Авторы

  • Азизбекх Ёлдошев
    Denau Institute of Entrepreneurship and Pedagogy
  • Абдулла Дурсоатов
    Denau Institute of Entrepreneurship and Pedagogy
  • Шохир Турсунов
    Denau Institute of Entrepreneurship and Pedagogy

DOI:

https://doi.org/10.71337/inlibrary.uz.imjrd.113739

Аннотация

 Complex coacervate hydrogels formed by sodium carboxymethyl cellulose (Na-CMC) and chitosan were prepared and analyzed for rheological behavior over broad ranges of concentration (1–4% w/v), pH (7–18), and temperature (30–60 °C). Viscosity measurements under steady shear revealed pronounced non-Newtonian (shear-thinning) flow behavior for all samples, with higher polymer concentrations yielding dramatically increased zero-shear viscosity and stronger shear-thinning character. Oscillatory tests showed that low concentrations (∼1%) behaved as viscous liquids (loss modulus G'' exceeding storage modulus G'), whereas at 4% a weak gel-like network formed with G' approaching or exceeding G'' at higher frequencies. The rheological properties were highly sensitive to pH: near-neutral pH enabled electrostatic intermolecular associations between the anionic Na-CMC and cationic chitosan, resulting in maximum viscosity and viscoelastic moduli. Increasing pH beyond ~8 caused a sharp drop in viscosity and G', as chitosan became deprotonated and the polyelectrolyte complex disintegrated, leaving a solution with properties approaching those of Na-CMC alone. Temperature elevations from 30 to 60 °C reduced solution viscosity by roughly 2–3 and slightly decreased viscoelastic moduli, consistent with thermal disruption of polymer interactions and decreased solvent viscosity. These results illustrate how polymer concentration, pH, and temperature co-modulate the flow and viscoelastic properties of Na-CMC/chitosan complexes. Mechanistically, the strong pH-responsiveness is attributed to the polyelectrolyte nature of the constituents: only within a certain pH window are both polymers sufficiently charged to form extended ionically-crosslinked networks. The findings deepen our understanding of biopolymer complexation and provide quantitative guidance for tuning the rheology of Na-CMC/chitosan systems in applications ranging from injectable hydrogels and controlled drug release matrices to sustainable thickening agents.


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

76

RHEOLOGICAL PROPERTIES BASED OF NA-CMC AND CHITOSAN

POLYELECTROLYTE COMPLEXES

Yoldoshev Azizbekh Abduganiyevich

1

, Dursoatov Abdulla Chori ugli

1

, Tursunov Shokhir

Toirovich

1

1*- Teacher of the Denau Institute of Entrepreneurship and Pedagogy

y.azizbek0711@gmail.com

dursoatovabdulla53@gmail.com

shokirtursunov04@gmail.com

.

Abstract:

Complex coacervate hydrogels formed by sodium carboxymethyl cellulose (Na-CMC)

and chitosan were prepared and analyzed for rheological behavior over broad ranges of

concentration (1–4% w/v), pH (7–18), and temperature (30–60 °C). Viscosity measurements

under steady shear revealed pronounced non-Newtonian (shear-thinning) flow behavior for all

samples, with higher polymer concentrations yielding dramatically increased zero-shear viscosity

and stronger shear-thinning character. Oscillatory tests showed that low concentrations (

1%)

behaved as viscous liquids (loss modulus

G''

exceeding storage modulus

G'

), whereas at 4% a

weak gel-like network formed with

G'

approaching or exceeding

G''

at higher frequencies. The

rheological properties were highly sensitive to pH: near-neutral pH enabled electrostatic

intermolecular associations between the anionic Na-CMC and cationic chitosan, resulting in

maximum viscosity and viscoelastic moduli. Increasing pH beyond ~8 caused a sharp drop in

viscosity and

G'

, as chitosan became deprotonated and the polyelectrolyte complex disintegrated,

leaving a solution with properties approaching those of Na-CMC alone. Temperature elevations

from 30 to 60 °C reduced solution viscosity by roughly 2–3 and slightly decreased viscoelastic

moduli, consistent with thermal disruption of polymer interactions and decreased solvent viscosity.

These results illustrate how polymer concentration, pH, and temperature co-modulate the flow

and viscoelastic properties of Na-CMC/chitosan complexes. Mechanistically, the strong pH-

responsiveness is attributed to the polyelectrolyte nature of the constituents: only within a certain

pH window are both polymers sufficiently charged to form extended ionically-crosslinked

networks. The findings deepen our understanding of biopolymer complexation and provide

quantitative guidance for tuning the rheology of Na-CMC/chitosan systems in applications

ranging from injectable hydrogels and controlled drug release matrices to sustainable thickening

agents.

Key words:

Na-CMC, hydrogels, chitosan, polyelectrolete complexe, biopolymer, rheometr.

Introduction

Hydrogels and viscous solutions formed by oppositely charged biopolymers have garnered

significant interest for applications in food, biomedical, and pharmaceutical fields[1]. Sodium

carboxymethyl cellulose (Na-CMC) is a water-soluble anionic cellulose derivative known for its

biocompatibility, abundance, and efficient thickening ability[3]. Chitosan, derived by partial

deacetylation of chitin, is a cationic polysaccharide that is soluble in mildly acidic aqueous media

and valued for its biodegradability and bioactivity[3]. When mixed together, anionic Na-CMC

and cationic chitosan can form polyelectrolyte complexes (PECs) via electrostatic attraction

between their oppositely charged groups[3]. Such Na-CMC/chitosan complexes combine the

favorable properties of both polymers and have been explored in applications including drug

delivery (e.g. microcapsule shells), tissue engineering scaffolds, and sustainable food packaging


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

77

films[3]. Notably, the complexation process and resulting material properties are highly sensitive

to environmental conditions – especially polymer concentration, solution pH, and temperature –

which alter the balance of ionic interactions and polymer conformations[1]. A systematic

understanding of how these parameters affect the rheological (flow and viscoelastic) properties of

