J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... 413–420 DISPERSION OF AQUEOUS Y 2 O 3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS BY COLLOIDAL PROCESSING VODNA DISPERZIJA NANO DELCEV Y 2 O 3 IN IZDELAVA PROSOJNE KERAMIKE S KOLOIDNIM POSTOPKOM Jiao He * , Xin Zhang, Jingbao Lian, Xue Zhang, Mingxia Lei School of Mechanical Engineering, Liaoning Petrochemical University, Fushun, 113001, P.R. China Prejem rokopisa – received: 2024-03-05; sprejem za objavo – accepted for publication: 2024-04-22 doi:10.17222/mit.2024.1129 In the present study, transparent Y2O3 ceramics were successfully fabricated via colloidal processing, employing polyethylenimine (PEI) as an effective dispersant. The effect of PEI on nanosized Y2O3 suspensions was characterized by zeta-potential, adsorption behavior, rheological properties, and sedimentation test. The addition of PEI shifted the IEP of the Y2O3 powder towards a more alkaline pH range, and the adsorption of PEI on the Y2O3 surfaces gradually increased with the amount of PEI until reaching the saturation adsorption at 1.5 w/%. The PEI amount of 1.5 w/% was superior for the preparation of Y2O3 suspension as a result of a lower viscosity or sedimentation behavior. The viscosity of the suspension depended on the solids loading, increasing with the amount of powder. The optimal rheological behavior was achieved with 29 /% Y2O3 in the suspension, enabling the centrifugal slip casting of complex-shaped green bodies with a packing density of 43 %. Including an additional CIP treatment boosted the packing density of the green compacts, achieving above 50 %. Vacuum sintering the com- pacts at 1700 °C for 5 h yielded high-density ceramics exhibiting an in-line transmittance of approximately 73 % at a wave- length of 1100 nm. Keywords: Y2O3, suspension, rheological behavior, transparent ceramics V ~lanku avtorji opisujejo uspe{no izdelavo prosojne Y2O3 keramike s koloidnim postopkom. Pri tem so uporabili Polietilenimin (PEI) kot u~inkovit dispergent. U~inek PEI na nanodelce Y2O3 v vodni suspenziji so dolo~ili z zeta potencialom, adsorpcijo, reolo{kimi lastnostmi in testom sedimentacije (posedanja), Dodatek PEI je premaknil izoelektri~no to~ko (IEP) nano delcev Y2O3 bolj v smeri alkalnega pH obmo~ja in adsorpcija PEI na povr{inah nanodelcev Y2O3 je postopoma nara{~ala z vsebnostjo PEI dokler ni bilo dose`eno adsorpcijsko nasi~enje pri 1,5 w/%. Vsebnost PEI pri 1,5 w/% je bila optimalna za pripravo Y2O3 suspenzije, ker je le-ta imela majhno viskoznost in majhno nagnjenost k posedanju nano delcev. Viskoznost suspenzije je bila odvisna od dele`a trdne faze oziroma je nara{~ala s pove~evanjem dele`a pra{nih nano delcev. Optimalne reolo{ke lastnosti so dosegli avtorji pri 29 /% Y2O3 v suspenziji. To je omogo~ilo postopek centrifugalnega nalivanja kerami~ne go{~e (angl.: cen- trifugal slip casting) kompleksnih oblik surovcev s 43 % gostoto pakiranja (nalivanja). Z dodatnim hladnim izostatskim stiskanjem (CIP; angl.: cold isostatic pressing) so gostoto surovcev pove~ali na pribli`no 50 %. Sledilo je pet urno vakuumsko sintranje surovcev pri 1700 °C za tvorbo goste Y2O3 keramike. Transparentnost keramike so dosegli pri cca 73 % za valovno dol`ino 1100 nm. Klju~ne besede: Y2O3, suspenzija, reologija, transparentna keramika 1 INTRODUCTION Transparent Y 2 O 3 ceramics have aroused increasing research interest for many decades, primarily due to their excellent optical transparency over a wide wavelength range (0.2–8 μm), high thermal conductivity and low phonon energy, making them very suitable for various optical materials, such as IR windows and domes, host materials for solid-state laser or scintillator, and bulb en- velopes, etc. 1–4 Widely accepted, achieving the fabrica- tion of high-quality transparent ceramics requires special attention to each key step in ceramic production, namely, the synthesis of starting powders with excellent sintering activity, the consolidation of high-density and uniform green compacts, and the sintering of near-full-dense ce- ramics with pore-free structures. 5–7 During the consolidation process, it is widely ac- knowledged that colloidal processing offers advantages in facilitating the production of large-sized and com- plex-shaped green compacts, exhibiting a homogeneous microstructure and the desired green density, whereas traditional dry uniaxial pressing methods often pose challenges in achieving such characteristics. 