Optimization of Chitin Extraction from Shrimp Shells

Biomacromolecules 2003, 4, 12-18

12

Articles
Optimization of Chitin Extraction from Shrimp Shells
Aline Percot, Christophe Viton, and Alain Domard*
Laboratoire des Mate´ riaux Polyme` res et des Biomate´ riaux, UMR-CNRS 5627, Baˆ timent ISTIL,
Domaine Scientifique de la Doua, 15 Bd. Andre´ Latarjet, 69622 Villeurbanne Cedex, France
Received June 25, 2002; Revised Manuscript Received October 4, 2002

The aim of this paper is to define optimal conditions for the extraction of chitin from shrimp shells. The
kinetics of both demineralization and deproteinization with, in the latter case, the role of temperature are
studied. The characterization of the residual calcium and protein contents, the molecular weights, and degrees
of acetylation (DA) allows us to propose the optimal conditions as follows. The demineralization is completely
achieved within 15 min at ambient temperature in an excess of HCl 0.25 M (with a solid-to-liquid ratio of
about 1/40 (w/v)). The deproteinization is conveniently obtained in NaOH 1 M within 24 h at a temperature
close to 70 °C with no incidence on the molecular weight or the DA. In these conditions, the residual
content of calcium in chitin is below 0.01%, and the DA is almost 95%.
Introduction
Sources of chitin are estimated to be as abundant as those
for cellulose with a yearly production of approximately 10101012 T.1 Chitin is a polysaccharide corresponding to linear
copolymers of β(1f4)-linked 2-amino-2-deoxy-D-glucan and
2-acetamido-2-deoxy-D-glucan. Chitin, especially its main
derivative chitosan, has numerous applications, for example,
in agriculture, biomedicine, paper making, and food industries.2 Some applications require specific architectures, and
the effectiveness of the polymers for these applications was
shown to depend on the molecular weight distribution and
the degree of acetylation (DA).3-5 A cost-effective, fast, and
easily controlled industrial process for producing chitins of
high molecular weight and DA still remains to be developed.

The main sources of raw material for the production of
chitin are cuticles of various crustaceans, principally crabs
and shrimps. However, chitin in biomass is closely associated
with proteins, minerals, lipids, and pigments. They all have
to be quantitatively removed to achieve the high purity
necessary for biological applications. Although many methods can be found in the literature for the removal of proteins
and minerals, detrimental effects on the molecular weight
and DA cannot be avoided with any of these extraction
processes.6 Therefore, a great interest still exists for the
optimization of the extraction to minimize the degradation
of chitin, while, at the same time, bringing the impurity levels
down to a satisfactory level for specific applications.
Demineralization is generally performed by acids including
HCl, HNO3, H2SO3, CH3COOH, and HCOOH but HCl
seems to be the preferred reagent and is used with a
concentration between 0.275 and 2 M for 1-48 h at
* To whom correspondence should be addressed.

temperatures varying from 0 to 100 °C.1 Madhavan and
Ramachandran Nair7 have shown that the viscosity of the
chitosan obtained decreases with the treatment time in HCl,
suggesting a decrease of the molecular weight with time.
Deproteinization of chitin is usually performed by alkaline
treatments, although other effective reagents have been
reported. Typically, raw chitin is treated with approximately
1 M aqueous solutions of NaOH for 1-72 h at temperatures
ranging from 65 to 100 °C.1 An interesting alternative method
involves the enzymatic degradation of proteins. However,
the residual protein satisfactory level, ranging approximately
from 1% to 7%, remains higher and the reaction time is

longer compared to that of the chemical way. These
drawbacks make the enzymatic method8 unlikely to be
applied industrially before progress is made in making the
process more efficient.
In this first paper on the study of the production of chitin,
we propose an optimized method of extraction for the
production of biological-grade chitin with both high molecular weights and DA. The kinetics of demineralization and
deproteinization using two classical methods have been
studied to achieve these objectives. The purity level of chitin
was followed by the evaluation of the calcium and protein
contents as a function of the reaction time. In parallel, the
influence of both treatments on the preservation of the chitin
structure was studied from the control of the molecular
weight and DA after each step.
Materials and Methods
Raw Materials and Preparation. Shells of marine shrimp
Parapenaeopsis stylifera were provided by France-Chitine.
The tiny brown shrimp, which is common all over the Indian
Ocean, originates in our case from the port of Jakham (India)

