Acetylated Starch Co2

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

RESEARCH ARTICLE

Process-product studies on starch acetylation


reactions in pressurised carbon dioxide
Henky Muljana, Francesco Picchioni, Hero J. Heeres and Leon P. B. M. Janssen
Department of Chemical Engineering, University of Groningen, Nijenborgh, Groningen, The Netherlands
An in-depth study on the effect of process conditions (pressure, temperature and type of
catalyst) on the acetylation of starch with acetic anhydride in pressurised carbon dioxide is
described. A total of 22 experiments were performed and the experimental data were
analysed using non-linear multivariable regression. The highest degree of substitution
(DS) value (0.46) was obtained using K
2
CO
3
as the catalyst at 908C, 15 MPa pressure,
and a catalyst to starch ratio of 0.5 mol/mol anhydroglucose units (AGUs). Important product
properties of the acetylated starch prepared in CO
2
like viscosity in water and relevant
thermal properties were determined and compared with typical products prepared in an
aqueous system.
Received: February 20, 2010
Revised: May 20, 2010
Accepted: May 21, 2010
Keywords:
Acetylation / Potato starch / Starch acetate / Supercritical CO
2
1 Introduction
Starch acetates are commercially produced on large scale
and used in many applications due to their favourable
product properties compared to native starch [1].
Examples are the use as thickening agents in a variety
of food products, as sizing agents for the textile (wrap
sizing) and paper (surface sizing) industry as well as
gummed tape adhesives [2].
The acetylation reactions are typically carried out in
water, though organic solvents also have shown to be
effective [36]. The reaction is catalysed by alkaline bases,
well known examples are sodium hydroxide (NaOH),
sodium acetate (NaOAc), sodium carbonate (Na
2
CO
3
)
or disodium hydrogen phosphate (Na
2
HPO
4
). Selectivity
is a key issue and considerable amounts of acetic acid may
be produced by the simultaneous hydrolysis reaction of the
anhydride with water.
An attractive alternative solvent for starch modication
reactions is pressurised CO
2
. It has proven to be an
excellent solvent and processing aid for polymer process-
ing [7, 8]. CO
2
is considered to be green, it is non-am-
mable, relatively non toxic and inert. Another advantage is
the ease of separation from the (reacting) system by simple
depressurisation [911]. Furthermore, when operation at
supercritical conditions is required, these conditions are
not very extreme (T
c
318C, P
c
7.38 MPa) and simplify
experimental studies on lab scale and subsequent indus-
trial processing.
Pioneering studies on the use of CO
2
for starch modi-
cations were reported by Harris et al. [6]. It was demon-
strated that, among others, acetylation of high amylose
starch (Hylon RTM VII, 70% amylose content) is possible
in supercritical CO
2
and products with DS values ranging
from 0.2 to 2.4 could be obtained [6]. However, systematic
studies for the acetylation reaction, particularly for potato
starch, and relevant product properties are not reported in
this paper. We have recently reported exploratory studies
on the synthesis of starch acetate in pressurised CO
2
using acetic anhydride as the reagents and sodiumacetate
as catalyst in sub- and supercritical CO
2
[12]. In addition,
thermodynamic properties of the system starch/CO
2
like
Correspondence: Professor Hero J. Heeres, Department of
Chemical Engineering, University of Groningen, Nijenborgh 4,
9747 AG Groningen, The Netherlands.
E-mail: h.j.heeres@rug.nl
Fax: 31-50-3634479
Abbreviations: AAH, analytical grade acetic anhydride; AGU,
anhydroglucose unit; DS, degree of substitution; GC, gas
chromatograph; TGA, thermogravimetric analysis.
DOI 10.1002/star.201000021 566 Starch/Sta rke 2010, 62, 566576
2010 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim www.starch-journal.com
CO
2
solubility [13] and effects of CO
2
on starch gelatini-
sation [14] have been explored.
The present work concerns process-product studies on
the base catalysed acetylation of starch in pressurised
CO
2
. The effects of temperature, catalyst ratio and pres-
sure on the product degree of substitution (DS), the anhy-
dride conversion (X
AAH
) and the selectivity of the reaction
(S
SA
) values were quantied for a broad experimental
window. The experimental results were modelled using
non-linear multivariable regression. Important product
properties of the acetylated starch prepared in CO
2
like
viscosity in water and relevant thermal properties were
determined and compared with typical products made in
an aqueous system.
