Elsevier Editorial System(tm) for Carbohydrate Polymers Manuscript Draft Manuscript Number: CARBPOL-D-10-00507R1 Title: "Synergistic interaction of Locust Bean Gum and Xanthan investigated by rheology and light scattering" Article Type: Research Paper Keywords: Polysaccharides; Synergistic interactions; Hydrogels; Rheology; Light Scattering; Gel point. Corresponding Author: Dr. Tonmmasina Coviello, Corresponding Author's Institution: University of Rome First Author: Chiara Sandolo, Dr. Order of Authors: Chiara Sandolo, Dr.; Donatella Bulone, Dr.; Maria R. Mangione; Silvia Margheritelli; Chiara Di Meo, Dr.; Franco Alhaique, Prof.; Pietro Matricardi, Dr.; Tonmmasina Coviello Abstract: The use of synergistic interactions between polysaccharides can represent an important approach for the preparation of physical networks with a wide range of industrial applications, such as in the food and biomedical fields. Locust Bean Gum and Xanthan, two biocompatible polymers, are able to form gels when mixed together. A detailed study was carried out on three samples, prepared at different weight ratio and two temperatures, by means of light scattering and rheological techniques. The solid-like character of each sample was evaluated by measurements of the viscoelastic spectrum and intensity autocorrelation function. The contribution of fast and slow fluctuations to the static light scattered intensity was also evaluated at different angles. The power law exponents for the sol-gel transitions were estimated together with the fractal dimensions. The scattering results were found in good agreement with the mechanical ones in revealing a strong dependence of the structural properties of the different samples on their preparation conditions. Indeed, results show that a fine tuning of the properties of the mixed gels is possible through the change of the temperature preparation and/or the polymer weight ratio.

Dear Editor

please find the revised version of the paper entitled “Synergistic interaction of Locust Bean Gum

and Xanthan investigated by rheology and light scattering” by Chiara Sandolo et al., that was

submitted for publication on Carbohydrate Polymers.

We are grateful to the referees for their comments and suggestions that helped us to improve the

manuscript.

Separately, we report the answers to the reviewers. Text variations are written in blue colour.

Looking forward to hear from you for further information, thank you in advance for the attention.

Yours sincerely,

Tommasina Coviello

Rome, May 20, 2010

Cover Letter

Answers to the reviewers (corrections in the text are in blue colour)

Reviewer #1: The manuscript contains an interesting comparison between experimental data which

were obtained with different techniques and converge to similar conclusions about the structural

effects of blending ratio and modes in LBG-xanthan mixtures, and this is not a frequent result. The

paper is well written and clear and does not need corrections. The only observation concerns the

number of figures: Figure 3 seems to be redundant and could be eliminated without any

consequence for the offered explanations and comments.

Fig. 3 is now omitted.

Reviewer #2: General: The abbreviation 'Xanth' for xanthan seems unnecessary. I recommend just

using 'xanthan'.

Xanthan is now used throughout the text.

p.2, l. 61: Tm depends strongly on the ionic strength (45C only under certain conditions), and the

content of pyruvate and acetate. True 'unentangled' random coils exist even well above Tm.

The reviewer is correct: for this reason the “hot samples” were prepared at a temperature (75 °C)

well above Tm in water. The corresponding sentence was changed and implemented.

p. 4. Since Mw of xanthan was obtained by LS, what were the estimates of Rg and A2? Should be

included for the sake of completeness

The Rg and the A2 values are now added in the text.

p. 4. l.113. 'Hot gelation': Were the samples cooled to RT after 15 min at 75C? Cooling rate?

The following explanation was added in the text: The “hot samples”, after heating at 75 °C for 15

minutes, were left to slowly cool down in the switched off thermostatted bath, until room

temperature was reached.

p.3-4: Please specify the temperature used for NMR analysis

The information was added.

p.4. It should be pointed out that the autoclaving of xanthan necessarily leads to a order-disorder

transition, and the reverse upon cooling. Thus, the xanthan is a renatured type rather than the native

type, as these may in fact differ (perfect double-strands are not necessarily formed)

The following explanation was added:

“It must be pointed out that the Xanthan chains undergo an order-disorder transition during

autoclaving and therefore a renaturated state (where not necessarily perfect double-strands are

formed) should be considered in the subsequent interactions with the LBG chains.”

Response to Reviewers

Fig 1: The graphical quality is unsatisfactory (unless this is related to problems reading the pdf file

on my Mac)

Probably the graphical quality depends on Mac (we verified quality on the pdf version).

P. 11, l. 270-270: This sentence sounds a bit unclear to me. Please rephrase/clarify.

The sentence was changed:

“Furthermore, the usual evolution with time of the two moduli (and correspondingly of the

mechanical spectra) observed for a chemical cross-linking reaction, is detected here by changing

the weight ratio between the two polymers (Fig. 1B).”

p. 11, l. 272: 'Random coil' only during heat treatment, presumably renaturated after cooling (and

during rheological measurements). Needs to be stated clearly.

The sentence was changed:

“A quite different situation occurs for the “hot samples”: at 75 °C LBG keeps its random coil

conformation (present at room temperature) but also Xanthan is dispersed as single chains.

Actually, Xanthan is not allowed to renaturate during cooling, because it is implied in the network

formation with the LBG chains.”

1

Synergistic interaction of Locust Bean Gum and Xanthan investigated by 1

rheology and light scattering 2

3

Chiara Sandoloa, Donatella Bulone

b, Maria R. Mangione

b, Silvia Margheritelli

a, Chiara Di Meo

a, 4

Franco Alhaiquea, Pietro Matricardi

a, Tommasina Coviello

a,* 5

6 a

Department of Drug Chemistry and Technologies, Faculty of Pharmacy, “Sapienza”, University 7

of Rome, 00185 Rome, Italy 8 b

Institute of Biophysics at Palermo Italian National Research Council Via Ugo La Malfa, 153 9

90146 Palermo (Italy)

10

11

12

13

* Corresponding author. Tel.: +39 06 49913557; fax: +39 06 49913133. 14

E-mail address: tommasina.coviello@uniroma1.it (T. Coviello). 15

16

17

18

19

ABSTRACT 20

21 The use of synergistic interactions between polysaccharides can represent an important approach for 22

the preparation of physical networks with a wide range of industrial applications, such as in the 23

food and biomedical fields. Locust Bean Gum and Xanthan, two biocompatible polymers, are able 24

to form gels when mixed together. A detailed study was carried out on three samples, prepared at 25

different weight ratio and two temperatures, by means of light scattering and rheological 26

techniques. The solid-like character of each sample was evaluated by measurements of the 27

viscoelastic spectrum and intensity autocorrelation function. The contribution of fast and slow 28

fluctuations to the static light scattered intensity was also evaluated at different angles. The power 29

law exponents for the sol-gel transitions were estimated together with the fractal dimensions. The 30

scattering results were found in good agreement with the mechanical ones in revealing a strong 31

dependence of the structural properties of the different samples on their preparation conditions. 32