10.1021/bm025602k CCC: $25.00
© 2003 American Chemical Society
Published on Web 11/05/2002


Biomacromolecules, Vol. 4, No. 1, 2003 13

Optimization of Chitin Extraction

located on the Arabian Sea. The shrimps were kept on ice

for 2 or 3 days before being peeled; the shells were scraped
free of loose tissue and washed individually in lightly salted
water. The following procedure was chosen by the producer
as an optimal treatment for a long time preservation of the
raw material. The shells were then separated from cephalothoraxes, salted (10 kg of NaCl per 500 g of shell), and
dried in the sun (25-30 °C) for 3 days. Prior to use, the
shrimp shells were washed thoroughly in distilled water until
the conductivity reached that of water. The shells were then
freeze-dried and cryo-ground under liquid nitrogen. The
powder thus obtained was sieved, and the fraction below 80
µm was used hereafter.
Characterization of Shrimp Shells and Intermediary
Products. The water content in the obtained powders was
estimated by thermogravimetric analysis using a Du Pont
Instrument TGA 2000 with 10-20 mg samples and a
temperature ramp of 2 °C/min.
The percentage of proteins remaining in chitin was
determined by two different methods. First of all, the nitrogen
content was measured by elemental analysis and the percentage of proteins calculated from the following equation
P% ) (N% - 6.9) × 6.25

(1)

where P% represents the percentage of proteins remaining
in the obtained powder and N% represents the percentage of
nitrogen measured by elemental analysis with 6.9 corresponding to the theoretical percentage of nitrogen in fully
acetylated chitin and 6.25 corresponding to the theoretical
percentage of nitrogen in proteins. In the case of the crude
shrimp shells, no accurate values could be obtained because
of the great amount of calcium carbonate present. For this

special case, the amount of proteins was measured using the
amino acid analysis. Two milligrams of sample was hydrolyzed with HCl 6 M in the presence of trifluoroacetic acid
for 45 min at 150 °C under vacuum. The samples were then
solubilized in a buffer, and an aliquot was used for analysis
on a Beckman system 6300 amino acid analyzer. The
percentage of proteins was calculated from the total amino
acid weight.
The lipid content was estimated after a Soxlhet extraction
with chloroform/methanol (2/1, v/v) and subsequent gravimetric analysis of the shrimp shells.
The ash content was determined by slowly heating a
sample to 900 °C with stages at 120 and 340 °C and
weighing the remaining product after cooling in a desiccator.
For the minerals, in addition to ashes analysis, levels of
calcium and magnesium were analyzed by inductively
coupled plasma atomic emission spectroscopy (ICP-AES)
and ICP mass spectrometry (ICP-MS). Prior to analysis, the
solid samples were digested in concentrated sulfuric acid in
a microwave reactor until complete dissolution had occurred.
Determination of the Degree of N-Acetylation (DA) of
Chitin. Chitin samples were dissolved in DCl/D2O (20%
w/w) with vigorous stirring for 8 h at 50 °C. These conditions
were necessary to sufficiently depolymerize chitin, thus
allowing the full solubilization of the polymer.
The spectra were recorded on a Bruker AC 200 spectrometer (200 MHz for 1H) at 298 K. The DA was calculated, as

proposed by Hirai et al.,9 from the ratio of the methyl proton
signal of (1f4)-2-acetamido-2-deoxy-β-D-glucan residues
with reference to the H-2 to H-6 proton signals of the whole
structure.
For samples containing proteins, solid-state 13C NMR

