Shaken but not Stirred: The Formation of Reversible Particle - Polymer Gels under

Shear

M. Mar Ramos-Tejada1* and Paul F. Luckham2

1Dep. Física. Universidad de Jaén. Escuela Politécnica Superior de Linares.

Universidad de Jaén. mmramos@ujaen.es

2Dept Chem. Eng. and Chem. Tech., Imperial College of Science, Technology and

Medicine. Prince Consort Road. London SW7 2BY. United Kingdom.

p.luckham01@imperial.ac.uk

ABSTRACT

Some dispersions of clay and silica particles in water in the presence of relatively high

molecular weight polyethyleneoxide (PEO) which are fluid when at rest, become solid-

like after a quick shake to such an extent that they can be held in the hand. On leaving

the dispersions for a certain period of time, minutes to days depending on the polymer

molecular weight and concentration, the dispersions become liquid-like again. These

dispersions have been called “shake gels”, and a number of physical variables are

determinant for producing the gel and controlling its behavior. In this work, we have

studied the effect of the shape and size of the particles and of the PEO molecular

weights. To that aim, we have mapped the “phase” behavior of silica (Ludox TM50),

montmorillonite and laponite dispersions in presence of PEO of different molecular

weight. Shake gels are formed under certain concentrations of particles and PEO. The

necessary degree of particle coverage for shake gel formation seems to depend on the

particle shape. Whereas in the case of disc-shaped particles this limit is around the

saturation concentration, in the case of spherical particles the limit is around 2/3 of the

particle surface saturation. On the other hand, we observe some differences between

montmorillonite (micrometer-size particles) and laponite (nanometric particles)

dispersions. When we shake the former we find in some cases an important and

irreversible phase separation; on the contrary, in the case of the laponite dispersions, the

phase separation is far less frequent and extensive. Finally, we have found that the

nature of the applied shear field has a profound effect on the sample behavior. When we

place the dispersions in a conventional rheometer and shear them at moderate shear rate,

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

for many minutes the gel is not formed. However, shaking it by hand or extruding it

through a syringe brings about this effect in a matter of seconds.

Keywords: Shake gel; Colloidal silica Ludox; Laponite; Montmorillonite;

Poly(ethyleneoxide).

Introduction

Polymer adsorption to particles as a method of enhancing colloid stability, due the role

of polymers in stabilizing particles against aggregation, has been extensively studied for

at least the last five decades. It is also well known that under certain circumstances

polymers can induce the aggregation of particles either by simultaneously adsorbing on

two particles (bridging flocculation) or due to the presence of non adsorbing polymer

giving rise to depletion flocculation. Certain formulations where particles are dispersed

in a relatively high molecular weight polyethyleneoxide (PEO) solution are perfectly

stable for many days, giving a low viscosity solution. The important point is that a gel is

formed upon shaking, reaching, in some cases, with the consistency of toffee. On

standing the gel relaxes back into the liquid state over the course of seconds to days,

depending on the formulation. This shear effect has been observed in watery mixtures of

laponite and PEO [1, 2] and silica particles in presence of adsorbing PEO [3, 4]. As a

possible explanation for this phenomenon, Pozzo and Walker, 2004 [1] suggested that

the PEO adsorbs on the particles; when the surface is not fully saturated, the PEO can

bridge several particles causing the formation of aggregates. Application of shake can

desorb some bonds of the PEO and reabsorb them onto other particles, causing the

aggregates to grow substantially.