2 Materials and methods
2.1 Materials
Potato starch was kindly supplied by AVEBE (The
Netherlands). The water content was determined by drying
the potato starch in a vacuum oven at 508C until constant
weight and was found to be 16.2% w/w. The acetylated
starch products prepared in water (DS 0.2 and DS 0.39,
synthesis at room temperature) were also supplied by
AVEBE (AV). Analytical grade acetic anhydride (AAH)
and alkaline base salt catalysts (potassium carbonate
(K
2
CO
3
), sodium carbonate (Na
2
CO
3
), magnesium
acetate (Mg(OAc)
2
, sodium acetate (NaOAc) and diso-
dium hydrogen phosphate (Na
2
HPO
4
)) were purchased
from Merck (Germany). DMSO-d
6
(DMSO-d
6
) was
obtained from Aldrich (Germany). All chemicals were
used as received without further purication. High purity
CO
2
(99.7% volume) and N
2
(99.9% volume) were
purchased from Hoekloos (The Netherlands).
2.2 Experimental set-ups
The high pressure reactor set-up consists of a stirred batch
reactor (Parr Instrument), an electrical heating element
with temperature controller, a high pressure pump unit,
and CO
2
and N
2
storage bottles (Fig. 1). The reactor
has a capacity of 100 mL and may be operated in a
temperature range from 10 to 3508C and a maximum
pressure of 35 MPa.
The CO
2
was supplied to the reactor using a membrane
pump (Lewa, Leonberg, Germany) with a capacity of
60 kg/h at a maximum pressure of 35 MPa. To prevent
cavitation in the pump, the CO
2
is rst cooled to 08C in a
heat exchanger (Huber, Offenburg, Germany). After pres-
surising, a second heat exchanger with hot-oil is used to
heat the CO
2
to the desired temperature.
2.3 Experimental section
2.3.1 Starch acetylation reactions in
supercritical CO
2
(scCO
2
)
Potato starch was acetylated in the high pressure set-up by
reacting native potato starch with acetic anhydride (AAH)
and an alkaline base salt as the catalyst. Native potato
starch (3.53 g), AAH (4.35 mol/mol anhydroglucose unit
(AGU)) and the salt (0.10.5 mol/mol AGU) were charged
to the reactor. The reactor was ushed with N
2
to remove
traces of air. Next, the autoclave was pressurised with CO
2
and heated till the desired temperature. CO
2
was added to
increase the pressure to the desired level. The reaction
was performed at a stirring rate of 1750 rpm. After reac-
tion, the reactor was cooled to room temperature and
depressurised and the solid products were separated from
the liquid phase using a syringe. The solid product was
washed several times with water (1.5 L) to remove acetic
Figure 1. Schematic drawing of high
pressure reactor set up.
Starch/Sta rke 2010, 62, 566576 Process-product studies on starch acetylation reactions 567
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
acid and unreacted acetic anhydride, ltered and dried in a
vacuum oven (5 mbar) at 508C until constant weight. The
products were collected for further analyses. The amount
of unreacted AAH in the liquid phase after reaction was
determined using a gas chromatograph (GC).
2.4 Analytical equipment
The product DS was determined using
1
H NMR measure-
ments performed on an Oxford NMR AS 400 spectrometer
(Varian, Palo Alto, CA, USA) operating at 400 MHz. The
products (25 mg) were dissolved in DMSO-d
6
(2 g). The
spectra were measured at 608C with 64 scans. The
residual protons of the NMR solvent (DMSO-d6) at d
2.5 ppm were used as the reference.
The chemical composition of the liquid phase after
reaction was determined using a GC HP 5890 series II
(Avondale, PA, USA) equipped with an FID detector and
HP ve column (length 30 m and id 0.25 mm). The
GC was operated at an injector temperature of 2808C, an
oven temperature of 1008C, a detector temperature of
3008C, and an inlet pressure of 86 kPa. Helium was used
as the carrier gas with a ow rate of 2.2 mL/min.
Thermogravimetric Analysis (TGA) measurements were
performed on a Perkin Elmer TGA 7 Thermogravimetric
Analyser (Perkin Elmer. Norwalk, CT, USA). The samples
were heated to 9008C in an inert atmosphere with a heating
rate of 108C/min. DSC analyses were performed using a TA
Instrument DSC 2920 (TA Instrument, New Castle, NJ,
USA). Prior the measurements, the DSC was calibrated
using Indium as the standard. About 10 mg of sample
was placed in an aluminium pan and sealed. The samples
were treated in two heating cycles. The rst heating run was
performed to erase the material thermal history. After that,
the samples were cooled to 08C and then heated again to
2008C(with heating/cooling rate of 108C/min). An empty pan
was used as the reference.
