Conformational analysis of oligo-1,3-dioxanylmethanes

Article abstract of DOI:10.1002/1099-0690(200204)2002:7<1292::AID-EJOC1292>3.0.CO;2-6

Stereoselective synthesis of a series of 1,3-dioxan-4-ylmethanes 1-9 has been achieved by use of solely substrate-based asymmetric induction. The simple C₂-symmetric bis(dioxanyl)methane 1 has a greater than 99% conformational preference at the two inter-ring bonds for the conformation 1a. The homologous structures 3-9 contain up to five dioxanylmethane units, maintaining a high conformational preference in each of the bis(dioxanyl)methane units. Thus, these flexible compounds reach a conformational preference in excess of 90% over up to eight rotatable interring bonds. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200204)2002:7<1292::AID-EJOC1292>3.0.CO;2-6

FULL PAPER
Conformational Analysis of oligo-1,3-Dioxanylmethanes[‡]
Thomas Trieselmann,[a] Reinhard W. Hoffmann,*[a] and Karsten Menzel[a]
Keywords: Asymmetric synthesis / Conformational analysis / Oxygen heterocycles
Stereoselective synthesis of a series of 1,3-dioxan-4-ylme- tional preference in each of the bis(dioxanyl)methane units.
thanes 1−9 has been achieved by use of solely substrate-
based asymmetric induction. The simple C2-symmetric
Thus, these flexible compounds reach a conformational pref-
erence in excess of 90% over up to eight rotatable inter-
bis(dioxanyl)methane 1 has a greater than 99% conforma- ring bonds.
tional preference at the two inter-ring bonds for the con-
formation 1a. The homologous structures 3−9 contain up to
(
Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
five dioxanylmethane units, maintaining a high conforma- 2002)
Nature has attained conformation ‘‘design’’ of flexible 0.98). To test the validity of this reasoning, we set out to
structures through evolution. This is evident in many bioac- study a set of compounds comprising combinations of up
tive natural products; examples are to be found in the flex- to four identical modules. In this manner we wanted to
ible linker units between diverse parts of certain pharmaco- evaluate how far conformation design on a modular basis
phore structures.[2] Conformation design of flexible struc- can be pushed in a real system.
tures therefore turns out to be a means to specify and separ-
ate distinct biological activities of multivalent drugs.[3]
Conformation design becomes important in materials sci-
ence as well, when conformation-dependent properties need
to be optimised.[4,5] In these contexts, a modular approach
to conformation design of larger flexible backbone struc-
tures appears attractive. Such an approach could rely on
small modules; molecules possessing only one or two rotat-
able bonds and to a great extent populating a single con-
formation. A combination of several modules of this kind
should then give rise to larger flexible molecules, which
should, despite their higher number of rotatable bonds,
nevertheless populate a single preferred conformation.
There is, however, an obvious limitation in such an ap-
proach. Consider a single module in which the conforma-
tional preference for population of the desired conforma-
tion is x (0 Ͻ x Ͻ 1). In a larger molecule, arising out of a
combination of n modules, the overall preference to popu-
late a single backbone conformation will be xn. This means
that the overall conformational preference one can attain
will decrease exponentially with the number of modules
that are combined. A sizeable conformational preference in
molecules composed of n Ͼ 1 modules can therefore only
be attained if the conformational preference in each module
is approaching 1.0 (in practice, it should be greater than
The module chosen for our studies was that of the
bis(dioxanyl)methanes 1 (cf. also 2).[6] Compound 1, for in-
stance, would be expected to populate conformation 1a
with a very high preference, on the grounds that any 120°
rotation about one of the inter-ring bonds would generate
conformations that were destabilised by two simultaneous
syn-pentane interactions. While such high-energy con-
formations relax into non-diamond-lattice-type conforma-
tions, they should nevertheless be destabilised by greater
than 5 kcal[7] with respect to the most stable conformation
1a.[8]
On this basis, the conformational preference in 1 should
approach x ϭ 1.0 and, hence, would augur well for en-
dowing molecules such as 3؊5, products of combination of
several modules of 1, with large overall conformational
preferences.
We report here on a study of such oligo-dioxanylme-
thanes. The study was extended to the arylidene acetals 6
and 8, as well as the mixed acetals 7 and 9, in order Ϫ
among other reasons Ϫ to facilitate conformational ana-
[‡]
Flexible Molecules with Defined Shape, XVII. Ϫ Part XVI:
3
Ref.[1]
lysis. This was based on the determination of vicinal JH,H
[a]
Fachbereich Chemie der Philipps-Universiät,
Hans-Meerwein-Straße, 35032 Marburg, Germany
Fax: (internat.) ϩ 49-6421/282-8917
constants along the backbone and therefore required that
each of the relevant proton signals could be individually
assigned and not be subject to severe signal overlap.
E-mail: rwho@chemie.uni-marburg.de
1292
WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0407Ϫ1292 $ 17.50ϩ.50/0 Eur. J. Org. Chem. 2002, 1292Ϫ1304

Conformational Analysis of oligo-1,3-Dioxanylmethanes
FULL PAPER
The same building block 10 can be used to enter into the
synthesis of the tetrakis(dioxanyl)methanes 4 and 8. The
diol 10 was converted into the bis(benzyl ether) 12 (71%).
We were unable to produce the diketone 15 by Wacker ox-
idation and so took a longer route by ozonolysis to afford
the dialdehyde 13, Grignard addition to give the diol 14
and oxidation to provide the desired diketone 15.
For the next bidirectional chain extension we relied on
the sizeable 1,5-asymmetric induction inherent in boron al-
dol additions.[16,17] Thus, the diketone was converted into
the bis(dicyclohexylboron) enolate and allowed to react
with the aldehyde 16[18] at Ϫ90 °C. This resulted in 65% of
a diastereomer mixture, in which the anti/anti/anti diastere-
omer 17 predominated over the anti/anti/syn diastereomer
by 4:1. The symmetric nature of the anti/anti/anti diastere-
omer 17 was readily apparent from its 13C NMR spectrum.
The resulting diastereomer mixture was subjected to stand-
ard 1,3-anti reduction[19] with tris(acetoxy)borohydride to
give the tetraol 18, together with its diastereomer. The
benzyl groups were removed by catalytic hydrogenation and
the resulting octaol 19 (80%) was converted into the tetrakis-
(acetonide) 4 (65%) or the tetrakis(benzylidene acetal) 8
(62%) by standard acetalisation reactions. Flash chromato-
graphy in both cases furnished diastereomerically pure
products.
Synthesis
The synthesis of the oligo-dioxanylmethanes has to take
into account that all the target compounds are either C2-
or σ-symmetric (meso) compounds. A simplification of the
syntheses might therefore be attained by use of a bidirec-
tional strategy.[12,13] For meso compounds, however, this im-
plies that stereogenic centres of opposite relative configura-
tion would have to be created in a single reaction, a fact
that would preclude any reagent-based asymmetric induc-
tion. The syntheses described below therefore had to rely
only on substrate-based asymmetric induction to create the
stereogenic centres (for a preliminary account see ref.[14]).
The synthesis of the prototype structure 1 started from
malondialdehyde, which was diprenylated by Imai’s
method[15] to give a 4:1 mixture of the rac- and meso-diols
10 and 11. The C2-symmetric diol 10 could be obtained in
25% yield by crystallisation. Ozonolysis, followed by reduct-
ive workup and acetalisation, provided 1 in 75% yield.
Transacetalisation with benzaldehyde dimethyl acetal fur-
nished compound 2.
Eur. J. Org. Chem. 2002, 1292Ϫ1304
1293

T. Trieselmann, R. W. Hoffmann, K. Menzel
FULL PAPER
The tris- and tetrakis(dioxanyl)methanes required the
meso compound 25 as a starting material for a bidirectional
synthesis. Its synthesis started with a double allylation of
dimethylmalondialdehyde (20), which afforded a 2:1 mix-
ture of 21 and 22.
that treatment of the 2:1 mixture of 21 and 22 with p-me-
thoxybenzaldehyde dimethyl acetal in the presence of cam-
phorsulfonic acid readily converted 21 into the acetal 31,
but left 22 essentially unchanged. The polarity difference
between 31 and 22 allowed for ready chromatographic sep-
aration.
While the structures could easily be assigned through the
number of the 13C NMR signals, separation of 21 and 22
could not be achieved, nor that of their derived acetonides.
This necessitated a somewhat lengthier route in order to
reach the meso-bis(benzyl ether) 25. The mixture of diols
21 and 22 was monobenzylated to give 85% of a 2:1 mixture
of 23 and 24. This latter mixture could be separated by
flash chromatography. The desired major diastereomer 23
was then benzylated to give the symmetrical bis(benzyl
ether) 25 (95%). The undesired diastereomer 24 may be con-
verted into 23 through an oxidation-reduction sequence.
High selectivities (16:1) in the reduction step were attained
by use of LiAlH4 in Et2O/THF at Ϫ120 °C.[20]
With the meso compound 25 in hand, ozonolysis of both
double bonds furnished the dialdehyde 27 (88%). The sub-
sequent chain-extension had to establish a 1,3-diol unit with
1,3-anti diastereoselectivity. By following a precedent from
Evans’ group,[21,22] this was achieved with the Mukaiyama
aldol procedure and a nonchelating Lewis acid (BF3·OEt2).
The anti/anti diastereomer 29 was obtained with a 6:1 pref-
erence over the anti/syn diastereomer 28, structure assign-
ment being based on the number of 13C NMR signals. That
the symmetrical isomer 29 was the anti/anti diastereomer
(as indicated) and not the syn/syn one could be established
by comparison of the signal positions of certain 13C NMR
signals, since a syn-β-hydroxy ether has the 13C signals of
its oxygen-bearing carbon atoms downfield from those of
the corresponding anti diastereomer.[23]
The further conversion of 31 into 6 and 7 followed the
route developed above for 3; ozonolysis of 31, followed by
Mukaiyama aldol addition, provided a 5:1 mixture of 34
and 33. Their structure assignment was again based on 13
C
The subsequent transformation of 29 into compound 3
proceeded uneventfully; reduction of 29 with DIBAH fur-
nished the tetraol 30, which was straightaway debenzylated
[H2/Pd(OH)2] and converted into the tris(acetonide) 3
(80%).
NMR spectroscopic data, as outlined above for 28 and 29.
Chromatographic purification furnished pure 34, which was
reduced with LiBH4 to give the tetraol 35 (90%). This could
be converted without problem into the acetals 6 (67%) and
When we attempted the synthesis of the mixed tris(diox- 7 (55%).
anyl) compounds 6 and 7, we took account of the unsatis-
Having gained experience in the generation of stereo-
factory diastereomer separation of 21 and 22. We found chemically defined skipped polyol sequences using bidirec-
1294
Eur. J. Org. Chem. 2002, 1292Ϫ1304