Na-CMC/chitosan complexes is important for optimizing their performance in practical

applications.
Both Na-CMC and chitosan are polyelectrolytes with pH-dependent charge states. Na-CMC

carries negatively charged carboxylate groups above its p

K_a

(~3.6), remaining highly ionized

and water-soluble at neutral and alkaline pH[1]. Chitosan, a weak polybase with p

K_a

~6.5 for its

amino groups, is predominantly protonated (positively charged) and soluble in acidic to near-

neutral conditions, but loses charge and precipitates above pH ~7[1]. Consequently, coacervation

or gelation of Na-CMC with chitosan is expected only within a moderate pH window where both

polymers carry charge of opposite sign. Indeed, recent studies have shown that chitosan/CMC

PEC formation occurs in roughly the pH 6–8 range, whereas more alkaline conditions (> pH 8)

abolish complexation as chitosan becomes deprotonated and unable to bind anionic CMC[1]. On

the acidic side, if pH is too low (approaching CMC’s p

K_a

), CMC itself loses charge and

solubility[1], which can also hinder complex formation. Thus, the interplay of chitosan’s and

CMC’s charge states gives the Na-CMC/chitosan complexes an inherent pH-responsive character.

Harnessing this property could enable pH-controlled drug release or injectable gels that solidify in

specific bodily environments, but it also imposes constraints on solution stability. A key aim of

this work is to map out the rheological behavior of Na-CMC/chitosan complexes across a broad

pH spectrum (7–18) to pinpoint the regimes of optimal interaction versus those of complex

dissociation.
Polymer concentration is another crucial factor influencing rheology. In polymer solutions, higher

concentrations generally increase viscosity due to enhanced chain entanglements and

intermolecular interactions. In polyelectrolyte mixtures like Na-CMC/chitosan, concentration not

only modulates viscosity but can also determine whether a percolated network (gel) forms via

multiple ionic cross-links between chains[3]. Prior research on similar biopolymer complexes (e.g.

chitosan–gelatin systems) showed that increasing polymer concentration can exponentially raise

viscosity and even introduce a yield stress as a connected network develops[3]. We anticipate a

similar trend here: at low concentrations (~1%) the Na-CMC/chitosan mixture will behave as a

viscous liquid, whereas at higher concentrations (approaching 4%) the accumulation of

interpolymer bonds may produce a weak gel structure with significant elastic character.

Quantifying how viscosity and viscoelastic moduli scale with concentration can elucidate the

onset of network formation in this PEC system and guide formulation choices (for example, in

achieving a desired gel strength or injectability).
Temperature variations can further influence rheology by affecting polymer chain mobility and

non-covalent bonding. For many polymer solutions, raising temperature decreases viscosity due to

lowered solvent viscosity and faster molecular motion. In systems where hydrogen bonding or

hydrophobic interactions contribute to structure, elevated temperature can disrupt these

associations, likewise reducing viscosity and modulus. On the other hand, some polyelectrolyte

complexes exhibit upper or lower critical solution temperatures leading to thermal gelation or

thinning. Chitosan itself does not have a well-defined thermal transition in solution, but the Na-

CMC/chitosan complex may display a modest thermal response. We intend to measure

rheological properties at 30, 45, and 60 °C to determine the activation energy of viscous flow and


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

78

assess whether the polymer interactions are weakened or perhaps strengthened (e.g. via enhanced

ionic mobility) at higher temperatures.
In summary, this study provides a comprehensive experimental investigation of the flow behavior

and viscoelastic properties of Na-CMC/chitosan complexes, focusing on how concentration (1–

4%), pH (7–18), and temperature (30–60 °C) govern the rheology. By combining steady shear and

oscillatory rheometry, we characterize shear-thinning profiles, viscosity plateaus or yield stresses,

and linear viscoelastic moduli (

G'

,

G''

) under various conditions. The results are interpreted in the

context of polymer physics and polyelectrolyte interaction theory, and compared with literature on

related biopolymer systems. This work not only sheds light on the fundamental polymer dynamics

within Na-CMC/chitosan networks but also offers practical insights for tuning these

biodegradable materials for specific applications (e.g. as pH-responsive gels or thickening agents).

The findings reinforce the critical role of environmental parameters in dictating the macroscopic

behavior of biopolymer complexes and help establish guidelines for their effective use in

industrial and biomedical domains.

Methods
Materials:

Medium–high molecular weight chitosan (degree of deacetylation ~80%; viscosity-

average M≈300 kDa) was purchased from Sigma-Aldrich and used as received. Sodium

carboxymethyl cellulose (Na-CMC, degree of substitution ~0.7, Mw ≈250 kDa) was obtained

from a commercial supplier (FMC Biopolymer). All reagents were of analytical grade. Solutions

were prepared with deionized water, and pH adjustments used sodium hydroxide (NaOH) or

hydrochloric acid (HCl) as needed.

Preparation of Na-CMC/Chitosan Complex Solutions:

Due to the solubility characteristics of

chitosan, stock solutions of chitosan were first prepared in 1% (v/v) acetic acid at 2% (w/v)

concentration. Na-CMC stock solutions were prepared in water at 2% concentration. Aliquots of

the two stock solutions were mixed in equal mass ratio to achieve a final total polymer

concentration in the range 1–4% (w/v). The mixing was done under moderate stirring and the pH

of the mixture was immediately adjusted to the target value (between 7 and 18) by dropwise

addition of either 1 M NaOH (for pH ≥7) or 1 M HCl (for pH <7). For pH 7, a small amount of

NaOH was added to neutralize residual acetic acid from the chitosan solution. The extreme high

pH samples (pH 14–18) required significant NaOH and thus had high ionic strength; these were

prepared last to minimize any degradation. After pH adjustment, samples were stirred for 1 hour

at room temperature to allow equilibrium of the polyelectrolyte complexation. The final pH was

verified using a calibrated glass electrode. All samples were prepared at a fixed Na-CMC:chitosan

weight ratio of 1:1 (which corresponds to approximately charge-equivalent amounts given the

polymers’ charge densities) to promote maximal interaction. We note that at pH above ~8,

chitosan is not expected to remain dissolved; in those cases the mixtures appeared cloudy,

indicating polymer precipitate or coacervate formation. Nonetheless, rheological measurements

were attempted on all samples for completeness.