8–10 The criti- cal technology of colloidal processing is to achieve a sta- ble ceramic suspension with a high solids loading, low viscosity, and good dispersibility. 11,12 This is crucial for both enhancing the ease of handling and attaining a final product with high reliability. In both aqueous and non-aqueous nanopowder suspensions, the presence of a high surface area and a reduced separation distance among the nanoparticles gives rise to attractive interac- tions, thereby causing rapid settling and aggregation of the suspensions. Therefore, to get well-dispersed and sta- ble nanopowder suspensions, the major stabilization Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 413 UDK 666.3:661.864.1 ISSN 1580-2949 Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 58(3)413(2024) *Corresponding author's e-mail: hejiao@lnpu.edu.cn (Jiao He) mechanisms, including electrostatic, steric, or electro- steric stabilization mechanisms should be introduced in ceramic colloidal processing to offset the van der Waals attraction. 13 This often necessitates the addition of suit- able dispersants that are adsorbed from solution onto the particle surfaces to control the interparticle forces, in- hibit aggregation, and achieve stability. 14 Mouzon et al. found that electrosteric stabilization in the presence of polyelectrolyte A40 was effective for obtaining well-dis- persed Y 2 O 3 suspensions. The optimal concentration of A40 was determined to be 1.0 w/%, yielding the lowest viscosity of the suspension. 15 Sun prepared a stable sus- pension with 81 w/% solids loading, using water-soluble copolymer (0.2 w/% Ib104 plus 0.3 w/% Ib600) as both a dispersing and gelling agent. 16 The transmittance mea- sured at 1100 nm of the resultant Y 2 O 3 ceramics was found to be 80.9 %. More recently, a novel aqueous col- loidal processing route has been explored to improve the stability and dispersion of Y 2 O 3 suspensions through the combined use of a ZrO 2 coating agent and polyelectro- lyte dispersant. 17 However, the use of the coating agent necessitates a higher sintering temperature or pres- sure-assisted sintering during the sintering process in or- der to densify the final products. Over the past few de- cades, a series of anionic polymers has often been utilized to effectively improve the stability of Y 2 O 3 sus- pensions. 18–20 However, fewer investigations of the prepa- ration of Y 2 O 3 aqueous suspension by adding cationic dispersants have been reported, and those available were not systematic. Polyethylenimine (PEI) has been employed as a cationic dispersant to enhance the stability of various ce- ramic powder suspensions such as ZrO 2 , 21 Y-TZP, 22 SiC, 23 and Si 3 N 4 . 24 In a neutral or acidic environment, the positively charged PEI could adsorb on ceramic particle surfaces, resulting in an electrosteric effect that hinders the flocculation of the ceramic suspension. 21 Xu et al. re- ported that PEI was used as a dispersant to improve the dispersion properties of Y 2 O 3 powders (average particle size of 200 nm) in ethanol. Adding 1.5 w/% PEI, a Y 2 O 3 alcoholic suspension with a solids loading of 20.8 /% and a viscosity of less than 0.1 Pa·s at a shear rate of 10 s –1 was obtained. 25 In this work, PEI was selected as a dispersant for aqueous Y 2 O 3 suspensions. The colloidal properties of Y 2 O 3 suspensions with regard to PEI amount are extensively elucidated. In addition, the effect of ceramic solids loading on the rheological properties of the PEI-based suspensions was also explored. Subse- quently, the optimal conditions for Y 2 O 3 suspensions were determined and used to produce cylindrical green compacts by centrifugal slip casting. Then the vacuum sintering was carried out to fabricate the Y 2 O 3 ceramics. Finally, optical transmission of the resultant sintered samples was examined. 2 EXPERIMENTAL PART 2.1. Raw materials In the present study, commercially available, high-purity Y 2 O 3 (99.99%) was utilized. This powder has a specific surface area of 8.17 m 2 /g measured through N 2 adsorption (Model TriStar II 3020, Micrometritics In- strument Corp., Norcross, GA) The morphology of the as-received powder, measured by scanning electron mi- croscopy (SEM), is illustrated in Figure 1, revealing that the primary particle size was approximately 96 nm. Be- fore preparing the suspension, the powder underwent pretreatment. Polyethyleneimine (PEI) (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), with an average molecular weight of 10,000 was taken as the dis- persant. 