Conformational Analysis of oligo-1,3-Dioxanylmethanes
FULL PAPER
tional techniques, we had become confident that we could
also master a synthesis of the pentakis(dioxanyl)methane
compounds 5 and 9. We used the dialdehyde 32 as a starting
point for this endeavour. Chain extension by the Mukai-
yama aldol technique involved the enol silyl ether 36[24] and
again the nonchelating Lewis acid BF3·OEt2. This reaction
afforded 73% of a 6:1 mixture of a symmetrical (38) and
a nonsymmetrical (37) bis(aldol) adduct, which could be
separated by chromatography. The structure assignment
was based, as before, on the relative positions of the charac-
teristic 13C NMR signals.[23]
With the relative configurations at all stereocenters in 42
established, we attempted a selective deprotection of the p-
methoxybenzyl ethers in the presence of the p-methoxyben-
zylidene acetal. After what was a short reaction time (16 h)
for a hydrogenolysis, we were able to secure substantial
amounts of the octaol 44. Subsequent acetalisation with 2-
methoxypropene furnished the differentially protected pen-
takis(dioxane) 9.
The free hydroxy groups in 38 were converted into p-me-
thoxybenzyl ethers to give 39 in 69% yield. This set the
stage for the next bidirectional chain extension using the
1,5-asymmetric induction inherent in the boron aldol meth-
odology. The required aldehyde 40 was generated in two
steps from 2,2-dimethylpropanol. Treatment of 40 with the
bis(dicyclohexylboron) enolate of 39 afforded the full-
length carbon skeleton (72%) with a 5:1 selectivity in favour
of a symmetrical product, to which we assigned (see below)
the desired structure 41. This mixture was subjected to an
1,3-anti-specific reduction of the aldol units with tris(ace-
toxy)borohydride[19] to give 96% of the tetraol 42 admixed
with its stereoisomer. To obtain definite proof of the ste-
reostructure of 42, all protecting groups were removed by
hydrogenolysis and the resulting decaol 43 was converted
into the pentakis(acetonide) 5. The diagnostic 13C NMR
chemical shifts of the acetal carbon atoms all fell in the
range of δ ϭ 99.5Ϫ99.9, showing that all acetonide rings
were derived from syn-1,3-diol units.[25,26] Moreover, there
is precedent to show that these characteristic 13C NMR sig-
nal positions are also found in 5,5-disubstituted 1,3-diox-
anes.[27] The fact that all the acetonides in 5 were based
on syn-1,3-diol units would require that the aldol addition
between 39 and 40 had given the anticipated 3,7-anti-diol
relationship. Otherwise, the subsequent 3,5-anti-selective re-
duction of di-epi-41 with tris(acetoxy)borohydride would
have resulted in a 5,7-anti relationship at the stage of 42.
Hence, upon conversion into the pentakis(acetonide), two
of the acetonide rings should then be derived from an anti-
1,3-diol unit.
We have described here a sequence of transformations
that allowed us to access skipped polyols with an anti/syn/
anti pattern of stereocentres. In order to achieve a bidirec-
tional approach we were restricted to the use only of sub-
strate-based asymmetric induction. The key to success is the
high 1,3-asymmetric induction in the Mukaiyama aldol re-
action, the high 1,5-asymmetric induction in the Paterson/
Evans enol borinate addition and in the reliable 1,3-anti
reduction with tris(acetoxy)borohydride. This efficiency, in
combination with a bidirectional approach, made it pos-
sible to access structures such as 5 or 9 in only eight steps
from malondialdehydes.
Force Field Calculations
Force field calculations were carried out with the MM3*
force field implemented in the MACROMODEL pro-
Eur. J. Org. Chem. 2002, 1292Ϫ1304
1295

T. Trieselmann, R. W. Hoffmann, K. Menzel
FULL PAPER
gram[28] (versions 4.5 or 7.0) in order to estimate conforma- tions fell to 92 and 89%. The difference in energy between
tional preferences for the compounds synthesised. For com- the most stable conformation of 6 and the one with a dis-
pound 1, the data shown in Table 1 were obtained by means torted inter-ring bond decreased from 15.5 kJ·molϪ1 in 2 to
of a Monte Carlo search,[29] with torsion allowed about the merely 7.5 kJ·molϪ1. Clearly, we had no clue why this
inter-ring and stereochemically indifferent intra-ring bonds. should be so. This was yet another reason to synthesise and
The structures generated were minimised to the nearest study compounds 6 and 7.
minimum, and of the resulting structures only those with
The calculations for the mixed acetal compound 7 pre-
energies lying Յ 25 kJ·molϪ1 higher than that of the lowest dicted conformational behaviour close to that for the tris(a-
energy conformer were considered further. The conformer cetonide) 3, the energy difference between the most stable
population was estimated by subjecting the ensemble of conformation and the one with a distorted inter-ring bond
conformations generated in this fashion to a Boltzmann was calculated to be 17 kJ·molϪ1, corresponding to a con-
distribution for 25 °C, taking the statistical degeneracy of formational preference of Ͼ 99%. The same applied to the
certain conformations into account. This is demonstrated mixed acetal 9, for which the lowest-energy non-diamond-
for compound 1 in Table 1.
lattice conformation involved a distortion next to a benzyli-
dene acetal ring. We were curious at this point as to which
of these features would stand the test of experimental con-
formational analysis.
Table 1. Conformer energies of 1 calculated with the MM3* force
field method
Conformational Analysis
Conformational analysis of the compounds of interest
here was based on 3JH,H NMR coupling constants.[31] Com-
pound 1 has only one spin system in which protons H1 and
H1Ј as well as H2 and H2Ј are homotopic (cf. 1a) due to the
C2 symmetry of the molecule. This results in a higher order
spin system. The individual coupling constants were estim-
ated to Ϯ0.2 Hz by simulation of the coupling patterns.
This indicated that the H1,H2 coupling constant at room
temperature amounted to 10.4 and the H1,H2Ј coupling to
1.7 Hz. The coupling constants in the corresponding
benzylidene acetal were approximated to 11.1 and 1.8 Hz,
respectively.
Conformer
Dihedral angles [°]
1Ϫ4 2Ϫ5 3Ϫ6 4Ϫ7 [kJ/mol] (%; 25 °C)
Energy Population
1
Ϫ176 Ϫ177 Ϫ175 Ϫ176 0.00
Ϫ157 Ϫ174 Ϫ175 Ϫ176 11.37
Ϫ176 Ϫ175 Ϫ174 Ϫ157 11.37
Ϫ156 Ϫ171 Ϫ172 Ϫ156 18.34
94.09
2.88
2.88
0.06
0.02
0.02
0.02
0.02
2[a]
3[a]
4[a]
5[a]
6
Ϫ74
Ϫ176 Ϫ177
Ϫ178
Ϫ175 Ϫ165
178 Ϫ177
176 20.70
178 Ϫ74 20.70
93 Ϫ165 Ϫ175 21.34
93 Ϫ178 21.34
7
8
[a]
Conformers 2Ϫ5 have a twist-boat conformation.
The global minimum conformation was found to be 1a,
corresponding to a diamond lattice arrangement. Con-
formers 2 and 3 each have a twist-boat conformation in
one of the dioxane rings, conformer 4 in both. Nevertheless,
conformers 2Ϫ4 maintain the same arrangement as con-
former 1 at the inter-ring bonds. Therefore, the predicted
conformational preference at the inter-ring bonds of 1
amounted to 99.9% (i.e. Ͼ 99%) at 25 °C. Similarly, the
Compounds 3؊9 each have more than one spin system.
preferences for compounds 3 (4) were predicted to Ͼ 99 (Ͼ These spin systems are separated and can be analysed indi-
99) % including twist-boat conformers at either or both of vidually. They reflect the local conformational preferences
the outer rings.[30]
in the individual segments. Within each of the tris(dioxanyl)-
The calculations therefore suggested that the energy dif- methanes 3, 6, and 7 the two spin systems are identical,
ference between the ground state conformation with an ex- due to the σ symmetry of the molecules. Each spin system
tended backbone and the higher energy conformations with comprises four protons H1 to H4 (cf. 45). As the coupling
distorted backbones was so high (Ͼ 20 kJ·molϪ1) for 1 and constants H1,H2; H3,H4 and H1,H3; H2,H4 turned out to
the related compounds 3, 4, and 5 that a conformational be pairwise identical for compounds 3, 6, and 7, there was
preference in excess of 99% should result irrespective of the no need for an individual assignment of the proton signals.
number of modules involved.
The coupling constants found are listed in Table 2.
The benzylidene acetal 2 appeared to fit nicely into this
In the pentakis(dioxanyl)methane 5, the signals of the
pattern, with a 99.0% predicted conformational preference four spin systems were completely superimposed. No res-
for a conformation of the 1a type at the inter-ring bonds. olution could be attained at 800 MHz by changing the solv-
Of course, there was a multitude of conformers differing in ent from CDCl3 to CD3OD, CD3COCD3, or C6D6. We
the torsion of the arylϪacetal bond. However, compounds therefore resorted to an analysis of the pentakis(dioxanyl)-
6 and 8 showed a deviating pattern, as the calculated con- methane 9, containing one benzylidene acetal. Because of
formational preferences for the fully extended conforma- the σ symmetry of the molecule, two of the four spin sys-
1296
Eur. J. Org. Chem. 2002, 1292Ϫ1304

Conformational Analysis of oligo-1,3-Dioxanylmethanes
FULL PAPER
3
Table 2. Summary of the relevant J coupling constants
Compound
Solvent
T [K]
Segment
3J [Hz]
H1,H2 ϭ H3,H4
H1,H3 ϭ H2,H4
1
2
3
4
6
7
C6D5CD3
CDCl3
CDCl3
CDCl3
CDCl3
C6D6
CDCl3
CDCl3
CDCl3
298
298
298
273
223
298
298
223
298
10.4
11.1
9.6
9.6; 9.9
9.6
8.6
9.9
9.9
9.0
9.9
9.3
1.7
1.8
2.4
2.3; 2.5
2.2
2.9
2.0
2.0
3.0
A
9
A
B
A
B
1.1
3.5
0.5
CDCl3
213
10.0
tems are pairwise identical, and so only two spin systems
If the ensemble of the data obtained (cf. Table 2) is exam-
have to be analysed. Assignment of the individual proton ined, some general features become evident. In the tetrakis-
signals was based on NOE experiments, as shown in 9a. (dioxanyl)methanes 3, 6, and 7, and also in the pentakis-
The sequences of NOE contacts originated either at protons (dioxanyl)methane 9, there are only two coupling constants
HA and HB in the terminal CH2ϪO group or at the acetal (instead of four) to characterise each spin system. This in-
proton HC of the benzylidene acetal.
dicates that local C2 symmetry in each of the modules is
retained when several modules are combined. It is a moot
point whether the four coupling constants found for the
spin system in the tetrakis(dioxanyl)methane 4 are really
different. The numerical values are pairwise so close to one
another that they are not indicative of an absence of local
C2 symmetry. Moreover, the coupling constants in 1, 3, and
7 were found to be temperature-invariant over a temper-
ature change of 75°C. This indicated that the local con-
former preference was higher than 95% even at room tem-
perature.[32]
Conformational analysis of the tetrakis(dioxanyl)me-
thane 4 was again hampered by signal overlap. There are
two identical spin systems at the outer methylene bridges
and one spin system in the centre of the molecule, which
because of the C2 symmetry should give a higher order mul-
tiplet. At 237 K the splitting pattern of two CHϪO group
protons could be recorded separately from the rest. None
of these multiplets appeared as a higher order pattern sim-
ilar to that recorded for Ϫ for instance Ϫ 1. We therefore
assigned both signals to spin system A. One signal could be
clearly assigned to H1, thanks to an NOE contact with HB.
The coupling constants to H1 amounted to 9.6 and 2.5 Hz,
the coupling constants to H4 were recorded as 9.9 and
2.3 Hz. Thus, only limited information was available for
compound 4.
One question to be addressed is whether the local con-
former preference in each module remains the same when
several modules are combined. The difference between the
large and small coupling constants (i.e., the variation of the
coupling constants) is a measure of the prevailing con-
formational preference.[31] This alteration decreases on go-
ing from 1 to 3 (or from 2 to 6), but then stays constant on
going to 4. This reflects the statistical argument outlined in
the introduction: the coupling constants in compounds 3
and 4 relate to the average over two modules, in 1 they
are characteristic for a single module. The alteration of the
coupling constants is, however, somewhat lower for the two
outer modules (segment A) in 9 than for the inner ones
(segment B). A lower overall conformational preference (Ͻ
95%) for the two outer segments in 9 is also indicated by
the change in the coupling constants observed on lowering
the temperature by 85 °C.
Regarding segment B in 9, the coupling constants of 9.9
and 1.1 Hz suggest a high local conformer preference. The
increased alteration of the coupling constants on lowering
the temperature by 85°C indicates that the conformer pref-
erence at room temperature does not exceed 95%. Actually,
the conformational behaviour of segment B in 9 should be
compared to 7, as a model compound that also contains
one central benzylidene acetal. The coupling constants are
Eur. J. Org. Chem. 2002, 1292Ϫ1304
1297

