Tocopherols by hydride reduction of dialkylamino derivatives

Article abstract of DOI:10.1002/ejoc.200600874

Aminomethylation with Mannich reagents derived from secondary amines and paraformaldehyde under improved conditions has been used to convert non-α-tocopherol homologues into α-tocopherol, the biologically most important vitamin E compound. Mono- and bis(aminomethylated) β-, γ-and δ-tocopherol were then subsequently transformed into the corresponding tocopherols (α- and β-tocopherol) by reductive deamination. As an alternative to classical catalytic hydrogenation in the last step, efficient laboratory protocols using complex hydrides have been derived and applied to the preparation of labelled vitamin E compounds. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejoc.200600874

FULL PAPER
DOI: 10.1002/ejoc.200600874
Tocopherols by Hydride Reduction of Dialkylamino Derivatives
Thomas Netscher,*[a] Francesco Mazzini,*[b] and Roselyne Jestin[a][‡]
Keywords: Natural products / Vitamins / Reduction / Aminomethylation / Deuteriated tocopherols
Aminomethylation with Mannich reagents derived from sec-
ondary amines and paraformaldehyde under improved con-
ductive deamination. As an alternative to classical catalytic
hydrogenation in the last step, efficient laboratory protocols
ditions has been used to convert non-α-tocopherol homo- using complex hydrides have been derived and applied to
logues into α-tocopherol, the biologically most important vi-
tamin E compound. Mono- and bis(aminomethylated) β-, γ-
and δ-tocopherol were then subsequently transformed into
the preparation of labelled vitamin E compounds.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
the corresponding tocopherols (α- and β-tocopherol) by re- Germany, 2007)
Generally, two-step alkylation/reduction sequences are
Introduction
used for this transformation, including halo-,[3] amino-,[4]
and hydroxymethylation reactions[5] (Scheme 1), but most
of them suffer from disadvantages relating to the selectivity
of the reactions and the reagents and reaction conditions
used. We previously disclosed our results on the efficient
preparation of 1 from lower homologues using a hydroxy-
methylation/hydrogenation[6] sequence in the presence of
boric acid.
Tocopherols play an essential role in biological systems
owing to their vitamin E and antioxidant activities.[1] All-
racemic α-tocopherol, an equimolar mixture of eight stereo-
isomers prepared by acid-catalyzed condensation of tri-
methylhydroquinone and racemic isophytol, is the product
of highest economic relevance.[2] The naturally occurring
(2R,4ЈR,8ЈR)-α-tocopherol (1) can be obtained on an indus-
trial scale from the processing of natural material. However,
in most sources used for this purpose, for example, soy-
beans and sunflowers, 1 is accompanied by the lower homo-
logues, β-, γ- and δ-tocopherol (2–4, Figure 1), which pos-
sess lower vitamin E activity. Therefore, improvement of the
chemical conversion of 2–4 into 1 is an attractive goal.
Figure 1. Naturally occurring tocopherols.
[a] Research and Development, DSM Nutritional Products
P. O. Box 3255, 4002 Basel, Switzerland
Fax: +41-61-687-22-01
Scheme 1. Two-step alkylation/reduction sequence generally used
in the conversion of non-α-tocopherols into 1.[7]
E-mail: thomas.netscher@dsm.com
[b] Dipartimento di Chimica e Chimica Industriale, Università di
Pisa
For the aminomethylation route, it is known from the
literature[4,7] that conversion of tocopherol mixtures of 2–4
into the corresponding aminomethyl derivatives is incom-
plete. The catalytic hydrogenation reaction, performed sub-
sequently to convert the incompletely aminomethylated
Via Risorgimento 35, 56126 Pisa, Italy
Fax: +39-050-2219-260
E-mail: lcap@dcci.unipi.it
[‡] Summer student (1996) from Ecole Nationale Supérieure de
Chimie de Montpellier, 34053 Montpellier Cedex, France
1176
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1176–1183

Tocopherols by Hydride Reduction of Dialkylamino Derivatives
FULL PAPER
product mixture into 1, leads to a mixture containing lower
In order to find an alternative deamination process with-
tocopherol homologues that then has to undergo an ad- out the use of catalytic hydrogenation for laboratory-scale
ditional reaction cycle of aminomethylation/catalytic re- preparations and, in particular, for the synthesis of labelled
duction for complete conversion into 1. This drawback has analogues, we carried out a series of experiments testing
been solved through the application of a protocol developed several reducing agents (mainly complex hydrides) in the
in our laboratories for efficient aminomethylation of vita- transformation of 7a into 1 (Table 1). The course of the
min E concentrates (Scheme 1).[7] Herein we report an reactions was monitored both by GC and HPLC analyses.
exhaustive study of the reductive deamination step carried The reagents used in Entries 1–11 gave essentially no or
out with complex hydrides, an alternative to the classic hy- very little conversion into 1. Moderate (Entry 12) to good
drogenation protocol. The deamination procedure disclosed yields (Entries 13 and 14) were obtained by acylation of the
is particularly useful for the preparation of selectively lab- phenolic hydroxy function followed by quaternization of
elled vitamin E analogues, as effective as previously re- the amino group, thus transforming it into a better leaving
ported by using NaCNBD3.[8]
group facilitating the subsequent reaction with the reducing
agent. However, we continued our investigations by looking
for an easier one-step reduction protocol. Therefore, several
hydrides were investigated (Table 1, 4.5–5 equiv. of hydride
used in all the experiments). LiAlH4 led to only decomposi-
tion of the starting material 7a, while no reaction occurred
with DIBAH and BH3. Zn(BH4)2, prepared in situ by
treacing ZnCl2 with NaBH4 in THF according to a re-
ported procedure,[9] proved to be too reactive in alcoholic
solvents, and even in iBuOH it was consumed quickly.
Changing the solvent to an aprotic medium like 1,4-dioxane
afforded no α-tocopherol, while TLC and HPLC analyses
showed the formation of a moderate amount of an un-
known compound. Excess nPrOH was then added to a sam-
ple of the reaction mixture which was then refluxed for 2 h,
showing complete disappearance of the unknown com-
pound and the presence of starting material only. We then
supposed that the observed compound was a complex
formed between Zn(BH4)2 and 7a, maybe favoured by the
coordinating solvent. A similar behaviour was observed
using LiBH4[10] in THF, with the formation of the supposed
hydride–7a complex in up to 80% yield without any ap-
preciable conversion into α-tocopherol. Addition of a few
millilitres of nPrOH to the reaction mixture also gave the
starting material quantitatively after 2 h reflux. LiBH4 was
slightly more resistant to hydrolysis in iBuOH than
Zn(BH4)2, affording 1 in 20% yield after 6 h. While NaBH-
(AcO)3 proved to be ineffective (Entry 23), the use of
NaBH4 in polar solvents gave interesting results. The use
of diglyme and DMF provided low and moderate yields,
respectively, accompanied by decomposition of 7a, whereas
reactions carried out in DMSO (Entry 26) afforded irrepro-
ducible data, although a yield of 86% was obtained in one
experiment, and are also less practical in the workup. Better
results were obtained by using alcoholic solvents (En-
tries 27–30, 35). Water was not tested because of the insolu-
bility of the substrate in this solvent, while MeOH was avo-
ided because of the rapid decomposition of NaBH4 in this
solvent.[11] Though such a decomposition is slower in
EtOH, the yield of reduction of 7a was no better than 50%.
