Carbohydrate Polymers 74 (2008) 813–821

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Polyelectrolyte complex formation using alginate and chitosan

Heidi Vogt Sther a, Hilde K. Holme b,c, Gjertrud Maurstad a, Olav Smidsrød b, Bjørn T. Stokke a,*
a
Biophysics and Medical Technology, Department of Physics, The Norwegian University of Science and Technology, NTNU, NO-7491 Trondheim, Norway
b
Norwegian BioPolymer Laboratory, Department of Biotechnology, The Norwegian University of Science and Technology, NTNU, NO-7491 Trondheim, Norway
c
FMC BioPolymer AS/NovaMatrix, Industriveien 33, N-1337 Sandvika, Norway

a r t i c l e i n f o a b s t r a c t

Article history: Polyelectrolyte complexes (PECs) of alginate and chitosan were formed by addition of 0.1% alginate solu-
Received 2 January 2008 tion (pH 6.5) to 0.1% chitosan solution (pH 4.0), and by adding the chitosan solution to the alginate
Received in revised form 21 April 2008 solution under high shearing conditions. Variations in the properties of the polymers and the preparation
Accepted 29 April 2008
procedure were studied, and the resultant PEC size, zeta potential (Zp), and pH were determined using
Available online 15 May 2008
dynamic light scattering (DLS), electrophoresis and by measuring turbidity and pH. Tapping mode atomic
force microscopy (AFM) was used to examine some of the complexes. The particle size was decreased as
Keywords:
the speed and diameter of the dispersing element of the homogenizer was increased. The net charge ratio
Alginate
Chitosan
between chitosan and alginate, and the molecular weights (MW) of both the alginate and chitosan sam-
Polyelectrolyte ples were the most significant parameters that influenced the particle size, Zp, and pH. The mixing order
Complexes also influenced the size of the PECs, however, the Zp and pH were not affected by the mixing order. The
Zeta potential stability of the complexes was investigated by incubation at an elevated temperature (37 °C), storage for
Particle size one month at 4 °C, alteration of the pH of the PEC mixture, and addition of salt to physiological ionic
strength (0.15 M NaCl). The properties of the PEC could be affected according to the molecular properties
of the polyelectrolytes selected and the preparation procedures used. The resultant PEC sizes and prop-
erties of the complex were rationalised using a core-shell model for the structure of the complexes.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction cean shells. Chitosan is a linear binary copolymer that consists of

b (1 ? 4)-linked 2-acetoamido-2-deoxy-b-D-glucopyranose (Glc-
Oppositely charged polysaccharides in aqueous solutions inter- NAc; A-unit) and 2-amino-2-deoxy-b-D-glucopyranose (GlcN; D-
act spontaneously to form polyelectrolyte complexes (PECs) when unit). The A- and D-type residues are randomly distributed along
they are mixed. The entropy gain associated with the release of the chitosan chain (Vårum, Anthonsen, Grasdalen, & Smidsrød,
counter-ions is one of the major driving forces for PEC formation 1991a, 1991b). The average linear charge density of chitosans is
(Thünemann, Müller, Dautzenberg, Joanny, & Löwen, 2004). Poly- determined by the degree of acetylation (FA) and also by the pH
saccharides are of particular interest due to their desirable biocom- of the solution. The degree of acetylation determines the density
patible, biodegradable, hydrophilic and protective properties of the amino groups that can carry a charge, and the pH determines
(Barichello, Morishita, Takayama, & Nagai, 1999; Chellat et al., the degree of protonization of the amino groups. Chitosans of dif-
2000). Interactions between cationic chitosan and anionic alginate ferent chemical compositions (FA values in the range of 0.01–
leads to PEC formation, and these PECs have potential applications 0.49) have similar dissociation constants (pKa) of 6.5–6.6 (Strand,
such as drug or gene delivery systems in biomedicine. It has been Tømmeraas, Vårum, & Østgaard, 2001).
demonstrated that linking bioactive alginate to particles increases Alginate is a natural anionic polysaccharide and can be ex-
biological stimulation (Berntzen et al., 1998). The alginate–chito- tracted from marine brown algae. Alginate is a linear binary
san PECs may also present an immune stimulating property arising copolymer that consists of (1 ? 4)-linked b-D-mannuronic acid
from the immunogenic properties of the constituting polymers (M) and a-L-guluronic acid (G) residues. The relative amount of
that can be affected by the form of presentation analogously to the two uronic acid monomers and their sequential arrangement
conjugation of alginates to particles. along the polymer chain vary widely, depending on the origin of
Chitosan is a natural cationic polysaccharide that is formed by the alginate. The uronic acid residues are distributed along the
the N-deacetylation of chitin, which is a product found in crusta- polymer chain in a pattern of blocks, where homopolymeric blocks
of G residues (G-blocks), homopolymeric blocks of M residues (M-
blocks), and blocks with an alternating sequence of M and G units
* Corresponding author. Fax: +47 73 59 77 10.
E-mail addresses: [email protected], [email protected] (B.T. (MG-blocks) can co-exist [for a review see Moe, Draget, Skjåk-
Stokke). Brk, & Smidsrød, 1995]. Alginates may be designed with other

0144-8617/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2008.04.048
814 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821

monomeric sequences and can be produced using biotechnological Table 1

methods (Bjerkan et al., 2004). The polyelectrolyte properties of Chitosans used for the formation of alginate–chitosan complexes