T. Trieselmann, R. W. Hoffmann, K. Menzel
FULL PAPER
(s, 6 H), 1.27 (m, 2 H), 1.34 (s, 6 H), 1.35 (s, 6 H), 3.24 (d, J ϭ
11.3 Hz, 2 H), 3.50 (d, J ϭ 11.3 Hz, 2 H), 3.58 (m, 2 H). 13C NMR
(125 MHz, CDCl3): δ ϭ 18.0, 18.8, 21.6, 28.7, 30.0, 32.4, 72.0, 72.1,
98.4. C17H32O4 (300.2): calcd. C 67.96, H 10.74; found C 67.91,
H 10.96.
essentially the same in 7 and 9. These findings show that
the conformational preference in a given module is by and
large, but not completely, independent of the nature and
number of the neighbouring modules.
We next wanted to learn whether there was any evidence
for a lower conformational preference in modules incorpor-
ating benzylidene acetals versus those with acetonides, as
suggested by the force field calculations. The coupling con-
stants in the monobenzylidene acetal 7 were almost the
same as those of the tris(acetonide) 3. On going from the
monobenzylidene acetal 7 to the tris(benzylidene acetal) 6,
there is indeed a decrease in the alteration of the coupling
constants. It remains a moot point whether these changes
reflect a decrease in the conformer preference in the benzyl-
idene vs. the acetonide systems, as suggested by the force
field calculations, or whether this is caused by slightly dif-
ferent torsional angles at the inter-ring bonds in the benzyli-
dene and acetonide systems. The direction of such changes
is not uniform (cf. the differences in the coupling constants
between 2 and 3). There are therefore at least grounds for
scepticism regarding the validity of the force field calcula-
tions.
The aim of this study was to evaluate to what extent high
overall conformational preferences could be attained by
combining an increasing number of modules 1. We report
that on ‘‘combining’’ up to four bis(dioxanyl)methane units
1 (cf. compound 9) a high local conformer preference still
prevails, endowing compound 9 with a sizeable (ca. 90%)
overall conformational preference over eight rotatable
bonds.
2.
(4R*,4ЈR*)-Bis(2-phenyl-5,5-dimethyl-1,3-dioxan-4-yl)methane
(2): Camphorsulfonic acid (ca. 10 mg) was added to a solution of
compound 1 (0.19 g, 0.63 mmol) and benzaldehyde dimethyl acetal
(241 mg, 1.58 mmol) in dichloromethane (5 mL). The mixture was
heated to 40 °C at 25 mbar in a rotary evaporator for 3 h. Flash
chromatography of the residue with tert-butyl methyl ether/pentane
(1:10) furnished compound 2 (0.155 g, 62%) as a colourless oil. 1H
NMR (300 MHz, CDCl3): δ ϭ 0.79 (s, 6 H), 1.15 (s, 6 H),
1.60Ϫ1.63 (m, 2 H), 3.62 (d, J ϭ 11.1 Hz, 2 H), 3.73 (d, J ϭ
11.1 Hz, 2 H), 3.80Ϫ3.83 (m, 2 H), 5.50 (s, 2 H), 7.35Ϫ7.42 (m, 6
H), 7.51Ϫ7.53 (m, 4 H). 13C NMR (75 MHz, CDCl3): δ ϭ 18.8,
21.6, 29.6, 32.8, 79.1, 80.9, 101.7, 126.2, 128.3, 128.8, 139.2.
C25H32O4 (396.2): calcd. C 75.73, H 8.13; found C 75.41, H 8.18.
3. (4S*,6S*)-4,6-Bis(benzyloxy)-3,3,7,7-tetramethyl-1,8-nonadiene
(12): Sodium hydride (60% in white oil, 1.26 g, 31.5 mmol) was
added to a solution of the diol 10[33] (3.20 g, 15.0 mmol) in DMF
(45 mL) at 0 °C. After this mixture had been stirred for 30 min at
0 °C, benzyl bromide (3.80 mL, 31.5 mmol) was added and stirring
was continued for 1.5 h. Tetrabutylammonium iodide (ca. 20 mg)
was added and stirring was continued for 15 h at room temperature.
Water (50 mL) and saturated aqueous NH4Cl solution (50 mL)
were added. The phases were separated and the aqueous phase was
extracted with tert-butyl methyl ether (4 ϫ 60 mL). The combined
organic phases were dried (Na2SO4) and concentrated. Flash chro-
matography of the residue with pentane/tert-butyl methyl ether
(98:2) furnished the product 12 (4.20 g, 71%) as a slightly yellowish
oil. 1H NMR (200 MHz, CDCl3): δ ϭ 0.91 (s, 12 H), 1.50 (m, 2
H), 3.14 (m, 2 H), 4.44 (m, 4 H), 4.87 (m, 4 H), 5.66 (m, 2 H), 7.19
(m, 10 H). 13C NMR (50 MHz, CDCl3): δ ϭ 25.9, 26.6, 37.1, 45.3,
77.6, 87.6, 114.6, 129.8, 129.9, 131.1, 142.1, 148.9. C27H36O2
(392.6): calcd. C 82.61, H 9.24; found C 82.52, H 9.18.
Experimental Section
General Remarks: All temperatures quoted are uncorrected. 1H
NMR, 13C NMR: Bruker ARX 200, AC 300, WH 400, AM 400,
AMX 500. Boiling range of petroleum ether: 40Ϫ60 °C. Flash
chromatography: Silica gel SI 60, E. Merck KGaA, Darmstadt,
40Ϫ63 µm. Buffer (pH ϭ 7): NaH2PO4·2H2O (56.2 g) and
Na2HPO4·4H2O (213.6 g) made up to 1 L with water.
4.
(2RS*,4S*,6S*,8RS*)-4,6-Bis(benzyloxy)-3,3,7,7-tetramethyl-
nonane-2,8-diol (14): A stream of ozone was introduced at Ϫ78 °C
to a solution of compound 12 (1.15 g, 2.90 mmol) in dichlorome-
thane (10 mL) until the blue colour persisted. Excess ozone was
purged with oxygen, and triphenylphosphane (4.60 g, 11.7 mmol)
1.
[(4R*,4ЈR*)-2,2,2Ј,2Ј,5,5,5Ј,5Ј-Octamethyl-4,4Ј-methanediylbis(1,3- over 3 h. Silica gel (5 g) was added, and the suspension was concen-
dioxane)] (1): A stream of ozone was introduced at Ϫ78 °C to a trated. The remaining solid was placed on top of a chromatography
solution of the diol 10[33] (1.06 g, 5.00 mmol) in methanol (10 mL) column. Flash chromatography with pentane/tert-butyl methyl
until the blue colour persisted. Excess ozone was purged with nitro- ether (9:1) furnished the bis(aldehyde) 13 (1.07 g, 85%) as a colour-
gen, and NaBH4 (1.51 g, 40.0 mmol) was added. The mixture was less oil. H NMR (200 MHz, CDCl3): δ ϭ 1.04 (s, 6 H), 1.10 (s, 6
(4R*,4ЈR*)-Bis(2,2,5,5-tetramethyl-1,3-dioxan-4-yl)methane was added. The mixture was allowed to reach room temperature
1
allowed to reach room temperature over 15 h. Aqueous hydro-
chloric acid (1 , 40 mL) was added, and the phases were separ-
ated. The aqueous phase was extracted with ethyl acetate (5 ϫ
H), 1.60 (m, 2 H), 3.99 (dd, J ϭ 6.5, 4.2 Hz, 2 H), 4.43 (d, J ϭ
11.3 Hz, 2 H), 4.57 (d, J ϭ 11.3 Hz, 2 H), 7.38 (m, 10 H), 9.55 (s,
2 H). 13C NMR (50 MHz, CDCl3): δ ϭ 20.1, 43.1, 45.8, 73.0, 78.0,
50 mL). The combined organic phases were washed with brine 127.7, 127.9, 128.4, 138.1, 201.2. A solution of methylmagnesium
(50 mL), dried (Na2SO4), and concentrated. The residue was taken chloride (2.68  in THF, 9.0 mL, 24 mmol) was added at Ϫ20 °C
up in a mixture of acetone and 2,2-dimethoxypropane (10:1; over 30 min to a solution of the bis(aldehyde) 13 (3.5 g, 8.9 mmol)
20 mL). p-Toluenesulfonic acid (ca. 10 mg) was added, and the mix-
ture was stirred for 2 d at room temperature. Saturated aqueous
in THF (18 mL). After stirring for 5 min, the mixture was allowed
to reach room temperature and poured into saturated aqueous
K2CO3 solution (50 mL) was added, the phases were separated, NH4Cl solution (50 mL). The phases were separated and the aque-
and the aqueous phase was extracted with pentane (3 ϫ 50 mL). ous phase was extracted with tert-butyl methyl ether (5 ϫ 30 mL).
The combined organic phases were dried (MgSO4) and concen-
trated. Flash chromatography of the residue with tert-butyl methyl
ether/pentane (1:10) furnished compound 1 (1.14 g, 75%) as a col-
The combined organic phases were dried (Na2SO4) and concen-
trated to give the diol 14 (3.8 g, quantitative) as a 1.2:1 mixture of
diastereomers. 1H NMR (200 MHz, CDCl3): δ ϭ 0.81 (m, 6 H),
ourless oil. 1H NMR (500 MHz, CDCl3): δ ϭ 0.67 (s, 6 H), 0.93 0.98 (m, 3 H), 1.10 (m, 6 H), 1.25 (m, 3 H), 1.83 (m, 2 H), 3.74
1298
Eur. J. Org. Chem. 2002, 1292Ϫ1304