Increasing the reaction temperature by using alcoholic sol-
vents with higher boiling points (Entries 28–30) led to a
substantial improvement, with yields ranging from 64 to
90% when refluxing in iBuOH for 12 h. We observed the
formation of a viscous white gel in the reaction medium
which increased with temperature and time of reaction,
Results and Discussion
Efficient procedures for aminomethylation using Man-
nich reagents have been developed in our laboratories
(Scheme 1).[7] The reagents can be prepared easily from pa-
raformaldehyde and secondary dialkylamines, for example,
diethylamine, morpholine (preferred) or piperidine, in a 1:1
ratio. Best results are obtained when the alkylation step is
carried out under solvent-free conditions at temperatures of
80–140 °C over several hours. Complete alkylation, includ-
ing double alkylation of δ-tocopherol (4), can be achieved
by using 4–10 mol-equiv. of Mannich reagent. If the aim is
selective monoalkylation (vide infra, for the preparation of
β-tocopherol, 2), or if more expensive labelled formalde-
hyde has to be used, the amount of reagent can be reduced
accordingly, while adapting the reaction conditions (tem-
perature and time). Details of the alkylation of β-, γ- and
δ-tocopherol (2–4) and subsequent standard catalytic hy-
drogenation, are specified in the Experimental Section.
In addition to the complete double alkylation of the aro-
matic moiety (Scheme 1, 4 Ǟ 8), being essential for the
preparation of 1 in high yields, the protocol devised also
allowed the highly regioselective monoalkylation 4 Ǟ 9
(Scheme 2), which represents a particular advantage of this
approach in the efficient synthesis of β-tocopherol (2). In
fact, by using roughly a stoichiometric amount of the Man-
nich reagent (1–1.2 equiv.) and by taking advantage of the
different reactivity of the 5-position of 4 compared with the
7-position, it was possible to obtain 9 regioselectively in
high yields (83%), which was then reduced (by catalytic hy-
drogenation or hydride treatment) to afford 2 quantitatively.
Scheme 2. Regioselective preparation of 2 from 4.
Eur. J. Org. Chem. 2007, 1176–1183
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1177

T. Netscher, F. Mazzini, R. Jestin
FULL PAPER
Table 1. Reduction trials for the transformation of 7a into 1.
Entry Reagent
Solvent
Temp.
Time
Result[a]
1
2
3
4
5
6
7
8
Li
Na
Na
Zn
NH3/THF
NH3/THF
EtOH
AcOH
THF
THF
THF
H2O
CH2Cl2
CH3CN
–80 °C Ǟ room temp.
–80 °C Ǟ room temp.
reflux
reflux
room temp.
reflux
reflux
100 °C
reflux
75 °C
2.5 h
16 h
22 h
1.5 h
5 h
5 d
3 d
only 7a
mainly 7a + dec.
7a (39%) + 11a
mainly 7a
LiDBB
Ph2PLi
TiCl3/Li
Ni/Al, KOH
HCOOH
HSiCl3/Et3N
Pd/C
mainly 7a
mainly 7a
7a (32%) + 1 (12%) + dec.
only 7a
only 7a
mainly 7a
5 h
9
16 h
16 h
2 d
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
HCO2H/MeOH room temp.
7a (23%) + 1 (12%) + n.i.
47% 1
1. Ac2O, 2. HCl, 3. SnCl2
1. Ac2O, 2. MeI/Ag+, 3. NaBH3CN
71% 1
1. diphosgene/Hünig’s base, 2. LiAlH4
DIBAH
BH3/THF
ZnCl2+NaBH4
ZnCl2+NaBH4
LiAlH4
LiBH4
LiBH4
LiEt3BH
NaBH(AcO)3
NaBH4
NaBH4
NaBH4
NaBH4
NaBH4
NaBH4
NaBH4
NaBH3CN
NaBH3CN
NaBH3CN
Bu4NBH4
NaBH4/OH–
88% 1
hexane
THF
1,4-dioxane
iBuOH
THF
THF
iBuOH
THF
THF
diglyme
DMF
DMSO
EtOH
iPrOH
tBuOH
iBuOH
DMSO
EtOH
reflux
room temp.
reflux
16 h
5 d
6 h
only 7a
only 7a
only 7a
mainly 7a+ 7% 1
mainly dec.
only 7a
reflux
reflux
70 °C
reflux
reflux
100 °C Ǟ 160 °C
reflux
120 °C
reflux
reflux
reflux
reflux
120 °C
reflux
reflux
reflux
reflux
2 d
12 h
6 h
2 d
22.5 h only 7a
7 h
7a + 20% 1
only 7a
mainly 7a + dec.
50% 1 + dec.
max. 86% 1
50% 1 + 50% 7a
50–60% 1 + 30–40% 7a
64% 1 + 36% 7a
90% 1 + 10% 7a
26% 1 + 37% 7a + 9% n.i.
56% 1 + 17% 7a + 5% n.i.
95% 1
2 h
6 h
21 h
2 d
21 h
12 h
4.5 h
22 h
4 h
iBuOH
iBuOH
iBuOH
4 h
10 h
77% 1 + 19% 11
ca. 100% 1
[a] Qualitative results from screening experiments are given for the cases in which no (only 7a) or low (mainly 7a) conversion of 7a or
major decomposition (dec.) and formation of unknown compounds (n.i. not identified) occurred.
quite evident using iBuOH, and disappeared on addition of proving the performances of the reactions with NaBH4 and
water. This behaviour was not observed by refluxing iBuOH Bu4NBH4. It was noticed that conversion of 7a into 1 in
and NaBH4 together. Surprisingly, NaCNBH3 gave better refluxing iBuOH proceeded relatively well in the first few
and reproducible results under these conditions after 4 h, hours (60–70% after 3 h) and then became slower and
affording 1 in almost quantitative yield (95%, Entry 33). slower (80% max. after 9 h) and was accompanied by an
Similar conversions were obtained by treating 7a with increase in gel formation. Increasing the initial loading of
Bu4NBH4 in refluxing iBuOH (Entry 34), the reaction me- NaBH4 (up to 8 equiv.) produced only limited improve-
dium remaining a homogeneous solution, but the desired α- ments in the early stages of the reaction (5–15%) without
tocopherol was accompanied by the formation of moderate significant changes in the subsequent reaction pattern. We
amounts (15–25%) of the corresponding n-butyl ether 11. supposed that decomposition of NaBH4 accounted for the
Formation of the n-butyl ether of the aminomethylated pre- observed behaviour and the different reactivity compared
cursor 7a was not observed.
with NaCNBH3 which is more resistant to hydrolysis.[12]
On the basis of the data gathered in this hydride screen- Supporting this hypothesis, a trial performed starting with
ing, we carried out further experiments with the aim of im- 2 equiv. of NaBH4 and by adding 2 equiv. every 3 h showed
1178
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1176–1183

Tocopherols by Hydride Reduction of Dialkylamino Derivatives
FULL PAPER
a rather constant conversion difference for each addition, with the devised solvent-free aminomethylation protocol.
eventually affording 1 in 96% yield after 18 h. Further evi- By combining the use of deuteriated or unlabelled para-
dence came from experiments carried out in the presence formaldehyde and NaBH4 it is possible to prepare mono-,
of NaOH. NaBH4 decomposition has been reported to be di- and trideuteriated analogues. In this way, [D1]-, [D2]-,
significantly slower in basic aqueous and alcoholic solu- [D3]-α-tocopherol and [D3]-β-tocopherol were successfully
tions.[13] Indeed, reduction of 7a in refluxing iBuOH using and easily prepared in high yields and isotopic purity
–
5 equiv. of NaBH4 (starting [BH4 ] = 0.6 ) and 2 equiv. of (Scheme 3).
