G Model

JASCER 27 1–7 ARTICLE IN PRESS

Journal of Asian Ceramic Societies xxx (2013) xxx–xxx

Journal of Asian Ceramic Societies

Journal of Asian Ceramic Societies

journal homepage: www.elsevier.com/locate/jascer

1 Poly(maleic acid) – A novel dispersant for aqueous alumina slurry

2 Q1 Saralasrita Mohanty, Bodhisatwa Das, Santanu Dhara ∗
3 Biomaterials and Tissue Engineering Laboratory, School of Medical Science and Technology, Indian Institute of Technology Kharagpur, Kharagpur 721302, India
4

5
20 a r t i c l e i n f o a b s t r a c t
6
7 Article history: Poly(maleic acid) (PMA) was investigated as an anionic long chain dispersant for the preparation of highly
8 Received 13 April 2013 loaded aqueous alumina slurry. Suspensions with 1 wt% alumina powder were prepared through varia-
9 Received in revised form 13 May 2013 tion of PMA concentrations at pH 9 to assess its dispersion efficacy in aqueous medium. Addition of PMA
10 Accepted 16 May 2013
increased the stability of 1 wt% alumina suspensions as evidenced by higher zeta potential and RSH (rel-
11 Available online xxx
ative suspension height). 55 vol% alumina slurries were prepared with different PMA concentrations at
12
pH 9. Based on the rheological studies, 3.8 mg of PMA per gram of alumina was obtained as the optimized
13 Keywords:
dispersant amount at pH 9. All the slurries had shear thinning behavior and the slurry with optimum
14 Alumina
15 Dispersant dispersant amount had insignificant thixotropy. Further, adsorption of dispersant was evident by Fourier
16 Zeta-potential transform infrared spectroscopy (FTIR) analysis and thermogravimetric analysis (TGA). The slurry at pH Q3
17 Thixotropy 9 with optimum dispersant amount had lowest viscosity and easy flowability. Microstructure of bisque
18 Viscosity fired and sintered body by SEM microscopy revealed homogeneous particle distribution and uniform
19 Microstructure grain growth with sintered density of 3.94 g/cm3 . The mechanical property evaluation revealed that the
samples prepared with optimum dispersant amount had the flexural strength of 483 ± 22 MPa.
© 2013 The Ceramic Society of Japan and the Korean Ceramic Society. Production and hosting by
Elsevier B.V. All rights reserved.

21 1. Introduction facilitates homogeneous microstructure of the ceramic component 35

during sintering and thus improves the mechanical properties [10]. 36

22 Colloidal processing techniques have been widely used in near In aqueous colloidal processing of ceramics, the nature and 37

23 net shape forming of ceramics for diverse applications due to dif- amount of dispersant are important in order to obtain stable and 38

24 ficulty in controlling the microstructure by melt processing [1–3]. uniform dispersed system [11–13]. Well dispersed slurry with easy 39

25 High powder loading is important for slurry based colloidal pro- flowability is achieved by optimum concentration of the dispersant. 40

26 cessing to obtain high packing density of the green body facilitating Optimum amount of dispersant provides highest surface charge 41

27 minimum drying and sintering shrinkage which eventually reduces resulting in a homogeneous and well-stabilized suspension. It was 42

28 the warpage of the final components [4]. However, high powder reported that ceramic suspensions containing insufficient or excess 43

29 loading is difficult to achieve for fine powders due to high tendency amounts of dispersant than those with optimum show a rela- 44

30 of agglomeration. During slurry based colloidal processing, disper- tively higher viscosity [14]. The increase in viscosity with less than 45

31 sant has important role in the preparation of ceramic green body optimum dispersant concentration is due to insufficient surface 46

32 with uniform green density and homogeneous particle distribution coverage of the particles which facilitates formation of agglom- 47

33 [5–9]. A stable and well-dispersed colloidal suspension facilitates erates in the suspension through bridging flocculation. While in 48

34 high packing density in the green body. Uniform green density case of dispersant amount above the optimum concentration, the 49

excess non-adsorbed dispersant molecules facilitate zeta poten- 50

tial reduction of the suspended particles through compression of 51

double layers by excess counter ions present in the solution. 52

Q2 ∗ Corresponding author. Tel.: +91 9434701616. In the context of slurry processing [15], Yasuda et al. [16] 53
E-mail addresses: sdhara@smst.iitkgp.ernet.in, santanudhara@gmail.com reported that the quantity of polymer dispersant has major 54
(S. Dhara). influence on microstructure and mechanical properties of bulk 55
Peer review under responsibility of The Ceramic Society of Japan and the Korean
hydroxyapatite during colloidal processing. Actually, high viscos- 56
Ceramic Society.
ity of the slurry will render difficulty in removal of air bubbles 57

entrapped during milling which eventually introduces defects in 58

the sample [17,18]. Takao et al. [19] studied influence of process- 59

ing defects on mechanical properties of slip cast alumina. Thus, 60

optimization of dispersant amount is important in controlling sta- 61

2187-0764 © 2013 The Ceramic Society of Japan and the Korean Ceramic Society.
Production and hosting by Elsevier B.V. All rights reserved. bility of aqueous alumina slurries with minimum viscosity in order 62

http://dx.doi.org/10.1016/j.jascer.2013.05.005 to achieve defect-free components. 63