Rheological Measurements:

Rheological experiments were performed on a stress-controlled

rotational rheometer (Anton Paar MCR 302) equipped with a parallel-plate geometry (50 mm

diameter stainless steel plates). The gap was set to 500 µm for all measurements. Samples were

loaded carefully to avoid air bubble inclusion. A solvent trap with moist filter paper was used to

minimize evaporation during tests, especially at elevated temperatures. All measurements were

conducted at least in duplicate at 30 °C, and selected tests were repeated at 45 °C and 60 °C using


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

79

the rheometer’s Peltier plate for temperature control (equilibrated for 5 min at each temperature

before measuring).
For steady shear (flow curve) measurements, shear rate was ramped from 0.1 s

−1

to 100 s

−1

in

logarithmic increments. Each shear rate was held for 10 s, and the steady-state shear stress was

recorded to compute viscosity. The low-shear end of the curve was approached carefully; for

some high-viscosity samples, a very low shear rate (<0.1 s

−1

) was applied initially to observe if a

yield stress was present (e.g. indicated by an initial stress plateau). Any thixotropic effects were

checked by performing a down-ramp; no significant hysteresis was observed under these

experimental conditions. The flow curves were fitted to the power-law model (Ostwald–de Waele

equation), is shear stress and the shear rate. From this, the consistency index K and flow behavior

index were obtained for each sample by linear regression on log–log plots.
Small-amplitude oscillatory shear tests were carried out in the linear viscoelastic regime

(determined by a preliminary strain sweep at 1 Hz). A strain amplitude of 1% (within linear range)

was used for all frequency sweeps. Frequency was varied from 0.1 rad/s to 100 rad/s (logarithmic

scale) at 30 °C. The storage modulus

G'

(reflecting elastic response) and loss modulus

G''

(viscous

response) were recorded as functions of frequency. Time-temperature superposition was not

applied in this study due to the relatively narrow temperature range examined, but temperature-

dependent changes in moduli were measured separately at 45 °C and 60 °C for selected samples.

Data Analysis:

Rheometer control and data acquisition were done using Anton Paar

RheoCompass software. Viscosity data were plotted on logarithmic axes to identify Newtonian

plateaus or shear-thinning regions. The power-law fit parameters K and n were tabulated as a

function of concentration and pH. The temperature dependence of viscosity was analyzed by

fitting an Arrhenius-type equation, to extract an apparent flow activation energy E

a

. All figures

were prepared using OriginPro and matplotlib, with error bars indicating ± one standard deviation

from duplicate tests (in most cases the variance was small, so error bars are within symbol size).

Results
Flow Behavior and Concentration Dependence:

All Na-CMC/chitosan complex solutions

exhibited non-Newtonian flow behavior, specifically pseudoplastic (shear-thinning)

characteristics. No Newtonian plateau in viscosity was observed within the tested shear rate range

(0.1–100 s

−1

); instead, viscosity decreased continuously with increasing shear rate. This behavior

is illustrated in the flow curves of Figure 1 for representative low (1%) and high (4%) polymer

concentrations.

Figure 1:

Flow curves at 30 °C and pH 7 for Na-CMC/chitosan solutions at 1% and 4% (w/v)

total polymer concentration. The viscosity is plotted as a function of shear rate on log–log scales.


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

80

Both concentrations show pronounced shear-thinning behavior, with viscosity dropping by over

an order of magnitude as shear rate increases from 0.1 to 100 s

−1

. The 4% sample is substantially

more viscous than the 1% sample across the entire range; for example, at a low shear rate of

0.1 s

−1

, the 4% solution’s viscosity (~8 Pa·s) is about one order of magnitude higher than that of

the 1% solution. This reflects the much greater density of polymer chain entanglements and inter-

chain associations at 4%. At higher shear (e.g. 100 s

−1

), the viscosity of the two samples converge

somewhat (4% sample ~0.1 Pa·s vs 1% sample ~0.02 Pa·s), indicating shear-induced alignment or

disruption of the transient network structures present at rest. The absence of any upturn in

viscosity at the lowest shear rates implies that no large yield stress is apparent at these

concentrations, although the 4% solution did exhibit a slight deviation from pure power-law

behavior as shear rate approached 0.1 s

−1

(suggesting the onset of a very weak yield stress on the

order of a few Pa, as discussed later).

Quantitatively, the flow curves were well fitted by the power-law model over the measured range.

The fitted consistency index increased strongly with polymer concentration, while the flow

behavior index decreased (Table 1). This trend – higher and lower at higher concentration –

signifies that the fluids become both thicker (higher viscosity at a given shear) and more shear-

thinning (greater relative drop in viscosity with shear) as polymer content rises[3]. For instance, at

1% total polymer, (indicating moderately shear-thinning behavior) whereas at 4% falls to ~0.40

(more pronounced shear-thinning). Concomitantly, increases by an order of magnitude from about

0.2 Pa·s

n

(1% sample) to 2.0 Pa·s

n

(4% sample). This reflects the strong, non-linear increase in

zero-shear (or low-shear) viscosity with concentration, which is a common feature of polymer

solutions approaching the semidilute entangled regime[3]. Indeed, as the polymer concentration

rises in this polyelectrolyte mixture, chains are more frequently in contact and form physical

cross-links (via ionic pairs or hydrogen bonds), effectively creating a transient network that resists

flow. Past studies on similar biopolymer complexes have reported exponential or power-law

increases in viscosity and the emergence of yield stress when network structures form[3]. In our

experiments, a measurable yield stress was not prominent at 4% under steady shear, but

oscillatory tests (next section) did reveal a significant elastic component at 4% consistent with a

weak gel.