The viscosity was measured on a RVA Super 4 instru-
ment fromNewport Scientic Pty. Ltd, Australia. The starch
suspensions (5 g of 3%w/w, dry based) were placed inside
the canister and stirred with a paddle speed of 160 rpm.
The suspensions were heated from 30 to 958C (heating
rate of 58C/min), subsequently the temperature was main-
tained at 958C for 6 min, cooled to 508C (cooling rate of
58C/min) and nally maintained at 508C for 10 min.
SEM pictures were obtained using a eld emission
scanning electron microscope from JEOL (type 6320F,
Jeol, Tokyo, Japan).
2.5 Determination of the product DS
The DS of the acetylated products was calculated from
1
H NMR spectra by comparing the unit area of the acetate
protons (A
H-ace
, at d 1.92.1 ppm) with the unit area of
the starch protons (A
H-agu
, at d 3.25.6 ppm) using the
following equation [15, 16]:
DS
7A
H-
ace
3A
H-
agu
(1)
where A
H-ace
is the area of the methyl signals and A
H-agu
is
the area of the proton signals of the AGU.
2.6 Determination of AAH conversion (X
AAH
)
and product selectivity (S
SA
)
The acetic anhydride conversion (X
AAH
) and the product
selectivity (S
SA
) were calculated on the basis of the chemi-
cal composition of the liquid phase after reaction (GC
analyses). Here, X
AAH
and S
SA
are dened as:
X
AAH

n
AAH;o
n
AAH
n
AAH;o
100%
n
AAH;reacted
n
AAH;o
100% (2)
S
SA

n
AAH;p
n
AAH;reacted
100% (3)
In Eq. 2 and Eq. 3, n
AAH,o
is the initial moles of AAH,
n
AAH
is the moles of unreacted AAH, and n
AAH,p
and n
AAH,
reacted
are the moles of acetate in the starch acetate prod-
uct and the total moles of reacted AAH, respectively.
2.7 Statistical modelling
The output variables (DS and X
AAH
) were modelled using
non-linear multivariable regression (Eq. 4) using the three
independent variables (temperature, pressure and catalyst
ratio).
y
k
b
o

X
3
i 1
b
i
x
i

X
3
i 1
b
ii
x
2
i

X
3
i 1
X
3
j 1
b
ij
x
i
x
j
(4)
Due to singularity issues in the matrix calculations, the
quadratic term of catalyst ratio (x
2
3
) was excluded from
the model. All regression coefcients (b
i
, b
ii
and b
ij
) and the
intercept (b
o
) were calculated using Mathcad
1
(Mathsoft)
software. A t-test (for statistical signicance) was per-
formed for regression coefcient and the non signicant
terms were deleted from the model [17].
3 Results and discussion
3.1 Catalyst screening experiments
A number of basic salts were screened at standard con-
ditions (15 MPa, 908C, catalyst to starch ratio of 0.1 mol/
mol AGU, and 1 h reaction time) to evaluate their catalytic
performance. After reaction, the product was washed
thoroughly with water to remove unreacted reagents
568 H. Muljana et al. Starch/Sta rke 2010, 62, 566576
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
and catalyst and dried at 508C and 5 mbar till constant
weight. After work-up, the products were obtained as white
granular powders. The DS of the products was sub-
sequently determined by
1
H NMR.
K
2
CO
3
is the best catalyst in the series and gives the
highest DS value (Fig. 2). The least active catalyst is
Mg(OAc)
2
with the lowest DS (0.08) value. The acetic
anhydride conversion was far from quantitative after 1 h
reaction time and was between 40 and 47% for all cata-
lysts. To obtain insights in catalyst structure-performance
relations, the DS of the product was related to the pK
b
[18]
of the catalyst in water. Although the system under study is
not an aqueous system, the acetylation reaction takes
place in the starch matrix, which, certainly at the beginning
of the reaction, contains considerable amounts of water.
The results are given in Fig. 2.