Indeed, results show that a fine tuning of the properties of the mixed gels is possible through the 33

change of the temperature preparation and/or the polymer weight ratio. 34

35

36

37

Keywords: Polysaccharides; Synergistic interactions; Hydrogels; Rheology; Light Scattering; Gel 38

point. 39

*ManuscriptClick here to view linked References

2

1. Introduction 40 41

Binary mixtures of certain polysaccharides often exhibit unexpected synergistic interactions, e.g. 42

the mixture may gel under conditions where the individual components are non-gelling. Striking 43

example of this behaviour occurs when the bacterial polysaccharide Xanthan gum (Scheme 1a) is 44

mixed with some plant galactomannans (Morris, 1990). It is well known that the synergistic 45

gelation occurs by direct binding between the two polymers and not by thermodynamic 46

incompatibility. Furthermore, X-ray diffraction pattern obtained from fibres prepared from the 47

synergistic gel, are new patterns and not the simple sum of the diffraction patterns of the component 48

polysaccharides (Cairns, Miles, Morris, & Brownsey, 1988). In particular, Xanthan and Locust 49

Bean Gum (LBG) (Scheme 1b), when mixed together, give a network whose strength depends on 50

the temperature preparation and on the weight ratio between the two components. A wide debate is 51

still open in elucidating the actual mechanism leading to the gel formation and, during the past 52

decades, several models have been proposed (Lundin, & Hermansson, 1995; Richter, Boyko, 53

Matzker, & Schröter, 2004; Richter, Brand, & Berger, 2005; Wang, Y. J., Wang, Z., & Sun, 2002; 54

Zhan, Ridout, Brownsey, & Morris, 1993). It is acquired that the presence of the acetyl substituents 55

on the Xanthan chains (Wang Y. J., Wang Z., & Sun, 2002; Ojinnaka, Brownsey, Morris, E. R., & 56

Morris, V. J., 1998) as well as the branching degree of the galactomannan are crucial for gel 57

formation (an increase of the number of galactose units hinders gel formation (Dea, Clark, & 58

McCleary,1986) ). 59

The possibility to modulate the mechanical properties of the gel by varying the relative amount of 60

the two polymers and taking also into account the different conformation assumed by the Xanthan 61

chains in distilled water at 25 °C (described in terms of a double-helix ordered structure) and at T > 62

45 °C (random coil conformation) (Coviello, Kajiwara, Burchard, Dentini, & Crescenzi, 1986, 63

Hacche, Washington, & Brant, 1987) appears to be particularly interesting. The temperature at 64

which the helix-coil transition occurs, Tm, deeply depends on several factors (ionic strength, content 65

of pyruvate and acetate minor components) and therefore for the “hot samples” preparation a rather 66

3

high temperature, T=75 °C, was adopted. In the present work a comparison among three samples 67

prepared at different weight ratio and two temperatures is reported: rheological characterization and 68

light scattering techniques were applied. Morphological features of the samples, acquired by means 69

of SEM spectroscopy, are also given. 70

The importance of a study on this binary mixture is also supported by the fact that these polymers 71

(both individually and mixed together) are already widely used in food industry (Dolz, Hernández, 72

Delegido, Alfaro, & Muňoz, 2007; Mandala, Kapetanakou, & Kostaropoulos, 2008; Pedersen,1980; 73

Sahin, & Ozdemir, 2004; Ramírez, Barrera, Morales, & Vázquez, 2002; Secouard, Malhiac, Grisel, 74

& Decroix, 2003). Furthermore, it must be pointed out that the single polymers are already used as 75

excipients in tablet formulations, and the LBG/Xanthan gels have been proposed in pharmaceutical 76

applications for slow release purposes (Baichwal, 2004; Fadden, Kulkarni, & Sorg, 2004; Mannion, 77

Melia, Mitchell, Harding, & Green, 1991; Rolf, 2003; Sugden, K. 1989; Ughini, Andreazza, Ganter, 78

& Bresolin, 2004; Venkataraju, Gowda, Rajesh, & Shivakumar, 2008;). The biocompatibility of the 79

two polysaccharides, together with their peculiar synergic behaviour, can still offer new interesting 80

applications. In this sense, even more significant appears to be any attempt to elucidate the 81

molecular structure and the spectroscopic characteristics of these mixed gels in order to improve 82

and to better modulate their potential uses. 83

84

2. Materials and Methods 85

2.1. Materials 86

Locust Bean Gum (LBG), from Ceratonia Campestris, was provided by Carbomer (USA). The 87

ratio between Mannose and Galactose, was estimated by means of 1H NMR (carried out at 70 °C 88

with a Brucker AVANCE AQS 600 spectrometer, operating at 600.13 MHz) and an M/G value of 89

3.4 was found. The molecular weight (Mw) of LBG, 5.0x105, was estimated from the usual Zimm 90

plot analysis of static light scattering measurements carried out at 25 °C in 0.1 M NaCl aqueous 91

4

solutions of sterilized LBG. The estimated radius of gyration was Rg = 102 nm and the second virial 92

coefficient A2 = -3.21 x 10-4

cm3mol/g

2. Procedure details are given in the literature (Coviello, 93

Kajiwara, Burchard, Dentini, & Crescenzi, 1986). 94

Xanthan Gum, from Xanthomonas campestris, was provided by Fluka (Italy). For each repeating 95

unit, 1.6 acetate and 2.7 pyruvate groups were estimated by 1H NMR measurements (carried out at 96

85 °C with a Brucker AVANCE AQS 600 spectrometer, operating at 600.13 MHz). The molecular 97

weight (Mw) of Xanthan, 1.25x106, was obtained by means of static light scattering measurements, 98

carried out at 25 °C in 0.1 M NaCl aqueous solutions of sterilized Xanthan, and reported according 99

to the classical Zimm plot representation. The estimated radius of gyration was Rg = 140 nm and the 100

second virial coefficient A2 = +6.09 x 10-4

cm3mol/g

2. 101

The two polysaccharides were used after sterilization, centrifugation and purification. A given 102

amount of Xanthan, sodium salt, (polymer concentration, cp = 0.5% w/V) was dissolved in distilled 103

water, under magnetic and mechanical stirring, at room temperature for 48 h, while LBG (cp = 0.5% 104

w/V) was dispersed in distilled water, under magnetic and mechanical stirring, at 80 °C for 24 h and 105

at room temperature for 24 additional hours (Kok, Hill, & Mitchell, 1999). The obtained solutions 106

were autoclaved at 121 °C for 20 minutes and then centrifuged at 8.000 rpm for 15 minutes at 25 107

°C. The supernatant solutions (the polymer weight loss was 15 % in the case of Xanthan and 30 108

% for LBG) were then exhaustively dialyzed at 4 °C against distilled water using dialysis 109

membranes with a cut-off 12,000–14,000. In order to convert the Xanthan polymer in the sodium 110

form, NaOH 0.2 N was added to the dialyzed solution up to pH = 7.0. Finally, the samples were 111

freeze-dried and stored in a desiccator until use. 112

All other products and reagents were of analytical grade. 113

114

2.2. LBG/Xanthan hydrogel preparations 115

116

5

The LBG/Xanthan hydrogels were prepared by mixing appropriate amounts of LBG and 117

