Direct conversion of xylan into alkyl pentosides

Article abstract of DOI:10.1016/j.carres.2010.09.003

Xylan has been used as a raw material in the synthesis of butyl, octyl and decyl glycosides. Mixtures of d-xylose-, l-arabinose- and d-glucose-based surfactants were obtained under smooth conditions with high yields in a one-pot process. The surface activities of octyl and decyl glycosides thus obtained have been studied and compared with that of pure alkyl d-xylosides. The results have confirmed that the new synthetic approach described in this paper is a potentially economical and efficient method for the preparation of environmentally friendly surfactants.

Full text of DOI:10.1016/j.carres.2010.09.003

Carbohydrate Research 345 (2010) 2469–2473
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Direct conversion of xylan into alkyl pentosides
Florent Bouxin a, Sinisa Marinkovic a, Jean Le Bras b, Boris Estrine a,
⇑
a Agro-Industrie Recherches et Développement ‘Green Chemistry Department’, F-51110 POMACLE, France
b Institut de Chimie Moléculaire de Reims, UMR 6229, CNRS-Université de Reims Champagne-Ardenne, UFR des Sciences Exactes et Naturelles, BP 1039, 51687 REIMS Cedex 2, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 April 2010
Received in revised form 1 September 2010
Accepted 2 September 2010
Available online 7 September 2010
Xylan has been used as a raw material in the synthesis of butyl, octyl and decyl glycosides. Mixtures of D-
xylose-,
yields in a one-pot process. The surface activities of octyl and decyl glycosides thus obtained have been
studied and compared with that of pure alkyl -xylosides. The results have confirmed that the new syn-
L-arabinose- and D-glucose-based surfactants were obtained under smooth conditions with high
D
thetic approach described in this paper is a potentially economical and efficient method for the prepara-
tion of environmentally friendly surfactants.
Keywords:
Surfactants
Ó 2010 Elsevier Ltd. All rights reserved.
Pentoses
Glycosylation
Alkyl glycosides
Surface activity
1. Introduction
pentoses remains problematic. Their extraction from various
sources of lignocellulosic biomass usually requires elevated tem-
The increasing requirements placed on today’s consumer prod-
ucts make the entry into the surfactant market rather difficult. A
new successful surfactant has to be toxicologically safe, readily
biodegradable, efficient, compatible with other components, and
finally available at a competitive price. Alkyl polyglucosides (APGs)
are non-ionic surfactants produced from renewable raw materials
peratures and pressures, as well as a low loading of biomass (less
than 20%).7 This raises the energy demand of the overall surfac-
tant-generating process.8 Applying lower temperature and atmo-
spheric pressure limits the equipment and energy costs, but
imposes the use of high concentrations of acid (up to 100% based
on carbohydrate material). This limits extraction yields and ren-
ders the overall cost of the process above the market value of
any other process.9
In this context, we have developed a new green and sustainable
route for the direct conversion of xylan (obtained from waste or
by-products resources) into glycoside surfactants. This route, un-
like those of previous work,10 does not use isolated and purified
pentose monomers. This new process fits with Green Chemistry
guidelines and Biorefinery concepts as it allows a substantial
improvement of the energy balance (loading of xylan 20–33%,
90 °C, atmospheric pressure) and a diminution of the chemical
amount required (only 10% of acid) to build the surfactants.11 Fur-
thermore, the long-chain surfactants can be manufactured directly,
without the use of a glycosylation–transglycosylation process, as is
the case using the literature process.10
such as fatty alcohols from vegetable oils and
D-glucose from
starch. They have become important as complements to ethoxy-
lates1 and have a good impact on the environmental profile of
detergents. Amongst these chemicals, butyl and decyl alcohol-
based APGs are used as hydrotropes in detergent products or as
foam boosters in toiletries and personal care products.2 Although
the APG industry gather most of the required specifications de-
scribed above, huge efforts have been aimed by them at using
low-cost raw materials such as starch or glucose syrups with low
degradation levels (polymeric form).3 The main consequences of
working with these raw materials are the plant equipment require-
ments and higher production costs.4 Pentose-based surfactants are
also gaining interest as ingredients prepared from agricultural
waste.5 Here, the innovative approach consists in using agricultural
raw materials, which previously had little or no value (such as
wheat straw or corn stover), for the preparation of glycosidic sur-
factants. Although, by the hydrolysis of wheat or corn co-products,
2. Results and discussion
previously unavailable quantities of sugars, especially pentoses (L-
First of all, we analysed the free sugar weight fraction after
hydrolysis of the xylan using a standard method13 and found a com-
arabinose and D
-xylose), can be obtained,5,6 the manufacture of
position in accordance with the literature data: 53% of
of -arabinose and 7% of
-glucose.14 The rest of the material was
mainly ash (4%) and lignin (see details in Supplementary data).