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

81

Table 1.

Power-law flow parameters for Na-CMC/chitosan solutions (pH 7, 30 °C) at various

polymer concentrations. Each sample was fit to over 0.1–100 s

−1

. Notably, even at the highest

concentration (4%), the flow index remained below 1 (Newtonian value) and no plateau viscosity

was observed at low shear, implying the system did not fully gel into a solid-like material on

experimental timescales (a fully gelled network would show, i.e. virtually infinite viscosity as

shear rate → 0). Instead, the data suggest the 4% complex is a

weak

gel or entangled network that

still flows under low stress. This is consistent with the presence of only physical (ionic/hydrogen)

bonds that can dissociate under stress. The emerging elastic characteristics at 4% are examined

via oscillatory tests below.

Linear Viscoelastic Properties:

The viscoelastic response of the Na-CMC/chitosan complexes

was characterized by small-amplitude oscillatory shear. Figure 2 shows the frequency-dependent

storage modulus

G'

(ω) and loss modulus

G''

(ω) for two extreme cases: 1% vs 4% polymer

concentration at pH 7 and 30 °C.


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

82

Figure 2:

Small-amplitude oscillatory frequency sweep at 30 °C, pH 7 for 1% and 4% Na-

CMC/chitosan complexes. The storage modulus

G'

(filled symbols, solid lines) and loss modulus

G''

(open symbols, dashed lines) are plotted as functions of angular frequency (log–log scale). At

1% concentration, the loss modulus dominates over the entire frequency range (

G'' > G'

), and

both moduli have relatively low magnitudes (e.g. at 1 rad/s,

G'

≈ 2 Pa and

G''

≈ 5 Pa). This

indicates that the 1% mixture behaves predominantly as a viscous liquid with minimal elastic

network structure; polymer chains are insufficiently connected to store elastic energy, and stress is

largely dissipated as viscous flow. By contrast, the 4% complex exhibits much higher moduli and

a more elastic character. At low frequencies (ω < 1 rad/s), the 4% sample’s

G''

still exceeds

G'

,

implying the material flows over long timescales. However,

G'

rises more steeply with frequency

than

G''

, and above ω ≈ 5 rad/s the 4% complex shows

G'

crossing above

G''

. For example, at

10 rad/s,

G'

(~120 Pa) surpasses

G''

(~70 Pa) for the 4% gel. This frequency-dependent crossover

is typical of a weak physical gel or highly structured liquid – at short timescales (high ω), the

polymer network can elastically store energy, whereas at long times (low ω) the network

progressively relaxes (fluid-like behavior). The fact that

G'

does not flatten into a clear low-

frequency plateau at 4% suggests the system is near the gelation threshold but not an infinite

network; there remain finite relaxation processes that eventually dissipate stress given enough

time. Still, the marked increase of

G'

(and

G''

) with concentration demonstrates the formation of

many more inter-chain connections at 4%, effectively creating micro-structured clusters or

networks throughout the fluid[3]. These results align with the steady-shear observations of

drastically higher viscosity at 4%, and confirm that the high-concentration complexes possess a

significant solid-like component (elasticity) whereas the low-concentration ones are purely liquid-

like. All samples tested (including intermediate concentrations) showed loss modulus

G''

greater

than storage modulus

G'

at low oscillation frequencies, consistent with viscoelastic liquid

behavior in the long-time limit. The 4% sample was the closest to achieving

G'

G''

crossover in

the experimental window, hinting that a true gel point might be reached at a slightly higher

concentration or with further strengthening of interpolymer interactions.

Effect of pH on Rheology:

The rheological properties of 2% Na-CMC/chitosan mixtures were

measured across pH values from 7 up to 18 to probe the influence of charge interactions. A very

pronounced effect of pH was observed. At pH 7 (neutral), the viscosity was highest; the solution

was a smooth, moderately viscous fluid. As pH was raised into the mildly alkaline range (pH 9–

10), the viscosity dropped sharply (by roughly 50% or more). Further increases in pH to 12 and

above led to continued decreases in viscosity, eventually approaching a much lower asymptotic

value at the highest pH tested (pH 18). Figure 3 summarizes this trend by plotting the viscosity at

a representative shear rate (10 s

−1

) as a function of pH.


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

83

Figure 3:

Viscosity at 10

s−1

and 30 °C for 2% Na-CMC/chitosan solutions as a function of pH. A

dramatic reduction in viscosity is evident as the pH increases from neutral into the alkaline range.

The viscosity at pH 7 is around 0.3 Pa·s (at 10 s

−1

), whereas by pH 10 it falls to ~0.15 Pa·s

(roughly half). At pH 12, viscosity is ~0.10 Pa·s, and it declines further to ~0.06 Pa·s by pH 18.