the monomer residues of alginates have previously been reported FA a

App. viscosityb g (mPas) MWc (kg/mol) Identification term
upon (Haug, 1964), and the pKa-values for mannuronic and gulu- 0.14 160 400 C0.14,400
ronic acid monomers were found to be 3.38 and 3.65, respectively. 0.14 88 343 C0.14,343
Polyelectrolyte complexes may be formed from alginate and 0.14 16 117 C0.14,117
chitosan using different methods. In this study, we used a one- 0.15 4 47 C0.15,47
0.47 5 50 C0.47,50
stage process, where one polymer solution is slowly added in a 0.61 n.d.d 115 C0.61,115
dropwise manner into the other solution (Daly & Knorr, 1988). 0.02 7 50 C0.02,50
Alternatively, a two-stage procedure may be used to form PECs. a
FA determined by NMR spectroscopy (Vårum et al., 1991a).
The first step in this process is the preparation of Ca-alginate gel b
Apparent viscosity g (mPas) measured in a 1% solution (w/w), Brookfield
beads (Hackel, Klein, Megenet, & Wagner, 1975; Skjåk-Brk & rotational viscosimeter, water bath at 20 °C and 20 rpm.
Martinsen, 1991).The beads are then transferred into a chitosan c
MW determined by SEC-MALLS (Size Exclusion Chromatography Multi Angle
solution, and a PEC membrane is formed on the surface of the Laser Light Scattering). The MW of the chitosans C0.47,50 and C0.02,50 are calculated
from viscosity measurements.
beads (Douglas & Tabrizian, 2005; Gåserød, Smidsrød, & Skjåk- d
Not determined.
Brk, 1998).
A more molecular based structure-function relationship for the
formation of PECs from alginate and chitosan and the parameters weight (Holme, Davidsen, Kristiansen, & Smidsrød, 2008). The
that influence the process are needed to in order to produce sample notation CFA,MW was used for the chitosans, where the first
high-quality, safe products that can be used in biomedical applica- subscript represents the FA, and the second, the weight average
tions (Skaugrud, Hagen, Borgersen, & Dornish, 1999). PEC forma- molecular weight (MW) in kg/mol. Alginates with various weight
tion has previously been reported to be affected by the ratio of average molecular weight, MW, and various fractions of guluronic
alginate to chitosan, the molecular weight of the polysaccharides acid (G) and mannuronic acid (M) units were used to investigate
and the pH of the solutions (Douglas & Tabrizian, 2005). The frac- the effects on the PECs (Table 2). The fraction of G in the alginates,
tion of acetylated monomers, FA, in the chitosan has also been FG, was determined by NMR spectroscopy (Grasdalen, 1983). The
shown to be a determining parameter in PEC formation (Lee, Park, MW of the alginates was determined by SEC-MALLS (Holme et al.,
& Ha, 1997). It is necessary to study PECs in physiological environ- 2003). The sample notation AlgFG,MW was adapted for the alginates,
ments in order to assess their suitability for use in pharmaceutical and the first subscript was used to describe the fraction of G-units
and biomedicinal applications. In this study, we investigated vari- (FG) and the second subscript referred to the MW in kg/mol.
ations in preparation procedures and polymer characteristics in or-
der to study their influence on PEC particle size and particle surface 2.2. Preparation of polyelectrolyte complexes
charge (zeta potential, Zp), and the final pH of the PEC solution in
more detail than has previously been revealed. The influence of Aqueous stock solutions (0.1% w/v) of sodium alginate and
biopolymeric properties such as charge ratio [alginate/chitosan ra- chitosan chloride were prepared by dissolving the polymer in
tio, K(/+)], the average molecular weights (MW) of chitosan and MQ-water and magnetically stirring (100 rpm) the solution over-
alginate, and chemical composition (FA of chitosan or fraction of night at 4 °C. The pH of the alginate solution was 6.5, and the pH
guluronic acid, FG of alginate) were studied, and some of the parti- of the chitosan chloride solution was 4.0 after the dissolution.
cles were visualized by AFM. The stability of the PECs was also These pHs ensured that the alginate was fully deprotonated and
investigated with respect to storage time, physiological pH and that the chitosan was fully protonated. The polyelectrolyte com-
temperature. Salt (NaCl) was added to the PEC solutions in order plexes (PECs) were prepared at room temperature. The PEC solu-
to obtain physiological conditions, and the resulting particle sizes tions were mixed using an Ultra-turrax homogenizer with high
and zeta potentials were studied. shearing conditions. The alginate solution was added in a dropwise
This study focused on PEC formation from alginate and chitosan manner to the chitosan solution at a flow rate of 1 ml/s, thus form-
using a one-stage process. A sodium alginate solution was added to ing alginate in chitosan PECs (AiC PECs). The PECs were prepared in
a chitosan chloride solution under high shearing conditions, or order to give a final charge ratio K = qalginate/qchitosan 2 (0.01, 17) for
chitosan chloride was added to sodium alginate. We describe complex formation, under the assumption that the chitosan was
how the PEC particle sizes and surface charge can be controlled fully protonated, using the calculation described by Danielsen,
by varying preparation procedures and polymer characteristics. Vårum, and Stokke (2004). The order in which the solutions were
added was also investigated in order to form chitosan in alginate
PECs (CiA PECs) in addition to the AiC PECs. To examine the influ-
2. Materials and methods ence of chitosan and alginate molecular parameters, PECs from dif-
ferent alginates and chitosans were prepared. The notation
2.1. Biopolymer samples (AlgFG,MW CFA,MW)K was adapted in order to describe the PECs,
and the subscript K represents the net polyanion–polycation
The sodium alginate and chitosan chloride samples were kindly
provided by FMC BioPolymer AS/NovaMatrix (Sandvika, Norway). Table 2
The chitosan samples (Table 1) varied in their degree of acetylation Alginates used for the formation of alginate–chitosan complexes
(FA) and in their MW. The chitosan FA was determined by NMR
FGa App. Viscosityb g (mPas) MWc (kg/mol) Identification term
spectroscopy (Vårum et al., 1991a), and the chitosan MW was
determined by SEC-MALLS (Size Exclusion Chromatography 0.46 127 170 Alg0.46,170
0.69 138 219 Alg0.69,219
equipped with a Multi Angle Laser Light Scattering detector) 0.69 5 41 Alg0.69,41
(Holme, Lindmo, Kristiansen, & Smidsrød, 2003) using the mobile
a
phase (0.2 M CH3COONH4, pH 4.5), and also from the apparent vis- FG determined by NMR spectroscopy (Grasdalen, 1983).
b
Apparent viscosity g measured in a 1% solution (w/w), Brookfield rotational
cosity (g/mPas) of a 1% chitosan aqueous solution (w/w) (Brook- viscosimeter, water bath at 20 °C.
field rotational viscometer 20 rpm, at 20 °C) using a previously c
MW determined by SEC-MALLS (Size Exclusion Chromatography Multi Angle
derived relationship between apparent viscosity and molecular Laser Light Scattering).
 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821 815