It is clear that the DS decreases with the basicity of the
salt. The only exception is NaOAc, which gives a much
higher DS than expected. A reduction of the DS at lower
basicity of the catalyst is expected on the basis of the
proposed reaction mechanism (Scheme 1) [1921]. The
rst step in the mechanism involves the deprotonation of a
starchOH group by the base to form the intermediate
starch alkoxide (STO

) [19-21]. This starch alkoxide sub-


sequently reacts with the acetic anhydride to form the
desired product (starch acetate). According to this
sequence, it is anticipated that the rate of the acetylation
reaction will be higher when using a stronger base catalyst,
in line with our experiments. At this stage it remains specu-
lative why the performance of NaOAc is much better than
Mg(OAc)
2
as both salts have a similar pK
b
. A possible
explanation is the involvement of cation effects (Mg
2
versus Na

). Such cation effects have been reported for


the reaction of starch with propylene oxide [20, 22]. Further
research outside the scope of this paper will be required to
evaluate cation effects for the acetylation of starch in
pressurised CO
2
. As the result fromthis catalyst screening
study, K
2
CO
3
was chosen as the catalyst for the system-
atic studies reported in the next paragraph.
3.2 Systematic studies on the acetylation of
starch in pressurised CO
2
In a subsequent systematic study, the effect of tempera-
ture (50908C), pressure (825 MPa) and catalyst to
Figure 2. The DS as a function of the basicity (pK
b
) of the
catalyst for the acetylation of starch in pressurised CO
2
.
Scheme 1. Representative scheme of the starch acetylation reaction using AAH as reagent and a basic catalyst. In this
scheme, NaOAc is taken as an example.
Starch/Sta rke 2010, 62, 566576 Process-product studies on starch acetylation reactions 569
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
starch ratio (0.10.5 mol/mol AGU) on the DS and X
AAH
were determined. The other process variables like the AAH
to starch ratio (4.35 mole/mole AGU), starch water content
(16.2%w/w), reaction time (1 h) and type of catalyst (K
2
CO
3
)
were kept constant for all experiments. An overview of the
experimental conditions and the results is given in Table 1.
Table 1. Overview of experiments
No T (8C) P (MPa)
K
2
CO
3
: AGU
(mol/mol) K
2
CO
3
(g) DS X
AAH
(%)
Critical
region (MPa)
a)
1 50 8 0.1 0.26 0.06 19 9.49.8
2 50 10 0.1 0.26 0.08 16
3 50 15 0.1 0.26 0.06 9
4 50 25 0.1 0.26 0.06 14
5 50 8 0.5 1.28 0.13 44
6 50 10 0.5 1.28 0.23 42
7 50 15 0.5 1.28 0.05 31
8 50 25 0.5 1.28 0.04 17
9 70 9 0.3 0.77 0.16 51 12.412.6
10 70 12.5 0.3 0.77 0.19 52
11 70 20 0.3 0.77 0.19 47
12 70 9 0.3 0.77 0.14 53
13 70 12.5 0.3 0.77 0.2 56
14 70 20 0.3 0.77 0.18 48
15 90 8 0.1 0.26 0.23 39 14.514.8
16 90 10 0.1 0.26 0.29 43
17 90 15 0.1 0.26 0.39 40
18 90 25 0.1 0.26 0.28 39
19 90 8 0.5 1.28 0.24 n.d.
b)
20 90 10 0.5 1.28 0.42 n.d.
b)
21 90 15 0.5 1.28 0.49 n.d.
b)
22 90 25 0.5 1.28 0.47 n.d.
b)
a) taken from (Muljana et al., submitted for publication).
b) n.d.: not determined.
Figure 3. The product morphology at
various DS values (DS of 0.23 (a), DS
of 0.29 (b), DS of 0.36 (c) and DS of
0.62 (d)).
570 H. Muljana et al. Starch/Sta rke 2010, 62, 566576
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
The products were obtained as white powders. Informa-
tion on particle morphology was obtained by SEM. Pictures
of the acetylated products (Fig. 3) show clear difference in
appearance as a function of the DS. At intermediate DS
values (DS 0.230.36) granular fusion occurs, at higher
DS values (0.62) the granular structure is lost (Fig. 3d).
The output variables (DS, X
AAH
and S
SA
) were modelled
using non-linear multivariable regression (Eq. 4) based on
the three independent variables (temperature, pressure
and catalyst ratio). Good results were obtained for the
DS and X
AAH
whereas it proved not possible to model
the S
SA
values properly for the given dataset. The results
Table 2. Regression coefficients of the DS and X
AAH
models
Variables DS X
AAH
Intercept b
o
0.586 0.011
Temperature (x
1
) b
1
0.021 2.013 10
3
Pressure (x
2
) b
2
7.979 10
4
2.833 10
4
K
2
CO
3
ratio (x
3
) b
3
0.079 0.305
x
2
1
b
11
1.658 10
4
6.693 10
6
x
2
2
b
22
6.322 10
6
n.s.
x
1
x
2
b
12
1.792 10
5
0.013
x
1
x
3
b
13
2.517 10
3
n.s.
x
2
x
3
b
23
n.s.
a)
1.23 10
3
a) n.s.: non-signicant terms according to the t-test evaluation.