Xanthan solutions (cp = 0.125 % w/V) previously autoclaved at 121 °C for 20 min and filtered 118

through cellulose nitrate membranes with pore sizes of 8 and 1.2 μm in sequence. It must be pointed 119

out that the Xanthan chains undergo an order-disorder transition during autoclaving and therefore a 120

renaturated state (where not necessarily perfect double-strands are formed) should be considered in 121

the subsequent interactions with the LBG chains. Different aliquots of the two polymer solutions 122

were mixed for 15 minutes at 75 °C (“hot gelation”) (where the Xanthan chains adopt a coil 123

conformation (T > Tm)) or at room temperature (“cold gelation”) (where the Xanthan chains are in 124

double helix conformation (T < Tm)), leading to samples with final different weight ratios. The “hot 125

samples”, after heating at 75 °C for 15 minutes, were left to slowly cool down in the switched off 126

thermostatted bath, until room temperature was reached. Three samples were prepared, namely 1:1 127

(50% of both, LBG and Xanthan solutions), 1:3 (25% of LBG solution and 75% of Xanthan 128

solution), and 1:9 (10% of LBG solution and 90% of Xanthan solution). 129

130

2.3. Rheological characterization 131

The rheological characterization of the LBG/Xanthan gels was performed by means of a 132

controlled stress rheometer, Haake Rheo-Stress RS300 model, with a Thermo Haake DC50 water 133

bath; a grained plate-plate device (Haake PP35 TI: diameter = 35 mm; gap between plates = 1mm) 134

was used in order to reduce the extent of the wall slippage phenomena (Lapasin, & Pricl, 1995). 135

Rheological properties were studied with oscillatory experiments; mechanical spectra were recorded 136

in the frequency range 0.001–10 Hz. The linear viscoelastic region was assessed, at 1 Hz, by stress 137

sweep experiments; a constant deformations, γ = 0.01, was used. All samples were analysed at 25 ± 138

1 °C, one day and 6 days after the mixing of the two polymer solutions. 139

140

2.4. Light Scattering 141

6

142

Each sample was placed into a thermostatted cell compartment of a Brookhaven Instruments 143

BI200-SM goniometer. The temperature was controlled to within 0.1 °C using a thermostatted 144

recirculating bath. The light scattered intensity and time autocorrelation function (TCF) were 145

measured by using a Brookhaven BI-9000 correlator and a 100 mW Ar laser (Melles Griot) tuned at 146

= 514.5 nm or a 50 mW He-Ne tuned at = 632.8 nm. Measurements were taken at different 147

scattering vector q = n0-1

sin(/2), where n is the refractive index of the solution, 0 is the 148

wavelength of the incident light, and is the scattering angle. To overcome the non-ergodicity, 149

typical of gelled systems (Pusey, & van Megen, 1989; Rodd, Dunstan, Boger, Schmidt, & 150

Burchard, 2001), the measurements were carried out by collecting the signal over different regions 151

of the specimen, automatically scanned by means of a motor-driven cell holder. In dynamic light 152

scattering (DLS) experiments the correlator operated in the multi- mode. About 150 153

measurements, lasting 7 min each, were taken over different sample positions; the intensity 154

autocorrelation function was then obtained by their sum normalized by the square of the total 155

photon number. The average scattered intensity factor and its dependence on q-vector were obtained 156

by collecting the signal over large scattering volumes while the samples were continuously rotating. 157

Static light scattering data were corrected for the background scattering of the solvent and 158

normalized by using toluene as calibration liquid. The static light scattered intensity was also 159

studied in terms of fluctuant and static components (Manno, Bulone, Martorana, & San Biagio, 160

2004; Barthès, Bulone, Manno, Martorana, & San Biagio, 2007). For this purpose about 700 161

intensity values were collected over different sample positions on a scattering volume 162

corresponding to a single speckle. The integration time (TI) for a single measurement was 1 s. 163

164

2.6. Data Analysis 165

166

7

Data distribution of the scattered intensity was analyzed with an expression (Manno, 167

Bulone, Martorana, & San Biagio, 2004; Barthès, Bulone, Manno, Martorana, & San Biagio, 2007) 168

derived by extending the Pusey and van Megen theory for non-ergodic media (Pusey, & van 169

Megen, 1989) to the case of no completely frozen-in systems. Indeed, the core of this expression is 170

the ratio between the coherence time of the scattering radiation and the integration time TI of the 171

measurement of the scattered intensity. As in the Pusey and van Megen approach, it was assumed 172

that the total scattered field E is the sum of a fast-fluctuating field EF due to small-size objects with 173

an average intensity IF and a coherence time F <TI, and a slowly decaying field ES due to large-size 174

slowly diffusing objects, with an average intensity IS and a coherence time S>TI. By taking into 175

account the ratio N = F /TI, it was shown that the total time-integrated intensity is given by two 176

statistically independent contributions: an exponentially slow contribution with average 177

NIII FSNG /2 , and a Gaussian distributed fast contribution with average NIII FFG /2 178

and variance NI G /22 . The expression for intensity distribution function is: 179

180

C

II

IIerf

I

IIIerf

I

II

IIP

NGNG

NGG

NG

NGG

NG

G

GTI

2exp

2

1

222exp

1)(

222

(1) 181

182

where dtexerfx

t

0

2

2)( is the error function and C1 is a normalization constant: 183

21

2

1

GI

erfC . 184

The expression is correct to the first order in F/TI and allows to distinguish between a fast 185

fluctuating and a slowly relaxing contribution even when their time scales are not completely 186

separated. 187

2.6. Scanning electron microscopy (SEM) 188

8

The SEM images were obtained using a FEI Quanta 400 FEG apparatus. Freeze-dried 189

hydrogels were mounted on appropriate stubs and examined under vacuum (50 Pa), with no need of 190

the gold coating technique, at an accelerating voltage of 15 KV. All images were acquired digitally 191

using xTmicroscope Control software and at 800x magnification. 192

193

3. Results and Discussion 194

195

3.1. Rheological investigations 196

197

The mechanical spectra of the LBG/Xanthan samples, prepared at different weight ratios in 198

“hot” and “cold conditions” recorded at 25 °C one day after and 6 days after the mixing of the two 199

polymer solutions (in the meanwhile the samples were kept at a constant temperature of 25 °C), are 200

shown in Fig. 1. 201

It is well known that both polymers, alone, are not capable to form gels, while, when mixed in 202

appropriate amounts, a three-dimensional network is formed. It is evident that, when the polymer 203

solutions are mixed in “cold conditions” the degree of entanglements and the number of effective 204

physical cross-links is deeply dependent on the weight ratio considered (see Fig. 1). If Xanthan is 205

the main component of the mixture (1:9), then the typical solution behaviour is observed: the loss 206

modulus is always higher than the storage modulus and both show a strong dependence on the 207

applied frequency. If the amount of Xanthan is reduced, but still predominant in comparison with 208

LBG (1:3), the characteristic gel point is observed. At the gel point the G’ and G” moduli overlap 209

and increase together, showing the typical power law behaviour (see below). Further decrease of the 210

Xanthan percentage in the mixture (1:1) leads to the formation of a gel, i.e., in the whole range of 211

explored frequencies (ca. four decades), G’ is always higher than G”. It must be underlined that, for 212

an appropriate comparison with LS experiments, the used polymer concentration is quite low (cp = 213

0.125 %). As a consequence the formed gel is a weak network: the difference between G’ and G” 214

9

values is less than a factor ten. Furthermore, the complete gel formation does not occur 215

instantaneously: consequently the time evolution of the moduli was followed and the measurements 216

were carried out also after six days (see Fig. 2A, B and C). For the 1:9 and 1:1 ratios an increase of 217