NaOH occurred to give 82% yield after 3 h and 96% yield
after 6 h. Moreover, the same very good yields were ob-
tained by using a three-fold lower starting concentration of
–
NaBH4 ([BH4 ] = 0.2, 4 equiv. NaBH4, 2 equiv. NaOH) and
complete conversion was reached in 9–10 h on addition of
one more equivalent of NaBH4 after 7 h. Using the same
conditions in the analogous transformation of 9 into 2 pro-
vided β-tocopherol in quantitative yield in 10 h. Conversely,
reduction of bis(morpholinomethylated) 8a to α-tocopherol
proceeded more slowly using this protocol, very likely due
to the increased steric hindrance and to the different reac-
tivity of the 5-position compared with the 7-position, as has
been observed in several other cases.[14] In particular, after
refluxing for 6 h, the resulting reaction mixture was com-
posed of 5% 1, 15% 7a, 60% 6a and 20% 8a. The addition
of 2 equiv. of NaBH4 changed the composition to 26% 1,
1–2% 7a, 62% 6a and 10% 8a after 9 h of overall reflux.
Further additions of hydride (11 equiv. total) allowed com-
plete conversion of 8a, but only 60% formation of 1 and
40% unreacted 6a. The higher resistance of 6a to hydride
Scheme 3. Preparation of [D1]-, [D2]-, [D3]-α-tocopherols and [D3]-
reduction compared with 7a was also observed in the prep-
β-tocopherol.
aration of [C7-13C]-α-tocopherol from 7-(morpholino-[13C]-
methyl)-β-tocopherol using NaCNBH3 (iBuOH reflux,
–
5 equiv., [BH3 ] = 0.8 ), in which the desired labelled α-
tocopherol was recovered in 34% yield after 6 h together
Conclusions
with the unreacted aminomethylated precursor.[15] These
Efficient protocols for the alkylation of the lower homo-
findings suggest that the hydride reduction proceeds
logues β-, γ- and δ-tocopherol (2–4) by aminomethylation/
through an o-quinomethide intermediate, with the 5-posi-
reduction sequences are now available. Moreover, high re-
tion being highly preferred over the 7-position, as reported
gioselectivity was obtained in the reaction between δ-to-
copherol and a stoichiometric amount of the aminomethyl-
ation reagent (Mannich reagent), providing an easy synthe-
sis of β-tocopherol. An exhaustive hydride reduction screen-
ing study was carried out in order to find alternatives to
classical hydrogenation for the conversion of aminomethyl-
ated derivatives into α-tocopherol. NaCNBH3 and NaBH4/
NaOH in iBuOH proved to be very effective in the re-
duction of 5-(aminomethylated) γ-tocopherol, but less ef-
ficient than hydrogenation in the reduction of bis(amino-
methylated) δ-tocopherol. However, the conditions iden-
tified for hydride reduction are useful in the laboratory-
scale preparation of multi-fold labelled α- and β-tocopherol
and tocotrienols, increasingly used for metabolic studies
and accurate quantitative analytical methods, and represent
a very efficient alternative to previously reported methods
for the synthesis of labelled vitamin E derivatives.[16]
previously.[14]
Interestingly, Bu4NBH4 provided the best hydride load-
ing/conversion ratio, but we did not find reaction condi-
tions to completely avoid the formation of n-butyl ether 11.
However, the ether can be easily separated from 1 by col-
umn chromatography [hexane/EtOAc, 8:1; Rf(11) = 0.78,
Rf(1) = 0.51]. A lower temperature led to a longer reaction
time and not less than 7–12% of 11. Addition of small
–
amounts of water (2–5%, starting [BH4 ] = 0.2 ) afforded
the best result in terms of byproduct formation (11, 5%),
but conversion was no higher than 86%. A similar result
was obtained by refluxing in nPrOH instead of iBuOH.
–
Using a lower initial loading of Bu4NBH4 (1 equiv., [BH4 ]
= 0.05 ) provided 77% of 1 and 6% of 11 after 7 h. Ad-
dition of a further equivalent of Bu4NBH4 afforded almost
quantitative conversion, yielding 84% of 1 and 14% of 11.
While the reduction conditions outlined above give po-
orer performances than classic hydrogenation in the conver-
sion of 7-(aminomethylated) vitamin E concentrates into α-
tocopherol, they represent a very efficient procedure with
which to prepare labelled (stable and unstable isotopes) α-
Experimental Section
General: All reactions were carried out under argon. Room tem-
and β-tocopherols[8b] and tocotrienols[8a] in association perature (room temp.) corresponds to 20–22 °C. All solvents were
Eur. J. Org. Chem. 2007, 1176–1183
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1179

T. Netscher, F. Mazzini, R. Jestin
FULL PAPER
of puriss. quality and purchased from Fluka, Merck or Aldrich and
were used without further purification. Paradeuterioformaldehyde
was obtained from CEA (France). Bu4NBH4 was prepared accord-
ing to the procedure reported by Brändström et al.[17] To isolate γ-
(3) and δ-tocopherol (4), we used tocopherol concentrate, “d-Mixed
Tocopherols” from Bizen Chemical Co., Ltd., Kumayama, Akaiwa,
Okayama 709-07 (Japan), containing 11.17% 1, 1.27% 2, 37.26%
3, 23.74% 4 (GC, acetate derivatives) and δ-tocopherol (Sigma)
containing 86.8% 4, 3.7% 3, Ͻ0.1% 1 (GC, acetate derivatives) as
starting materials, according to the purification procedure de-
scribed below. Column chromatography: SiO2, Merck, particle size
usually 0.063–0.2 mm, 0.04–0.063 mm where indicated, 0–0.2 bar
Ar. TLC: silica gel plates, Merck, SiO2 60 F254. Substances were
detected with UV (254/366 nm) and ammonium molybdate/
Ce(SO4)2 in H2O/H2SO4 with subsequent heating. Supercritical
fluid chromatography (SFC): Instrument Lee Scientific, model 600
(Dionex, Salt Lake City, UT, USA), FID at 380 °C, fused silica
capillary column with biphenyl-30, length 10 m, i.d. 50 µm, mobile
phase CO2, 100 °C, 0.2–0.75 gmL–1. HPLC: Lichrosorb S 160,
5 µm and LUNA C18(2), 5 µm, 250ϫ4.6 mm (Phenomenex),
CH3CN/MeOH (50:50) 1.5 mLmin–1, λ = 295 nm. Melting points
(uncorrected): Büchi 510 apparatus. IR spectra: Nicolet 7199 FT-
IR spectrometer. 1H and 13C NMR spectra: CDCl3, chemical shifts
expressed on the δ scale (ppm), TMS as internal standard, Bruker
Spectrospin WM-250, BrukerAC 300, Bruker AM-400. EI-MS and
ISP-MS: SSQ 7000 (Finnigan MAT), EI; API IIl (SCIEX Perkin-
Elmer), ISP; API 300 (SCIEX Perkin-Elmer), ISP; MAT 95 (Finni-
gan MAT) spectrometer, ISP or EI; indication of characteristic
peaks, m/z (%).
chromatography as described for the purification of 4. The desired
fraction was obtained after 6 L. Concentration of the collected
fractions and distillation in vacuo (185 °C, 0.035 mbar, bulb-to-
bulb) afforded 3 (20.0 g) as a yellowish oil. Purity Ͼ99% (GC, ace-
tate derivatives). TLC (hexane/AcOEt, 9:1), Rf(3) = 0.29. 1H NMR
spectral data are identical to published values.[19] Stereochemical
purity (GLC, methyl ether derivative): 100%.