charge ratio (/+) in the final PEC dispersion, AlgFG,MW and CFA,MW after 1–2 min of incubation, the samples were dried in a stream
reflect the alginate and chitosan used for PEC formation, respec- of N2 gas. An AFM (Digital Instrument Multimode IIIa) was used
tively, and the order of AlgFG,MW and CFA,MW in the brackets repre- for imaging. The instrument was operated in tapping mode, as pre-
sents the mixing order [alginate into chitosan (AlgFG,MW CFA,MW) or viously described (Maurstad, Danielsen, & Stokke, 2003; Stokke,
chitosan into alginate (CFA,MW AlgFG,MW)]. Using this notation, Falch, & Dentini, 2001) using silicon nitride cantilevers (PPP-
(Alg0.46,170 C0.14,400)0.3 complexes were prepared by adding alginate NCH-W, Pointprobe plus, Nanosensors) with nominal spring con-
solution into chitosan solution to yield an average charge ratio (/ stants of 10–130 N/m. The topographs were obtained at scan sizes
+) of 0.3, and the use of an alginate with FG = 0.46 and a MW of in a range of 0.5 lm  0.5 lm to 15 lm  15 lm (512  512 pix-
170 kg/mol, and a chitosan with an FA = 0.14 and a MW of 400 kg/ els). The topographs were flattened line by line excluding the poly-
mol. electrolyte complexes to correct for the tube-like scanner
movement on a small sector of a sphere in presenting the height
2.3. Homogenization over the supporting surface, and use only the determined height
of the supporting mica as the basis for this procedure.
The PECs were generally prepared by mixing polyelectrolyte
solutions with an Ultra-turrax homogenizer (Ika Werke) at 2.6. PEC stability upon changes in pH, temperature and storage time
11,000 rpm, using a dispersing element with a 25 mm diameter.
However, the influence of the blending conditions on the proper- In order to investigate PEC stability with respect to pH, the solu-
ties of the AiC PECs (mass ratio = 1, corresponding to K = 1.2) was tions of the complexes were characterized by following changes in
investigated, using dispersing elements that varied in size (diame- pH. After the (Alg0.69,5 C0.47,50)0.2 complexes were prepared, the pH
ters of the dispersing element = 5, 17 and 25 mm) and speed (11, of the PEC solution was adjusted (pH 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 6.4,
000–24,000 rpm). The details of the different homogenizers are 6.7, 7.0, 7.4, 8.0) by addition of 0.1 M HCl or 0.1 M NaOH while the
shown in Table 3. solution was magnetically stirred. Experiments were also carried
out in order to examine the effects of temperature and storage time
2.4. Characterization of the complexes upon PEC stability. PEC (Alg0.69,41 C0.47,50)0.2 (pH not adjusted) was
incubated at 37 °C using a water bath, and the PECs were charac-
In order to characterize the complexes, the z-average diameter terized after 30 min, 4 h and 8 h incubation periods. PEC
(sz), zeta potential (Zp), pH and transmittance of the PEC solutions (Alg0.69,41 C0.14,177)0.3, (Alg0.69,41 C0.15,47)0.3 and (Alg0.69,41 C0.47,50)0.3
were measured 1–3 h after the complexes were prepared. The pH were characterized immediately after formation, and also after
was measured (Radiometer Copenhagen, PHM 92 Lab pH meter) 1 month storage at 4 °C.
while the PEC solution was magnetically stirred. In order to obtain
a qualitative measure of the sizes of the PECs, the transmittance of
the PEC solutions was measured at k=800 nm (UV-1201, Shimad- 3. Results
azu). At this wavelength, neither alginate nor chitosan absorb light,
therefore, the size and number of particles is the only variables 3.1. Influence of homogenization parameters
that affect the degree of transmittance. MQ-water was used as a
standard, and was set to 100% transmittance. A dynamic light scat- The polyelectrolyte complexes were prepared using different
tering (DLS) measurement was made using a Zetasizer Nano homogenizers, and the speed and diameter of the dispersing ele-
instrument (Zetasizer Nano, Malvern instrument, USA), and used ments of the homogenizers were varied in order to examine the ef-
to determine sz. Based on measurements of the Brownian motion fects of homogenizer parameters upon the PECs obtained (Table 3).
of the particles in the solution, the z-average diameters (sz) of the PEC (Alg0.69,41 C0.15,50)1.2 was prepared at a mass ratio of 1, which
particles were determined using the Stokes–Einstein equation. corresponded to a charge ratio, K (/+) 1.2. The particle size dis-
The Zp of the solutions was measured by laser Doppler electropho- tributions showed that the PEC diameters decreased as homoge-
resis (Zetasizer Nano, Malvern instrument, USA). The sz and Zp val- nizer speed was increased (Fig. 1). At a homogenizer speed of
ues are the mean values with standard deviations from three to six 24,000 rpm, particles with an average diameter (sn) of 100 nm
replicate measurements of each sample. For K  1, most PEC diam- were formed, whereas at 11,000 rpm, particles of sn  1000 nm
eters had values of sz > 6 lm. This is beyond the range of the Nano were formed. Use of a homogenizer speed of 11,000 rpm led to a
sizer instruments (range 0.6 nm–6 lm according to the manufac- more narrow particle diameter distribution than when a homoge-
turer), therefore these measurements were only used in a qualita- nizer speed of 16,000 rpm (data not shown) or 24,000 rpm (Fig. 1)
tive manner, and the 6 lm level was marked on the figures. was used. Furthermore, the diameter of the dispersing element
was also shown to influence the particle size. The smallest diame-
2.5. AFM imaging ter of the dispersing element [T8 (5 mm, 25,000 rpm)] resulted in
the production of large particles sn  6 lm, whereas the largest
Samples for AFM imaging were prepared by transferring 10 lL diameter of the dispersing element [T25 (25 mm, 24,000 rpm)]
PEC solution to a freshly cleaved 5 mm diameter mica disk, and led to the production of smaller particles sn  100 nm (Fig. 1).

3.2. Influence of charge ratio K(/+) and mixing order

Table 3
Ultra-turrax homogenizers used in the experiments
Fig. 2 shows the influence of variations in charge ratio (K rang-
Homogenizer Diameter, dispersing Speed Term used for
ing from 0.01 to 17) on PEC formation by adjusting the amount of
element (mm) (1000 rpm) identificationa
alginate and chitosan. The effect of altering the order of polyelec-
T8 (IKA Werke) 5 5–25 T8(5 rpm) trolytes mixing is also shown in Fig. 2. Significant alterations in
Janke & Kunkel 17 20 JK(17 rpm)
(IKA Werke)
sz (Fig. 2a), transmittance (Fig. 2b), Zp and pH (Fig. 2c) values occur
T25 (IKA Werke) 25 11–24 T25(25 rpm) at a stoichiometric charge ratio between the polyanion and polyca-
a
tion. A significant increase in sz and pH, and a concomitant reduc-
To identify the Ultra-turrax homogenizers, the first number in the parenthesis
indicates the diameter of the dispersing element, while the second number indi-
tion of transmittance and Zp was observed when K  1 (Fig. 2). The
cates the speed (1000 rpm) used to mix the polyelectrolyte solutions. sz increased from 470 nm for (Alg0.69,41 C0.47,50)0.3 to sz > 6 lm for
816 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821

Fig. 1. The effect of homogenizer size and speed on the size-distribution of the
alginate in the chitosan polyelectrolyte complex (AiC PEC) formed at a mass ratio of
1 (K = 1.2), using (Alg69,41 C0.15,50). The mean number (%) vs. size (diameter)
distributions when using homogenizers with dispersing elements with diameters of
25 mm (T25), 17 mm (JK17) or 5 mm (T8) and homogenizer speeds ranging
between 11,000 and 24,000 rpm is shown. The nominal upper limit for the
determination of a particle size of 6 lm is shown as a dotted line.