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005
 G Model
JASCER 27 1–7 ARTICLE IN PRESS
2 S. Mohanty et al. / Journal of Asian Ceramic Societies xxx (2013) xxx–xxx

64 Among various dispersant types, anionic electrosteric disper- highly loaded slurries. Fourier transform infrared spectroscopy 129

65 sant such as poly(acrylic acid) [20], poly(methacrylic acid) [21], (FTIR) analysis and thermogravimetric analysis (TGA) were per- Q4 130

66 pyrophosphate [22], citric acid [23], etc. are mostly studied for formed to validate adsorption of the PMA on alumina powder 131

67 preparation of highly loaded ceramics (alumina, zirconia, silica, sil- surface. Density, microstructure and mechanical properties of the 132

68 icon carbide, etc.) based colloidal suspensions. The performance of sintered alumina prepared using 55 vol% slurry were evaluated for 133

69 the polyelectrolyte [24] dispersants is mainly influenced by sev- optimized dispersant amount at pH 9. 134

70 eral factors including charge density, type of monomers, functional

71 groups, molecular weight, pKa values, etc. In this context, polyelec- 2. Experimental procedures 135

72 trolytes containing carboxylic group are found to be an effective

73 dispersant for stabilization of highly loaded aqueous alumina slur- 2.1. Materials 136

74 ries. Chain length and number of carboxylic groups per unit length
75 of poly anionic dispersant have influence on dispersion efficacy Alumina powder (RG 4000, Almatis Alumina Private Limited, 137

76 during slurry preparation. The carboxylic acid groups attached to Falta, India; d50 ∼ 0.7 ␮m; surface area, 7 m2 /g) based aqueous slur- 138

77 the polymer chain are deprotonated at pH above pKa value, result- ries were prepared using PMA (PM-200, Aquapharm Chemicals 139

78 ing in an increase of surface charge of alumina in the suspension Pvt. Ltd., Pune, India) for dispersion and tetramethyl ammonium 140

79 due to adsorption of the negatively charged dispersant. At optimum hydroxide (25% solution, Merck, Mumbai, India) was used for pH 141

80 dispersant concentration, these molecules adsorb on suspended adjustment. 142

81 powder surface forming monolayer and part of the long moiety

82 remains in the solution facilitating eletrosteric stabilization [25]. 2.2. Stability of suspension with different concentrations of PMA 143

83 Further, the extended chain in the solution hinders formation of

84 any powder agglomerates and thereby stabilized the ceramics sus- Different amount of PMA (average molecular weight 500–1000) 144

85 pension. was added to distilled water for the preparation of 1 wt% alumina 145

86 It is reported by Hanaor et al. [26] that small quantity of car- suspension at pH 9. The suspensions with different dispersant con- 146

87 boxylic acid reagent imparted high negative zeta potential values centrations (Table 1) were ultrasonicated for 10 min to breakdown 147

88 and stabilized ZrO2 suspension electrosterically over a wide pH the agglomerates. The suspensions were poured into graduated 148

89 range. Hidber et al. [27] successfully used citric acid as dispersant cylinder and were placed vertically for over a period of one month 149

90 for the deflocculation of concentrated alumina slurries. The interac- followed by measurement of turbidity and sediment height at dif- 150

91 tion of citric acid with alumina powder particles leads to adsorption ferent time intervals. Further, zeta potentials of all the suspensions 151

92 of citrate ions on the hydroxylated alumina powder surface result- with different dispersant amounts were measured by Zetasizer 152

93 ing in development of negative charge. The three carboxylic groups (Malvern instrument, Malvern, UK) at pH 9. 153

94 attached to each citrate ion impart sufficient magnitude of negative

95 charge on the powder surfaces. The adsorbed citrate ions also form 2.3. Slurry preparation 154

96 steric barrier which offer repulsive force between individual parti-

97 cles. Davies and Binner [28] have reported a novel method for the For optimization of dispersant amount with highly loaded aque- 155

98 preparation of high solid loading ceramic slurries using ammonium ous alumina slurries, PMA was added initially to a poly(propylene) 156

99 polyacrylate as electrosteric dispersant at alkaline pH. Cesarano bottle containing predetermined amount of distilled water and alu- 157