A number of physical variables are necessary for producing the gel and controlling its

behavior. Liu et al. [3] and Cabane et al. [4] studied how depending on the adsorption of

the number of silica particles per PEO macromolecular chain, and the free polymer

equilibrium concentration, quite different rheological behaviours can be observed in the

silica PEO mixtures such as rheopectic shear-induced gelation, Newtonian flow or

thixotropic shear-thinning. Ye et al [5] studied the effect of de particle polydispersity on

the structure and dynamics of silica-PEO mixtures. They found that mixtures of PEO

with polydisperse silica particles showed strong shear-thickening upon shaking, while

mixtures of monodisperse silica did not. In addition, there are some studies [1, 2] about

2

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

the “phase behaviour” of laponite-PEO system. Shake-gels are observed for PEO

concentrations slightly below the threshold of complete saturation of the laponite

particles by the polymer. Can and Okay [6] investigated the effect of the molecular

weight of PEO on the properties of laponite-PEO shake gels. They found that increasing

molecular weight shifts the critical range of average number of polymer chains adsorbed

by a laponite particle for the gel formation towards smaller values. Moreover, at high

molecular weight PEOs only weak gels are form. Saito et al. [7] studied the rheological

behavior of silica-PEO suspensions. They found that suspensions of silica particles in

PEO solution show shear-thickening profiles in steady shear, and highly elastic

responses in oscillatory shear, under large strains.

The nature of the applied shear field has an important effect on the shake gel formation.

When the samples are subjected to simple shear in Couette flow it is found that the gels

are not induced up to shear rates of thousands of s-1 [3]. However, the gelation phase

transition is observed with moderate manual agitation, which corresponds to a

significantly smaller shear rate (< 100 s-1) [1].

In this work, we study the effect of the shape and size of the particles on the shake gel

formation. To that aim, we have mapped the “phase” behavior of silica (Ludox TM50),

montmorillonite and laponite dispersions in presence of PEO. Moreover, we also

analyze the fact that extruding the dispersion through a syringe brings about the gel

formation in a reproducible manner. We use these data to try to advance the

understanding of the underlying mechanism that causing the shake gel formation.

Experimental

Materials

For our experiments, we have used colloidal silica, Ludox ®TM-50 (Aldrich Chemical

Company) with according to the manufacturer, specific surface area of ~140 m2/g and

average diameter of 30-34 nm [8, 9]. Laponite RD is a synthetic hectorite-like clay

manufactured by Laporte Industries Ltd (UK). It was used throughout this work without

further purification. Laponite consists of small platelets of approximately 30 nm x 1 nm

dimensions [10, 11]. Its surface area is approximately 350 m2/g, as determined by BET

Argon adsorption measurements according to the manufacturer. Sodium

3

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

montmorillonite was a commercial material supplied by Laporte Absorbents, sold under

the trade name Gelwhite H-NF. The montmorillonite consists of plates of particles

approximately 1 micron in diameter and 0.5 nm thick. The surface area for

montmorillonite, estimate by methylene blue adsorption was found to be ~1000 m2/g

[12].Poly(ethylene oxide), of average molecular weights 200 000, 300 000, 400 000,

900 000, 4 000 000 g/mol was purchased from the Aldrich Chemical Company.

Water was purified using a Milli-Q Academic System (Millipore Cooperation), USA.

Methods

Appropriate proportions of polymer solutions and particles were mixed and gently

stirred in a bottle and let to stay for 24 h to ensure solutions homogeneity. They were

then shaken, varying between a couple of shakes and ten minutes.

Some samples were poured into a syringe and squirted through a hypodermic needle of

dimensions 10 cm in length and 1 mm diameter.

Some samples were poured into the concentric cylinder geometry of a Paar Physica

Universal rheometer (Austria) and the viscosity at different shear rates (from 200 s-1 to

1000 s-1) and the complex modulus (at 1Hz and 1 Pa) were monitored as a function of

time.

4

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

Results and discussion

Laponite suspensions.

Fig. 1

The “phase diagram” for laponite-PEO (average molecular weight 400 000 g/mol) dispersions as a function of clay concentration in weight per cent and the mixture concentration given as Γt (total mass of PEO per total area of clay surface) is shown in Fig. 1. At low Γt no shake gel is formed, but as Γt, increases, there is a narrow range of Γt, shake gel effect is observed. At higher Γt, shake gels no longer form. The critical value of Γt, below which no shake gel is formed is quite independent of the laponite concentration and seems to be related only to the available amount of PEO per total area of clay surface.

