Kinetic and chemical studies on the isomerization of monosaccharides in N-methylmorpholine-N-oxide (NMMO) under Lyocell conditions

Article abstract of DOI:10.1016/j.carres.2004.06.004

The Lyocell process is a modern and environmentally fully compatible industrial fiber-making technology. Cellulosic pulp is dissolved without chemical derivatization in a melt of N-methylmorpholine-N-oxide monohydrate (NMMO). In the present work, the reactions of monosaccharides under Lyocell conditions were investigated in detail, using capillary zone electrophoresis as the analytical technique to clarify the composition of reaction mixtures and to follow the kinetics. Under Lyocell conditions, xylose and glucose undergo two competitive reactions: rapid conversion to nonreducing products, and complete isomerization involving the whole carbohydrate backbone, via ketose intermediates. Sugar acids are present in minor amounts only, as demonstrated by employing isotopically labeled material for NMR techniques.

Full text of DOI:10.1016/j.carres.2004.06.004

Carbohydrate
RESEARCH
Carbohydrate Research 339 (2004) 1899–1906
Kinetic and chemical studies on the isomerization of
monosaccharides in N-methylmorpholine-N-oxide (NMMO)
under Lyocell conditions
*
€
Immanuel Adorjan, John Sjoberg, Thomas Rosenau, Andreas Hofinger and Paul Kosma
Christian-Doppler-Laboratory, Department of Chemistry, University of Natural Resources and Applied Life Sciences (BOKU),
Muthgasse 18, A-1190 Vienna, Austria
Received 17 March 2004; accepted 10 June 2004
Available online 2 July 2004
Abstract—The Lyocell process is a modern and environmentally fully compatible industrial fiber-making technology. Cellulosic
pulp is dissolved without chemical derivatization in a melt of N-methylmorpholine-N-oxide monohydrate (NMMO). In the present
work, the reactions of monosaccharides under Lyocell conditions were investigated in detail, using capillary zone electrophoresis as
the analytical technique to clarify the composition of reaction mixtures and to follow the kinetics. Under Lyocell conditions, xylose
and glucose undergo two competitive reactions: rapid conversion to nonreducing products, and complete isomerization involving
the whole carbohydrate backbone, via ketose intermediates. Sugar acids are present in minor amounts only, as demonstrated by
employing isotopically labeled material for NMR techniques.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Keywords: Capillary electrophoresis; Isomerization; Degradation; NMMO; Monosaccharides; Aldoses
1. Introduction
and the oxidative nature of the solvent.1 These chemical
processes are reflected for instance in a pronounced
chromophore formation in the Lyocell spinning dope,
which might even cause discoloration of the fiber
products. Resulting from multiple pathways, a large
number of chromophores is formed, of which all con-
tribute to overall discoloration or yellowing. In a recent
work, most of the identified chromophores have been
shown to be degradation and condensation products of
monosaccharides, which in turn must have originated
from degradation reactions of the pulp polymers.2 The
rates of chromophore formation in NMMO solutions
from different carbohydrates and pulps have also been
studied, finding pentoses to react faster than hexoses.3
Carbonyl groups were determined as the main influen-
tial factor regarding chromophore formation.3 In the
present study, the behavior of monosaccharides under
Lyocell conditions will be addressed, and kinetic and
chemical studies will be presented.