100 and Aksay [29] described the dispersion of sufficiently highly con- mina powders were added in steps. In the present study, 55 vol% 158

101 centrated (>50 vol%) ␣-alumina powder (particle size 0.2–1.0 ␮m) alumina slurries were prepared by addition of alumina in three 159

102 slurries using sodium poly(methacrylic acid) and poly(acrylic acid) steps in presence of varied amount of PMA and slurries were 160

103 at different pH for the fabrication of dense ceramic components. adjusted at pH 9. The slurries were milled by roll mill with addition 161

104 Poly(acrylic acid) is widely used dispersant in ceramic processing of 1:1 ratio of powder to ZrO2 milling media (3 mm diameter) at 162

105 due to its eletrosteric stabilization. 40 rpm. Finally, the slurries were filtered using sieve to separate the 163

106 In this context, poly(maleic acid) (PMA), being a poly car- milling media. For deairing of slurries, 100 ␮l of n-octanol per 100 g 164

107 boxylic anionic macromolecules, is adsorbed on metallic oxides slurry was added followed by rolling at 20 rpm for 30 min using the 165

108 (like Goethites), carbonates (calcites) and forms a Langmuir mono- procedure discussed by Dhara et al. [18]. The deaired slurries were 166

109 layer as reported by Wang et al. [30] and Euvrard et al. [31]. It has ready for rheological measurement and casting. 167

110 also been utilized for boiler treatment as it stabilizes gypsum par- The effect of slurry pH on suspension stability was evaluated for 168

111 ticles in aqueous solution and inhibits crystal growth by adhering a constant dispersant concentration. Based on optimization study, 169

112 to the surface [32]. There are few preliminary reports on improve- 55 vol% alumina slurries with 3.8 mg PMA per gram of alumina 170

113 ment of rheology and zeta potential in chalcocite slurries even at powder were prepared at three different pH levels (viz., 6, 9 and 171

114 very high powder loading by addition of PMA [33]. Interestingly, 12, respectively). Flow behavior test of these slurries was carried Q5 172

115 PMA has similar structure to poly(acrylic acid) and this polymer has out to evaluate influence of slurry pH on viscosity at constant shear 173

116 double charge density than that of poly(acrylic acid) per unit chain rate. Further, slurries were diluted to 1 wt% by adding deionized 174

117 length owing to presence of one carboxylic acid group in each car- water and final suspension pH was maintained at their native pH 175

118 bon atom in the main back bone of the polymer chain. Maleic acid using TMAH and HCl. The zeta potentials of these suspensions were 176

119 has pKa values within 1.8–6 (pK1 = 1.83; pK2 = 6.07) [34,35] and its also measured to evaluate surface charge of alumina with constant 177

120 macromolecule (PMA) highly ionizes above pH 7 (pKa between 5 dispersant concentration at the above mentioned pH. 178

121 and 7) in aqueous medium [36]. This polymer has the potential to
122 disperse oxide ceramics for the preparation of colloidal suspension 2.4. Rheological measurement 179

123 in aqueous medium [37].

124 In the present study, PMA is explored as an anionic dispersant To measure slurry viscosity, 11.5 mL of alumina slurry was 180

125 for the stabilization of highly loaded alumina slurry. Dispersion poured into cylindrical cup of the rotational viscometer (Visco Elite, 181

126 ability of the PMA was primarily evaluated by RSH (relative sus- Fungilab, UK) along the wall and a spindle was placed carefully to 182

127 pension height), zeta potential measurement. Optimization of avoid entrapment of air bubbles. To prevent water loss during mea- 183

128 dispersant dose was carried out through rheological studies of surement, 5 drops of n-octanol were added on top the slurries. Prior 184

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005
 G Model
JASCER 27 1–7 ARTICLE IN PRESS
S. Mohanty et al. / Journal of Asian Ceramic Societies xxx (2013) xxx–xxx 3

Table 1
Effect of PMA concentrations on RSH, zeta potential and viscosity of alumina powders in aqueous medium at pH 9.

PMA concentrations (mg/g of alumina powder) 1 wt% suspension 55 vol% slurry

RSH (%) after one month Zeta potential (mV) Viscosity (Pa s) at (5.6 s−1 )

2.3 29.8 ± 0.5 −28.7 ± 0.5 5.1

3.0 36.3 ± 0.2 −42.3 ± 1.4 2.2
3.8 59.0 ± 0.1 −55.1 ± 2.5 1.5
4.6 52.6 ± 0.4 −50.8 ± 1.8 1.9
5.3 48.2 ± 0.3 −44.0 ± 1.4 2.6

185 to measurement, the slurries were subjected to stirring by rotat- out and amount of PMA dispersant adsorbed on the surface of alu- 229

