418C;
1H NMR (400 MHz, CDCl3): d=1.23 and 1.37 (2ꢅs, 2ꢅ9H; 17-H, 19-H),
6.08 (d, J=11.7 Hz, 1H; 14-H), 6.12–6.19 (m, 2H; 12-H, 2-H), 6.25–6.37
(m, 2H; 10-H, 8-H), 6.41–6.57 (m, 3H; 4-H, 6-H, 7-H), 6.53–6.57 (m,
2H; 9-H, 11-H), 6.72 (dd, J=14.7, 11.3 Hz, 1H; 5-H), 6.91 (dd, J=14.1,
12.0 Hz, 1H; 13-H), 7.14 (dd, J=15.0, 11.3 Hz, 1H; 3-H), 9.56 ppm (d,
J=7.9 Hz, 1H; 1-H); 13C NMR (100 MHz, CDCl3): d=193.38 (d, C1),
159.24 (C15), 151.81 (d, C3), 142.80 (d, C5), 139.11, 137.00, 136.20,
134.37, 132.04, 131.77, 131.68, 131.20, 130.62, 129.54 (d, C2, C4, C6, C7,
C8, C9, C10, C11, C12, C13), 123.94 (d, C14), 39.07 and 38.01 (2ꢅs; C16,
C18), 33.77 and 31.70 ppm (2ꢅq, C17, C19); IR (KBr): n˜ =2960 (m),
1957 (m), 1667 (s), 1613 (m), 1542 (s), 1482 (w), 1392 (w), 1114 (s), 1105
(s), 1007 cmꢀ1 (vs); UV/Vis (chloroform): lmax (loge)=422 nm (4.80); MS
(EI, 70 eV): m/z (%): 324 (100) [M]+, 289 (35), 267 (88), 240 (20), 162
(100), 147 (42), 131 (58), 105 (44), 91 (100), 77 (66), 72 (66), 57 (79). The
elemental analysis of this aldehyde was unsatisfactory, possibly because
of the increased reactivity of 27e.
1
zation. H NMR (400 MHz, CDCl3): d=1.23 and 1.37 (2ꢅs, 2ꢅ9H; 23-H,
25-H), 6.10–6.95 (m, 16H; 2-H, 4-H, 6-H, 7-H, 8-H, 9-H, 10-H, 11-H, 12-
H, 13-H, 14-H, 15-H, 16-H, 17-H, 18-H, 19-H, 20-H), 7.11–7.23 (dd, 1H;
3-H), 9.61 ppm (d, J=7.9 Hz, 1H; 1-H); UV/Vis (chloroform): lmax
(loge)=468 nm (4.81). The nature of the compound was confirmed by
Knoevenagel condensation with malononitrile, which gave the expected
dinitrile in 30% yield. This derivative is much more stable than the start-
ing material 27h and could be characterized fully[42]
.
McMurry reaction of aldehydes 23, 26, and 27a–d
Compound 29: Under a nitrogen atmosphere at 08C, zinc dust (4.67 g,
71.38 mmol) and anhydrous pyridine (2.72 g, 2.67 mL, 34.5 mmol) were
added to a solution of TiCl4 (6.77 g, 3.92 mL, 35.69 mmol) in anhydrous
THF (90 mL). The mixture was stirred for 30 min at 08C before a solu-
tion of aldehyde 23 (1.00 g, 5.95 mmol) in THF (50 mL) was added. The
reaction mixture was stirred for 1.5 h at RT and subsequently heated at
reflux temperature for 1.5 h. Hydrolysis with ice water, diethyl ether, and
half-concentrated hydrochloric acid gave a clear aqueous phase from
which the organic phase was separated. After extraction of the aqueous
phase with diethyl ether (2ꢅ), the organic phases were combined, neu-
tralized with saturated aqueous sodium bicarbonate solution, and dried
over anhydrous Na2SO4. The solvent was removed in vacuo and the oily
residue was purified by column chromatography on silica gel (pentane).
Thermal gradient sublimation at 0.05 mbar provided two fractions: Frac-
tion 1: (Z)-29 (0.21 g, 23%); fraction 2: (E)-29 (0.42 g 47%). Rinsing the
chromatography column with dichloromethane provided a small amount
of diol 28 as a mixture of diastereomers.
Compound (Z)-29: M.p. 1368C; 1H NMR (400 MHz, CDCl3): d=1.25
and 1.38 (2ꢅs, 2ꢅ18H; 1-H, 10-H, 12-H, 14-H), 6.50–6.55 ppm (AA’XX’,
4H; 4-H, 5-H, 6-H, 7-H); 13C NMR (100 MHz, CDCl3): d=157.11 (C3,
C8), 126.22 and 117.92 (d, C4, C5, C6, C7), 39.48 and 37.67 (C2, C9, C11,
C13), 33.54 and 31.74 ppm (q, C1, C10, C12, C14); IR (KBr): n˜ =2955
(vs), 1625 (w), 1387 and 1364 cmꢀ1 (m); UV/Vis (acetonitrile): lmax
(loge)=306 (4.38, sh), 294 (4.56), 284 nm (4.50); MS (EI, 70 eV): m/z
(%): 304 (36) [M]+, 247 (10), 191 (64), 57 (100).
