Carbohydrate Polymers 68 (2007) 708–717

www.elsevier.com/locate/carbpol

Rheological study of interactions between non-ionic surfactants

and polysaccharide thickeners used in textile printing
a,*
Rebeka Fijan , Sonja Šostar-Turk a, Romano Lapasin b

a
Institute of Engineering Materials and Design, Faculty of Mechanical Engineering, University of Maribor, Smetanova ulica 17, 2000 Maribor, Slovenia
b
Department of Chemical, Environmental and Raw Materials Engineering, University of Trieste, piazzale Europa 1, 34127, Italy

Received 10 January 2006; received in revised form 8 August 2006; accepted 8 August 2006
Available online 27 September 2006

Abstract

The influence of four non-ionic surfactants (isododecyl and cetyl polyoxyethylene ethers) on aqueous polysaccharide solutions (sodi-
um alginate, guar gum, and sodium carboxymethyl guar), applicable for textile printing pastes, were studied via rheological
measurements.
Rheology of polysaccharide–surfactant solutions in aqueous matrices is primarily governed by polymer content, which imparts
marked shear-thinning and viscoelastic character to the system. Such properties are modulated in moderate but sensible way by changes
in surfactant concentration or type. Above 3% surfactants addition to non-substituted guar gum solutions results in a significant impact
leading to phase separation and a particular strongly associated phase is formed due to hydrogen bonds between ethylene oxy units from
the surfactant and primary hydroxyl groups in guar.
A satisfactory fitting of viscosity data is obtained with both the Cross equation and the Roberts–Barnes–Carew model. The experi-
mental results of mechanical spectra can be described quite satisfactory with both the Friedrich–Braun and the generalized Maxwell
models.
 2006 Elsevier Ltd. All rights reserved.

Keywords: Polysaccharide; Textile printing; Non-ionic surfactant; Polymer–surfactant interactions; Viscoelasticity

1. Introduction Color pastes employed in the machine printing of tex-

tiles are prepared by mixing a thickening agent solution
From the variety of industrial application of polysac- and a dye solution or dispersion. The role of thickening
charides, it is probably true to say that textile printing is agents in the formulation of printing pastes is of para-
one of the largest exploitations of the thickening properties mount importance since they impart adequate rheological
of carbohydrate polymers. Thickeners are used in textile properties to the pastes in the different flow conditions
printing to modulate the rheological properties of the encountered in the printing process. During this process,
printing pastes during application and to obtain sharp, the materials are moved by the sequence across the mask
clean drawing patterns by preventing dye migration. Algi- and through the screen openings, to the substrate. Thus,
nates, guar gum and its derivatives, methyl and carboxy- the pastes are subjected to flows with shear or extensional
methylcellulose, some exudate gums and xanthan are component, or both, at different strain rates. High strain
excellent thickeners for this application, since they can rate conditions are present during the first steps of the
impart high viscosities at low concentrations and possess application process, while after being forced through the
the adequate rheological behavior. screen openings and deposited on the fabric, the paste will
continue to flow at very low shear rates. Hence, the thick-
ening agents must ensure at the same time both a homoge-
*
Corresponding author. Tel.: +386 22207887; fax.: +386 22207990. neous distribution of the printing paste on the screen and
E-mail address: rebeka.fi[email protected] (R. Fijan). its uniform flow through the screen openings. In other

0144-8617/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2006.08.006
 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717 709