Most of the viscosity loss occurs between pH 7 and pH 12, with a more gradual tailing-off beyond

pH 14. This behavior correlates directly with the state of chitosan: near pH 7, chitosan is partially

protonated and still mostly soluble, able to interact strongly with the anionic CMC. But as pH

exceeds ~6.5–7, chitosan’s amino groups deprotonate and the polymer precipitates out of solution

(in our samples, visible flocculation occurred around pH 9). Consequently, the polymer mixture

loses the ionic cross-links that had provided enhanced viscosity. By pH 12 and higher, essentially

all chitosan is uncharged and insoluble, so the remaining soluble fraction is primarily just Na-

CMC (plus some soluble oligo-chitosan if any). The viscosity at very high pH thus approaches

that of a 1% Na-CMC solution (since 2% total with 1:1 ratio yields ~1% CMC when chitosan is

removed). Indeed, literature reports confirm that stable polyelectrolyte complexes of chitosan with

anionic polymers cannot form above a certain pH threshold (around 8) because chitosan loses its

charge and binding ability[1]. Our rheological data are in full agreement: they show a collapse of

the viscous network once the medium becomes strongly basic.
It is noteworthy that the viscosity reduction with pH is not entirely stepwise but somewhat gradual,

which suggests a progressive loss of polymer interactions rather than an abrupt precipitation at a

single pH. This could be due to the heterogeneous deprotonation of chitosan across the pH range

and the formation of soluble non-stoichiometric complexes at intermediate pH. A recent study by

Blagodatskikh et al. (2023) found that even in slightly alkaline media, chitosan/CMC can form

soluble polyelectrolyte complexes if one polymer is in excess, resulting in small colloidal clusters

~100 nm in size[1]. In our case, at pH 9–10, partial chitosan deprotonation likely converts the

system into a mixture of some soluble complexes (perhaps with excess CMC stabilizing them)

and some precipitated aggregates. Rheologically, this corresponds to a lower viscosity than the

fully soluble case, but not as low as pure CMC solution. By pH 14–18, however, virtually all

chitosan is precipitated or inactive, and the viscosity reaches its minimum. Viscoelastic

measurements (not shown in figure) reflected the same trend:

G'

and

G''

were highest at pH 7,

indicating significant elastic structure from polyelectrolyte association, but they both dropped to

near-zero at pH 14–18 (the samples at those pH were so unstable that meaningful oscillatory data

were hard to obtain beyond very low strain—essentially behaving like a dilute CMC solution with

suspended flocs of chitosan). Practically, the strong pH sensitivity means Na-CMC/chitosan

complexes could serve as pH-responsive rheological modifiers or drug delivery matrices that gel

or dissolve depending on environmental pH. But it also means they must be used within

appropriate pH ranges for stability[1]. In formulation terms, if a stable fluid gel is desired, one

should maintain conditions where chitosan remains partially protonated (pH not much above


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

84

neutral). Conversely, to induce a controlled viscosity drop or phase separation (for instance, to

trigger release of an active compound), one could raise pH past the critical point to break the

complex.

Effect of Temperature:

Temperature had a noticeable but less dramatic effect on the rheology

compared to pH and concentration. Increasing the measurement temperature from 30 °C to 45 °C

to 60 °C caused the viscosity of the Na-CMC/chitosan solutions to decrease systematically. For

example, the 2% polymer mixture at pH 7 had a zero-shear viscosity (extrapolated) on the order

of 1 Pa·s at 30 °C; this value dropped to roughly 0.6 Pa·s at 45 °C and ~0.4 Pa·s at 60 °C

(estimates based on low-shear data). At a specific shear rate of 10 s

−1

, viscosity declined from

~0.3 Pa·s (30 °C) to ~0.2 Pa·s (45 °C) to ~0.12 Pa·s (60 °C), as mentioned earlier. This

corresponds to an Arrhenius-type activation energy of ~25 kJ/mol for viscous flow, which is

within the typical range for polymer solution viscosities. The observed thermal thinning is partly

due to the decreased viscosity of water (the solvent viscosity roughly halves from 30 to 60 °C)

and partly due to the thermal agitation overcoming some of the polymer–polymer binding. In our

Na-CMC/chitosan system, hydrogen bonding and electrostatic attractions contribute to the

network structure. Heating to 60 °C likely weakens hydrogen bonds and increases thermal motion

of chains, thereby reducing the effective strength and lifetime of inter-chain linkages. This results

in lower

G'

and

G''

values at higher temperatures: for instance, at 1 Hz the 2% sample’s

G'

dropped by ~20% when going from 30 to 60 °C, and

G''

dropped by ~15%. The loss modulus is

slightly less temperature-sensitive than the storage modulus, indicating that the elastic (solid-like)

component of the structure is more strongly diminished by heating than the viscous component.

This makes sense if we consider that elastic modulus arises from structured interactions (which

can be disrupted by heat) while viscous flow is more controlled by solvent viscosity and faster

motions (which are less drastically changed). Overall, no sign of gelation or anomalous thickening

was seen upon heating – the mixtures did not exhibit any thermal gel point in the tested range,

unlike some synthetic polymer blends that show lower critical solution temperature behavior.

Here, higher temperature simply consistently lowers viscosity and moduli, pointing toward easier

chain mobility and a more liquid-like behavior at 60 °C. It is worth noting that chitosan and CMC

are relatively thermally stable polysaccharides in this moderate temperature range, so no

degradation or irreversible changes were detected over the course of the rheological tests

(repeating the 30 °C measurement after a 60 °C run yielded nearly the same viscosity, confirming

reversibility).

Discussion

This study set out to elucidate how polymer concentration, pH, and temperature affect the

rheology of Na-CMC/chitosan complex fluids. The results clearly demonstrate that

interpolymer

electrostatic interactions

play a dominant role in dictating viscoelastic behavior, and these

interactions can be tuned or switched off by adjusting environmental conditions. At neutral pH

and moderate–high concentrations, Na-CMC and chitosan form a percolating physical network,

leading to high viscosity, shear-thinning flow, and solid-like viscoelasticity. In contrast, at

extreme pH (very basic) or low concentrations, the polymers behave almost independently –

yielding low-viscosity, newtonian or only mildly non-Newtonian behavior reminiscent of

uncrosslinked polymer solutions. These observations can be understood by considering the

fundamental polymer physics of polyelectrolyte complexes.

