Selective mixed Tishchenko reaction via substituted 1,3-dioxan-4-ols

Article abstract of DOI:10.1021/op0100652

Monoesters of 1,3-diols can be prepared with the mixed Tishchenko reaction from β-hydroxy aldehydes and another aldehyde. These two aldehydes form a diastereomeric mixture of 1,3-dioxan-4-ol hemiacetal derivatives which can be further converted to monoesters with suitable catalysts. Limitations in the formation and esterification of this hemiacetal intermediate have been investigated in this work and the formation and stability of 1,3-dioxan-4-ols was found to be aldehyde-, temperature-, and solvent-dependent. A new method was developed for selective preparation of monoesters of 1,3-diols with this mixed Tishchenko reaction via 1,3-dioxan-4-ols without any significant side products. During the development of this method a possibility to scale up the reactions to reach a selective and economical process was one of the main targets in this work.

Full text of DOI:10.1021/op0100652

Organic Process Research & Development 2002, 6, 125−131
Selective Mixed Tishchenko Reaction via Substituted 1,3-Dioxan-4-ols
Olli P. To¨rma¨kangas,† Pauli Saarenketo,‡,§ and Ari M. P. Koskinen*,†
Laboratory of Organic Chemistry, Helsinki UniVersity of Technology, P.O. Box 6100, FIN-02015 HUT, Finland, and
Department of Chemistry, UniVersity of JyVa¨skyla¨, SurVontie 9, FIN-40351 JyVa¨skyla¨, Finland
Scheme 1. Tishchenko and homoaldol-Tishchenko
reactions with enolizable aldehydes
Abstract:
Monoesters of 1,3-diols can be prepared with the mixed
Tishchenko reaction from â-hydroxy aldehydes and another
aldehyde. These two aldehydes form a diastereomeric mixture
of 1,3-dioxan-4-ol hemiacetal derivatives which can be further
converted to monoesters with suitable catalysts. Limitations in
the formation and esterification of this hemiacetal intermediate
have been investigated in this work and the formation and
stability of 1,3-dioxan-4-ols was found to be aldehyde-, tem-
perature-, and solvent-dependent. A new method was developed
for selective preparation of monoesters of 1,3-diols with this
mixed Tishchenko reaction via 1,3-dioxan-4-ols without any
significant side products. During the development of this method
a possibility to scale up the reactions to reach a selective and
economical process was one of the main targets in this work.
The conventional Tishchenko reaction of aldehydes gives
simple esters with moderate-to-excellent yield (Scheme 1;
1f2)1 typically in the presence of Lewis acidic catalysts
such as aluminium alcoholates.2 Several Lewis acidic transi-
tion metal complexes have also been found effective.3 For
enolizable aldehydes, sequential aldol-Tishchenko reaction
can become competing if the catalyst is sufficiently basic
(Scheme 1; 1f3f5).4 The basic catalyst first accomplishes
the aldol reaction which is followed by Tishchenko esteri-
fication by the Lewis acidic nature of the same catalyst. In
the mixed Tishchenko reaction between different aldehydes
the product distribution is difficult to control.5 1,3-Dioxan-
4-ols 4 have been reported as reaction intermediates in the
homoaldol-Tishchenko reaction (Scheme 1) with only one
enolizable aldehyde.6 Dioxanol 4 type intermediates react
further to the monoester 5. In other reports the formation of
4 has not been disclosed, but a direct Tishchenko step (3f5)
via a [6,6]-membered bicyclic transition state has been
suggested.7
Our interest in this topic was originally triggered by the
potential versatility of the mixed aldol-Tishchenko reaction
and the possible applications of mixed esters in more
complex natural product syntheses where the target molecules
bear monoester moieties of diols (e.g., polyketides). How-
ever, the most important application of 1,3-diol monoesters
is their use as the most common coalescing agents in the
paint and coating industry, and this was also the main focus
of our work.8 Herein we report our results on the formation
and stability of the mixed 1,3-dioxan-4-ols of type 8 and
their further esterification to a variety of 1,3-diol monoesters.
* To whom correspondence may be sent. Telephone: +358 9 451 2526.
Fax: +358 9 451 2538. E-mail: ari.koskinen@hut.fi.
† Helsinki University of Technology.
‡ University of Jyva¨skyla¨.
§ X-ray crystal structure analysis.
(1) (a) Tischtschenko, W. J. Russ. Phys. Chem. 1906, 38, 355, 482. (b)
Tischtschenko, W. E. Chem. Zentr. 1906, 77, I, 1309, 1554, 1556. For
mechanism, see: (c) Ogata, Y.; Kawasaki, A.; Kishi, I. Tetrahedron 1967,
23, 825-830.
(2) (a) Ogata, Y.; Kawasaki, A. Tetrahedron 1969, 25, 2845-2851. (b) Saegusa,
T.; Hirota, K.; Hirasawa, E.; Fujii, H. Bull. Chem. Soc. Jpn. 1967, 40, 967-
972. (c) Child, W. C.; Adkins, H. J. Am. Chem. Soc. 1923, 45, 3013-
3023. Fast Tishchenko reaction has been reported recently with bidentate
aluminum alcoholate catalysts: (a) Ooi, T.; Miura, T.; Takaya, K.; Maruoka,
K. Tetrahedron Lett. 1999, 40, 7695-7698.
(3) (a) Menashe, N.; Shvo, Y. Organometallics 1991, 10, 3885-3891. (b) Ito,
T.; Horino, H.; Koshiro, Y.; Yamamoto, A. Bull. Chem. Soc. Jpn. 1982,
55, 504-512.
(4) (a) Villani, F. J.; Nord, F. F. J. Am. Chem. Soc. 1947, 69, 2605-2607. (b)
Kuplinski, M. S.; Nord, F. F. J. Org. Chem. 1943, 8, 256-270.
(5) (a) Ogata, Y.; Kawasaki, A. Tetrahedron 1969, 25, 929-935. (b) Lin, I.;
Day, A. R. J. Am. Chem. Soc. 1952, 74, 5133-5135. (c) For transition
metal-catalyzed crossed Tishchenko reaction, see: Morita, K.-I.; Nishiyama,
Y.; Ishii, Y. Organometallics 1993, 12, 3748-3752.