D-xylose, 7%
L
D
⇑
Corresponding author. Tel.: +33 326 054 284; fax: +33 326 054 289.
E-mail address: b.estrine@a-r-d.fr (B. Estrine).
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2010.09.003

2470
F. Bouxin et al. / Carbohydrate Research 345 (2010) 2469–2473
The sugar composition thus obtained has been used to calculate the
respective available amounts of -xylose, -arabinose and -glucose
sides were obtained. Other water concentrations evaluated did
not afford such results (Table 1, entries 6 and 8).
D
L
D
from the xylan and to deduce the maximum yield of each
Overall, the results of Table 1 have proven that the reactivity of
xylan in the synthesis of glycosides is better than that of starch, as
the latter requires more drastic conditions.4 The difference of the
reactivity between starch and xylan is probably due to the nature
glycoside.12
The initial evaluation of the reaction was carried out with a dis-
persion of oat spelt xylan in n-butanol (20% xylan based on n-buta-
nol weight) at 90 °C and atmospheric pressure with only 5%
sulfuric acid (Table 1, entries 1–4). Under these conditions, no pro-
duction of furfural or hydroxymethyl furfural was observed in any
of the trials.15 This could be explained by the smooth temperature
and pressure conditions applied compared with those required for
of the oligosaccharide chain, since
a-(1?4)-glucosidic bonds are
more stable than b-(1?4)-«xylosidic» linkages.18 Furthermore,
the higher proportion of nonlinear chains for starch and the ab-
sence of gelatinisation for xylan can also contribute to this differ-
ence. Finally, the fact that the chains of starch are longer (1000–
100,000 glucose units) than those of xylan (500–3000 sugar units)
probably contributes to a lower accessibility to the monomers dur-
ing the hydrolysis step.
the synthesis of n-butyl
-glucosides from starch (165 °C, 20 bars).4
D
We recovered 58% of the starting xylan when varying amounts of
water were added to the reaction medium (Table 1, entries 2–4)
and 72% without water addition (Table 1, entry 1). These results
all showed an incomplete conversion of the starting material. Nev-
ertheless, we observed a quantitative formation of n-butyl L-arabi-
nosides (Table 1, entries 1–4). This can be explained by the good
availability and known fast hydrolysis rate of the arabinofuranose
Having demonstrated the favourable behaviour of xylan in a di-
rect synthesis of butyl glycosides, we next attempted to improve
the efficiency of this methodology by performing the reaction in a
more concentrated medium (33% xylan in alcohol) and keeping
the temperature and pressure conditions of the tests reported in Ta-
ble 1. We evaluated the results at 1, 2 and 3 h of reaction time
moieties along xylooligosaccharide chains.16
A
D
-xyloside yield of
42% was reached when the amount of water was up to 2.5% (Ta-
ble 1, entries 2–4). The -arabinoside/ -xyloside ratios were in
the range of 0.31 (Table 1, entry 4) to 0.53 (Table 1, entry 1) and
can be compared to the molecular ratio between the -arabinose
and the
-xylose present in the starting xylan (0.13).17 It is also
(Fig. 1). As previously observed (Table 1), L-arabinosides were
L
D
formed quickly at the beginning of the reaction (Fig. 1). The best bu-
tyl pentoside yields were obtained after 3 h of reaction. The global
yield of glycosides remained lower than that obtained in Table 1,
L
D
entry 7, which was particularly due to the yield of D-glucosides.
interesting to point out that GC analysis of the filtrates indicated
the presence of very low concentrations of monomeric sugars
We also attempted to apply these reaction conditions to higher
alcohols. The results obtained with n-octanol are presented in Fig-
ure 2. The yields of the octyl pentosides were quasi-quantitative in
only 2 h, although a lower accessibility of the sugar material was
expected in the n-octanol phase. We finally carried out the reaction
in n-decanol with equal success.
