Chiral building blocks from biomass: 2,5-diamino-2,5-dideoxy-1,4-3,6- dianhydroiditol

Article abstract of DOI:10.1016/j.tet.2010.11.031

An efficient route towards the synthesis of 2,5-diamino-2,5-dideoxy-1,4-3, 6-dianhydroiditol 4 has been developed resulting in significant improvements in both isolated yields and purity when compared to literature procedures. As a consequence, resin-grad

Full text of DOI:10.1016/j.tet.2010.11.031

Tetrahedron 67 (2011) 383e389
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Chiral building blocks from biomass: 2,5-diamino-2,5-dideoxy-1,4-3,6-
dianhydroiditol
,
,
,
,b
Shanmugam Thiyagarajan a b, Linda Gootjes a b, Willem Vogelzang a b, Jing Wu a
,
Jacco van Haveren a b, Daan S. van Es a
,
,b,
*
a Food & Bio-based Research, Wageningen University and Research Centre, P.O. Box 17, 6700 AA Wageningen, The Netherlands
b Dutch Polymer Institute, P.O. Box 902, 5600 AX, Eindhoven, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 July 2010
Received in revised form 21 October 2010
Accepted 8 November 2010
Available online 13 November 2010
An efficient route towards the synthesis of 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol 4 has been
developed resulting in significant improvements in both isolated yields and purity when compared to
literature procedures. As a consequence, resin-grade 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol 4
has become available for laboratory scale step-growth polymer synthesis. Additionally, an interesting
renewable chiral 2-amino-2-deoxy-1,4-3,6-dianhydroiditol 10, has been isolated.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
2,5-Diamino-2,5-dideoxy-1,4-3,6-
dianhydroiditol
Bio-based
Building blocks
Renewable resources
1. Introduction
These rigid bicyclic diols have several interesting properties in
polymer applications.3 Thiem and Lueders for instance have shown
Recently, the sharp increase in crude oil prices and increasing
concerns about rising carbon dioxide levels and associated global
warming effects have sparked a growing interest towards bio-
based building blocks for the polymer industry. One group of re-
newable building blocks that gained considerable attention is the
anhydro sugar polyols, also known as isohexides.1 These isohexides
can be obtained from C6-alditols such as sorbitol or mannitol via
acid catalysed cyclodehydration (Scheme 1).2 Of the three known
that incorporation of isohexides into polyesters such as PET results
in a significant increase in glass transition temperature (Tg), which
could widen the scope of applications of these materials.4 Although
the first investigations intothe use of isohexides as buildingblock for
polymers have been reported as far back as in 1965,5 it was the more
recent extensive work by Dupont and Roquette on the incorporation
of high purity isosorbide 1b in polyesters that triggered renewed
interest in these molecules.6 Our group has also been actively in-
volved in isosorbide chemistry for over ten years, with recently an
increasing focus on applications in performance materials.7
isohexide isomers, only dianhydro-1,4-3,6-
D-sorbitol or isosorbide
is currently produced on (small) industrial scale.
Whereas the parent isohexides [isomannide 1a (endoeendo
conformation), isosorbide 1b (endoeexo conformation), and to
a lesser extent isoidide 1c (exoeexo conformation)] have been
studied extensively as bio-based monomers, only few reports have
been made on the use of the corresponding dideoxy-dia-
minoisoidides.9,13,19,20 This is undoubtedly due to the limited syn-
thetic accessibility of these substances, combined with inadequate
purities obtained with the known procedures. During our on-going
investigations towards the viability of bio-based polyamides and
polyurethanes with performance polymers characteristics, we re-
quired substantial amounts of high purity amino isohexides. This
prompted us to reinvestigate the known literature procedures.
To date there are only three known synthetic routes8,11,13 to
unsubstituted isohexide diaminoisoidides. All of these procedures
HO
H
hydrogenation
acid catalysed
cyclodehydration
O
C-6 Aldose/Ketose
C-6 Alditol
O
H
1
OH
C2/C5 endo = isomannide 1a
C2 endo/C5 exo = isosorbide 1b
C2/C5 exo = isoidide 1c
Scheme 1.
* Corresponding author. Tel.: þ31 317 481160; fax: þ31 317 483011; e-mail
address: daan.vanes@wur.nl (D.S. van Es).
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.11.031

384
S. Thiyagarajan et al. / Tetrahedron 67 (2011) 383e389
rely on introducing amine functionality by nucleophilic displace-
ment of tosylated or mesylated hydroxyl groups. Since SN2 re-
actions on an exo-leaving group are only effective with very small
nitrogen nucleophiles (e.g., ammonia or azide anion),2 often iso-
mannide, with two endo hydroxyl functionalities is used in these
reactions. The resulting symmetrical 2,5-diamino-2,5-dideoxy-
1,4-3,6-dianhydroiditol will have the idide (exoeexo) conformation,
which is expected to have advantageous effects when incorporated
into step-growth polymers.7
The first known route involves conversion of the isohexide to
the corresponding bismesylate 5, followed by nucleophilic sub-
stitution with azide 6.8 Hydrogenation of the isohexide azides
yields the corresponding 2,5-diamino-2,5-dideoxy-1,4-3,6-dia-
nhydroiditols 4 in 88% yield (Scheme 2).9 The major drawback of
this route is the potentially explosive character of the diazide,
which prohibits its application on larger scale.10 The second route
consists of reacting isohexide bistosylates 2 with methanolic am-
monia at 170 ꢀC in a sealed vessel for 30 h (Scheme 2).11 This
method is severely hampered by the necessity to use a large
volume autoclave in order to obtain reasonable amounts of
2. Results and discussion
The first step in the reaction sequence, the tosylation of iso-
mannide, is virtually quantitative.11 After recrystallisation from eth-
anol, the product can be obtained in high purity as colorless needles.
The subsequent introduction of a nitrogen functionality via a nucle-
ophilic displacement of the tosylate group is however less trivial.