words, the paste must be characterized by a good screena- molecular weight thickeners are suitable only for mono-
bility and a complete and uniform penetrability into the functional reactive dyes (Šostar & Schneider, 1999). To
cloth; moreover, the best sharpness of definition must be preclude chemical reaction of the guar gum with the reac-
achieved and flushing out prevented. It follows that the tive dye, an addition of an appropriate additive to the
favorable properties for an easy application and a good printing paste can alternatively reduce such undesired
performance of a printing paste are generally low viscosity interactions (Kokol, 2002; Oblonšek, Šostar-Turk, &
values at high shear rates and high viscosities at low shear Lapasin, 2003). On the basis of the screening data, using
rates (or a presence of a yield stress), respectively. non-substituted guar gum and bifunctional reactive dyes
Elasticity affects both the flow behavior of the paste in printing pastes, reasonable additives have found to be
through the screen openings to the fabric and the following ethoxylated stearyl alcohols and oleyl alcohols as well as
step (flow through the fibers) by governing possible recov- some alkylpolyglycosides and some surfactants with ami-
ery of the paste immediately after its application. The selec- no-groups (Schneider & Šostar-Turk, 2003), but in some
tion of the thickening agent, which in most cases is cases the results are still not satisfactory, since some sur-
confined to polysaccharides and their derivatives with high factants form gels with guar gum. The rheological investi-
molecular weight, is determined by the fabric to be printed, gation of interactions between polysaccharide polymer
the printing conditions, and, above all, the type of dye thickener and reactive dye in the presence of polyoxyeth-
used. Depending on their chemical structure, dyes may ylene stearylethers has shown that the breakup of the pri-
interact with thickening agents, to form complexes or to mary polymer cross-linked structure leads to stronger
give a chemical reaction, so causing a variation of the rhe- shear thinning and elasticity of the paste due to hydrogen
ological properties of the printing pastes and, hence, of and covalent ether bonds of surfactants as individual mol-
their application characteristics. However, in most cases ecules (below critical micelle concentration), or as micelle
the addition of a dye does not produce a qualitative change aggregates (above critical micelle concentration) with the
of the flow behavior, but only a limited increase in the rhe- hydroxyl groups of the thickener (Kokol, 2002). The sur-
ological quantities. factant molecules tend to accumulate at polysaccharide
Alginates and modified starches are used as thickeners, interfaces and build an oriented monomolecular film on
surface penetration and spreading controlling agents in tex- its interfaces (Kokol, Schneider, & Šostar, 1999). When
tile printing with reactive dyes. Printing operations with the surfactant addition is contained at sufficiently low lev-
this type of dye involve true chemical reactions of the dye els, only slight but different changes are observed for lin-
with primary hydroxyl groups of the cotton fibers under ear viscoelastic properties, ascribed to surfactant-induced
alkaline conditions. Therefore, the ideal thickener must intramolecular interactions and, consequently to slight
possess two essential features: its viscosity must be insensi- conformational changes of polymer chains, whereas at
tive to high pH and the polymer itself must not interact higher surfactant concentrations the intermolecular con-
with the dye. Alginate is the only hydrocolloid to have nections promoted by micelles can prevail depending on
these requisites: besides being stable to alkali it has no free micelle conformations and interactions, so leading to a
primary hydroxyl groups which can compete with cotton moderate increase in viscosity and characteristic relaxa-
for dye fixation. The net result is that less dye is required tion times (Oblonšek et al., 2003).
to obtain a predetermined color yield and a better defini- In the present research, the influence of isododecyl and
tion of the drawing (Lapasin & Pricl, 1995). cetyl polyoxyethylene ethers on the rheological properties
The application of guar gum in cellulose printing with of aqueous polysaccharide solutions is studied under
reactive dyes causes direct or indirect (via reactive dye) steady and oscillatory shear conditions, in order to evalu-
chemical covalent binding of guar gum macromolecules ate possibilities of an extended usage of a wider range of
with the cellulose substrate (Kokol, 1998). The occurred non-ionic surfactants in combination with guar gums and
chemical cross-linking of guar gum on the fiber surface their derivatives for textile printing on cellulose.
hinders washing-off and thus causes increased fabric stiff-
ness and color changes of the printed substrate. Modifica- 2. Experimental
tion of guar gum, namely substitution of free primary
hydroxyl groups with different substituents, decreases 2.1. Polysaccharides
the chemical cross-linking. On the basis of high and low
viscosity guar gum of different substitution levels, it was For the rheological investigation of polysaccharide–sur-
shown that the fabric stiffness strongly depends on the factant interactions three types of industrial polysaccha-
solid content. While high viscosity guar gum is suitable rides (Fig. 1), produced by specialized manufacturer of
for nearly all reactive dyes, it is imperative to increase natural thickeners for textile printing Lamberti S.p.a. (Ita-
the substitution level of guar gum with increased solid ly), were used, namely sodium alginate (95% active content
content. The behaviour of monofunctional reactive dyes and 5% moisture), guar gum (85% active content, 5% pro-
is more favourable than the behaviour of dyes with more teins and 10% moisture) and sodium carboxymethylated
reaction centres. High molecular weight guar gum can be guar gum (70% active content, 15% salts, 5% proteins
used for printing with most reactive dyes, whereas low and 10% moisture). Alginate is sodium salt of alginic acid
710 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717

A
COO Na+
O

OH
COO Na+ O HO O
O HO
OH
OH
O COO Na+ OH O
OH O O COO Na+ O

OH OH
O
HO COO Na+ O
O
O OH

O
OH COO Na+
G M G G M M

B C
OX CH3 CH (CH3) (CH2)9 O (CH2 CH2 O)n H
OH
CH2 O Polyoxyethylene 7 isododecyl ether: n = 7
Polyoxyethylene 11 isododecyl ether: n = 11
HO OH
O
D
HO O CH2 O CH3 (CH2)14 CH2 O (CH2 CH2 O)n H
OH
OH Polyoxyethylene 10 cetyl ether: n = 10
O CH2 O HO Polyoxyethylene 20 cetyl ether: n = 20
OX
n
Sodium carboxymethyl guar: X = CH2 COO Na+
Guar gum: X = H

Fig. 1. (A) Idealized structure of sodium alginate with G-block, M-block and MG-block (B) idealized structure of guaran, the principal of guar gum and
its derivatives, (C) monomer unit of polyoxyethylene isododecyl ethers and (D) monomer unit of polyoxyethylene cetyl ethers.