spectroscopy was used. Indeed, in this case, the too high
amount of proteins makes too difficult the interpretation of
the proton NMR spectra. The spectra were obtained on
lyophilized samples with CP-MAS techniques (cross polarization, magic angle spinning) using a Bruker DSX400
instrument working at 100.6 MHz. Typical conditions were
as follows: 90 RF pulse, 4.5 µs; contact time, 2.5 ms; pulse
repetition, 2s; MAS rate, 5 kHz; 4096 scans were acquired.
The measurements were performed at room temperature. The
DA was calculated by comparison between the integrated
areas of the methyl group carbon (δ 24 ppm) and the C2C6 signals (δ 56-105 ppm).9
Determination of the Intrinsic Viscosity of Chitin.
Chitin samples were solubilized at about 0.25 mg/mL in N,Ndimethylacetamide (DMAc) containing 5% lithium chloride
(LiCl).10,11 The viscosity was measured using an automatic
capillary viscometer, Viscologic TI 1 SEMATech (diameter
0.8 mm), at 25 °C.
Kinetics of Demineralization. Demineralization was
carried out in dilute HCl solutions. Typically, the shrimp
shells were soaked in HCl 0.25 M at ambient temperature
with various solution-to-solid ratios under constant stirring.
The demineralization kinetics were followed by monitoring
the pH as a function of time in the supernatant. The
characteristics of the obtained chitin as a function of the
demineralization time were studied by retrieving a representative sample of known volume from the dispersion of
shrimp shell particles at 2, 6, 13, 30, 60, 180, 360, and 1440
min using a syringe with a large needle. The heterogeneous
samples were then filtered under vacuum on paper, and the
supernatant was analyzed for the pH and the calcium content
(by ICP-AES). The recovered demineralized shrimp shell
powder was washed to neutrality and freeze-dried. Following
the demineralization step, the partly demineralized shrimp

shells were deproteinized with a solution of NaOH 1 M under
vigorous stirring using 30 mL of solution per gram of
demineralized shells. After 24 h of reaction, the solid samples
were washed to neutrality and freeze-dried. The calcium
content in the purified chitin was measured by ICP-AES.
Kinetics of Deproteinization. The kinetic studies of
deproteinization were performed on demineralized shrimp
shell powder (Figure 1). Quantitative demineralization was
carried out in HCl 1 M at a room temperature corresponding
to 22 ( 1 °C with a solution-to-solid ratio of 10 mL/g. After
24 h, the demineralized shrimp shell powder was removed
by filtration, washed to neutrality, and freeze-dried. Deproteinization kinetics studies were then performed by addition
of NaOH 1 M to decalcified powder at a solution-to-solid
ratio of 15 mL/g, and a representative sample was taken at
5, 20, 60, 180, 300, and 1440 min. The partly deproteinized
shrimp shell powder was removed in each sample by
filtration, washed to neutrality, and freeze-dried, while the
supernatant was analyzed for the protein content.


14

Biomacromolecules, Vol. 4, No. 1, 2003

Percot et al.

Figure 1. Overall process for the preparation of chitin from salted shrimp shells.

The release of proteins in the supernatant was followed
by the measurement of the absorbance at 280 nm, characteristic of the tryptophan residues present in the protein

composition. The absorbance was measured on an Uvikon
(UV-vis.) spectrophotometer 941 (Kontron Instruments).
Shrimp shell proteins were recovered by a classical process
of deproteinization on demineralized shrimp shells in NaOH
1 M for 24 h with a solution-to-solid ratio of 15 mL/g (Figure
1). After filtration, the supernatant containing the proteins
was dialyzed against pure water for 1 week and lyophilized.
The protein content in the recovered powder was determined
by amino acid analysis and found to be close to 60% (w/w).
These proteins were used to plot a calibration curve giving
a straight line with the following equation
abs ) Cp × 1.7

r2 ) 0.999 73

(2)

where abs corresponds to the UV absorbance measured at
280 nm, Cp corresponds to the protein concentration in mg/
mL, and r corresponds to the correlation coefficient.
The weight of proteins released could be then expressed
as a weight percentage of the initial weight of decalcified
shrimp shells.
The deproteinization was also studied as a function of
temperature. In this case, NaOH 1 M was added to demineralized shrimp shells with a solution-to-solid ratio of
15 mL/g. The deproteinization was carried out for 24 h at
various temperatures. The protein content was measured in
the supernatant and in the obtained chitin.
Results and Discussion
The overall process for the preparation of shrimp chitin

is given in Figure 1. The raw shrimp shells contain about
20% of chitin and other components reported in Table 1.
While CaCO3 is the major inorganic component, some
magnesium is also present in a low proportion. Demineralization (usually performed in concentrated acid) and deproteinization (in aqueous NaOH) are therefore the critical steps