According to the procedure described by Cope and Shen (DMF,110 ꢀC,
40 h) the desired 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhy-
droiditol 3 (Scheme 2) was obtained in an unsatisfactory yield of only
34%.13 In contrast, Kuszmann and Medgyes reported an isolated yield
of62%of the 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol
3 starting from 1,4-3,6-dianhydro-2,5-di-O-mesyl-
However, while we obtained 1,4-3,6-dianhydro-2,5-di-O-p-tosyl-
mannitol 2 in an isolated yield of >90% and in >99% purity after re-
crystallization, the 1,4-3,6-dianhydro-2,5-di-O-mesyl- -mannitol 5
was obtained in only 83% yield. Furthermore, crude 2,5-diph-
thalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3 was obtained in
D
-mannitol 5.14
D-
D
94% yield from the 1,4-3,6-dianhydro-2,5-di-O-p-tosyl-
while in case of the 1,4-3,6-dianhydro-2,5-di-O-mesyl-
D-mannitol 2,
D-mannitol 5
product (50 g of 1,4:3,6-dianhydro-2,5-di-O-p-tosyl-
D-mannitol 2
the crude yield of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhy-
in 1.5 L methanolic ammonia). Although in the original paper no
isolated yields were given, preliminary results in our own labo-
ratories indicate that the isolated yields are in very low range
(<20%).12
droiditol 3 did not exceed 80%. Based on this study, we used 1,4-
3,6-dianhydro-2,5-di-O-p-tosyl- -mannitol 2 as a starting material.
D
In order to obtain insight in the selectivity of the reaction with
potassium phthalimide, the reaction was first performed on small
O
O
N
TosO
O
O
e
K-phthalimide / DMF
n
H
i
d
2 N
i
r
110 oC, 40 h
34%
-
y
h
N
O
p
o
O
8
H
/
1
l
2
,
C
N
/ E
OTos
1
3
C
2
s
O
%
O
T
%
t
O
5
8
0
3
e
H
t
u
o
r
3
HO
TosO
H2N
TsCl / pyridine
24 h, rt
O
O
O
NH3 / CH3OH
30 h, 170 o
route 2
C
O
O
O
2
OTos
1a OH
NH2
4
H
O
e
M
r
o
M
/
u
s
C
t
H 2
;
e
l
i
1
e
s
/ p
C
/
p
MsO
N3
d
P
yr
0
4
i
di
;
h
%
O
%
8
O
5
n
NaN3 / DMF
2 h, 120 oC
86%
2
1
8
O
O
OMs
N3
6
5
Scheme 2. Synthetic routes to unsubstituted isohexide diaminoisoidides.
The third route comprises three steps starting from isomannide
1a; tosylation, nucleophilic substitution with potassium phthali-
mide, and subsequent deprotection with hydrazine hydrate
(Scheme 2).13 This route also has severe drawbacks, such as low
overall isolated yield in the second step (34% of the 2,5-diph-
thalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3), and low isolated
yield of high purity 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhy-
droiditol 4 (<13%). However, this procedure was the most preferred
with regard to scalability in order to obtain lab-scale quantities
(>50 g) of resin-grade 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhy-
droiditol 4 on short term. Obviously, more catalytic routes towards
diamino isohexides are required to make these building blocks
economically viable, the development of which is currently in
progress in our laboratories.
scale (1.5 g) in deuterated solvents (DMF-d6 and DMSO-d6, 20 mL),
and the progress of the reaction was followed by 1H and 13C NMR
spectroscopy (Table 1). When the reaction was performed in DMF,
the potassium phthalimide had only limited solubility, which se-
verely hampered NMR analysis. As the solubility of the starting
material (1,4-3,6-dianhydro-2,5-di-O-p-tosyl- -mannitol 2) and
D
the desired product (2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dia-
nhydroiditol 3) was excellent in DMSO, this became the solvent of
choice. Furthermore, DMSO is known for its ability to solvate cat-
ions, an increase in the reactivity of the potassium phthalimide was
anticipated.15 A first series of small-scale experiments showed that
at a temperature of 130 ꢀC all of the NMR signals corresponding
with the starting material had disappeared within 24 h. The
resulting reaction mixture consisted mainly of 2,5-diphthalimido-

S. Thiyagarajan et al. / Tetrahedron 67 (2011) 383e389
385
Table 1
Optimization of the synthesis of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3
Starting material
(g, mmol)
Potasssium phthalimide
(g, mmol, equiv)
Time (h)
Temp (ꢀC)
Solvent
Molar yield ratio % in situ NMR
(end of reaction) 3:7
2 (1.64, 3.6)
2 (0.55, 1.2)
2 (1.64, 3.6)
(1.47, 7.9, 2.2)
(0.50, 2.7, 2.2)
(1.48, 8.0, 2.2)
48
42
25
130
110
130
DMSO-d6
DMF-d7 (wet)
DMSO-d6(dry)
35:65
a
67:33
a
The ratio between the 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3 and phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7 could not be calculated because of
huge amount of impurities present in the crude mixture.
2,5-dideoxy-1,4-3,6-dianhydroiditol 3, phthalimido-2-deoxy-1,
4-3,6-dianhydroiditol 7, and minor amounts of other carbohydrate
based species. Work-up of the crude reaction mixture with water
(D2O) and chloroform (CDCl3), followed by NMR analysis showed
that the aqueous phase contained a small amount of isoidide be-
sides the expected potassium tosylate. The identification of isoidide
was based on 1H and 13C NMR spectra of an authentic sample.
NMR analysis of the organic phase showed that it contained
predominantly 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydro-
iditol 3 and phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7 in the
ratio of 2:1. The exoeexo conformation of 3 and 7 was confirmed by
1H-, HH-COSY and NOESY NMR analysis. Variation of reaction pa-
rameters such as time and temperature had little effect on the 2:1
ratio (Table 1). Optimal reaction conditions in DMSO thus far were
stirring at 130 ꢀC for 24 h, using 2.2 equiv of potassium phthalimide.
Based on the optimized conditions, the reaction was performed on
50 g and 100 g scales (Table 2).