generally derived from brown seaweeds (Phaeophyceae). able for derivatization on every sugar unit in guar, so a
The alginate polymer is composed of two building blocks maximum degree of hydroxyl group substitution (DS) of
of monomeric units, namely 1–4-linked b-D-mannuronic three can be achieved. Carboxymethylated guar gum used
acid (M) and 1–4-linked a-L-guluronic acid (G) in varying in this research has DS = 0.3.
proportions and sequential arrangements, depending on An estimate of the viscosity average molecular weight
the seaweed source. These monomers can be organized in (Mv) of the polysaccharides was established, using intrinsic
blocks of consecutive G-residues (G-blocks), consecutive viscosity [g] data and the appropriate Mark–Houwink rela-
M-residues (M-blocks) or alternating M and G (MG- tionship, given as ½g ¼ KM av , where K and a are constants
blocks) (Fig. 1, A). The dyadic frequency composition of depending on the polymer–solvent pair. Since the pub-
sodium alginate thickener (FGG,FMM, FGM + MG) was lished relationships are known to be reasonably robust,
determined via circular dichroism (Donati et al., 2003) at we can assume that the weight average molecular weight,
Department of Biochemistry, Biophysics and Macromolec- M w  M v.
ular Chemistry, University of Trieste, Italy. The M/G For the determination of intrinsic viscosity the samples
ratio, determined from the dyadic frequency composition were previously purified as described in Azero and And-
(FGG = 0.15, FMM = 0.41, FGM + MG = 0.44) is 1.7, typical rade (2002) and Beer, Wood and Weisz (1999). The efflux
of alginates derived from Macrocystis pyrifera (Mancini, times of the dilute polysaccharide solutions were measured
Montanari, Peresini, & Fantozzi, 2002; Panikkar & Brasch, using the Ubbelohde capillary viscometer (k = 0–0109).
1996). The viscometer was immersed in a precision water bath
Guar gum is a galactomannan, obtained from the seed (thermostat Petrotest, Petrotest Instrument GmbH & Co.
of the legume Cyamopsis tetragonolobus. Guaran (Fig. 1, KG, Germany) to maintain constant temperature. The
B), the functional polysaccharide in guar gum is a chain results are shown in Table 1.
of (1 fi 4)-linked b-D-mannopyranosyl units with single
a-D-galactopyranosyl units connected by (1 fi 6) linkages 2.2. Surfactants
to, on the average, every second main chain unit (Whistler
& BeMiller, 1993). Carboxymethylated guar gum is guar Two branched non-ionic polyoxyethylene surfactants
ether, generally prepared at elevated temperature under (Fig. 1C) produced by Lamberti S.p.a. (Italy), namely
aqueous conditions using strong NaOH and monochloro- TR 8 (polyoxyethylene 7 isododecyl ether) and TR 12
acetic acid. An average of three hydroxyl groups is avail- (polyoxyethylene 11 isododecyl ether); and two linear
 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717 711

Table 1
Intrinsic viscosity, g, and estimated weight average molecular weight, Mw, of sodium alginate, guar gum and sodium carboxymethylated guar gum
Sample Conditions Reference K · 106 a [g] Mw
(dL/g) (g/mol)
Sodium alginate 0.1 M NaCl, 25 C K and a values are derived 122.8 0.963 5.78 7.11 · 104
from experiments of
Donnan and Rose (1950)
Guar gum aqueous solution, 25 C Robinson, Ross-Murphy, and Morris (1982) 380 0.723 10.61 1.41 · 106
Sodium carboxymethyl 0.1 M NaCl, 25 C Risica, Dentini, and Crescenzi (2005) 193 0.739 7.12 1.51 · 106
guar gum (DS = 0.3)a
a
For carboxymethylated guar gum (DS = 0.3) Mark–Houwink constants are not available in the literature, therefore the weight average molecular
weight was estimated using the Mark–Houwink parameters for methyl guar gum (DS = 0.3) determined by Risica et al. (2005).

non-ionic polyoxyethylene surfactants (Fig. 1D) produced to rheological properties on the polymer solution depends
by Sigma–Aldrich (Germany), namely Brij(R) 56 (polyoxy- on surfactant concentration and is most evident in G sys-
ethylene 10 cetyl ether) and Brij(R) 58 (polyoxyethylene 20 tems where phase separation due to polymer–surfactant
cetyl ether), were used. interactions appears at higher surfactant concentrations.
Such behaviour could be ascribed to the saturation of the
2.3. Sample preparation polymer surface with surfactant molecules. Stronger inter-
actions between hydrophilic ethylene oxy (EO) units from
All aqueous systems were prepared by adding a proper the surfactant and primary hydroxyl groups of the G poly-
amount of polymer (marks: A (sodium alginate), G (guar), mer could cause the orientation of the hydrophobic part of
CG (sodium carboxymethyl guar)) to a corresponding the surfactant toward the aqueous medium. Such polymer-
solution of single surfactant under stirring conditions. surfactant complexes become hydrophobic on the surface
The surfactants, marked as S7 (TR 8), S11 (TR 12), S10 and consequently separation into two phases appears.
(Brij(R) 56) and S20 (Brij(R) 58), were dissolved in distilled Looking at the typical composition of guar gum thick-
water with slight heating, prepared at three different con- eners they are generally contaminated with 5–6% of impu-
centrations for each polymer system (w = 1%, 3% and rities such as proteins (Whistler & BeMiller, 1993), which
5%) and the same polysaccharide concentration (4% for could be considered as an indication to strong interactions
A and CG solutions and 1% for G systems). All the systems between surfactant and proteins in G-surfactant systems. It
were left in the refrigerator to swell overnight. A phase sep- is known that when ionic charged surfactants or micellar
aration was observed for G-S7 (wS7 = 3% and 5%), G-S11 systems are added to protein solutions, generally the pro-
(wS11 = 5%), G-S10 (wS10 = 5%) and for G-S20 (wS20 = 3% tein is denatured or biologically inactivated as a result of
and 5%), whereas slight inhomogeneity and turbidity was binding. Nonetheless, it has been found that non-ionic
displayed by CG-S7 (wS7 = 3% and 5%). (non-charged) surfactants do not bind to these biomole-
cules to a significant extent and do not denaturate proteins
2.4. Apparatus (Helenius & Simons, 1972; Makino, Reynolds, & Tanford,
1973; Sahito, 1987). Moreover, taking this into consider-
Rheological measurements (flow curves, stress sweep ation, it must also be emphasized that substituted guar,
and frequency sweep) were carried out at 25 C and per- used in this research, also contains 5% of protein impurities
formed with a rotational controlled stress rheometer (as provided by the producer), but the rheological effects
Haake RS150 with different measuring devices: cone and produced by the addition of non-ionic surfactants in CG
plate (C60/1) and parallel plates with serrated surfaces systems, do not seem to be quantitatively important (see
(PP35 Ti). The latter device was used to investigate flow Fig. 4) and indicate that the interactions between the sur-
properties and linear viscoelastic behavior of the poly- factant and protein contents are not so strong, thus sup-
mer-rich phase of the G systems, where phase separation posed of secondary importance.
was evident, and also of inhomogeneous CG systems to Oblonšek et al. (2003) investigated changes in the rheo-
avoid wall slippage effects. logical properties after the elimination of proteins from
non-substituted guar, with purification by extraction
3. Results and discussion method, carried out by the producer. The application of
the purified guar to reactive printing using standard print-
3.1. Steady shear properties ing paste recipe, prepared with the thickener and addition
of non-ionic surfactant, bifunctional reactive dye, urea,
In most cases the results obtained from steady shear sodium carbonate and reduction substance, showed that
tests indicate the shear-thinning properties peculiar to printing results on viscose were still not satisfactory,
ordinary polymer solutions. The surfactant contribution due to highly elastic printing pastes resulting from an
712 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717