Conformational Analysis of oligo-1,3-Dioxanylmethanes
FULL PAPER
(m, 2 H), 3.90 (m, 2 H), 4.60 (m, 4 H), 7.32 (m, 10 H). C27H40O4
(428.6): calcd. C 75.66, H 9.41; found C 75.62, H 9.19.
with dichloromethane (5 ϫ 40 mL) and the combined organic
phases were dried (Na2SO4). Cyclohexane (5 mL) was added, and
the solution was concentrated. Flash chromatography with pent-
ane/tert-butyl ether (1:1) furnished the tetraol 18 (213 mg, 80%)
5. (4S*,6S*)-4,6-Bis(benzyloxy)-3,3,7,7-tetramethylnonane-2,8-di-
one (15): A solution of dimethyl sulfoxide (3.70 mL, 51.8 mmol) in
dichloromethane (6 mL) was added dropwise at Ϫ78 °C to a solu-
tion of oxalyl chloride (2.26 mL, 25.8 mmol) in dichloromethane
(10 mL). After this mixture had been stirred for 45 min, a solution
of the diol 14 (3.70 g, 8.60 mmol) in dichloromethane (21 mL) was
added. After this in turn had been stirred for a further 70 min at
Ϫ78 °C, triethylamine (8.30 mL, 60.2 mmol) was added dropwise,
resulting in the formation of a precipitate. The mixture was stirred
for 30 min at Ϫ78 °C and allowed to reach room temperature over
90 min. It was poured into saturated aqueous NH4Cl solution
(150 mL). The phases were separated and the aqueous phase was
extracted with tert-butyl methyl ether (3 ϫ 50 mL). The combined
organic phases were washed with aqueous hydrochloric acid (1 ,
100 mL) and brine (100 mL), dried with Na2SO4, and concen-
trated. Flash chromatography of the residue with pentane/tert-butyl
1
as a 4:1 diastereomer mixture and as a yellowish resin. H NMR
(400 MHz, CDCl3): δ ϭ 0.82 (s, 6 H), 0.89 (s, 6 H), 0.90 (s, 6 H),
0.97 (s, 6 H), 1.50 (m, 4 H), 1.91 (m, 2 H), 3.33 (m, 6 H), 3.67 (m,
2 H), 3.89 (m, 2 H), 4.47 (m, 4 H), 4.65 (m, 4 H), 7.30 (m, 20 H).
13C NMR (125 MHz, CDCl3): δ ϭ 19.1, 19.3, 19.6, 22.4, 32.8, 33.2,
38.0, 42.8, 72.2, 73.2, 74.4, 74.5, 79.2, 83.8, 127.1, 127.3, 128.0,
128.2, 137.7, 137.8, 138.6, 138.7. C51H72O8 (813.1): calcd. C 75.33,
H 8.93; found C 75.27, H 8.93.
8. (4R*,6S*)-2,2,5,5-Tetramethyl-4-[(4R*)-2,2,5,5-tetramethyl-1,3-
dioxan-4-ylmethyl]-6-{(4S*,6R*)-2,2,5,5-tetramethyl-6-[(4R*)-
2,2,5,5-tetramethyl-1,3-dioxan-4-ylmethyl]}-1,3-dioxan-4-ylmethyl]-
1,3-dioxane (4): Palladium hydroxide on carbon (ca. 30 mg) was
added to
a solution of the tetrabenzyl ether 18 (213 mg,
0.262 mmol) in THF (3 mL). The mixture was stirred under 1 bar
methyl ether (4:1 to 1.5:1) furnished the diketone 15 (3.17 g, 86%) of hydrogen for 1 d. Dichloromethane (10 mL) was added, and the
1
as a yellowish oil. H NMR (200 MHz, CDCl3): δ ϭ 1.04 (m, 12 mixture was filtered through kieselguhr. The filtrate was concen-
H), 1.21 (m, 6 H), 1.45 (m, 2 H), 3.85 (m, 2 H), 4.54 (m, 4 H), 7.38 trated to give the octaol 19 (95 mg, 80%) as a yellow resin. 1H
(m, 10 H). 13C NMR (50 MHz, CDCl3): δ ϭ 20.7, 26.3, 34.3, 52.7, NMR (500 MHz, CH3OD): δ ϭ 0.74 (s, 6 H), 0.83 (s, 3 H), 0.85
74.4, 74.9, 81.7, 127.1, 127.2, 128.8, 138.4, 212.9. C27H36O4 (424.6): (s, 3 H), 0.89 (s, 6 H), 1.50 (m, 6 H), 3.34 (d, J ϭ 10.8 Hz, 2 H),
calcd. C 76.38, H 8.55; found C 76.54, H 8.34.
3.42 (d, J ϭ 10.8 Hz, 2 H), 3.67 (dd, J ϭ 10.6, 1.7 Hz, 2 H), 3.77
(m, 4 H). 13C NMR (50 MHz, CH3OD): δ ϭ 19.5, 20.1, 21.0, 21.7,
34.0, 34.1, 34.3, 41.4, 42.6, 71.0, 74.4, 74.5, 75.7. The octaol 19
(25 mg, 0.055 mmol) was taken up in DMF (0.5 mL), and 2-me-
thoxypropene (80 mg, 1.1 mmol) and p-toluenesulfonic acid (ca.
10 mg) were added. After this mixture had been stirred for 15 h
at room temperature, water (10 mL) was added. The phases were
separated, and the aqueous phase was extracted with diethyl ether
(5 ϫ 10 mL). The combined organic phases were dried (Na2SO4)
and concentrated. Flash chromatography with pentane/tert-butyl
methyl ether (9:1) furnished compound 4 (22 mg, 65%), diastere-
omerically pure and as a colourless resin. 13C NMR (100 MHz,
CDCl3): δ ϭ 13.0, 18.1, 19.0, 19.4, 20.5, 21.7, 28.5, 28.6, 29.7, 30.3,
32.6, 35.1, 72.4, 72.6, 73.3, 73.7, 98.2, 98.5. For the other physical
data see ref.[14]
6. (3R*,7S*,9S*,13R*)-1,7,9,15-Tetrakis(benzyloxy)-3,13-dihy-
droxy-2,2,6,6,10,10,14,14-octamethylpentadecane-5,11-dione (17):
Triethylamine (230 µL, 2.00 mmol) was added at 0 °C to a solution
of chlorodicyclohexylborane (320 µL, 1.50 mmol) in diethyl ether
(2.5 mL). After addition of a solution of the diketone 15 (212 mg,
0.50 mmol) in diethyl ether (1 mL), a white precipitate formed.
After stirring for 3 h at 0 °C, the mixture was cooled to Ϫ90 °C.
A
solution of 3-benzyloxy-2,2-dimethyl-1-propanal (16)[18]
(384 mg, 2.00 mmol) in diethyl ether (1 mL) was added dropwise.
After stirring for 3 d at Ϫ90 °C, the mixture was poured into a
mixture of buffer (pH ϭ 7; 5 mL), methanol (5 mL), and water
(2 mL). Hydrogen peroxide (30% in water, 5 mL) was added at 0
°C and the mixture was stirred for 3 h. The phases were separated,
and the aqueous phase was extracted with tert-butyl methyl ether
(5 ϫ 50 mL). The combined organic phases were dried (Na2SO4)
and concentrated. Flash chromatography with pentane/tert-butyl
9. (2R*,4R*,6S*)-5,5-Dimethyl-4-[(2S*,4R*)-5,5-dimethyl-2-phenyl-
1,3-dioxan-4-ylmethyl]-6-{(2R*,4S*,6R*)-6-[(2S*,4R*)-5,5-
methyl ether (9:1 to 1.5:1) provided the bis(aldol) 17 (263 mg, 65%) dimethyl-2-phenyl-1,3-dioxan-4-ylmethyl]-5,5-dimethyl-2-phenyl-1,3-
as a 4:1 diastereomer mixture. Major diastereomer: 1H NMR
dioxan-4-ylmethyl}-2-phenyl-1,3-dioxane (8): Benzaldehyde di-
(400 MHz, CDCl3): δ ϭ 0.78 (s, 6 H), 0.84 (s, 6 H), 1.07 (m, 6 H), methyl acetal (42 mg, 0.28 mmol) and camphorsulfonic acid (ca.
1.22 (m, 6 H), 1.51 (m, 2 H), 1.71 (broad signal, 2 H), 1.72 (m, 2 10 mg) were added to a solution of the octaol 19 (60 mg,
H), 1.89 (m, 2 H), 2.70 (m, 2 H), 3.26 (m, 2 H), 3.60 (m, 2 H), 3.91 0.074 mmol) in dichloromethane (1 mL). The mixture was heated
(m, 2 H), 4.43 (m, 4 H), 4.52 (m, 4 H), 7.28 (m, 20 H). 13C NMR to 40 °C at 25 mbar for 4 h in a rotary evaporator. Flash chromato-
(50 MHz, CDCl3): δ ϭ 19.9, 20.5, 21.2, 21.9, 34.3, 38.1, 40.7, 53.1, graphy of the residue with pentane/tert-butyl methyl ether (9:1) fur-
72.7, 73.3, 74.6, 78.0, 81.9, 127.3, 127.4, 127.5, 128.2, 128.3, 138.4, nished diastereomerically pure 8 (28 mg, 62%) as a yellowish oil.
138.5, 216.1. C51H68O8 (809.1): calcd. C 75.71, H 8.47; found C
75.36, H 8.23.
1H NMR (500 MHz, CDCl3): δ ϭ 0.79 (s, 6 H), 0.81 (s, 6 H), 0.98
(s, 6 H), 1.14 (s, 6 H), 1.65 (m, 2 H), 1.68 (m, 4 H), 3.63 (d, J ϭ
11.2 Hz, 2 H), 3.73 (d, J ϭ 11.2 Hz, 2 H), 3.77 (m, 4 H), 3.82 (m,
2 H), 5.51 (s, 2 H), 5.58 (s, 2 H), 7.36 (m, 12 H), 7.51 (m, 8 H).
13C NMR (125 MHz, CDCl3): δ ϭ 14.0, 18.7, 20.8, 21.6, 25.3, 29.4,
32.7, 35.3, 79.0, 81.3, 81.4, 82.0, 100.7, 101.5, 126.0, 126.1, 128.0,
128.1, 128.4, 128.5, 139.0, 139.4. C51H64O8 (exact mass, FAB):
calcd. 804.4601; found 804.4619.
7.
(3R*,4R*,7S*,9S*,11R*,13R*)-1,7,9,15-Tetrakis(benzyloxy)-
2,2,6,6,10,10,14,14-octamethylpentadecane-3,5,11,13-tetraol (18):
Acetic acid (3.20 mL, 54.8 mmol) was added to a suspension of
tetramethylammonium
tris(acetoxy)borohydride
(7.21 g,
27.4 mmol) in 4 mL of acetone. The resulting clear solution was
cooled to 0 °C, and a solution of the bis(aldol) 17 (263 mg,
0.325 mmol) in acetone (4 mL) was added. After this mixture had 10. (4R*,6S*)- and (4R*,6R*)-5,5-Dimethyl-1,8-nonadiene-4,6-diol
been stirred for 1 h at 0 °C and 24 h at room temperature, saturated
aqueous sodium potassium tartrate solution (3 mL) was added, re-
sulting in vigorous gas evolution. After stirring for 20 min, the mix-
(21, 22): Dimethylmalondialdehyde (20) (9.00 g, 88.1 mmol) was
added to a solution of allyl chloride (23.0 mL, 282 mmol), tin
dichloride dihydrate (79.5 g, 352 mmol), and sodium iodide (58.1 g,
ture was poured into saturated aqueous NaHCO3 solution (40 mL) 388 mmol) in DMF (440 mL). After stirring for 6 d, the mixture
and the phases were separated. The aqueous phase was extracted
was poured into 10% aqueous NH4F solution (500 mL). The mix-
1299
Eur. J. Org. Chem. 2002, 1292Ϫ1304