Mannich Reagent Derived from Morpholine: Paraformaldehyde
(30.0 g, 1 mol) was added portionwise to morpholine (87.1 mL,
1 mol) while stirring at 70 °C over 20 min in such a manner that
the temperature rose to a maximum of 80 °C. After stirring at 80 °C
for an additional 2 h, the reaction finished with the formation of a
colourless liquid. According to the 1H NMR spectrum, the reagent
contained morpholinomethanol (A), dimorpholinomethane (B)
and minor amounts of other unidentified components. Characteris-
1
tic data: H NMR (400 MHz, CDCl3): δ = 2.14 (s, 1 H, OH, A),
2.50 (t, J = 4.6 Hz, 8 H, NCH2CH2O, B), 2.68 (m, 4 H,
NCH2CH2O, A), 2.91 (s, 2 H, NCH2N, B), 3.71 (m, 12 H
NCH2CH2O, A+B), 4.13 (s, 2 H, NCH2OH, A) 4.24–4.98 (not
allocated signals) ppm. 13C NMR (100 MHz, CDCl3): δ = 49.76
(NCH2CH2O, A), 52.00 (NCH2CH2O, B), 66.81 (NCH2CH2O, A),
67.00 (NCH2CH2O, B), 81.65 (NCH2N, B), 87.23 (NCH2OH, A)
ppm. Signals were assigned on the basis of 13C,1H COSY NMR
investigations and comparison with 13C and 1H NMR spectro-
scopic data of pure B. ISP-MS: m/z (%) = 187.3 (6) [B + H]+, 159.3
(12), 129.3 (40), 118.2 (100) [A + H]+.
(2R,4ЈR,8ЈR)-7-(Morpholinomethyl)-β-tocopherol (6a): The Man-
nich reagent derived from morpholine (9.37 g, 80 mmol) was added
to mechanically stirred 2 (8.4 g, 20 mmol) at room temp. Then the
mixture was heated to 125 °C within about 15 min and became
tarnished. After 1.5 h, the aminomethylation was complete [control
by GC, silylated derivatives: tr(2) = 5.91 min, tr(6a) = 10.67 min].
After cooling to room temp., the mixture was diluted with TBME
(250 mL) and washed with H2O until the wash solution became
neutral (about 5–6 times with 50 mL of H2O each). The organic
layer was dried with K2CO3. Removal of the solvent in vacuo af-
forded 6a (9.89 g) as a yellowish oil. Purity 97.7% (GC, silylated
derivative). For further characterization, 6a (1 g) was dissolved in
MeOH (25 mL). While stirring the solution and slowly cooling to
0 °C, 6a was obtained as a colourless precipitate. The solid was
filtered off and recrystallized from MeOH. Colourless crystals.
Analytical Procedures
Tocopherols 1–4: The chemical purity was determined by GLC of
the acetate derivatives. Squalane (50.2 mg, Fluka, 99.6%), DMAP
(10 mg), pyridine and acetic anhydride (0.5 mL each) were added
at room temp. to 1–4 (100 mg). The mixture was allowed to stand
for 15 min at room temp. and then 1 µL was injected into the GLC
apparatus. Values are given as wt.-% and were determined by the
internal standard method. GLC was conducted on a PS-086 capil-
lary column. The optical purity of 1–4 was determined by GLC of
the methyl ether derivative.[18]
Mono- and Bis(alkylaminomethyl)tocopherols 6–10
Chemical Purity via Trimethysilyl Ether Derivatives: Pyridine and
N,O-bis(trimethylsilyl)trifluoroacetamide (BSTAF) (0.1 mL each)
were added to a crude mixture of an aminomethylation reaction
(10 µL) or a crude product 6–10 (ca. 10 µg). The mixture was
heated at 90 °C for 1 min and then diluted with 3 parts of AcOEt.
The resulting solution was used directly for GC analysis. The values
are given as area-% of trimethylsilyl ether derivatives 6–OTMS to
10–OTMS. GLC: capillary column PS-086 unless stated otherwise.
M.p. 33–36 °C. IR (KBr): ν = 3438 (s), 2952 (vs, sh), 2926 (vs),
˜
2867 (s, sh), 1622 (m), 1457 (vs), 1417 (s), 1377 (s), 1261 (vs), 1118
(vs, sh), 1109 (vs), 1064 (s), 1005 (m), 915 (m) cm–1. 1H NMR
(250 MHz, CDCl3): δ = 0.83–1.65 (m, 36 H), 1.78 (m, 2 H), 2.1 (s,
6 H), 2.60 (m, 6 H), 3.69 (s, 2 H), 3.74 (m, 4 H), 10.63 (s, 1 H)
ppm. EI-MS: m/z (%) = 515.5 (35) [M]+·, 428 (100) [M –
C4H9NO]+, 165 (45) [C10H13O2]+. C33H57NO3 (515.82): calcd. C
76.84, H 11.14, N 2.72; found C 76.59, H 11.17, N 2.75. The crude
residue can be purified by column chromatography (hexane/Ac-
OEt, 8:1). Alternatively, 6a can be converted into the corresponding
hydrochloride (6a·HCl): 6a (3.7 g, 6.7 mmol) was dissolved in
TBME (250 mL) and gaseous HCl (180 mL, 7.5 mmol) was passed
through the solution to yield a colourless precipitate. Removal of
the solvent in vacuo and recrystallization of 6a·HCl (0.5 g) from
AcOEt (10 mL) afforded 6a·HCl as colourless crystals in the form
Purification of Raw Materials
δ-Tocopherol (4): δ-Tocopherol concentrate (150 g, Sigma) was di-
luted with eluent (150 mL hexane/AcOEt, 88:12) and purified
by chromatography (column dimension: 90/10.5 cm; flow:
40 mLmin–1, fractions were checked by TLC every 200 mL). Toc-
opherol 4 was obtained after 6.5 L. Concentration of the collected
fractions and distillation in vacuo (178–185 °C, 0.035 mbar, bulb-
to-bulb) afforded 4 (97.0 g) as a yellow oil. Purity 98.58% (GC,
acetate derivative): TLC (hexane/AcOEt, 9:1), Rf(4) 0.24. Shifts and
allocations were in accordance with results of Baker and Myers.[19]
Stereochemical purity (GLC, methyl ether derivative): 100%.
of druses. M.p. 152–161 °C. IR (KBr): ν = 3430 (s), 2952 (vs, sh),
˜
2928 (vs), 2869 (s, sh), 1455 (vs), 1261 (s), 1120 (s) cm–1. 1H NMR
(250 MHz, CDCl3): δ = 0.83–1.65 (m, 36 H), 1.79 (m, 2 H), 2.17
(s, 3 H), 2.20 (s, 3 H), 2.61 (t, J = 6.8 Hz, 2 H), 3.03 (m, 2 H), 3.44
(d, J = 11.7 Hz, 2 H), 3.92 (d, J = 13 Hz, 2 H), 4.28–4.42 (m, 4
H), 7.44 (s, 1 H), 11.30 (s, 1 H) ppm. ISP-MS: m/z (%) = 516.5
(100) [M – HCl + H]+, 429.5 (6.5) [M – HCl – C4H9NO + H]+.