6-dianhydroiditol 3 was more effective from chloroform than from
ethanol.13 All of these improvements resulted in an increase in
isolated yield in this step to 84% (from 34%). Purification of the
chloroform soluble fraction by column chromatography resulted in
the isolation of phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7 in
high purity (95% based in 1H NMR).
The third step in the original synthesis is the deprotection
of the 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol
3
with hydrazine hydrate. Although this reaction is very efficient, the
separation of the desired product 4 from the by-product (phthalhy-
drazide) resulted in very poor isolated yields (<20%) and purity
(<95%) of 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol 4.16
Thus, we explored an alternative method for the deprotection of
phthalimides; refluxing in 6 M HCl/glacial acetic acid.17 Complete
conversion of 8 was achieved after 24 h at reflux temperature. Upon
cooling the reaction mixture, phthalic acid crystallizes from the so-
lution, and can be efficiently removed by suction-filtration. The
resulting mother liquor is subsequently evaporated under reduced
pressure giving the crude 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhy-
droiditol HCl salt 8 in very highyield (92%). Further purification of the
crude 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol HCl salt 8
was achieved by dissolution/precipitation from methanolic/etha-
nolic HCl. Subsequent transformation of the HCl salt into the free 2,5-
diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol 4 was achieved by
treatment with a (purified) basic ion exchange resin (AmberlystÒ
A26), which also removes residual phthalic acid. As a result, the
Apparently phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7 as
well as isoidide were formed by hydrolysis of the tosylate group
under the reaction conditions. Consequently, we investigated the
effect of water in the reaction medium. Surprisingly, 1,4-3,6-dia-
nhydro-2,5-di-O-p-tosyl-D-mannitol 2 remained intact in wet
DMSO, and even the addition of a mild base (K2CO3) did not lead to
any significant changes. However, when potassium hydroxide was
added, fast degradation of the 1,4-3,6-dianhydro-2,5-di-O-p-tosyl-
D-mannitol 2 was observed (together with significant darkening of
the solution). After 2 h, the NMR signals of the starting material had
disappeared; resulting in a multitude of unresolved carbohydrate
and unsaturated signals, which were observed in the 13C NMR
spectrum, indicating complete degradation of the sample. Curi-
ously, it was observed that even after a short time at room tem-
perature the methyl signal of the tosylate group had disappeared,
presumably due to perdeuteration, given the appearance of a small
multiplet in the 13C NMR spectrum at the original chemical shift
value of the methyl signal.
desired 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol 4 was
obtained in high yield (95%) and with high purity (99%). The
overall isolated yield of the 2,5-diamino-2,5-dideoxy-1,4-3,6-dia-
nhydroiditol 4 in four steps was thus increased from 14% to 47%.
Inordertoimprovethetotalvalorizationofthebiomass feed-stock
we investigated the viability of our deprotection strategy on the
phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7. Despite the fact that
isohexides havebeenreported toundergo ring-openingwhen treated
with strong acids for prolonged time,18 hydrolysis of the phthalimido-
Table 2
Synthesis of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3 and optimal reaction conditions
Starting material
(g, mol)
Potasssium phthalimide
(g, mol, equiv)
Time (h)
Temp (ꢀC)
Solvent (dry)
Crude yield % 3 and 7
(g, mol)
Ratiob
2 (50, 0.11)
2a (50, 0.11)
2 (100.1, 0.22)
5 (15, 0.05)
(44.3, 0.24, 2.2)
(44.3, 0.24, 2.2)
(88.7, 0.48, 2.2)
(20, 0.10, 2.2)
24
24
48
24
130
130
110
130
DMSO
DMSO
DMSO
DMSO
94 (41.5, 0.10)
96 (42.7, 0.10)
95 (85, 0.21)
80 (16, 0.04)
1:0
1:0.15
1:0.2
1:0.2
a
Refer synthesis of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3, Route:2 in the experimental procedure.
The ratio between the 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3 and phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7 in isolated crude mixture, according
b
to 1H NMR.
It was envisaged that even a small amount of water in the
2-deoxy-1,4-3,6-dianhydroiditol 7 successfully yielded 2-amino-2-
deoxy-1,4-3,6-dianhydroiditol HCl salt 9 in moderate yield.
These results prompted us to further simplify our procedure.
reaction mixture will influence the hydrolysis of product 3 lead to
by-products. Hence, in order to avoid the by-product formation by
hydrolysis, the reaction was performed in fresh anhydrous
DMSO (<0.005% water), which resulted in a significant increase
in the isolated yield of the 2,5-diphthalimido-2,5-dideoxy-
1,4-3,6-dianhydroiditol 3. Furthermore, it was found that
selective precipitation of the 2,5-diphthalimido-2,5-dideoxy-1,4-3,
Starting from 50 g 1,4-3,6-dianhydro-2,5-di-O-p-tosyl-D-mannitol
2, the substitution reaction with potassium phthalimide was per-
formed in DMSO for 24 h at 130 ꢀC. After completion, the reaction
mixture was diluted with water and extracted with chloroform.
The crude product mixture was subsequently hydrolysed in 6 M

386
S. Thiyagarajan et al. / Tetrahedron 67 (2011) 383e389
HCl/glacial acetic acid without prior separation of the bisph-
thalimide. After removal of the phthalic acid by filtration, the crude
product contained the HCl salts of the 2,5-diamino-2,5-dideoxy-
1,4-3,6-dianhydroiditol 8, the 2-amino-2-deoxy-1,4-3,6-dianhy-
droiditol 9 and residual phthalic acid (according to NMR analysis).