(6) (a) To¨rma¨kangas, O. P.; Koskinen, A. M. P. Org. Process Res. DeV. 2001,
5 (4), 421-425. (b) Villani, F. J.; Nord, F. F. J. Am. Chem. Soc. 1946, 68,
1674-1675.
(7) (a) Mascarenhas, C. M.; Duffey, M. O.; Liu, S.-Y.; Morken, J. P. Org.
Lett. 1999, 1, 1427-1429. (b) Bodnar, P. M.; Shaw, J. T.; Woerpel, K. A.
J. Org. Chem. 1997, 62, 5674-5675. (c) Umekawa, Y.; Sakaguchi, S.;
Nishiyama, Y.; Ishii, Y. J. Org. Chem. 1997, 62, 3409-3412.
(8) Kirk, R. E.; Othmer, D. F. Encyclopedia of Chemical Technology, 3rd ed.;
Wiley-Interscience: New York, 1984; Vol. 11, p 966-967.
10.1021/op0100652 CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/05/2002
Vol. 6, No. 2, 2002 / Organic Process Research & Development
•
125

Scheme 2. Aldol-Tishchenko type-mixed Tishchenko
reaction
Scheme 3. Preparation of monoesters of 1,3-diols with
mixed Tishchenko reaction via 1,3-dioxan-4-ol intermediates
Results and Discussion
The purpose of our studies was to gain insight into the
formation and properties of mixed dioxanols 8 as well as to
develop a selective process where their Tishchenko reaction
to 1,3-diol monoesters takes place while competing reactions
are avoided.9 One goal was to create an economically and
environmentally advantageous process which would be
possible to scale up to industrial scale without any hazardous
or expensive steps. However, the work reported in this paper
describes the main features of the method including the
possibilities and the limitations. For synthetic versatility,
differing from typical aldol-Tishchenko reactions, a mixed
aldol-Tishchenko reaction was investigated (Scheme 2).
The Tishchenko reaction of 1,3-dioxan-4-ols has been
reported typically for dimeric aldol products.10 Mixed dimers
from a â-hydroxyaldehyde and a second aldehyde have been
suspected for instability, but detailed investigations on their
esterification have not been reported.11 Very little is known
about the effects of different substituents on the formation,
stability, stereochemistry, and reactivity of dioxanols similar
to 8. 2,2-Dimethyl-3-hydroxypropanal 6 (3-hydroxypivala-
ldehyde [HPA])12 was chosen as the starting material because
of its stability and easy detection. Aldehyde 6 was isolated
as the dimer 10, which by 1H NMR (CDCl3) was a mixture
of diastereomers in 60:40 ratio (Scheme 3). By 1H NMR in
D2O, at room temperature the equilibrium consisted of 56%
6 and 44% 10, and at 35 °C, the equilibrium rapidly shifted
completely to monomer 6. In contrast, in CDCl3 at 35 °C
only 9% was in monomeric form, and heating at 50 °C for
90 min was required to shift equilibrium to monomer 6. Slow
dimerisation of 6 was observed in CDCl3 when the solution
was cooled back to room temperature.
new “mixed” 1,3-dioxan-4-ols dimeric, 10 was first mono-
merized in the presence of a large excess (10 mol equiv) of
7b by heating at 65 °C without solvent (Scheme 3). At higher
temperatures 6 started to decompose via retro-aldol reaction.
Heating (monomerisation) should be continued for at least
3 h at 65 °C to monomerize all the dimeric 10. When the
solution was cooled to room temperature, slow formation
of 8b was observed, and a stable equilibrium was reached
only in 2.5-3 days, and monomer 6 was still present (1H
NMR in CDCl3). When the cooling was repeated at 0 °C,
the formation of dioxanol 8 was already complete in less
than 3 h, and only a trace of monomeric 6 was observed
1
with H NMR. The product mixture contained the desired
dimer 8b, dimer 10, free aldehyde 7, and sometimes a trace
of monomeric 6.13
The role of the free aldehyde 7 was also studied to
uncover its effect on the stereochemistry and the rate of
formation of dioxanols 8 (Table 1). Both steric and electronic
effects of the aldehydes 7 were studied by varying the
substituents R to obtain a mixture of diastereomers of
compounds 8a-f. All dimers were prepared with the method
described above. The product distribution was measured
directly after step 6f8 by 1H NMR. Due to the vulnerability
of the products to decomposition, the product mixture was
acetylated or esterified directly to the corresponding mo-
noesters 9 for further analysis.
The two dioxanol diastereomers formed in practically
identical ratios regardless of the substituent R. The identity
of the cis/trans isomers was secured through an X-ray
crystallographic analysis of the crystalline major diastereomer
of 8c, separable by careful chromatography and crystallisa-
tion from MeOH:H2O. Dimer 10 is rather stable even at room
temperature, and there are no problems in isolation and
storing, but most of the dioxanols 8 prepared here were found
to be relatively unstable. The reason for the better stability
of dimer 10 compared to that of 8 was studied by means of
molecular modelling. The calculations (MacroModel 6.0;
Monte Carlo, solvent CDCl3) gave some evidence for the
possibility of intramolecular hydrogen bonding in the dimeric
We next optimized the reaction conditions and require-
ments for preparation of mixed dimers. The first “mixed”
dioxanol 8 we tried to prepare was the dimer of 6 and
2-methylpropanal 7b. In our normal procedure to produce
(9) Aldol reaction: (a) Casiraghi, C.; Zanardi, F.; Appendino, G.; Rassu, G.
Chem. ReV. 2000, 100, 1929-1972. (b) Machajewski, T. D.; Wong, C.-H.;
Lerner, R. A. Angew. Chem., Int. Ed. 2000, 39, 1352-1374. (c) Mahrwald,
R. Chem. ReV. 1999, 99, 1095-1120. (d) Cowden, C.; Paterson, I. Org.
React. 1997, 51, 1-200. (e) Sawamura, M.; Ito, Y. Catal. Asymmetric Synth.
1993, 367-388. (f) Bednarski, M. D. Appl. Biocatal. 1991, 1, 87-116.