Figure 4. Parity plot for the experimental and predicted values of the DS(a) and X
AAH
(b).
Table 3. Analysis of variance for the DS and X
AAH
models
DS
SS DF MS F p-value R
2
values
Model 0.36 7 0.051 25.00 7.027 10
7
R
2
0.93
Error 0.029 14 2.06 10
3
R
2
adjusted
0.89
Total 0.39 21 R
2
PRESS
0.78
X
AAH
Model 0.38 6 0.064 73.91 3.53 10
9
R
2
0.98
Error 8.82 10
3
12 7.35 10
4
R
2
adjusted
0.97
Total 0.39 18 R
2
PRESS
0.85
Starch/Sta rke 2010, 62, 566576 Process-product studies on starch acetylation reactions 571
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
for the regression coefcients for the DS and X
AAH
models
are given in Table 2.
The R
2
and R
2
adjusted
values for both models were very
satisfactorily (DS: R
2
0.93, R
2
adjusted
0.89, X
AAH
:
R
2
0.97, R
2
adjusted
0.96), thus demonstrating that the
models are a good representation of the experimental
data. This was conrmed by parity plots for both models
(Fig. 4). In addition, the relatively high R
2
PRESS
values
(0.780.87) indicate a good predictive capability of both
models [17].
The results of the analysis of variance (ANOVA) are
given in Table 3. The very low p-values (7.027 10
7
for
DS and 1.675 10
8
for X
AAH
) indicate that both models
are statistically signicant.
The predicted effects of the process variables on the
DS and X
AAH
are illustrated in Fig. 5. The temperature has
by far the largest effect on the product DS. The DS
increases with temperature for all pressures and catalyst
intakes. The highest DS value (0.46) was obtained at a
temperature of 908C, a pressure of 15 MPa and a catalyst
to starch ratio of 0.5 mol K
2
CO
3
/mol AGU.
The effect of the pressure on the DS is considerably
lower than the temperature, though still relevant. For
all temperatures and catalyst intakes, the DS shows a
clear maximum at a pressure of about 15 MPa. This
pressure is the critical point of the mixture (Table 1)
[Muljana et al., submitted for publication; 23], a clear
indication that the highest overall rates for the starch acety-
lation reaction are at or near the critical point of the mixture.
Similar observations are known for a range of reactions
carried out in supercritical CO
2
. For the systemunder study,
the optimum pressure of 15 MPa is likely the results of two
opposing effects, improved solubility of CO
2
and associ-
ated higher mass transfer rates of AAHin the starch matrix
when increasing the pressure to the critical point and a
compressive effect at higher pressures leading to a
reduction in the free starch volume and lower mass transfer
rates [Muljana et al., submitted; 13, 14, 2427].
As expected on the basis of the chemistry, a higher
catalyst intake leads to higher reaction rates and associ-
ated DS values. The shape of the curves is not affected by
the catalyst intake (Fig. 5a).
When the starch acetylation reaction is the sole reaction
occurring in the system, the shape and qualitative trends
for the DS and the X
AAH
curves are expected to be similar.
This is clearly not the case (Fig. 5), and this is due to the
occurrence of the competing AAH hydrolysis reaction with
its own kinetic features ( Scheme 1). In an earlier
Figure 5. 3D visualisation of DS (a) and X
AAH
(b) as function of reaction temperature, pressure and catalyst to starch ratio.
Figure 6. DSC measurements on various products pre-
pared in scCO
2
(DS 0.390.96) and in water (AVDS 0.39),
(a.u., arbitrary unit).
572 H. Muljana et al. Starch/Sta rke 2010, 62, 566576
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
contribution, we have already shown that the hydrolysis
reaction is considerably faster than the starch acetylation
reaction and dominates the chemistry when water is
present (Muljana et al., submitted).
With the models available, it is possible to tune the
reaction conditions in such a way to obtain the product
with the desired DS for the foreseen application. The
temperature is the main parameter though it is also clearly
possible to tune the DS with the pressure. This is a remark-
able additional feature for DS optimisation, which is not
possible when performing the reaction in the classical
aqueous systems. Within our experimental range of con-
ditions, the highest starch acetylation rates and thus
highest DS are achieved at 908C, 15 MPa and
0.5 mol K
2
CO
3
/mol AGU resulting in a DS of 0.46.