Phase diagram of laponite-PEO mixtures at lower PEO molecular weight (~300 000

g/mol) has been presented previously by Pozzo and Walker [1] and Zebrowski et al.

[2].Their results agree fairly well with ours and it seems that there are no significant

differences between the phase diagram of the two PEO molecular weights.

Fig. 2 shows how the complex modulus of laponite suspension (concentration of

particles 1%) changes as a function of time for different PEO concentrations. As we can

see, the complex modulus shows a maximum at 0.5 mg/m2 and then decreases. This

result is in accordance with Pozzo and Walker discussion about the “phase diagram” of

the laponite-PEO mixtures [1]. At very low Γt we have found that the complex modulus

is low and they suggest that, under those conditions, there is too little polymer to result

5

109

110

111

112

113114115116117118119120

121

122

123

124

125

126

127

128

129

130

in any observable flocculation. Under these conditions, the macromolecules bind as

many particles as possible, and there are still excess particles. If some segments of a

macromolecule are still available for adsorption, they will bind preferentially to a free

particle rather than to a particle already bound to another. [4]. When Γt is increased, it is

observed that the complex modulus increases, indicating that the polymer forms, in

absence of shear, stable bridges between the necklaces to share the available polymer

segments and free laponite surfaces, and that this causes the macroscopic gelation of the

samples At higher Γt the complex modulus decreases. Pozzo and Walker [1] suggested

that near the surface saturation value, there is little free clay surface and the aggregates

are smaller because the number of bridging chains is reduced and the complex modulus

decreases. However, the aggregates are large enough to be influenced by experimental

shear rates and they can be stretched or deformed. This results in the exposure of

additional free surface and creates the necessary conditions for further aggregation. This

is where the shear gelation or shake-gel effect is observed. When the values of Γt are

substantially above the surface saturation, the free polymer in solution reduces the size

of the aggregates by further saturating the particle surface. The dispersion is made of

small necklaces or chains, containing saturated particles, or isolated saturated particles.

The bridging is prevented by steric stabilization.

Fig. 2

6

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

shows the effect of varying the molecular weight of the polymer on the “phase”

behavior of the PEO-laponite system. We have observed that low molecular weight

polymers require considerable shaking in order to form a weak gel, sometimes of the

order 10 minutes. This may be due to the fact that at low molecular weight the length of

the polymer chains is short and this will make the formation of bridges between the

particles more difficult than at higher molecular weight, and additionally the bridges

formed will be weaker. On the other hand, it is hard to be sure whether the very high

molecular weight polymers form a shake gel as the viscosity of the mixed suspension

itself is high. The same observation can be made about mixtures where the degree of

covered of the laponite surfaces are medium. On the other hand, at moderate molecular

weight and low particle concentration we have observed that the dispersion separates

into a “solid like” phase and a little excess water. These small phase separation could be

due to the fact that the bound necklaces form macroscopic “flocs” that are separated

from an excess aqueous phase containing nearly all the particles and macromolecules

[4]. The range of concentrations to which the shake gel is form is quite independent of

the PEO molecular weight. A shake gel is formed for PEO concentration ranging

between 0.5 and 1 mg/m2.

The polydispersity of the polymers used in the current work was unknown and is likely

to be relatively high, as we use commercial samples. However, polydispersity does not

seem to have a large influence on the range of concentrations over which the shake gel

is formed although it could modify slight it. (In one experiment, we manufactured a

more polydisperse polymer of average molecular weight 900,000 by blending PEO

400,000 and 4,000000 in appropriate concentrations and observed that the PEO

concentrations when a shake gel was formed decreased slightly from 0.1% for the

original PEO to 0.05% for the blended samples). This suggests that polydispersity

favors the formation of bridges between the polymer chains and the silica, probably due

the presence of higher molecular weight molecules in the blended PEO, which would

shift the critical range of PEO concentrations for the gel formation towards slightly

lower values.