G’ is observed. For the 1:1 ratio the frequency dependence remains the same but the difference 218

between the two moduli increases with time indicating an increase of the number of the physically 219

active bonds. A similar behaviour is detected for the 1:9 ratio: the frequency dependence does not 220

change with time, the G” values are almost the same at 1 and 6 days while an increase with time is 221

observed for G’. Nevertheless this increase does not change the nature of the system that shows 222

always a solution behaviour (G’<G”). 223

On the other side, when the two polymer solutions are mixed in “hot conditions”, remarkable 224

differences are registered in the mechanical spectra (Fig. 1C and D and Fig. 2D, E and F). All 225

absolute values of the moduli are higher than those of the “cold samples” indicating that, when 226

Xanthan is in a disordered conformation, a more entangled network with a higher number of 227

effective cross-links can be formed in the presence of LBG. (see Fig. 2). For the sample with the 1:1 228

weight ratio, after one day, a gel is already formed, although a slight dependence from the 229

frequency can be observed for both moduli (Fig. 2D). After 6 days, an almost independence on the 230

frequency is monitored (Fig. 2D): this is an indication that the network evolves and more effective 231

physical cross-links are formed with time, which contribute to the elastic modulus. 232

Also for the network prepared at 1:9 ratio in “hot conditions”, an evolution is observed (Fig. 2F); 233

but, while for the “hot sample” a gel point occurs already after one day (Fig. 2F) (similar spectra 234

were registered for the 1:3 “cold sample”, Fig. 2B) the corresponding 1:9 “cold sample” still shows 235

mechanical spectra typical of solutions after 6 days (Fig. 2C). Clearly in the 1:9 “hot sample” the 236

Xanthan chains, being in the disordered conformation, are more capable to form physical 237

entanglements with the LBG chains in comparison with the 1:9 “cold sample” where the solution 238

behaviour is always observed (Fig. 2C). When a lower amount of Xanthan is present (1:3 ratio), for 239

10

the “hot samples” (Fig. 2E) the evolution of a very weak gel is monitored with time, while, for the 240

“cold samples”, a gel point transition is always observed (Fig. 2B). 241

Summarizing: the variations with time of the moduli are more significant in “hot samples” (Fig. 2D, 242

E and F). For the “cold samples” weak gel behaviour is monitored for the 1:1 ratio while gel 243

behaviour is monitored for the corresponding “hot samples” (Fig. 2A and D). The 1:9 ratio “hot 244

samples” show a gel point transition with a negligible evolution with time while, for the “cold 245

samples”, the corresponding spectra are those typical of solutions (Fig. 2C and F). For the 1:3 ratio 246

the “hot samples” are weak gels while the “cold samples” are at the gel point (Fig. 2B and E). 247

The 1:1 ratio is the most effective leading to a gel for both temperature preparation conditions 248

(tough weak gel is formed in the case of the “cold samples”). The difference detected between the 249

1:3 (gel point for the “cold samples”) and 1:9 ratios (gel point for the “hot samples”), indicates the 250

crucial role played by the conformation of the Xanthan chains. When a high amount of Xanthan 251

chains is present in double helix conformation (1:9, “cold samples”), a predominant liquid-like 252

behaviour is observed. On the other side, when the Xanthan chains are in the disordered 253

conformation, also for a high amount of the polysaccharide (1:9, “hot samples”), a gel point 254

transition is monitored. 255

It is interesting to point out that the critical exponents for gelation, n, are very close to the 256

theoretical one 0.5 (Chambon, Petrovic, MacKnight, & Winter, 1986) for both 1:3, “cold samples” 257

and 1:9 , “hot samples” (Table 1), i.e. for those samples showing a gel-point behaviour. 258

Actually such theoretical 0.5 value was firstly hypothesized by Winter and Chambon (Winter, & 259

Chambon,1986) as the universal parameter for the sol-gel transition, nevertheless later works 260

clarified that there is no universal value for n that can assume any value between 0 and 1, depending 261

on several parameters, such as cross-linker amount, polymer concentration, molecular weight, and 262

the specific pre-gel history (Coviello, & Burchard, 1992; Grisel, & Muller, 1998; Hsu, & Jamieson, 263

1993; Matricardi, Dentini, Crescenzi, 1993; Michon, Cuvelier, & Launay, 1993; Richter, Boyko, & 264

Schröter, 2004; Richter, Boyko, Matzker, & Schröter, 2004; Sandolo, Matricardi, Alhaique, & 265

11

Coviello, 2009; Winter, Izuka, & De Rosa, 1994). Usually, in a cross-linking reaction, just beyond 266

the gel point, G' levels off as the frequency decreases, and G" is proportional to . At later stages of 267

the reaction, the equilibrium modulus increases and G' and G" depart from each other. This is 268

actually what we observe for our “cold samples” at different weight ratios: the solution behaviour 269

for the 1:9, the gel point behaviour for the 1:3 and the gel behaviour for the 1:1 samples (Fig. 1B). 270

In the case of “hot samples” the solution behaviour is not detectable but gel point, weak gel and gel 271

behaviours are monitored for the 1:9, for the 1:3 and for 1:1 samples, respectively (Fig. 1D). 272

The critical exponent value of 0.5 detected for the two cases above reported (1:9 “hot samples” and 273

1:3 “cold samples”) suggests that, in accordance with the critical exponent found for stoichiometric 274

chemical crosslinkings, the networks are based on “one to one” interactions between the LBG and 275

Xanthan chains. In the case of the 1:9 “cold sample” there are actually too few LBG chains able to 276

build up a three-dimensional network (Fig. 2C). As LBG increases (1:3 “cold sample”) a gel point 277

is observed indicating that a sufficient number of LBG chains can interact with Xanthan (Fig. 2B). 278

Finally, for the 1:1 “cold sample” a real weak gel is obtained (Fig. 2A). In conclusion it can be 279

assumed that, for the “cold samples”, the network is mainly dominated by LBG and the network 280

formation is mediated by the addition of an appropriate amount of Xanthan chains. Furthermore, the 281

usual evolution with time of the two moduli (and correspondingly of the mechanical spectra) 282

observed for a chemical cross-linking reaction, is detected here by changing the weight ratio 283

between the two polymers (Fig. 1B). 284

A quite different situation occurs for the “hot samples”: at 75 °C LBG keeps its random coil 285

conformation (present at room temperature) but also Xanthan is dispersed as single chains. 286

Actually, Xanthan is not allowed to renaturate during cooling, because it is implied in the network 287

formation with the LBG chains. In this case no solution behaviour is ever detected: the 1:9 “hot 288

sample” is already at the gel point (Fig. 2F), thus showing that the different conformation assumed 289

by Xanthan plays a crucial role in the building up of the network. For the other ratios, 1:3 and 1:1, 290

weak gel and gel behaviours are monitored (Fig. 2E and 2D). 291

12

For the LBG/Xanthan cross-linking reaction it is also possible to assume an analogy between the 292

weight ratio and the reaction time and to report the loss tangent (tan ) as a function of the weight 293

ratio (instead of the reaction time). In fact, according to the definition of the gel point the loss 294

tangent is given (Kramers-Kronig relation) by tan = tan (n/2) = G"/G' (Ferry, 1980), indicating 295

that tan , at the sol-gel transition, is independent on and its value depends only on n. Fig. 3A 296

shows the results obtained by dynamic frequency sweep measurements for some selected ratios 297