entanglement of the polymer with other components of the molecules participate in the polymer intermolecular associ-
printing paste. ations, binding via hydrogen bonds with their hydrophilic
In the present research, the removal of soluble proteins units towards the guar gum hydroxide groups, giving origin
from guar with purification by precipitation in ethyl alco- to the formation of polymer–surfactant complexes, so
hol as described in Beer et al. (1999) and Azero and And- replacing the intermolecular interactions and associations
rade (2002), was performed. As in the cases observed for promoted by hydrogen bonds between individual polymer
unpurified guar the addition of surfactant to aqueous solu- chains. The application of higher shear stresses causes the
tions of purified guar likewise leads to phase separation. To break-up of such complexes, the consecutive uncoiling of
summarize the previous observations, it can be concluded polysaccharide chains and decrease in viscosity.
that strong interactions in G-surfactant systems, where Comparing flow curves of the G polymer-rich phases
phase separation appears, are primarily governed by (Fig. 3) makes evident that their properties result from
hydrogen bonding between G and surfactant, as discussed the combination of several condition factors, such as guar
further. gum concentration, surfactant concentration, EO chain
The investigation of the polymer-rich phases show a length and hydrophobic group as well. In the particular
dramatic increase of viscosity (Fig. 2) and, corresponding- case of G-S10 and G-S20 polymer-rich phases with 5% sur-
ly, much longer relaxation times, as it will be deduced later factant addition, it can be seen that the contribution of sur-
from the analysis of the mechanical spectra reported in factant to the properties of the polymer-rich phases
Fig. 6. depends on the number of EO groups in the surfactant
As shown in Fig. 2, the flow curves of G-S7 systems molecule, causing a more concentrated phase and, conse-
(black symbols) are compared with those of the corre- quently, higher viscosity values at higher EO number.
sponding pure G aqueous solutions (open symbols) having The properties of systems G(w=1%)–S10(w = 5%)-conc. phase
the same polysaccharide concentration and it can be seen and G(w = 1%)–S20(w = 3%)-conc. phase coincide in the linear
that the asymptotic values of the viscosity at zero shear regime at low shear stresses, whilst at higher shear stresses
stress nearly coincide. the profile shape of complex G(w=1%)–S10(w = 5%)-conc.
Looking at the guar solutions (Fig. 2, open symbols), an phase shows an anticipated viscosity drop, indicating the
appreciable increase in the low shear viscosity of almost influence of surfactant structure and concentration.
three orders of magnitude is observed passing from 1% to Conversely, in the case of the systems prepared with
3% polymer concentration. Increasing polymer concentra- branched non-ionic surfactants, an increase in the number
tion leads not only to an increase of the topological con- of EO groups leads to lower viscosity values and anticipat-
straints posed to each individual chain by the temporary ed viscosity drop (compare the polymer-rich phases of
polymer network but also enhances the role of hydroxyl G(w = 1%)–S11(w = 5%) and G(w = 1%)–S7(w = 5%)).
groups and the resulting hydrogen-bond intermolecular A satisfactory fitting of viscosity data is provided with
interactions. the Cross equation (Cross, 1965), using shear rate as the
It can be seen from Fig. 2 that the flow curves of G-S7 independent variable:
polymer-rich phases show an anticipated viscosity drop in g  g1
the middle shear region with respect to the corresponding g ¼ g1 þ 0 n ð1Þ
1 þ ðk_cÞ
pure G aqueous systems. It is obvious that the surfactant
where g0 and g1 are the asymptotic values of the viscosity
at zero and infinite shear rates, respectively, k is the charac-
10000 teristic time and n rules the shear dependence in the power
law region.
1000

100 100000
h (Pa s)

10000
10
1000
η (Pa s)

1 100

10
0,1
1

0,01 0,1
1 10 100 1000
0,01 0,1 1 10 100 1000 σ (Pa)
s (Pa)
Fig. 3. Steady shear viscosity vs. shear stress for polymer rich phases (s,
Fig. 2. Steady shear viscosity vs. shear stress for G, G-S7 (d, G(w = 1%)– G(w = 1%)–S7(w = 3%)-conc. phase; d, G(w=1%)–S7(w = 5%)-conc. phase; j,
S7(w=1%); s, G(w=1%); ¤, G(w=1%)–S7(w=3%)-conc. phase; }, G(w = 2%); m, G(w = 1%)–S11(w = 5%)-conc. phase; m, G(w=1%)–S10(w = 5%)-conc. phase; },
G(w = 1%)–S7(w = 5%)-conc. phase and n, G(w=3%)). G(w=1%)–S20(w = 3%)-conc. phase and ¤, G(w=1%)–S20(w = 5%)-conc. phase).
 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717 713