7

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

Fig. 3

Bentonite suspensions

The analysis of the effect of the size of the particles was investigated in PEO-

montmorillonite systems. Montmorillonite and laponite particles have plate-like shape

but laponite particles are typically two orders of magnitude smaller than

montmorillonite. Fig. 4 shows in the effect of varying the molecular weight of the

polymer on the “phase” behaviour of PEO-montmorillonite systems. The trends are

similar to those found in laponite samples but some differences can be found between

them. The gels formed upon shaking montmorillonite samples are weaker than in

laponite suspensions. Besides, when phase separation is produced under shaking, this is

most noticeable in the case of montmorillonite, where the formation of a single “floc”

can be observed, see Fig.5.a, where the clay and aqueous phases separate out. This floc

can be partially relaxed, see Fig.5.b, and this partial relaxation can be reversed by

shaking the sample again. It can be argued that only a small fraction of the polymer

segments are needed to establish chains capable of bridging small laponite particles [7,

14, 15], as compared to the larger montmorillonite units. Here, bonding points of the

polymers will be much more numerous for each particle. This explains that flocs in the

montomorillonite-based gels will be more difficult to break down, and the will partially

relax but not disappear when shaking has ceased.

8

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

Fig. 4

Fig. 5

Silica suspensions

Having found that shake gels can be formed with montmorillonite and not just with the

(smaller size) laponite particles, we decided to study the shape effect on their formation.

Liu et al. [3] observed weak (or soft) gels based on silica beads and PEO (Mv ~ 2·106).

In our case, gels are also found, but, under certain concentrations of silica nanoparticles

and PEO, the systems obtained can be much stronger. It is illustrated in Fig. 6, a strong

gel is formed by shaking in a sample contains 25% LudoxTM-50 and 0.4% PEO Mv ~

4·106. Manipulation of the gel can be performed by hand, as the structures formed can

withstand strong stretching and rolling deformations, without breaking.

9

200

201

202

203

204

205

206

207

208

209

210

211

212

Fig. 6

In order to explore the shake gel formation in silica-PEO mixtures, we vary the relative

concentration of PEO (average molecular weight 400 000 g/mol) in the mixture and

determine if the shake-gels is formed for each composition. Shown in Fig. 7 is the

“phase diagram” for Ludox®TM-50/PEO (Mv~400 000 g/mol) dispersions as a function

of silica concentration in weight per cent and the mixture concentration given as the

total mass of PEO per total area of silica surface. We can see that it is necessary a

Ludox particle concentration higher than 10% (weight fraction) to form a gel.

Moreover, there is a limiting value of Γt, below which shake gels are formed. At low

and moderate PEO molecular weights this upper limit of Γt is around 0.2 mg/m2 , and it

decreases when the Ludox particle concentration increases just up to 25%. That

behavior is related to the fact that the viscosity of the samples increases with both

particle and polymer concentrations and in those conditions it becomes more difficult to

produce the relative movement between them, and therefore the shear-induced polymer

bridging between the particles is hindered.

An upper limit of Γt has been found for laponite and silica, although the actual value is

lower in the former case. Furthermore,: in the case of laponite gels are formed for PEO

ranging between 0.5 and 1.5 mg/m2, whereas polymer surface concentrations between

0.02 and 1.5 mg/m2 are required in silica gels. Such differences can be ascribed to

particle shape effects: in laponite, it can be expected that the polymer chains lay flat on

the planar surfaces, and few chains arrange themselves perpendicularly [13]. The small

spheres of Ludox, in contrast, can only accommodate a limited amount of polymer,

most chains extending away from the surfaces increasing the chances of bridging [7, 17,

18]. In addition, no specific orientations can be expected in the latter configuration, and

this makes it easier the stretching and deformation of gels based on the spherical silica

than on the planar laponite particles.