Virtually pure 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol
HCl salt 8 was obtained by selective precipitation from ethanolic
HCl. Evaporation of the filtrate gave the crude 2-amino-2-deoxy-
1,4-3,6-dianhydroiditol HCl salt 9 (Scheme 3). After treatment
with AmberlystÒ A26 resin and decolourization with activated
ꢁ99.9%, anhydrous), pyridine (Merck, p.a.), hydrochloric acid (re-
agent grade 37%, Sigma Aldrich), acetic acid (Merck; glacial; 100%
p.a.), chloroform (Merck, p.a.), chloroform-d (99.8 atom%
D
Aldrich), DMSO-d6 (99.9 atom% D, Aldrich) were stored on dried
ꢀ
molecular sieves 4 A. D2O (>99.8%, Merck) was used as received,
ethanol (Merck, p.a.), methanol (p.a. Merck), ethanolic HCl
1.25 N (Fluka), Silica gel (Alfa Aesar, 60; 0.040e0.063 mm;
ꢀ
230e400 mesh), Molecular sieves (Acros, 5 A, 8e12 mesh), diethyl
ether (Merck, 99.7%), magnesium sulfate (Acros Organics, 99% extra
pure, dried, contains 3e4 mol of water), CeliteÒ 545 coarse (Fluka),
Sicapent (with indicator, Merck). AmberlystÒ A26 (Aldrich)
hydroxide form (Strongly basic, macroreticular resin with quater-
nary ammonium functionality from Rohm & Haas Co): prior to use,
the resin was washed with demineralised water (100 mL) by
sonification in an ultrasonic bath at room temperature for 10 min.
The water layer was subsequently removed by decantation. This
carbon, 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol
obtained in an overall yield of 47% (starting from the 1,4-3,6-dia-
nhydro-2,5-di-O-p-tosyl-
-mannitol 2) and in >99% purity (1H
4 was
D
NMR). Analogously, 2-amino-2-deoxy-1,4-3,6-dianhydroiditol HCl
salt 9 was converted to pure 2-amino-2-deoxy-1,4-3,6-dianhy-
droiditol 10.
O
O
TosO
HO
N
N
O
O
O
O
K-Phthalimide
-TosCl
p
O
O
+
Pyridine
3 h, rt
DMSO
O
O
O
O
2
24 h, 130 oC
O
1a
OTos
OH
N
OH
7
(90%)
O
(16%)
3
(84%)
6N HCl/AcOH (4:1)
24 h, 135 oC
ClH3N
HO
HO
H2N
O
O
O
O
Amberlyst A26-OH
H2O
1 h, 30 o
Amberlyst A26-OH
+
H2O
1 h, 30 oC
O
O
O
O
C
8
NH3Cl
NH3Cl
9
10 NH2
4
NH2
(94%)
(29%)
(70.5%)
(70%)
Scheme 3. Synthesis of 2,5-diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol 4 and 2-amino-2-deoxy-1,4-3,6-dianhydroiditol 10.
3. Conclusions
procedure was repeated five times, until the water layer remained
colourless. Activated carbon (Norit SA2 94007-6, steam activated
carbon with a high adsorptive capacity) was a gift from Norit
Nederland BV. Isoidide (R&D product) was a gift from Roquette
Freres. All the chemicals were used as received, unless denoted
otherwise.
Fourier transform infrared (FT-IR) spectra were obtained on
a Varian Scimitar 1000 FT-IR spectrometer equipped with a Pike
MIRacle ATR Diamond/ZnSe single reflection plate and a DTSG-
detector. The measurement resolution was set at 4 cmꢂ1, and the
spectra were collected in the range 4000-650 cmꢂ1 with 32 co-
added scans. NMR spectra were recorded on a Bruker Avance III
spectrometer operating at 400.17 MHz (1H) and 100.62 MHz (13C).
Gas chromatography was performed on an Interscience Focus GC
equipped with an AS 3000 series auto sampler. Injection volume
Driven by a need for resin-grade 2,5-diamino-2,5-dideoxy-
1,4-3,6-dianhydroiditol 4 we decided to re-design the unsatisfactory
literature procedures. By using in situ NMR analysis of the reaction
mixtures in the most critical step in the reaction sequence, we
were able to significantly improve the yield. Furthermore, imple-
mentation of an alternative deprotection procedure for the phtha-
limide groups, followed by a highly selective purification procedure
resulted in a significant increase in the isolated yield of high purity 4.
A radical integration of the synthetic sequence, by-passing time
consuming separation/purification steps, further increased the
overall isolated yield of 4 to 47% in four steps. Furthermore, we have
shown the possibility to isolate another interesting chiral renewable
building block, the 2-amino-2-deoxy-1,4-3,6-dianhydroiditol 10.
Currently, we are exploring more catalytic routes towards
isohexide based diamines, as well as more selective routes to
2-amino-2-deoxy-1,4-3,6-dianhydroiditol 10.
1
m
L. Injector temperature 275 ꢀC. Split ratio 1:33. Column flow (at
275 ꢀC) 50 mL/min helium. GC column: Restek Rxi-5ms,
m. GC program (2,5-FDA.mth); hold 2 min
30 mꢃ0.25 mmꢃ0.25
m
at 70 ꢀC, ramp 10 ꢀC/min, final temperature 300 ꢀC, hold 2 min.
Total run time 27 min. Detector; FID at 300 ꢀC. GCeMS analysis was
performed on a Finnigan Mat GC8000 Top GC (He carrier gas, flow
1 mL/min, split flow 25 mL/min; Chrompack CP-Sil 5CB column,
4. Experimental section
4.1. General information
30 mꢃ0.25 mmꢃ1.0
m
m; GC program hold 2 min at 40 ꢀC, ramp
10 ꢀC/min, final temperature 300 ꢀC) connected to a Finnigan Mat
Isomannide (Aldrich, 95%), p-toluene sulphonyl chloride (Fluka,
99%), potassium phthalimide (Fluka, ꢁ99%), DMSO (Aldrich,
Automass II quadrupole mass selective detector (EI, mass range

S. Thiyagarajan et al. / Tetrahedron 67 (2011) 383e389
387
40e400 Da, 150 ms sample speed). Electrospray Ionisation (ESI)
mass spectrometry was carried out using a Bruker micrOTOF-Q
instrument in positive ion mode (capillary potential of 4500 V).