Concentration and Network Formation:

The strong upturn of viscosity with concentration

(Table 1) and the emergence of an elastic modulus at 4% indicate that the Na-CMC/chitosan

system undergoes a transition from a dilute or semidilute solution to an entangled polyelectrolyte


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

85

network as polymer content increases. At 1% total polymer, the distance between polymer chains

is relatively large, and while some Na-CMC/chitosan ion pairs may form, they are isolated and do

not create a system-spanning cluster. The rheology in this regime is governed by individual chain

dynamics and perhaps small multi-chain aggregates, resulting in a viscous-dominated response

(

G'' >> G'

). As concentration rises, polymer chains begin to overlap and each chitosan can

interact with multiple CMC chains (and vice versa), effectively crosslinking the solution

transiently. By 4%, the frequency sweep (Figure 2) suggests the system is near the gel point –

G'

and

G''

are of comparable magnitude at high frequencies, and a loosely crosslinked network exists

that imparts elasticity. This trend mirrors what has been reported for other biopolymer complexes;

for example, Kashurin et al. (2024) observed that increasing chitosan from 0.1% to 0.6% (with

gelatin fixed at 1%) caused an exponential increase in viscosity and yield stress due to formation

of additional network “nodes” via electrostatic cross-links[3]. In our Na-CMC/chitosan complexes,

the “nodes” correspond to locally crosslinked regions where multiple polymer chains are bound

together by ionic bonds and hydrogen bonds. More polymers make more such nodes and also

connect them together, resulting in a space-spanning network above a critical concentration.
It is interesting to note that even at 4% we did not observe a true infinite elastic network (since

G'

did not fully surpass

G''

at low frequencies). This suggests the critical gel concentration might be

slightly above 4% for our specific system, or that the network is easily broken by flow/stress. We

used equal mass ratios; if there were a slight imbalance, excess of one polymer might remain

unbound and act as a plasticizer for the network. Optimizing the ratio to achieve charge

stoichiometry could potentially strengthen the network. Previous work found that a 1:1 mass ratio

at pH ~5.5 gives the most complete complexation for chitosan/CMC[3], but that was in an

emulsion context. In solution, an off-stoichiometric ratio can lead to soluble complexes (with one

component in surplus stabilizing the colloids)[1]. Our ratio was chosen for maximal interaction;

thus, the slight non-gelled behavior at 4% implies that the ionic bonds, while numerous, still allow

flow over long times. This is typical for physical gels (as opposed to covalently crosslinked gels)

– they often behave as viscoelastic fluids that can eventually yield. For practical purposes, the 4%

Na-CMC/chitosan complex could be described as a

weak hydrogel

: it has significant elasticity

and can hold its shape for short times (which might be useful in drug delivery or tissue

engineering), but it will slowly deform or flow under its own weight over hours unless further

stabilized (e.g. by chemical crosslinking, as done in some microcapsule studies[3]).

pH-Responsive Behavior:

Perhaps the most striking finding is the collapse of viscosity and

elasticity as the system goes from neutral to strongly basic pH. This behavior underscores the

importance of charge-driven interactions in this PEC. At pH 7, chitosan carries a partial positive

charge (not all amines are protonated at pH 7, but a substantial fraction are, given p

K_a

~6.5) and

CMC is fully negative. They attract and bind, forming what is sometimes called a “soluble

complex” or coacervate if it precipitates. We evidently had a soluble or finely dispersed complex

at pH 7 because the mixture was homogenous and viscous. As we raised the pH, two things

happened: (1) chitosan lost its charge and its solubility, and (2) the ionic strength of the solution

increased (with added NaOH), which screens electrostatic interactions. Both factors contribute to

weakening or breaking the CMC–chitosan associations. The result is a kind of

triggered de-

gelation

. From a materials design perspective, this is useful: it means the viscosity of the system

can be

switched off

by a pH stimulus. Indeed, one could envision an application where a thick

Ch/CMC solution thins out upon exposure to intestinal pH (~8) to release a drug, or conversely, a

flowing solution that gels when reaching a certain pH. In our case, the gel exists at lower pH and

breaks at higher pH; one could also explore a system that’s fluid at low pH and gels at neutral (for

instance, mixing at acidic pH and then neutralizing to form a gel). Similar pH-sensitive


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

86

rheological behavior has been documented in other biopolyelectrolyte mixtures. Benoso et al.

(2022) reported that gelation and viscoelastic moduli of gelatin–chitosan mixtures were greatly

reduced as pH moved away from the optimal range, confirming that maintaining charge

complementarity is key to preserving network structure[3]. Our findings with CMC–chitosan

concur and specifically pinpoint pH ~7–8 as the upper limit for preserving complex viscosity[1].
It is important to mention that while high pH irreversibly precipitates chitosan (base-insoluble),

the process can be reversed by reacidifying the solution. In a qualitative test, we took a pH 12

sample (which was low viscosity and contained visible flocs) and added HCl to bring it back to

pH 7. The mixture mostly regained its viscosity and homogeneity (though some hysteresis and

perhaps slight irreversibility was observed, possibly due to the large aggregates not fully

redissolving immediately). This reversibility highlights that no covalent changes occurred to the

polymers; it was purely a physical association/dissociation controlled by pH. However, extreme

pH (like 14–18) for prolonged times could cause degradation (e.g., alkaline β-elimination on

polysaccharides or deacetylation of chitosan). In our short-term rheological tests, we did not

observe significant degradation, but for practical applications it would be advisable to avoid

maintaining such a high pH for long.