Within the past decade,
a
new, environmentally
compatible cellulose-fiber technology has emerged, the
Lyocell process, which employs a melt of N-meth-
ylmorpholine-N-oxide monohydrate (NMMO) for di-
rect dissolution of cellulosic pulp without chemical
derivatization. Other benefits with regard to environ-
mental compatibility are the use of air and water as the
only spinning bath ‘chemicals’, and the nearly complete
recycling of the amine oxide solvent. Although the dis-
solution process is supposed to be entirely physical
according to theory, manifold chemical side reactions
occur in reality due to the elevated process temperatures
* Corresponding author. Tel.: +43-1-36006-6071; fax: +43-1-36006-
6059; e-mail: trosenau@edv2.boku.ac.at
0008-6215/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.06.004

1900
I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
2. Results and discussion
Glucose
120
100
80
60
40
20
0
Mannose
other sugars
2.1. Isomerization and degradation kinetics of monosac-
charides under Lyocell conditions
While the reactions of carbohydrates and cellulose un-
der aqueous alkaline conditions have been intensively
investigated in the past,4–6 as well as recently,7–10 the
behavior of low-molecular-weight carbohydrates under
Lyocell conditions––that is in an NMMO melt at about
110 ꢁC––remained unclear so far. In the present study,
we used an optimized CZE method,11 based on reductive
amination of aldoses with aminobenzoic acid ethyl ester
(ABEE) carried out as pre-column derivatization. In the
case of carbohydrates dissolved in the cellulose solvent
N-methylmorpholine-N-oxide monohydrate (NMMO),
no negative influence of residual NMMO present in the
aqueous saccharide solutions on the reductive amination
reaction was observed.
Monosaccharides under Lyocell conditions undergo
isomerization reactions similar to the reactions under
aqueous alkaline conditions, even though the pH value
of the NMMO/carbohydrate reaction mixtures, which is
about 6–7 after dissolution in excess water, is much
lower than that of the alkaline media used. The iso-
merization was largely complete and proceeded along
the whole carbon chain of the sugar. Starting for
instance with galactose, isomerization proceeded to give
glucose, and all eight isomeric aldohexoses were found
(Fig. 1).
0
50
100
time [min]
150
200
250
(a)
1.5
1.0
0.5
0.0
Altrose
Allose
Galactose
Talose
Gulose
Idose
0
50
100
time [min]
150
200
250
(b)
Figure 2. Isomerization of glucose in NMMO at 110 ꢁC: (a) formation
of the C-2 epimer mannose; (b) build-up and degradation of the C-3,
C4, and C-5 epimers of glucose. Note the different scales of the y-axes.
The isomerization processes are only a relatively small
part of the overall chemical conversions in the system.
The major part are degradations and reactions to non-
reducing products (including ketoses), which are thus
not detectable by the capillary electrophoresis method
used,11 as the presence of a reducing end is required for
attaching the UV label by reductive amination. In the
case of glucose, after 4 h reaction time only 20% of the
starting material was still present, along with about 10%
of isomerized aldoses (Fig. 2a). Since each isomerization
step involves the formation of an enediol structure as
intermediate and these strongly reducing structures
(‘reductones’) are supposed to be the starting points of
the degradation reactions in case of alkaline treat-
ment,4–9 similar behavior can be expected in the presence
of the reactive NMMO especially at the elevated reac-
tion temperature used.
Figure 2 presents the kinetics of
D-glucose isomeri-
zation/degradation under Lyocell conditions. An expo-
nential decay according to pseudo first-order kinetics
was found for the starting material (Fig. 2a); formation
of the isomeric sugars showed the typical course of
intermediates, their generation being superimposed by
consumption in degradation/isomerization processes.
From all reducing intermediates, the C-2 epimer man-
nose was found in highest concentration, with a maxi-
mum at approx. 150 min reaction time, followed by
allose and altrose (Fig. 2a and b). The isomers galactose,
talose, gulose, and idose were formed only in rather
small amounts, as plotted in Figure 2b.
Gal in NMMO at 110°C after 2.5 h
40
Gal
Int. std. Rib
20
Tal
Gul
Glc
Ido
Man All
Alt
0
5
6
7
8
9
Also in the case of
D-galactose an exponential decay
time [min]
was observed, but the reaction rate was significantly
higher as compared to glucose. The concentration of the
C-2 epimer talose reached its maximum after 60 min
Figure 1. Electropherogram of the reaction mixture of galactose after
2.5 h in NMMO at 110 ꢁC.

I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
1901
already, whereas glucose was still in the build-up phase
after 3 h reaction time.