Cannizzaro reaction: (g) Kharasch, M. S.; Snyder, R. H. J. Org. Chem.
1949, 14, 819-835. (h) Pfeil, E. Ber. 1951, 84, 229-45. Meerwein-
Ponndorf-Verley reduction/Oppenauer oxidation: (i) de Graauw, C.
F.; Peters, J. A.; van Bekkum, H.; Huskens, J. Synthesis 1994, 1007-1017.
(j) Pickart, D. E.; Hancock, C. K. J. Am. Chem. Soc. 1955, 77, 4642-
4643. Tollens reaction: (k) March, J. AdVanced Organic Chemistry;
Reactions, Mechanisms and Structure, 4th ed.; John Wiley & Sons: 1993;
p 955.
(10) Fouquet, G.; Merger, F.; Platz, R. Liebigs Ann. Chem. 1979, 46, 1591-
1601.
(11) Spa¨th, E.; Szilaˆgui, I. v. Ber. Dtsch. Chem. Ges. 1943, 76, 949-956.
(12) Merger, F.; Platz, R.; Fuchs, W. (Badische Anilin- und Soda-Fabrik A.-
G.). Ger. Offen. Pat. 1957301, 1971; Chem. Abstr. 1971, 75, 76165b.
(13) Similar results have been obtained under rather vigorous reaction conditions
(8 h, 160 °C, under 3.5 MPa pressure): Duke, R. B.; Perry, M. A. (Eastman
Kodak Company). FR Pat. 1414216, 1964; Chem. Abstr. 1966, 64, 11090c.
126
•
Vol. 6, No. 2, 2002 / Organic Process Research & Development

Table 1. Product distribution for the preparation of new dimers
8
8
(mol %)a,b
cis
10
6
(mol %)a,b
trans
(mol %)a
entry
aldehyde 7
propanal
2-methylpropanal
pivalaldehyde
2-ethylhexanal
crotonaldehyde
benzaldehyde
cyclohexylcarboxaldehyde
R
(mol %)a
(diastereomers)
1
2
3
4
5
6
7
a
b
c
d
e
f
Et
i-Pr
t-Bu
0
0
0
22
25
32
38
0
42
34
48
50
60
49
0
58
66
30 (40:60)
25 (41:59)
8 (40:60)
13 (31:69)
38 (41:59)
34 (40:60)
18 (40:60)
1′-Et-pentyl
CHdCHCH3
Ph
0
62
51
0
g
C6H11
a The product distribution was determined by 1H NMR. b Relative stereochemistry has been detemined by X-ray crystallography of acetylated 8c.
Figure 3.
in different deuterated solvents. The decomposition was
followed as a function of time at room temperature. After 3
days 8b was partly decomposed in d6-benzene (33%) and
CDCl3 (61%). On the other hand, in d6-DMSO the decom-
position was not observed (<1%). The reason for this is
believed to be the hydrogen bonding between the hydroxyl
Figure 1. X-ray structure of the cis-enantiomers of 4-acetoxy-
5,5-dimethyl-2-tert-butyl-1,3-dioxane 8c, crystallized in a cen-
trosymmetric space group P-1. Unit cell contained both enan-
tiomers of 8c which are shown in the figure.
group of the dioxanol and DMSO which stabilizes the acetal-
type dioxanol structure.
In our earlier studies we have shown that the catalyst in
the Tishchenko esterification of such dimers should bear
sufficient basicity to deprotonate the hydroxyl proton and
Lewis acidity for the intramolecular hydride shift from ring
position 2 to position 4.6a,14 Esterification of dioxanol 8b to
the corresponding monoester 9 (R ) i-Pr) was first carried
Figure 2.
out with 30-40 mol % of traditional metal hydroxide
catalysts such as LiOH (4.55 M) or Ba(OH)2‚H2O with low
isolated yield (0-35%). Monoester 12 was formed with
Tishchenko esterification of 10 (formed due to equilibration,
Scheme 3). Products 13, 14, and 15 were formed by
irreversible hydrolysis of monoesters 9b and 12. Use of 1,3-
diol-based alkali metal monoalcoholates as catalysts was then
investigated.14a Gratifyingly, the esterification of dioxanol
8b occurred almost quantitatively in the presence of 30 mol
% of 0.1 M solution (in THF) of monolithium alcoholate of
diol 13. Monoester 9b was obtained in 86% isolated yield
after 50 min. Monoesters 9a and 9c were obtained with the
similar manner in 41 and 60% isolated yield, respectively.
However, these two latter experiments were carried in
several-times-smaller scales. The reaction was quenched with
the addition of a catalytic amount of 2 M HCl to avoid side
reactions during the workup. In larger scale the products can
be isolated by fractional distillation under reduced pressure
â-hydroxyl aldehydes between oxygen at ring position 1 and
the hydroxyl group in the side chain at ring position 2. Also,
the differences in calculated minimum energies between the
diastereomers were compared to the ratio of diastereomers
obtained with NMR. In the case of dioxanol 8b the
calculations gave the axial anomers 8b the energy minimum
of ∆G ) -177.6 kJ, and for the equatorial anomers 8b′ ∆G
) -166.7 kJ, corresponding to a ratio of 80:20, in good
agreement with the observed ratio 73:27. After the formation
of the new dioxanal-type mixed dimer, the 2-methylpropanal
(7b) excess was removed under reduced pressure (0.1
mmHg) at 0 °C. The product distribution was followed with
1H NMR during the evaporation, and the new dimer 8b was
observed to slowly decompose. The reaction was found to
be limited to low-boiling aldehydes 7. With higher-boiling
aldehydes such as 2-ethylhexanal (7d) the reaction had to
be heated to room temperature (under reduced pressure of
0.1 mmHg) to evacuate an excess of free aldehyde. This
caused significant decomposition of dioxanol 8d. The effect
of the solvent on the stability of mixed dimer 8b was studied
(14) (a) To¨rma¨kangas, O. P.; Koskinen, A. M. P. Tetrahedron Lett. 2001, 42,
2743-2746. (b) See also: Abu-Hasanayn, F.; Streitwieser, A. J. Org. Chem.
1998, 63, 2954-2960.