2.2 Preparation process A series of Y 2 O 3 suspensions at a solids loading of 5 /%, containing increasing amounts of dispersant (ex- pressed as the total mass of Y 2 O 3 powder), were pre- pared. In a typical procedure, the dispersant was dis- solved in distilled water in a polyethylene jar containing zirconia balls. Then Y 2 O 3 powder was gradually added to the mixture. The resultant suspensions were ball-milled for6hinaplanetary mill. Various suspensions at the optimal dispersant amount with a solids loading ranging from 10 /% to 31 /% for different ball-milling times were prepared following the same procedure. Y 2 O 3 green bodies were fabricated by a centrifugal slip-casting method. Selected, concentrated Y 2 O 3 sus- pensions were poured into the molds and left for consoli- dation using a laboratory centrifuge at constant speed of 3000 min –1 for 40 min. After 24 h at room temperature, the green bodies were extracted from the mold, and fur- ther oven dried at 80 °C for over 12 h. Then, the green bodies were pre-sintered at 1000 °C for4hinamuffle furnace to remove the organic additives, and subse- quently sintered at 1700 °C for5hi nav a c u u mo f J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... 414 Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 Figure 1: SEM micrograph of the Y 2 O 3 powder 1.0×10 –3 Pa in a furnace with a molybdenum heating el- ement (VSF-7, Shenyang, China). 2.3 Characterization and test The zeta-potential of the aqueous Y 2 O 3 suspension was estimated by an acoustic and electroacoustic spec- trometer (DT-1202, Dispersion Technology Inc., Bedford Hills, USA). In this measurement, 5 w/% Y 2 O 3 suspen- sion was employed and the pH was adjusted using 1 mol/L HCl and 1 mol/L KOH aqueous solution. To study the adsorption behavior of the PEI, Y 2 O 3 suspensions with 5 /% solids loading were utilized. The as-received suspensions were centrifuged at 3000 min –1 for 40 min, the supernatants were poured out and the sediments were dried in the oven at 100 °C for 5 h. Af- terwards, the adsorbed amount of PEI was determined using a DTA/TG analyzer (Model SETSYS Evolu- tion-16, Setaram, Lyons, France) by measuring the weight loss of the dried powders during heating in a flowing oxygen atmosphere. Rheological behavior mainly focuses on the viscosity curve of Y 2 O 3 suspensions that were evaluated using a cone-plate viscometer (Brookfield DV-II+Pro, Brook- field Engineering Laboratories, USA) at room tempera- ture. The stability of the Y 2 O 3 suspension was easily char- acterized by sedimentation tests, which offer an approxi- mation of colloidal stability. The suspensions were kept in glass graduated cylinders for more than 60 h and the volume of the sedimentation as a percentage of the total suspension volume (Normalized volume) was measured as a function of settling time. Archimedes’immersion method was applied to deter- mine the densities of the obtained green bodies. The rela- tive density was determined by calculating the ratio of the actual density to the theoretical density of Y 2 O 3 ,us - ing distilled water as the immersion medium. The microstructure of the green bodies was observed using a field-emission scanning electron microscope (FE-SEM). The samples were cut into fractured slices, and subse- quently coated with gold to enhance their conductivity before being examined. Subsequently, polished pellets with a thickness of 1.2 mm, which were cut from the sintered specimens, were employed to measure the opti- cal transmittance using an ultraviolet/visible/near-infra- red spectrophotometer (Model Lambda 750S, Perkin Elmer, CT, USA). 3 RESULTS AND DISCUSSION 3.1 Optimization of PEI amount As previously mentioned, achieving effective colloi- dal processing and maximizing uniformity in the green bodies depends on the ability to produce a suspension that is not only well-dispersed, but also possesses a rela- tively high solids loading. 