T. Trieselmann, R. W. Hoffmann, K. Menzel
FULL PAPER
ture was extracted with tert-butyl methyl ether (5 ϫ 100 mL). The in diethyl ether/THF (9:1; 1 mL) was added dropwise by cannula
combined organic phases were dried (Na2SO4) and concentrated. at Ϫ120 °C to a solution of lithium aluminium hydride (0.096 g,
Flash chromatography with pentane/tert-butyl methyl ether (7:3) 2.53 mmol) in diethyl ether/THF (9:1; 25 mL). After stirring for 6 h
furnished a 2:1 mixture of 21 and 22 (11.1 g, 69%) as a colourless
at Ϫ120 °C, the mixture was allowed to reach Ϫ60 °C over 18 h.
liquid. For analysis a sample was distilled at 104 °C/5 mbar. Saturated aqueous NH4Cl solution (20 mL) was added, the phases
C11H20O2 (184.3): calcd. C 71.70, H 10.94; found C 71.57, H 11.04. were separated, and the aqueous phase was extracted with tert-
1
21: H NMR (300 MHz, CDCl3): δ ϭ 0.79 (s, 3 H), 0.91 (s, 3 H), butyl methyl ether (4 ϫ 20 mL). The combined organic phases were
1.97Ϫ2.45 (m, 4 H), 2.99 (broad s, 2 H), 3.51Ϫ3.62 (m, 2 H),
dried (Na2SO4) and concentrated. Flash chromatography of the
5.07Ϫ5.19 (m, 4 H), 5.75Ϫ5.97 (m, 4 H). 13C NMR (75 MHz, residue with pentane/tert-butyl ether (20:1) gave 24 (0.032 g, 14%)
CDCl3): δ ϭ 14.8, 20.9, 21.1, 36.6, 40.7, 78.1, 117.8, 136.1. The and 23 (0.190 g, 82%) as colourless oils.
following signals of 22 could be recorded: δ ϭ 21.1, 36.5, 40.1,
14. (4R*,6S*)-4,6-Dibenzyloxy-5,8-dimethyl-1,8-nonadiene (25):
Benzyl trichloroacetimidate (0.90 mL, 3.8 mmol) was added drop-
77.4, 117.6, 136.2.
11. (4R*,6S*)- and (4R*,6R*)-6-Benzyloxy-5,5-dimethyl-1,8-nona-
dien-4-ol (23, 24): Sodium hydride (80% in white oil, 0.100 g,
3.44 mmol) was added in small portions at 0 °C to a solution of
21 and 22 (2:1, 577 mg, 3.13 mmol) in DMF (0.73 mL) and THF
(8 mL). After this mixture had been stirred for 30 min, benzyl
bromide (0.39 mL, 3.3 mmol) was added dropwise. Tetrabutylam-
monium iodide (ca. 20 mg) was added, and the mixture was al-
lowed to reach room temperature over 15 h. It was poured into
saturated aqueous NH4Cl solution (20 mL). The phases were sep-
arated, and the aqueous phase was extracted with tert-butyl methyl
ether (4 ϫ 20 mL). The combined organic phases were dried
(Na2SO4) and concentrated. Flash chromatography of the residue
with pentane/tert-butyl methyl ether (95:5) furnished 24 (0.24 g,
28%) and 23 (0.485 g, 57%) as colourless oils. 24: 1H NMR
(400 MHz, CDCl3): δ ϭ 0.80 (s, 3 H), 0.94 (s, 3 H), 1.96 (m, 1 H),
2.12 (m, 1 H), 2.36 (m, 2 H), 3.35 (dd, J ϭ 7.6, 4.0 Hz, 1 H), 3.58
(d, J ϭ 2.2 Hz, 1 H), 3.63 (dt, J ϭ 10.2, 2.5 Hz, 1 H), 4.43 (d, J ϭ
10.9 Hz, 1 H), 4.62 (d, J ϭ 10.9 Hz, 1 H), 5.00 (m, 4 H), 5.87 (m,
2 H), 7.22 (m, 5 H). 13C NMR (100 MHz, CDCl3): δ ϭ 20.1, 22.1,
35.3, 36.4, 41.6, 74.5, 75.1, 88.2, 116.4, 116.8, 127.5, 127.6, 127.7,
wise to a solution of 23 (0.52 g, 1.9 mmol) in diethyl ether (2 mL).
Trifluoromethanesulfonic acid (1 drop) was added, and the mixture
was stirred for 15 h. The mixture was poured into water (10 mL)
and the phases were separated. The aqueous phase was extracted
with tert-butyl methyl ether (4 ϫ 10 mL). The combined organic
phases were dried (Na2SO4) and concentrated. Flash chromato-
graphy of the residue with pentane/tert-butyl methyl ether (98:2)
furnished 25 (0.64 g, 92%) as a colourless oil. 1H NMR (200 MHz,
CDCl3): δ ϭ 0.95 (s, 3 H), 1.10 (s, 3 H), 2.36 (m, 2 H), 2.43 (m, 2
H), 3.49 (dd, J ϭ 8.1, 3.2 Hz, 2 H), 4.56 (d, J ϭ 11.3 Hz, 2 H),
4.71 (d, J ϭ 11.3 Hz, 2 H), 5.10 (m, 4 H), 5.98 (m, 2 H), 7.37 (m,
10 H). 13C NMR (50 MHz, CDCl3): δ ϭ 19.6, 20.0, 35.8, 44.2,
73.9, 84.0, 116.2, 127.3, 127.4, 128.2, 137.4, 139.0. C25H32O2
(274.4): calcd. C 82.37, H 8.85; found C 82.47, H 8.81.
15. (3R*,5S*)-3,5-Bis(benzyloxy)-4,4-dimethylheptane-1,7-dial (27):
The diene 25 (0.365 g, 1.00 mmol) was ozonised as described under
2. Flash chromatography of the crude product with pentane/tert-
butyl methyl ether (7:3) furnished the product 27 (0.32 g, 88%) as
1
a colourless oil, which was used as obtained. H NMR (200 MHz,
1
CDCl3): δ ϭ 0.91 (s, 3 H), 1.04 (s, 3 H), 2.69 (m, 4 H), 3.99 (dd,
J ϭ 6.6, 4.2 Hz, 2 H), 4.43 (d, J ϭ 11.3 Hz, 2 H), 4.57 (d, J ϭ
11.3 Hz, 2 H), 7.31 (m, 10 H), 9.76 (t, J ϭ 1.7 Hz, 2 H). 13C NMR
(50 MHz, CDCl3): δ ϭ 20.1, 43.1, 45.8, 73.0, 78.0, 127.7, 127.9,
128.4, 138.0, 201.2.
136.6, 136.8, 137.9. 25: H NMR (200 MHz, CDCl3): δ ϭ 0.87 (s,
3 H), 1.01 (s, 3 H), 2.05 (m, 1 H), 2.33 (m, 2 H), 2.54 (m, 1 H),
2.67 (s, 1 H), 3.48 (dd, J ϭ 7.4, 3.7 Hz, 1 H), 3.56 (dd, J ϭ 10.2,
1.2 Hz, 1 H), 4.60 (d, J ϭ 11.1 Hz, 1 H), 4.92 (d, J ϭ 11.1 Hz, 1
H), 5.10 (m, 2 H), 5.14 (m, 2 H), 5.92 (m, 2 H), 7.33 (m, 5 H). 13
C
NMR (50 MHz, CDCl3): δ ϭ 17.4, 20.4, 35.5, 36.6, 42.7, 73.5,
76.0, 86.0. 116.3, 117.4, 127.5, 127.6, 128.3, 136.5, 137.1, 138.4.
C18H26O3 (274.4): calcd. C 78.79, H 9.55; found C 78.61, H 9.40.
16. Diethyl (3R*,5R*,7S*,9S*)-5,7-Bis(benzyloxy)-3,9-dihydroxy-
2,2,6,6,10,10-hexamethylundecanedioate (29): BF3·OEt2 (62 µL,
0.49 mmol) was added dropwise at Ϫ90 °C over 15 min to a solu-
tion of the dialdehyde 27 (90 mg, 0.24 mmol) and 1-ethoxy-2-
methyl-1-(trimethylsilyloxy)propene[35] in 1 mL of dichlorome-
thane. After this had been stirred for 20 h at Ϫ90 °C, saturated
aqueous NaHCO3 solution (10 mL) was added, the phases were
separated, and the aqueous phase was extracted with tert-butyl
methyl ether (5 ϫ 10 mL). The combined organic phases were dried
(Na2SO4) and concentrated. Flash chromatography with pentane/
tert-butyl methyl ether (7:3) furnished the bis(aldol) products 29
and 28 as a 6:1 mixture. C35H52O8 (600.8): calcd. C 69.97, H 8.72;
12.
6-Benzyloxy-5,5-dimethyl-1,8-nonadien-4-one
(26):
DessϪMartin periodinane[34] (0.575 g, 1.36 mmol) was added to a
solution of 24 (0.248 g, 0.904 mmol) and pyridine (0.219 mL,
2.71 mmol) in dichloromethane (4.5 mL). After stirring for 2 h at
room temperature, the mixture was poured into a combination of
saturated aqueous NaHCO3 solution (10 mL), saturated aqueous
Na2S2O3 solution (10 mL) and tert-butyl methyl ether (20 mL).
After the mixture had been stirred vigorously for 30 min, the
phases were separated, and the aqueous phase was extracted with
tert-butyl methyl ether (3 ϫ 20 mL). The combined organic phases
were dried (Na2SO4) and concentrated. Flash chromatography of
the residue with pentane/tert-butyl methyl ether (98:2) furnished
the ketone 26 (0.236 g, 95%) as a colourless oil. 1H NMR
(200 MHz, C6D6): δ ϭ 1.00 (s, 3 H), 1.11 (s, 3 H), 2.20 (m, 2 H),
3.01 (dt, J ϭ 6.7, 1.4 Hz, 2 H), 3.84 (dd, J ϭ 7.1, 4.7 Hz, 1 H),
4.31 (d, J ϭ 11.2 Hz, 1 H), 4.55 (d, J ϭ 11.2 Hz, 1 H), 5.04 (m, 4
H), 5.85 (m, 1 H), 6.13 (m, 1 H), 7.24 (m, 5 H). 13C NMR
(50 MHz, C6D6): δ ϭ 20.6, 21.2, 36.0, 43.4, 52.5, 74.0, 84.2, 116.8,
117.9, 127.4, 127.5, 128.2, 131.5, 136.1, 138.5, 211.2. C18H24O2
(272.4): calcd. C 79.37, H 8.88; found C 79.35, H 8.99.
1
found C 70.15, H 8.92. 29: H NMR (400 MHz, CDCl3): δ ϭ 0.92
(s, 3 H), 1.07 (s, 3 H), 1.13 (s, 6 H), 1.17 (s, 6 H), 1.23 (t, J ϭ
7.0 Hz, 6 H), 1.57 (m, 2 H), 1.65 (m, 2 H), 2.67 (d, J ϭ 7.0 Hz, 2
H), 3.78 (m, 4 H), 5.13 (q, J ϭ 7.0 Hz, 4 H), 4.70 (s, 4 H), 7.35 (m,
10 H). 13C NMR (100 MHz, CDCl3): δ ϭ 14.1, 19.8, 20.3, 21.5,
22.8, 33.3, 44.7, 46.8, 60.7, 73.7, 74.9, 81.0, 127.2, 127.6, 128.2,
139.4, 178.0.
17.
(3R*,5R*,7S*,9S*)-5,7-Bis(benzyloxy)-2,2,6,6,10,10-hexa-
methylundecane-1,3,9,11-tetraol (30): A solution of diisobutylalum-
inium hydride (1  in petroleum ether, 1.5 mL, 1.5 mmol) was ad-
ded dropwise at Ϫ78 °C to a solution of the diesters 28 and 29
(90 mg, 0.15 mmol) in dichloromethane (1.5 mL). After stirring for
20 h at Ϫ78 °C, the mixture was kept for 20 h at room temperature.
13. (4R*,6S*)-6-Benzyloxy-5,5-dimethyl-1,8-nonadien-4-ol (23): A
pre-cooled (Ϫ120 °C) solution of ketone 26 (0.230 g, 0.844 mmol)
1300
Eur. J. Org. Chem. 2002, 1292Ϫ1304