3.3 Product properties
The thermal properties of the starch acetates produced in
pressurised CO
2
with various DS values (0.290.96) were
investigated. The acetylated product with the highest DS
(0.96) in the range was not obtained in the set of exper-
iments described above but prepared at a longer reaction
time (24 h) at a temperature of 908C, a pressure of 15 MPa
and a K
2
CO
3
to starch ratio of 0.1 mol/mol AGU. The
properties of the products are compared with starch
acetates produced in water (DS 0.29 and 0.39, supplied
by AVEBE).
The DSC results are shown in Fig. 6. The glass tran-
sition temperature (T
g
) of the acetylated products obtained
in pressurised CO
2
are present between 166 and 1758C.
Figure 7. TGA measurements on acetylated products in scCO
2
(DS 0.290.96) (a) and acetylated products in water
(AVDS 0.290.39) (b) and their derivatives (c), (a.u., arbitrary unit).
Starch/Sta rke 2010, 62, 566576 Process-product studies on starch acetylation reactions 573
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
The T
g
values are reduced when the DS increases, in
agreement with available literature data [28, 29]. The
introduction of acetyl groups in the starch granules
reduces the strength of the intermolecular bonds and
increases the chain mobility, leading to lower T
g
value
[29]. Remarkably, the DSC spectrum for the product pro-
duced in water (DS 0.39) did not show a T
g
value in the
temperature range of the DSC analysis. A possible expla-
nation is the difference in the temperature for the prep-
aration of the samples (258C for water versus 908C for
pressurised CO
2
), which will likely have an effect on the
extent of gelatinisation and thus on the thermal history of
the sample.
Further insights in the thermal stability of the products
were obtained from TGA measurements (Fig. 7). The
acetylated products in pressurised CO
2
(DS range of
0.290.96) show a decrease in the initial degrada-
tion temperature (2582608C) compared to native
potato starch (2708C). This effect is probably due to a
reduction of molecular interactions such as hydrogen
bonds in the starch chain when introducing acetyl groups
[30, 31].
Although the initial degradation temperature of the
products is reduced, the acetylated products prepared in
pressurised CO
2
show an increase in thermal stability
compared to native potato starch as indicated by a broad-
ening shoulder in the thermograms (Fig. 7a). From the
corresponding rst derivatives (Fig. 7c), it is clear that this
is related to the appearance of a second degradation step
at a considerably higher temperature than native starch.
As reported in the literature, the shoulder in the starch
degradation peaks is likely related to different degradation
rates of the amylose and amylopectin regions within the
starch particles [32].
The starch acetates produced in water showed a higher
thermal stability compared to native starch (Fig. 7b) [33].
Remarkable differences in thermal stability were observed
for the products prepared in water and in pressurised CO
2
.
The samples made in water only showed a single peak
whereas two peaks were observed for pressurised CO
2
,
particularly when the DS is higher than 0.29. The peak
maximum for the samples made in water is intermediate
between the two peaks observed in the samples made in
pressurised CO
2
. The exact reasons for these differences
in TGA results for the samples prepared in water and CO
2
,
even at similar DSvalues, is not yet clear and needs further
investigations.
The viscosity proles of two samples made in
pressurised CO
2
and two in water with a similar DS
(0.29 and 0.39) were determined using an RVA analysis
and the results are shown in Fig. 8. In all cases, the
acetylated products show a signicant decrease in the
gelatinisation temperature and the viscosity of the pastes
as compared with native potato starch (Fig. 8) [3, 34, 35].
This is likely due to the introduction of the acetyl
group leading to a disruption of H-bonds in the native
starch (reducing the interactions between the individual
macromolecules) and an increase in the starch hydro-
phobicity [3, 34, 35].
Moreover, a clear difference in the pasting behaviour of
the samples synthesised in pressurised CO
2
(DS 0.29
0.39) with respect to the products synthesised in water
(AVDS 0.290.39) was observed (Fig. 8). All products
made in pressurised CO
2
have lower viscosities than those
prepared in water. This is indicative for a higher extent of
disruption of inter- and the intra-molecular hydrogen bonds
between the starch chain for the products synthesised in
pressurised CO
2
[14, 26].
Figure 8. Viscosity profile of the acetylated products synthesized in pressurised CO
2
(DS0.29) and in water (AVDS0.29) (a)
and of the acetylated products with higher DS (DSAVDS 0.39) (b).