1
by H NMR. Dioxanol 8b was prepared at 0 °C, dissolved
Vol. 6, No. 2, 2002 / Organic Process Research & Development
•
127

Scheme 4
monoalcoholate catalysts consistently giving high yields.
Thus, excess aldehyde has to be evaporated before addition
of the catalyst to avoid unfavorable and exothermic side
reactions such as homo aldol-Tishchenko. The evaporated
aldehyde can be recycled directly to the next batch. Fur-
thermore, the process developed is possible to scale up to
large (industrial) scale. However, this paper focused on the
main features of the method, and the largest scale was the
10-mmol scale. The scale-up work to larger (possibly to pilot)
has not been carried out yet. In industrial scale a possible
need for cryogenic condenser cooling would raise the total
costs of the method considerably. Finally, in comparison to
alternative methods this is the most inexpensive way to
prepare 1,3-dioxan-4-ols16 and, further, 1,3-diol monoesters.13
The method developed allows a design and preparation of
various monoesters to be used as the coalescing agents for
different purposes.
after workup. In these experiments (10 mmol of monomeric
HPA) column chromatography was the preferred method for
purification.
Tishchenko esterifications have also been studied previ-
ously with milder catalysts yielding the 1,3-diol monoester
with excellent anti-selectivity, high yields and without any
side reactions.15 However, we found SmI2 alone to be
ineffective in the esterification of 1,3-dioxan-4-ols due to
its low basic nature.
It has been claimed by Merger et al. that â-hydroxy
aldehydes form both an acyclic hemiacetal and a cyclic 1,3-
dioxan-4-ol and that these react to the corresponding diol
monoester.10 We found no evidence of such an open-ring
hemiacetal by 1H NMR. We believe that in the case of 1,3-
dioxan-4-ols the metal first coordinates to the hydroxyl group
and to the ring oxygen at position 3 which activates the
hydride shift from ring carbon 2 to carbon 4. The reaction
mixture was also acetylated, and no traces of open-ring forms
were observed. However, the final stages of the mechanism
are believed to be similar to the mechanism reported in the
case of â-hydroxyketone and simple aldehyde in the presence
of SmI2.15
Experimental Section
All aldehydes were purchased from commercial sources
(Aldrich) but (except formalin) were further distilled (purity
>99.5%). TLC was performed on Merck silica gel 60-F
plates staining with 1% anisaldehyde in acidic ethanol
solution. Flash chromatography was performed on Merck
silica gel 60 (230-400 mesh). NMR spectra were recorded
on Bruker AM-200 and Varian Gemini 400 or Bruker DPX
400 instruments. CDCl3 was filtered through basic alumina
right before use to remove traces of HCl. Mass spectra were
recorded on a Kratos MS80 RF Autoconsole instrument. All
melting points were measured with Gallenkamp GMP
(capillary) apparatus and are uncorrected. IR spectra were
obtained on a Perkin-Elmer Spectrum One apparatus. Perkin-
Elmer 8420 gas chromatograph was used with OV-1701
silica column. The 4-hydroxy-1,3-dioxane derivatives 8 were
very unstable and decomposed during flash chromatography,
distillation under atmospheric and reduced pressure and even
when kept under reduced pressure. The derivatives were
Conclusions
The formation of 1,3-dioxan-4-ol type acetal between a
â-hydroxyaldehyde and a second aldehyde was shown to be
especially temperature-dependent. This also explains the need
for low reaction temperature in the case of Evans-Tish-
chenko and aldol-Tishchenko reactions. At 0 °C and in the
presence of excess monofunctional aldehyde all the mono-
meric â-hydroxyaldehyde dimerized in 3 h to a diastereo-
meric mixture of 1,3-dioxan-4-ols. Also, the electronic effects
of the aldehyde 7 were found to be important, for example,
in the case of electron-withdrawing substituent R (crotonal-
dehyde) the formation of the initial hemiacetal is disfavored.
With donor substituents (alkyl) the formation of dioxanols
8 is favored. Several dioxanol-type “mixed” dimers are rather
unstable and sometimes even impossible to purify, and
protection of the ring hydroxyl group may be needed for
isolation and purification. The structure can be stabilized by
intramolecular hydrogen bonding in the case of dimeric
â-hydroxy aldehydes. The Tishchenko esterification can be
favourably carried out in the presence of 1,3-diol-based
1
identified directly from the product mixture with H NMR
and usually protected at the free hydroxyl group. In the
following Experimental Section the dioxanols were first
prepared and analyzed as a mixture followed by possible
acetylation or direct Tishchenko esterification to monoesters
and purification by means of column chromatography.
5,5-Dimethyl-4-hydroxyl-2-(1′,1′-dimethyl-2′-hydroxy-
lethyl)-1,3-dioxane (10). Triethylamine (5.06 mL, 50 mmol)
and formaldehyde (37.4 mL, 500 mmol, 10 mol equiv, 37%
aqueous solution) were placed into a reaction vessel kept
under argon. Isobutyraldehyde 7b (45.4 mL, 500 mmol, 10
mol equiv) was added dropwise (over 10 min), and the
temperature was raised to +75 °C. The reaction was stirred
at +75 °C for 1 h and then cooled to room temperature. A
white precipitate started to form at +45 °C. Isooctane (10
mL) was added, and the product was allowed to crystallise
(16) Alternative ways to produce 1,3-dioxan-4-ols: (a) Reduction of 1,3-dioxan-
4-ones: Dahanukar, V. H.; Rychnovsky, S. D. J. Org. Chem. 1996, 61,
8317-8320. (b) m-CPBA oxidation of cyclic vinyl ethers (1,3-dioxane type
acetals): Wattenbach, C.; Maurer, M.; Frauenrath, H. Synlett 1999, 303-
306. (c) Treatment of acrolein in the presence of TiCl4-n-Bu4NI and 2 equiv
of aldehyde: Uehira, S.; Han, Z.; Shinokubo, H.; Oshima, K. Org. Lett.
1999, 1 (9), 1383-1385.
(15) (a) Evans, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1990, 112, 6447-
6449. See also: (b) Lu, L.; Chang, H.-Y.; Fang, J.-M. J. Org. Chem. 1999,
64, 843-853.