26 This objective was realized through the addition of the PEI dispersant to a wa- ter-based solution, effectively stabilizing the suspension. The optimization of the operational parameters was iden- tified through a sequence of analyses and assessments. Zeta-Potential Analysis Figure 2 illustrates the rela- tionship between the zeta-potential value and the amount of PEI for individual fresh dilute suspensions of Y 2 O 3 . For Y 2 O 3 powder without PEI, increasing the pH value from 6.2 to a significantly basic range, the surface charges of the particles experienced a notable shift, transitioning from a positive charge to a negative charge. This shift was a clear sign of an isoelectric point (IEP) of approximately 9.7, where the suspension was in its least stable state in terms of particle agglomeration. With the increasing of the pH, a steep increase is observed in the absolute value of the zeta potential. In the case of Y 2 O 3 containing 0.5 w/% PEI, the values of the zeta potential changed to positive across the pH range < 10.9, indicat- ing PEI successfully adsorbed on the surface of Y 2 O 3 particles. This phenomenon can be attributed to the fact of polyethyleneimine (PEI) being a cationic poly- electrolyte. When acid was introduced to a suspension containing PEI, it reacted with the basic –NH– groups in PEI, leading to their neutralization and the formation of a positively charged polymer structure (–NH 2+ –). 23 Conse- quently, these positively charged groups exhibited a strong affinity for adsorption onto negatively charged Y 2 O 3 surfaces through electrostatic interactions, causing a shift in the IEP towards a more alkaline range. Then, at a higher PEI amount of 1.5 w/%, the IEP moved to a pH of 11 due to the absorption of PEI on the Y 2 O 3 particles. With the further increasing amount of PEI, there was no notable alteration in the zeta-potential, indicating a state of saturation adsorption. Additionally, both within the selected amount range of PEI and across the entire pH range, the absolute value of the potential remained below 30 mV, indicating a lack of sufficient electrostatic repul- sion to disperse the powder particles. J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 415 Figure 2: Zeta-potential of Y 2 O 3 particles in the absence and presence of various amounts of PEI Adsorption of PEI onto Y 2 O 3 particles Figure 3 presents the variation in the adsorption amount of PEI on the surface of Y 2 O 3 powder particles as a function of the initial PEI addition. As evident from Figure 3, the adsorption of PEI on the surface of Y 2 O 3 particles gradually rose with an increase in PEI amount from 0.1 w/% to 1.5 w/%. Afterwards, a saturation ad- sorption amount of 0.5 mg/m 2 was observed. The adsorp- tion amount was kept almost constant as the PEI amount exceeded 1.5 w/%, revealing that PEI on the surface of Y 2 O 3 particles had reached its adsorption limit, in agree- ment with the results of the zeta-potential measurement. Suspension Rheological Properties Dispersants play a role in improving the rheological and dispersion stability of suspensions, both of which are very dosage-sensitive. Figure 4 depicts the variations in the viscosity of the Y 2 O 3 suspensions with varying PEI amounts in the range from 0 w/% to 2.5 w/%, at the shear rate of 1.32–264 s –1 .AllY 2 O 3 suspensions exhibited clear shear thinning behavior, indicative of a non-Newto- nian rheological behavior. As is obvious in Figure 4a, the inclusion of PEI dispersant notably enhanced the sta- bility of the Y 2 O 3 suspension, exhibiting a rapid reduc- tion in viscosity as the increased PEI amount in the range of not exceeding 1.5 w/%, as a result of the absorption of PEI on the surfaces of powder particles. Upon regulating the PEI amount from 0 w/% to 1.5 w/%, the viscosity of Y 2 O 3 suspensions presented a substantial transformation, shifting from 134.7 mPa·s to 2.7 mPa·s at a shear rate of 13.2 s –1 , as seen in Figure 4b. Then, at higher PEI con- tents of up to 2.0 w/% and 2.5 w/%, a minor increase in the viscosity of the suspension was detected. This obser- vation was explained by the mutual cross-linking effect of the PEI after reaching adsorption saturation on the powder surfaces, which subsequently led to a deteriora- tion in the stability of the suspension. The suspension containing 1.5 w/% PEI exhibited the lowest viscosity value, which is indicative of a superior suspension stabil- ity, enabling easy particle movement. Sedimentation test A sedimentation plot versus the time of Y 2 O 3 suspen- sions with various PEI amounts is presented in Figure 5. It is observed from Figure 5 that the suspension in the absence of PEI settled rapidly within 30 min. After 120 min, the normalized volume of the suspensions at- J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... 