574 H. Muljana et al. Starch/Sta rke 2010, 62, 566576
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
4 Conclusions
We performed an in-depth experimental study to evaluate
the effect of process conditions and particularly the pres-
sure on the acetylation of starch in pressurised CO
2
using
a basic catalyst and AAHas the reagent. The experimental
data of 22 experiments were modelled using a statistical
approach to quantify the effects of process conditions on
important reaction parameters like the DS and X
AAH
. The
models predict a positive inuence of temperature and
catalyst ratio on the product DS, whereas an optimum is
observed for the pressure. The highest DS value (0.49)
after 1 h for the acetylation reaction in scCO
2
within our
experimental conditions was obtained at a temperature of
908C, a pressure of 15 MPa, and at 0.5 mol K
2
CO
3
/mol
AGU. A further increase to a DS of 0.96 was possible by
performing the reaction at 24 h instead of 1 h reaction
time.
Relevant product properties (viscosity, thermal stability)
were determined and compared with samples prepared in
water. Remarkable differences in thermal stability and
viscosity proles were observed between those made in
water and pressurised CO
2
. The potential of the products
prepared in pressurised CO
2
and possible advantages
compared to products made in water needs to be estab-
lished in follow up application research.
The authors thank G. O. R Alberda van Ekenstein
(Polymer Department, University of Groningen) for his help
with the DSC and TGA measurements and Harry Nijland
(Polymer Department, University of Groningen) for the
SEM analyses. We also would like to thank AVEBE for
providing the materials (potato starch, potato starch
acetates) and Michael Polhuis (AVEBE) for RVA measure-
ments and stimulating discussions. Henky Muljana
acknowledges the University of Groningen for providing
an Ubbo Emmius scholarship.
The authors have declared no conict of interest.
5 References
[1] Huang, J. R., Schols, H. A., Klaver, R., Jin, Z. Y., et al., Acetyl
substitution patterns of amylose and amylopectin popu-
lations in cowpea starch modied with acetic anhydride
and vinyl acetate. Carbohydr. Polym. 2007, 67, 542550.
[2] Singh, N., Chawla, D., Singh, J., Inuence of acetic anhydride
on physicochemical, morphological and thermal properties
of corn and potato starch. Food Chem. 2004, 86, 601
608.
[3] Jarowenko, W., in: Wurzburg, O. B. (Ed.), Acetylated Starch
And Miscellaneous Organic Esters, CRC Press, Inc., Boca
Raton, Florida 1986, pp. 5577.
[4] Dicke, R. E., A straight way to regioselectively functionalised
polysaccharide esters. Cellulose 2004, 11, 255263.
[5] Shogren, R. L., Rapid preparation of starch esters by high
temperature/pressure reaction. Carbohydr. Polym. 2003, 52,
319326.
[6] U.S. Patent 5977348, Harris, R., Jureller, S. H., Kerschner, J.
L., Trzasko, P. T. et al., 1999.
[7] Yalpani, M., Supercritical uids puissant media for the
modication of polymers and biopolymers. Polymer 1993,
34, 11021105.
[8] Nalawade, S. P., Picchioni, F., Janssen, L., Supercritical car-
bon dioxide as a green solvent for processing polymer melts:
Processing aspects and applications. Prog. Polym. Sci.
2006, 31, 1943.
[9] Eckert, C. A., Knutson, B. L., Debenedetti, P. G., Supercritical
uids as solvents for chemical and materials processing.
Nature 1996, 383, 313318.
[10] Beckman, E. J., Supercritical and near-critical CO2 in green
chemical synthesis and processing. J. Supercrit. Fluids
2004, 28, 121191.
[11] Savage, P. E., Gopalan, S., Mizan, T. I., Martino, C. J., et al.,
Reactions at supercritical conditions applications and fun-
damentals. AIChE J. 1995, 41, 17231778.
[12] Muljana, H., Picchioni, F., Heeres, H. J., Janssen, L. P. B. M.,
International Symposium on Chemical Reaction Engineering
20, Kyoto, Japan, 710 September 2008, pp 2628.
[13] Muljana, H., Picchioni, F., Heeres, H. J., Janssen, L. P. B. M.,
Experimental and modelling studies on the solubility of sub-
and supercritical carbon Dioxide (scCO2) in potato starch
and derivatives. Polym. Eng. Sci. (in press), DOI: 10.1002/
pen.21786.
[14] Muljana, H., Picchioni, F., Heeres, H. J., Janssen, L. P. B. M.,
Supercritical carbon dioxide (scCO2) induced gelatinisation
of potato starch. Carbohydr. Polym. 2009, 78, 511519.
[15] Junistia, L., Sugih, A. K., Manurung, R., Picchioni, F., et al.,
Synthesis of higher fatty acid starch esters using vinyl laurate
and stearate as reactants. Starch/Sta rke 2008, 60, 667675.