It is interesting to note that the isomerization of
-sorbose proceeded faster, and that sorbose as well as
D
In the course of the isomerization steps, keto-sugars
must have formed in accordance with the isomerization
schemes in Figures 6 and 7. Even though these species
cannot be detected and quantified by the CZE method
used,11 their aldose isomerization products should be
detectable, which would provide strong support for the
suggested mechanism. Therefore, the ketohexoses D-
fructose and D-sorbose were subjected to the NMMO
treatment. Further direct evidence for the intermediacy
of ketohexoses in the isomerization reactions was pro-
vided by 13C NMR spectroscopy, see Chapter 2.2.
the main products gulose and idose were less stable
under the prevailing conditions. The sum of all aldoses
reached its maximum already after 40 min, with 18% of
the starting material having been converted into aldoses
at this time. Due to the instability of the keto sugar and
its quick conversion, the build-up phase for the two
main products gulose and idose was so fast, that only a
decline was monitored. On the other hand, the more
stable glucose was still in the build-up phase after a
reaction time of 2 h.
In the next step, the isomerization of pentoses in
NMMO was investigated, since the chromophore for-
mation rate under Lyocell conditions had been shown to
be much higher for monomeric pentoses as compared to
hexoses, and also for polymeric pentoses (xylan) in
comparison to their hexose counterpart (cellulose).3
In Figure 3a and b, the
D-fructose isomerization
kinetics are presented. In contrast to the aldoses glucose
and galactose, where epimerization could only progress
‘in one direction’, the keto group in fructose can iso-
merize ‘in two directions’. The main products are glu-
cose followed by mannose (Fig. 3a), in accordance with
the expectations. Further products in the order of
decreasing intermediate concentration were allose,
altrose, galactose, talose, gulose, and idose (Fig. 3b).
After approx. 60 min, the sum of all aldoses reached its
maximum corresponding to 34% of the consumed
starting material.
D
-Xylose, as the starting material, was consumed
according to pseudo-first-order kinetics (Fig. 4a). The
140
120
100
Xylose
22
20
18
16
14
12
10
8
80
Lyxose
Xylose calcd.
60
Lyxose calcd.
degradation products
40
20
calcd.
Glucose
Mannose
other sugars
0
6
4
2
0
20
40
60
80
100
120
(a)
time [min]
0
0
20
40
60
80
100
120
Lyxose
14
12
10
8
(a)
Arabinose
Ribose
time [min]
Lyxose calcd.
Arabinose calcd.
Ribose calcd.
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Allose
Altrose
Galactose
Talose
Gulose
Idose
6
4
2
0
0
20
40
60
80
100
120
0
20
40
60
80
100
120
time [min]
(b)
(b)
time [min]
Figure 4. (a) Isomerization of xylose in NMMO at 110 ꢁC; (b) build-up
and degradation of the C-2 epimer lyxose and the C-3 isomers arabi-
nose and ribose. Note the different scales of the y-axes. The symbols
represent the experimental values, the full lines the mathematically
modeled fits.
Figure 3. Isomerization of fructose in NMMO at 110 ꢁC: (a) course of
the main intermediates glucose and mannose; (b) profile of the minor
isomerization products allose, altrose, galactose, talose, gulose, and
idose. Note the different scales of the y-axes.

1902
I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
Figure 5. Simplified model of the isomerization of pentoses and corresponding differential rate laws.
firmed the first-order kinetics for all individual reaction
steps in the isomerization scheme.
As expected, isomerization occurred also in the case
of D-arabinose, and the reaction proceeded much faster
than in the case of hexoses. Consumption of the starting
sugar followed a pseudo-first-order exponential decay,
and the isomerization products showed the typical
course of intermediates.
The complete set of isomerization reactions is sum-
marized in Figure 6 for hexoses and Figure 7 for pen-
toses.