Compound 27 f: According to the general procedure above, reaction of
27e (0.19 g, 0.58 mmol) in THF (7 mL), 24 (0.63 g, 1.47 mmol) in THF
(15 mL), tBuOK (0.16 g, 1.41 mmol), and oxalic acid (0.52 g, 4.10 mmol)
in water (4 mL), followed by silica-gel chromatography (dichlorome-
thane), afforded 27 f as a red solid (0.18 g, 90%). M.p. 141–1438C;
1H NMR (400 MHz, CDCl3): d=1.23 and 1.37 (2ꢅs, 2ꢅ9H; 19-H, 21-H),
6.08 (d, J=11.8 Hz, 1H; 16-H), 6.12–6.18 (m, 2H; 12-H, 14-H), 6.25–6.57
(m, 10H; 2-H, 4-H, 6-H, 7-H, 8-H, 9-H, 10-H, 11-H, 12-H, 13-H), 6.72
(dd, J=11.2, 14.6 Hz, 1H; 5-H), 6.89 (dd, J=11.6, 13.0 Hz, 1H; 15-H),
7.14 (dd, J=11.3, 15.0 Hz, 1H; 3-H), 9.56 ppm (d, J=7.9 Hz, 1H; 1-H);
13C NMR (100 MHz, CDCl3): d=193.37 (d, C1), 158.78 (C17), 151.76 (d,
C3), 142.76 (d, C5), 139.05 (d, C15), 136.96, 135.46, 133.94, 132.19,
132.16, 132.02, 131.91, 131.37, 130.66, 129.62 (d, C2, C4, C6, C7, C8, C9,
C10, C11, C12, C13, C14), 123.98 (d, C16), 39.04 and 37.98 (2ꢅs; C18,
C20), 33.75 and 31.70 ppm (2ꢅq, C19, C21); IR (KBr): n˜ =2957 (m),
1668 (s), 1600 (s), 1367 (w), 1139 (s), 1109 (vs), 1104 cmꢀ1 (m); UV/Vis
(chloroform): lmax (loge)=444 nm (4.83). During spectroscopic analysis
the compound polymerized.
10518
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
Compound (E)-29: M.p. 1858C; 1H NMR (400 MHz, CDCl3): d=1.23 (s,
18H; 1-H, 10-H), 1.38 (s, 18H; 12-H, 14-H), 6.08–6.11 and 6.63–6.65 ppm
(AA’XX, 4H; 4-H, 5-H, 6-H, 7-H); 13C NMR (100 MHz, CDCl3): d=
155.93 (C3), 131.72 (d, C5, C6), 124.15 (d, C4, C7), 38.81 (C2, C9), 37.72
(C11, C13), 33.74 (q, C12, C14), 31.85 ppm (q, C1, C10); IR (KBr): n˜ =
2956 (vs), 1598 (w), 1386, 1366 and 1355 (m-s), 984 cmꢀ1 (m); UV/Vis
(acetonitrile): see Figure 4; MS (EI, 70 eV): m/z (%): 304 (24) [M]+, 247
(10), 191 (56), 57 (100); HRMS: m/z: calcd for C22H40: 304.313 [M]+;
found: 304.312; elemental analysis calcd (%) for C22H40 (304.54): C 86.76,
H 13.24; found: C 86.54, H 13.32.
Compound 28 (mixture of diastereomers): M.p. 1088C; 1H NMR
(400 MHz, CDCl3): d=1.19 (s, 18H; 1-H, 10-H), 1.34 (s, 18H; 12-H, 14-
H), 1.99 (s, 2H; 5-H, 6-H), 4.86 and 5.43 ppm (AA’XX’, 4H; 4-H, 5-H, 6-
H, 7-H); 13C NMR (100 MHz, CDCl3): d=158.92 (C3, C8), 122.04 (d, C4,
C7), 71.14 (d, C5, C6), 39.06 (C2, C9), 37.45 (C11, C13), 34.12 (q, C12,
C14), 31.62 ppm (C1, C10); IR (KBr): n˜ =3465 (br, OH), 2957 (vs), 1624
(m), 1392, 1369 and 1363 (m), 1217 (m), 1011 cmꢀ1 (m); UV/Vis (acetoni-
trile): lmax (loge)=198 nm (4.29); MS (EI, 70 eV): m/z (%): 339 (12)
[M+H]+, 338 (50) [M]+, 321 (100), 247 (84); elemental analysis calcd
(%) for C22H42O2 (338.56): C 78.03, H 12.51; found: C 78.29, H 12.75.
C17), 124.04 (d, C4, C19), 39.98 and 37.92 (C2, C21, C23, C25), 33.72 (q,
C1, C12), 31.74 ppm (q, C24, C26); IR (KBr): n˜ =2954 (m), 1215 (m),
1005 cmꢀ1 (vs); UV/Vis (acetonitrile): lmax (loge)=460 (4.97), 432 (4.98),
408 (4.79), 390 (4.48, sh), 372 (4.15, sh), 330 (3.76), 266 nm (4.02); MS
(EI, 70 eV): m/z (%): 460 (100) [M]+, 403 (14), 346 (22), 57 (76);
HRMS: m/z: calcd for C34H52: 460.407 [M]+; found: 460.406.