Conformational Analysis of oligo-1,3-Dioxanylmethanes
FULL PAPER
Hydrochloric acid (1 , 10 mL) was added, the phases were separ- tert-butyl methyl ether/pentane (7:3) furnished diastereomerically
ated, and the aqueous phase was extracted with tert-butyl methyl
ether (5 ϫ 10 mL). The combined organic phases were dried
pure 34 (0.90 g, 65%) as a colourless oil. 1H NMR (400 MHz,
CDCl3): δ ϭ 0.78 (s, 3 H), 0.90 (s, 3), 1.16 (s, 6 H), 1.21 (s, 6 H),
(Na2SO4) and concentrated. Flash chromatography of the residue 1.24 (t, J ϭ 7.1 Hz, 6 H), 1.47 (ddd, J ϭ 13.9, 10.1, and 1.5 Hz, 2
with ethyl acetate furnished the tetraol 30 (53 mg, 69%) as a colour-
H), 1.58 (ddd, J ϭ 13.9, 10.1, and 1.5 Hz, 2 H), 3.79 (s, 3 H), 3.81
(dd, J ϭ 10.1, 1.5 Hz, 2 H), 3.87 (m, 2 H), 4.15 (q, J ϭ 7.1 Hz, 4
less oil, which was used as obtained. 1H NMR (200 MHz, CDCl3):
δ ϭ 0.83 (s, 6 H), 0.85 (s, 6 H), 0.99 (s, 3 H), 1.11 (s, 3 H), 1.68 H), 5.56 (s, 1 H), 6.86 (m, 2 H), 7.40 (m, 2 H). 13C NMR
(m, 4 H), 3.40 (d, J ϭ 10.7 Hz, 2 H), 3.52 (d, J ϭ 10.7 Hz, 2 H),
(100 MHz, CDCl3): δ ϭ 14.0, 14.1, 20.5, 22.5, 31.5, 35.0, 46.7, 55.2,
3.66 (m, 2 H), 3.74 (dd, J ϭ 8.8, 1.8 Hz, 2 H), 4.69 (s, 4 H), 7.35 60.6, 72.8, 81.9, 100.6, 113.3, 127.2, 131.9, 159.5, 177.7. C29H46O9
(m, 10 H). 13C NMR (50 MHz, CDCl3): δ ϭ 19.1, 20.1, 21.8, 22.4, (600.8): calcd. C 64.66, H 8.61; found C 64.49, H 8.70.
33.1, 38.4, 44.2, 72.3, 74.7, 76.4, 81.4, 127.5, 127.6, 128.3, 139.2.
21.
(2s*,4R*,6S*)-2-(4-Methoxyphenyl)-4-[(4R*)-2-(4-methoxy-
18. (4R*,5S*)-2,2,5,5-Tetramethyl-4-[(4R*)-2,2,5,5-tetramethyl-1,3-
dioxan-4-ylmethyl]-6-[(4S*)-2,2,5,5-tetramethyl-1,3-dioxan-4-yl-
methyl]-1,3-dioxane (3): Palladium hydroxide on carbon (ca. 30 mg)
was added to a solution of the tetraol 30 (40 mg, 0.074 mmol) in
THF (1 mL). The mixture was stirred under 1 bar hydrogen for
35 h. Dichloromethane (10 mL) was added, and the mixture was
filtered through kieselguhr. Concentration of the filtrate resulted in
a crystalline solid (the hexaol). The residue was taken up in DMF
(0.5 mL), and 2-methoxypropene (48 mg, 0.67 mmol) was added,
followed by p-toluenesulfonic acid (ca. 10 mg). The mixture was
stirred for 4 h at room temperature. Buffer solution (pH ϭ 7;
10 mL) was added. The phases were separated, and the aqueous
phase was extracted with dichloromethane (5 ϫ 10 mL). The com-
bined organic phases were dried (Na2SO4) and concentrated. Flash
chromatography of the residue with pentane/tert-butyl methyl ether
(9:1) furnished diastereomerically pure 3 (28 mg, 80%) as a colour-
less resin. For the spectroscopic and physical data, see ref.[14]
phenyl)-5,5-dimethyl-1,3-dioxan-4-ylmethyl]-6-[(4S*)-2-(4-methoxy-
phenyl)-5,5-dimethyl-1,3-dioxan-4-ylmethyl]-5,5-dimethyl-1,3-
dioxane (6): Lithium borohydride (58 mg, 2.67 mmol) was added at
room temperature to a solution of the diester 34 (240 mg,
0.446 mmol) in dichloromethane (2.2 mL). After this had been
stirred for 18 h, saturated aqueous NH4Cl solution (10 mL) was
added. The phases were separated and the aqueous phase was ex-
tracted with ethyl acetate (6 ϫ 10 mL). The combined organic
phases were dried (Na2SO4) and concentrated. Flash chromato-
graphy with ethyl acetate/methanol (98:2) furnished the tetraol 35
1
(182 mg, 90%) as a colourless oil, which was used as obtained. H
NMR (500 MHz, CDCl3): δ ϭ 0.83 (s, 6 H), 0.84 (s, 6 H), 0.92 (s,
3 H), 0.97 (s, 3 H), 1.57 (s, 4 H), 3.46 (d, J ϭ 10.8 Hz, 2 H), 3.55
(d, J ϭ 10.8 Hz, 2 H), 3.70 (m, 2 H), 3.78 (s, 3 H), 3.89 (m, 2 H),
5.55 (s, 1 H), 6.88 (m, 2 H), 7.39 (m, 2 H). 13C NMR (50 MHz,
CH3OD): δ ϭ 19.9, 20.9, 21.8, 22.3, 32.3, 36.2, 40.0, 55.7, 70.8,
74.9, 83.7, 102.6, 115.3, 128.7, 133.0, 161.3. The obtained tetraol
35 and p-methoxybenzaldehyde diethyl acetal (279 mg, 1.33 mmol)
were taken up in dichloromethane (5 mL). After addition of cam-
phorsulfonic acid (ca. 20 mg), the mixture was heated to 40 °C at
25 mbar in a rotary evaporator for 4 h. Flash chromatography of
the residue with pentane/tert-butyl methyl ether (9:1) furnished 6
(184 mg, 67%) as a yellowish oil. 1H NMR (500 MHz, C6D6): δ ϭ
0.50 (s, 6 H), 0.68 (s, 3 H), 1.06 (s, 3 H), 1.15 (s, 6 H), 1.69 (m, 4
H), 3.35 (s, 6 H), 3.36 (s, 3 H), 3.33 (d, J ϭ 11.1 Hz, 2 H), 3.54 (d,
J ϭ 11.1 Hz, 2 H), 3.84 (m, 2 H), 3.89 (m, 2 H), 5.47 (s, 2 H), 5.65
(s, 1 H), 6.79 (m, 4 H), 6.89 (m, 2 H), 7.61 (m, 4 H), 7.66 (m, 2
H). 13C NMR (100 MHz, CDCl3): δ ϭ 13.9, 18.6, 20.5, 21.5, 29.5,
32.6, 35.2, 55.3, 78.9, 81.1, 81.2, 100.5, 101.5, 113.5, 113.6, 127.2,
127.3, 131.5, 132.0, 159.7, 159.8. C41H54O9 (690.4): calcd. C 71.28,
H 7.88; found C 71.43, H 8.11.
19.
(2s*,4R*,6S*)-4,6-Diallyl-2-(4-methoxyphenyl)-5,5-dimethyl-
1,3-dioxane (31): Camphorsulfonic acid (ca. 20 mg) was added to
a solution of the diols 21 and 22 (ca. 2:1, 2.97 g, 16.1 mmol) and
p-methoxybenzaldehyde diethyl acetal (3.23 g, 15.4 mmol) in
dichloromethane (5 mL). The mixture was heated to 40 °C at
25 mbar in a rotary evaporator. Flash chromatography of the res-
idue with pentane/tert-butyl methyl ether (98:2) furnished com-
pound 31 (3.29 g, quantitative relative to 21) as a yellowish oil. 1H
NMR (200 MHz, CDCl3): δ ϭ 0.83 (s, 3 H), 0.98 (s, 3 H), 2.27 (m,
4 H), 3.45 (dd, J ϭ 6.7, 5.5 Hz, 2 H), 3.78 (s, 3 H), 5.06 (m, 4 H),
5.50 (s, 1 H), 5.96 (m, 2 H), 6.87 (m, 2 H), 7.42 (m, 2 H). 13C NMR
(50 MHz, CDCl3): δ ϭ 13.8, 20.7, 33.7, 35.7, 55.2, 85.7, 100.8,
113.4, 116.0, 127.2, 131.5, 136.3, 159.5. C19H26O3 (302.4): Cald. C
75.46, H 8.67; found C 75.48, H 8.44.
20. Ethyl (3R*)-4-{(2s*,4R*,6S*)-6-[(2S*)-3-Ethoxycarbonyl-2-hy-
22.
(2s*,4R*,6S*)-2-(4-Methoxyphenyl)-5,5-dimethyl-4-[(4R*)-
droxy-3-methylbutyl]-2-(4-methoxyphenyl)-5,5-dimethyl-1,3-dioxan- 2,2,5,5-tetramethyl-1,3-dioxan-4-ylmethyl]-6-[(4S*)-2,2,5,5-tetra-
4-yl}-3-hydroxy-2,2-dimethylbutanoate (34): The diene 31 (0.85 g,
2.8 mmol) was ozonised as described under 2. Flash chromato-
methyl-1,3-dioxan-4-ylmethyl]-1,3-dioxane (7): 2-Methoxypropene
(48 mg, 0.66 mmol) and p-toluenesulfonic acid (ca 5 mg) were ad-
graphy with pentane/tert-butyl methyl ether (9:1 to 3:7) furnished ded at 0 °C to a solution of the tetraol 35 (60 mg, 0.13 mmol) in
the dialdehyde 32 (0.79 g, 92%) as a colourless oil, which was used
as obtained. H NMR (200 MHz, CDCl3): δ ϭ 0.78 (s, 3 H), 1.00
DMF (1 mL). After this had been stirred for 8 h, buffer (pH ϭ 7)
solution (10 mL) was added. The phases were separated, and the
1
(s, 3 H), 2.48 (m, 2 H), 2.65 (ddd, J ϭ 16.6, 9.5, and 1.9 Hz, 2 H), aqueous phase was extracted with dichloromethane (5 ϫ 10 mL).
3.76 (s, 3 H), 4.17 (dd, J ϭ 9.4, 2.7 Hz, 2 H), 5.62 (s, 1 H), 6.85 The combined organic phases were dried (Na2SO4) and concen-
(m, 2 H), 7.35 (m, 2 H), 9.79 (broad s, 2 H). 13C NMR (50 MHz, trated. Flash chromatography of the residue with pentane/tert-butyl
1
CDCl3): δ ϭ 14.0, 20.3, 34.6, 43.2, 55.1, 80.1, 101.1, 113.4, 127.2, methyl ether (9:1) furnished 7 (39 mg, 55%) as a colourless oil. H
130.2, 159.9, 200.6. The aldehyde obtained above and 1-ethoxy-2-
methyl-1-trimethylsilyloxypropene[35] (1.07 g, 5.70 mmol) were H), 1.09 (s, 6 H), 1.38 (s, 6 H), 1.53 (s, 6 H), 1.61 (ddd, J ϭ 13.9,
taken up in dichloromethane (5 mL). BF3·OEt2 (0.656 mL, 9.7, and 2.2 Hz, 2 H), 1.63 (ddd, J ϭ 13.9, 9.7, and 2.2 Hz, 2 H),
NMR (500 MHz, C6D6): δ ϭ 0.51 (s, 6 H), 0.73 (s, 3 H), 1.06 (s, 3
0.518 mmol) was added dropwise at Ϫ78 °C over 15 min. After this 3.22 (d, J ϭ 11.3 Hz, 2 H), 3.26 (s, 3 H), 3.46 (d, J ϭ 11.3 Hz, 2
had been stirred for 2 d at Ϫ78 °C, buffer (pH ϭ 7; 20 mL) was H), 3.89 (dd, J ϭ 9.7, 2.2 Hz, 2 H), 3.97 (dd, J ϭ 9.7, 2.2 Hz, 2
added, the phases were separated, and the aqueous phase was ex-
tracted with dichloromethane (5 ϫ 10 mL). The combined organic
phases were dried (Na2SO4) and concentrated to give a 5:1 mixture
H), 5.80 (s, 1 H), 6.87 (m, 2 H), 7.68 (m, 2 H). 13C NMR (50 MHz,
C6D6): δ ϭ 14.2, 18.3, 19.2, 20.5, 21.6, 29.6, 30.1, 32.7, 35.3, 54.7,
72.2, 73.2, 82.0, 98.8, 102.0, 113.9, 128.5, 131.8, 160.4. C31H50O7
of the diesters 33 and 34 (1.09 g, 78%). Flash chromatography with (exact mass, FAB): calcd. 534.3557; found 534.3566.
Eur. J. Org. Chem. 2002, 1292Ϫ1304
1301