[16] Elomaa, M., Asplund, T., Soininen, P., Laatikainen, R., et al.,
Determination of the degree of substitution of acetylated
starch by hydrolysis,
1
H NMR and TGA/IR. Carbohydr.
Polym. 2004, 57, 261267.
[17] Montgomery, D. C. (Ed.), Design and Analysis of
Experiments. 5, John Wiley and Sons, Inc, New York 2001.
[18] Lide, D. R. (Ed.), Handbook of Chemistry and Physics, CRC
Press, New York 1998.
[19] deGraaf, R. A., Broekroelofs, G. A., Janssen, L., Beenackers,
A., The kinetics of the acetylation of gelatinised potato starch.
Carbohydr. Polym. 1995, 28, 137144.
[20] Junistia, L., Sugih, A. K., Manurung, R., Picchioni, F., et al.,
Experimental and modelling studies on the synthesis and
properties of higher fatty esters of corn starch. Starch/Sta rke
2009, 61, 6980.
[21] Tijsen, C. J., Kolk, H. J., Stamhuis, E. J., Beenackers, A., An
experimental study on the carboxymethylation of granular
potato starch in non-aqueous media. Carbohydr. Polym.
2001, 45, 219226.
[22] Villwock, V. K., BeMiller, J. N., Effects of salts on the reaction
of normal corn starch with propylene oxide. Starch/Sta rke
2005, 57, 281290.
[23] Calvo, L., de Loos, T. W., High pressure vapour-liquid
equilibria of the binary and some of the ternary and
multicomponent mixtures of the carbon dioxideacetic
anhydridealpha-methylbenzyl alcoholacetic acidalpha-
methylbenzyl acetate system experimental and modelling
results. Fluid Phase Equilib. 2006, 244, 179187.
Starch/Sta rke 2010, 62, 566576 Process-product studies on starch acetylation reactions 575
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
[24] Hoshino, T., Nakamura, K., Suzuki, Y., Adsorption of carbon-
dioxide to polysaccharides in the supercritical region. Biosci.
Biotechnol. Biochem. 1993, 57, 16701673.
[25] Chen, K. H. J., Rizvi, S. S. H., Measurement and prediction of
solubilities and diffusion coefcients of carbon dioxide in
starchwater mixtures at elevated pressures. J. Polym.
Sci., Part B: Polym. Phys. 2006, 44, 607621.
[26] Francisco, J. D., Sivik, B., Gelatinisation of cassava, potato
and wheat starches in supercritical carbon dioxide. J.
Supercrit. Fluids 2002, 22, 247254.
[27] Briscoe, B. J., Kelly, C. T., The plasticisation of a polyurethane
by carbon-dioxide at high pneumatic stresses. Polymer 1995,
36, 30993102.
[28] Shogren, R. L., Preparation, thermal properties, and extru-
sion of high-amylose starch acetates. Carbohydr. Polym.
1996, 29, 5762.
[29] Wang, X., Gao, W. Y., Zhang, L. M., Xiao, P. G., et al., Study on
the morphology, crystalline structure and thermal properties
of yam starch acetates with different degrees of substitution.
Sci. China, Ser. B-Chem. 2008, 51, 859865.
[30] Aburto, J., Hamaili, H., Mouysset-Baziard, G., Senocq, F., et
al., Free-solvent synthesis and properties of higher fatty
esters of starch Part 2. Starch/Sta rke 1999, 51, 302
307.
[31] Ren, J. L., Xu, F., Sun, R. C., Peng, B., et al., Studies of the
lauroylation of wheat straw hemicelluloses under heating. J.
Agric. Food Chem. 2008, 56, 12511258.
[32] Mano, J. F., Koniarova, D., Reis, R. L., Thermal properties of
thermoplastic starch/synthetic polymer blends with potential
biomedical applicability. J. Mater. Sci. Mater. Med. 2003, 14,
127135.
[33] Xu, Y. X., Miladinov, V., Hanna, M. A., Synthesis and charac-
terisation of starch acetates with high substitution. Cereal
Chem. 2004, 81, 735740.
[34] Saartrat, S., Puttanlek, C., Rungsardthong, V., Uttapap, D.,
Paste and gel properties of low-substituted acetylated canna
starches. Carbohydr. Polym. 2005, 61, 211221.
[35] Shon, K. J., Yoo, B., Effect of acetylation on rheological
properties of rice starch. Starch/Sta rke 2006, 58, 177
185.
576 H. Muljana et al. Starch/Sta rke 2010, 62, 566576
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com

You might also like