This article was downloaded by: [University of Toronto Libraries]

On: 30 December 2014, At: 00:16

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Carbohydrate Chemistry

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lcar20

Hydrothermolysis of Cellulose Under

Static and Dynamic Conditions at High
Temperatures
a a
W. Schwald & O. Bobleter
a
Institute of Radiochemistry University of Innsbruck Innrain 52a ,
6020, Innsbruck, Austria
Published online: 05 Oct 2006.

To cite this article: W. Schwald & O. Bobleter (1989) Hydrothermolysis of Cellulose Under Static and
Dynamic Conditions at High Temperatures, Journal of Carbohydrate Chemistry, 8:4, 565-578, DOI:
10.1080/07328308908048017

To link to this article: http://dx.doi.org/10.1080/07328308908048017

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 J. CARBOHYDRATE CHEMISTRY, 8 ( 4 ) , 5 6 5 - 5 7 8 (1989)

HYDROTHERMOLYSIS OF CELLULOSE UNDER STATIC AND DYNAMIC

CONDITIONS AT HIGH TEMPERATURES

Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

W. Schwald and 0. Bobleter

Institute of Radiochemistry
University of Innsbruck
Innrain 52a, 6020 Innsbruck, Austria

R e c e i v e d January 20, 1988 - Final Form February 27, 1989

ABSTRACT

Cellulose (cotton) was hydrolyzed under high tempera-

ture conditions without any addition of acidic or alkaline
catalysts. Kinetic parameters of this hydrothermal process
were established in batch experiments at temperatures
ranging from 200 -
30OoC. The degradation of cellulose
followed a first-order reaction kinetic with an activation
energy of 129 kJ mol' .
Relatively small concentrations of
monomeric sugars were detected in the reaction solutions
due to the rapid decomposition into furfurals under static
conditions. In dynamic experiments, cellulose was hydro-
lyzed at various percolation rates. Several reaction pro-
ducts were analyzed quantitatively. Application of opti-
I 'pized parameters allowed almost complete solubilization of
,zellulose, with more than 60 % of the initial material
recovered as sugars.

INTRODUCTION

Hydrothermolysis is known as an autohydrolytic

process for the conversion of lignocellulosic materials

565

Copyright 0 1989 by Marcel Dekker, Inc.

 566 SCHWALD AND BOBLETER

into soluble products of different molecular weight. 1-3 A s

an alternative to processes based on acidic or alkaline
~atalysis~- the
~ reaction under hydrothermal conditions
uses only pure water at elevated temperatures. The required
temperature and the reaction time depend on the objective
and the plant material which is investigated.
Generally, a direct conversion of cellulose into
glucose in one step can only be accomplished at relatively
high temperatures. Lignocellulosic biomass, such as
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

aspenwood, birchwood or wheat straw is, preferably,

hydrolyzed in two steps,7'8 whereby in the first step the
hemicelluloses and the easily hydrolyzable lignin are
removed at temperatures between 180-22OOC. The resulting
water-soluble fraction contains oligo- and polymeric
hydrolysis products of hemicellulose, as well as low-
molecular weight degradation products derived from lignin. '
Subsequent enzymatic or acidic treatment of the remaining
cellulosic fraction allows the conversion of the insoluble
residue into glucose. It was shown that at temperatures
between 260 - 270' C also hydrothermal treatment gives
saccharification yields (approx. 50 % glucose) comparable
to acid catalysis.'' However, these results are based on
the main fraction, and therefore not representative for the
total conversion of the starting material. Moreover, a
further increase of the amount of sugars formed during the
reaction is desired.
With respect to the results of Saemanll and Fagan et
al., l2 who described the temperature dependance of acid hy-
drolysis of cellulose, an increase in reaction temperature
is expected to result in improved saccharification yields
obtained after hydrothermolysis. They have shown that, in
equation I, the reaction rate constant kl (hydrolysis of
cellulose) is affected more by increasing temperatures,
than is k2 (decomposition of glucose).

kl k2
Cellulose -Glucose -Decomposition products (1)
 HYDROTHERMOLYSIS OF CELLULOSE 56 7

Therefore, in the present work, we have investigated

the hydrothermal degradation of cellulose over a wide range
of temperatures up to 30OoC. Kinetic parameters were
established from studying the hydrolysis reaction under
.
static conditions (batch experiments) For dynamic experi-
ments a hydrothermolysis apparatus was constructed, which
allowed operation at pressures required for the high tem-
perature treatment. Reaction parameters were varied, in
order to optimize the saccharification yields for this per-
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

colysis process. The course of reaction was studied by ana-

lyzing a number of hydrolysis and degradation products.