Since the concentrated G systems, that result after phase the average percentage deviation (APD):
separation, due to strong interactions between guar and X N  

surfactant, show a gradual fall-off in the viscosity before APD ¼ 1  gi;cal   100 ð3bÞ
 gi;exp  N
the precipitous drop, a quite satisfactory data fitting can i¼1
also be obtained with the Roberts–Barnes–Carew (RBC) and number of runs is defined as:
model (Roberts, Barnes, & Carew, 2001). The model repre-
N 
X 

sents a modified version of the Ellis model which is more
no:runs ¼ signðgi;cal  gi;exp Þ  signðgi1;cal  gi1;exp Þ
flexible in virtue of the increased number of adjustable  2 
i¼2
parameters. The previous parameters g0 and g1 are rede-
ð3cÞ
fined so that they become functions (g00 and g01 Þ of the
shear stress, r, as follows: In order to examine surfactant contribution, viscosity
g00 g01 data (calculated with the Cross equation) of the
g ¼ g01 þ m ð2aÞ polymer–surfactant systems were compared with poly-
1 þ ðr=rc Þ
mer aqueous systems with no surfactant addition
where (Fig. 4).
g0 In the case of CG-S10 systems (triangles) the surfactant
g00 ¼ ð2bÞ gradually reduces the viscosity of the polysaccharide
1 þ ðr=r1 Þp
s matrix over the whole shear region. The effects of poly-
g01 ¼ g1 ð1 þ ðr=r2 Þ Þ ð2cÞ
mer-surfactant interactions in the case of CG systems add-
and rc is a critical shear stress. The number of adjustable ed with S7 (3% or 5%, circles) are not quantitative
parameters can be reduced, setting s equal to 1,5. Even important but appear evident only under steady shear con-
if the RBC model provides adequate data fitting also for ditions where inhomogeneous flow is observed in the non-
some G-surfactant solutions, the Cross equation appears linear region and the behaviour becomes boundary
to be more suitable and convenient to describe the shear dependent. A viscosity reduction is evident above 1 Pa,
thinning behavior of these systems. Table 2 shows a com- especially for the systems with higher surfactant concentra-
parison of the model parameters and the quality of the fit- tion, which indicates that the different effects produced by
ting with the Cross and RBC models, respectively. The the surfactant addition are strictly related to the different
objective function (O.F.) used in the minimization proce- structural features of the polymer matrix under different
dure is given as: shear conditions. Above the critical stress the polymer–sur-
!2 factant interactions are substantially broken setting a sig-
X N
gi;cal nificant fraction of polysaccharide molecules free in
O:F: ¼ 1 ð3aÞ
i¼1
gi;exp solution.

Table 2
Model parameters of cross and RBC model and fitting quality parameters for the concentrated systems G-S7, G-S11, G-S10 and G-S20 after phase
separation
G-S7 conc. phase. G-S11 conc. phase G-S10 conc. phase G-S20 conc. phase
(wS7 = 3%) (wS7 = 5%) (wS11 = 5%) (wS10 = 5%) (wS20 = 3%) (wS20 = 5%)
Cross-model
g0(Pa s) 494.9 8890 6740 1848 1646 26526
g1(Pa s) 0.043 0.203 0.487 0.176 0.055 0.066
k (s) 18.3 72.6 93.9 40.4 33.4 69.1
n 0.870 0.943 0.972 0.980 0.890 1.034
O.F. 0.0457 0.0447 0.0157 0.5427 0.7357 0.6309
APD 3.47 4.49 4.06 18.73 16.96 11.63
No. runs 5 4 6 4 6 8
RBC model
g00 (Pa s) 1023 8414 6064 1539 1465 26098
g01 (Pa s) 0.001 0.002 0.001 0.256 0.001 0.121
rc(Pa) 0.623 56.8 37.0 29.5 29.6 174.4
m 0.339 2.06 2.48 11.79 4.62 3.60
r1(Pa) 24.5 157.9 76.6 1909.2 1909.2 275.8
p 5.48 29.5 23.9 19.8 19.8 61.0
r2(Pa) 7293 35616 30844 181 6405 689
s 1.5 1.5 1.5 1.5 1.5 1.5
O.F. 0.8976 0.0281 0.0161 1.9229 1.4846 0.4335
APD 14.50 2.95 5.70 9.25 12.80 5.29
No. runs 6 7 8 6 6 8
714 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717

1000 10000

100 1000

G r', G r'' (Pa)

10 100

G r', G r'' (Pa)