416 Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 Figure 3: Adsorption amount of PEI in Y 2 O 3 suspensions versus the amount of added PEI Figure 5: Variation of sedimentation volume versus time of Y 2 O 3 sus- pensions with various amounts of PEI Figure 4: Evolution of the viscosity of Y 2 O 3 suspensions as a func- tion of PEI amount tained a value of 0.94. Further extending the settling time, it was noted that the sedimentation volume of the suspension remained unchanging, indicating the occur- rence of severe flocculation. Y 2 O 3 nanoparticles with a high surface energy easily gave rise to severe agglomera- tion among the particles. This agglomeration, in turn, triggered flocculation within the suspension, ultimately leading to the complete loss of fluidity among the parti- cles in the suspension. In the case of containing PEI, the starting settling time of the slurry was delayed, and set- tling was observed to be much slower, due to the PEI ad- sorbed on the Y 2 O 3 powder. PEI is a highly branched macromolecule with a general chemical formula of [–CH 2 –CH 2 –NH–] n . 23 The partially ionized or non-ion- ized PEI could adsorb onto the negatively charged sur- face and thus provide an electrosteric or steric stabiliza- tion for ceramic suspensions. It was reported that the degree of PEI dissociation ( ) exhibits an increasing trend as the pH decreases, and partially dissociated PEI with = 0.1 at pH 10 can be obtained. 24 The pH of the suspension without pH adjustment was approximately 9.4, at which value of partially dissociated PEI was 0.1. In all cases of suspensions with PEI, the low abso- lute potential indicated that positively charged polymer groups adsorbed onto the surface of powder could not provide sufficient electrostatic repulsive force to prevent the aggregation of particles, As seen from the results of the zeta-potential measurement in Figure 2. Moreover, lower viscosity and sedimentation of suspensions with PEI were achieved, revealing that the stabilization of sus- pensions may be dominated by incorporation of electro- steric and steric effects developed from the ionized and unionized PEI. The higher sedimentation can be ex- plained in view of the lack of enough electrosteric or steric forces to overcome the Van der Waals attraction force for PEI amounts less than 1.0 w/%, due to incom- plete coverage of the Y 2 O 3 particle surfaces. However, after a certain amount of 1.5 w/%, the settling volume in- creased again due to excessive PEI molecules in solu- tion. A mass of free polymers in solution entangled with each other, thus causing depletion flocculation. 22 The optimum PEI amount that can be added to a Y 2 O 3 sus- pension to achieve suspension stability was found to be 1.5 w/%, which possessed the lowest viscosity and small sedimentation. Optimization of dispersion time The viscosity presented in Figure 6 shows the effi- ciency of ball-milling time on the stability of suspen- sions containing 1.5 w/% PEI, as a function of shear rate. It is evident that as the ball-milling time increased, the viscosity of the suspension decreased significantly. At a ball milling time of 1 h, the viscosity measured at a shear rate of 1.32 s –1 was 60.55 mPa·s. Moreover, upon reach- ing a ball-milling time of 6 h, the viscosity attained its minimal value of 24 mPa·s at 1.32 s –1 , which indicated the finest homogeneity in the Y 2 O 3 system. Further ex- tending the ball milling time to 8 h and 10 h resulted in a slight increase in viscosity to 25.90 mPa·s and 29.66 mPa·s, respectively. The purpose of ball milling is to achieve a uniformly dispersed suspension through the collisions between the milling balls and Y 2 O 3 particles. In the initial stages of ball milling, specifically at1hand 2 h, the process focused on refining and homogenizing the powder particles. The application of external force through ball milling could enhance the penetration of the dispersing medium, resulting in its uniform dispersion among the Y 2 O 3 