Table 1. Calculated reaction constants for pentose isomerization in
NMMO at 110 ꢁC using -xylose as the starting material
D
Reaction constants k [sÀ1
]
k1
k2
k3
k4
k5
k6
k7
k8
4.41E)02
3.45E)02
2.57E)01
1.28E)01
2.75E)01
5.68E)02
8.05E)02
2.03E)01
3.36E)01
4.26E)01
6.58E)02
1.02E)01
k9
k10
k11
k12
2.2. 13C NMR investigations of [1-13C]glucose under
Lyocell conditions
Another issue in reactions of monosaccharides under
Lyocell conditions is the oxidation of aldoses to the
corresponding aldonic acids, especially as NMMO is
considered a rather strong oxidant. Preliminary CZE
runs indicated aldonic acids to be present in very small
amounts only. The drawback of the CZE method was its
low sensitivity, since direct UV detection at 200 nm was
used. Ion chromatography further supported that the
amount of aldonic acids in the monosaccharide reaction
mixtures was well below 5% (data not shown). To clarify
this issue and to finally confirm the intermediacy of
keto-sugars (which were not detectable by the used CZE
method,11 [1-13C]-glucose was subjected to the usual
isomerization conditions. A reaction time of 1.5 h was
chosen, because CZE experiments had shown that after
this time approx. half of the glucose starting material
had been consumed. After cooling, the reaction mixture
was dissolved in D2O and analyzed by quantitative 13C
NMR spectroscopy.12 Due to the selective 13C enrich-
ment, C-1 produces a signal about 100 times stronger
than the other carbon atoms present, which are thus
formed amount of ‘degradation products’, calculated as
the difference between starting material and detectable
aldohexoses, gave the expected delayed exponential in-
crease. As shown in the zoomed plot in Figure 4b, the
main intermediate lyxose reached its concentration
maximum already after 20 min, ribose after about
30 min, and arabinose after approx. 40 min.
The experimental kinetic data were fitted by a math-
ematical model. For that purpose, the quite complex
overall isomerization processes (cf. Fig. 7) were simpli-
fied using the chemical model in Figure 5. The keto and
enediol structures were combined into two virtual
intermediates I1 and I2 as indicated in Figure 7. Fur-
thermore, the model was based on the assumptions that
all reactions are of first order, all equilibria are estab-
lished, and degradation reactions start exclusively from
the intermediates I1 and I2. The resulting set of differ-
ential equations was solved, and the kinetic rate con-
stants were determined by iterative fit optimization
(Table 1).
Considering that the data points in Figure 4 show
the experimental values and the full lines represent the
solutions of the differential equations in Figure 5, the
match between experimental and calculated values is
excellent. This outcome proved the validity of the
assumptions used in the model. In particular, it con-
At present, it can be stated that the kinetics are actually of pseudo-
first-order. Assuming both the sugar and NMMO to participate in
the rate-determining step, the NMMO concentration would be much
larger and would thus remain nearly constant over time.

I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
1903
O
O
OH
OH
OH
L-Series
HO
D-Fructose
D-Sorbose
CH2OH
O
OH
OH
OH
OH
OH
CH2OH
D-Glucose
O
HO
CH2OH
O
OH
CH2OH
OH
OH
OH
HO
HO
CH2OH
OH
CH2OH
CH2OH
OH
D-Gulose
HO
HO
OH
OH
CH2OH
OH
OH
CH2OH
O
HO
HO
OH
CH2OH
OH
CH2OH
O
OH
OH
O
OH
OH
HO
HO
HO
HO
HO
OH
OH
CH2OH
OH
OH
CH2OH
CH2OH
OH
CH2OH
OH
CH2OH
CH2OH
OH
CH2OH
HO
HO
HO
D-Idose
D-Mannose
O
OH
OH
CH2OH
O
OH
CH2OH
CH2OH
O
CH2OH
CH2OH
OH
OH
OH
OH
HO
HO
HO
HO
HO
HO
OH
OH
OH
O
OH
OH
OH
OH
OH
OH
CH2OH
D-Galactose
OH
CH2OH
O
CH2OH
O
OH
CH2OH
CH2OH
D-Allose
O
OH
OH
OH
CH2OH
CH2OH
HO
HO
HO
HO
OH
OH
O
HO
HO
OH
CH2OH
OH
CH2OH
OH
CH2OH
OH
CH2OH
HO
OH
OH
OH
HO
D-Tagatose
OH
CH2OH
D-Psicose
CH2OH
D-Talose
D-Altrose
Figure 6. Isomerization of hexoses.