We isolated the glycosides prepared from xylan after neutrali-
sation of the acid and distillation of the excess of alcohol. The crude
glycoside mixtures were recovered without any further purifica-
tion step, and their purity was evaluated by GC (Table 2). The gly-
coside purities of the crude mixtures were 65% for butyl glycosides,
83% for octyl glycosides, and 75% for the decyl glycosides. Such re-
sults are of interest considering the sugar purity of the starting
material (67% sugars). The lower purity of butyl glycosides is
potentially due to solubilisation of ‘non-sugar’ materials during
the glycosylation step. This would be explained by the higher
(mainly
served as their
D
-xylose). Furthermore, the n-butyl pentosides were ob-
and b pyranoside and furanoside forms. In view
a
of these results, two possible mechanisms can be proposed for
the direct synthesis of alkyl pentosides from xylan. The first one
would be a transglycosylation reaction followed by anomeric
equilibration under acidic conditions. A second possibility would
be a one-pot hydrolysis–glycosylation mechanism. We pursued
our study by raising the sulfuric acid concentration (10%). The
weight of recovered xylan became lower than those of trials with
5% of acid (Table 1, entries 5–8). Without any addition of water
(Table 1, entry 5), a D-xyloside yield of 57% was obtained. This yield
can be compared to the 42% obtained in an experiment done at 5%
acid without addition of water (Table 1, entry 1). At a water con-
centration of 5% (Table 1, entry 7), optimal yields of butyl glyco-
Table 1
Synthesis of n-butyl glycosides starting from xylan and starch*
O
HO
O
O
HO
O
HO
O
OH
H2SO4 (cat.) / H2O (cat.)
Butanol /
O
O
O
HO
+
OH
OH
HO
HO
OH
O
OH
Δ
O
O
O
HO
OH
OH
Entry
H2SO4/H2Oa
Xylan (%)b
n-Butyl glycosides yieldsc (%)
AraC4
XylC4
GluC4
1
2
3
4
5
6
7
8
5/0
5/2.5
5/5
5/10
10/0
10/2.5
10/5
10/10
72
58
58
58
48
19
18
18
100
100
100
100
75
100
100
75
25
42
42
42
57
88
96
88
0
0
0
0
25
75
100
50
a
Sulfuric acid wt % compared to xylan weight. Water wt % compared to n-butanol weight.
Xylan weight fraction recovered by filtration. Unrecovered fraction consists of solubilised and hydrolysed xylan and xylan converted into
b
glycosides
c
Yield of monoglycosides determined by GC method,12 and based on
Mixture of glycosides obtained: furanosides and glucosides are not represented for clarity.
D-xylose,
L-arabinose and D-glucose available in starting xylan.
*

F. Bouxin et al. / Carbohydrate Research 345 (2010) 2469–2473
2471
100
90
80
70
60
50
40
30
20
10
0
Ara. % yield
Xyl. % yield
Glu. % yield
Xylan % conv.
1 h.
2 h.
3 h.
Figure 1. Production of n-butyl glycosides from xylan as a function of time.
100
90
80
70
60
50
40
30
20
10
0
Ara. % yield
Xyl. % yield
Glu. % yield
Xylan % conv.
1 h.
2 h.
3 h.
Figure 2. Production of n-octyl glycosides from xylan as a function of time.