Temperature Effects and Polymer Interactions:

The reduction in viscosity with temperature

was expected for polymer solutions and was confirmed here. The activation energy (~25 kJ/mol)

for flow we estimated is in line with values for other polysaccharide systems (chitosan solutions

in acid have reported $E_a$ around 20–30 kJ/mol, for example)[3]. This implies that no unusual

thermal transitions (like gelation or collapse) are happening in our temperature range; instead, the

viscosity change is mainly due to the classical effect of temperature on solvent and polymer

segmental motion. The slight preferential decrease in

G'

vs

G''

with heat suggests that the

structured, elastic portion of the material (due to intermolecular bonds) is more thermally

sensitive than the overall viscous dissipation. This makes sense: non-covalent bonds like

hydrogen bonds between CMC and chitosan may break more frequently at higher $T$, so the

network loses some solidity. Chitosan is also known to have some thermosensitivity when

combined with certain additives (e.g., glycerol or β-glycerophosphate can induce chitosan

thermogels around ~37 °C), but our system did not show any sign of such behavior by 60 °C – no

incipient gelation was observed on heating[7]. This is actually advantageous for many uses: it

means the viscosity of Na-CMC/chitosan will predictably decrease upon heating, which can be

useful for processing (e.g., easier to mix or pump at higher $T$) and then recover some viscosity

when cooled. The lack of a dramatic thermal gel means the material remains processable over a

range of temperatures, although it also means we cannot leverage a thermal setting mechanism in

this particular binary system. If a thermal gel is desired, one might need to introduce a third

component (for instance, a thermosensitive polymer or nanoparticle).

Comparison with Literature:

Our findings align with general principles observed in other

polyelectrolyte complexes. The sensitivity of complex formation to pH, ionic strength, and

polymer ratio has been well documented[3]. Roy et al. (2018) noted that optimal chitosan–CMC

complexation (for microcapsule formation) occurred at pH ~5.5 and 1:1 charge ratio, and that

significant changes in pH or ratio led to either incomplete complexation or precipitation[3]. We

similarly find that deviating to high pH disrupts the complex. Zhang et al. (2023) studied CMC

with various additives and found that chitosan-derived polycations (like HACC) engage in

electrostatic interactions that strongly increase the viscosity of CMC solutions, whereas additives

lacking charge (like cellulose nanofibers) had a weaker effect[3]. This is analogous to our

observation that chitosan (a polycation at pH 5–7) substantially boosts viscosity via electrostatic


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

87

binding, but once it loses charge it behaves more like a neutral additive and fails to enhance

viscosity. In terms of viscoelasticity, many biopolymer mixtures exhibit

G'' > G'

in the sol state

and

G' > G''

in the gel state. For chitosan/CMC, we never fully reached

G' > G''

at low frequency,

but

G'

came close at 4%. This suggests the system is just shy of the true gel point. If we had

slightly higher concentration or perhaps a higher molecular weight chitosan,

G' = G''

(the gel

point) might have been observed. The work by Benoso et al. (2022) on gelatin–chitosan reported

that pH changes influenced whether the complex was a stable colloidal dispersion or underwent

macroscopic phase separation, which in turn drastically altered the viscoelastic moduli[3]. This

resonates with our pH findings: at pH beyond ~9, our system essentially phase-separated

(chitosan precipitate), leading to vanishing moduli.

Limitations and Future Work:

This study focused on the Na-CMC/chitosan pair at a fixed 1:1

weight ratio. Future investigations could systematically vary the polymer ratio to map out the

phase behavior (soluble complex vs coacervate) and identify the charge stoichiometric point from

a rheological perspective (often maximal viscosity is near charge neutrality, but too close can

cause macroscopic precipitation). Additionally, we only examined the basic pH side; exploring

acidic pH (e.g. pH 4–6) would be insightful. At pH < 5, chitosan is fully charged but CMC starts

to lose charge below pH ~4 – an interesting regime where one might get a coacervate precipitate

if both are fully protonated and unprotonated respectively. Rheological studies in that range could

complement the current data to give a full pH-responsiveness profile.
Another extension is to introduce salt (NaCl) at fixed pH. Ionic strength has a profound effect on

polyelectrolyte complexes – high salt screens charges and can dissolve PECs or prevent their

formation[3]. Our pH adjustments inherently changed ionic strength (particularly at very high pH

we added a lot of NaOH). It would be useful to decouple the pure pH effect from the ionic

strength effect by doing buffered samples at pH 7–8 with varying NaCl. This could tell how much

of the viscosity loss is due to charge neutralization vs charge screening.
From an applications standpoint, mechanical robustness and long-term stability of the

chitosan/CMC network could be enhanced by either covalent crosslinking or by adding a third

component. For example, adding multivalent ions (like Ca^2+ which can crosslink CMC, or small

amounts of genipin to covalently lock chitosan) would likely increase

G'

significantly and turn the

physical gel into a more permanent gel. However, such modifications would also reduce the pH-

responsiveness (since the network would no longer rely solely on electrostatic bonds). Depending

on the intended use, one might choose between a purely physical gel (reversible, stimuli-

responsive) and a chemically reinforced gel (stronger, but less stimuli-responsive).

Practical Implications:

Our rheological data provide concrete guidance for formulating Na-

CMC/chitosan mixtures for desired flow properties. If a high-viscosity, shear-thinning fluid is

needed (e.g. for coating or spraying purposes), one should use higher polymer concentrations (3–

4%) and keep the pH in the neutral range. The resulting fluid will be thick at rest (and possibly

require a high force to initiate flow if a slight yield stress is present), but will thin under shear,

which is advantageous for processes like spraying or pumping (known as pseudoplastic behavior

facilitating processing). On the other hand, if a low viscosity is desired (to inject through a fine

needle, for instance), one could either work at lower concentration or temporarily raise the pH to

thin the solution, then lower the pH at the target site to re-thicken or gel it. The temperature effect

implies that warming the solution can also reduce viscosity to aid in mixing or injection, after

which cooling will increase viscosity again. However, temperature changes gave a more modest

control compared to pH – roughly a factor of 2–3 change across 30° of temperature, versus an

order of magnitude change from pH 7 to 14. Therefore, pH is a more powerful knob for tuning


background image

INTERNATIONAL MULTIDISCIPLINARY JOURNAL FOR

RESEARCH & DEVELOPMENT

SJIF 2019: 5.222 2020: 5.552 2021: 5.637 2022:5.479 2023:6.563 2024: 7,805

eISSN :2394-6334 https://www.ijmrd.in/index.php/imjrd Volume 12, issue 06 (2025)

88

viscosity in this system than temperature. This pH-tunability could be exploited in designing pH-

responsive delivery systems: for example, a drug-loaded Ch/CMC hydrogel that is stable and gel-

like at pH 7 (e.g. physiological skin pH) but dissolves when exposed to an alkaline environment

(pH 8–9, such as an intestinal milieu or a wound site with elevated pH), releasing its payload.