γ-Tocopherol (3): “d-Mixed Tocopherols” (80 g, Bizen) were diluted
with eluent (80 mL hexane/AcOEt, 88:12) and purified by
1180
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1176–1183

Tocopherols by Hydride Reduction of Dialkylamino Derivatives
FULL PAPER
C33H58ClNO3 (552.28): calcd. C 71.77, H 10.59, Cl 6.42, N 2.54;
found C 71.70, H 10.47, Cl 6.70, N 2.49. Addition of 5% aq.
NaOH to 6a·HCl followed by extraction with TMBE provided 6a
quantitatively.
line dihydrochloride 8a·2HCl was obtained as a colourless precipi-
tate. The solid was filtered off and washed with TBME. Tocopherol
salt 8a·2HCl (1 g) was recrystallized from EtOH/Et2O (15 mL) as
colourless crystals. M.p. 144–147 °C. IR (KBr): ν = 3235 (w), 2951
˜
(vs, sh), 2924 (vs), 2698 (m), 2675 (m), 2619 (m), 1256 (m), 1600
(m), 1498 (m), 1461 (vs), 1442 (vs, sh), 1428 (s, sh), 1405 (m), 1377
(s), 1354 (m), 1331 (m), 1268 (s), 1238 (m), 1203 (m), 1141 (s), 1125
(s), 1078 (s), 1071 (m), 1057 (m), 1020 (m), 1000 (m), 950 (m), 898
(m), 875 (m) cm–1. 1H NMR (250 MHz, CDCl3): δ = 0.85–1.58 (m,
36 H), 1.86 (t, J = 6.5 Hz, 2 H), 2.21 (s, 3 H), 2.82–3.30 (m, 6 H),
3.48 (m, 4 H), 3.95 (m, 4 H), 4.14–4.50 (m, 8 H), 9.28 (s, 1 H),
11.30 (s, 1 H), 11.46 (s, 1 H) ppm. ISP-MS: m/z (%) = 601.5 (100)
[M + H – 2 HCl]+. C37H66Cl2N2O4 (673.85): calcd. C 65.95, H
9.87, Cl 10.52, N 4.16; found C 65.84, H 9.77, Cl 10.42, N 4.11.
Addition of 5% aq. NaOH to 8a·2HCl followed by extraction with
TMBE provided 8a quantitatively.
(2R,4ЈR,8ЈR)-5-(Morpholinomethyl)-γ-tocopherol (7a): The Man-
nich reagent derived from morpholine (9.37 g, 80 mmol) was added
to mechanically stirred 3 (8.4 g, 20 mmol) at room temp. While
adding morpholine, the yellowish mixture became tarnished. Then
the mixture was heated to 80 °C within 15 min. After 0.5 h, the
aminomethylation was complete [control by GC, silylated deriva-
tive; tR(3) = 5.94 min, tR(7a) = 10.01 min]. Workup as described
for 6a afforded 7a (11.1 g) as a faint yellowish oil forming colour-
less crystals at room temp., purity 97.9% (GC, silylated derivative).
For further characterization, 7a (1 g) was dissolved in MeOH
(22 mL) and recrystallized while stirring the solution and slowly
cooling to 0 °C, precipitating colourless crystals. M.p. 40–41 °C. IR
(KBr): ν = 3444 (w), 2954 (vs, sh), 2927 (vs), 2861 (s, sh), 1461 (vs),
(2R,4ЈR,8ЈR)-5-(Morpholinomethyl)-δ-tocopherol (9): The Mannich
˜
1420 (m), 1379 (m), 1328 (m), 1302 (m), 1265 (s), 1114 (vs), 994 reagent derived from morpholine (2.8 g, 24 mmol, 1.2 equiv.) was
(m), 912 (m), 867 (m) cm–1. 1H NMR (250 MHz, CDCl3): δ = 0.84– added to mechanically stirred 4 (8.2 g, 20 mmol) at room temp.
1.6 (m, 36 H), 1.76 (m, 2 H), 2.11 (s, 3 H), 2.14 (s, 3 H), 2.60 (m,
During addition of the morpholine reagent, the yellowish mixture
6 H), 3.64 (s, 2 H), 3.74 (m, 4 H), 10.59 (s, 1 H) ppm. EI-MS: m/z became tarnished. Then the mixture was heated to 80 °C within
(%) = 515 (21) [M]+·, 428 (100) [M – C4H9NO]+, 203 (26), 165 (26) about 15 min. After 1.5 h, the aminomethylation was almost com-
[C10H13O2]+. C33H57NO3 (515.82): calcd. C 76.84, H 11.14, N 2.72;
found C 76.93, H 11.19, N 2.74. The crude residue can be purified
by column chromatography (hexane/AcOEt, 8:1). Alternatively, 7a
can be converted into the corresponding hydrochloride as described
for the synthesis of 6a·HCl. Recrystallization of 7a·HCl (0.5 g)
from AcOEt (10 mL) afforded 7a·HCl as colourless crystals in the
plete (control by GLC, silylated derivative). Workup as described
for 6a afforded 9 (10.3 g) as a yellowish oil, 90.0% (GLC, silylated
derivative). IR (film): ν = 2952 (vs, sh), 2926 (vs), 2853 (s, sh), 1468
˜
(vs), 1378 (s), 1302 (m), 1223 (s), 1158 (m), 1120 (vs), 991 (m), 913
1
(m) cm–1. H NMR (400 MHz, CDCl3): δ = 0.70–1.64 (m, 36 H),
1.76 (m, 2 H), 2.11 (s, 3 H), 2.57–2.61 (6 H), 3.64 (s, 2 H), 3.74 (m,
4 H), 6.53 (s, 1 H), 10.4 (s, 1 H) ppm. EI-MS: m/z (%) = 501
(42) [M]+·, 414 (100) [M – C4H8O]+, 189 (24), 43 (13). C32H55NO3
form of druses. M.p. 164–166 °C. IR (KBr): ν = 3179 (m), 2954
˜
(vs, sh), 2922 (vs), 2869 (s, sh), 2618 (m), 1461 (vs), 1429 (s, sh),
1379 (s), 1350 (m), 1342 (m, sh), 1295 (s), 11 74 (s), 11 30 (s), 1118 (501.80): calcd. C 76.60, H 11.05, N 2.79; found C 76.40, H 11.09,
(s), 1091 (s), 1080 (s) cm–1. 1H NMR (250 MHz, CDCl3): δ = 0.82–
1.66 (m, 36 H), 1.81 (t, J = 6.8 Hz, 2 H), 2.11 (s, 3 H), 2.22 (s, 3
H), 2.69 (t, J = 6.8 Hz, 2 H), 3.04 (m, 2 H), 3.42 (m, 2 H), 3.89
N 2.94. This raw product was dissolved in TBME (400 mL) and
gaseous HCl (30 mmol) was passed through the solution. A colour-
less precipitate was obtained. Removal of the solvent in vacuo and
(m, 2 H), 4.23 (m, 2 H), 4.40 (m, 2 H), 7.35 (s, 1 H), 11.30 (s, 1 H) recrystallization of the solid residue from warm acetone (120 mL)
ppm. EI-MS: m/z (%) = 515.5 (24) [M – HCl]+·, 428.3 (100) [M –
afforded 9·HCl (8.9 g, 83% yield) as colourless crystals in the form
HCl – C4H9NO]+, 203 (14), 165 (7) [C10H13O2]+. C33H58ClNO3 of druses. M.p. 152–154 °C. IR (KBr): ν = 3112 (s), 2951 (vs, sh),
˜
(552.28): calcd. C 71.77, H 10.59, Cl 6.42, N 2.54; found C 71.71,
H 10.69, Cl 6.52, N 2.57. Addition of 5% aq. NaOH to 7a·HCl
followed by extraction with TMBE provided 7a quantitatively.