10

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

Fig. 7

Rheological observations

In order to reach a deeper understanding of the shake gel formation mechanisms, we

have studied the rheological behavior of silica shake gels. Fig. 8.a shows the viscosity

of mixtures containing 25% Ludox and 0.3% PEO (average molecular weight 400 000

g/mol, Γt =0.09 mg m-2) as a function of time for different shear rate applied. As

observed, the values of viscosity are almost constant over time, and only a slight

increase is observed when the shear rate is increased (see Fig.9.b). These results differ

to some extent from those results found by Liu et al [3] for low silica concentrations

(between 10-40 g/l). They reported that for a coverage between 0.35 and 0.8 mg m-2 of

polymer the viscosity of the mixtures is low, Newtonian and independent of time, as

long as the shear rate is below a critical value (between 600 and more than 1500 s-1).

However, when the critical shear rate just exceeds this value, the apparent viscosity

abruptly increases with shearing time. The critical shear rate decreased when decreasing

coverage or increasing concentration.

11

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

Fig. 8

The nature of the applied shear field has a profound effect on the sample behavior. As

mentioned above, when the same samples were subjected to simple shear in Couette

flow it was found that gels were not induced up to shear rates of thousands s-1 (see Fig.

8). However, the gelation phase transition is observed with moderate manual agitation

which corresponds to a significantly smaller shear rate (< 102 l/s) [1]. Pozzo and

Walker [1] suggest that this observation indicates that particles aggregate and become

oriented in the shear field as they form, and this prevents their further aggregation in

simple flow fields by reducing the number of effective collisions. We have found that

the sample becomes a gel when we squirt it through a hypodermic needle (see Fig. 9.a).

Under these conditions, the shear rate is increased and the polymer chains are extended

in the direction of flow. Particle-polymer necklaces can be formed and aligned along the

flow and this contributes to form a continuous thread of gel.

So, it seems that the alignment and chaining of particles in elongation could be an

important mechanism underlying the unusual response of shake gels. In this regard,

Wang et al. [20] have found that PEO solutions get stiffer with the rate of extension.

This stiffening is greatly enhanced by the presence of silica nanoparticles in solution.

These authors proposed that alignment and chaining of particles seems to be favored in

elongation and in these conditions longer necklaces can be formed. This fact can favor

the number of crosslinks between necklaces and increase the size of the aggregates. In

this way, a large composite macromolecule is created and the resulting solution has

enormous extensional stresses and exceptional siphoning properties. The formation of

12

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

long necklaces could explain the formation of a long and continuous thread when the

shake gel formed is left to drop (see Fig. 9.c).

Fig. 9

Comparison of polymer size and particle size

We shall now compare the size of the particles used in this work to the size of the

polymers.

The z-average radius of gyration of the polymers can estimate with the expression

suggested by Can and Okay [6];

<S2>z1/2 = (4.08*10-4M w1.16)1/2 (nm)

Table 1 shows the z-average radius of gyration of PEOs used in this study together with

their overlap concentrations c* calculated using the equation:

C*= 3M w·100/(4𝜋<S2>z3/2) (w/v, %)Table 1. The radius of gyration and c* of PEO of various molecular weights

Mv <S2>z1/2(nm) C*(w/v, %)

200,000 24 0,57300,000 30 0,43400,000 36 0,34900,000 57 0,19

4,000,000 136 0,06

We recall that the diameter of the silica and laponite is around 30 nm in diameter, so

that these particles are somewhat smaller than the diameter of the polymer coil, however

the montmorillonite has a diameter of around 1000 nm (although it is only around 0.5

nm thick). This may explain why the montmorillonite suspensions only form very weak

13

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

shake gels compared to the laponite and silica, with the larger montmorillonite particles

probably being bridged by the polymer, whilst the smaller particles may be more likely

to act as points for crosslinking the polymer chains rather than the polymer chains

bridging between the particles.