(1:9, 1:3 and 1:1 “cold samples”) by plotting tan measured at various frequencies as a function of 298

the LBG weight ratio. According to the Winter-Chambon criterion (Winter, Morganelli, & 299

Chambon, 1988) the weight ratio at which tan curves converge is defined as the gel point. For the 300

samples, prepared in “cold conditions”, this value is practically coincident with the 1:3 weight ratio 301

and is equal to 1. The frequency-dependent measurement at the gel point should exhibit the same 302

slope for G’ and G”, as shown in Fig. 1B and reported in Table 1. Thus, for the LBG/Xanthan 303

weight ratios corresponding to the gel point, 1:3 “cold sample” and 1:9 “hot sample”, tan ≈ 1 (and 304

therefore n ≈ 0.5), as shown in Fig. 3B, for both systems, in the whole explored frequency domain. 305

Furthermore, the power law observed at the gel point implies a selfsimilar structure of the network 306

(Muthukumar,1985; Vilgis, & Winter, 1988); thus, it can be assumed that also the LBG/Xanthan 307

networks assume a selfsimilar structure when the gel point is detected (1:3 “cold sample” and 1:9 308

“hot sample”), and such connectivity can be related to the fractal dimension, df. In fact, according to 309

an important, and still valid work (Muthukumar, 1989), in the case that the excluded volume effect 310

is fully screened the following relation holds for a polydisperse system: f

f

dd

dddn

22

22, where 311

d is the space dimension (i.e. 3). For the tested systems (see Table 1) at the gel point (1:3 “cold 312

sample” and “1:9 “hot sample”) n ≈ 0.5 and the corresponding df value is ≈ 2, the same of an 313

isolated linear chain of infinite length under conditions (Muller, Gérard, Dugand, Rempp, & 314

Gnanou, 1991). This df value could be surprising for a chemical cross-linking reaction where at the 315

gel point the polymer is a highly branched macromolecule, but it can be reasonable for a physical 316

13

gel. In this case the critical state is not due to the simultaneous chemical reaction of many 317

monomers, but to the physical interactions among chains of quite long contour length (order of 318

hundred of nm). As a consequence the build up of peculiar structures is observed, as those typical of 319

biological networks that rarely can be strictly identified within the theoretical frameworks of 320

electrical analogue or tree-models. 321

322

3.2. Light Scattering investigations 323

324

325

The intensity autocorrelation functions (TCF) of LBG/Xanthan samples prepared at the three 326

different weight ratios in “hot” and “cold conditions”, were acquired 6 days after the preparation. 327

DLS measurements were carried out by collecting the signal on different regions of each sample to 328

take into account the non-ergodicity of gel systems. Results are shown in Fig. 4. Samples with a 329

higher content of LBG display higher values of a no-decay component, which is an indication of the 330

presence of a higher fraction of scattered light coming from scatterers unable to relax over the 331

applied experimental time scale. For a given weight ratio of LBG/Xanthan, “hot samples” appear 332

always more dynamically frozen than the cold ones. By comparing the time dependence of TCFs it 333

can be noted that the dynamics of samples 1:1, either “hot” or “cold”, and 1:3 “hot”, appears to be 334

almost entirely frozen, whereas the autocorrelation function of sample 1:9 “cold” displays a liquid-335

like character. Furthermore, samples 1:3 “cold” and 1:9 “hot” follow a power-law dependence, g(2) 336

t-

, with an exponent of 0.48 and 0.44, respectively. This power-law behavior is usually taken as a 337

mark of the gelation threshold (Shibayama, & Norisuye, 2002). The same two samples were seen to 338

be at the gelation point with a critical exponent close to 0.5 by rheological results (see the 339

“Rheological investigations” section). Although a relation between critical exponents obtained by 340

DLS and rheology is predicted by the Doi and Onuki theory (Richter, 2007), the critical exponents 341

obtained by means of light scattering measurements were, in the present study, lower than those 342

derived by rheology, as already observed by other researchers for several different systems (Richter, 343

2007). 344

14

The dynamic properties of the samples were also investigated by looking at the statistical properties 345

of the scattered light. Indeed, gel systems are characterized by the presence of very slow 346

fluctuations in the average intensity reflecting the reduced speckle mobility (Rodd, Dunstan, Boger, 347

Schmidt, & Burchard, 2001) and due to either dynamical or structural network inhomogeneities 348

(Richert, 2002). For each scattering angle the intensity from single speckles was collected with an 349

integration time of 1 s over different sample positions. Data were then normalized by their mean 350

value and the data distribution was fitted to the convolution of a Gaussian with an exponential 351

function (see Eq. (1)), which allowed to distinguish the fast fluctuating and slowly relaxing 352

contribution. The fraction of the fast contribution to the scattered light at different q-vector values is 353

shown in Fig 5A. A lower fraction of the fast relaxing contribution, XF, is observed at higher LBG 354

content, with the “hot samples” being more frozen than the “cold” ones, consistently with the DLS 355

results. The decrease of XF with q lowering implies that density fluctuations occurring over larger 356

length scale are more hindered. This was observed in all samples, including LBG 1:9 “cold”, that 357

appears liquid in the range of several ten nanometers. 358

Information on the structural properties can be provided by the dependence of the scattered light 359

intensity on q wave-vector. Indeed, for a system of molecules in solution, the q-dependence of the 360

scattered light is given by: 361

I(q) P(q) S(q) 362

where P(q) is the form factor of the single molecule and S(q) is the structure factor giving the 363

intensity contribution due to the spatial correlation between molecules in solution. In the case of 364

fractal aggregates, S(q) follows a power-law (Thill, Lambert, Moustier, Ginestet, Audic, & Bottero, 365

2000) of the form fdqqS

)( . The law is valid only over a length scale smaller than the aggregate 366

dimension (R) and larger than the dimension of the primary particles assembled in the aggregate 367

(r0), where P(q) 1. This leads to the well known relation fdqqI

)( with r0 << 1/q << R often 368

used for determining the fractal dimension. The q-dependence of the scattered intensities for 369

15

LBG/Xanthan samples was obtained by averaging the signal over different regions of the sample. 370

Data for each sample, normalized for the weight average molecular weight, are shown in Fig. 5 B. 371

Except for the absolute value, all samples display a very similar I(q) profile, despite their different 372

dynamical behavior. A sign of curvature can be envied in S(q) profile at about 6x104 cm

-1 373

(corresponding to 170 nm) that could signal the approaching to some characteristic length of the 374

system. At higher q values, I(q) follows a power-law with a slope of 2.4 ± 0.2. This value of fractal 375

dimension indicates a self-similar and quite compact structure, somehow larger, but anyhow 376

comparable, than that obtained from rheology and DLS for samples 1:3 “cold” and 1:9 “hot”. 377