(Alg0.69,41 C0.47,50)1.0 (Fig. 2a). When K was increased from 0.3 to

1.0, the transmittance of these complexes at 800 nm decreased
from over 0.98 to below 0.88 (Fig. 2b), the pH increased from
4.0–4.5 to 6.0–6.5, and the Zp decreased from +40 to 40 mV
(Fig. 2c).
Changing the order of polyelectrolyte mixing from polyanion in
a polycation solution to polycation in a polyanion solution affected
the properties of the PECs obtained. For cationic PECs (K < 1), the sz
for (C0.15,47 Alg0.69,41) was larger than the sz for (Alg0.69,41 C0.15,47).
For anionic PECs (hence K > 1, an excess of alginate), sz(C0.15,47
Alg0.69,41) < sz(Alg0.69,41 C0.15,47). These observations indicated that
the order of polymer addition affects the size of the PEC particles
produced.

3.3. Influence of alginate and chitosan molecular parameters

The preparation of alginate in chitosan polyelectrolyte com-

plexes using alginates and chitosans that had different molecular Fig. 2. Parameters of the alginate–chitosan polyelectrolyte complexes (PEC)
parameters and K values of 0.3, 1 and 3 indicated that the most obtained at charge ratios K from 0.01 to 17. (a) The z-average diameter sz, (open
symbols), (b) the transmittance (open symbols) and (c) Zp (open symbols) and pH
important parameter that influences PEC properties is the charge
(filled symbols) for AiC and CiA PECs vs. the alginate to chitosan charge ratio. AiC
ratio K(/+) (Fig. 3). For K  1, the complexes were unstable, which PECs were (Alg0.69,41 C0.15,47) (s, d) and (Alg0.69,41 C0.47,50) (4, N), whereas the CiA
was indicated by fluctuations in Zp, pH (Fig. 3c) and transmittance PEC was (C0.15,47 Alg0.69,41) (h, j).
measurements (Fig. 3b). The variations of sz, Zp and pH with re-
spect to FA are small. The smallest PECs were made when chitosan K = 0.3 to 0 mV at K = 1, and a further decrease in Zp < 40 mV
was in excess, and when the alginate solution was added to the was observed when K = 3. The pH of the PEC dispersion was 4.0–
chitosan solution. At K  0.3, the sz values of (Alg0.69,41 C0.02,50)0.3, 5.0 at K = 0.3, and increased to pH 6.5–7.5 when K = 3. The size of
(Alg0.69,41 C0.15,47)0.3 and (Alg0.69,41 C0.47,50)0.3 were 748 nm, complexes at K  1 was sz > 6 lm (Fig. 3a) and the transmittance
526 nm and 467 nm, respectively. In an excess of alginate, and (800 nm) < 0.88 (Fig. 3b).
K = 3, the sz values for (Alg0.69,41 C0.02,50)3, (Alg0.69,41 C0.15,47)3 and Selection of polymers with various MW values influenced the
(Alg0.69,41 C0.47,50)3 were 2.4 lm, 1.2 lm and 1.5 lm, respectively. resulting sz of the complexes (Figs. 4 and 5). Increasing the MW of
Altering the FA of the chitosan did not affect the pH or the Zp of the polymer in excess, either alginate (K  3) or chitosan
the PEC formed (Fig. 3c). The Zp decreased from +40 mV at (K  0.3), led to an evident increase in PEC size. This effect was
 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821 817

Fig. 4. The z-average diameter for alginate in chitosan polyelectrolyte complexes

vs. the molecular weight of the chitosan (FA = 0.14) used to prepare the complexes.
The alginate Alg0.69,41 was used to prepare the polyelectrolyte complexes using
alginate to chitosan charge ratios of 0.3 (open symbols) and 3 (filled symbols).

Fig. 3. (a) The z-average diameter sz (open symbols), (b) transmittance (800 nm, Fig. 5. The z-average diameter for alginate in chitosan polyelectrolyte complexes
open symbols), (c) Zp (open symbols) and pH (filled symbols) for different alginate vs. the molecular weight of alginate when a chitosan C0.15,47 was used to prepare
to chitosan charge ratios (K = 0.3, 1 and 3) for AiC PEC made from C0.02,50 (s), C0.15,47 the complexes. Complexes were prepared using alginates of two different G-
(4) and C0.47,50 (h). The alginate was constant Alg0.69,41. fractions, FG = 0.69 (circle) and FG = 0.46 (triangle), and alginate to chitosan charge
ratios of 0.3 (open symbols) and 3 (filled symbols).

not as obvious when the MW of the deficient polymer was in-

creased. Alginates with different FG values were used to prepare structures, and instead larger complexes with various shapes,
complexes, and the chitosan (C0.15,47) was kept the same. There flocks, and aggregates were observed (Fig. 6b). In an excess of algi-
was no systematic variation in the properties of the complexes nate, i.e. for PEC (Alg0.69,41 C0.14,343)3, both complexes and aggre-
according to the G-content of the alginate (Fig. 5). An overall gates were observed, and a chequered layer was also observed in
assessment of these data (Figs. 3–5) indicates that chitosans and the background (Fig. 6c).
alginates with low MW values give the smallest complexes, how-
ever, altering the molecular parameters FG and FA does not signifi- 3.5. Stability of the PECs
cantly change the PEC properties.
The PECs were subjected to changes in pH, temperature and
3.4. AFM imaging were also stored for one month at 4 °C. One month storage at
4 °C or incubation at 37 °C for 8 h did not cause any changes in
The PECs were visualized by AFM. Three samples prepared at PECs diameter or Zp (Fig. 7).
different conditions and with varying scan sizes are shown in When the pH in the PEC (Alg0.69,41 C0.47,50)0.2 solution was ad-
Fig. 6. Complexes exist in variable shapes and forms, such as round justed within an interval of pH 3–6.4 by addition of HCl or NaOH,
spherical particles, linear chains, segments and aggregating flocks the particle diameter stayed almost constant (Fig. 8). As the pH was
(Fig. 6). Fig. 6a shows PEC (Alg0.69,41 C0.47,50)0.3 (excess of chitosan) increased between 6.4 and 6.7, the diameter of the PEC particles
appears as small segments with spherical, round and linear shapes. significantly increased. As shown in Fig. 8, the observed particle
The sizes of these complexes were determined to be sz  500 nm diameter at pH 7.0 is about 50 times the size observed at pH 6.
using DLS. For PEC (Alg0.69,41 C0.14,343)1, i.e. at K  1, sz > 6 lm, it The particle size levels out at pH 7.0. A decrease in the Zp of the
was not possible to observe clearly distinguishable isolated complexes was observed as the size of the particles increased.
818 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821

Fig. 6. AFM height topographs obtained from three different PECs (a–c). (a) PEC (Alg0.69,41 C0.47,50)0.3, (b) PEC (Alg0.69,41 C0.14,343)1 and (c) PEC (Alg0.69,41 C0.14,343)3.