Compound 30d: According to the procedure described for 29, reaction
of TiCl4 (0.79 g, 0.46 mL, 4.16 mmol) in THF (40 mL), zinc dust (0.54 g,
8.32 mmol), pyridine (0.29 g, 0.28 mL, 3.67 mmol), and aldehyde 27c
(0.190 g, 0.71 mmol) in THF (20 mL), followed by chromatographic
workup on silica gel (pentane/dichloromethane) and recrystallization (di-
chloromethane/methanol), yielded 30d as
a dark-red solid (0.080 g,
44%). M.p. 2578C; 1H NMR (400 MHz, CDCl3): d=1.17 (s, 18H; 1-H,
26-H), 1.31 (s, 18H; 28-H, 30-H), 5.90 (d, J=11.7 Hz, 2H; 4-H, 23-H),
6.01–6.07 and 6.21 (m, 16H; 6-H, 7-H, 8-H, 9-H, 10-H, 11-H, 12-H, 13-H,
14-H, 15-H, 16-H, 17-H, 18-H, 19-H, 20-H, 21-H), 6.70–6.77 ppm (m, 2H;
5-H, 22-H); IR (KBr): n˜ =2958 (m), 1689 (m), 1392 and 1367 (m),
1007 cmꢀ1 (vs); UV/Vis (acetonitrile): see Figure 5; MS (EI, 70 eV): m/z
(%): 512 (80) [M]+, 455 (10), 398 (12), 57 (100); HRMS: m/z: calcd for
C38H56: 512.438 [M]+; found: 512.437.
Compound 30e: According to the procedure described for 29, reaction
of TiCl4 (0.60 g, 0.35 mL, 3.12 mmol) in THF (45 mL), zinc dust (0.41 g,
6.32 mmol), pyridine (0.22 g, 0.23 mL, 2.97 mmol), and aldehyde 27d
(0.130 g, 0.44 mmol) in THF (20 mL), followed by chromatographic
workup on silica gel (pentane/dichloromethane), yielded 30e as a dark-
red oil (0.022 g, 18%), which contained several inseparable impurities.
1H NMR (400 MHz, CDCl3): d=1.23 (s, 18H; 1-H, 30-H), 1.37 (s, 18H;
22-H, 34-H), 6.08 (d, J=11.5 Hz, 2H; 4-H, 27-H), 6.11–6.18 (dd, J=11.1,
13.9 Hz, 2H) and 6.25–6.39 (m, 20H; 6-H, 7-H, 8-H, 9-H, 10-H, 11-H, 12-
H, 13-H, 14-H, 15-H, 16-H, 17-H, 18-H, 19-H, 20-H, 21-H, 22-H, 23-H,
24-H, 25-H), 6.82–6.85 ppm (m, 2H; 5-H, 26-H); IR (KBr): n˜ =2953 (m),
1458 (m), 1388 and 1367 (m), 1216 (m), 1008 cmꢀ1 (vs); UV/Vis (di-
chloromethane): see Figure 5; MS (EI, 70 eV): m/z (%): 564 (44) [M]+,
507 (10), 450 (4), 57 (100); HRMS: m/z: calcd for C42H60: 564.468 [M]+;
found: 564.468.
Compound 30a: According to the procedure described for 29, reaction
of TiCl4 (5.76 g, 3.34 mL, 30.38 mmol) in THF (130 mL), zinc dust
(3.97 g, 60.76 mmol), pyridine (2.12 g, 2.16 mL, 26.78 mmol), and alde-
hyde 26 (1.00 g, 5.15 mmol) in THF (60 mL), followed by chromato-
graphic workup on silica gel (pentane), yielded 30a as a pale-yellow solid
(0.60 g, 65%). M.p.1848C; 1H NMR (400 MHz, CDCl3): d=1.23 (s, 18H;
1-H, 14-H), 1.37 (s, 18H; 16-H, 18-H), 6.07 (d, J=11.7 Hz, 2H; 4-H, 11-
H), 6.09–6.15 (AA’XX’, 2H; 6-H, 9-H), 6.30–6.33 (AA’XX’, 2H; 7-H, 8-
H), 6.78–6.85 ppm (dd, J=11.7, 14.2 Hz, 2H; 5-H, 10-H); 13C NMR
(100 MHz, CDCl3): d=157.26 (C3, C12), 132.72 (d, C7, C8), 132.46 (d,
C6, C9), 132.25 (d, C5, C10), 124.00 (d, C4, C11), 38.92 and 37.86 (C2,
C13, C15, C17), 33.71 (q, C1, C14), 31.8 ppm (q, C16, C18); IR (KBr):
n˜ =2956 (s), 1389, 1367 and 1355 (s), 995 cmꢀ1 (vs); UV/Vis (acetonitrile):
lmax (loge)=370 (4.95), 352 (4.99), 334 nm (4.79); MS (EI, 70 eV): m/z
(%): 356 (66) [M]+, 299 (28), 242 (24), 57 (100); HRMS: m/z: calcd for
C26H44: 356.344 [M]+; found: 356.344; elemental analysis calcd (%) for
C26H44 (356.28): C 87.56, H 12.44; found: C 87.21, H 12.31.