However, sample polydispersity may severely affect the accuracy of the fractal dimension 378

determination by means of this type of data, since aggregates of different size can contribute to 379

scattered intensity with a different form factor. The similarity of I(q) profile for the different 380

samples suggests that the structure is almost the same, at least on the length-scale ranging from two 381

hundred down to few nanometers. A further comment can be made on the variation of the intensity 382

absolute value for the different samples. Actually, we observe that the intensity increases with the 383

increase of LBG content in the samples and the increase is even higher for those systems behaving 384

as “strong” gels, although the average spatial coordination inside the aggregates does not 385

appreciably change, suggesting the formation of ticker or denser junction regions inside the same 386

type of aggregates. 387

388

3.3. Scanning electron microscopy characterization 389

390

The SEM micrographs, taken six days after preparation on freeze-dried samples, support the 391

considerations above discussed. In Fig. 6 the photographs of the two single polymers (LBG and 392

Xanthan) and of the two samples prepared at low and high temperature with a weight ratio of 1:1 393

are shown. The last two systems are both gels, as indicated by the frequency spectra and by the 394

TCFs, but the different strength of the networks, estimated with the absolute values of the elastic 395

16

moduli and by the decay behaviour of the two TCFs, is clearly evidenced by the different texture 396

observed in the pictures. 397

It is interesting to appreciate the variations between the initial aspect of the single polymers, and the 398

final situations, when the synergistic effects, deeply influenced by the preparation temperature, 399

change dramatically also the macroscopic morphology of the mixed systems. 400

401 402

4. Conclusions 403 404

The LBG/Xanthan system leads to different types of network depending on the ratio of the two 405

polymers and on the preparation temperature. In particular, for the 1:1 ratio a gel is always 406

obtained; for the 1:3 ratio a weak gel is observed in “hot conditions” while a gel point is detected in 407

“cold conditions”; for the 1:9 ratio the gel point is monitored in “hot conditions” while the solution 408

is present in “cold conditions”. These results indicate how a fine tuning of the properties of the 409

mixed gels is possible through the variation of the preparation temperature and/or the weight ratio 410

between the two polymers. 411

Interesting is the qualitative agreement between the two approaches: the TCFs of 1:1 samples (“hot” 412

and “cold” samples) do not relax in the explored time window and similarly the corresponding 413

mechanical spectra are those of a gel in the whole explored frequency domain. Also for the other 414

two weight ratios a correspondence between the rheology and the DLS can be appreciated. 415

Furthermore, looking at the values of Table 1, it is possible to see the good accordance between the 416

fractal dimensions, obtained with the rheology approach, and the DLS technique for the two 417

systems that show sol-gel transition in the mechanical spectra and power law behaviour in DLS. 418

The slight discrepancy between the two critical exponents is reasonable: the theory mainly deals 419

with chemical network formation while in the present case physical interactions lead to the gel: 420

furthermore, the light scattering is a non perturbative technique while a periodic deformation is 421

applied in rheology partly disturbing the crosslinking process with a possible shift in the 422

determination of the infinite cluster formation and detection. 423

17

424

Acknowledgements 425

Financial supports from FIRB, Fondo per gli Investimenti della Ricerca di Base, Research Program: 426

Ricerca e Sviluppo del Farmaco (CHEM-PROFARMA-NET), grant no. RBPR05NWWC_003 is 427

acknowledged. 428

429

430

References 431 432

433

Baichwal, A. R. (2004). Sustained-release matrix for high-dose insoluble drugs. US Patent Office, 434

Pat. No. 2004121012. 435

Barthès, J., Bulone, D., Manno, M., Martorana, V., & San Biagio, P.L. (2007). A statistical light 436

scattering approach to separating fast and slow dynamics. European Biophysical Journal, 36, 437

743-752. 438

Cairns, P., Miles, M. J., Morris, V. J., & Brownsey, G. J. (1988). X-Ray fibre-diffraction studies of 439

synergistic, binary polysaccharide gels. Carbohydrate Research, 176, 329-334. 440

Chambon, F., Petrovic, Z. S., MacKnight, W. J., & Winter, H. H. (1986). Rheology of model 441

polyurethanes at the gel point. Macromolecules, 19, 2146-2149. 442

Coviello, T., Kajiwara, K., Burchard, W., Dentini, M., & Crescenzi, V. (1986). Solution properties 443

of Xanthan. 1. Dynamic and static light scattering from native and modified Xanthans in 444

dilute solutions. Macromolecules, 19, 2826-2831. 445

Coviello, T., & Burchard, W. (1992). Criteria for the point of gelation in reversibly gelling systems 446

according to dynamic light scattering and oscillatory rheology. Macromolecules, 25, 1011-447

1012. 448

Dea, I. C. M., Clark, A. H., & McCleary, B. V. (1986). Effect of galactose-substitution-patterns on 449

the interaction properties of galactomannans. Carbohydrate Research, 147, 275-294. 450

18

Dolz, M., Hernández, Delegido, J., Alfaro, M. C., & Muňoz, J. (2007). Influence of xanthan gum 451

and locust bean gum upon flow and thixotropic behaviour of food emulsions containing 452

modified starch. Journal of Food Engineering, 81, 179-186. 453

Fadden, D. J., Kulkarni, N. M., & Sorg, A. F. (2004). Fast dissolving orally consumable films 454

containing a modified starch for improved heat moisture resistance. US Patent Office, Pat. No. 455

2004247648. 456

Ferry, J. D. (1980). Viscoelastic Properties of Polymers. (3rd

ed.) New York: Wiley. 457

Grisel, M., & Muller, G. (1998). Rheological properties of the schizophyllan-borax system. 458

Macromolecules, 31, 4277-4281. 459

Hacche, L. S., Washington, G. E., & Brant, D. A. (1987). Light-scattering investigation of the 460

temperature-driven conformation change in Xanthan. Macromolecules, 20, 2179-2187. 461

Hsu, S. H., & Jamieson, M. (1993). Viscoelastic behaviour at the thermal sol-gel transition of 462

gelatin. Polymer, 34, 2602-2608. 463

Kok, M. S., Hill, S. E., & Mitchell, J. R. (1999). Viscosity of galactomannans during high 464

temperature processing: influence of degradation and solubilisation. Food Hydrocolloids, 13, 465

535–542. 466

Lang, P., & Burchard, W. (1991). Dynamic light scattering at the gel point. Macromolecules, 24, 467

814-815. 468

Lapasin, R, & Pricl, S. (1995). Rheology of industrial polysaccharides: theory and applications. 469

London: Blackie Academic & Professional. 470

Lundin, L., & Hermansson, A. M. (1995). Supermolecular aspects of Xanthan-locust bean gum gels 471

based on rheology and electron microscopy. Carbohydrate Polymers, 26, 129-140. 472

Mandala, I., Kapetanakou, A., & Kostaropoulos, A. (2008). Physical properties of breads containing 473

hydrocolloids stored at low temperature: II-Effect of freezing. Food Hydrocolloids, 22, 1443-474

1451. 475

19

Mannion, R.O., Melia, C. D., Mitchell, J. R., Harding, S. E., & Green, A. P. (1991). Effect of 476

xanthan/locust bean gum synergy on ibuprofen release from hydrophilic matrix 477

tablets. Journal of Pharmacy and Pharmacology, Supplement, 43, 79P-79P. 478

Manno, M., Bulone, D., Martorana, V. & San Biagio, P.L. (2004). Ergodic to non-ergodic transition 479

monitored by scattered light intensity statistics. Physica A, 341, 40-54. 480

Matricardi, P., Dentini, M., & Crescenzi, V. (1993). Rheological gel-point determination for a 481

polysaccharide system undergoing chemical cross-linking. Macromolecules, 26, 4386-4387. 482