2924 (vs), 2856 (s, sh), 2600 (m), 2555 (m, sh), 1460 (vs), 1423 (m),
1377 (s), 1366 (m, sh), 1237 (m), 1226 (s), 1125 (vs), 1104 (m), 918
1
(m) cm–1. H NMR (250 MHz, CDCl3): δ = 0.84–1.65 (m, 36 H),
1.79 (t, J = 6.5 Hz, 2 H), 2.09 (s, 3 H), 2.78 (t, J = 6.5 Hz, 2 H),
3.09 (m, 2 H), 3.40 (d, J = 12.0 Hz, 2 H), 3.89 (d, J = 12.0 Hz, 2
H), 4.22 (m, 4 H), 6.90 (s, 1 H), 8.04 (s, 1 H), 11.49 (s, 1 H) ppm.
ISP-MS: m/z (%) = 515.5 (100) [M – HCl]+. C32H56ClNO3 (538.26):
calcd. C 71.41, H 10.49, Cl 6.59, N 2.60; found C 71.45, H 10.45,
Cl 6.63, N 2.80. The mother liquor of the above crystallization was
treated with aq. NaOH and extracted with TBME as described
below. Concentration under reduced pressure afforded a yellow oil
containing 1.81% 4, 42.92% 9, 33.58% 10 and 13.24% 8a. Com-
pound 10 was identified by 1H NMR spectroscopy (400 MHz,
CDCl3, only characteristic and identified signals): δ = 1.23 (s, 3 H),
2.10 (s, 3 H), 3.70 (s, 2 H), 6.41 (s, 1 H) ppm. Assignment of the
signals was possible because of the known content of the mother
liquor and by comparison with the 1H NMR data of 9. Tocopherol
salt 9·HCl (8.9 g, 16.6 mmol) was dissolved in TBME (100 mL) and
5% aq. NaOH (50 mL) was added. The resulting suspension was
extracted five times with 50 mL of TBME each and the collected
organic layers were dried (K2CO3). Concentration to dryness af-
forded 9 (8.3 g, 100% yield) as a slightly yellow oil.
(2R,4ЈR,8ЈR)-5,7-Bis(morpholinomethyl)-δ-tocopherol (8a): The
Mannich reagent derived from morpholine (37.5 g, 320 mmol) was
added to mechanically stirred 4 (16.4 g, 40.8 mmol) at room temp.
During the addition, the yellowish mixture became tarnished. Then
the mixture was heated to 135 °C within about 60 min as described
for the preparation of 6a. After 6 h, the aminomethylation was
complete [control by GC, silylated derivatives; tr(4) = 5.37 min,
tr(9) = 9.03 min, tr(8a) = 16.41 min]. Workup as described for 6a
afforded 8a (24.5 g) as a yellowish oil. Purity 96.3% (GC, silylated
derivative). IR (film): ν = 2926 (vs), 2852 (s, sh), 1466 (vs), 1378
˜
(m), 1302 (m), 1222 (s), 1159 (m), 1119 (vs), 990 (m), 864 (m) cm–1.
1H NMR (400 MHz, CDCl3): δ = 0.70–1.64 (m, 36 H), 1.76 (m, 2
H), 2.15 (s, 3 H), 2.51 (m, 8 H), 2.75 (t, J = 6.8 Hz, 2 H), 3.56 [s,
2 H, NCH2C(5)], 3.63 [s, 2 H, NCH2C(7)], 3.75 (m, 8 H), 10.6 (s,
1 H) ppm; NCH2C(5) and NCH2C(7) were assigned on the basis
of NOE investigations involving the irradiation of CH3C(8). EI-
MS: m/z (%) = 600.5 (10) [M]+, 513 (98) [M – C4H8O]+, 426 (100)
[M – 2 C4H8O]+, 203 (29), 165 (65) [C10H13O2]+, 43 (33).
C37H64N2O4 (600.93): calcd. C 73.95, H 10.74, N 4.66; found C
74.0, H 10.84, N 4.65. The crude residue can be purified by column
chromatography (hexane/AcOEt, 4:1 Ǟ AcOEt). Alternatively, 8a
can be converted into the corresponding hydrochloride as described
for the synthesis of 6a·HCl. From the TBME solution the crystal-
Catalytic Hydrogenation of Aminomethylated Tocopherol Homo-
logues:[7] The aminomethylated products (6–10) were hydrogenated
in a 380-mL steel autoclave at 180 °C and 28–34 bar for 6–24 h
using 5% palladium on carbon as the catalyst in TBME (5 wt.-%
Eur. J. Org. Chem. 2007, 1176–1183
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1181

T. Netscher, F. Mazzini, R. Jestin
FULL PAPER
in the normal case). After the hydrogenation, the catalyst was fil-
tered off through Speedex (filter aid), rinsed with TBME and the
solvent removed under reduced pressure. The liquid residue was
distilled bulb-to-bulb in vacuo (180–200 °C/0.03 mbar).
(2R,4ЈR,8ЈR)-[C5-2H]-α-Tocopherol (13): Reduction of 7a (1.03 g,
2 mmol) by NaBD4/NaOH as described above for the transforma-
tion of 7a to 1 afforded 13 (810 mg, 94% yield) as a yellowish oil.
1H NMR (250 MHz, CDCl3): δ = 0.7–1.9 (m, 38 H), 2.11 (br. s, 5
H), 2.16 (s, 3 H), 2.61 (t, J = 7 Hz, 2 H), 4.17 (br. s, 1 H) ppm. 13
C
Reduction of 7a to 1 Using NaCNBH3: Compound 7a (516 mg,
1 mmol) was dissolved in iBuOH (4 mL) and NaBH3CN (283 mg,
4.5 mmol, 4.5 equiv.) was added. The colourless suspension was re-
fluxed (108 °C) and stirred for a further 4 h. Afterwards, the mix-
ture was cooled to room temp., diluted with Et2O (10 mL) and
acidified to pH = 1 by addition of 2  HCl (12 mL). The aqueous
layer was separated and washed with Et2O (2ϫ5 mL). The organic
layers were washed with satd. aq. NaHCO3 (10 mL), satd. aq. NaCl
(10 mL) and dried (MgSO4). Concentration under reduced pressure
to dryness afforded raw 1 (412 mg, 95.9% pure, GC, acetylated
derivative). Further purification of 335 mg of raw 1 by column
chromatography (20 g SiO2, hexane/AcOEt, 9:1) gave pure 1
(332 mg, 89.6% yield) as a yellowish oil. Purity approx. 100% (GC,
acetylated derivative). 1H and 13C NMR data are in accordance
with reported data.[19]
NMR (75 MHz, CDCl3): δ = 11.3 (m), 11.8, 12.2, 19.9, 21.0, 21.2,
22.6, 22.7, 23.8, 24.6, 24.9, 28.1, 31.5, 32.7, 32.8, 37.3, 37.5, 37.6,
37.7, 39.4, 39.8, 74.6, 117.3, 118.7, 121.1, 122.6, 144.4, 145.6 ppm.