ConclusionsA shake-induced transition from fluid to solid is observed in mixtures of clay (laponite

or montmorillonite) or silica nanoparticles (Ludox) and PEO for certain concentrations

of polymer and particles. This behavior has been found in particles of different shapes

(spherical or disc) and size (between 30 nm and 1 micron) although some differences

have been found in the phase diagrams of the different particles. The degree of polymer

coating of the particles necessary for shake gel formation seems to depend of the

particle shape. Whereas in the case of the disc-shaped particles this is around the value

of the particle surface saturation, in the case of spherical particles it is around 2/3 of that

limit. Moreover, shake gel behaviour is found to form over a narrower composition

regime in the case of laponite than in the case of Ludox. The main difference between

montmorillonite (platelets of approximately 1000 nm x 1 nm dimensions) and laponite

(platelets of approximately 30 nm x 1 nm dimensions) dispersions are that when we

shake the former, at some PEO concentration, we found weak gels or, in most cases, an

important and irreversible phase separation instead of a shake gel; on the contrary, in

the case of the laponite dispersion we obtain stronger gels and, when the phase

separation is produced, this is not extensive.

While, it is possible to reach strong gel formation upon a moderate shaking of the

samples studied or if we squirt them through a hypodermic needle but not to simple

shear in Couette flow at shear rates up to thousands of inverse seconds. This behavior

suggests that the nature of the applied shear play an important role in the shake gel

formation and that extensional shear, rather than simple shear may be important. A

quantitative model explaining this is the subject of further investigation.

Figure captionsFig. 1. Phase diagram (regions of shake gel formation) for laponite-PEO (Mv ~ 400 000

g/mol) systems as a function of particle and polymer concentration (mg PEO per m2

laponite surface). The line indicates the approximate value of surface coating saturation.

14

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

Fig. 2 (a) Complex modulus, G*, as a function of time (at 1Hz and 1 Pa) at different

PEO concentrations and 1% laponite. (b) Complex modulus for the plateau of part (a),

G*plateau, as a function of polymer concentration (mg PEO per m2 laponite surface). The

error bars represent the standard deviation of the data on the plateau.

Fig. 3 Effect of varying the molecular weight of the PEO on the “phase” behaviour of

the PEO-laponite system (laponite concentration 2%).

Fig. 4 Effect of varying the molecular weight of the polymer on the “phase” behaviour

of the PEO-montmorillonite system (montmorillonite concentration 2%).

Fig. 5 Aqueous Bentonite (3%) (left) and laponite (3%) (right) -PEO mixtures (contains

0.4% PEO Mv~4 000 000); Top: after completion of shaking; bottom: after 1h

relaxation.

Fig. 6 Aqueous Ludox-PEO mixture contains 25% LudoxTM-50 and 0.4% PEO

Mv~4000000. (A) before shake it; (B) after completion of shaking; (C) after stretching

sample (B).

Fig. 7 Shake gel formation diagram of Ludox TM 50-PEO (average molecular weight of

PEO 400 000 g/mol) systems as a function of particle and polymer concentrations (mg

PEO per m2 Ludox surface). The line indicates the approximate value of the surface

saturation.

Fig. 8 Viscosity as a function of (a) the time at several shear rates, and (b) shear rate at

several times. The mixture contains 25% Ludox and 0.3% PEO of M v= 400 000 g/mol.

The error bars represent the standard deviation of the data on the plateau.

Fig. 9 Mixture containing 25% Ludox and 0.3% PEO of M v=400 000 g /mol squirted

through a hypodermic needle (a); dropping before shaking (b); droping after shaking

(c).

Acknowledgement

The work was supported by Junta de Andalucía, Spain, thorugh the Programme of Researchers

Support in Foreign Institutions.

References

15

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

[1] D.C. Pozzo, L.M. Walker, Reversible Shear Gelation of Polymer-Clay Dispersions, Colloids Surf., A, 240 (2004) 187-198.