Table 4
Comparison of 0 M NaCl and 0.15 M NaCl in PEC solution
0 M NaCl 0.15 M NaCl in PEC solution
Size (nm) Zp (mV) Size (nm) Zp (mV)
(Alg0.69,41 C0.15,47)0.2 1458 ± 120 58 ± 1 501 ± 7 26 ± 1.1
(Alg0.69,41 C0.47,50)0.2 567 ± 22 43 ± 0.3 1538 ± 161 16 ± 1.4
(Alg0.69,41 C0.61,115)0.2 1755 ± 194 39 ± 1 1090 ± 18 12 ± 0.8
(Alg0.69,41 C0.15,47)1.8 1781 ± 56 49 ± 0.8 >6 lm (18390 ± 1168) 24 ± 2.4

Table 4 shows that the absolute value of the Zp of the PECs is

reduced from 40–50 mV to 15–25 mV after salt is added to a final
concentration of 0.15 M NaCl. No systematic variation in particle
size was observed when salt (0.15 M NaCl) was added to the PEC
mixture.

4. Discussion

4.1. The influence of homogenizer parameters

The homogenization parameters used when blending the poly-

anions and polycations together appeared to be an important fac-
tor in controlling the particle size of the resulting polyelectrolyte
complexes. The data (Fig. 1) show that the resulting sizes of the
Fig. 7. The z-average diameter sz (a), and zeta potential Zp, (b) for alginate in polyelectrolyte complexes prepared from alginate and chitosan
chitosan polyelectrolyte complexes prepared using alginate Alg0.69,41 and chitosans are influenced by both the size- and speed parameters of the
C0.14,117 (s, d), C0.15,47 (4, N) and C0.47,0.50 (h, j), respectively. The measurements homogenizer. Homogenization at a high speed and use of a dis-
were performed immediately after the PEC was formed (solid line, open symbols)
persing element with a large diameter produces the smallest par-
and after one month storage at 4 °C (dashed line, filled symbols).
ticles. These data are in agreement with the size of the particles
that are described by the Ultra-turrax supplier when different dis-
persing elements are used; a large dispersing element diameter
should produce smaller particles. These observations are also in
accordance with effects on the size of polycaprolactone-coated
and chitosan-coated epichlorohydrin-crosslinked alginate micro-
spheres (Lillard, Palaniappan, & Koritala, 2004). High shear mixers
create an intense, concentrated energy input that can disperse par-
ticles significantly quicker than traditional mixing methods. The
size of the PECs indicates that several molecules are involved in
each complex, and therefore, polymer degradation is not expected
to contribute to the size of the PECs at the highest shear rates.
Based on these initial observations, a constant set of preparation
conditions were selected for further studies. Importantly, we found
that experiments carried out at stoichiometric charge ratio (K  1)
led to the production of PECs that were out of the range of the par-
ticle size characterization instruments, and the data under these
conditions was not reproducible from experiment to experiment.

4.2. DLS measurements and intensity size distribution sz

Fig. 8. The z-average diameter sz (s) and zeta potential Zp (4) for polyelectrolyte The polyelectrolytes used in this study are polydisperse, there-
complexes (Alg0.69,41 C0.47,50)0.2 vs. pH. The pH was adjusted in the PEC solution. fore, the given size always represents an average value of the par-
 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821 819