Table 2
Composition of isolated crude surfactants and starting xylan
xylosides that favourably impacted the purity of the crude mix-
tures recovered. This is probably due to the impurities of xylan that
are not solubilised in the crude mixture when octanol and decanol
are used, which resulted in a higher purity of the glycoside prod-
ucts. Finally, the weights of each of the glycoside mixtures recov-
ered (starting from 5 g of xylan at 67% sugar purity) are also
indicated in Table 2 and can be compared to the results reported
Weight (g)
Alcohol
Sugars (%)a
Alkyl monoglycosidesa (%)
Ara
Xyl
Glu
6
7
7
n-Butanol
n-Octanol
n-Decanol
6
0.5
1
54
5
5
73
65
53b
5
5
5
Xylan
67
7b
7b
in the literature, where only 103 g of butyl D-glucosides were ob-
tained from 100 g of starch under drastic conditions.4
a
Concentration of free sugars (total of arabinose, xylose and glucose) and
monoglycosides in the crude mixtures determined by GC method.13
Long-tailed glycosidic surfactants are known to be good foam
boosters in cosmetics and impart exceptional degreasing properties
to detergents. That’s why the surface activity of crude octyl and de-
cyl glycosides from xylan were studied and compared to pure glyco-
sides obtained by the Fischer glycosidation reaction. Foam
properties were also evaluated using the Ross–Miles foam height
test (Table 3 and Section 3.4). First of all, concerning a CMC that is
b
Arabinose, xylose and glucose concentrations in xylan material determined by
GC after hydrolysis.
polarity of n-butanol. Nevertheless, the distribution of butyl glyco-
sides is similar to those of sugars in the starting xylan. For the
other glycoside mixtures, we observed higher concentrations of

2472
F. Bouxin et al. / Carbohydrate Research 345 (2010) 2469–2473
Table 3
xylan), 25 g of n-butanol or 15 g of n-decanol, were progressively
heated to 90 °C in an oil bath at atmospheric pressure. Sulfuric acid
(5–10% of sugar weight) and distilled water (2.5–10% of alcohol
weight) were then added dropwise, and the reaction medium
was maintained under stirring (800 rpm) for 3 h. The remaining so-
lid was filtered off over a cellulose filter. The residual sugar mate-
rial was washed with acetone, dried at 90 °C under vacuum and
weighed. The filtrate was neutralised by sodium hydroxide solu-
tion and adjusted to pH 9 (measured in a 10 weight % solution of
water and 2-propanol 1:1 v/v).
Surface and foam properties of glycosides’ compositions obtained from xylan and D-
xylose
Surfactant
composition
CMC
c
cmc
Foam volume Foam at
at t = 0 (mL) t = 20
min (mL)
(mg LÀ1
)
(mN mÀ1
)
Octyl glycosides from
xylan (Table 2)
200
953
35
28
120
280
0
Octyl
D
-xylosides from
20
D
-xylose13
Decyl glycosides from
xylan (Table 2)
230
301
27
28
490
480
370
360
As the reaction products are only rarely commercially available,
alkyl glycoside standards were prepared by the Fischer glycosida-
Decyl
-xylose13
Decyl xylosides from
-xylose Na2SO4 solutiona
D-xylosides from
D
tion of D-xylose, and L-arabinose in an excess of n-butanol, n-octa-
212
28
480
360
nol and n-decanol.21 Each product was purified using column
chromatography to give a mixture of glycoside anomers. Their puri-
ties were checked by 13C and 1H NMR spectroscopy. The GC calibra-
tion was performed afterwards using these purified products.
D
a
Concentration of Na2SO4 is 1 mol %.
a good cursor of surfactant efficiency, it is difficult to explain the dif-
ferences between -xylosides obtained from xylan or -xylose
3.3. Surface tension measurements
D
D
sources. Obviously, impurities such as electrolytes (sodium sulfate)
could play a significant role. Sodium sulfate arises from the neutral-
isation of sulfuric acid and induces salting-out effects. This is dis-
The critical micelle concentration (CMC) of the surfactants was
determined with a KRUSS Processor tensiometer (K100) by the
Wilhelmy method (KSV, Finland) using a platinum plate sensor
in a jacketed cell. The temperature was controlled using a Julabo
thermocontroller with an accuracy of 25 0.5 °C. The surfactant
solutions were stirred and allowed to stand at equilibrium at the
set temperature before measuring the surface tension. The surface
tension of the solutions was obtained as a mean value of ten mea-
surements. The CMC was graphically determined at the break of
the curve of the surface tension measured versus the concentration
of surfactants in solution.22
played by the change in the CMC of decyl
addition of 1 mol % of sodium sulfate.19 Potential synergistic behav-
iour between -xylosides, -arabinosides and -glucosides could
D-xylosides after the
D
L
D
also explain the fact that the CMC values of xylan-based surfactants
are lower. Except for octyl glycosides from xylan that exhibit a
hydrotrope-like behaviour and poor foaming properties,20 all com-
positions showed similar effectiveness as shown by surface tensions
between 26 and 28 mN mÀ1 at CMC. Interestingly, the foam power
and stability of the decyl glycosides were not affected by impurities
and were very similar to the properties of corresponding alkyl
xylosides.