186 ing the spindle at 100 rpm to maintain similar condition prior to mina powder for suspensions with different PMA concentrations 230

187 measurement. Viscosity of the slurries was measured at different was calculated by the percentage weight loss during decomposition 231

188 shear rates ranging from 0.14 and 16.8 s−1 at 25 ◦ C from upward of organics by heat treatment. 232

189 to downward sweep cycle. Thixotropic behavior of the slurries

190 was also measured to examine influence of dispersant amount on 2.8. Microstructure analysis 233
191 time dependent aging behavior at corresponding shear rate during
192 upward and downward sweep viscosity measurement. Thixotropy The samples were placed in furnace for bisque firing and sinter- 234
193 of alumina slurry was expressed in terms of difference in viscosity ing. The samples were bisque fired at 1000 ◦ C in air by holding for 235
ῃdw ῃuw 2 h in a chamber furnace (N. R. Enterprise, Kolkata, India). Fracture 236

surface of bisque fired samples was gold coated and analyzed via 237
194 ( ) at constant shear rate during downward ( ) and upward SEM (SEM; EVO ZEISS, Carl Zeiss SMT AG, Germany) at an accelerat- 238
(ῃdw) ing voltage of 10–20 kV for particles distribution in the green body. 239

Further, the green samples were also sintered at 1550 ◦ C for 2 h in 240
195 ( ) viscosity against shear rate measurement.
air by a chamber furnace (Bysakh, Kolkata, India) for microstructure 241

and mechanical properties analysis. 242

196 2.5. Green body preparation
2.9. Mechanical properties 243
197 The deaired slurry with optimized dispersant amount at pH 9
198 was used for the preparation of green body. Prior to the slurry cast- The flexural strength (3-point bending configuration) of sin- 244
199 ing, all the molds were coated with white petroleum jelly. Silicon tered rectangular bars, each of dimensions 40 mm × 5 mm × 4 mm, 245
200 rubber molds were used to prepare rectangular bars for flexural were measured by UTM (Model HEK25, Hounsfield, UK) with a span 246
201 strength measurement under 3-point bending mode. The samples of 25 mm at a cross-head speed of 0.5 mm/min using 5000 N load 247
202 were dried under controlled humidity and vacuum dried samples cell. All the failures occurred on the tensile side of the bar samples 248
203 were removed from the mold carefully. The polished samples were while performing the bending test. Ten specimens were tested for 249
204 ready for sintering. The density of the vacuum dried green and flexural strength measurements of the sintered alumina. 250
205 sintered samples was measured via Archimedes principle.

3. Results and discussion 251

206 2.6. FTIR study of alumina powders for adsorption of PMA
The pH of 1 wt% alumina suspensions was adjusted above the 252
207 FTIR analysis of different samples was carried out using pKa value of PMA (at pH 9) for high degree of ionization of polymer 253
208 spectrophotometer (NEXUS-870, Thermo Nicolet Corporation, molecules. Stability of the aqueous alumina suspensions at pH 9 254
209 Madison, WI, USA) controlled by Spectrum software. To prepare was increased significantly with addition of PMA as evidenced by 255
210 samples for FTIR analysis, slurries were diluted with deionized RSH measurement. The RSH data in Table 1 clearly depicted that 256
211 water to 1 wt% and centrifuged to collect the alumina precipitate. stability of the suspensions increased with increasing PMA con- 257
212 The vacuum dried precipitates were made into pellets after mixing centrations. Alumina powders in suspension took more than thirty 258
213 with KBr (1:10) and used for FTIR analysis. Also, the freeze dried days to settle down in presence of PMA, whereas the suspension 259
214 PMA was crushed into powder and made into palate by mixing without PMA (at pH 9) sedimented within an hour. This suspen- 260
215 with KBr (1:10) for analysis. To validate adsorption of PMA on alu- sion stability is also reflected by the zeta potential values (Table 1). 261
216 mina powder surface during preparation of aqueous slurries, the All the suspensions had zeta potential values higher than that of 262
217 vacuum dried as received alumina powder was analyzed by FTIR −25 mV at pH 9 with addition of different concentrations of PMA 263
218 spectroscopy using KBr palate (1:10) and compared. ranging between 2.3 and 5.3 mg/g of alumina powder. 264

Based on the above results, 55 vol% aqueous alumina slurries 265

219 2.7. Thermogravimetric analysis were prepared using different concentrations of PMA at pH 9 and 266

their flow behavior was studied by viscosity measurement at dif- 267

220 TGA was carried out for analyzing quantitative adsorption of ferent shear rates. The shear rate sweep measurement of all the 268

221 PMA on alumina surface by Thermo-gravimetric analyzer (Perkin slurries indicated non-Newtonian flow behavior (shear thinning) 269