Table 1. Composition of Crude Shrimp Shells, Demineralized
Shrimp Shells (24 h in HCl 1 M), and Chitin (24 h in NaOH 1 M)
composition

shrimp shells
(wt %)

water
crude protein
crude fat
ash (as oxide)
calcium
magnesium

11.3 ( 0.4
7c
6(2
35.49 ( 0.04
19a
1a

demineralized shrimp
shells (wt %)
6.3
20d

chitin
(wt %)
6.9 ( 0.1
<1c-d

0.01b

0.01b

a Measured by ICP-MS. b Measured by ICP-AES. c Measured by amino
acid analysis. d Measured by elemental analysis.

of chitin extraction. It is generally agreed that the processing
conditions strongly affect the molecular weight and DA of
chitin. As a rule, as the acidic conditions for demineralization
(pH, time, and temperature) become harsher, the molecular
weight of the products thus obtained becomes lower. Indeed,
chitin is an acid-sensitive material and can be degraded by
several pathways: hydrolytic depolymerization, deacetylation, and heat degradation leading to physical property
modifications.
Although HCl may be the cause of detrimental effects on
the intrinsic properties of the purified chitin, it remains the
most commonly used decalcifying agent in both laboratory
and industrial scale production of chitin. It is generally used
at a concentration of 1 M. Chang and Tsai12 determined
optimal demineralization conditions for the shell of pink
shrimp (Solenocera melantho) measuring the calcium level
in the obtained product but without taking into account the
chitin characteristics that they obtained. In this paper, a

concentration of HCl of 0.25 M, below all values reported
in the literature was tested to minimize the hydrolysis of the
polymer. The solution-to-solid ratio was kept above 10 mL/g
to obtain a homogeneous mixture with a large excess of
solution. The demineralization occurs when the acid reacts
with the calcium carbonate according to the following simple
equation:
CaCO3 + 2HCl f CO2v + CaCl2 + H2O

(3)


Optimization of Chitin Extraction

Figure 2. Kinetics of demineralization: (A) HCl was initially in default,
and then, 10 mL of HCl 0.25 M was successively added at 120 and
180 min; (B) HCl was initially in excess, and then, 10 mL of HCl 0.25
M was added after 18 h.

Therefore, pH increases till the end of the reaction. Rhazi et
al.13 have proposed the use of acidimetric titration to follow
the demineralization process, and the end of the reaction is
related to the persistence of the acidity in the medium. In
Figure 2A, we can see a fast increase of pH as a function of
the reaction time. In this experiment, 20 mL of HCl 0.25 M
was initially added to 1 g of shrimp shells. After 30 s, 98%
of the added acid had already reacted. After 2 h, the pH of
the solution was neutral, and 10 mL of HCl 0.25 M was
further added to the solution, leading successively to a
decrease of pH and then a rapid increase. The added acid

thus reacted with the remaining calcium carbonate, and after
3 h, the pH of the solution reached a plateau at 5.5. Still
another 10 mL of acid was added to the solution. We may
now consider that the acid added was in excess because the
pH remained acidic and stable at 1.8 even after 4 h. From
these results, we may estimate the amount of acid necessary
to carry out the reaction to completion and we may also
follow the kinetics of the reaction. Then, a similar experiment
was carried out from a unique addition of acid in excess:
40 mL of HCl 0.25 M was added to 1 g of shrimp shell
powder, and from Figure 2B, we can see that the reaction
was complete after only 15 min (pH remained acidic and
constant with time). After 18 h, another addition of 10 mL
of HCl 0.25 M was made to make sure that the reaction was
indeed complete. pH remained acidic at a value of 1.4 in a
similar manner as in Figure 2A after the second addition of
HCl. To verify that the measure of pH could actually be
used to accurately follow the demineralization kinetics, we
performed an experiment in the same conditions while
samples of the mixture were collected periodically with a
large syringe. HCl 0.25 M in excess was then added to a
shrimp shell powder.