RESULTS AND DISCUSSION

The course of reaction for the hydrothermal degra-

dation of cellulose (cotton) is shown in Fig. 1. As can be
seen the reaction follows a first-order kinetic. In con-
trast to earlier studies14, the calculation of reaction
rate constants kl for the hydrolysis of cellulose resulted
in considerably lower values. However, the observed diffe-
rence was to be expected, due to modifications in the expe-
rimental set-up. Whereas the studies inI4 were carried out
in glass ampules, the results described here were obtained
by using stainless steel autoclaves. Thus, any catalytic
effect caused by the glass wall (increased alkalinity in
the solution) could be avoided. Another difference was
that, in the present study, cotton cellulose was treated
hydrothermally, rather than cellulose from oat plants,
which were grown under a 14C02-containing atmosphere, where
consequently radiation induced crosslinking occurs.14
Evaluation of the Arrhenius-plot shown in Fig. 2 results
in an activation energy of

E, = 129.1 kJ x m01-l
 568 SCHWALD AND BOBLETER

'/o CELLULOSE

100
215O
0
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

50

- 274'

TIME (min)
I I I I I 1 b

Fig.1: Hydrothermal degradation of cellulose; percentage

of original cellulose as a function of time and
temperature ("C).

kjmin

Fig.2: Arrhenius-plot of the hydrolysis of cellulose

under hydrothermal conditions.
 HYDROTHERMOLYSIS OF CELLULOSE 569

Surprisingly, the value for this heterogenous reaction is

comparable with the activation energy obtained after
hydrothermal degradation of cellobiose in a homogenous
system.15 A similar Ea-value (133 kJ x mol'l) was obtained
for the hydrolysis of cellobiose under acidic conditions.
Ea-values for the acid catalyzed hydrolysis of cellulosic
substrates at temperatures between 170 and 240 OC are
reported to be ca. 180 kJ x mol-1.17'18 In comparison to
the activation energy for the high temperature hydrothermal
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

degradation of cellulose these values are considerably

higher. The fact that the Ea-values for both, the
homogeneous and the heterogeneous hydrothermal reactions
are virtually identical, suggests that at temperatures
between 200 and 300' C the hydrogen bonds are weakened to
such an extent that the cellulose under the given
conditions shows the reaction behaviour of a dissolved
carbohydrate.
When the experimentally determined kl-values are used
together with the reaction rate constants k2 for the
degradation of glucose, which are reported by Bobleter and
Pape," the glucose build up (G)t can be calculated
according to equation 111,

-klt -k2t
kl
( G I t = (C), FL ( e - e 1
k2 - kl

where (G)t represents the concentration of glucose at the

time t and (C), the initial amount of cellulose. The
concentration of glucose at t = 0 is zero and p is the
stoichiometric factor that corrects for the weight
increase, when cellulose is converted into glucose. The
corresponding curves for glucose production and degradation
are shown in Fig. 3. It is apparent that under isothermal
batch condiditons up to 4 % of the initial cellulose should
be recovered as glucose. The actual glucose concentrations,
 570 SCHWALD AND BOBLETER

GLUCOSE in O h of orig.
Cellulose

'""j
5.0
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

Fig.3: Calculated curves for the formation of glucose from

hydrothermolysis of cellulose as a function of time
and temperature ( " C ) according to equation 111.

however, found in the corresponding reaction mixtures were

substantially lower. This can be explained by the fact,
that under the given conditions isomerization products of
glucose are formed according to the Lobry-de-Bruyn-Alberda-
.
van-Ekenstein-rearrangement Similar to earlier studies15
a considerable amount of fructose, but no mannose, could be
detected in the respective solutions. Especially at the
reaction temperatures of 230 and 24OoC, the concentration
of fructose accounted for ca. 50 % of the glucose, present
in the solution. At higher temperatures and prolonged
reaction times the fast decomposition of both, glucose and
fructose, is reflected in relatively high concentrations of
HMF and, to a lesser extent, furfural.
 HYDROTHERMOLYSIS OF CELLULOSE 571

SOLUBLE P R O D U C T S

"i kG
32

/
/ /
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

10

2 4 6 8 1 0

Fig.4: Elution curves for the hydrothermolysis of cellulose

under dynamic conditions at 295OC with a percolation
rate of 12.5 ml/min;
G...glucose, C...cellobiose, Fru ...
fructose,
A...anhydro-glucose, H...HMF, F...furfural.