‘blinded out’. The amount of carboxylic acid oxidation
products, seen in the region of 170–180 ppm for C-1
signals, was found to be very low as compared to the a-
and b-pyranose forms of glucose (Fig. 8). The extremely
small signal at 179.99 ppm (see the inset in Fig. 8) is
likely to originate from an aldonic acid, the signal at
171.59 ppm from another, hitherto unidentified C-1
carboxylic degradation product. In addition, a CH3
signal at 24.07 ppm was observed.
The chemical shifts and coupling constants (JC1;C2) of
the a- and b-anomers of glucose as well as the ratio
between these anomers were in complete agreement with
published values.13–15 Anomeric signals of mannose (a-
and b-pyranose form) were also detected. Three forms
of fructose (b-pyranose and a,b-furanose; the a-fura-
nose form being very small) were found, thus directly
confirming the presence of keto-sugars. Also here,
chemical shifts and the ratios of the different hemiacetal/
hemiketal isomers fully corresponded with literature
data.13;16;17 Compared to glucose, a rather high amount
of fructose (33%) was present. The content of mannose
was 9%, which corresponded roughly to the CZE re-
sults. Thus, the 13C NMR study supported not only the
outcome of the CZE experiments, it moreover provided
additional information about the formation of fructose,
which was not accessible with the CZE method used.
Nevertheless no significant amounts of degradation
products were found, though––according to CZE
experiments––about half of the glucose had been con-
sumed, that is, the integrals of the 1-13C signals of glu-
cose should be roughly the same as the integrals over all
other 13C-signals. In the present case, considerable 13C
intensity (about 40%) was missing. This could be due to
loss of C-1 via decarboxylation reactions. Further
investigations using [2-13C] and [6-13C]glucose are
underway.
2.3. Isomerization of carbohydrates in other N-oxides
Finally the question was addressed whether the induc-
tion of isomerization in monosaccharides is a unique
ability of NMMO, or whether it is a property of tertiary
amine N-oxides in general.
D-Xylose was chosen as the
test carbohydrate because of its quite high reactivity,
and N,N-dimethyloctylamine-N-oxide as well as N,N-
dimethyldodecylamine-N-oxide were used as the reac-
tion media. A reaction time of 1.5 h was scheduled.

1904
I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
O
OH
HO
OH
CH2OH
D-Xylulose
OH
k2
CH2OH
OH
CH2OH
O
D-Xylose
O
k1
OH
HO
HO
HO
k4
k3
OH
OH
OH
CH2OH
CH2OH
CH2OH
CH2OH
HO
HO
HO
I1
O
OH
OH
CH2OH
L-Series
CH2OH
D-Lyxose
k5
k7
CH2OH
OH
CH2OH
OH
O
OH
CH2OH
O
OH
OH
OH
HO
HO
I2
k10
OH
CH2OH
k9
CH2OH
OH
CH2OH
O
CH2OH
OH
OH
OH
D-Ribose
O
HO
OH
OH
OH
CH2OH
CH2OH
CH2OH
k8
HO
OH
OH
D-Ribulose
k7
CH2OH
D-Arabinose
Figure 7. Isomerization of pentoses. The simplifications of intermediates I1 and I2 chosen for mathematical modeling of the rate constants are
indicated by circles.
β-Glcp
NMMO
α-Glcp
Dioxane
β-Frup
180
178
176
ppm
174
172
β-Fruf
α + β -Manp
200
180
160
140
120
100
ppm
80
60
40
20
0
Figure 8. 13C NMR spectrum of the reaction mixture of [1-13C]glucose in NMMO at 110 ꢁC after 1.5 h. Dioxane was used as the internal reference
(67.40 ppm).
Isomerization occurred with all amine oxides tested,
even though the reactions were slower than in the ref-
erence case of NMMO. Thus, the results confirm that
other tertiary amine-N-oxides besides NMMO possess

I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
1905
(95%) were obtained from Sigma (St. Louis, MO, USA).