ISP-MS: m/z (%) = 432.7 (100) [M + H]+; deuterium content
Ͼ99%. C29H49DO2 (431.71): calcd. C 80.68, H 11.91; found C
80.65, H 11.94.
(2R,4ЈR,8ЈR)-[C5-2H2]-α-Tocopherol (14): Reduction of 12 (777 mg,
1.5 mmol) using NaBH4/NaOH as described above for the transfor-
mation of 7a to 1 afforded 14 (610 mg, 94% yield) as a dark yellow-
1
ish oil. H NMR (250 MHz, CDCl3): δ = 0.7–1.9 (m, 38 H), 2.09
(s, 1 H), 2.11 (s, 3 H), 2.16 (s, 3 H), 2.58 (t, J = 7 Hz, 2 H), 4.16
(br. s, 1 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 11.3 (m), 11.8,
12.2, 19.9, 21.0, 21.2, 22.6, 22.7, 23.8, 24.6, 24.9, 28.1, 31.5, 32.7,
32.8, 37.3, 37.5, 37.6, 37.7, 39.4, 39.8, 74.6, 117.3, 118.7, 121.1,
122.6, 144.4, 145.6 ppm. ISP-MS: m/z (%)
= 433.7 (100)
Reduction of 7a to 1 Using NaBH4/NaOH: Compound 7a (516 mg,
1 mmol) was dissolved in iBuOH (20 mL) and NaBH4 (160 mg,
4 mmol, 4 equiv.) and then NaOH (80 mg, 2 mmol, 2 equiv.) was
added. The light yellow suspension was dipped in an oil bath at
120 °C and refluxed and stirred for 7 h. Afterwards, the mixture
was cooled to room temp., further NaBH4 was added (1 equiv.) and
the mixture dipped again in the heated oil bath (120 °C) and re-
fluxed for an additional 2 h. Afterwards, the mixture was cooled to
room temp., diluted with Et2O (10 mL) and acidified to pH = 4 by
addition of 2  HCl (ca. 11 mL). The aqueous layer was separated
and washed with Et2O (2ϫ10 mL). The organic layers were washed
with satd. aq. NaHCO3 (10 mL), satd. aq. NaCl (10 mL) and dried
(MgSO4). After concentration under reduced pressure, the residue
(raw 1, purity Ͼ99%, HPLC) was purified by column chromatog-
raphy (20 g SiO2, hexane/AcOEt, 8:1), affording 1 (408 mg, 95%
yield) as a yellowish oil. Purity approx. 100% (GC, acetylated de-
[M + H]+; deuterium content 98%. C29H48D2O2 (432.72): calcd. C
80.49, H 12.11; found C 80.46, H 12.09.
(2R,4ЈR,8ЈR)-[C5-2H3]-α-Tocopherol (15): Reduction of 12 (800 mg,
1.54 mmol) using NaBD4/NaOH as described above for the trans-
formation of 7a to 1 afforded 15 (610 mg, 95% yield) as a yellow
1
oil. H NMR (250 MHz, CDCl3): δ = 0.7–1.9 (m, 38 H), 2.11 (s, 3
H), 2.18 (s, 3 H), 2.60 (t, J = 7 Hz, 2 H), 4.19 (br. s, 1 H) ppm. 13
C
NMR (75 MHz, CDCl3): δ = 11.3 (m), 11.8, 12.2, 19.9, 21.0, 21.2,
22.6, 22.7, 23.8, 24.6, 24.9, 28.1, 31.5, 32.7, 32.8, 37.3, 37.5, 37.6,
37.7, 39.4, 39.8, 74.6, 117.3, 118.7, 121.1, 122.6, 144.4, 145.6 ppm.
ISP-MS: m/z (%) = 434.7 (100) [M + H]+; deuterium content
98.5%. C29H47D3O2 (433.72): calcd. C 80.31, H 12.32; found C
80.34, H 12.34.
(2R,4ЈR,8ЈR)-5-(Morpholino-[2H2]methyl)-δ-tocopherol (16): Man-
nich reagent (510 mg, 2.8 mmol, 1.3 equiv.) was added to δ-tocoph-
erol (870 mg, 2.16 mmol) and the resulting mixture was stirred at
80 °C for 2 h, TLC (hexane/EtOAc, 9:1) control confirming the end
of the reaction. After cooling to room temp., the mixture was di-
luted with TBME (15 mL) and washed with H2O until the wash
solution became neutral. The organic layer was dried with K2CO3.
Concentration to dryness gave 16 (860 mg, 80% yield) as a yellow
1
rivative). H and 13C NMR data are in accordance with reported
data.[19]
Characterization of n-Butyl α-Tocopheryl Ether (11): The n-butyl
ether 11 was purified by column chromatography (hexane/AcOEt,
9:1) of the crude residue resulting from Bu4NBH4 reduction of 7a
carried out as described above for the reduction of 7a to 1 by
NaCNBH3. Characteristic 1H NMR signals in comparison with
1
oil. H NMR (250 MHz, CDCl3): δ = 0.7–1.9 (m, 38 H), 2.11 (s, 3
1
H), 2.54–2.61 (m, 6 H), 3.74 (m, 4 H), 6.50 (s, 1 H), 10.39 (br. s, 1
H) ppm. ISP-MS: m/z (%) = 504.8 (100) [M + H]+; deuterium con-
tent 98.5%. C32H53D2NO3 (503.80): calcd. C 76.29, H 11.40, N
2.78; found C 76.25, H 11.41, N 2.77.
the H NMR spectrum of α-tocopherol: absence of signal of OH;
δ = 3.63 (t, J = 6.9 Hz, 2 H) ppm. 13C NMR (75 MHz, CDCl3): δ
= 11.3, 11.8, 12.2, 14.0, 19.4, 19.6, 19.7, 20.7, 21.0, 22.6, 22.7, 23.8,
24.6, 24.9, 28.1, 31.4, 32.5, 32.7, 32.8, 37.3, 37.5, 37.6, 37.7, 39.4,
39.8, 72.8, 74.7, 117.4, 122.7, 125.8, 127.8, 147.6, 148.4 ppm. ISP-
MS: m/z (%) = 487.8 (100) [M + H]+.
(2R,4ЈR,8ЈR)-[C5-2H3)-β-Tocopherol (17): Reduction of 16 (800 mg,
1.59 mmol) using NaBD4/NaOH as described above for the trans-
formation of 7a to 1 afforded 17 (600 mg, 90% yield) as a yellow
β-Tocopherol (2) by NaBH4/NaOH Reduction of 9: Reduction of 9
(500 mg, 1 mmol) using NaBH4/NaOH as described above for the
reduction of 7a to 1 afforded 2 (380 mg, 95% yield). 1H and 13C
NMR data are in accordance with reported data.[19]
1
oil. H NMR (250 MHz, CDCl3): δ = 0.7–1.9 (m, 38 H), 2.11 (s, 3
H), 2.59 (t, J = 7 Hz, 2 H), 4.25 (br. s, 1 H), 6.50 (s, 1 H) ppm. 13
C
NMR (75 MHz, CDCl3): δ = 11.1 (m), 15.8, 19.6, 19.8, 21.0, 21.1,
22.7, 22.8, 23.8, 24.5, 24.9, 28.0, 31.3, 32.7, 32.8, 37.3, 37.5, 37.6,
37.7, 39.4, 39.6, 74.5, 115.3, 119.1, 120.3, 124.1, 145.7, 146.0 ppm.