Compound 31: Under nitrogen atmosphere and ice cooling, a solution of
phosphorus tribromide (40.1 g, 0.148 mol) in ether (60 mL) was added
slowly to a solution of allyl alcohol 21 (10.0 g, 58.72 mmol) and anhy-
drous pyridine (11.71 g, 12 mL, 0.148 mol) in ether (30 mL). The reaction
mixture was stirred for 30 min at 08C and then for 1 h at RT. After hy-
drolysis with ice water, the aqueous phase was extracted with diethyl
ether. The organic phases were combined, neutralized with saturated
aqueous sodium bicarbonate, and dried over anhydrous Na2SO4. The sol-
vent was removed in vacuo and the residual oil was purified by fractional
distillation. The product 31 (6.35 g, 46%) distilled as a colorless oil at 59–
628C and 0.8 mbar (100–1058C/15 mmHg)[39]. In addition, the rearrange-
ment product 32 (0.37 g, 4%) was isolated as a colorless oil (428C/
0.8 mbar).
Compound 31: 1H NMR (400 MHz, CDCl3): d=1.20 and 1.33 (2ꢅs, 2ꢅ
9H; 5-H, 7-H), 4.28 (d, J=8.8 Hz, 2H; 1-H), 5.59 ppm (t, J=8.8 Hz, 1H;
2-H); 13C NMR (100 MHz, CDCl3): d=158.62 (C3), 121.34 (d, C2), 38.79
and 37.34 (C4, C6), 33.22 (t, C1), 33.27 and 31.54 ppm (2ꢅq, C5, C7); IR
(film): n˜ =2960 (vs), 1483 (m), 1608 (m), 1392 and 1367 (m), 1200 (m),
672 cmꢀ1 (m); UV/Vis (acetonitrile): lmax (loge)=228 nm (4.14); MS (EI,
70 eV): m/z (%): 234/232 (36) [M]+, 153 (4), 96 (100), 81 (86), 57 (98).
Compound 32: 1H NMR (400 MHz, CDCl3): d=0.92 (s, 9H; 7-H), 1.17
(s, 3H; 9-H), 1.83 (s, 3H; 8-H), 4.79–4.96 (m, 2H; 1-H), 4.96–5.05 (m,
2H; 5-H), 6.31–6.38 ppm (dd, J=11.0, 17.5 Hz, 1H, 4-H); 13C NMR
(100 MHz, CDCl3): d=149.89 (C2), 144.28 (d, C4), 113.67 (t, C1), 111.82
(t, C5), 49.39 (C3), 36.02 (C6), 26.66 (q, C7), 24.31 (q, C8), 18.86 ppm (q,
C9); IR (film): n˜ =2960 (vs), 1627 (m), 1395 and 1376 (m), 912 (m),
896 cmꢀ1 (m); UV/Vis (acetonitrile): lmax (loge)=194 nm (3.92); MS (EI,
70 eV): m/z (%): 152 (2) [M]+, 137 (4), 96 (92), 81 (77), 67 (31), 57 (100).
Compound 30b: According to the procedure described for 29, reaction
of TiCl4 (7.63 g, 4.42 mL, 40.19 mmol) in THF (180 mL), zinc dust
(5.26 g, 80.39 mmol), pyridine (2.81 g, 2.87 mL, 35.52 mmol), and alde-
hyde 27a (1.02 g, 4.61 mmol) in THF (90 mL), followed by chromato-
graphic workup on silica gel (pentane), yielded 30b as orange plates
(0.60 g, 64%) (recrystallized from dichloromethane/methanol). M.p.
1
1788C; H NMR (400 MHz, CDCl3): d=1.23 (s, 18H; 1-H, 18-H), 1.37 (s,
18H; 20-H, 22-H), 6.07 (d, J=11.7 Hz, 2H; 4-H, 15-H), 6.11–6.17 (dd,
J=10.5, 14.3 Hz, 2H; 6-H, 13-H), 6.25–6.35 (m, 6H; 7-H, 8-H, 9-H, 10-
H, 11-H, 12-H), 6.80–6.87 ppm (dd, J=11.8, 14.2 Hz, 2H; 5-H, 14-H);
13C NMR (100 MHz, CDCl3): d=157.79 (C3, C16), 133.60, 132.96, 132.39
(d, C7, C8, C9, C10, C11, C12), 132.76 (d, C5, C14), 132.44 (d, C6, C13),
124.04 (d, C4, C15), 38.97 (C2, C17), 37.91 (C19, C21), 33.73 (q, C20,
C22), 31.76 ppm (q, C1, C18); IR (KBr): n˜ =2955 (s), 1392 and 1363 (m),
1217 (m), 998 cmꢀ1 (s); UV/Vis (acetonitrile): see Figure 5<xfigr5; MS
(EI, 70 eV): m/z (%): 408 (100) [M]+, 351 (21), 294 (38), 121 (17), 57
(91); HRMS: m/z: calcd for C30H48: 408.375 [M]+; found: 408.374; ele-
mental analysis calcd (%) for C30H48 (408.79): C 83.35, H 11.22; found: C
83.49, H 11.37.