ticles in the solution. The aim of the current study was to select a ponent that is titrated into the mixture, the segregation of hydro-
few representative parameters that conveyed the essential features phobic segments lead to the production of particles which are
of the particles prepared under various conditions. The size distri- electrostatically stabilized by a large hydrophilic corona of the ex-
butions of the complexes were mostly monomodal. In cases with cess component. However, when the component in excess is smal-
more than one peak, the largest peak (i.e. most of the particles) ler than the component that is titrated in, the outcome of the
generally occurred at the largest particle diameter. Selection of process depends on the capacity of the excess polymer to neutral-
the average particle size ignores the possible presence of smaller ize charges. The ‘‘scrambled” egg model (Michaels & Miekka, 1961;
PECs within the distribution. The PEC particle size is an important Philipp, Dautzenberg, Linow, Kötz, & Dawydoff, 1989) may de-
parameter when determining locations for drug and gene delivery scribe some essential morphological features of polyelectrolyte
and entrapment (Douglas & Tabrizian, 2005). Therefore, accurate complexes made from flexible polymers. Within the ‘‘scrambled”
measurements of the particle size and the homogeneous PEC frac- egg model a possible interaction model where alignment of
tions are valuable for biological, medical and industrial applica- charges yielding aligned (/+) charge neutralization in a lattice-
tions. The AFM images (Fig. 6) and the standard deviations of sn like type of interaction, is not realized due to various type of mis-
(Fig. 1) and sz (Figs. 2a–5) indicate that there are variations in match arising from e.g. structural (charge spacing and distribu-
the sizes and shapes of PECs within one sample. The AFM images tions) or differences in polymer flexibility. Basically, all of the
indicate that a variety of particle shapes, which range from spher- charges on the molecules are not accessible for electrostatic inter-
ical and round segments via linear chains to larger aggregates, may actions for steric reasons within the ‘‘scrambled” egg model. Such
occur. The AFM images in a previous study that visualized PECs of features may also occur locally within a core-shell model, where
alginate and chitosan synthesized at lower polyelectrolyte concen- the focus is more on the net local charge balance in various regions
trations indicated that the PECs were mostly present as globular of the complexes. The recently reported persistence length of
species, and there was possibly also a poorly-resolved internal 12 nm alginates in an unperturbed state (Vold, Kristiansen, &
structure (Danielsen, Maurstad, & Stokke, 2005). Despite the short- Christensen, 2006), and in the order of 5 nm for chitosans (Rina-
comings associated with only selecting the mean value and the udo, Milas, & Le Dung, 1993; Schatz et al., 2004), along with a re-
associated standard deviation for discussion of the major trends port that alginate–chitosan PECs do not adapt regular rod-like or
in the data, this provide an indication from the DLS measurements toroidal morphologies (Maurstad et al., 2003) all support the no-
and the associated physical appearance within the distributions as tion that ‘‘scrambled” egg regions also may exist in the complexes
obtained from the AFM topographs. The concerns of possible ef- described in this study. This is also in line with the reported frac-
fects of preparation procedures, size of analysed ensemble, and tion of carboxylic acid of alginate being involved in complexation
artefacts (Stokke & Brant, 1990; Stokke, Elgsaeter, Hara, Kitamura, with chitosan being much less than one (Becherán-Marón, Peniche,
& Takeo, 1993) should also be included when judging the corre- & Argülles-Monal, 2004).
spondence between the data obtained by AFM and other physical In line with the basic features of the core-shell model, we ob-
measurements, as DLS applied for the present samples. served that the zeta potentials of the alginate–chitosan PECs reflect
the excess component, and therefore the value of the Zp changes
4.3. A core-shell polyelectrolyte complex model sign at K = 1. In an excess of either chitosan or alginate, the Zp
has a large positive or negative value and small, non-aggregating
We suggest that a core-shell model can account for some main particles are formed. The smallest particles are obtained in an ex-
features of the alginate–chitosan complexes obtained using differ- cess of chitosan at K = 0.3 (AiC) and in an excess of alginate at
ent blending conditions with respect to the molar ratio in the final K = 1.2 (CiA). This non-aggregating behaviour is suggested to arise
blend and the polymer properties (see Fig. 9). This model is analo- due to an excess of the major component, which forms a stabilizing
gous to the model proposed for polyelectrolyte complex formation shell around the particles.
for chitosan–dextran sulphate (Schatz, Domard, Viton, Pichot, & At the crossover from a negative to positive Zp (which occurs at
Delair, 2004). In the case of chitosan–dextran sulphate, PEC forma- K  1), the data indicate that there is an increase in the mean com-
tion resulted in particle flocculation at balanced charge ratios of plex size (sz > 6 lm; Figs. 2a,b, 3a,b, 7a and 8), and there are also
the components, whereas in an excess of cations (chitosan) or an- relatively large differences in size between parallel preparations.
ions (dextran sulphate), dense and small non-aggregating particles These observations probably reflect the formation of flocks and
were formed (Schatz et al., 2004). When the component in excess aggregates, as indicated in the AFM images (Fig. 6b.) These findings
acts as a host and can accommodate several molecules of the com- are in line with similar findings that were observed near a net
charge neutrality for DNA and cationic polymers (Tang & Szoka,
1997), chitosan and dextran sulfate (Drogoz, David, Rochas, Dom-
ard, & Delair, 2007) or the aggregation of poly(diallyldimethyldam-
monium)chloride and sodium poly(ethane sulfonate) at K = 1.1
(Kekkonen, Lattu, & Stenius, 2001), and other similar results (Daut-
zenberg, 2000; Schatz et al., 2004). The aggregation behaviour ob-
served around K  1 can be rationalized within the core-shell
model, since local patches that have a net charge that is different
from zero can participate in electrostatic-mediated interactions
with oppositely charged patches on other clusters. At an increasing
distance from K  1, the probability of having local patches of
opposite net charge on different clusters is expected to diminish.
The capability to build larger aggregates/agglomerates therefore
mainly occurs close to K  1. Although, the core-shell model
description of alginate–chitosan, as suggested here, accommodates
some main global trends of the PECs, it is expected that it needs to
Fig. 9. Schematic illustration of a core-shell model for polyelectrolyte complexes of
be elaborated when even more detailed characterization of the
alginate (polyanion) and chitosan (polycation) at net charge ratios less than one, PECs will be available, e.g., similar to that already reported by
equal to one and larger than one. Drogoz et al. (2007).
820 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821