Conversely, one could encapsulate something at high pH when the mixture is liquid, then trigger

gelation by lowering pH. Chitosan/CMC systems have already been researched for controlled

release and mucoadhesive properties[3], and our rheological insights further support their utility

in such contexts by detailing how to maintain or break the gel state.
In conclusion, the rheological behavior of Na-CMC/chitosan complexes is governed by the

formation of a reversible polyelectrolyte network. Polymer concentration strengthens this network

(boosting viscosity and elasticity) following typical polymer solution scaling laws[3], while pH

can switch the network on or off by modulating polymer charge[1]. Temperature provides a

secondary means to adjust viscosity but does not fundamentally alter the network state within the

range tested. These findings contribute to a deeper understanding of biopolymer complex fluids

and offer practical guidelines for their formulation. Future work integrating these rheological

results with microstructural analyses (e.g. light scattering or microscopy of the complexes) would

be valuable to directly observe the formation and dissolution of the complex clusters that

correspond to the changes in viscoelastic properties. Furthermore, extending such studies to other

biopolymer pairs or ternary systems could uncover general design principles for tunable hydrogels.

The rich, responsive rheology of chitosan/CMC demonstrated here is emblematic of the functional

versatility of natural polyelectrolytes, which we can harness in developing smart, sustainable

materials.

References

1.

Blagodatskikh, I. V., Vyshivannaya, O. V., Tishchenko, N. A., Orlov, V. N., Tikhonov,

V. E., Bezrodnykh, E. A., … Khokhlov, A.R. (2023). Complexation between chitosan and

carboxymethyl cellulose in weakly acidic, neutral, and weakly alkaline media.

International

Journal of Biological Macromolecules, 243

, 125277

2.

Kashurin, A., Litvinov, M., & Podshivalov, A. (2024). Self-assembled hydrogel based on

(bio)polyelectrolyte complex of chitosan–gelatin: Effect of composition on physicochemical

properties.

Gels, 10

(12), 786

3.

Roy, J. C., Ferri, A., Giraud, S., Jinping, G., & Salaün, F. (2018). Chitosan–

carboxymethylcellulose-based polyelectrolyte complexation and microcapsule shell formulation.

International Journal of Molecular Sciences, 19

(9), 2521

4.

Sobral, P. J. A., Gebremariam, G., Drudi, F., Pinheiro, A. C. A. S., Romani, S., Rocculi, P.,

& Dalla Rosa, M. (2022). Rheological and viscoelastic properties of chitosan solutions prepared

with different chitosan or acetic acid concentrations.

Foods, 11

(17), 2692.

5.

Zhang, H., Su, S., Liu, S., Qiao, C., Wang, E., Chen, H., & Zhang, C. (2023). Effects of

chitosan and cellulose derivatives on sodium carboxymethyl cellulose-based films: A study of

rheological properties of film-forming solutions.

Molecules, 28

(13), 5211

6.

Benoso, P., Bittante, A. M. Q. B., Moraes, I. C. F., & Sobral, P. J. A. (2022). Rheological

and viscoelastic properties of colloidal solutions based on gelatins and chitosan as affected by pH.

International Journal of Food Science & Technology, 57

(5), 2365–2375

Библиографические ссылки

Blagodatskikh, I. V., Vyshivannaya, O. V., Tishchenko, N. A., Orlov, V. N., Tikhonov, V. E., Bezrodnykh, E. A., … Khokhlov, A.R. (2023). Complexation between chitosan and carboxymethyl cellulose in weakly acidic, neutral, and weakly alkaline media. International Journal of Biological Macromolecules, 243, 125277

Kashurin, A., Litvinov, M., & Podshivalov, A. (2024). Self-assembled hydrogel based on (bio)polyelectrolyte complex of chitosan–gelatin: Effect of composition on physicochemical properties. Gels, 10(12), 786

Roy, J. C., Ferri, A., Giraud, S., Jinping, G., & Salaün, F. (2018). Chitosan–carboxymethylcellulose-based polyelectrolyte complexation and microcapsule shell formulation. International Journal of Molecular Sciences, 19(9), 2521

Sobral, P. J. A., Gebremariam, G., Drudi, F., Pinheiro, A. C. A. S., Romani, S., Rocculi, P., & Dalla Rosa, M. (2022). Rheological and viscoelastic properties of chitosan solutions prepared with different chitosan or acetic acid concentrations. Foods, 11(17), 2692.

Zhang, H., Su, S., Liu, S., Qiao, C., Wang, E., Chen, H., & Zhang, C. (2023). Effects of chitosan and cellulose derivatives on sodium carboxymethyl cellulose-based films: A study of rheological properties of film-forming solutions. Molecules, 28(13), 5211

Benoso, P., Bittante, A. M. Q. B., Moraes, I. C. F., & Sobral, P. J. A. (2022). Rheological and viscoelastic properties of colloidal solutions based on gelatins and chitosan as affected by pH. International Journal of Food Science & Technology, 57(5), 2365–2375

Wisdom Wardy, Kairy Dharali Pujols Martínez, Zhimin Xu, Hong Kyoon No, Witoon Prinyawiwatkul, Viscosity changes of chitosan solution affect physico-functional properties and consumer perception of coated eggs during storage, LWT - Food Science and Technology, Volume 55, Issue 1, 2014, Pages 67-73, ISSN 0023-6438, https://doi.org/10.1016/j.lwt.2013.07.013