1 10

0.1 1

Fig. 4. Ratios between Cross viscosities of A-S7 and A systems; CG-S7

and CG systems; CG-S10 and CG systems at the same polysaccharide 0.01 0.1
content and at different surfactant concentration (A-S7: wS7 = 1% (h), 3% 0.01 0.1 1 10 100 1000 10000
( ), 5% (j); CG-S7: wS7 = 1% (s), 3% ( ), 5% (d); CG-S10: wS10 = 1% ω r (rat s-1)
(n), 3% ( ), 5% (m)).
Fig. 5. Relative master curves of the mechanical spectra of A (d, G 0 ; s,
G00 ), G(j, G 0 ; h, G00 ) and CG(m, G 0 ; n, G00 ) systems at different surfactant
additions, obtained by horizontal shift along the frequency axis; reduced
The alginate–surfactant systems (Fig. 4, squares) show frequency (xr) is identical to the applied frequency for the single
slight decrease of the steady shear viscosity due to surfac- polysaccharide solutions without surfactant.
tant addition over the whole shear region. These effects
can be ascribed to reduction in the connectivity of the poly-
complexes leads to a significant increase of both moduli
meric matrix induced by the entrapped surfactant micelles,
and relaxation times as well, as it can be evinced by double
hindering intermolecular interactions and associations
shifting of the crossover condition towards higher values
between the polysaccharide chains.
and lower frequencies.
The experimental data from oscillatory tests, namely fre-
3.2. Linear viscoelastic properties
quency sweep, can be described quite satisfactorily with the
Friedrich–Braun model (Friedrich & Braun, 1992), based
From the preliminary step in the analysis of the oscilla-
on fractional derivatives, as well as with the classical gener-
tory flow data, namely stress sweep tests, the limit of the
alized Maxwell model (composed of 4 Maxwell elements in
viscoelastic regime was determined. The Soskey–Winter
parallel). The following relationships can be derived from
equation is used for the description of viscoelastic moduli
the Friedrich–Braun fractional model in order to describe
vs. deformation (Soskey & Winter, 1984):
the frequency dependence of the storage G 0 and loss G00
G 1 moduli, respectively:
 ¼ ð4Þ
G0 1 þ bcn
where G* is the complex modulus, G0 is the complex mod-
ulus in the linear viscoelastic regime, c is the strain and b 10000
and n are adjustable parameters. The critical strain value
c0c marking the upper limit of the linear viscoelastic regime
1000
is arbitrarily set in correspondence with G =G0 ¼ 0:95.
For CG and A systems the critical strain values are typ-
ical of ordinary polymer solutions, lying in the ranges 0.2– 100
G ', G '' (Pa)

0.4 and 0.6–0.8, respectively, while c0c varies between 0.01

and 0.07 for the polymer-rich phases (G systems), so indi-
cating a proximity to weak gel condition. 1
The patterns of the mechanical spectra of A-surfactant
and CG-surfactant systems are typical of ordinary polymer
10
solutions and do not change substantially with increasing
surfactant concentration, so that, a generalized representa-
tion can be easily derived by superposing the individual 0,1
profiles through simple shifting procedure (Fig. 5). 0,01 0,1 1 10 100 1000
Slight deviations from the ordinary profiles are observed ω (rad s–1)
for G systems (polymer-rich phases after phase separation), Fig. 6. Mechanical spectra for G and G-S20 systems (G(w = 1%) (r, G 0 ; },
reasonably due to interactions between surfactant and guar G00 ); G(w = 1%)–S20(w = 1%) (j, G 0 ; h, G00 ); G(w = 1%)–S20(w = 3%)-conc.
molecules (Fig. 6). The formation of polymer–surfactant phase (m, G 0 , n, G00 ); G(w = 1%)–S20(w = 5%)-conc. phase (d, G 0 ; s, G00 )).
 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717 715

d c 
0 ðkxÞ cos p2 d þ ðkxÞ cos p2 ðd  cÞ
G ðxÞ ¼ DG ð5aÞ
1 þ 2ðkxÞc cos p2 c þ ðkxÞ2c
d c 
ðkxÞ sin p2 d þ ðkxÞ sin p2 ðd  cÞ
G00 ðxÞ ¼ DG ð5bÞ
1 þ 2ðkxÞc cos p2 c þ ðkxÞ2c
where k is a characteristic time, c and d are related to the
derivation orders of the differential operators, and DG rules
the magnitude of both moduli. The Friedrich–Braun model
parameters obtained from data fitting for A-S11 systems
(Fig. 7) show no significant discrepancies. This indicates
that no physical interactions via hydrogen bonds between
alginate and surfactant appear in these systems, being the
surfactant micelles hosted inside the polysaccharide
network.
The constitutive equation for the corresponding general-
ized Maxwell fluid is obtained by summing the contribu-
tion sk of each of the N Maxwell elements: Fig. 8. Relaxation spectra of G-S7, G, A and CG systems (G(w = 1%)–S7:
wS7 = 1% (s), 3%-conc. phase , 5%-conc. phase (d); G(w = 1%) (m);
X
N A(w = 4%) (r), CG(w = 4%) (j)).
s¼ sk ð6aÞ
k¼i

where each contribution is given by the constitutive equa- contribution of higher relaxation times, due to a more com-
tion of the Maxwell element: plex and strongly associated structure produced by the ef-
osk fect of higher concentrations of surfactant addition and
s k þ kk ¼ Gk c_ ð6bÞ higher contribution of thickener in the polymer-rich
otk
phases.
Accordingly, the model yields the following expressions for Comparing experimental data obtained under continu-
the viscoelastic quantities under small-amplitude oscillato- ous and oscillatory flow conditions, it can be seen that in
ry motion (storage G 0 and loss G 0 0 moduli, respectively): the case of pure G and CG solutions the curve of g*(x)
X
N
Gk k2k x2 essentially coincides with that of g(_c) over nearly two dec-
G0 ¼ 2
ð6cÞ ades of x and c_ (Fig. 9, squares and diamonds). The empir-
k¼1 1 þ ðkk xÞ
ical Cox–Merz rule (Cox & Merz, 1958) states that the
X
N
Gk kk x shear-rate dependence of the steady-state viscosity, g(c·)
G00 ¼ ð6dÞ is equivalent to the frequency dependence of the complex
k¼1 1 þ ðkk xÞ2
viscosity, g*(x):
From the results of data fitting using the generalized Max- qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2
well model expressed through the corresponding relaxation gð_cÞ ¼ jg ðxÞjx¼_c ¼ ½ðG0 =xÞ þ ðG00 =xÞ x¼_c ð7Þ
time distribution Gk (kk) (Fig. 8), canonical profiles can be
recognized for relaxation spectra of A and CG systems, but where G 0 and G00 are the storage and loss modulus,
the profiles of the spectra for G systems show increasing respectively. Discrepancies between the surfactant effects
on steady and oscillatory shear properties are underlined
by the failure of the rule for G-surfactant systems
(Fig. 9, triangles) or, as in the case of CG-surfactant sys-
tems, a divergence between the curves are noticed as
both x and c_ increase. In the case of A systems discern-
ible deviations from the prediction of the Cox-Merz rule
appear (Fig. 9, circles). The functional dependence of
g*(_c) is substantially different from that of g*(x), and
steady and oscillatory shear quantities show only slight
tendency to converge.