In order to avoid the rapid decomposition of the

glucose formed from cellulose, the reaction products,
mainly monomeric and some oligomeric sugars, have to be
removed from the reaction zone quickly. After a number of
preliminary experiments with the newly constructed
hydrothermolysis apparatus, and based on the kinetic
parameters described above, the reaction temperature was
increased even further. At 295OC a series of percolysis
experiments was performed with flow-rates ranging from 5.0
- 15.0 ml/min. This corresponds roughly to a 0.5- 1.5-fold
 572 SCHWALD AND BOBLETER

exchange of the reaction volume per minute. A typical

elution profile for the hydrothermal degradation of
cellulose is shown in Fig. 4 .
Under the given conditions (flow-rate = 12.5 ml/min)
the percentage of glucose obtained in the main fraction
(based on the solids content) exceeded 60 %. The con-
centration of glucose in the same fraction reached 3.6 %.
Although at lower flow-rates substantially higher glucose
concentrations can be achieved (i.e. more than 7.6 %
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

glucose at a flow-rate of 5.0 ml/min) , Table 1 shows, that

at the same time the over-all solubilization of cellulose
decreased markedly. As a result, more than 20 % of the
initial material remains insoluble and is, in part,
converted into tarry products. However, at flow-rates
higher than 10 ml/min the cellulose is hydrolyzed almost
quantitatively, with ca. 90 % being recovered as soluble
products. The difference is lost due to the formation of
volatile compounds, such as furfural and organic acids. Of
these soluble products close to 8 0 % have been identified
in this study. The distribution of various characterized
substances is shown in Fig. 5 . It is apparent that the
formation of glucose reaches a maximum at a flow-rate of
ca. 12.5 ml/min. While glucose accounts for ca. 50 % of the
starting material, (somewhat less than in the main fraction
alone) cellobiose and the reversion products fructose and
anhydro-glucose2* add another 13 % to the potentially
available sugars. Interestingly enough, the amount of HMF
remains virtually unchanged with increasing flow-rate.
The integral curves for the production of glucose at
the various flow-rates are shown in Fig. 6. The steep slope
of these curves indicates that the major part of the
cellulose is hydrolyzed within a relatively short time. It
should be noted that the total amount of glucose decreases
when the reaction vessel is percolated with the highest
flow-rate of ca. 15 ml/min. However, the apparent loss in
potential glucose is compensated by an increased amount of
cellobiose and higher oligomers. This points to the fact,
 HYDROTHERMOLYSIS OF CELLULOSE 573

TABLE 1

Mass Balance and Maximum Concentrations after Hydrother-

mal Degradation of Cellulose under High Temperature Per-
colysis Conditions ( 2 9 5 O C ) as a Function of the Flow-rate

F1 OW- Water sol ubl es Insoluble max . max.

rate Total Identified residue Losses Solidsa) Glucosea)
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

(ml/mi n 1
in % of orig. Cellulose in mg/ml

5.7 67.6 47.8 21.7 10.7 149.7 77.9

7.6 71.0 52.6 16.5 12.5 132.7 77.8
10.1 86.9 64.9 5.4 7.7 83.1 43.3
12.9 88.5 68.8 2.4 9.1 60.2 35.9
15.6 87.0 67.9 2.0 11.0 48.6 28.3

"Concentration of the fraction collected at the elution peak maximum

SOLUBLE PRODUCTS;
in '/, of orig. Cellulose

'i
40

- Fru

5
I &- , 10 15
FLOW
(mI/min)

Fig.5: Hydrothermolysis of cellulose at 295OC;

distribution of reaction products in percent of the
original cellulose as a function of the percolation
rate: product identification...same as in Fig.4.
 5 74 SCHWALD AND BOBLETER

GLUCOSI in OO/ of orig.

GLUCOSE;
Cellulose
12.9 mllmin

"1 J
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

Fig.6: Integral curves for the formation of glucose by

hydrothermal degradation of cellulose at 2 9 5 ° C as
a function of the percolation rate.

that due to the high liquid velocity the residence time

inside the reaction vessel is too short to allow complete
hydrolysis of the oligomers which are formed in the first
place. The situation is clearly reflected in the relative
concentrations of glucose and cellobiose. Whereas at a
flow-rate of only 5 ml/min the ratio of glucose :
cellobiose in the main fraction is 25.8, the corresponding
ratio of the 15 ml/min-experiment is merely 7.3. Although
the cellodextrins with higher molecular weight than
cellobiose could not be determined quantitatively, the low
ratio at 15 ml/min implies that under these conditions a
considerable amount of higher oligomers is present in the
reaction solutions.
 HYDROTHERMOLYSIS OF CELLULOSE 575

It is concluded that under optimized conditions

hydrothermolysis at high temperatures allows sacchari-
fication of cellulose with total sugar yields exceeding 60
% of the initial material. The evaluation of the kinetic
parameters indicates that hydrothermolysis of cellulose
exhibits a reaction mechanism which is distinctively
different from that of acid hydrolysis.
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