128
•
Vol. 6, No. 2, 2002 / Organic Process Research & Development

3
3
overnight at room temperature. The white solid was filtered
and triturated with isooctane. The product was filtered again,
washed with 8 × 50 mL of cold isooctane, and dried to give
10 (44.4 g, 87%) as a mixture of diastereomers.17 TLC: Rf
) 0.39 (EtOAc:hexanes, 7:3). 1H NMR (CDCl3, 200 MHz)
(Diastereomer 1, 40%) δ 4.91 (s, 1H), 4.81 (d, 1H, 4J ) 1.1
Hz), 3.85 (d, 1H, 2J ) 11.1 Hz), 3.46 (s, 2H), 3.39 (dd, 1H,
2J ) 11.1 Hz, 4J ) 1.1 Hz), 1.17 (s, 3H), 0.95 (s, 3H), 0.92
(s, 3H), 0.82 (s, 3H). (Diastereomer 2, 60%) δ 4.60 (s, 1H),
4.42 (s, 1H), 3.62 (d, 1H, 2J ) 11.4 Hz), 3.49 (s, 2H), 3.30
3H), 0.96 (d, 3H, J ) 6.9 Hz), 0.95 (d, 3H, J ) 6.9 Hz),
0.77 (s, 3H).
Lithium Monoalcoholate of 2,2-Dimethyl-1,3-propane-
diol.14a 2,2-Dimethyl-1,3-propanediol 13 (0.312 g, 3.0 mmol)
was dissolved in 28.5 mL of dry THF under argon. The
solution was cooled to 0 °C, and n-BuLi (1.41 mL, 3.0 mmol,
2.13 M) was added dropwise to give a 0.1 M solution of the
catalyst. The colourless solution was allowed to warm to
room temperature over 1 h and then cooled back to 0 °C
before use.
4
4
(dd, 1H, J ) 11.4 Hz, J ≈ 0.1 Hz), 1.06 (s, 3H), 0.97 (s,
3H), 0.96 (s, 3H), 0.83 (s, 3H). CI-MS (monomer when
ionized with NH3) 120 (M + 1 + NH3), 72, 56, 41, 39.
5,5-Dimethyl-4-hydroxyl-2-isopropyl-1,3-dioxane (8b).
Dimer 10 (1.022 g, 5.0 mmol) was dissolved in isobutyral-
dehyde 7b (4.54 mL, 50 mmol, 1000 mol equiv) under Ar.
The colourless solution was stirred for 3 h at +65 °C ((3
°C) to monomerize 6. The solution was cooled to 0 °C and
stirred for another 3 h. Excess 7b was evaporated at 0 °C
(60 min at 0.1 mmHg), yielding an oily product. TLC: Rf
(2,2-Dimethyl-3-hydroxylpropyl)-2-methylpropionate
(9b). Dioxanol 8b were prepared as described above. The
catalyst solution described above (30 mL, 0.1 M, 30 mol
%) was added at 0 °C, and after 45 min the reaction was
quenched with 1.5 mL of 2 M HCl. The solvents were
evaporated, and the oily residue was dissolved in 15 mL of
CH2Cl2 and washed with 1.5 mL of water. The water phase
was saturated with NaCl and washed twice with 5 mL of
CH2Cl2. Organic phases were combined, dried over Na2SO4,
filtered, evaporated, and purified by column chromatography
(EtOAc:hexane, 40:60). Compound 9b was obtained as a
colourless oil (0.753 g, 86%). TLC: Rf ) 0.55 (EtOAc:
hexane, 70:30). 1H NMR (CDCl3, 200 MHz) δ 3.93 (s, 2H),
1
) 0.59 (EtOAc:hexane, 8:2). H NMR (CDCl3, 200 MHz)
(trans-isomer 2S*,4S*-8b, 27%) δ 4.86 (d, 1H, OCH(i-Pr)O,
4
3J ) 4.4 Hz), 4.82 (d, 1H, -OCHOH, J ≈ 1 Hz), 3.89 (d,
2
2
1H, OCH2C, J ) 11.0 Hz), 3.40 (d-d, 1H, OCH2C, J )
3
3.29 (s, 2H), 2.63 (broad s, 1H, OH), 2.59 (sept, 1H, J )
4
3
11.0 Hz, J ) 1.3 Hz), 1.76 (dq, 1H, -CHMe2, J ) 7.0
3
7.0 Hz), 1.22 (d, 6H, J ) 7.0 Hz), 0.98 (s, 6H). 13C NMR
3
Hz, J ) 4.4 Hz), 1.21 (s, 3H, ring CH3), 0.94 (d, 6H, 2×
(CDCl3, 50 MHz) 177.8, 69.1, 68.1, 36.5, 34.1, 21.4, 19.0.
EI-MS: 174 (M), 199, 101, 73, 56, 41.
(i-Pr) CH3, 3J ) 7.0 Hz), 0.82 (s, 3H, ring CH3). (cis-Isomer
2S*,4R*-8b, 73%) δ 4.60 (s, 1H, OCHOH), 4.36 (d, 1H,
OCH(i-Pr)O, 3J ) 4.6 Hz), 3.60 (d, 1H, OCH2C, 2J ) 11.5
5,5-Dimethyl-4-hydroxyl-2-propyl-1,3-dioxane (8a).
Dimeric HPA 10 (0.204 g, 1.0 mmol) was dissolved in
propanal 7a (0.75 mL, 10 mmol, 1000 mol equiv) under
argon. The solution was refluxed for 3 h, cooled to 0 °C,
followed with additional stirring for 3 h. The product mixture
2
Hz), 3.35 (d, 1H, OCH2C, J ) 10.9 Hz), 1.87 (dq, 1H,
-CHMe2, 3J ) 6.9 Hz, 3J ) 4.6 Hz), 1.18 (s, 3H), 0.97 (d,
3
6H, 2× (i-Pr) CH3, J ) 6.9 Hz), 0.83 (s, 3H, ring CH3).
The compound decomposed on CI-MS analysis.
1
was not acetylated but was analyzed with H NMR. TLC:
4-Acetoxy-5,5-dimethyl-2-isopropyl-1,3-dioxane (11b).