D-
3.4. Foam properties
In conclusion, we have shown that the conversion of xylan into
alkyl monoglycosides in high yields is possible via a single-step
operation. We carried out this reaction under smooth conditions
reducing at the maximum the potential difficulties for scaling-up
The foaming power is determined according to the Ross–Miles
protocol derived from standard ISO 696 and AFNOR NFT 73-404,
diluting the surfactant compositions (0.1 wt % in demineralised
water) so as to represent the real conditions of foam creation dur-
ing the use of a shampoo.23 The method consists in measuring the
volume of foam obtained after dropping, from a height of 450 mm,
500 mL of a solution of composition to be tested, onto a liquid sur-
face of the same solution (100 mL). When the solution reaches the
graduation mark (600 mL), the chronometer is then started. The
volume of the foam is measured upon starting the chronometer
and after 20 min. The volume of the foam is measured between
the foam/liquid horizontal interface and the base of the foam/air
interface.
the methodology. The yields of n-butyl D-xylosides and L-arabino-
sides thus obtained represent a clear improvement over earlier re-
sults from the conversion of starch into butyl glucosides.4 These
yields are also much higher than those reported by enzymatic pro-
cesses for the conversion of xylan into alkyl D
-xylosides.19 The sur-
face activities of crude octyl and decyl pentoside mixtures obtained
using this new methodology were studied and compared with pure
alkyl glycosides obtained from pure D-xylose. Although the pento-
sides obtained from xylan were impure materials, their behaviour
is indicative of their potential use in the industrial sector as deter-
gents. Furthermore, we have confirmed that the new synthetic ap-
proach described in this paper is a potentially economical and
efficient method for the preparation of powerful environmentally
friendly surfactants that fit within Biorefinery guidelines.11
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.carres.2010.09.003.
3. Experimental
References
1. Biermann, M.; Schmid, K.; Shulz, P. Henkel Referate 1994, 30, 7–15.
2. Rybinski, W.; Hill, K. Angew. Chem., Int. Ed. 1998, 37, 1328–1345.
3. (a) Roth, C. D.; Moser, K. B.; Bomball, W. A. Eur. Patent 35,589, 1980.; (b) Van
der Burgh, L. F.; Sommer, S. J. Eur. Patent 92,355, 1983.; (c) Lüders, H. Chemical
Properties and Derivatization of Surface-Active Alkyl Glycosides and Alkyl
Polyglucosides. In Nonionic Surfactants/Alkyl Polyglucosides, Surfactants Science
Series; Balzer, D., Lüders, H., Eds.; Marcel Dekker Ltd: New York, 2000; Vol. 91,
pp 77–83; (d) Lüders H. Eur. Patent 253,996, 1987.
3.1. Materials and chemicals
n-Butanol, n-octanol, n-decanol, H2SO4 (96%) and NaOH (50%
solution) were of analytical grades and were used as received (Ac-
ros). Xylan from oat spelt (Fluka) and starch (Chamtor) were used
as received.
4. Eskuchen, R.; Nitsche, M. Technology and Production of Alkyl Polyglycosides. In
Alkyl Polyglycosides: Technology, Properties and Applications; Hill, K., von
Rybinski, W., Stoll, G., Eds.; VCH: Weinheim, 1997.
3.2. Glycosylation experiments and product analysis
5. (a) Estrine, B.; Bouquillon, S.; Hénin, F.; Muzart, J. Eur. J. Org. Chem. 2004, 13,
2914–2922; (b) Estrine, B.; Bouquillon, S.; Hénin, F.; Muzart, J. Green Chem.