Compound 30c: According to the procedure described for 29, reaction
of TiCl4 (1.80 g, 1.04 mL, 9.48 mmol) in THF (100 mL), zinc dust (1.04 g,
15.83 mmol), pyridine (0.56 g, 0.57 mL, 7.08 mmol), and aldehyde 27b
(0.250 g, 1.02 mmol) in THF (30 mL), followed by chromatographic
workup on silica gel (pentane), yielded 30c as a dark-red solid (0.194 g,
64%). M.p. 1728C; 1H NMR (400 MHz, CDCl3): d=1.23 (s, 18H; 1-H,
22-H), 1.37 (s, 18H; 24-H, 26-H), 6.07 (d, J=11.7 Hz, 2H; 4-H, 19-H),
6.11–6.17 (dd, J=10.7, 14.3 Hz, 2H; 6-H, 17-H), 6.29–6.38 (m, 10H; 7-H,
8-H, 9-H, 10-H, 11-H, 12-H, 13-H, 14-H, 15-H, 16-H), 6.81–6.88 ppm (dd,
J=11.8, 14.2 Hz, 2H; 5-H, 18-H); 13C NMR (100 MHz, CDCl3): d=
157.91 (C3, C20), 133.92, 133.45, 133.23, 132.91, 132.35 (d, C7, C8, C9,
C10, C11, C12, C13, C14, C15, C16), 132.96 (d, C5, C18), 132.41 (C6,
Compound 35: A mixture of allylbromide 31 (6.00 g, 25.74 mmol) and tri-
phenylphosphine (7.87 g, 30 mmol) was stirred at 708C for 4 h. The re-
sulting colorless crude product was washed with diethyl ether and dried
under high vacuum to give 35 as colorless needles (10.45 g, 82%). M.p.
223–2358C; 1H NMR (400 MHz, CDCl3; H–P decoupled): d=1.04 and
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10519

H. Hopf et al.
1.18 (2ꢅs, 2ꢅ9H; 1-H, 13-H), 4.66 (d, J=7.5 Hz,
2H; 5-H), 5.26 (t, J=7.5 Hz, 1H; 4-H), 7.70–7.98
(m, 15H; Ar-H); 13C NMR (100 MHz, CDCl3; C–P
decoupled): d=162.86 (C3), 135.21, 134.96, 133.89,
133.64, 130.37, 130.26 (d, Ar-C), 117.59 (q, C6, C10,
C11), 108.28 (d, C4), 39.80 and 37.20 (C2, C12),
32.67 and 31.19 (2ꢅq, C1, C13), 26.74 ppm (t, C5);
IR (KBr): n˜ =2959 (s), 1588 (m), 1456 (vs), 1390 and
1368 (s), 1113 (vs), 746 cmꢀ1 (vs); UV/Vis (acetoni-
trile): lmax (loge)=196 (5.05), 268 nm (3.47); MS
(EI, 70 eV): m/z (%): 414 (8) [M]+ꢀHBr, 399 (42),
262 (100), 183 (72), 57 (32); elemental analysis calcd
(%) for C29H36PBr (495.48): C 70.30, H 7.32, Br
16.13; found: C 69.70, H 7.18, Br 16.19.
Wittig reactions
Compound 30 f: Under nitrogen atmosphere,
tBuOK (1.79 g, 16.0 mmol) was added to a solution
of 35 (7.92 g, 16.0 mmol) in anhydrous THF
(320 mL). After stirring the mixture for 30 min at
RT, a solution of aldehyde 26 (1.61 g, 8.06 mmol) in
THF (60 mL) was added. The reaction mixture was
stirred for 2.5 h and then hydrolyzed with ice water.
The aqueous phase was extracted with diethyl ether
(ꢅ2). The combined organic phases were dried over
anhydrous Na2SO4 and the solvent was removed in
vacuo. The residue was filtered through a pad of
silica gel (pentane) and the crude solid was recrys-
tallized (dichloromethane/methanol) to give 30 f as
colorless needles (0.51 g, 19%). M.p. 1918C;
1H NMR (400 MHz, CDCl3): d=1.22 (s, 18H; 1-H
12-H), 1.36 (s, 18H; 14-H, 16-H), 6.07 (d, J=
11.6 Hz, 2H; 4-H, 9-H), 6.19–6.23 and 6.77–
6.83 ppm (AA’XX’, 4H; 5-H, 6-H, 7-H, 8-H);
13C NMR (100 MHz, CDCl3): d=156.73 (C3, C10),
132.69 (d, C6, C7), 131.78 (d, C5, C8), 123.97 (d, C4,
C9), 38.86 and 37.79 (C2, C11, C13, C15), 33.66 (q,
C14, C16), 31.77 ppm (q, C1, C12); IR (KBr): n˜ =
2954 (vs), 1609 (m), 1388, 1365 and 1351 (m), 1215
(m), 988 cmꢀ1 (s); UV/Vis (acetonitrile): see
Figure 4; MS (EI, 70 eV): m/z (%): 330 (44) [M]+,
273 (22), 216 (40), 109 (46), 57 (100); HRMS: m/z:
calcd for C24H42: 330.329 [M]+; found: 330.328; ele-
mental analysis calcd (%) for C24H42 (330.58): C
87.20, H 12.80; found: C 86.81, H 13.11.