Differential Scanning Calorimetry (DSC) measurements were con-
ducted on a PerkineElmer Diamond series DSC. The temperature
range used was 20 ꢀC up to 200 ꢀC at a heating rate of 10 ꢀC/min.
(Aldrich, ꢁ99.9%, anhydrous) (500 mL) were charged into the reactor
under a continuous flow of nitrogen. The initially yellow suspension
was subsequently stirred at 130 ꢀC (internal temperature) for 24 h.
Duringthecourse of thereactionthesuspensionbecame dark brown.
After termination of the reaction, thedark brown solutionwas cooled
downto roomtemperature, andpouredintocold water(1L)(pH7/8).
The resulting DMSO/water mixture was extracted with CHCl3
(4ꢃ500 mL). The combined CHCl3 layers were then subsequently
washed with water (2ꢃ500 mL), dried over MgSO4 and decolorized
with activated carbon (30 min at room temperature). The resulting
solution was then filtered over a G-3 filter funnel containing Celite.
The brown filtrate was evaporated at reduced pressure using a rota-
tory evaporator, giving a crude solid product. The crude product was
further purified by selective precipitation of 2,5-diphthalimido-
2,5-dideoxy-1,4-3,6-dianhydroiditol 3 from 250 mL of chloroform.
Route 2: Potassium phthalimide (44.4 g, 0.24 mol) and DMSO
(650 mL) was charged into the reactor under a continuous flow of
nitrogen. The mixturewasstirred at 130ꢀC (internaltemperature)for
4 h. 150 mL of DMSO was distilled off under reduced pressure. The
reaction mixture was cooled down to room temperature and 1,4-3,6-
4.2. Synthesis of 1,4-3,6-dianhydro-2,5-di-O-p-tosyl-D-
mannitol 2
A 2-necked round-bottom flask was charged with a solution of
isomannide (50 g, 0.34 mol) in 150 mL of pyridine under nitrogen
atmosphere. The mixture was cooled to 0e5 ꢀC. A solution of
p-toluene sulphonyl chloride (130.5 g, 0.68 mol) in pyridine
(350 mL) was added drop wise over 30 min. After 3 h, the reaction
mixture was placed in a refrigerator over night. Then, the mixture
was poured onto ice-water (2.5 L) and stirred for 1 h. The obtained
sandy solid was subsequently grounded and washed thoroughly
with water, dilute HCl (0.2 M), and finally water followed by suc-
tion-filtration. The crude product was recrystallized from ethanol
yielding 2 as white needles. Yield: 140 g, 90%; mp 93e94 ꢀC (lit.20
mp 92e93 ꢀC); nmax (neat) 1595, 1370, 1192, 1171, 1113, 1054,
dianhydro-2,5-di-O-p-tosyl- -mannitol 2 (50 g, 0.11 mol) was added.
D
Again the mixture was stirred at 130 ꢀC (internal temperature) for
24 h. After the completion of reaction, the dark brown solution was
cooled down to room temperature, and poured into cold water (1 L)
(pH 7/8). The resulting DMSO/water mixture was extracted with
CHCl3 (4ꢃ500 mL). The combined CHCl3 layers were then sub-
sequently washed with water (2ꢃ500 mL), dried over MgSO4 and
decolorizedwithactivatedcarbon(30minatroomtemperature). The
resulting solutionwasthen filtered overa G-3 filterfunnel containing
Celite. The brown filtrate was evaporated at reduced pressure using
rotatory evaporator, giving a yellow solid. NMR analysis showed the
presence of both bisphthalimide and monophthalimide in the crude
mixture. The crude product was further purified by selective pre-
cipitation of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol
3 from 250 mL of chloroform. Yield: 38 g, 84%; mp 244e245 ꢀC (lit.13
mp 243.4e243.6 ꢀC); nmax (neat) 3393, 3280, 1772, 1709, 1610, 1466,
996 cmꢂ1
; dH (400.17 MHz, CDCl3) 7.79 (4H, d, J 8.0 Hz), 7.33 (4H, d, J
8.0 Hz), 4.86e4.79 (2H, m), 4.45 (2H, d, J 3.7 Hz), 3.89 (2H, dd, J 9.3,
6.8), 3.72 (2H, d, J 9.0 Hz), 2.43 (6H, s); dC (100.62 MHz, CDCl3) 145.3,
133.0, 129.9, 127.9, 79.9, 77.8, 69.9, 21.5; HRMS (ESI): MHþ, found
455.0829. C20H23O8S2 requires 455.0834.
4.3. Synthesis of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-
dianhydroiditol 3 from 1,4-3,6-dianhydro-2,5-di-O-mesyl-D-
mannitol 5
1,4-3,6-Dianhydro-2,5-di-O-mesyl-
sized according to the procedure reported previously.19 1,4-
3,6-Dianhydro-2,5-di-O-mesyl- -mannitol (15 g, 0.05 mol),
D
-mannitol 5 was synthe-
D
5
potassium phthalimide (20 g, 0.10 mol) and DMSO (150 mL) were
charged to the reactor under a continuous flow of nitrogen. The
initially yellow suspension was subsequently stirred at 130 ꢀC
(internal temperature) for 24 h. During the course of the reaction
the suspension became dark brown. After termination of the re-
action, the dark brown solution was cooled down to room tem-
perature, and poured into cold water (1 L) (pH 7/8). The resulting
DMSO/water mixture was extracted with CHCl3 (2ꢃ500 mL). The
combined CHCl3 layers were then subsequently washed with water
(2ꢃ500 mL), dried over MgSO4 and decolorized with activated
carbon (30 min at room temperature). The resulting solution
was then filtered over a G-3 filter funnel containing Celite. The
brown filtrate was evaporated at reduced pressure using a rotatory
evaporator, giving a crude solid product of 16.2 g (80%). The crude
product was further purified by selective precipitation of
1380, 1301, 1173, 1122, 1087, 1008 cmꢂ1
; dH (400.17 MHz, CDCl3)
7.90e7.84 (4H, m), 7.78e7.73 (4H, m), 5.39 (2H, s), 4.92 (2H, dd, J 7.3,
6.4 Hz), 4.37 (2H, dd, J 9.2, 7.7 Hz), 4.01 (2H, dd, J 9.3, 6.1 Hz); dC
(100.62 MHz, CDCl3) 167.5, 134.1, 131.6, 123.3, 88.0, 71.0, 55.6; HRMS
(ESI): MHþ, found 405.1081. C22H17N2O6 requires 405.1087.