2005, 7, 219–223; (c) Hadad, C.; Damez, C.; Bouquillon, S.; Estrine, B.; Hénin, F.;
In a 100-mL three-necked round-bottomed flask, equipped with
a magnetic stirring apparatus, 5 g of sugar material (starch or

F. Bouxin et al. / Carbohydrate Research 345 (2010) 2469–2473
2473
Muzart, J.; Pezron, I.; Komunjer, L. Carbohydr. Res. 2006, 341, 1938–1944; (d)
Estrine, B.; Bouquillon, S.; Hénin, F.; Muzart, J. Appl. Organomet. Chem. 2007, 21,
945–946.
14. Sun, H. J.; Yoshida, S.; Park, N. H.; Kusakabe, I. Carbohydr. Res. 2002, 337, 657–
661.
15. Liu, C.; Wyman, C. E. Ind. Eng. Chem. Res. 2004, 43, 2781–2788.
16. Sun, R.; Lawther, J. M.; Banks, W. B. Cellul. Chem. Technol. 1997, 31, 199–212.
17. Kontturi, E.; Vuorinen, T. Holzforschung 2006, 60, 691–693.
18. (a) Bochkov, A. F.; Zaikov, G. E. The Chemistry of the O-Glycosidic Bond;
Pergamon Press: Oxford, 1979; (b) Bemiller, J. N. Adv. Carbohydr. Chem. 1967,
22, 25–108; (c) Szejtli, J. Säurehydrolyse Glycosidischer Bindungen; Budapest:
Académiai Kaldo, 1976.
6. Klass, D. L.. Biomass for Renewable Energy and Fuels. In Encyclopedia of Energy;
Elsevier: New York, 2004.
7. (a) Rakman, S. H. A.; Choudhury, J. P.; Ahmad, A. L. Biochem. Eng. J. 2006, 30, 97–
103; (b) Nguyen, Q. A.; Tucker, M. P.; Keller, F. A.; Eddy, F. P. Appl. Biochem.
Biotechnol. 2000, 84, 561–576; (c) Deguchi, S.; Tsujii, K.; Horikoshi, K. Green
Chem. 2008, 10, 623–626.
8. Hirsinger, F. Life-Cycle Inventory. In Alkyl Polyglycosides: Technology, Properties
and Applications; Hill, K., von Rybinski, W., Stoll, G., Eds.; VCH: Weinheim, 1997.
9. (a) Harmer, M. A.; Fan, A.; Liauw, A.; Kumar, R. K. Chem. Commun. (Cambridge)
2009, 43, 6610–6612; (b) Torget, R.; Hatzis, C.; Hayward, T. K.; Hsu, T. A.;
Philippidis, G. P. Appl. Biochem. Biotechnol. 1996, 57, 85–101; (c) Lloyd, T. A.;
Wyman, C. E. Bioresour. Technol. 2005, 96, 1967–1977.
10. Bertho, J. N.; Mathaly, P.; Dubois, V.; De Baynast, R. U.S. Patent 5,688,930, 1997.
11. Kamm, B.; Kamm, M.; Schmidt, M.; Hirth, T.; Schulze, M. Biorefinery Systems—
An Overview. In Biorefineries—Industrial Processes and Products; Kamm, B.,
Gruber, P. R., Kamm, M., Eds.; VCH: Weinheim, 2006.
19. Gibbs, J. W.. In The Collected Works of J. W. Gibbs; Longmans: London, 1928;
Vol. I. pp 119–120.
20. Matero, A.; Mattson, A.; Svensson, M. Journal. Surfactants Deterg. 1998, 1, 485–
489.
21. For
Suling, W. J.; Gurcha, S. S.; Besra, G. S.; Reynolds, R. C. Bioorg. Med. Chem. 2001,
9, 3145–3151; For -xylosides standards see (b) Matsumura, S.; Kinta, Y.;
L-arabinoside standards see: (a) Pathak, A. K.; Pathak, V.; Maddry, J. A.;
D
Sakiyama, K.; Toshima, K. Biotechnol. Lett. 1996, 18, 1335–1340.
22. Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 3rd
ed.; Marcel Dekker: New York, 1997.
12. Landreat, C. Ph.D. Thesis, University of Reims, 1996.
23. Standard method AFNOR NFT 73-047.
13. Boeriu, C. G.; Bravo, D.; Gosselink, R. J. A.; van Dam, J. E. G. Ind. Crops Prod. 2004,
20, 205–218.