The product mixture (8b, 100 mol %) from above was cooled
to 0 °C under argon. Pyridine (8.09 mL, 100 mmol, 1000
mol equiv) and freshly distilled acetic anhydride (9.44 mL,
100 mmol, 1000 mol equiv) were added, and the solution
was stirred at 0 °C for 4 h and then allowed to warm to
room temperature overnight. Excess pyridine was removed
by evaporation with toluene (3 × 20 mL). Salts were
removed with filtration through a silica column (height 15
cm, diameter 3 cm) with EtOAc. After evaporation the crude
product mixture was purified with column chromatography
(CH2Cl2:hexane, 80:20), and a mixture of diastereomers (ratio
26:74) of 11b was obtained in 44% overall yield (starting
from 11) (0.949 g). TLC: Rf ) 0.66 (EtOAc:hexane, 70:
Rf ) 0.60 (EtOAc:hexane, 7:3). 1H NMR (CDCl3, 200 MHz)
(trans-isomer 2S*,4S*-8a; 28%) δ 5.05 (t, 1H, OCH(n-Pr)O,
3J ) 5.1 Hz), 4.82 (s, 1H, -OCHOH), 3.86 (d, 1H, OCH2C,
2
4
2J ) 11.1 Hz), 3.40 (dd, 1H, OCH2C, J ) 10.8 Hz, J )
3
3
1.3 Hz), 1.70 (qd, 1H, J ) 7.5 Hz, J ) 5.0 Hz), 1.19 (s,
3H, ring CH3), 0.94 (t, 3H, J ) 7.5 Hz), 0.83 (s, 3H, ring
3
CH3). (cis-isomer 2S*,4R*-8a, 72%) δ 4.81 (s, 1H, OCHOH),
4.55 (t, 1H, OCH(n-Pr)O, J ) 5.1 Hz), 3.60 (d, 1H, OCH2C,
2
2J ) 11.5 Hz), 3.37 (d, 1H, OCH2C, J ) 11.5 Hz), 1.61
(qd, 1H, 3J ) 7.5 Hz and 3J ) 5.0 Hz), 1.08 (s, 3H), 0.97 (t,
3
3H, CH3, J ) 7.5 Hz), 0.84 (s, 3H, ring CH3).
(2,2-Dimethyl-3-hydroxylpropyl)-propionate (9a). Di-
oxanol 8a was prepared as above. Excess propanal was
evaporated at 0 °C from the product mixture. (30 min at 0.5
mmHg), yielding an oily product. The catalyst solution (30
mol %) of monolithium alcoholate of 2,2,dimethyl-1,3-
propandiol (0.1 M in THF) was added in one portion. The
reaction was complete after 15 min at 0 °C and was quenched
with 0.5 mL of 2 M HCl. The solvent was evaporated to
give white residue which was taken up with 15 mL of CH2-
Cl2. Organic phase was washed with 3 mL of brine which
was then washed with 2 × 5 mL of CH2Cl2. All organic
phases were combined, dried over Na2SO4, filtered, and
evaporated to give 0.423 g of yellowish, oily product mixture.
Column chromatography purification (CH2Cl2:MTBE, 9:1)
1
30). H NMR (CDCl3, 200 MHz) (trans-isomer 2S*,4S*-
4
3
11b, 26%) δ 5.76 (d, 1H, J ) 1.5 Hz), 4.66 (d, 1H, J )
4.2 Hz), 3.81 (d, 1H, 2J ) 11.1 Hz), 3.48 (dd, 1H, 2J ) 11.1
Hz, 4J ) 1.5 Hz), 2.141 (s, 3H), 1.79 (d sept, 1H, 3J ) 4.2,
3J ) 6.9 Hz), 1.24 (s, 3H), 0.93 (d, 3H, J ) 6.9 Hz), 0.92
3
(d, 3H, 2.9 Hz), 0.77 (s, 3H). (cis-Isomer 2S*,4R*-11b, 74%)
δ 5.57 (s, 1H), 4.41 (d, 1H, 3J ) 4.8 Hz), 3.63 (d, 1H, 2J )
2
4
11.4 Hz), 3.44 (dd, 1H, J ) 11.4 Hz, J ) 0.6 Hz), 2.135
3
3
(s, 3H), 1.88 (d sept, 1H, J ) 4.8, J ) 6.9 Hz), 1.13 (s,
(17) Some 2,2-dimethyl-1,3-propanediol 14 is formed during the reaction. This
can be removed by dissolving the product in ether and washing with small
amount of water.
Vol. 6, No. 2, 2002 / Organic Process Research & Development
•
129

gave 0.131 (41% yield) of monoester 9a. TLC: Rf ) 0.55
5,5-Dimethyl-4-hydroxyl-2-phenyl-1,3-dioxane (8f).
Dimer 10 (1.022 g, 5.0 mmol) was dissolved in benzaldehyde
7f (5.08 mL, 50 mmol, 1000 mol equiv) under argon. The
solution was stirred for 3 h at +65 °C and then cooled to 0
°C and stirred for another 3 h. TLC: Rf ) 0.58 (EtOAc:
1
(EtOAc:hexane, 7:3). H NMR (CDCl3, 400 MHz) δ 3.94
(s, 2H), 3.30 (s, 2H), 2.37 (q, 2H, 3J ) 7.6 Hz), 1.16 (t, 3H,
3J ) 7.0 Hz), 0.92 (s, 6H). 13C NMR (CDCl3, 100 MHz) δ
175.2, 69.2, 68.2, 36.4, 27.6, 21.4, 18.6.
1
5,5-Dimethyl-4-hydroxyl-2-t-butyl-1,3-dioxane (8c).