Michon, C., Cuvelier, G., & Launay, B. (1993). Concentration dependence of the critical 483

viscoelastic properties of gelatin at the gel point. Rheologica Acta, 32, 94-103. 484

Morris, E.R. (1990). Mixed polymer gels. In P. Harris (Ed.). Food Gels (pp 291-359). London, UK: 485

Elsevier. 486

Muller, R., Gérard, E., Dugand, P., Rempp, P., & Gnanou, Y. (1991). Rheological characterization 487

of the gel point: a new interpretation. Macromolecules, 24, 1321-1326. 488

Muthukumar, M. (1985). Dynamics of polymeric fractals. Journal of Chemical Physics, 83, 3161-489

3166. 490

Muthukumar, M. (1989). Screening effect on viscoelasticity near the gel point. Macromolecules, 22, 491

4656-4658. 492

Ojinnaka, C., Brownsey, J. G., .Morris, & E. R., Morris, V. J. (1998). Effect of deacetylation on the 493

synergistic interaction of acetan with locust bean gum or konjac mannan, Carbohydrate 494

Research, 305, 101-108. 495

Pedersen, J. K. (1980). Carrageenan, pectin and xanthan/locust bean gum gels. Trends in their food 496

use. Food Chemistry, 6, 77-88. 497

Pusey, P., & van Megen, W. (1989). Dynamic light scattering by non-ergodic media. Physica A, 498

157, 705-741. 499

Ramírez, J. A., Barrera, M., Morales, O. G., & Vázquez, M. (2002). Efefct of xanthan and locust 500

bean gums on the gelling properties of myofibrillar protein. Food Hydrocolloids, 16, 11-16. 501

20

Richert, R. (2002). Heterogeneous dynamics in liquids: fluctuations in space and time. Journal of 502

Physics: Condensed Matter, 14, R703–R738. 503

Richter, S., Boyko, V., & Schröter, K. (2004). Gelation studies on a radical chain cross-linking 504

copolymerization process: comparison of the critical exponents obtained by dynamic light 505

scattering and rheology. Macromolecular Rapid Communication, 25, 542-546. 506

Richter, S., Boyko, V., Matzker, R., & Schröter, K. (2004). Gelation studies: comparison of the 507

critical exponents obtained by dynamic light scattering and rheology, 2a A thermoreversible 508

gelling system: mixtures of xanthan gum locust bean gum. Macromolecular Rapid 509

Communication, 25, 1504-1509. 510

Richter, S., Matzker, R., & Schröter, K. (2005). Gelation studies, 4 Why do “classical” methods like 511

oscillatory shear rheology and dynamic light scattering for characterization of the gelation 512

threshold sometimes not provide identical results especially on thermoreversible gels? 513

Macromolecular Rapid Communication, 26, 1626-1632. 514

Richter, S., Brand, T., Berger, S. (2005). Comparative monitoring of the gelation process of a 515

Thermoreversible gelling system made of xanthan gum and locust bean gum by dynamic light 516

scattering and 1H NMR spectroscopy. Macromolecular Rapid Communication, 26, 548-553. 517

Richter, S. (2007). Comparison of Critical Exponents Determined by Rheology and Dynamic Light 518

Scattering on Irreversible and Reversible Gelling Systems. Macromolecular Symposia, 256, 519

88-94. 520

Rodd, A. B., Dunstan, D. E., Boger, D. V., Schmidt, J. & Burchard, W. (2001). Heterodyne and 521

Nonergodic Approach to Dynamic Light Scattering of Polymer Gels: Aqueous Xanthan in the 522

Presence of Metal Ions (Aluminum(III)). Macromolecule, 34, 3339-3352. 523

Rolf, D. (2003). Aqueous gel wound dressing and package. US Patent Office, Pat. No. 6620436. 524

Sahin, H., & Ozdemir, F. (2004). Effect of some hydrocolloids on the rheological properties of 525

different formulated ketchups. Food Hydrocolloids, 18, 1015-1022. 526

21

Sandolo, C., Matricardi, P., Alhaique, F., & Coviello, T. (2009). Effect of temperature and cross-527

linking density on rheology of chemical cross-linked guar gum at the gel point. Food 528

Hydrocolloids, 23, 210-220. 529

Secouard, S., Malhiac, C., Grisel, M., & Decroix, B. (2003). Release of limonene from 530

polysaccharide matrices: viscosity and synergy effect. Food Chemistry, 82, 2227-234. 531

Shibayama, M., & Norisuye, T. (2002). Gel Formation Analyses by Dynamic Light Scattering. 532

Bulletin of the Chemical Society of Japan, 75, 641-659. 533

Sugden, K. (1989). Buccal etorphine tablets with xanthan and locust bean gums. US Patent Office, 534

Pat. No. 4829056. 535

Takenata, M., Kobayashi, T., Hashimoto, T., & Takahashi, M. (2002). Time evolution of dynamic 536

shear moduli in a physical gelation process of 1,3:2,4-bis-O-(p-methylbenzylidene)- D-537

sorbitol in polystyrene melt: critical exponent and gel strength. Physical Review E, 65, 538

041401-1-7. 539

Thill, A., Lambert, S., Moustier, S., Ginestet, P., Audic, J. M., & Bottero, J. Y. (2000). Structural 540

Interpretations of Static Light Scattering Patterns of Fractal Aggregates. Journal of Colloid 541

and Interface Science, 228, 386–392. 542

Ughini, F., Andreazza, I. F., Ganter, J. L. M. S., & Bresolin, T. M. B. (2004). Evaluation of 543

Xanthan and highly substituted galactomannan from M. scabrella as a sustained release 544

matrix. International Journal of Pharmaceutics, 271, 197-205. 545

Venkataraju, M. P., Gowda, D. V., Rajesh, K. S., & Shivakumar, H. G. (2008). Xanthan and locust 546

bean gum (from Ceratonia siliqua) matrix tablets for oral controlled delivery of metoprolol 547

tartrate. Current Drug Therapy, 3, 70-77. 548

Vilgis, T. A., & Winter, H.H. (1988). Mechanical selfsimilarity of polymers during chemical 549

gelation. Colloid and Polymer Science, 266, 494-500. 550

Wang, E., Wang, Y. J., & Sun, Z. (2002). Conformational role of xanthan in its interaction with 551

locust bean gum. Journal of Food Science, 67, 2609-2614. 552

22

Winter, H. H., & Chambon, F. (1986). Analysis of linear viscoelasticity of a crosslinking polymer 553

at the gel point. Journal of Rheology, 30, 367-382. 554

Winter, H. H., Morganelli, P., & Chambon, F. (1988). Stoichiometry effects on rheology of model 555

polyurethanes at the gel point. Macromolecules, 21, 532-535. 556

Winter, H. H., Izuka, A., & De Rosa, M. E. (1994). Experimental observation of the molecular 557

weight dependence of the critical exponents for the rheology near the gel point. Polymer Gels 558

and Networks, 2, 239-245. 559

Zhan, D. F., Ridout, M. J., Brownsey, G. J., & Morris, V. J. (1993). Xanthan-locust bean gum 560

interactions and gelation. Carbohydrate Polymers, 21, 53-58. 561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