Compound 30g: According to the procedure de-
scribed for 30 f, reaction of 35 (2.01 g, 4.06 mmol) in
THF (100 mL), tBuOK (0.46 g, 4.06 mmol), and al-
dehyde 27b (0.50 g, 2.03 mmol) in THF (30 mL) af-
forded 30g (0.35 g, 45%) as a yellow solid. M.p.
1
898C; H NMR (400 MHz, CDCl3): d=1.23 (s, 18H;
1-H, 16-H), 1.37 (s, 18H; 18-H, 20-H), 6.07 (d, J=
11.7 Hz, 2H; 4-H, 13-H), 6.11–6.17 (dd, J=10.6,
14.3 Hz, 2H; 6-H, 11-H), 6.24–6.35 (AA’XX’, 4H; 7-
H, 8-H, 9-H, 10-H), 6.79–6.86 ppm (dd, J=11.7,
14.2 Hz, 2H; 5-H, 12-H); 13C NMR (100 MHz,
CDCl3): d=157.54 (C3, C14), 133.28 and 132.39 (d,
C7, C8, C9, C10), 132.58 (d, C5, C12), 132.42 (d, C6,
C11), 124.02 (d, C4, C13), 38.95 (C2, C15), 37.88
(C17, C19), 33.71 (q, C18, C20), 31.75 ppm (q, C1,
C16); IR (KBr): n˜ =2954 (m), 1587 (m), 1390 and
1364 (m), 998 cmꢀ1 (vs); UV/Vis (acetonitrile): lmax
(loge)=398 (5.00), 376 (5.00), 356 nm (4.80); MS
(EI, 70 eV): m/z (%): 382 (100) [M]+, 325 (32), 268
(46), 57 (88); HRMS: m/z: calcd for C28H46: 382.359
[M]+; found: 382.359; elemental analysis calcd (%)
for C28H46 (382.31): C 78.38, H 10.71; found: C
78.13, H 10.98.
10520
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522

Fully Terminally tert-Butylated Linear Polyenes
FULL PAPER
Compound 30h: According to the procedure described for 30 f, reaction
of 35 (0.332 g, 0.67 mmol) in THF (15 mL), tBuOK (0.075 g, 0.34 mmol),
and aldehyde 27d (0.100 g, 0.34 mmol) in THF (2.5 mL) afforded an in-
separable E/Z mixture of 30h as an orange solid (0.062 g, 43%). We
were successful in obtaining single crystals (recrystallization from chloro-
benzene) suitable for X-ray structural analysis. The sample contained
crystals with both the E and Z configuration at the C3=C4 double bond,
hence the structure is disordered in this part of the molecule (see below).
1H NMR (400 MHz, CDCl3): d=0.88, 1.08, 1.23, 1.37 (s, 4H; 1-H, 20-H,
22-H, 24-H), 6.02, 6.07 (d, J=11.42, J=11.7 Hz, 2H; 4-H, 17-H), 6.11–
6.17 (dd, J=10.7, 14.2 Hz, 2H; 6-H, 15-H), 6.20–6.36 (m, 8H, 7-H, 8-H,
9-H, 10-H, 11-H, 12-H, 13-H, 14-H), 6.49–6.55 and 6.81–6.87 ppm (dd,
J=11.3, 14.2 Hz, 1H and dd, J=11.7, 14.2 Hz, 1H; 5-H, 16-H); 13C NMR
(100 MHz, CDCl3): d=157.78, 146.15 (C3, C18), 133.71, 133.36, 133.13,
132.96, 132.84, 132.68, 132.43, 132.39, 132.03, 130.49, 127.17, 124.04 (d,
C4, C5, C6, C7, C8, C9, C10, C11, C12, C13, C14, C15, C16, C17), 44.37,
38.97, 37.91, 36.93 (C2, C19, C21, C23), 33.72, 31.74, 26.64, 23.76 ppm (q,
C1, C20, C22, C24); IR (KBr): n˜ =2955 (s), 1391 and 1366 (m), 1216 (m),
1001 cmꢀ1 (vs); UV/Vis (acetonitrile): lmax (loge)=250 (4.05), 356 (4.17),
374 (4.51), 392 (4.80), 416 (4.98), 442 nm (4.98); MS (EI, 70 eV): m/z
(%): 434 (100) [M]+, 337 (15), 320 (21), 57 (68); HRMS: m/z: calcd for
C32H50: 434.391 [M]+; found: 434.390
[7] a) H. Shichi, Biochemistry of Vision, Academic Press, New York,
1983; b) Israel J. Chem. 1995, 35, special issue on “Photophysics and
Photochemistry of Retinal Proteins”.
[8] See Nobel lecture of: H. Shirakawa, Angew. Chem. 2001, 113, 2642–
2648; Angew. Chem. Int. Ed. 2001, 40, 2574–2580.
[9] Whereas there are hundreds of references to conjugated trienes in
the literature, this number drops by two orders of magnitude for the
respective tetraenes and even more for the pentaenes and higher
oligomers. For some leading references concerning spectroscopic
properties and molecular structure, see: a) 1,3,5-hexatriene (1a, n=
1): F. W. Langkilde, R. Wilbrandt, O. F. Nielsen, D. H. Christensen,
F. M. Nicolaisen, Spectrochim. Acta 1987, 43A, 1209–1230; M. Træt-
teberg, Acta Chem. Scand. 1968, 22, 628–640 and M. Traetteberg,
Acta Chem. Scand. 1968, 22, 2294–2304; S. J. Cyvin, G. Hagen, M.