In this study, altering the chitosan FA did not significantly influ- (Cherng et al., 1999). Moreover, a lower degree of protonation also
ence the properties of the resultant complexes. When the chitosan resulted in particles with lower absolute value of Zp, which ap-
FA is large, the mass ratio of chitosan/alginate is larger than when pears to cause aggregate formation.
chitosans with a low FA for a given K are used. However, our results The effect of adding salt into the PEC solution is shown in Table
still indicate that the particle sizes are not affected by FA. In this 4. As expected, the increase in ionic strength due to 0.15 M NaCl re-
context, chitosans with an FA value between 0.02 and 0.15 when sulted in a decrease of Zp (15–25 mV), clearly demonstrating the
fully protonated will have the same effective charge density due screening effect of counter-ions. Furthermore, reducing the charge
to counter-ion condensation. The chitosan with an FA = 0.47 has a of the surfaces should increase aggregate formation. Variations in
lower maximum charge density. The fact that the charge density the resulting PEC sizes were found, and this reflects an unstable
of the chitosan does not appear to affect the properties of the poly- system. When the resulting Zp values (15–25 mV) are compared
electrolyte complexes may indicate that not all of the charges are with the results presented in Fig. 8, it is clear that the PECs are
accessible to interact with the polyanions, which may be caused in a state where aggregate formation begins. Incubation at 37 °C
by conformational features. for 8 h (data not shown) did not affect the particle size, transmit-
Similarly, the data presented in Fig. 5 indicates that there is no tance, Zp or pH, and one month storage at 4 °C also did not affect
systematic influence of the G-content of the alginate on the prop- particle size or Zp (Fig. 7). Storage and temperature effects on
erties of the polyelectrolyte complexes. This result was expected PEC particles have been studied by several groups, and different re-
since the linear charge densities of alginates are almost indepen- sults have been obtained from different studies (Danielsen et al.,
dent of the G-content of the alginates. Previous reports have sug- 2005; Maurstad & Stokke, 2004; Romøren, Aaberge, Smistad, Thu,
gested that there are some differences in the molecular chain & Evensen, 2004; Vijayanathan, Lyall, Thomas, Shirahata, & Tho-
flexibility between the alginates (Moe et al., 1995), but more recent mas, 2005). Chitosan-based polyplexes were stored for one year
reports do not support an increase in the chain stiffness of algi- at 4 °C, 25 °C and 45 °C (Romøren et al., 2004). No changes were
nates that is associated with increasing G-content (Vold et al., observed in the polyplexes stored at 4 °C, and minor changes were
2006). Therefore, a correlation between alginate G-content and observed in the polyplex kept at 25 °C. The polyplex stored at 45 °C
the properties of the polylelectrolyte complexes mediated by var- changed within a short period of time. Another study reported that
iation in chain stiffness is also not expected to exist. incubation at 50 °C for 5.5 h led to annealing and the formation of a
The data (Figs. 4 and 5) show that PEC made from chitosans and denser chitosan–DNA complex(Danielsen et al., 2005), whereas
alginates with low MW form smaller complexes when compared to xanthan–chitosan PECs have been shown to increase in size when
chitosans and alginates with high molecular weights. The Zp and heated (Maurstad & Stokke, 2004). Stable DNA-polyamine particles
pH curves follow almost equal trends, and are independent of vari- were observed at temperatures <50 °C, while particle size signifi-
ations in the polymer MW. It has been reported that the molecular cantly increased for temperatures >60 °C (Vijayanathan et al.,
weight of the polymers influences PEC formation. The smallest par- 2005). From these reports, it is likely that storage beyond the per-
ticles were found using a combination of low MW alginate and low iod reported in these studies may affect the properties of PECs.
MW chitosan and a two-stage process (Douglas & Tabrizian, 2005).
A possible scenario that could account for the observed effects of 5. Conclusions
molecular weight is that an increased MW supports interactions
with accessible chain segments that are not engaged in other inter- This study demonstrated that the rotational speed and diameter
actions (analogously to tail and loop segments) to a larger extent, of the dispersion element of the homogenizer, and molecular prop-
and therefore facilitate the formation of PECs with larger sizes. This erties of the polyelectrolytes should be carefully chosen in order to
is also in agreement with the observation that an increase in the optimize the particle size and the particle surface charge (Zp) of
MW of the polymer in excess resulted in the largest increase in alginate–chitosan complexes. The charge ratio and the molecular
PEC size. weight were shown to be parameters that affected particle size,
Zp and pH in the mixture. At charge ratios close to one (neutral),
4.4. Stability the sizes of the largest particles were less reproducible. The small-
est particles were obtained by adding one of the two polymer solu-
The AiC PEC stability was tested with respect to changes in pH, tions into an excess amount of the other solution and by using
ionic strength and exposure to elevated temperatures. The most forms of the component polymers with low molecular weights.
significant changes were observed between pH 6.4 and 7.0, where Increasing the pH of the PEC mixture above 7 led to a significant
the PEC size increased 50 times and the Zp changed from +30 mV increase in particle size, while the PECs were very stable over a
at pH 6 to ±2.1 mV at pH 7.4 (Fig. 8). At pH > pKa, the amine groups temperature range of 4–37 °C. Increasing the ionic strength in
of the chitosan are less protonated and the remaining protonated the PEC mixture (0.15 M NaCl) led to PECs with less reproducible
groups available for interaction would result in overall weaker structures, and the PECs had a reduced zeta potential.
electrostatic interactions (Fig. 8), leading to the observed changes
in the particle size and zeta potential. A similar trend has been re- Acknowledgements
ported for alginate–chitosan PEC prepared by gelling (Douglas &
Tabrizian, 2005). Another study showed that there were strong This work was supported by The Norwegian Research Council,
plasmid-polymer [poly{(2-dimethylamino)ethyl methacrylate}- Grant No. 166794/V30 (GM) and FMC BioPolymer AS/NovaMatrix.
plasmid] interactions, and this resulted in the formation of smaller
complexes at pH 5. Increasing the pH close to the pKa value (7.5) References
of the polymer resulted in increased particle size and formation of
aggregates at pH 8 (Cherng, Talsma, Verrijk, Crommelin, & Henn- Barichello, J. M., Morishita, M., Takayama, K., & Nagai, T. (1999). Encapsulation of
hydrophilic and lipophilic drugs in PLGA nanoparticles by the nanoprecipitation
ink, 1999). This is in agreement with our observations that increas-
method. Drug Development and Industrial Pharmacy, 25, 471–476.
ing the pH close to the pKa value of chitosan of 6.5 (Strand et al., Becherán-Marón, L., Peniche, C., & Argülles-Monal, W. (2004). Study of the
2001) resulted in an increased particle size. It has been suggested interpolyelectrolyte reaction between chitosan and alginate: Influence of
that strong polymer interactions occur when most side groups of alginate composition and chitosan molecular weight. International Journal of
Biological Macromolecules, 34, 127–133.
the polymer are protonated, resulting in a complex of high charge Berntzen, G., Flo, T. H., Medvedev, A., Kilaas, L., Skjak-Braek, G., Sundan, A., et al.
density where the formation of small particles are favoured (1998). The tumor necrosis factor-inducing potency of lipopolysaccharide and
 H.V. Sther et al. / Carbohydrate Polymers 74 (2008) 813–821 821