4.5. Synthesis of phthalimido-2-deoxy-1,4-3,
6-dianhydroiditol 7
The chloroform soluble fraction of the aforesaid reaction was
purified by column chromatography over silica gel (ethyl acetate/
petroleum ether (2:3)), yielding the phthalimido-2-deoxy-1,4-
3,6-dianhydroiditol 7. Yield: 4.8 g, 16%; mp 122e123 ꢀC; nmax (neat)
3393, 3280, 1772, 1709, 1610, 1466, 1380, 1301, 1173, 1122, 1087,
2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol
3
from
1008 cmꢂ1
; dH (400.17 MHz, CDCl3) 7.89e7.82 (4H, m), 7.77e7.72
200 mL of chloroform. Yield: 15 g, 75% (purity 99% based on NMR);
(4H, m), 5.16 (2H, dd, J 4.4, 2.2 Hz), 4.82 (2H, d, J 4.4 Hz), 4.78 (2H, td,
J 7.7, 2.2 Hz), 4.38 (2H, s), 4.21e4.13 (2H, m), 4.05 (2H dd, J 10.1,
3.4 Hz), 3.97e3.87 (4H, m); dC (100.62 MHz, CDCl3) 166.8, 133.4,
130.8, 122.6, 88.4, 84.9, 72.9, 69.2, 55.6; HRMS (ESI): MHþ, found
276.0866. C14H14NO5 requires 276.0872.
mp 244e245 ꢀC (lit.13 mp 243.4e243.6 ꢀC); nmax (neat) 3393, 3280,
1772, 1709, 1610, 1466, 1380, 1301, 1173, 1122, 1087, 1008 cmꢂ1
; dH
(400.17 MHz, CDCl3) 7.90e7.84 (4H, m), 7.78e7.73 (4H, m), 5.39 (2H,
s), 4.92 (2H, dd, J 7.3, 6.4 Hz), 4.37 (2H, dd, J 9.2, 7.7 Hz), 4.01 (2H, dd,
J 9.3, 6.1 Hz); dC (100.62 MHz, CDCl3) 167.5, 134.1, 131.6, 123.3, 88.0,
71.0, 55.6; HRMS (ESI): MHþ, found 405.1081. C22H17N2O6 requires
405.1087.
4.6. Synthesis of 2,5-diamino-2,5-dideoxy-1,4-
3,6-dianhydroiditol dihydrochloride 8
4.4. Synthesis of 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-
dianhydroiditol 3 from 1,4-3,6-dianhydro-2,5-di-O-p-tosyl-D-
mannitol 2
A two-necked round-bottom flask was charged with crude
mixture of 3 (41.2 g, 0.10 mol) and 250 mL 6 M HCl/AcOH (4:1). The
stirred slight brownish suspension was then heated at 135 ꢀC (oil
bath temperature) for 24 h. After completion of the reaction, the
slightly brown solution was cooled down to room temperature.
During cooling crystallization of phthalic acid occurred. The
Route 1: 1,4-3,6-Dianhydro-2,5-di-O-p-tosyl-D-mannitol 2 (50 g,
0.11 mol), potassium phthalimide (45.4 g, 0.25 mol) and DMSO

388
S. Thiyagarajan et al. / Tetrahedron 67 (2011) 383e389
resulting suspension was filtered over a G-3 filter funnel. The
brownish filtrate was washed thoroughly with diethyl ether
(3ꢃ300 mL) to remove residual phthalic acid, and the aqueous
phase was subsequently evaporated to dryness using a rotary film
evaporator yielding a brown solid. This was dissolved in methanol
(50 mL) at 50 ꢀC. Next, ethanol (50 mL) was added, and the tem-
perature was raised to distill off the methanol. Subsequently, an
ethanolic HCl solution (100 mL) was added to precipitate the de-
sired product. The resulting suspension was stirred for 10 min at
50 ꢀC, filtered over a G-3 filter funnel and the precipitate was dried
in a vacuum oven (<5 mbar, 24 h) over Sicapent yielding 8 as
slightly brown solid in 99% purity according to NMR analysis. Yield
8: 15.3 g, 70%; nmax (neat) 3370, 2894, 1638, 1525, 1268, 1525, 1084,
completion of reaction, the aqueous solution was filtered over
a glass filter (G-3) and the IEX-resin was washed thoroughly with
water (3ꢃ50 mL). The combined clear colourless aqueous phases
were evaporated to dryness using a rotary film evaporator to afford
an off-white viscous-solid 10. The product thus obtained was fur-
ther dried in vacuum oven (25 ꢀC, <5 mbar) over sicapent for 24 h.
Yield: 0.5 g, 70.5%; nmax (neat) 3365, 2885, 1624, 1565, 1235, 1080,
1025 cmꢂ1
; dH (400.17 MHz, D2O) 4.68 (H, d, J 1.6 Hz), 4.62 (H, s),
4.38 (H, s), 3.93 (3H, d, J 13.5 Hz), 3.77 (H, d J 9.6 Hz), 3.58 (H, s); dC
(100.62 MHz, D2O) 87.9, 86.6, 74.9, 74.1, 73.9, 56.8; HRMS (ESI):
MHþ, found 146.0812. C6H12NO3 requires 146.0817.