23

595

Legends to the figures 596

597

Scheme1. Repeating units of LBG (left) and Xanthan (right). 598

599

Fig. 1. Comparison among the mechanical spectra of LBG/Xanthan samples prepared at different 600

weight ratio (1:1, 1:3 and 1:9), prepared at two different temperatures (A, B : T=25 °C; C, D: T= 75 601

°C) and recorded at 25 °C one day (A and C) and six days (B and D) after the preparation. 602

603

Fig. 2. Comparison among the mechanical spectra of LBG/Xanthan recorded at 25 °C one day and 604

six days after the preparation at different weight ratio. 605

“Cold samples”: A: 1:1; B: 1:3 and C: 1:9. “Hot samples”: D: 1:1; E: 1:3 and F: 1:9. 606

607

Fig. 3. A) tan of LBG/Xanthan samples prepared in “cold conditions” and recorded at 25 °C six 608

days after the preparation at different frequencies as a function of the content of LBG (e.g., LBG = 609

0.50 corresponds to 1:1 weight ratio). B) tan of LBG/Xanthan samples prepared in “hot” and 610

“cold conditions” as a function of frequency for the 1:3 and 1:9 ratios. 611

612

Fig. 4. Intensity autocorrelation functions of LBG/Xanthan samples obtained 6 days after their 613

preparation. 614

615

Fig. 5. Fraction of the fast relaxing contribution to scattered light vs. wave-vector q estimated for 616

the LBG/Xanthan studied at different weight ratio (A). Log-log plot of the average scattered light 617

intensity, measured for the LBG/Xanthan at different weight ratio, vs. q-vector. Data were 618

normalized by toluene scattering and the weight average molecular weight (B). 619

620

Fig. 6. Scanning electron micrograph of surface morphology of freeze-dried samples of LBG, 621

Xanthan (top), and LBG/Xanthan 1:1 (bottom) prepared at 75 and 25 °C (magnification 800x). 622

623

Table 1

C. Sandolo et al.

Table 1

Apparent exponents relative to the two moduli, n’ and n”, evaluated for two weight ratios, power

law exponents from the TCFs and the corresponding fractal dimensions, df.

------------------------------------------------------------------------------------------------------------------------

Rheology Light Scattering

------------------------------------------------------------------------------------------------------------------------

“cold samples” “hot samples” “cold samples” “hot samples”

1:3 1:9 1:3 1:9

------------------------------------------------------------------------------------------------------------------------

time (days) n’ n” n’ n”

------------------------------------------------------------------------------------------------------------------------

1 0.56 0.54 0.53 0.59

6 0.47 0.43 0.50 0.56 0.48 0.44

time (days) df (n’) df (n”) df (n’) df (n”) df df

------------------------------------------------------------------------------------------------------------------------

1 1.92 1.95 1.96 1.89

6 2.04 2.08 2.00 1.93 2.08 2.27

------------------------------------------------------------------------------------------------------------------------

Table 1

Scheme 1

C. Sandolo et al.

Locust Bean Gum Xanthan

O

O OH

HO

OO

HOOH

O

HO

O

O

HO

HO

O

O

-O2C

O

HOOH

-O2C

OO HO

OH

O

O

O

OHO

HO

O

HO

O

HO

HOOH

O

OHO

HO

O

O

O

HO

HOOH

O

OH

HO

OH

OH

Scheme 1

Figure 1

C. Sandolo et al.

1 day, cold

0.001

0.01

0.1

1

10

0.001 0.01 0.1 1 10f (Hz)

G', G'' (Pa)

G' 1:1

G'' 1:1

G' 1:3

G'' 1:3

G' 1:9

G'' 1:9

A 6 days, cold

0.001

0.01

0.1

1

10

0.001 0.01 0.1 1 10f (Hz)

G', G'' (Pa)

G' 1:1

G'' 1:1

G' 1:3

G'' 1:3

G' 1:9

G'' 1:9

B

1day, hot

0.01

0.1

1

10

100

0.001 0.01 0.1 1 10f (Hz)

G', G'' (Pa)

1:1 G'1:1 G''1:3 G'1:3 G"1:9 G'1:9 G"

C 6 days, hot

0.01

0.1

1

10

100

0.001 0.01 0.1 1 10f (Hz)

G', G'' (Pa)

1:3 G'1:3 G''1:1 G'1:1 G"1:9 G'1:9 G"

D

Figure 1

Figure 2

C. Sandolo et al.

1:1 cold sample

0.1

1

10

0.001 0.01 0.1 1 10f (Hz)

G', G'' (Pa)

G' (1:1) 6 days

G'' (1:1) 6 days

G' (1:1) 1 day

G'' (1:1) 1 day

A

1:3 cold sample

0.01

0.1

1

10

0.001 0.01 0.1 1 10f (Hz)

G', G''(Pa)

G' (1:3) 6 days

G'' (1:3) 6 days

G' (1:3) 1 day

G'' (1:3) 1 day

B

1:9 cold sample

0.001

0.01

0.1

1

0.001 0.01 0.1 1 10f (Hz)

G', G'' (Pa)

G' (1:9) 6 days

G'' (1:9) 6 days

G' (1:9) 1 day

G'' (1:9) 1 day

C

1:1 hot sample

0.1

1

10

100

0.001 0.01 0.1 1 10f (Hz)

G', G''(Pa)

G', 6 days G'', 6 days

G', 1 day G", 1 day

D

1:3 hot sample

0.1

1

10

0.001 0.01 0.1 1 10f (Hz)

G',G'' (Pa)

G', 1 day G'', 1 day

G', 6 days G'', 6 days

E

1:9 hot sample

0.01

0.1

1

0.001 0.01 0.1 1f (Hz)

G',G''(Pa)

G', 1 day G'', 1 day

G', 6 days G'', 6 days

F

Figure 2

Figure 3

C. Sandolo et al.

6 days, cold

0

1

2

3

4

5

6

7

8

9

0.0 0.1 0.2 0.3 0.4 0.5 LBG (%)

tan

1 Hz

0.63 Hz

0.40 Hz

0.25 Hz

0.16 Hz

0.1 Hz

0.0631 Hz

0.0398 Hz

A

B

0

1

2

3

4

0.001 0.01 0.1 1 10f (Hz)

tan1:3 cold 6 days

1:9 hot 6 days

1:3 cold 1 day

1:9 hot 1 day

Figure 3

Figure 4

C. Sandolo et al.

(s)

100 101 102 103 104 105 106 107

g(2

) ()

-1

0.01

0.1

1

1:1 hot

1:1 cold

1:3 hot

1:3 cold

1:9 hot

1:9 cold

Figure 4

Figure 5

C. Sandolo et al.

q (cm-1)

105 106

I N (

arb

. un.)

101

102

103

104

105

q(cm-1)

1e+5 2e+5 3e+5

XF

0.0

0.2

0.4

0.6

0.8

1.0

1:1 hot

1:1 cold

1:3 hot

1:3 cold

1:9 hot

1:9 cold

A B

Figure 5

Figure 6

C. Sandolo et al.

LBG Xanth

LBG/Xanth 1:1 hot sample LBG/Xanth 1:1 cold sample

Figure 6