T. Trieselmann, R. W. Hoffmann, K. Menzel
FULL PAPER
23. (4R*)-4-Hydroxy-5-{(2s*,4R*,6S*)-6-[(2S*)-2-hydroxy-3,3-di-
(s, 2 H), 3.26 (s, 2 H), 3.62 (s, 3 H), 4.25 (s, 2 H), 6.66 (m, 2 H),
methyl-4-oxopentyl]-2-(4-methoxyphenyl)-5,5-dimethyl-1,3-dioxan- 7.05 (m, 2 H). Dimethyl sulfoxide (13.6 mL, 162 mmol) was added
4-yl}-3,3-dimethylpentanone (37, 38): BF3·OEt2 (1.76 mL, dropwise at Ϫ78 °C to a solution of oxalyl chloride (10.5 mL,
13.9 mmol) was added dropwise at Ϫ78 °C to a solution of the
bis(aldehyde) 32 (2.13 g, 6.96 mmol) and 3-methyl-2-trimethylsily-
83.0 mmol) in dichloromethane (150 mL). After this had been
stirred for 15 min at Ϫ78 °C, a solution of the alcohol obtained
loxy-2-butene[24] (3.13 g, 15.3 mmol). After this had been stirred for above (12.3 g, 54.0 mmol) in dichloromethane (20 mL) was added.
12 h at Ϫ78 °C, buffer (pH ϭ 7) solution (50 mL) was added, the After this mixture had in turn been stirred for 45 min at Ϫ78 °C,
phases were separated, and the aqueous phase was extracted with
diethyl ether (5 ϫ 20 mL). The combined organic phases were dried
(Na2SO4) and concentrated to give a 6:1 mixture of compound 38
and 37 according to the 13C NMR spectrum. Flash chromato-
graphy with pentane/tert-butyl methyl ether (6:4) furnished com-
triethylamine (38.1 mL, 367 mmol) was added dropwise, resulting
in the formation of a precipitate. After stirring for 30 min at
Ϫ78 °C, the mixture was allowed to reach room temperature. Satur-
ated aqueous NH4Cl solution (100 mL) was added, the phases were
separated, and the aqueous phase was extracted with diethyl ether
(3 ϫ 50 mL). The combined organic phases were washed with
pound 38 (2.12 g, 63%) and compound 37 (0.30 g, 10%) as colour-
1
less oils. 38: H NMR (500 MHz, CDCl3): δ ϭ 0.79 (s, 3 H), 0.89 aqueous hydrochloric acid (1 , 100 mL) and brine (100 mL). The
(s, 3 H), 1.13 (s, 6 H), 1.58 (s, 6 H), 1.49 (m, 4 H), 2.15 (s, 6 H), solution was dried (Na2SO4) and concentrated. Bulb-to-bulb distil-
2.85 (d, J ϭ 6.4 Hz, 2 H), 3.79 (m, 2 H), 3.80 (s, 3 H), 3.95 (m, 2 lation furnished the aldehyde 40 (9.72 g, 81%) as a colourless liquid.
H), 5.56 (s, 1 H), 6.87 (m, 2 H), 7.38 (m, 2 H). 13C NMR (50 MHz, 1H NMR (200 MHz, CDCl3): δ ϭ 1.04 (s, 6 H), 3.40 (s, 2 H), 3.74
CDCl3): δ ϭ 14.0, 19.8, 20.5, 21.7, 26.2, 31.4, 35.1, 51.8, 55.2, 72.6,
(s, 3 H), 4.41 (s, 2 H), 6.87 (m, 2 H), 7.20 (m, 2 H), 9.36 (s, 1 H).
81.9, 100.7, 113.3, 127.2, 131.7, 159.6, 215.5. C27H42O7 (478.6): 13C NMR (50 MHz, CDCl3): δ ϭ 21.6, 49.6, 57.8, 75.6, 77.3, 116.3,
calcd. C 67.76, H 8.84; found C 67.74, H 8.85.
129.0, 131.9, 161.8, 207.9. C13H18O3 (222.3): calcd. C 70.24, H 8.16;
found C 70.19, H 8.33.
24. (4R*)-4-(4-Methoxybenzyloxy)-5-{(2s*,4R*,6S*)-6-[(2S*)-2-(4-
methoxybenzyloxy)-3,3-dimethyl-4-oxopentyl]-2-(4-methoxyphenyl)- 26.
5,5-dimethyl-1,3-dioxan-4-yl}-3,3-dimethylpentane-2-one (39): A so-
lution of 4-methoxybenzyl trichloroacetimidate (0.500 g,
0.175 mmol) in diethyl ether (1 mL) was added at Ϫ20 °C to a
(2R*,6S*)-6-Hydroxy-1-{(2s*,4R*,6S*)-6-[(2S*,6R*)-6-hy-
droxy-2,8-bis(4-methoxybenzyloxy)-3,3,7,7-tetramethyl-4-oxooctyl]-
2-(4-methoxyphenyl)-5,5-dimethyl-1,3-dioxan-4-yl}-2,8-bis(4-meth-
oxybenzyloxy)-3,3,7,7-tetramethyloctan-4-one (41): Triethylamine
solution of 38 (0.837 g, 1.75 mmol) in diethyl ether (7 mL). One (0.56 mL, 4.0 mmol) was added at 0 °C to a solution of chlorodicy-
drop of dilute trifluoromethanesulfonic acid in diethyl ether was
added, and the mixture was stirred for 2 h at Ϫ20 °C. After it had
reached room temperature, the addition of 4-methoxybenzyl tri-
clohexylborane (0.657 mL, 3.0 mmol) in diethyl ether (10 mL).
After this had been stirred for 5 min, a solution of the diketone 39
(0.723 g, 1.01 mmol) in diethyl ether (1 mL) was added, resulting
chloroacetimidate and of a trace of trifluoromethanesulfonic acid in the formation of a white precipitate. After stirring for 3 h at 0
was repeated four times. Buffer solution (pH ϭ 7; 10 mL) was
added, the phases were separated and the aqueous phase was ex-
tracted with dichloromethane (4 ϫ 10 mL). The combined organic
phases were dried (Na2SO4) and concentrated. By addition of cold
dichloromethane, the trichloroacetamide was brought to crystallis-
°C, the mixture was cooled to Ϫ90 °C. A solution of the aldehyde
40 (0.778 g, 3.0 mmol) in diethyl ether (3 mL) was added dropwise
at Ϫ90 °C. After stirring for 5 d at this temperature, the mixture
was poured into a combination of buffer (pH ϭ 7; 10 mL), meth-
anol (20 mL), and water (10 mL). Hydrogen peroxide (30% in
ation and filtered. Silica gel (2 g) was added to the mother liquor water, 10 mL) was added at 0 °C, and the mixture was stirred for
and the suspension was concentrated. The residue was placed on 3 h. The phases were separated and the aqueous phase was ex-
top of a flash chromatography column, which was eluted with pent- tracted with dichloromethane (4 ϫ 30 mL). The combined organic
ane/tert-butyl methyl ether (9:1 to 6:4) to give compound 39 (0.86 g,
phases were dried (Na2SO4) and concentrated. Flash chromato-
graphy with pentane/tert-butyl methyl ether (1:1) furnished a 5:1
69%) as a yellowish oil. 1H NMR (200 MHz, [D6]acetone): δ ϭ
0.76 (s, 3 H), 0.87 (s, 3 H), 1.06 (s, 6 H), 1.18 (s, 6 H), 1.43 (ddd, mixture of aldol (0.85 g, 72%), in which product 41 predominated.
J ϭ 14.3, 10.2, and 1.9 Hz, 2 H), 1.62 (ddd, J ϭ 14.3, 9.7, and 1H NMR (500 MHz, CDCl3): δ ϭ 0.74 (s, 3 H), 1.00 (s, 6 H), 1.03
1.6 Hz, 2 H), 2.13 (s, 6 H), 3.59 (dd, J ϭ 10.2, 1.6 Hz, 2 H), 3.73
(s, 3 H), 1.09 (s, 6 H), 1.23 (s, 6 H), 1.30 (s, 6 H), 1.73 (m, 2 H),
(s, 6 H), 3.81 (s, 3 H), 3.97 (dd, J ϭ 9.7, 1.9 Hz, 2 H), 4.57 (m, 4 1.85 (m, 2 H), 2.74 (m, 2 H), 2.99 (m, 2 H), 3.33 (d, J ϭ 14.0 Hz,
H), 5.35 (s, 1 H), 6.89 (m, 4 H), 6.96 (m, 2 H), 7.31 (m, 4 H), 7.48 2 H), 3.34 (d, J ϭ 14.0 Hz, 2 H), 3.39 (s, 6 H), 3.40 (m, 2 H), 3.41
(m, 2 H). 13C NMR (50 MHz, [D6]acetone): δ ϭ 13.9, 15.6, 20.8, (s, 6 H), 3.32 (s, 3 H), 3.76 (m, 2 H), 4.24 (m, 2 H), 4.37 (m, 4 H),
21.7, 26.9, 32.3, 36.0, 55.4, 55.5, 66.1, 75.4, 81.0, 82.9, 101.5, 114.1, 4.71 (m, 4 H), 5.51 (s, 1 H), 6.92 (m, 4 H), 6.88 (m, 4 H), 6.98 (m,
114.5, 128.2, 130.2, 132.0, 132.7, 160.2, 160.7, 212.4. C43H58O9
(exact mass, FAB): calcd. 718.4081; found 718.4089.
2 H), 7.42 (m, 8 H), 7.77 (m, 2 H). 13C NMR (50 MHz, CDCl3):
δ ϭ 13.7, 15.2, 19.8, 20.5, 21.2, 22.0, 31.3, 35.3, 38.0, 40.6, 53.1,
55.1, 55.2, 55.3, 72.9, 73.0, 74.9, 77.9, 80.2, 82.0, 100.6, 113.5,
113.6, 113.7, 127.2, 129.0, 129.2, 130.1, 130.4, 131.6, 159.0, 159.2,
159.7, 215.9. C69H94O15 (exact mass, FAB): calcd. for M ϩ Naϩ:
1185.6490; found 1185.6482.
25. 3-(4-Methoxybenzyloxy)-2,2-dimethyl-1-propanal (40): Sodium
hydride (60% in white oil, 5.00 g, 130 mmol) was added to a solu-
tion of 2,2-dimethyl-1,3-propanediol (13.6 g, 130 mmol) in THF
(100 mL) at 0 °C. After this had been stirred for 30 min, 4-
methoxybenzyl bromide (26.1 g, 130 mmol) was added at 0 °C.
Tetrabutylammonium iodide (ca. 20 mg) was added, and the mix-
27.
(3S*,5S*,7R*)-8-{(2s*,4R*,6S*)-6-[(2S*,4R*,6R*)-4,6-Dihy-
droxy-2,8-bis(4-methoxybenzyloxy)-3,3,7,7-tetramethyloctyl]-2-(4-
ture was stirred for 18 h. Saturated aqueous NH4Cl solution methoxyphenyl)-5,5-dimethyl-1,3-dioxan-4-yl}-1,7-bis(4-methoxy-
(100 mL) was added, the phases were separated, and the aqueous benzyloxy)-2,2,6,6-tetramethyloctane-3,5-diol (42): Acetic acid
phase was extracted with tert-butyl methyl ether (3 ϫ 50 mL). The (2.23 mL, 39.0 mmol) was added to a suspension of tetramethylam-
combined organic phases were dried (MgSO4) and concentrated. monium tris(acetoxy)borohydride (5.09 g, 19.3 mmol) in acetone
Bulb-to-bulb distillation furnished 20.4 g (70%) of 3-(4-methoxy- (4.5 mL). After this had been stirred for 30 min, a clear solution
benzyloxy)-2,2-dimethyl-1-propanol, which was used as obtained. resulted, and this was cooled to 0 °C. A solution of the aldol 41
1H NMR (200 MHz, CDCl3): δ ϭ 0.71 (s, 6 H), 2.83 (s, 1 H), 3.07 (0.45 g, 0.39 mmol) in acetone (2.5 mL) was added, and the mixture
1302
Eur. J. Org. Chem. 2002, 1292Ϫ1304