particles. Additionally, this refined pro- cess effectively eliminated particle agglomeration and promoted the uniform adsorption of PEI onto the Y 2 O 3 powder surfaces. Thus, increasing the ball-milling time could enhance the stability and fluidity of the Y 2 O 3 sus- pensions, manifesting as a decrease in viscosity. How- ever, the powder particles may become excessively grinded and refined under prolonged mechanical impact of8hor10h,causing the higher specific surface area. These increases made the decrease in available volume of the dispersing medium within the system, narrowing the distance between the particles. Consequently, the Van der Waals attractive force increased and favored particle agglomeration, leading to an elevation in suspension vis- cosity and a subsequent decline in ball-milling effi- ciency. Furthermore, the higher specific surface area of the Y 2 O 3 particles indicated that more of their surfaces were exposed into the liquid phase, causing the 1.5 w/% PEI dosage to give inadequate coverage of the Y 2 O 3 par- ticle surfaces. Besides, a prolonged dispersion time can induce severe hydrolysis behavior of the nanosized Y 2 O 3 powders and cause a significant amount of trivalent yttria cations and their hydroxo complexes in the suspension, strongly decreasing the stability of the suspension, shown by the following equations : 27 OH – + + H 2 O–Y=Y–OH+H 2 O= =Y–O – +H 3 0 + (1) Y 3+ + 3OH – = Y(OH) 3 (s) (2) Y 3+ +OH – = Y(OH) 2+ (3) J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 417 Figure 6: Variation of viscosity for Y 2 O 3 suspensions as a function of ball-milling times 2Y 3+ + 2OH – =Y 2 (OH) 2 4 + (4) Therefore, in this experiment, ball milling for6hwas determined as the optimum dispersion condition. Rheological study to optimize solids loadings of Y 2 O 3 suspension In colloidal processing, carefully controlling the ap- propriate solids loading of the suspension is crucial for achieving optimal particle packing in the green body, while also maintaining a suitable viscosity. Viscosity measurements were conducted at varying shear rates to investigate the impact of solids loading on the rheologi- cal behavior, as depicted in Figure 7. The solids loading of Y 2 O 3 was varied from 15 /% to 31 /%. From Fig- ure 7, it is observed that all the suspensions showed shear thinning behavior. Under the tested range of shear rates, the suspension containing 15 /% solids loading exhibited lower viscosity than the other suspensions. This suspension exhibiting a remarkably low viscosity demonstrated exceptional fluidity, yet it was insuffi- ciently viscous to form perfect green bodies. Therefore, it is advisable to elevate the solids loading to attain green bodies with a higher density. As Figure 7 shows, the vis- cosity of the suspension substantially boosted by increas- ing the solids loading. In all the cases, the viscosity of the suspension changed from 84.0 mPa·s to 1401.8 mPa·s at a shear rate of 1.32 s –1 . when the solids loading ex- ceeded 31 /%, it became too stiff to meet the test- ing-range requirements. This is attributed to the fact of higher hydrodynamic interaction resulting from the in- creasing number of particles as the solids loading in- creased. Additionally, by elevating the solids loading, the mean distance between the particles decreased, as a re- sult, an intensified attractive force among them, leading to an increased viscosity. In the case of 29 /%, the sus- pension exhibited a viscosity of less than 1000 mPa·s over a broad range of shear rates. This level of solids loading proved advantageous not only for achieving a high density of green bodies but also for casting. Consolidation of Y 2 O 3 suspensions and sintering of transparent ceramics Consolidation of the green bodies using suspensions with various solids loading and 1.0 w/% PEI added was performed by centrifugal slip casting. Figure 8 demon- strates the influence of solids loading on the green rela- tive density. As evident from Figure 8, the sample with a solids loading of 29 /% exhibited the highest packing density of 43%. As the solids loading increased, the green relative density also increased. However, an excep- tion was observed for the sample with 31 /%, which ex- hibited a slightly lower density than the sample with 29 /%. This deviation can be attributed to unsuitable rheological properties of the suspension, which aligned with the rheological behavior in Figure 7. Figure 9 shows the fracture surfaces of Y 2 O 3 green bodies with different solids loadings. There was a differ- ence in the homogeneity among these samples. The green compacts for 15 /% showed a poor dispersion of particles due to excessively low solids loading. At a 29 /% solids loading, a more homogeneous and com- pact microstructure with small open pores can be ob- served. In sharp contrast, serious agglomeration and large pores were discovered in the green bodies of 31 /% suspension, thereby decreasing their density and ho- mogeneity. 