[2] Zebrowski, V. Prasad, W. Zhang, L.M. Walker, D.A. Weitz, Shake-Gels: Shear-Induced Gelation of Laponite-PEO Mixtures, Colloids Surf., A, 213 (2003) 189-197.

[3] S.F. Liu, F. Lafuma, R. Audebert, Rheological Behavior of Moderately Concentrated Silica Suspensions in the Presence of Adsorbed Poly(Ethylene Oxide), Colloid Polym. Sci., 272 (1994) 196-203.

[4] B. Cabane, K. Wong, P. Lindner, F. Lafuma, Shear Induced Gelation of Colloidal Dispersions, J. Rheol., 41 (1997) 531-547.

[5] L. Ye, X. Chu, Z.D. Zhang, Y. Kan, Y. Xie, I. Grillo, J. Zhao, C.A. Dreiss, D. Qiu, Effect of particle polydispersity on the structure and dynamics of complex formation between small particles and large polymer, Rsc Adv., 4 (2014) 14896-14903.

[6] V. Can, O. Okay, Shake Gels Based on Laponite-PEO Mixtures: Effect of Polymer Molecular Weight, Des. Monomers Polym., 8 (2005) 453-462.

[7] Y. Saito, Y. Hirose, Y. Otsubo, Shear-induced reversible gelation of nanoparticle suspensions flocculated by poly(ethylene oxide), Colloids Surf., A, 384 (2011) 40-46.

[8] G. Deželić, M. Wrischer, Z. Devidé, J.P. Kratohvil, Electron Microscopy of Ludox Colloidal Silica, Kolloid-Z., 171 (1960) 42-45.

[9] H. Bergna and W. Roberts, Colloidal Silica: Fundamentals and Applications, Delaware: CRC Press, 2006.

[10] B.S. Neumann, Behaviour of a synthetic clay in pigment dispersions, Rheol. Acta, 4 (1965) 250-255.

[11] N. Willenbacher, Unusual thixotropic properties of aqueous dispersions of Laponite RD, J. Colloid Interface Sci., 182 (1996) 501-510.

[12] S. Rossi, P.F. Luckham, T.F. Tadros, Influence of non-ionic polymers on the rheological behaviour of Na+-montmorillonite clay suspensions—I Nonylphenol–polypropylene oxide–polyethylene oxide copolymers, Colloids Surf., A, 201 (2002) 85-100.

[13] A. Nelson, T. Cosgrove, A small-angle neutron scattering study of adsorbed poly(ethylene oxide) on laponite, Langmuir, 20 (2004) 2298-2304.

[14] M. Kamibayashi, H. Ogura, Y. Otsubo, Rheological behavior of suspensions of silica nanoparticles in associating polymer solutions, Ind. Eng. Chem. Res., 45 (2006) 6899-6905.

[15] Y. Otsubo, Y. Nakane, Simulation of bridging flocculation and elastic percolation in suspensions, Langmuir, 7 (1991) 1118-1123.

16

358359

360361362

363364365

366367

368369370371

372373

374375376

377378

379380

381382

383384

385386387388

389390

391392393

394395

[16] F. Lafuma, K. Wong, B. Cabane, Bridging of Colloidal Particles Through Adsorbed Polymers, J. Colloid Interface Sci., 143 (1991) 9-21.

[17] G.P. Van der Beek, M.A.C. Stuart, The Hydrodynamic Thickness of Adsorbed Polymer Layers Measured by Dynamic Light-Scattering - Effects of Polymer Concentration and Segmental Binding Strength, J. Phys., 49 (1988) 1449-1454.

[18] P.A. Pincus, C.J. Sandroff, T.A. Witten, Polymer Adsorption on Colloidal Particles, J. Phys., 45 (1984) 725-729.

[19] J. Wang, R. Bai, D.D. Joseph, Nanoparticle-laden tubeless and open siphons, J. Fluid Mech., 516 (2004) 335-348.

17

396397

398399400

401402

403404405

406