Biomacromolecules, Vol. 4, No. 1, 2003 15

Figure 3. Kinetics of demineralization in HCl 0.25 M at ambient
temperature: (A) determination of the pH (9) and the calcium
concentration (0) in the supernatant as a function of time; (B) calcium
content in the demineralized shrimp shell (b) and in the corresponding
chitin (O) as a function of time.

Figure 3A shows that pH measurement increases in the
supernatant with the calcium concentration. Because the latter
was very high in the supernatant, the samples had to be
diluted before ICP-AES analysis. This might have caused
additional experimental errors on the results. These results
were used to calculate the number of molecules of H3O+
(nH3O+) having reacted with the calcium carbonate in relation
with the number of calcium ions (nCa) released from the
shrimp shell powder. The following ratio was obtained:
nH3O+/nCa ) 1.9 ( 0.1

(4)

As expected, the experimental result is about 2 and corroborates eq 3. The difference between the theoretical and
obtained value results from the uncertainty of the pH
measurement and of the calcium concentration. We also
determined the amount of calcium remaining in the decalcified shrimp shell as a function of time. For the mineralization
of the sample, several methods were tested: solubilization
of dry ashes in acid, mineralization of the sample in acid
with heating at 900 °C, and digestion of the sample in sulfuric
acid in a microwave reactor. With the first two methods,
there was always an insoluble fraction and the obtained
results were not reproducible. Only the digestion in a
microwave oven gave accurate results. Both the decalcified
samples and the obtained chitin were mineralized in this
reactor. In Figure 3B are reported the kinetics of demineralization in relation with the calcium content in the material.
Two minutes after the beginning of the process, the calcium
content in the samples is about 168 µg/g of powder, while
the lowest amount possible to be reached with the process

is near 108 ( 11 µg/g after 24 h of treatment. All of the
samples obtained were deproteinized in NaOH 1 M for 24 h
at ambient temperature. The calcium content in the chitin
obtained was measured (Figure 3B). The nondemineralized
chitin sample (t ) 0) contains 0.28 g of calcium per gram


16

Biomacromolecules, Vol. 4, No. 1, 2003

Percot et al.

Table 2. Characterization of the Extracted Chitin upon
Demineralization (HCl 0.25 M) at Ambient Temperaturea
demineralization
time (min)

[η] (mL/g)

DA (%)b

2
6
13
30
60
180
360
1440

2700 ( 200
3300 ( 300
4000 ( 200
4100 ( 200
4400 ( 200
4200 ( 200
3700 ( 200
3300 ( 200

97 ( 2
98 ( 2
95 ( 2
98 ( 2
97 ( 2
99 ( 2
94 ( 2
98 ( 2

a

The characterization was performed after a deproteinization proceeded for 24 h in NaOH 1 M under vigorous stirring using 30 mL of
solution per gram of demineralized shells. The intrinsic viscosity [η] and
DA are given as a function of time. b Measured by liquid NMR.

of deproteinized shrimp shells. If we consider the percentage
of impurities (proteins and lipids) eliminated during the
deproteinization step, we can estimate that nCa eliminated
during the demineralization step corresponds to nCa recovered
in the supernatant more or less 10%. In Figure 3B, we can