4.10. Experimental procedures for the synthesis of 2,5-
diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3 reaction
followed by NMR
1037 cmꢂ1
; dH (400.17 MHz, D2O) 5.04 (2H, s), 4.25 (2H, dd, J 10.7,
4.8 Hz), 4.08 (4H d, J 13.4 Hz); dC (100.62 MHz, D2O) 87.2, 72.7, 58.5.
4.7. Synthesis of 2,amino-2-deoxy-1,4-3,6-dianhydroiditol
hydrochloride 9
4.10.1. Reaction performed in DMSO-d6. 1,4-3,6-Dianhydro-2,5-di-
O-p-tosyl- -mannitol 2 (1.64 g, 3.6 mmol) and potassium phthali-
D
mide (1.47 g, 7.9 mmol) were added to 20 mL DMSO-d6 at room
temperature, giving a yellow suspension. The temperature was
raised to 130 ꢀC, resulting in an orange colored solution. The reaction
was stirred under nitrogen. NMR samples were taken at several time
intervals. After 48 h, the reaction was cooled down to room tem-
perature. Tothe solution100 mLwater wasadded, this was extracted
with 3ꢃ30 mL CHCl3. The combined organic layers were washed
with 2ꢃ25 mL water. The organic layer was then dried over MgSO4
and evaporated to dryness using a rotary film evaporator, giving
yellow solid (0.725 g, 59% yield). NMR analysis showed that the ratio
of products 2,5-diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol
3 and phthalimido-2-deoxy-1,4-3,6-dianhydroiditol 7 in 35:65%.
The filtrate of the aforesaid reaction was concentrated using
a rotary evaporator, giving clear brown viscous homogeneous oil
(1.4 g). Ethyl acetate was added tothe oil, and afterstirring for 10 min
at room temperature the product precipitated. The mono hydro-
chloride amine salt 9 was isolated by filtration, and subsequently
dried in a vacuum oven (<5 mbar, 24 h) over Sicapent yielding 9 as
brownish solid in 98% purity according to NMR analysis. Yield 9:
0.9 g, 29%; nmax (neat) 3370, 2894,1638,1525,1268,1084,1037 cmꢂ1
;
dH (400.17 MHz, D2O) 4.69 (H, d, J 4.0 Hz), 4.61 (H, d, J 4.0 Hz), 4.39 (H,
s), 4.00e3.89 (3H, m), 3.78 (H, dd, J 9.5, 1.8 Hz), 3.57 (H, dd, J 2.4,
1.8 Hz); dC (100.62 MHz, D2O) 87.9, 86.6, 74.9, 74.1, 73.9, 56.8.
4.8. Synthesis of 2,5-diamino-2,5-dideoxy-1,4-3,6-
dianhydroiditol 4
4.10.2. Reaction performed in DMF-d7. 1,4-3,6-Dianhydro-2,5-di-O-
p-tosyl- -mannitol 2 (0.55 g, 1.2 mmol) and potassium phthalimide
D
(0.50 g, 2.7 mmol) were added to 7.7 mL DMF-d7 at room temper-
ature, giving a white suspension. The temperature was raised to
110 ꢀC, giving a brown/orange solution with a brown precipitate.
The reaction was stirred under nitrogen. NMR samples were taken
at several time intervals. After 42 h, the reaction was cooled down
to room temperature. To the solution 100 mL water was added, this
was extracted with 3ꢃ30 mL CHCl3. The combined organic layers
were washed with 2ꢃ25 mL water. The organic layer was then
dried over MgSO4 and evaporated to dryness using a rotary film
evaporator, giving a yellow solid (0.5 g, >100% yield) NMR analysis
showed that there were too many impurities. Therefore the ratio
could not be determined.
2,5-Diamino-2,5-dideoxy-1,4-3,6-dianhydroiditol HCl salt
8
(15.3 g, 0.07 mol) was dissolved in demineralised water (200 mL)
giving a slightly brown solution. To this solution activated carbon
(100 mg) was added, and the resulting suspension was stirred for
30 min at 35 ꢀC. Filtration over a glass filter (G-3) containing Celite,
resulted in a clear colourless filtrate. To this filtrate freshly washed
Amberlyst A-26-OH (60 g, 0.075 mol) was added, and the resulting
suspension was sonified in an ultrasonic bath for 1 h at 30 ꢀC. After
completion of reaction, the aqueous solution was filtered over
a glass filter (G-3) and the IEX-resin was washed thoroughly with
water (3ꢃ50 mL). The combined clear colourless aqueous phases
were evaporated to dryness using a rotary film evaporator yielding
an off-white semi-solid 4, which was further dried in vacuum oven
(25 ꢀC, <5 mbar) over sicapent for 24 h. The crude product was
recrystallized from ethanol/methanol mixture (1:0.25) yielding 4 as
white needles. Yield: 9.5 g, 94% (99% purity based on NMR); mp
60e62 ꢀC (lit.13 mp 59e60 ꢀC); nmax (neat) 3351, 2825, 1597, 1537,
4.10.3. Reaction performed in DMSO-d6 (dry). 1,4-3,6-Dianhydro-
2,5-di-O-p-tosyl-D-mannitol 2 (1.64 g, 3.6 mmol), potassium
phthalimide (1.48 g, 8.0 mmol) were added to 20 mL DMSO-d6
(distilled, high vacuum distillation, 0.05 mbar) at room tempera-
ture, giving a yellow suspension. The temperature was raised to
130 ꢀC, giving an orange coloured solution. The reaction was stirred
under nitrogen. NMR samples were taken at several time intervals.
After 25 h, the reaction was cooled down to room temperature. To
the solution 100 mL water was added, this was extracted with
3ꢃ30 mL CHCl3. The organic layer was washed with 2ꢃ25 mL
water. The organic layer was then dried over MgSO4 and evaporated
to dryness using a rotary film evaporator, giving yellow solid (1.0 g,
77% yield). NMR analysis showed that the ratio of products 2,5-
diphthalimido-2,5-dideoxy-1,4-3,6-dianhydroiditol 3 and phthali-
mido-2-deoxy-1,4-3,6-dianhydroiditol 7 in 67:33%.