uronic acid polymers is increased when they are covalently linked to particles. Maurstad, G., & Stokke, B. T. (2004). Metastable and stable states of xanthan
Clinical and Diagnostic Laboratory Immunology, 5, 355–361. polyelectrolyte complexes studied by atomic force microscopy. Biopolymers, 74,
Bjerkan, T. M., Lillehov, B. L., Strand, W. I., Skjåk-Brk, G., Valla, S., & Ertesvåg, H. 199–213.
(2004). Construction of hybrid Azotobacter vinelandii epimerases with new Michaels, A. S., & Miekka, R. G. (1961). Polycation–polyanion complexes:
epimerisation pattern characteristics. Biochemical Journal, 381, 813–821. Preparation and properties of poly-(vinylbenzyltrimehylammonium) poly-
Chellat, F., Tabrizian, M., Dumitriu, S., Chornet, E., Magny, P., Rivard, C.-H., et al. (styrenesulfonate). Journal of Physical Chemistry, 65, 1765–1773.
(2000). In vitro and in vivo biocompatibility of chitosan–xanthan polyionic Moe, S. T., Draget, K. I., Skjåk-Brk, G., & Smidsrød, O. (1995). Alginate. In A. Stephen
complexes. Journal of Biomedical Materials Research, 51, 107–116. (Ed.), Food Polysaccharides and their applications (pp. 245–286). Marcel Dekker
Cherng, J.-Y., Talsma, H., Verrijk, R., Crommelin, D. J. A., & Hennink, W. E. (1999). The NY: Marcel Dekker N.
effect of formulation parameters on the size of poly-((2-dimethylamino)ethyl Philipp, B., Dautzenberg, H., Linow, K.-J., Kötz, J., & Dawydoff, W. (1989).
methacrylate)-plasmid complexes. European Journal of Pharmaceutics and Polyelectrolyte complexes – Recent developments and open problems.
Biopharmaceutics, 47, 215–224. Progress in Polymer Science, 14, 91–172.
Daly, M. M., & Knorr, D. (1988). Chitosan–alginate complex coacervate capsules – Rinaudo, M., Milas, M., & Le Dung, P. (1993). Characterization of chitosan. Influence
Effects of calcium-chloride, plasticizers, and poly-electrolytes on mechanical of ionic strength and degree of acetylation on chain expansion. International
stability. Biotechnology Progress, 4, 76–81. Journal of Biological Macromolecules, 15, 281–285.
Danielsen, S., Maurstad, G., & Stokke, B. T. (2005). DNA-polycation complexation Romøren, K., Aaberge, A., Smistad, G., Thu, B. J., & Evensen, O. (2004). Long-term
and polyplex stability in the presence of competing polyanions. Biopolymers, 77, stability of chitosan-based polyplexes. Pharmaceutical Research, 21, 2340–2346.
86–97. Schatz, C., Domard, A., Viton, C., Pichot, C., & Delair, T. (2004). Versatile and efficient
Danielsen, S., Vårum, K. M., & Stokke, B. T. (2004). Structural analysis of chitosan formation of colloids of biopolymer-based polyelectrolyte complexes.
mediated DNA condensation by AFM: Influence of chitosan molecular Biomacromolecules, 5, 1882–1892.
parameters. Biomacromolecules, 5, 928–936. Skaugrud, O., Hagen, A., Borgersen, B., & Dornish, M. (1999). Biomedical and
Dautzenberg, H. (2000). Light scattering studies on polyelectrolyte complexes. pharmaceutical applications of alginate and chitosan. Biotechnology and Genetic
Makromolekulare Chemie, 162, 1–21. Engineering Reviews, 16, 23–40.
Douglas, K. L., & Tabrizian, M. (2005). Effect of experimental parameters on the Skjåk-Brk, G., & Martinsen, A. (1991). Applications of some algal polysaccharides
formation of alginate–chitosan nanoparticles and evaluation of their potential in biotechnology. In M. D. Guiry & G. Blunden (Eds.), Seaweed resources in
application as DNA carrier. Journal of Biomaterials Science, Polymer Edition, 16, Europe: Uses and potential (pp. 219–257). Chichester: John Wiley & Sons.
43–56. Stokke, B. T., & Brant, D. A. (1990). The reliability of wormlike polysaccharide chain
Drogoz, A., David, L., Rochas, C., Domard, A., & Delair, T. (2007). Polyelectrolyte dimensions estimated from electron micrographs. Biopolymers, 30, 1161–1181.
complexes from polysaccharides: Formation and stoichiometry monitoring. Stokke, B. T., Elgsaeter, A., Hara, C., Kitamura, S., & Takeo, K. (1993). Physicochemical
Langmuir, 23, 10950–10958. properties of (1 ? 6) branched (1 ? 3)-ß-D-glucans. 1. Physical dimensions
Gåserød, O., Smidsrød, O., & Skjåk-Brk, G. (1998). Microcapsules of alginate– estimated from hydrodynamics and electron microscopy. Biopolymers, 33,
chitosan – IA quantitative study of the interaction between alginate and 561–573.
chitosan. Biomaterials, 19, 1815–1825. Stokke, B. T., Falch, B. H., & Dentini, M. (2001). Macromolecular triplex zipping
Grasdalen, H. (1983). High-field, 1H NMR spectroscopy of alginate: Sequential observed in derivatives of fungal (1 ? 3)-b-D-glucan by electron and atomic
structure and linkage conformation. Carbohydrate Research, 118, 255–260. force microscopy. Biopolymers, 58, 535–547.
Hackel, U., Klein, J., Megenet, R., & Wagner, F. (1975). Immobilization of microbial Strand, S. P., Tømmeraas, K., Vårum, K. M., & Østgaard, K. (2001). Electrophoretic
cells in polymeric matrices. European Journal of Applied Microbiology, 1, light scattering studies of chitosans with different degrees of N-acetylation.
291–293. Biomacromolecules, 2, 1310–1314.
Haug, A. (1964). Composition and properties of alginates. Composition and Tang, M. X., & Szoka, F. C. (1997). The influence of polymer structure on the
properties of alginates. Thesis. Norwegian Institute of Technology, Trondheim. interactions of cationic polymers with DNA and morphology of the resulting
Holme, H. K., Davidsen, L., Kristiansen, A., & Smidsrød, O. (2008). Kinetics and complexes. Gene Therapy, 4, 823–832.
mechanisms of depolymerization of alginate and chitosan in aqueous solution. Thünemann, A. F., Müller, M., Dautzenberg, H., Joanny, J.-F., & Löwen, H. (2004).
Carbohydrate Polymers, 73, 656–664. Polyelectrolyte complexes. Advances in Polymer Science, 166, 113–171.
Holme, H. K., Lindmo, K., Kristiansen, A., & Smidsrod, O. (2003). Thermal Vårum, K. M., Anthonsen, M. W., Grasdalen, H., & Smidsrød, O. (1991a).
depolymerization of alginate in the solid state. Carbohydrate Polymers, 54, Determination of the degree of N-acetylation and the distribution of N-acetyl
431–438. groups in partially N-deacetylated chitins (chitosans) by high-field NMR
Kekkonen, J., Lattu, H., & Stenius, P. (2001). Adsorption kinetics of complexes formed spectroscopy. Carbohydrate Research, 211, 17–23.
by oppositely charged polyelectrolytes. Journal of Colloid and Interface Science, Vårum, K. M., Anthonsen, M. W., Grasdalen, H., & Smidsrød, O. (1991b). 13C-NMR
234, 384–392. studies of the acetylation sequences in partially N-deacetylated chitins
Lee, K. L., Park, W. H., & Ha, W. S. (1997). Polyelectrolyte complexes of sodium (chitosans). Carbohydrate Research, 217, 19–27.
alginate with chitosan or its derivatives for microcapsules. Journal of Applied Vijayanathan, V., Lyall, J., Thomas, T., Shirahata, A., & Thomas, T. J. (2005). Ionic,
Polymer Science, 63, 425–432. structural, and temperature effects on DNA nanoparticles formed by natural
Lillard, J. W., Palaniappan, R. & Koritala, P. R. (2004). PACE-A microspheres for and synthetic polyamines. Biomacromolecules, 6, 1097–1103.
delivery of antigens. United States Patent Application 20040071780. Vold, I. M. N., Kristiansen, K. A., & Christensen, B. E. (2006). A study of the chain
Maurstad, G., Danielsen, S., & Stokke, B. T. (2003). Analysis of compacted stiffness and extension of alginates, in vitro epimerized alginates, and
semiflexible polyanions visualized by atomic force microscopy: Influence of periodate-oxidized alginates using size-exclusion chromatography combined
chain stiffness on the morphologies of polyelectrolyte complexes. The Journal of with light scattering and viscosity detectors. Biomacromolecules, 7,
Physical Chemistry B, 107, 8172–8180. 2136–2146.