222 Elmer Pyris Diamond Model, MA) using alumina crucible under air owing to their high powder loading (Fig. 1a). It was also found that 270

223 atmosphere in the temperature range of 50–650 ◦ C with a heat- viscosities of the slurries were strongly influenced by the amount 271

224 ing rate of 10 ◦ C/min. Different slurries were diluted to 1 wt% using of PMA. As viscosity is function of shear rate, while viscosity values 272

225 deionized water and centrifuged to collect the precipitate. The alu- of all slurries were plotted against different PMA amounts at con- 273

226 mina precipitates after centrifugation and the as received alumina stant shear rate, all the curves passed through a minima at 3.8 mg 274

227 powders were vacuum dried at 40 ◦ C for a period of 24 h in a vacuum of polymer/g of alumina (Fig. 1b). At this PMA concentration, vis- 275

228 oven. Thermal degradation behavior of all the samples was carried cosity of the slurry was 1.6 Pa s at 5.6 s−1 shear rate. Further, above 276

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005
 G Model
JASCER 27 1–7 ARTICLE IN PRESS
4 S. Mohanty et al. / Journal of Asian Ceramic Societies xxx (2013) xxx–xxx

Fig. 1. Flow behavior and change in viscosity of 55 vol% alumina slurries (at pH 9) (a) at different shear rates; (b) with different PMA concentrations.

277 this dose, viscosity of the slurries increased gradually with increase showed shear thinning behavior against shear rate sweep mea- 302

278 in polymer concentrations. surement at room temperature. Viscosities of all the slurries were 303

279 The pseudoplastic behavior of highly loaded slurries was due to strongly influenced by pH with same dispersant concentration. 304

280 formation of inter-particle network at relatively lower shear rate. Among all the slurries, minimum viscosity was evident for slurry 305

281 With gradual increase in shear rate, this inter-particle network was prepared at pH 9. The slurry with pH 6 showed highest viscos- 306

282 transformed into small flowable clusters and resulted in reduction ity, whereas moderate viscosity was obtained for slurry with pH 307

283 of slurry viscosity. For different PMA concentrations, viscosity of the 12. The viscosity values of the slurries at pH 6, 9, and 12 were 308

284 slurries reduced gradually with increase in polymer concentrations 3.18 Pa s, 1.34 Pa s, and 2.46 Pa s, respectively, at constant shear rate 309

285 up to 3.8 mg/g of alumina mainly due to increase in zeta potential of 5.6 s−1 . This result was also supported by lower zeta potential 310

286 value associated with higher surface coverage. At lower concen- value of suspension prepared at pH 9 (Fig. 2a). Thixotropic behavior 311

287 tration of PMA, there is actually incomplete coverage of polymer of highly loaded alumina slurries (55 vol%) was strongly depen- 312

288 molecules leading to strong network formation via inter-particle dent on PMA concentrations (Fig. 3). In this study, three different 313

289 adsorption of long chain molecules. At 3.8 mg of PMA/g of alu- PMA concentrations, such as 2.3 mg (below optimum), 3.8 mg (opti- 314

290 mina, the suspended powder surface had the highest surface charge mum) and 5.3 mg (above optimum) per gram of alumina, were 315

291 (−55 mV) owing to monolayer coverage of polymers. Beyond this chosen to investigate their effect on thixotropic behavior. From the 316

292 amount, the excess un-adsorbed polymer molecules remained in plot in Fig. 3, it is evident that the slurry with optimum amount of 317

293 the aqueous solution and their counter ions minimized the sur- dispersant showed minimum thixotropy in comparison to below 318

294 face charge by compression of the double layers which resulted and above the optimum amount of dispersant. However, the slurry 319

295 in reduction of zeta potential value. From this fact, it can be con- with less than optimum amount of dispersant had pronounced 320

296 cluded that 3.8 mg of PMA/g of alumina is the optimum amount of thixotropy in comparison to more than optimum amount, even 321

297 dispersant for preparation of highly loaded slurry. These data are with same magnitude of deviation from the optimum dispersant 322

298 also supported by RSH and zeta potential values. concentration. Actually, tendency of inter-particle network forma- 323

299 Further, 55 vol% slurries with 3.8 mg/g of alumina were also tion is more with less than optimum amount of dispersant owing 324

300 prepared at different pH (6, 9 and 12) to evaluate effect of pH to their less surface coverage with respect to optimum and above 325

301 on flowability (Fig. 2b). All the slurries prepared at different pH optimum dispersant concentrations. However, slurry with more 326

Fig. 2. Influence of pH on (a) zeta potential and (b) viscosity of 55 vol% alumina slurries prepared using optimum amount of dispersant (3.8 mg PMA/g alumina).