Traetteberg, Acta Chem. Scand. 1969, 23, 3285–3289. An X-ray
structural analysis of trans-1,3,5-hexatriene has been performed but
the results are unpublished so far: R. Boese, H. Hopf, C. Mlynek,
unpublished work; b) 1,3,5,7-octatetraene (1a, n=2): H. Yoshida,
M. Tasumi, J. Chem. Phys. 1988, 89, 2803–2809; R. H. Baughman,
B. E. Kohler, I. J. Levy, C. Spangler, Synth. Met. 1985, 11, 37–52; B.
Kohler, J. C. Woehl, J. Phys. Chem.
A 1999, 103, 2435–2445;
c) 1,3,5,7,9-decapentaene (1a, n=3): B. E. Kohler, V. Terpongov, J.
Chem. Phys. 1998, 108, 9586–9593; J. T. Lurito, I. R. Dunkin, L. An-
drews, J. Mol. Spectrosc. 1985, 114, 54–59.
X-ray structure determinations: Numerical details are presented in
Table 2.
Data collection and reduction: Crystals were mounted in inert oil on glass
fibers and transferred to the cold gas stream of the diffractometer (27d,
30c, cis-30h: Bruker SMART 1000 CCD; 27c: Stoe STADI-4; others:
Siemens P4, with appropriate low-temperature attachments). Measure-
ments were performed with monochromated MoKa irradiation (l=
0.71073 ꢄ). No absorption corrections were performed.
[10] a) F. Sondheimer, D. A. Ben-Efraim, R. Wolovsky, J. Am. Chem.
Soc. 1961, 83, 1675–1681; b) F. Sondheimer, D. A. Ben-Efraim, Y.
Gaoni, J. Am. Chem. Soc. 1961, 83, 1682–1685.
[11] P. Nayler, M. C. Whiting, J. Chem. Soc. 1955, 3037–3047.
[12] F. Bohlmann, H. J. Mannhardt, Chem. Ber. 1956, 89, 1307–1315.
[13] Review: R. Kuhn, Angew. Chem. 1937, 50, 703–718.
[14] a) G. Mayer, S. Pfriem, U. Schꢁfer, R. Matusch, Angew. Chem. 1978,
90, 552–553; Angew. Chem. Int. Ed. Engl. 1978, 17, 520–521; b) G.
Maier, S. Pfriem, U. Schꢁfer, R. Matusch, Chem. Ber. 1981, 114,
3965–3987 and references quoted therein; c) G. Maier, F. Fleischer,
Tetrahedron Lett. 1991, 32, 57–60.
[15] M. Trætteberg, P. Bakken, H. Hopf, R. Hꢁnel, Chem. Ber. 1994, 127,
1469–1478.
[16] M. Trætteberg, H. Hopf, H. Lipka, R. Hꢁnel, Chem. Ber. 1994, 127,
1459–1467.
Structure refinement: The structures were refined anisotropically against
F2 (program SHELXL-97[41]). Hydrogen atoms were included as rigid
methyl groups or with a riding model.
Exceptions/special features of refinement: For all compounds that crystal-
lized with solvent, the solvent was disordered; appropriate similarity re-
straints were used to improve refinement stability. A second structure of
30b was determined as a deuteriochloroform hydrate and proved to be
essentially isostructural to the first form; the data were of very limited
quality and have been deposited as indicated below. Compound cis-30h
was contaminated, both in solution (see discussion) and in the crystal, by
a few percent of the trans isomer (seen as a difference in electron density
of max. 0.75 eꢄ3); no appropriate disorder model could be refined. One
tert-butyl group was also disordered over two positions.
[17] W. R. Roth, O. Adamczak, R. Breuckmann, H.-W. Lennartz, R.
Boese, Chem. Ber. 1991, 124, 2499–2521.
[18] For early studies on conjugated oligoenes with one terminal tert-
butyl substituent, see: C. W. Spangler, T. P. Jondahl, B. Spangler, J.
Org. Chem. 1973, 38, 2478–2484. For the more recent preparation
of polyolefins bearing functional groups at the terminal positions,
see: a) F. Babudri, F. Naso, J. Org. Chem. 1997, 62, 3291–3298; b) E.
Shirakawa, H. Yoshida, Y. Nakao, T. Hiyama, J. Am. Chem. Soc.
1999, 121, 4290–4291; c) S. BouzBouz, C. Roche, J. Cossy, Synlett
2009, 803–807, and references therein.
CCDC-726529 (27a), 726530 (27c), 726531 (27d), 726532 ((E)-29),
726533 ((Z)-29), 726534 (30a), 726535 (30b), 726536 (30c), 726537 (30 f),
726538 (30g), 726539ACHTUNGTRENNUNG(30h), and 726540 (30b (second form)) contain the
supplementary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
[19] R. R. Schrock, K. Knoll, J. Am. Chem. Soc. 1989, 111, 7989–8004.
[20] For the Raman scattering of the olefins with 3 to 12 consecutive
double bonds, see: H. E. Schaffer, J. Chem. Phys. 1991, 94, 4161–
4170; for the UV/Vis and Raman-scattering spectra of the hydrocar-
bons 10 (n=1–11), see: R. R. Chance, H. Schaffer, K. Knoll, R.