Conclusions

The rheological investigation of the influence of four

non-ionic surfactants on different polysaccharides in gener-
Fig. 7. Concentration dependence of the Friedrich–Braun model param- al shows marked shear thinning and viscoelastic character
eters for A-S11 systems. of the systems.
716 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717

Fig. 9. Cox–Merz rule for A (s, g*; d, g), CG (}, g*; r, g); G (h, g*; j, g) G(w = 1%)–S7(w=3%)-conc. phase (n, g*; m, g) and G(w=1%)–S11(w = 3%) ( : g*,
: g).

Since alginate polymer has no free primary hydroxyl Beer, M. U., Wood, P. J., & Weisz, J. (1999). A simple and rapid method
groups, the slight changes in rheological properties are for evaluation of Mark–Houwink–Sacurada constants of linear
random coil polysaccharides using molecular weight and intrinsic
the result of reduction in the connectivity of the cross- viscosity determined by high performance size exclusion chromatog-
linked polysaccharide matrix induced by incorporation of raphy: application to guar galactomannan. Carbohydrate Polymers,
surfactant micelles that become entrapped in the associa- 39(4), 377–380.
tive structure of the polymer. Cox, W. P., & Merz, E. H. (1958). Correlation of dynamic and steady-flow
The different effects produced by the surfactant addition viscosities. Journal of Polymer Science, 28, 619–622.
Cross, M. M. (1965). Rheology of non-Newtonian fluids: a new flow
in the case of carboxymethyl guar systems are strictly relat- equation for pseudoplastic systems. Journal of Colloid Science, 20(5),
ed to the different structural features of the polysaccharide 417–423.
matrix under different shear conditions, but in general the Donnan, F. G., & Rose, R. C. (1950). Osmotic pressure, molecular weight,
rheological behaviour is governed by polymer content. and viscosity of sodium alginate. Canadian Journal of Research,
To avoid possible interactions via hydrogen bonds between 28(Sect. B), 105–113.
Donati, I., Gamini, A., Skjåk-Bræk, G., Vetere, A., Campa, C., Coslovi,
carboxymethyl guar and non-ionic surfactant, a higher A., et al. (2003). Determination of the diadic composition of alginate
substitution degree of free primary hydroxyl groups in by means of circular dichroism: a fast and accurate improved method.
carboxymethyl guar is recommendable for its usage as Carbohydrate Research, 338(10), 1139–1142.
thickening agent in reactive printing. Friedrich, C., & Braun, H. (1992). Generalized Cole-Cole behavior and its
Significant effects, ascribed to the physical interactions rheological relevance. Rheologica Acta, 31(4), 309–322.
Helenius, A., & Simons, K. (1972). The binding of detergents to lipophilic
(via hydrogen bonds) and associations between the poly- and hydrophilic proteins. The Journal of Biological Chemistry, 247(11),
mer chains induced by surfactant micelles, lead to a seques- 3656–3661.
tration of the guar gum polymer in a ‘gel-like’ phase with a Kokol, V. (1998). The consequence of interactions between guar thickener
particularly extended and strong associated structure, char- and different types of reactive dyes. Master’s Thesis, Faculty of
acterised with high viscosity, low critical strain value and Mechanical Engineering, University of Maribor, Slovenia.
Kokol, V. (2002). Interactions between polysaccharide polymer thick-
higher relaxation times. The possible usage of the investi- ener and bifunctional bireactive dye in the presence of non-ionic
gated non-ionic surfactants in non-substituted guar gum surfactants. Part 1: surface tension and rheological behaviour of
based reactive printing pastes is limited to lower surfactant different polysaccharide solutions. Carbohydrate Polymers, 50(3),
content. 227–236.
Kokol, V., Schneider, R., & Šostar, S. (1999). The influence of surfactants
Acknowledgements on rheological properties of polysaccharide solutions. In: Proceedings:
Southern Europe conference on rheology (pp. 182–187). Calabria, Italy,
Eurorheo 99-3; 7–11th September.
The authors thank Lamberti S.p.a. for supplying the Lapasin, R., & Pricl, S. (1995). Rheology of industrial polysaccharides:
thickeners and surfactants used in this research. The Theory and applications (1st ed.). London: Blackie Academic &
authors are grateful to Dr. Ivan Donati (Department of Professional.
Biochemistry, Biophysics and Macromolecular Chemistry, Makino, S., Reynolds, J. A., & Tanford, C. (1973). The binding of
deoxycholate and Triton X-100 to proteins. The Journal of Biological
University of Trieste, Italy) for performing dyadic frequen-
Chemistry, 248(14), 4926–4932.
cy composition via circular dichroism. Mancini, F., Montanari, L., Peresini, D., & Fantozzi, P. (2002). Influence
of alginate concentration and molecular weight on functional prop-
References erties of mayonnaise. Lebensmittel-Wissenschaft und-Technologie,
35(6), 517–525.
Azero, E. G., & Andrade, C. T. (2002). Testing procedures for Oblonšek, M., Šostar-Turk, S., & Lapasin, R. (2003). Rheological studies
galactomannan purification. Polymer Testing, 21(5), 551–556. of concentrated guar gum. Rheologica Acta, 42(6), 491–499.
 R. Fijan et al. / Carbohydrate Polymers 68 (2007) 708–717 717