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005
 G Model
JASCER 27 1–7 ARTICLE IN PRESS
S. Mohanty et al. / Journal of Asian Ceramic Societies xxx (2013) xxx–xxx 5

Fig. 3. Thixotropy of 55 vol% alumina slurries expressed as difference in viscosity Fig. 5. TGA–DTA of vacuum dried alumina powder and powder residues of dis-
persed alumina slurries with different concentrations of PMA, viz. below optimum,
(ῃuw)
optimum and above optimum.

during downward ( ) and upward (

(ῃdw) ) shear rate sweep for slurries with
different amounts of PMA dispersant.
to symmetric stretching vibration. For alumina powder, the peak at 337

1636 cm−1 is mainly associated with physisorbed water on the sur- 338

327 than optimum amount of dispersant had lowering of zeta potential face, and the peaks at 1401 cm−1 and 1384 cm−1 are due to surface 339

328 which facilitated relatively higher inter-particle interaction due to hydroxide [39]. For FTIR spectra of alumina with different amount 340

329 compression of double layer. of PMA, the peak at ∼1636 cm−1 and additional peak at ∼1573 cm−1 341

330 FTIR spectra of dried PMA, as received alumina and alumina are associated with asymmetric stretching vibration of carboxylate 342

331 with different PMA concentrations were plotted under transmit- ions [23] of the adsorbed PMA. There were also two more peaks 343

332 tance mode as shown in Fig. 4. Being a polycarboxylic acid, PMA has at 1401 cm−1 and 1384 cm−1 associated with symmetric stretch- 344

333 major characteristic peak at 1723 cm−1 for C O stretching vibra- ing vibration of carboxylate ion [23]. However, carbonyl stretching 345

334 tion of free molecules [38]. Additionally for COO− group, peaks at vibration peak at 1723 cm−1 for dried PMA was absent in case of 346

335 1633 cm−1 and 1557 cm−1 were assigned for asymmetric stretch- alumina containing adsorbed PMA. It is noteworthy that the peaks 347

336 ing vibration and peaks at 1487 cm−1 and 1400 cm−1 appeared due associated to carboxylate symmetric and asymmetric vibrations 348

were broadened for optimum and above optimum (3.8 mg/g and 349

above) dispersant concentrations due to the presence of adsorbed 350

and unadsorbed carboxylate groups of the long chain polymers. 351

Fig. 5 shows weight loss profile of vacuum dried alumina with 352

different PMA concentrations of slurries and as received alumina 353

during thermogravimetric analysis in air. From the TGA curve of as 354

received alumina powder, 0.1% weight loss was evident mainly due 355

to evaporation of adsorbed water and surface hydroxide. Appear- 356

ance of two endothermic peaks at ∼100 ◦ C and ∼400 ◦ C was mainly 357

due to evaporation of adsorbed water and removal of surface 358

hydroxide, respectively [40]. In case of alumina with different con- 359

centrations of PMA, percent weight loss was relatively higher in 360

comparison to as received alumina due to presence of adsorbed 361

dispersant. Specifically, percent weight loss of 0.68, 0.98 and 1.0 362

was evidenced for below optimum, optimum and above optimum 363

amount of dispersant, respectively. These data clearly evidenced 364

that there was a significant increase in %weight loss with increase 365

in dispersant amount from below optimum to optimum level, 366

whereas the increase in %weight loss was insignificant with beyond 367

the optimum level. Actually, with addition of more than optimum 368

amount of dispersant, surface coverage did not increase signifi- 369

cantly and unadsorbed PMA remained in the aqueous medium. 370

Further, presence of exothermic peak in DTA curve was also evident 371

for different dispersant concentrations within 239–400 ◦ C mainly 372

due to thermal decomposition of adsorbed PMA. 373

The green alumina bodies were prepared through casting of 374

slurry into silicone rubber mold with optimum PMA concentra- 375
Fig. 4. FTIR spectra of (a) freeze dried PMA; (b) vacuum dried alumina powder;
vacuum dried powder residues from alumina slurries with (c) below optimum, (d) tion (3.8 mg/g) at pH 9. The vacuum dried samples were ready for 376

optimum and (e) above optimum amount of PMA (at pH 9), respectively. sintering and did not have any visual crack on the surface. All the 377

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005
 G Model
JASCER 27 1–7 ARTICLE IN PRESS
6 S. Mohanty et al. / Journal of Asian Ceramic Societies xxx (2013) xxx–xxx

Fig. 6. SEM micrographs of fracture surface of (a) green, (b) bisque fired and (c) sintered alumina at pH 9 with optimum dispersant amount.