Conformational Analysis of oligo-1,3-Dioxanylmethanes
was stirred for 1 d at 0 °C and 5 d at room temperature. Saturated 2-Methoxypropene (93 mg, 1.3 mmol) and p-toluenesulfonic acid
FULL PAPER
aqueous sodium potassium tartrate solution (6 mL) was added, re-
sulting in vigorous evolution of gas. The mixture was stirred for
(5 mg) were added at 0 °C. After this had been stirred at room
temperature for 4 h, buffer (pH ϭ 7) solution (10 mL) was added,
20 min and poured into saturated aqueous NaHCO3 solution the phases were separated, and the aqueous phase was extracted
(50 mL). The phases were separated, and the aqueous phase was
with dichloromethane (5 ϫ 10 mL). The combined organic phases
extracted with dichloromethane (5 ϫ 50 mL). The combined or-
were dried (Na2SO4) and concentrated. Flash chromatography of
ganic phases were dried (Na2SO4). Cyclohexane (5 mL) was added the residue with pentane/tert-butyl methyl ether (85:15) furnished
and the solution was concentrated in vacuum. Flash chromato-
graphy of the residue with pentane/tert-butyl methyl ether (3:7) fur-
compound 9 (81 mg, 90%) as a colourless oil. 1H NMR (500 MHz,
CDCl3): δ ϭ 0.71 (s, 12 H), 0.77 (s, 3 H), 0.83 (s, 6 H), 0.91 (s, 3
nished the tetraol 42 as a 5:1 diastereomeric mixture (0.44 g, 96%) H), 0.99 (s, 6 H), 1.34 (m, 4 H), 1.36 (s, 6 H), 1.36 (s, 6 H), 1.37
1
as a yellowish resin. H NMR (500 MHz, CDCl3): δ ϭ 0.77 (s, 3
(s, 6 H), 1.38 (s, 6 H), 1.45 (ddd, J ϭ 13.8, 9.9, and 1.1 Hz, 2 H),
1.54 (ddd, J ϭ 13.8, 9.9, and 1.1 Hz, 2 H), 3.27 (d, J ϭ 11.3 Hz, 2
H), 0.80 (s, 6 H), 0.87 (s, 6 H), 0.89 (s, 6 H), 0.93 (s, 3 H), 0.95 (s,
6 H), 1.41 (m, 4 H), 1.71 (m, 4 H), 3.29 (m, 4 H), 3.47 (m, 2 H), H), 3.58 (dd, J ϭ 9.9, 1.1 Hz, 2 H), 3.59 (d, J ϭ 11.3 Hz, 2 H),
3.72 (s, 3 H), 3.73 (s, 6 H), 3.74 (m, 4 H), 3.67 (s, 6 H), 3.83 (m, 2 3.62 (dd, J ϭ 9.0, 3.0 Hz, 2 H), 3.69 (dd, J ϭ 9.0, 3.0 Hz, 2 H),
H), 4.50 (m, 8 H), 5.27 (s, 1 H), 6.84 (m, 10 H), 7.29 (m, 10 H). 3.73 (dd, J ϭ 9.9, 1.1 Hz, 2 H), 3.82 (s, 3 H), 5.43 (s, 1 H), 6.91
13C NMR (100 MHz, CDCl3): δ ϭ 13.8, 18.2, 19.6, 20.1, 20.6, 22.8, (m, 2 H), 7.41 (m, 2 H). 13C NMR (125 MHz, CDCl3): δ ϭ 13.0,
30.9, 33.2, 38.0, 41.0, 42.5, 55.1, 72.7, 73.1, 74.3, 75.0, 79.5, 82.2, 14.0, 18.1, 18.9, 19.6, 20.5, 21.7, 28.6, 29.0, 29.7, 30.3, 32.6, 35.1,
83.4, 100.7, 113.5, 113.6, 113.7, 127.1, 129.1, 129.2, 129.7, 129.8, 35.2, 55.3, 72.3, 72.6, 73.2, 74.0, 81.9, 98.2, 98.5, 100.8, 113.4,
131.0, 159.0, 159.1, 159.8. C69H98O15 (1167.5): calcd. C 70.98, H 127.2, 132.3, 159.6. C49H82O11 (exact mass, FAB): calcd. 846.5857;
8.46; found C 70.70, H 8.57.
found 846.5826.
28.
(4S*,6R*)-2,2,5,5-Tetramethyl-4-[(4R*,6S*)-2,2,5,5-tetrameth-
yl-6-{(4S*,6R*)-2,2,5,5-tetramethyl-6-[(4R*)-2,2,5,5-tetramethyl-1,3-
dioxan-4-ylmethyl]-1,3-dioxan-4-ylmethyl}-1,3-dioxan-4-ylmethyl]-
6-[(4S*)-2,2,5,5-tetramethyl-1,3-dioxan-4-ylmethyl]-1,3-dioxane (5):
Palladium hydroxide on carbon (ca. 10 mg) was added to a solution
of compound 42 (70 mg, 0.060 mmol) in methanol (3 mL). After
this had been stirred under 1 bar of hydrogen for 4 d, dichlorome-
thane (10 mL) was added and the mixture was filtered through kies-
elguhr. The filtrate was concentrated to give the decaol 43 (31 mg,
90%) as a yellow resin, which was used as obtained. 1H NMR
(500 MHz, CD3OD): δ ϭ 0.77 (s, 6 H), 0.87 (s, 6 H), 0.88 (s, 6 H),
0.89 (s, 3 H), 0.90 (s, 3 H), 0.93 (s, 6 H), 1.52 (m, 6 H), 3.38 (d,
J ϭ 11.0 Hz, 2 H), 3.46 (d, J ϭ 11.0 Hz, 2 H), 3.38 (dd, J ϭ 10.2,
1.2 Hz, 2 H), 3.80 (m, 6 H). 13C NMR (125 MHz, CD3OD): δ ϭ
18.8, 19.4, 19.5, 20.1, 21.8, 34.0, 34.1, 40.1, 41.4, 42.7, 71.0, 74.4,
74.5, 74.6, 74.7. 2-Methoxypropene (80 mg, 1.1 mmol) and p-tolu-
enesulfonic acid (ca. 5 mg) were added to a solution of the decaol
43 (20 mg, 0.11 mmol) in DMF (0.5 mL). After this had been
stirred for 18 h, water (10 mL) was added, the phases were separ-
ated, and the aqueous phase was extracted with diethyl ether (5 ϫ
10 mL). The combined organic phases were dried (Na2SO4) and
concentrated. Flash chromatography of the residue with pentane/
tert-butyl methyl ether (95:5) furnished compound 5 (21 mg, 52%)
as a colourless oil. For the spectroscopic and analytical data see
ref.[14]
Acknowledgments
We thank the Deutsche Forschungsgemeinschaft (Graduierten-
Kolleg ‘‘Metallorganische Chemie’’), the Volkswagen-Stiftung, and
the Fonds der Chemischen Industrie for support of this study. We
thank Dr. B. C. Kahrs for initial experiments towards the synthesis
of compounds 1 and 3.
[1]
R. W. Hoffmann, K. Menzel, Eur. J. Org. Chem. 2001,
2749Ϫ2755.
[2]
R. W. Hoffmann, Angew. Chem. 2000, 112, 2134Ϫ2150; Angew.
Chem. Int. Ed. 2000, 39, 2054Ϫ2070.
[3]
S. Gabriels, D. Van Haver, M. Vandewalle, P. De Clercq, A.
Verstuyf, R. Bouillon, Chem. Eur. J. 2001, 7, 520Ϫ532.
B. J. B. Folmer, R. P. Sijbesma, E. W. Meijer, J. Am. Chem.
[4]
Soc. 2001, 123, 2093Ϫ2094.
R. W. Hoffmann, T. Brandl, P. Kirsch, K. Harms, Synlett
[5]
2001, 960Ϫ963.
[6]
R. W. Hoffmann, B. C. Kahrs, Tetrahedron Lett. 1996, 37,
4479Ϫ4482.
[7]
S. Tsuzuki, L. Schäfer, H. Goto, E. D. Jemmis, H. Hosoya, K.
Siam, K. Tanabe, E. Osawa, J. Am. Chem. Soc. 1991, 113,
4665Ϫ4671.
[8]
The conformation control prevalent in 1 is a consequence of
29. (2s*,4S*,6R*)-2-(4-Methoxyphenyl)-5,5-dimethyl-4-[(4S*,6R*)-
2,2,5,5-tetramethyl-6-{(4S*,6R*)-2,2,5,5-tetramethyl-6-[(4R*)-
2,2,5,5-tetramethyl-1,3-dioxan-4-ylmethyl]-1,3-dioxan-4-ylmethyl}-
1,3-dioxan-4-ylmethyl]-6-[(4S*)-2,2,5,5-tetramethyl-1,3-dioxan-4-
ylmethyl]-1,3-dioxane (9): Hydrogenolysis of compound 42 (0.26 g,
0.26 mmol) was carried out as described under 28. for 16 h. Flash
chromatography of the crude product with dichloromethane/meth-
anol (9:1) containing 1% of concd. aq. NH3 furnished diastereom-
erically pure octaol 44 (37 mg, 48%) as a yellow resin, which was
used as obtained. 1H NMR (500 MHz, CD3OD): δ ϭ 0.83 (s, 6
H), 0.88 (s, 9 H), 0.90 (s, 6 H), 0.98 (s, 6 H), 1.00 (s, 3 H), 1.49 (m,
2 H), 1.57 (m, 4 H), 1.77 (m, 2 H), 3.40 (d, J ϭ 10.8 Hz, 2 H), 3.47
(d, J ϭ 10.6 Hz, 2 H), 3.71 (m, 2 H), 3.81 (s, 3 H), 3.85 (m, 6 H),
3.62 (s, 1 H), 6.91 (m, 2 H), 7.45 (m, 2 H). 13C NMR (125 MHz,
CD3OD): δ ϭ 13.9, 17.7, 18.9, 19.3, 20.2, 21.0, 31.5, 33.5, 35.7,
39.4, 41.9, 55.0, 70.3, 73.1, 73.8, 74.0, 83.2, 101.7, 113.7, 127.7,
131.7, 159.9. The octaol obtained was taken up in DMF (1 mL).
the ‘‘tert-butyl effect’’.[9Ϫ11]
[9]
R. W. Alder, C. M. Maunder, A. G. Orpen, Tetrahedron Lett.
1990, 31, 6717Ϫ6720.
R. W. Alder, P. R. Allen, K. R. Anderson, C. P. Butts, E. Khos-
[10]
ravi, A. Martin, C. M. Maunder, A. G. Orpen, C. B. St. Pur-
cain, J. Chem. Soc., Perkin Trans. 2 1998, 2083Ϫ2107.
R. W. Alder, P. R. Allen, D. Hnyk, D. W. H. Rankin, H. E.
[11]
Robertson, B. A. Smart, R. J. Gillespie, I. Bytheway, J. Org.
Chem. 1999, 64, 4226Ϫ4232.
[12]
C. S. Poss, S. L. Schreiber, Acc. Chem. Res. 1994, 27, 9Ϫ17.
[13]
S. R. Magnuson, Tetrahedron 1995, 51, 2167Ϫ2213.
T. Trieselmann, R. W. Hoffmann, Org. Lett. 2000, 2,
1209Ϫ1212.
T. Imai, S. Nishida, Synthesis 1993, 395Ϫ399.
I. Paterson, K. R. Gibson, R. M. Oballa, Tetrahedron Lett.
[14]
[15]
[16]
1996, 37, 8585Ϫ8588.
[17]
´
D. A. Evans, P. J. Coleman, B. Cote, J. Org. Chem. 1997, 62,
788Ϫ789.
Eur. J. Org. Chem. 2002, 1292Ϫ1304
1303

T. Trieselmann, R. W. Hoffmann, K. Menzel
FULL PAPER
[18]
[30]
M. C. Stumpp, R. R. Schmidt, Tetrahedron 1986, 42,
While the twist-boat conformers amount to 5Ϫ6% of the total
conformer population, they still have the same overall extended
backbone conformation of the molecular skeleton. Since only
the latter is of interest in the current context, the twist-boat
conformers may as well be neglected totally, without significant
change in the predicted conformer population (e.g., calcula-
tions using a systematic torsion of the inter-ring bonds, main-
taining chair conformation in the dioxane rings give a pre-
dicted conformational preference of Ͼ 99% for both 1 or 3).
R. Göttlich, B. C. Kahrs, J. Krüger, R. W. Hoffmann, Chem.
Commun. 1997, 247Ϫ251.
If the conformer equilibrium lies more than 95% to one side,
a temperature increase of less than 100 °C should change the
conformer population by less than 5%. The resultant effect on
the experimental coupling constants will be smaller than the
uncertainty in the measured data.
5941Ϫ5948.
[19]
D. A. Evans, K. T. Chapman, E. M. Carreira, J. Am. Chem.
Soc. 1988, 110, 3560Ϫ3578.
F. Matsuda, N. Tomiyoshi, M. Yanagiya, T. Matsumoto, Tetra-
[20]
hedron 1990, 46, 3469Ϫ3488.
[21]
D. A. Evans, J. L. Duffy, M. J. Dart, Tetrahedron Lett. 1994,
35, 8537Ϫ8540.
D. A. Evans, M. J. Dart, J. L. Duffy, M. G. Yang, J. Am. Chem.
[22]
[31]
[32]
Soc. 1996, 118, 4322Ϫ4343.
[23]
R. W. Hoffmann, U. Weidmann, Chem. Ber. 1985, 118,
3980Ϫ3992.
[24]
H. O. House, L. J. Czuba, M. Gall, H. D. Olmstead, J. Org.
Chem. 1969, 34, 2324Ϫ2336.
[25]
S. D. Rychnovsky, D. J. Skalitzky, Tetrahedron Lett. 1990, 31,
945Ϫ948.
[26]
[33]
[34]
[35]
D. A. Evans, D. L. Rieger, J. R. Gage, Tetrahedron Lett. 1990,
31, 7099Ϫ7100.
K. Pihlaja, M. Kivimäki, A.-M. Myllyniemi, T. Nurmi, J. Org.
R. W. Hoffmann, B. C. Kahrs, J. Schiffer, J. Fleischhauer, J.
Chem. Soc., Perkin Trans. 2 1996, 2407Ϫ2414.
[27]
D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113,
7277Ϫ7287.
Chem. 1982, 47, 4688Ϫ4692.
F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp,
[28]
´
E. Juaristi, J. S. Cruz-Sanchez, J. Org. Chem. 1988, 53,
M. Lipton, C. Caufield, G. Chang, T. Hendrickson, W. C. Still,
3334Ϫ3338.
J. Comput. Chem. 1990, 11, 440Ϫ467.
M. Lipton, W. C. Still, J. Comput. Chem. 1988, 9, 343Ϫ355.
Receiced October 25, 2001
[O01521]
[29]
1304
Eur. J. Org. Chem. 2002, 1292Ϫ1304