Panikkar, R., & Brasch, D. J. (1996). Composition and block structure of Auxiliaries, Textile Chemical Technologies, Textile Materials and Pollu-
alginates from New Zealand brown seaweeds. Carbohydrate Research, tion, Eco Printing, Technological Waters and Waste, Spotting, and Textile
293(1), 119–132. and Garment Care. She is head of Textile Printing and Textile Care
Risica, D., Dentini, M., & Crescenzi, V. (2005). Guar gum methyl ethers. Laboratory and Textile Care Centre and member of Society of Colorists of
Part I. Synthesis and macromolecular characterization. Polymer, Slovenia, of The Society of Engineers and Textile Technicians of Slovenia,
46(26), 1247–12255. of The Society of Rheology, of The Slovenian Society of Rheology, of
Roberts, G. P., Barnes, H. A., & Carew, P. (2001). Modelling the flow ICTC (International Technical Committee for Textile Care), of SDC (The
behaviour of very shear-thinning liquids. Chemical Engineering Society of Dyers and Colorists), of IWA (International Water Associa-
Science, 56(19), 5617–5623. tion), of EWA (European Water Association) and SDZV (The Slovenian
Robinson, G., Ross-Murphy, S. B., & Morris, E. R. (1982). Viscosity- Association for Water Protection). In collaboration with Forschungin-
molecular weight relationships, intrinsic viscosity, and dynamic stitut Hohenstein, Germany she developed a new study course at profes-
solution properties of guar galactomannan. Carbohydrate Research, sional higher study progamme Textile: Textile Care and has participated
107(1), 17–32. at research tasks in the fields of finishing of textiles, textile and garment
Sahito, S. (1987). Polymer–surfactant interactions. In M. J. Schick (Ed.), care, in basic and applicative research, development and applicative pro-
Surfactant science series. Nonionic surfactants physical chemistry (Vol. jects in Republic of Slovenia and in the European Union.
23). New York: Marcel Dekker. Her bibliography contains more than 300 bibliographic items, where she is
Schneider, R., & Šostar-Turk, S. (2003). Good quality printing with author or co-author. Her research interests include: the technology of
reactive dyes using guar gum and biodegradable additives. Dyes and textile printing, inkjet printing, rheology of polysaccharides, textile aux-
Pigments, 57(1), 7–14. iliaries, textile care, and laundry wastewater treatment.
Soskey, P. R., & Winter, H. H. (1984). Large step shear strain experiments
with parallel disk rotational rheometers. Journal of Rheology, 28(5),
625–645. Romano Lapasin graduated in Chemical Engineering at the University of
Šostar, S., & Schneider, R. (1999). A study of fabric stiffness with guar Trieste, Italy, in 1971. He is presently professor of Rheology and lecturer
gum in reactive printing. Dyes and Pigments, 41(3), 167–175. of Transport Phenomena at the Faculty of Engineering, University of
Whistler, R. L., & BeMiller, J. N. (1993). Industrial gums: Polysaccharides Trieste. Other courses held since 1975 concerned Separation Processes,
and their derivatives (3rd ed.). San Diego: Academic Press. Thermodynamics for Chemical Engineering, Measurements of Thermo-
dynamical and Transport Properties, Principles of Chemical Engineering.
He is presently Dean of the Course in Chemical Engineering, and Head of
Rebeka Fijan graduated in Chemical Engineering at the Faculty of the Rheology and Polymer Laboratory (DICAMP), member of the British
Chemistry and Chemical Engineering, University of Maribor, Slovenia, in Society of Rheology, of the Groupe Francaise de Rheology, of the Society
2002. She is currently working towards a Ph.D. degree in the field of textile of Rheology (USA), of GRICU (Gruppo Ricercatori di Ingegneria Chi-
technology. Her research includes investigation of purification of reactive mica dell‘Università) and honorary member of AITIVA. He was President
printing pastes based on natural polysaccharides. of Rheotech, past President and International delegate of the Italian
Society of Rheology.
Sonja Šostar-Turk received the B.Sc. degree in Textile Technology from He is author or co-author of more than 300 papers and congress com-
the Faculty of Natural Sciences and Technology, University of Ljubljana, munications, and of the book ‘‘Rheology of Industrial Polysaccharides;
Slovenia, in 1984, M.Sc. degree in textile technology from Technical Theory and Applications’’ edited by Chapman & Hall.
Faculty of University of Maribor, Department of Textile Chemistry, in At present, his research interests are located in the field of transport
1990 and Ph.D. degree from the Faculty of Mechanical Engineering, properties and mainly concern: (a) rheology of polysaccharide systems,
University of Maribor, Department of Textile Chemistry, in 1995. She is and other structured fluids, such as concentrated emulsions and colloidal
an Associate Professor for Textile Chemistry at the University of Maribor, dispersions in polymeric matrices, as well as its application to the analysis
Faculty of Mechanical Engineering, Department of textiles and lecturer of and solution of industrial problems, (b) study and modelling of the phe-
Technology of Printing, Finishing Agents, Textile Printing, Textile nomena involved in release processes from pharmaceutical systems.