378 samples were polished and sintered at 1550 ◦ C for 2 h. The sintered Conflict of interest statement 424

379 alumina prepared using optimum slurry did not develop any sig-
380 nificant warpages or cracks during sintering. All the samples had The authors declare that there are no conflicts of interest. 425

381 ∼15% linear shrinkage associate with densification during sinter-

382 ing. The green density of the sample was ∼65% of the theoretical
Acknowledgements 426
383 density. Sintered density of the sample was ∼99.5% for optimum
384 amount of dispersant. The green body microstructure of the frac-
Funding of Department of Biotechnology (DBT), Council of Sci- 427
385 ture surface had no significant defects associated to agglomerates
entific & Industrial Research (CSIR) and Department of Science and 428
386 and entrapped air bubbles (Fig. 6a). The particle distribution of
Technology (DST) Govt. of India is acknowledged. Financial sup- 429
387 bisque fired samples prepared using optimum dispersant amount
port of Indian Institute of Technology Kharagpur is acknowledged 430
388 was homogeneous as shown in Fig. 6b. The sintered microstruc-
for Mr. Saralasrita Mohanty. 431
389 ture of alumina showed homogeneous grain growth with minimum
390 porosity (Fig. 6c). The average grain size of the sintered alumina
391 is ∼2 ␮m. Flexural strength of alumina with optimum concentra- References 432

392 tion of dispersant showed 483 ± 22 MPa. Here it is to be noted that

[1] J.A. Lewis, J. Am. Ceram. Soc., 83, 2341–2359 (2000). 433
393 Tsetsekou et al. [41] carried out optimization study of polyacrylic [2] J.S. Reed, 2nd ed., Wiley, New York (1995). Q6 434
394 acid dispersant (Duramax D 3005) for alumina (mean particle [3] F.F. Lange, J. Am. Ceram. Soc., 72, 3–15 (1989). 435
395 size 3 ␮m and specific surface area 4.5 m2 /g). They have reported [4] M. Michálková, K. Ghillányová and D. Galusek, Ceram. Int., 36, 385–390 (2010). 436
[5] A.R. Studart, V.C. Pandolfelli and J. Gallo, Am. Ceram. Soc. Bull., 81, 36–44 (2002).
396 bending strength of ∼330 MPa for alumina with sintered den- 437
[6] U. Paik, V.A. Hackley and H.W. Lee, J. Am. Ceram. Soc., 82, 833–840 (1999). 438
397 sity of 3.7 g/cm3 . Similarly, Volceanov et al. [42] reported bending [7] D.J. Shanefield, Kluwer Academic Publishers, Dordrecht, Netherlands (1995). 439
398 strength of 470 MPa for alumina (average particle size 0.8 ␮m) [8] S. Sumita, W.E. Rhine and H.K. Bowen, J. Am. Ceram. Soc., 74, 2189–2196 (1991). 440
[9] J.D. Yates and S.J. Lombardo, J. Am. Ceram. Soc., 84, 2274–2280 (2001).
399 with sintered density 3.8 g/cm3 using polyelectrolyte dispersant 441
[10] S. Dhara and P. Bhargava, Int. J. Appl. Ceram. Technol., 3, 382–392 (2006). 442
400 (Dolapix). Thus, PMA can be used for dispersion of alumina at pH 9 [11] B.J. Briscoe, A.U. Khan and P.F. Luckham, J. Eur. Ceram. Soc., 20, 2141–2147 443
401 for the preparation of highly loaded slurry in aqueous medium. (1998). 444
[12] W.D. Teng, M.J. Edirisinghe and J.R.G. Evans, J. Am. Ceram. Soc., 80, 486–494 445
(1997). 446
[13] S. Dhara and P. Bhargava, J. Am. Ceram. Soc., 88, 547–552 (2005). 447
402 4. Conclusions [14] K. Nagata, J. Ceram. Soc. Jpn., 100, 1271–1275 (1992). 448
[15] S. Mohanty, A.P. Rameshbabu and S. Dhara, J. Am. Ceram. Soc., 95, 1234–1240 449
(2011). 450
403 PMA was successfully used as an anionic dispersant for the [16] H.Y. Yasuda, S. Mahara, Y. Umakoshi, S. Imazato and S. Ebisu, Biomaterials, 21, 451
404 preparation of aqueous alumina slurry. Stability of 1 wt% alu- 2045–2049 (2000). 452

405 mina suspensions in presence of different PMA concentrations [17] C. Zha and H. Sun, J. Colloid Interface Sci., 162, 52–58 (1994). 453
[18] S. Dhara, P. Bhargava and K.S. Ramakanth, Am. Ceram. Soc. Bull., 83, 9201–9206 454
406 (2.3–5.3 mg PMA per gram of alumina powder) at pH 9 was (2004). 455
407 increased significantly as evident by RSH measurement due to [19] Y. Takao, T. Hotta, K. Nakahira, M. Naito, N. Shinohara, M. Okumiya and K. 456