ISP-MS: m/z (%) = 420.8 (100) [M + H]+; deuterium content
98.5%. C28H45D3O2 (419.70): calcd. C 80.13, H 12.25; found C
80.19, H 12.23.
(2R,4ЈR,8ЈR)-5-(Morpholino-[2H2]methyl)-γ-tocopherol (12): Prepa-
ration from 3 (4.21 g, 10 mmol) and the dideuteriated Mannich rea-
gent derived from morpholine (4.77 g) and paradeuterioformal-
dehyde as described above for 7a afforded 12 (5.14 g, 98.3% yield)
as colourless crystals. M.p. 39–41 °C. Purity: 97.4% (GLC, silylated
derivative). 1H NMR (250 MHz, CDCl3): δ = 0.83–1.65 (m, 36 H),
1.78 (m, 2 H), 2.10 (s, 3 H), 2.14 (s, 3 H), 2.60 (m, 6 H), 3.76 (m,
4 H), 10.57 (s, 1 H) ppm. ISP-MS: m/z (%) = 518.4 (100)
[M + H]+; deuterium content Ͼ99%. C33H55D2NO3 (517.84):
calcd. C 76.54, H 11.48, N 2.70; found C 76.68, H 11.47, N 2.70.
Acknowledgments
We thank Mrs. Franziska Bähler, Mr. Jörg Schneider and Mr.
Heinz Schneider for experimental assistance and our collegues
1182
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1176–1183

Tocopherols by Hydride Reduction of Dialkylamino Derivatives
FULL PAPER
R. Jestin in Fifth International Electronic Conference on Syn-
thetic Organic Chemistry (ECSOC-5), “Molecular Diversity
Preservation International”, Basel, Switzerland, 2001; http://
www.mdpi.net/ecsoc/ecsoc-5/Papers/c0007/c0007; c) R. K.
Müller, H. Schneider (F. Hoffmann–La Roche AG), Eur. Pat.
Appl. 0735033B1, 1996, priority 1995 (CH 87895); Chem.
Abstr. 1996, 125, 301271.
from F. Hoffmann–La Roche, Basel, for spectroscopic analyses.
The University of Pisa and Consorzio Pisa Ricerche are also ac-
knowledged for financial support.
[1]
[2]
[3]
a) W. A. Skinner, R. M. Parkhurst, J. Scholler, J. Med. Chem.
1969, 12, 64–66; b) H. J. Kayden, M. G. Traber, J. Lipid Res.
1993, 34, 343–358; c) R. Brigelius-Flohe, M. G. Traber, FASEB
J. 1999, 13, 1145–1155.
a) K. U. Baldenius, L. von dem Bussche-Hünnefeld, E. Hilge-
mann, P. Hoppe, R. Stürmer in Ullmann’s Encyclopedia of In-
dustrial Chemistry, VCH, Weinheim, 1996, vol. A27, pp. 478–
488, 594–597; b) T. Netscher, Chimia 1996, 50, 563–567.
a) L. Weisler (Distillation Products Inc., Rochester, NY), U.S.
Pat. 2486539, 1949; Chem. Abstr. 1950, 44, 10278; b) L. Weisler
(Eastman Kodak, Rochester, NY), U.S. Pat 2519863, 1950;
Chem. Abstr. 1951, 45, 3760.
[8] a) F. Gu, T. Netscher, J. Atkinson, J. Labelled Cpd. Ra-
diopharm. 2006, 49, 733–743; b) E. Alpi, F. Mazzini, T.
Netscher, P. Salvadori, 4th Italian-Spanish Symposium on Or-
ganic Chemistry (ISSOC-4), 31st August to 3rd September,
2002 Perugia, Italy, contribution PO-44.
[9] S. Narasimhan, R. Balakumar, Aldrichim. Acta 1998, 31, 19–
26, and references cited herein.
[10] H. C. Brown, S. Narasimhan, Y. M. Choi, J. Org. Chem. 1982,
47, 4702–4708.
[11] H. C. Brown, S. Krishnamurthy, Tetrahedron 1979, 35, 567–
607.
[4] a) T. Nakamura, S. Kijima, Chem. Pharm. Bull. 1971, 19, 2318–
2324; b) W. S. Baldwin, S. M. Willging, M. B. Siegel (Henkel
Corporation), Eur. Pat. Appl. 0159018, priority 1984 (U.S. Pat.
601194); Chem. Abstr. 1986, 104, 95446; c) J. G. Baxter (East-
man Kodak, Rochester, NY), U.S. Pat. 2592531, 1952; Chem.
Abstr. 1953, 47, 4875.
[5] a) L. Weisler (Eastman Kodak, Rochester, NY), U.S. Pat.
2640058, 1951; Chem. Abstr. 1954, 48, 42536.
[6] a) K. Brüggemann, J. R. Herguijuela, T. Netscher, J. Riegl
(Hoffmann–La Roche AG), Eur. Pat. Appl. 0769497A1, 1997,
priority 1995 (CH 2951/95); Chem. Abstr. 1997, 126, 317507;
[12] J. R. Berschied, K. F. Purcell, Inorg. Chem. 1970, 9, 624–629.
[13] E. H. Jensen, A Study on Sodium Borohydride, NytNordisk
Forlag Amold Busck, Copenhagen, 1954.
[14] T. Rosenau, G. Ebner, A. Stanger, S. Perl, L. Nuri, Chem. Eur.
J. 2005, 11, 280–287, and references cited herein.
[15] T. Rosenau, L. Gille, E. Kloser, F. Mazzini, T. Netscher, unpub-
lished results.
[16] a) K. U. Ingold, L. Hughes, M. Slaby, G. W. Burton, J. Lab-
elled Cpd. Radiopharm. 1987, 24, 817–831; b) L. Hughes, M.
Slaby, G. W. Burton, K. U. Ingold, J. Labelled Cpd. Ra-
diopharm. 1990, 28, 1049–1057.
b) J. Riegl, K. Brüggemann, T. Netscher in Proc. ECSOC-3 [17] A. Brändström, U. Junggren, B. Lamm, Tetrahedron Lett.
(1999), Proc. ECSOC-4 (2000) (Ed.: E. Pombo-
Villar), MDPI, Basel, Switzerland, 2000, pp. 1542–1547;
http://www.mdpi.org/ecsoc-4.htm; http://pages.unibas.ch/mdpi/
ecsoc-4/c0035/c0035.htm.
1972, 13, 3173–3176.
[18] a) W. Walther, T. Netscher, Chirality 1996, 8, 397–401; b) N.
Cohen, C. G. Scott, C. Neukom, R. J. Lopresti, G. Weber, G.
Saucy, Helv. Chim. Acta 1981, 64, 1158–1173.
[19] J. K. Baker, C. W. Myers, Pharmaceutical Res. 1991, 8, 763–
770.
[7] a) T. Netscher, R. K. Müller, J. Schneider, H. Schneider, P.
Bohrer, R. Jestin in 12th European Symposium on Organic
Chemistry, Groningen, 13–18th July 2001, poster; b) T.
Netscher, R. K. Müller, J. Schneider, H. Schneider, P. Bohrer,
Received: October 4, 2006
Published Online: January 8, 2007
Eur. J. Org. Chem. 2007, 1176–1183
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1183