observe a slight decrease of the calcium content with the
demineralization time, and the calcium content in the
obtained chitin is about 75 µg/g after complete demineralization.
All of these results show that the demineralization times
reported in the literature are too long. Thus, with an excess
of HCl 0.25 M (with a solid-to-solvent ratio of about 1/40
(w/v) corresponding to 10 times more H3O+ than necessary),
the reaction of demineralization is mostly complete within
15 min. Then, the excess time will only contribute to the
degradation of chitin.
Table 2 shows the variation of the intrinsic viscosity of
chitin as a function of demineralization time. A rapid increase
of [η] was observed with time. This increase parallels the
demineralization kinetics as can be seen in Figure 3B. The
samples recovered before the end of the extraction process
(2 and 6 min) are characterized by a substantial amount of
minerals remaining in the chitin, insoluble in the solvent used
for viscometric experiments. As a consequence, this added
weight causes a significant error on the true chitin concentration thus reducing the apparent viscosity of the mixture
compared to a pure sample of a similar chitin. After 13 min,
the calcium ratio in chitin becomes negligible, and then, [η]
remains stable for 3 h, followed by a decrease in the
viscosity. In the latter case, the acid hydrolysis of chitin is
the only mechanism occurring in the media. Table 2 also
shows the variation of DA as a function of time. The values
obtained were above 95% for all the samples, and no
significant decrease was observed with time. Then, we may
consider that in HCl 0.25 M the treatment has no particular
effect on the DA of chitin. As a consequence, we may
conclude that in native shrimp shells the DA should be close

to 96% ( 3%. This value is approximately 6% over that of
squid pens (to be published).
Deproteinization. Deproteinization by alkaline treatment
was shown to be much less damaging to the chitin structure
compared to the acidic treatment involved in the demineralization.1 For this reason, we decided to use the conventional
procedure for the deproteinization of our samples. Thus, it

Figure 4. Kinetics of deproteinization in NaOH 1 M at ambient
temperature. Variation of the protein concentration in the supernatant,
expressed in mg/mL (O), and remaining in the obtained chitin
(measured by elemental analysis), expressed in protein % (9) is
shown. The measured protein concentration in the supernatant can
also be used to calculate the percentage of proteins remaining in the
obtained chitin (b).

was carried out in NaOH 1 M using a solution-to-solid ratio
of 15 mL/g. The amount of proteins extracted from the
demineralized shrimp shell powder was followed as a
function of time from the variation of the absorbance at 280
nm, characteristic of tryptophan. This method differs from
the conventional method that involves the determination of
proteins remaining in chitin (using the nitrogen content) or
the determination of the protein content in the supernatant
by a colorimetric method such as the Lowry assay.14 Hunt
and Nixon15 also examined the ultraviolet absorption spectra
of the alkaline supernatants for tryptophan. To plot a more
reliable calibration curve, we extracted proteins or peptides
from demineralized shrimp shells and used these proteins to
plot a calibration curve. As expected, the absorbance at 280
nm increases linearly with the amount of proteins and the

equation of the curve (see Experimental Section) can be used
to calculate the concentration of proteins in the supernatant.
NaOH 1 M was added to a demineralized shrimp shell
powder (demineralization proceeded in HCl 1 M at room
temperature with a solution-to-solid ratio of 10 mL/g) with
a solution-to-solid ratio of 15 mL/g at ambient temperature.
With a large syringe, we took samples from the mixture,
filtered the solution, and measured the absorbance in the
supernatant. Figure 4 shows the increase of the amount of
proteins released in the supernatant at ambient temperature
as a function of the deproteinization time. At the same time,
we can observe a decrease of the percentage of proteins in
the recovered chitin. To compare the results obtained by both
methods, we transformed the protein concentration in the
supernatant into a protein weight and then into the percentage
of remaining proteins assuming that after 24 h the whole of
the proteins was extracted (Figure 4). These calculated
percentages are similar to those obtained by elemental
analysis. Although both methods gave the same results,
taking into account the experimental errors, the absorbance
method was a much simpler method to follow the deproteinization process as a function of time. The reaction is
complete (percentage of proteins below 2%) after 6 h.
We also characterized the chitins obtained to evaluate the
effect of deproteinization on the molecular weight of chitin
and its DA. In Table 3, we notice the rapid increase of the
intrinsic viscosity of chitin with time. The phenomenon is