Dimer 10 (0.205 g, 1.0 mmol) was dissolved in pivalaldehyde
7c (1.08 mL, 10 mmol, 1000 mol equiv) under argon. The
solution was stirred for 3 h at +65 °C ((3 °C) and then 3
h at 0 °C. At the end a precipitation of pivalaldehyde was
hexane, 7:3). H NMR (CDCl3, 400 MHz) (trans-isomer
2S*,4S*-8f; 42%) δ 7.92-7.86 (m, 2H), 7.68-7.46 (m, 3H),
6.08 (s, 1H, -OCH(Ph)O), 4.98 (s, 1H, -OCHOH), 4.12
(d, 1H, OCH2C, 2J ) 11.2 Hz), 3.42 (dd, 1H, OCH2C, 2J )
4
11.1 Hz, J ) 1.2 Hz), 1.22 (s, 3H, ring CH3), 0.91 (s, 3H,
1
obtained. TLC: Rf ) 0.64 (EtOAc:hexane, 8:2). H NMR
ring CH3). (cis-Isomer 2S*,4R*-8f, 58%) δ 7.92-7.86 (m,
2H), 7.68-7.46 (m, 3H), 5.58 (s, 1H, OCH(Ph)O), 4.85 (s,
1H, OCH(Ph)O), 3.58 (d, 1H, OCH2C, 2J ) 11.3 Hz), 3.33
(d, 1H, OCH2C, 2J ) 11.3 Hz), 1.15 (s, 3H, ring CH3), 0.92
(s, 3H, ring CH3). The compound decomposed on CI-MS
analysis.
(CDCl3, 400 MHz) (trans-isomer 2S*,4S*-8c, 35%) δ 4.81
(d, 1H, -OCHOH, 4J ) 0.7 Hz), 4.82 (s, 1H, OCH(t-Bu)O),
3.85 (d, 1H, OCH2C, 2J ) 10.8 Hz), 3.39 (dd, 1H, OCH2C,
2J ) 10.8 Hz, 4J ) 1.1 Hz), 1.16 (s, 3H, ring CH3), 0.91 (s,
9H, CH3 of t-Bu), 0.82 (s, 3H, ring CH3). (cis-Isomer
2S*,4R*-8c, 65%) δ 4.58 (s, 1H, OCHOH), 4.19 (s, 1H,
OCH(t-Bu)O), 3.59 (d, 1H, OCH2C, 2J ) 11.2 Hz), 3.33 (d,
1H, OCH2C, 2J ) 11.2 Hz), 1.04 (s, 1H, ring CH3), 0.94 (s,
9H, CH3 of t-Bu), 0.82 (s, 3H, ring CH3).
Unstable 8c and 8d were acetylated successfully:
4-Acetoxy-5,5-dimethyl-2-t-butyl-1,3-dioxane (11c).
Dimer 10 (0.511 g, 2.5 mmol) and pivalaldehyde (2.75 mL,
25 mmol, 1000 mol %) were placed in a 25-mL two-necked
flask under argon and stirred for 3 h at 65 °C. The solution
was cooled to 0 °C and stirred for another 3 h. Pyridine (8.0
mL, 99 mmol, 4000 mol %) and acetic anhydride (9.5 mL,
101 mmol, 4000 mol %) were added at 0 °C, and the mixture
was stirred overnight, allowing the temperature to rise to
room temperature. Excess pyridine was removed by evapora-
tion with toluene (3 × 20 mL). The salts were removed with
filtration through a silica column. Solvents were evaporated
and the products purified by column chromatography (EtOAc:
(2,2-Dimethyl-3-hydroxylpropyl)-2,2-dimethylpropi-
onate (9c). The product mixture prepared above containing
mainly 8c was heated to +10 °C because of precipitation of
pivalaldehyde and the excess aldehyde was removed under
reduced pressure (60 min at 0.5 mmHg). The catalyst solution
of monolithium alcoholate of diol 13 (30 mol %, 0.1 M in
THF) was added at 0 °C, and the reaction was stirred for 60
min at 0 °C. The reaction was quenched with 0.5 mL of 2
M HCl, and the workup was carried out as in the preparation
of 9a. Column chromatography purification (CH2Cl2:MTBE,
90:10) gave 0.223 g (60% yield). TLC: Rf ) 0.58. 1H NMR
1
hexane, 5:95). TLC: Rf ) 0.64 (EtOAc:hexane, 7:3). H
NMR (CDCl3, 400 MHz) (trans-isomer 2S*,4S*-11c, 35%)
δ 4.81 (d, 1H, 4J ) 0.7 Hz), 4.68 (s, 1H), 3.85 (d, 1H, 2J )
10.8 Hz), 3.39 (dd, 1H, 2J ) 10.8 Hz, 4J ) 1.1 Hz), 1.16 (s,
3H), 0.91 (s, 6H), 0.82 (s, 3H). (cis-Isomer 2S*,4R*-11c,
3
(CDCl3, 400 MHz) δ 3.93 (s, 2H), 3.28 (d, 2H, J ) 6.6
3
Hz), 2.31 (t, 1H, J ) 6.6 Hz), 1.22 (s, 9H), 0.93 (s, 6H).
13C NMR (CDCl3, 100 MHz) δ 179.3, 69.2, 68.2, 39.0, 36.7,
27.2, 21.3.
2
65%) δ 4.58 (s, 1H), 4.19 (s, 1H), 3.59 (d, 1H, J ) 11.2
Dioxanols 8d, 8f, and 8g from high-boiling aldehydes 7
decomposed during evaporation of excess of 7 and thus their
esterification was not possible in this work. The Tishchenko
reaction of 8g was studied without the evaporation of excess
aldehyde, but a large number of side products were obtained
due to aldol-Tishchenko reaction, and we were not able to
isolate monoester 9g.
Hz), 3.33 (d, 1H, 2J ) 11.2 Hz), 1.04 (s, 3H), 0.94 (s, 6H),
0.82 (s, 3H). IR (KBr-disk) 2974.5, 2875.0, 2854.8, 1756.4,
1466.2, 1397.5, 1373.8, 1363.9, 1230.6, 1122.6, 1008.3, 922.
Analysis: Calculated for C12H22O4 C 62.58, H 9.63; Found
C 62.73, H 9.57.
4-Acetoxy-5,5-dimethyl-2-(1′-ethylpentyl)-1,3-diox-
ane (11d). Dimer 10 (2.046 g, 10 mmol) was dissolved in
2-ethylhexanal (9.17 mL, 101 mmol, 1000 mol-%) under
argon. The colourless clear solution was heated for 3 h at
+63 °C. Hexane (2 mL) was added and the solution cooled
immediately to 0 °C. The solution was stirred for 3 h at 0
°C. The ratio of 10 to 8d was 14:86. Only two diastereomers
of HPA-2-ethylhexanal dimer were observed (ratio 69:31).