Schrock, R. Silbey, Synth. Met. 1992, 49–50, 271–280.
[1] For part XIII of this series, see: O. Klein, H. Hopf, J. Grunenberg,
Eur. J. Org. Chem. 2009, 2141–2148.
[2] “The Chemistry of Dienes and Polyenes”: The Chemistry of Func-
tional Groups, Vol. I (Eds.: S. Patai, Z. Rappoport), Wiley, New
York, 1997.
[3] H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH, Weinheim
2000, Chapter 7, pp. 103–112.
[4] K. Weissermel, H.-J. Arpe, Industrial Organic Chemistry, 4th ed.,
Wiley-VCH, Weinheim, 2003.
[5] a) M. B. Sporn, A. B. Roberts, D. S. Goodman, The Retinoids, Vol. I
& II, Academic Press, Orlando, 1984; b) Chemistry and Biology of
Synthetic Retinoids (Eds.: M. I. Dawson, W. H. Okamura), CRC
Press, Boca Raton, 1990.
[21] Th. Bally, K. Roth, W. Tang, R. R. Schrock, K. Knoll, L. Y. Park, J.
Am. Chem. Soc. 1992, 114, 2440–2446. For the IR spectra of potassi-
um doped a,w-di-tert-butylated polyolefins, see: G. Zerbi, Chem.
Phys. Lett. 1994, 231, 70–74; cf. M. Rumi, G. Zerbi, Chem. Phys.
Lett. 1995, 242, 639–643 and M. Del Zoppo, C. Castglioni, M. Vero-
nelli, G. Zerbi, Synth. Metals 1993, 57, 3919–3926. For Pd complexes
of oligoenes end-capped by one tert-butyl substituent each: T. Mura-
hashi, J. Am. Chem. Soc. 2006, 128, 4377–4388.
[22] T. Swager, R. H. Grubbs, J. Am. Chem. Soc. 1988, 110, 2973–2974.
[23] A. Kiehl, A. Eberhardt, K. Mꢀllen, Liebigs Ann. Chem. 1995, 223–
230.
[6] Carotinoids—Handbook (Eds.: G. Britton, S. Liaaen-Jensen, H.
Pfander), Birkhꢁuser, Basel, 2004.
[24] H. B. Hass, M. L. Bender, J. Am. Chem. Soc. 1949, 71, 1767–1769.
Chem. Eur. J. 2010, 16, 10507 – 10522
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10521

H. Hopf et al.
[25] H. Hopf, C. Mlynek, D. Klein, M. Traetteberg, P. Bakken, Eur. J.
Org. Chem. 2001, 1385–1391; P. Kilickiran, S. Sankararaman, H.
Hopf, Ind. J. Org. Chem. B 2001, 40, 781–785.
[26] T. Mukaiyama, T. Sato, J. Hanna, Chem. Lett. 1973, 1041–1044.
[27] E. J. Corey, R. L. Danheiser, S. Chandrasekaran, J. Org. Chem.
1976, 41, 3929–3930.
[34] K. W. Hausser, R. Kuhn, G. Seitz, Z. Physik. Chem. 1935, 29, 391–
417.
[35] J. Catalꢇn, H. Hopf, Eur. J. Org. Chem. 2004, 4694–4702.
[36] Version 5.30: F. H. Allen, Acta Crystallogr. 2002, 58, 380–388.
[37] F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen,
R. Taylor, J. Chem. Soc. Perkin Trans. 2 1987, 1–19.
[38] F. C. Whitmore, E. E. Strahly, J. Am. Chem. Soc. 1933, 55, 4135–
4157.
[39] M. Andrac, Ann. Chim. 1964, 287–315.
[40] T. M. Cresp, M. V. Sargent, P. Vogel, J. Chem. Soc. Perkin Trans. 1
1974, 37–41.
[28] D. Lenoir, Synthesis 1977, 553–554.
[29] R. Kuhn, A. Winterstein, Helv. Chim. Acta 1928, 11, 87–116.
[30] E. R. Blout, M. Fields, J. Am. Chem. Soc. 1948, 70, 189–193.
[31] a) D. Soullez, G. Plꢆ, L. Duhamel, J. Chem. Soc. Perkin Trans. 1
1997, 1639–1645; b) Y. Ramondenc, G. Plꢆ, Tetrahedron 1993, 49,
10855–10876.
[32] P. K. Das, R. S. Becker, J. Phys. Chem. 1978, 82, 2081–2093.
[33] a) R. M. Silverstein, G. C. Bassler, C. M. Terence, Spectrometric
Identification of Organic Compounds, 5th ed., Wiley, New York,
1991; b) M. Klessinger, J. Michl, Lichtabsorption und Photochemie
organischer Molekꢁle, VCH, Weinheim, 1989; c) H. Kuhn, J. Chem.
Phys. 1949, 17, 1198–1212.
[41] G. M. Sheldrick, Acta Crystallogr. Sect. A 2002, 64, 112–122.
[42] H. Hopf, P. KiliÅkiran, unpublished results.
Received: January 5, 2010
Revised: May 5, 2010
Published online: July 21, 2010
10522
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10507 – 10522