Biomacromolecules, Vol. 4, No. 1, 2003 17

Optimization of Chitin Extraction
Table 3. Characterization of the Extracted Chitin upon
Deproteinization in NaOH 1 M at Room Temperaturea
deproteinization
time (min)
0
5
20
60
180
300
1440

[η] (mL/g)
2100 ( 300
2400 ( 300
2800 ( 200
3100 ( 200
3100 ( 200
3100 ( 200

Table 4. Variation of the Intrinsic Viscosity [η] and DA of the
Extracted Chitins upon Deproteinization (NaOH 1 M) for 24 h at
Various Temperatures

DA (%)b
113 ( 2
104 ( 2
116 ( 2
97 ( 2

95 ( 2
98 ( 2

a This step was performed after a demineralization proceeded for 24 h
in HCl 1 M at room temperature with a solution-to-solid ratio of 10 mL/g.
The intrinsic viscosity [η] and DA are given as a function of time.
b Measured by solid NMR.

Figure 5. Deproteinization in NaOH 1 M for 24 h at various
temperatures. Measurement of the protein concentration in the
supernatant at the end of the reaction (O) and determination of the
amount of proteins remaining in the obtained chitin (measured by
elemental analysis) (9) as a function of the deproteinization temperature is shown.

similar to that for the demineralization step. Thus, until 3 h,
chitin is polluted by a nonnegligible amount of partially
hydrolyzed proteins remaining insoluble in water. This
phenomenon affects the viscosity by reducing the amount
of chitin really present in solution and certainly also by a
lower contribution to the viscosity of these proteins. Beyond
3 h, the protein content becomes negligible and the intrinsic
viscosity is then stable, even after 24 h of treatment. Because
NaOH is a deacetylating agent, the DA of chitin was
measured as a function of time by solid-state NMR. Because
of the presence of proteins, the integrations of the various
carbons were overestimated and the DAs measured for times
below 1 h were erroneous. For high deproteinization times
and fully deproteinized samples, no significant difference
could be observed among all of the samples (taking into
account the experimental errors).

The role of temperature on the deproteinization process
was estimated from reactions performed for 24 h at various
temperatures. In Figure 5, the protein concentration in the
supernatant is plotted as a function of temperature. We
observe an increase of the amount of proteins extracted with
temperature, while the percentage of proteins remaining in
the obtained chitin decreases. Temperature seems to be a
critical parameter for the deproteinization with respect to the
purity of the obtained chitin. Differences between both results
can be expected because the deproteinization process carries

a

deproteinization
temperature (°C)

[η] (mL/g)

DAa (%)

15
22
50
70

3380 ( 200
3300 ( 200
3300 ( 200
3300 ( 200

106 ( 2
107 ( 2
99 ( 2
91 ( 2

Measured by solid NMR.

on during the washing step. In the case of elemental analysis,
negative values were obtained for almost pure products
because of the low precision of the method.
In Table 4, the intrinsic viscosity of the various chitin
samples obtained after 24 h of deproteinization in NaOH 1
M are given as a function of the deproteinization temperature.
The product obtained at 15 °C remains polluted by nonhydrolyzed or very weakly hydrolyzed proteins, and as a
consequence, the determination of the viscosity is wrong.
For the deproteinizations carried out at higher temperatures,
the obtained chitin is purer and the viscosities are stable with
time. As expected, the deproteinization conditions do not
involve the hydrolysis of the chain even at 70 °C. However,
the alkaline treatment can induce the hydrolysis of the
N-acetyl linkage and decrease the DA. The DA was estimated
by solid NMR (data not shown). No deacetylation was
noticed, even for the reaction performed at 70 °C. Then, it
is possible to increase the rate of deproteinization by an
increase of temperature without any influence on DA up to
70 °C. In a paper in preparation, we will further describe
the kinetics and thermodynamic parameters of the deproteinization steps.
Conclusion
The demineralization process of shrimp shells can be
simply followed by the measurement of the variation of pH