EXPERIMENTAL

Materials. Air-dry cotton cellulose with a moisture

content of 5% (oven-dry basis) was used for hydrolysis
experiments under static conditions. Cotton cellulose from
fabric clippings were used for hydrothermolysis in
percolation experiments.
Batch Experiments. Cellulose was placed into small
stainless steel autoclaves (180 x 8 mm I.D.). Destilled
water was added to give a final concentration of 10 mg/ml.
The autoclaves were kept in a metallic bath at temperatures
between 200 and 3OO0C for a predetermined period of time.
The reaction was then quenched immediately in a mixture of
ice/water. The reaction solution was recovered by
filtration through a sintered glass funnel. The solid
residue was washed for several times, dried at 105OC and
then weighed.
Percolation Experiments. Semi-continuous hydrother-
molysis was carried out in an apparatus, shown in Fig. 7.
Water without any additives is delivered by a high-pressure
pump, through a preheating unit into an electrically heated
reaction vessel with a reaction volume of 11 cm3. The
cellulose, which was placed inside the vessel, is
hydrolyzed by the action of hot water and the reaction
products are eluted continuously. After passing a cooling
unit and a control valve the reaction solution is collected
and fractionated. The control valve has to be adjusted
 576 SCHWALD AND BOBLETER
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

I FC
'a;
W

Fig.7: Flow-sheet and reaction vessel of the hydrothermo-

lysis apparatus:
HPP... high pressure pump, CV...check valve, SV...
safety valve, PHU...preheating unit, T...thermo-
couples, P. ..manometers, EH...heat exchangers,
...
RC...reaction cell, EfIC electrical heating coils,
NV... metering valves, W...waste, FC...fraction
collector.

such, that the pressure inside the system exceeds the

saturation pressure at a given temperature.

Analysis. Quantification of reaction products was

performed by HPLC. Carbohydrates were separated on an HPX-
87P-column (Bio-Rad Labs, California, USA) with water as
eluent.13 A refractive index detector was used for monitor-
ing the substances. Decomposition products (5-hydroxy-
methylfurfural (HMF) and furfural) were separated on a re-
versed-phase column (Nucleosil 5 c18, Machery, Nagel & Co.,
Duren F.R.G.) with methanol/water (35 : 65, v/v) as eluent.
For identification a W-detector was operated at 254 nm.
 HYDROTHERMOLYSIS OF CELLULOSE 577

ACKNOWLEDGEMENT

The authors wish to thank Ms. B. Hammerle for

skillful technical assistance.

REFERENCES

1. 0. Bobleter, and H. Binder, Holzforschunq, 34, 45

Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

(1980).
2. 0. Bobleter, R. Niesner, and M. Rohr, J. Annl. Polvm.
m. , 20, 2083 (1976).
3. * 0. Bobleter, and R. C o n c h , Cellul. Chem. Technol.,
13, 583 (1979).
4. R. Torget, M. Himmel, J . D . Wright, and K. Grohman,
ADD^. Biochem. Biotechnol., 17, 89 (1988).
5. A.H. Brennan, W. Hoagland, and D . J . Schell,
.
BiotechnoL. Bioenq S v m ~ ,
. No. 17, 53 (1986).
6. J.M. MacLeod, 2. Wood Chem.Technol., 2 (21, 187
(1982).
7. 0. Bobleter, W. Schwald, G. Bonn, and H. Esterbauer,
Austrian Patent No.A 102/85.
8. H.F. Hormeyer, G. Bonn, D.W. Kim, and 0. Bobleter,
-J. .
pood Chem. Technol , 7 ( 2 ) , 269 (1987).
9. E. Burtscher, 0. Bobleter, W. Schwald, R. C o n c h and
H. Binder, 2. Chromatosy., 390, 401 (1987).
10. G. Bonn, R. C o n c h , and O.Bobleter, wood m.
Technol. , 17, 195 (1983).
11. J . Saeman, m. m. Chem., 37, 43 (1945).
12. R.D. Fagan, H.E. Grethlein, Converse, and A.
Porteous, m. Sci. Technol. 5 ( 6 1 , 545 (1971).
13. W. Schwald, R. C o n c h , G. Bonn, and 0. Bobleter,
chromatoaranhia, 20 (l), 35 (1985).
14. G. Pape, Thesis, Univ. of Vienna, 1965.

15., 0. Bobleter and G. Bonn, Carbohvd. w., 124, 185

(1983).
 5 78 SCHWALD AND BOBLETER

16. 0. Bobleter, W. Schwald, R. Concin, and H. Binder,

-J.Carbohvdr. Chem., 5 (3), 387 (1986).
17. J.P. Franzidis, A. Porteous and J. Anderson,
215 (1983).
Conservation Becvcl., 5 (I),
18. H. Grethlein, z. m. Chew. Biotech&& ., 28, 296
(1978).
19. 0. Bobleter and G. Pape, Monatshefte, 99, 1560
(1968).

m.,4
Downloaded by [University of Toronto Libraries] at 00:16 30 December 2014

20. G. Bonn, 2. Carbohvdy. (31, 405 (1985).