The water used was purified (>17 MX/cm) with an HQ5
filter apparatus (REWA, Gladbeck, Germany).
Xylose in NMMO 1.5 h
Xyl
Ion chromatography: For the determination of aldonic
acids, a Dionex CarboPac PA10 4 · 250 mm column
fitted with a Dionex CarboPac PA10 Guard 4 · 50 mm,
0.6 M sodium hydroxide solution as eluant, a flow of
0.5 mL/min and pulsed amperometry (Dionex ED40) for
detection was used.
20
Int. Std. Gal
Ara
Lyx
Rib
13C NMR spectroscopy: 13C NMR spectra were
recorded on a Bruker Avance DPX instrument operat-
ing at 75.47 MHz for 13C using D2O as the solvent. 13C
NMR spectra were measured at 293 K and referenced to
1,4-dioxane (d 67.40). Quantitative 13C NMR spectro-
scopy was performed with Bruker standard software
using inverse gated decoupling with 4.0 s delay.12
Kinetic studies: Twenty milligrams of the respective
sugar was mixed with 200 mg NMMOÁH2O in a 4 mL
vial and immersed into an oil bath at 110 ꢁC with
magnetic stirring. After a certain reaction time, the
vessel was quickly cooled in an ice/water bath and the
reaction mixture was dissolved in water (2 mL). A
200 lL aliquot was taken and analyzed by capillary zone
electrophoresis (CZE) after pre-column derivatization
by reductive amination with ABEE and NaCNBH3. Pre-
column derivatization and analytical procedure have
been described in detail previously.11
0
6
8
10
(a)
time [min]
40
Xylose in N,N-Dimethyloctylamine N-oxide 1.5 h
Xyl
20
Lyx
Int Std. Gal
Rib
Ara
0
6
8
10
For modeling and experimental data fitting the soft-
ware Scientistꢂ for Windows Version 2.0 by MICRO-
MATH Scientific Software was used.
time [min]
(b)
Figure 9. Electropherograms of xylose heated for 1.5 h at 110 ꢁC in (a)
N-methylmorpholine-N-oxide monohydrate (NMMO) and (b) N,N-
dimethyloctylamine-N-oxide.
4. Conclusions
isomerizing capabilities as well. In Figure 9, a compar-
ison of electropherograms of the reaction mixtures is
presented.
Isomerization of monosaccharides in NMMO and other
amine N-oxides under Lyocell conditions proceeds quite
fast and affects the whole carbon chain, producing all
four isomeric aldopentoses from xylose and all eight
isomeric aldohexoses from glucose. Similar isomeriza-
tion processes start also from ketosugars. The iso-
merization reactions involve the formation of enediol
intermediates, which are likely to be the starting points
for degradation/oxidation reactions finally leading to
chromophores. Oxidation of aldoses to the corre-
sponding aldonic acids is of minor importance only. The
degradation or isomerization to ketoses seem to domi-
nate over the isomerization to CZE-detectable aldoses.
The isomerization kinetics of glucose, galactose, fruc-
tose, sorbose, arabinose and xylose were fully recorded,
that of xylose was mathematically modeled in addition.
Besides fructose and mannose as isomerization
products no significant amounts of degradation prod-
ucts were observed in the 13C NMR spectrum of
[1-13C]glucose after NMMO treatment. Thus, in con-
trast to the clarified isomerization, further investigations
The electropherogram of the reaction mixture in N,N-
dimethyldodecylamine-N-oxide was similar to Figure 9b
with the amount of isomerized pentoses being even
lower than in N,N-dimethyloctylamine-N-oxide.