408 increase in zeta potential. Highly loaded alumina slurries with Uematsu, J. Eur. Ceram. Soc., 20, 389–395 (2000). 457
[20] V.A. Hackley, J. Am. Ceram. Soc., 80, 2315–2325 (1997). 458
409 55 vol% powder loading were prepared using PMA as dispersant. [21] J. Cesarano, I.A. Aksay and A. Bleier, J. Am. Ceram. Soc., 71, 250–255 (1988). 459
410 Flow behavior of 55 vol% alumina slurries with different concen- [22] R. Jewad, C. Bentham, B. Hancock, W. Bonfield and S.M. Best, J. Eur. Ceram. Soc., 460
411 trations of PMA showed shear thinning behavior. The slurry with 28, 547–553 (2008). 461
[23] P.C. Hidber, T.J. Graule and L.J. Gauckler, J. Am. Ceram. Soc., 79, 1857–1867
3.8 mg PMA/g of alumina had minimum viscosity (1.5 Pa s at 5.6 s−1 )
462
412
(1996). 463
413 and was the optimum dispersant concentration owing to high zeta [24] L.C. Guo, Y. Zhang, N. Uchida and K. Uematsu, J. Am. Ceram. Soc., 81, 549–556 464
414 potential value (−55 mV), minimum thixotropy and better flowa- (1998). 465
[25] J. Davies and J.G.P. Binner, J. Eur. Ceram. Soc., 20, 1555–1567 (2000). 466
415 bility. Adsorption of PMA on alumina powder surface was evident
[26] D. Hanaor, M. Michelazzi, C. Leonelli and C.C. Sorrell, J. Eur. Ceram. Soc., 32, 467
416 by FTIR spectroscopy. Further, differential quantitative adsorption 232–244 (2012). 468
417 of PMA on alumina powder surface was evident by thermogravi- [27] P.C. Hidber, T.J. Graule and L.J. Gauckler, J. Am. Ceram. Soc., 78, 1775–1780 469

418 metric analysis. SEM microstructure of the fracture surface of (1995). 470
[28] J. Davies and J.G.P. Binner, J. Eur. Ceram. Soc., 20, 1539 (2000). 471
419 green body prepared using optimum level of dispersant revealed [29] J. Cesarano and I.A. Aksay, J. Am. Ceram. Soc., 71, 1062–1067 (1988). 472
420 homogeneous microstructure with negligible defects. The sintered [30] L. Wang, Y.P. Chin and S.J. Traina, Geochim. Cosmochim. Acta, 61, 5313–5324 473

421 samples achieved ∼99.5% of the theoretical density while sintered (1997). 474
[31] M. Euvrard, A. Martinod and A. Neville, J. Cryst. Growth, 317, 70–78 (2011). 475
422 at 1550 ◦ C for 2 h. The alumina samples with optimum dispersant [32] M.P.C. Weijnen and G.M.V. Rosmalen, Desalination, 54, 239–261 (1985). 476
423 amount had flexural strength of 483 ± 22 MPa. [33] M. He, J. Addai-Mensah and D. Beattie, Chem. Eng. J., 152, 471–479 (2009). 477

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005
 G Model
JASCER 27 1–7 ARTICLE IN PRESS
S. Mohanty et al. / Journal of Asian Ceramic Societies xxx (2013) xxx–xxx 7

478 [34] G.C. Chitanu, M. Rinaudo, J. Desbriares, M. Milas and A. Carpov, Langmuir, 15, [39] Z. Sarbak, Appl. Catal. A: Gen., 177, 85–97 (1999). 484
479 4150–4156 (1999). [40] M.M. Martin-Ruiz, L.A. Perez-Maqueda, T. Cordero, V. Balek, J. Subrt, N. Musafa 485
480 [35] N. Pekel and O. Güven, Polym. Int., 57, 637–643 (2008). and J. Pascual-Cosp, Ceram. Int., 35, 2111–2117 (2009). 486
481 [36] D. Sigrid, L. Alexander and H. Martina, WIPO Patent No. WO/2010/069519 [41] A. Tsetsekou, C. Agrafiotis and A. Milias, J. Eur. Ceram. Soc., 21, 363–373 487
482 (2010). (2001). 488
483 [37] S. Mohanty and S. Dhara, Indian Pat. 588/KOL/2012 (2012). [42] E. Volceanov, A. Volceanov and S. Stoleriu, J. Eur. Ceram. Soc., 27, 759–762 489
[38] A.N. Hess and Y.P. Chin, Colloids Surf. A, 107, 141–154 (1996). (2007). 490

Please cite this article in press as: S. Mohanty, et al., Poly(maleic acid) – A novel dispersant for aqueous alumina slurry, J. Asian Ceram.
Soc. (2013), http://dx.doi.org/10.1016/j.jascer.2013.05.005