in the supernatant, and the increase of pH is easily related
to the calcium release. The study of the variation of pH also
allows us to follow the kinetics of demineralization and then
to estimate the optimal reaction time and to foresee the exact
amount of acid necessary to perform a complete reaction
thus minimizing the hydrolysis of the glycosidic bonds. We
notice that the demineralization times reported in the
literature are too long. Indeed, the reaction is mostly complete
within 15 min in the conditions used in this paper. Another
interesting result is that the DA of obtained chitin remains
stable with this mild acidic treatment.
In NaOH 1 M as reactive, the deproteinization process is
slow and several hours are necessary to perform a satisfying
deproteinization. This behavior has to be related both to the
R chitin structure of difficult access to the reactives and to
the various kinds of possible interactions with proteins (paper
in preparation). The measurement of the absorbance of the
supernatant at 280 nm is a reliable, simple method to follow
the protein release. The reaction rate can be increased by an
increase of temperature. The effect of temperature on the
intrinsic characteristics of the obtained chitin was studied,


18

Biomacromolecules, Vol. 4, No. 1, 2003

and as expected, the deproteinization time and temperature
have no influence on both the molecular weight and DA
using NaOH 1 M with a temperature and a reaction time

below 70 °C and 24 h, respectively.
Acknowledgment. This work belongs to the CARAPAX
project financially supported by the EC through the 5th
PCRD.
References and Notes
(1) Roberts, G. A. F. In Chitin chemistry; Roberts, G. A. F., Ed.;
Macmillan Press Ltd.: London, 1992.
(2) Ravi Kum, M. N. V. React. Funct. Polym. 2000, 45, 1.
(3) Muzzarelli, R. A. A.; Tanfani, F.; Emanuelli, M.; Chiurazzi, E.; Piani,
M. In Chitin in Nature and Technology; Muzzarelli, R. A. A.,
Jeuniaux, C., Gooday, G. W., Eds.; Plenum Press: New York, 1986;
p 469.
(4) Vander, P.; Vårum, K. M.; Domard, A.; El Gueddari, N. E.;
Moerschbacher, B. M. Plant Physiol. 1998, 118, 1353.
(5) Ho¨rner, V.; Pittermann, W.; Wachter, R. In AdVances in Chitin
Science; Proceedings of the 7th International Conference on Chitin/

Percot et al.

(6)
(7)
(8)

(9)
(10)
(11)
(12)
(13)
(14)
(15)

Chitosan; Domard, A.; Roberts, G. A. F.; Vårum, K. M., Eds; Jacques
Andre´ Publisher: Lyon, France, 1997; p 671.
Kurita, K.; Tomita, K.; Tada, T.; Ishii, S.; Nishimura, S.; Shimoda,
K. J. Polym. Sci., Part A: Polym. Chem. 1993, 31, 485.
Madhavan, P.; Ramachandran Nair, K. G. Fish. Technol. 1974, 11,
50.
Shirai, K.; Palella, D.; Castro, Y.; Guerrero-Legaretta, I.; SaucedoCastan˜eda, G.; Huerta-Ochoa, S.; Hall, G. M. In AdVances in Chitin
Science; Proceedings of the third Asia-Pacific Chitin and Chitosan
Symposium; Hirano, S.; Tokura, S.; Kurita, K.; Hsing-Chen, C.;
Rong, H. C., Eds; Rong H. C., Hsing, C. C. Publisher: 1998, p 103.
Hirai, A.; Odani, H.; Nakajima, A. Polym. Bull. 1991, 26, 87.
Teng, W. L.; Khor, E.; Tan, T. K.; Lim, L. Y.; Tan, S. C. Carbohydr.
Res. 2001, 332, 305.
Austin, P. R. German Patent 2,707,164, 1977.
Chang, K. L. B.; Tsai, G. J. Agric. Food Chem. 1997, 45, 19001904.
Rhazi, M.; Desbrie`res, J.; Tolaimate, A.; Alagui, A.; Vottero P. Polym.
Int. 2000, 49, 337.
Peterson, G. L. Anal. Biochem. 1977, 83, 346.
Hunt, S.; Nixon, M. Comp. Biochem. Physiol. 1981, 68B, 535.

BM025602K