1200, 1080, 1035 cmꢂ1
; dH (400.17 MHz, D2O) 4.74 (2H, s), 4.03 (2H,
dd, J 10.3, 4.9 Hz), 3.84 (2H, dd, J 10.3, 2.5 Hz), 3.72 (2H, dd, J 4.7,
2.5 Hz); dC (100.62 MHz, D2O) 86.2, 72.1, 56.3; HRMS (ESI): MHþ,
found 145.0972. C6H13N2O2 requires 145.0977.
4.9. Synthesis of 2-amino-2-deoxy-1,4-3,6-dianhydroiditol 10
2-Amino-2-deoxy-1,4-3,6-dianhydroiditol HCl salt 9 (0.9 g,
0.005 mol) was dissolved in demineralised water (100 mL) giving
a slightly brown solution. To this solution was added activated
carbon (100 mg) and stirred for 30 min at 35 ꢀC. The suspension
was filtered over a glass filter (G-3) containing Celite, resulting
a clear colourless filtrate. To this filtrate freshly washed Amberlyst
A-26-OH (4 g, 0.005 mol) was added, and the resulting suspension
was sonified in an ultrasonic bath for 1 h at 30 ꢀC. After the
Acknowledgements
Dr. G. Frissen and W. Teunissen are gratefully acknowledged for
NMR and GCeMS analyses. Roquette Freres and Norrit are

S. Thiyagarajan et al. / Tetrahedron 67 (2011) 383e389
389
7. (a) Sablong, R.; Duchateau, R.; Koning, C. E.; de Wit, G.; van Es, D. S.; Koele-
wijn, R.; van Haveren, J. Biomacromolecules 2008, 9, 3090e3097; (b) Noord-
over, B. A. J.; van Staalduinen, V. G.; Duchateau, R.; Koning, C. E.; van Benthem,
R. A. T. M.; Mak, M.; Heise, A.; Frissen, A. E.; van Haveren, J. Biomacromolecules
2006, 7, 3406e3416; (c) van Haveren, J.; Oostveen, E. A.; Micciche, F.;
Noordover, B. A. J.; Koning, C. E.; van Benthem, R. A. T. M.; Frissen, A. E.;
Weijnen, J. G. J. J. Coat. Technol. Res. 2007, 4, 177e186; (d) van Es, D.S.; Frissen,
A.E.; Luitjes, H. WO2001083488, 2001; (e) Luitjes, H.; Jansen, J.C. WO9945060,
1999.
8. Kuszmann, J.; Medgyes, G. Carbohydr. Res. 1980, 85, 259e269.
9. Archibald, T. G.; Baum, K. Synth. Commun. 1989, 19, 1493e1498.
10. For safe handling of organic azides (NCþNO)/NN>3, in the case of isohexide
diazides the formula amount to 8/6 (!); see also Kolb, H. C.; Finn, M. G.;
Sharpless, K. B. Angew. Chem., Int. Ed. 2001, 40, 2004e2021.
11. (a) Montgomery, R.; Wiggins, L. F. J. Chem. Soc. 1946, 393e396; (b) Bashford, V.
G.; Wiggins, L. F. J. Chem. Soc. 1950, 371e374.
acknowledged for supplying isoidide and activated carbon re-
spectively and finally, this work is part of the research programme
of the Dutch Polymer Institute (DPI), Eindhoven, the Netherlands,
project nr. # 653 and # 656.
Supplementary data
Copies of 1H and 13C NMR spectra of 2e4, 7e10. Supplementary
data related to this article can be found online version, at
doi:10.1016/j.tet.2010.11.031.
References and notes
12. Van Es, D.S.; Gootjes, L. Unpublished results.
13. Cope, A. C.; Shen, T. Y. J. Am. Chem. Soc. 1956, 78, 3177e3182.
14. Kuszmann, J.; Medgyes, G. Carbohydr. Res. 1978, 64, 135e142.
15. Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456e463.
16. Van Es, D.S.; Koelewijn R. unpublished results.
17. E.g. Dong, Y.; Niranjan, N.; Ablaza, S. L.; Yu, S. X.; Bolvig, S.; Forsyth, D. A.; Le
Quesne, P. W. J. Org. Chem. 1999, 64, 2657e2666.
1. Fenouillot, F.; Rousseaua, A.; Colominesd, G.; Saint-Loupe, R.; Pascault, J. P. Prog.
Polym. Sci. 2010, 35, 578e622.
2. Stoss, P.; Hemmer, R. Adv. Carbohydr. Chem. Biochem. 1991, 49, 93e173.
3. Kricheldorf, H. R. J. Macromol. Sci., Rev. Macromol. Chem. Phys.1997, C37, 599e631.
4. (a) Thiem, J.; Lueders, H. Starch/Staerke 1984, 36, 170e176; (b) Thiem, J.;
Lueders, H. Polym. Bull. 1984, 11, 365e369.
18. Montgomery, R.; Wiggins, L. F. J. Chem. Soc. 1948, 237e241.
19. Coster, G. D.; Vandyck, K.; Eycken, E. V. D.; Eycken, J. V. D.; Elseviers, M.; Roper,
H. Tetrahedron: Asymmetry 2002, 13, 1673e1679.
20. Muri, E. M. F.; Gomes, M., Jr.; Albuquerque, M. G.; da Cunha, E. F. F.; de
Alencastro, R. B.; Williamson, J. S.; Antunes, O. A. C. Amino Acids 2005, 28,
413e419.
5. (a) GB1012563, 1965; (b) Tate, C.W.; Scruggs, J.G. U.S. Patent 3,223,752, 1965.
6. (a) Charbonneau, L.F.; Johnson, R.E.; Witteler, H.B.; Khanarian, G. U.S. Patent
5,959,066,1999; (b) Charbonneau, L.F.; Khanarian, G.; Johnson, R.E.; Witteler, H.B.;
Flint, J. A. U.S. Patent 6,063,495, 2000; (c) Charbonneau, L.F.; Johnson, R.E. U.S.
Patent 6,063,465, 2000; (d) Brandenburg, C.; Hayes, R.A. U.S. Patent 6,818,730,
2002; (e) Kurian, J.V.; Ling, Y. U.S. Patent 6,737,481, 2004.