3. Experimental
Chemicals: NMMO monohydrate was recrystallized
from acetone and chloroform (mp 76–78 ꢁC). 4-Amino-
benzoic acid ethyl ester (ABEE) 98%, sodium cyano-
borohydride 95%, boric acid (99.5+%),
(99%), -xylose (99+%), and -glucose (A.S.C. grade)
were obtained from Sigma-Aldrich (Milwaukee, WI,
USA). -Lyxose (99+%), -galactose (99+%), -allose
(99+%), -altrose (99+%), -talose (99+%) were
obtained from Fluka (Buchs, Switzerland). -Mannose
(98+%) was obtained from Merck E. (Darmstadt, Ger-
many). -Ribose (99+%), -gulose (95%), and -idose
D-arabinose
D
D
D
D
D
D
D
D
D
D
L

1906
I. Adorjan et al. / Carbohydrate Research 339 (2004) 1899–1906
using [2-13C] and [6-13C]glucose are needed to study the
degradation reactions under Lyocell conditions.
The results of this study help to improve the under-
standing of the very complex topic of color generation
under Lyocell conditions, especially by narrowing the
gap between the polymer cellulose and the monosac-
charide-derived chromophores. The missing link––the
mechanism of monosaccharide release from the poly-
meric starting materials (cellulose/xylan)––will be ad-
dressed in further studies.
(Lyocell). Part 2: Isolation and Identification of Chromo-
phores, Cellulose, in press.
3. Adorjan, I.; Potthast, A.; Rosenau, T., Sixta, H.; Kosma,
P. Chromophore Formation in Cellulose Solutions in N-
Methylmorpholine-N-Oxide (Lyocell). Part 1: Studies on
Model Compounds and Pulps, Cellulose, submitted for
publication.
4. Theander, O. Adv. Carbohydr. Chem. Biochem. 1988, 46,
273–326.
5. De Wit, G.; Kieboom, A. P. G.; van Bekkum, H.
Carbohydr. Res. 1979, 74, 157–175.
6. De Bruijn, J. M.; Kieboom, A. P. G.; van Bekkum, H.
Recl. Trav. Chim. Pays-Bas 1987, 106, 35–43.
7. Yang, B. Y.; Montgomery, R. Carbohydr. Res. 1996, 280,
27–45.
8. Yaylayan, V. A.; Ismail, A. A. Carbohydr. Res. 1995, 276,
253–265.
Acknowledgements
The assistance of Mr. Walter Milacher, R&D depart-
ment at Lenzing AG, with the ion chromatography
measurements is highly appreciated. The authors would
like to thank the Austrian Christian-Doppler Research
9. Ellis, A. V.; Wilson, M. A. J. Org. Chem. 2002, 67, 8469–
8474.
10. Knill, C. J.; Kennedy, J. F. Carbohydr. Polym. 2003, 51,
281–300.
€
11. Sjoberg, J.; Adorjan, I.; Rosenau, T.; Kosma, P. Carbo-
€
Society, the Austrian Fonds zur Forderung der wis-
hydr. Res., 2004, 339, in this issue. doi:10.1016/j.car-
res.2004.06.003.
12. Field, L. D.; Sternhall, S. Analytical NMR; Wiley:
Chichester, 1989; pp 41–63.
senschaftlichen Forschung (Project P-14687), and Len-
zing AG, Austria, for financial support.
13. Zhu, Y.; Zajicek, J.; Serianni, A. S. J. Org. Chem. 2001,
66, 6244–6251.
References
14. King-Morris, M. J.; Serianni, A. S. J. Am. Chem. Soc.
1987, 107, 3501–3508.
15. Bock, K.; Pedersen, C. Adv. Carbohydr. Chem. Biochem.
1983, 41, 27–66.
16. Goux, W. J. J. Am. Chem. Soc. 1985, 107, 4320–4327.
17. King-Morris, M. J. Carbohydr. Res. 1986, 154, 29–36.
1. Rosenau, T.; Potthast, A.; Sixta, H.; Kosma, P. Prog.
Polym. Sci. 2001, 26, 1763–1837.
2. Rosenau, T.; Potthast, A.; Millacher, W.; Adorjan, I.;
Hofinger, A.; Sixta, H.; Kosma, P. Discoloration of
Cellulose Solutions in N-Methylmorpholine-N-Oxide