1H NMR (CDCl3, 200 MHz) (Diastereomer 1, 31%) δ 5.07
(d, 1H, OCHRO, 3J ) 2.7 Hz), 4.81 (d, 1H, OCHOH, 4J )
1.6 Hz), 3.86 (d, 1H, OCH2C, 2J ) 10.6 Hz), 3.40 (dd, 1H,
5,5-Dimethyl-4-hydroxyl-2-cyclohexyl-1,3-dioxane (8g).
Dimer 10 (1.030 g, 5.0 mmol) was dissolved in cyclohexyl-
carboxaldehyde 7g (6.1 mL, 50 mmol, 1000 mol equiv) under
argon. The solution was stirred for 3 h at +65 °C and then
cooled to 0 °C and stirred for another 3 h. TLC: Rf ) 0.65
1
(EtOAc:hexane, 7:3). H NMR (CDCl3, 400 MHz) (trans-
3
isomer 2S*,4S*-8g, 34%) δ 4.82 (d, 1H, OCH(C6H11)O, J
) 5.0 Hz), 4.79 (s, 1H, -OCHOH), 3.84 (d, 1H, OCH2C,
2
4
2J ) 10.8 Hz), 3.37 (dd, 1H, OCH2C, J ) 10.8 Hz, J )
1.2 Hz), 2.28-2.16 (m, 1H), 1.94-1.58 (m, 5H), 1.42-1.06
(m, 5H), 1.16 (s, 3H, ring CH3), 0.81 (s, 3H, ring CH3). (cis-
Isomer 2S*,4R*-8g, 66%) δ 4.56 (broad s, 1H, OCHOH),
2
4
OCH2C, J ) 10.6 Hz, J ) 0.5 Hz), 1.80-1.40 (m, 1H,
n-Bu-CH-Et), 1.40-1.10 (m, 4H, 2× CH2, CH2CHCH2),
1.1-0.8 (m, 12H, 4× CH3). (Diastereomer 2, 69%) δ 4.59
(s, 1H, OCHOH), 4.56 (d, 1H, OCHRO, 3J ) 2.3 Hz), 3.61
(d, 1H, OCH2C, 2J ) 12.0 Hz), 3.34 (dd, 1H, OCH2C, 2J )
3
4.33 (d, 1H, OCH(C6H11)O, J ) 5.0 Hz), 3.58 (d, 1H,
OCH2C, 2J ) 11.3 Hz), 3.33 (d, 1H, OCH2C, 2J ) 11.3 Hz),
2.28-2.16 (m, 1H), 1.94-1.58 (m, 5H), 1.42-1.06 (m, 5H),
1.04 (s, 3H), 0.81 (s, 3H, ring CH3).
4
11.4 Hz, J ) 0.6 Hz), 1.80-1.40 (m, 1H, n-Bu-CH-Et),
1.40-1.10 (m, 4H, 2× CH2, CH2CHCH2), 1.1-0.8 (m, 12H,
130
•
Vol. 6, No. 2, 2002 / Organic Process Research & Development

4× CH3). Acetylation was carried out as with 8b and column
chromatographic purification (EtOAc:hexane) gave a dia-
stereomeric mixture of 11d in 34% isolated yield (1.86 g).
TLC: Rf ) 0.40 (EtOAc:hexane, 20:80). 1H NMR (CDCl3,
400 MHz) (cis-isomer 2S*,4R*-11d, 65%) δ 5.56 (d, 1H, 2J
) 0.9 Hz), 4.60 (dd, 1H, 3J ) 5.9 Hz, 3J ) 2.2 Hz), 3.63 (d,
1H, 2J ) 11.4 Hz), 3.43 (d, 1H, 2J ) 11.4 Hz), 2.14 (s, 3H),
1.60-1.25 (m, 9H), 1.13 (s, 3H), 0.76 (s, 3H), 0.95-0.85
(m, 6H), 0.76 (s, 3H). (trans-Isomer 2S*,4S*-11d, 35%) δ
72.34(1)°, â ) 89.02(1)°, γ ) 75.20(1)°, V ) 654.36(14)
Å3, Fcalc ) 1.169 g cm-3, µ ) 0.086 mm-1, no absorption
correction, Z ) 2, triclinic, space group P-1 (No. 2), λ )
0.71073 Å, T ) 173 K, ω and æ scans, 4970 reflections
collected, [(sin θ)/λ] ) 0.65 Å-1, 2971 independent (Rint
)
0.024) and 2170 observed reflections [I g 2σ(I)], 146 refined
parameters, R ) 0.047, wR2 ) 0.104, largest diff. peak and
hole 0.21 (-0.23) e Å-3, hydrogens calculated and refined
as riding atoms.
3
3
3
5.77 (d, 1H, J ) 1.5 Hz), 4.87 (dd, 1H, J ) 3.3 Hz, J )
1.1 Hz), 3.80 (d, 1H, 2J ) 10.8 Hz), 3.49 (dd, 1H, 2J ) 11.1
Hz, 4J ) 1.7 Hz), 2.12 (s, 3H), 1.60-1.25 (m, 9H), 1.24 (s,
3H), 0.93-0.85 (m, 6H), 0.76 (s, 3H).
Acknowledgment
Neste Chemicals Co., Tekes (National Technology Agency,
Finland) and Neste Research Foundation are gratefully
acknowledged for financial support and cooperation in this
work.
X-ray crystal data of 11c:18 formula C12H22O4, M )
230.30, colorless crystal 0.58 mm × 0.13 mm × 0.13 mm,
a ) 6.077(1) Å, b ) 9.311(1) Å, c ) 12.580(1) Å, R )
(18) Crystallographic data (excluding structure factors) for 8c has been deposited
with the Cambridge Crystallographic Data Centre as supplementary publica-
tion nos. CCDC 162042. Copies of the data can be obtained, free of charge,
on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax:
+44 1223 336033 or e-mail: deposit@ccdc.cam.ac.uk).
Received for review August 15, 2001.
OP0100652
Vol. 6, No. 2, 2002 / Organic Process Research & Development
•
131