28 It is well known that green compacts exhib- iting a narrow open pore size distribution and a uniform microstructure demonstrate superior sinterability, en- abling facile densification at relatively lower sintering temperatures. According to the above analysis, the green bodies, which contained 1.5 w/% PEI and possessed a solids loading of 29 /%, were selected for the fabrica- tion of transparent ceramics. However, the packing den- sity of 43 % achieved was inadequate for subsequent pressureless sintering due to their porous structure. To further boost the packing density of the consolidated green bodies, an additional CIP treatment was utilized at a pressure of 200 MPa. The obtained green bodies dem- onstrated exceptional compressibility. Following the CIP treatment, the densities of the samples increased to over J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... 418 Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 Figure 7: Effect of different solids loading on the rheological property of Y 2 O 3 suspensions containing 1.5 w/% PEI Figure 8: Relative density of cast green bodies from various solids loading suspensions 50 %, indicating a significant improvement in their com- paction properties. Vacuum sintering was carried out on the consolidated green bodies. The relative density of the centrifugally cast ceramics was 99.9 %. Figure 10 shows the optical transmission spectra of the mirror-polished Y 2 O 3 ceramic specimen with a thickness of 1.2 mm. At the wave- lengths of 1100 nm and 2000 nm, the in-line transpar- ency approached 73 % and 78 %, respectively. The inset in Figure 10 is the corresponding image of the as-ob- tained ceramic specimen, exhibiting an approximate di- ameter of 9 mm and a length extending to 20 mm. Nota- bly, the letters below the ceramic specimen were clearly visible, further emphasizing their remarkable optical transparency. 4 CONCLUSIONS Transparent Y 2 O 3 ceramics were prepared using a colloidal processing technique, employing nano-sized Y 2 O 3 powders as the starting material. The effectiveness of the PEI in enhancing the colloidal stability of the aqueous Y 2 O 3 suspension was investigated. The intro- duction of PEI shifted the IEP of Y 2 O 3 powder towards a higher pH value. The adsorption of PEI on the Y 2 O 3 sur- face increased with the amount of PEI and reached saturation adsorption at 1.5 w/%. A well-dispersed Y 2 O 3 suspension was obtained at an amount of 1.5 w/%, at which the viscosity and sedimentation volume of the sus- pension decreased to a minimum value. By optimizing the ball-milling time, the stability of the suspension can be significantly improved. The viscosity of the suspen- sion exhibited a dependence on solids loading, demon- strating an increase as the amount of powder boosted. A higher solids loading with suitable rheological behavior was achieved in the case of 29 /% of Y 2 O 3 in the sus- pension, which for centrifugal slip casting led to com- plex-shaped green bodies with a packing density of 43 %. By incorporating an additional CIP treatment, the packing density of the green compacts can be elevated beyond 50 %. Subsequently, vacuum sintering at 1700 °C for 5 h resulted in high-density ceramics with an in-line transmittance of approximately 73 % at 1100 nm. Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant No. 51802136), Foundation of Liaoning Educational Committee (Grant No. L2020043). 5 REFERENCES 1 H. M. Oh, Y. J. Park, H. N. Kim, J. W. Ko, H. K. Lee, Effect of mill- ing ball size on the densification and optical properties of transparent Y2O3 ceramics, Ceram. Int., 47 (2021) 4, 4681–4687, doi:10.1016/ j.ceramint.2020.10.035 2 D. Permin, O. Postnikova, S. Balabanov, A. Belyaev, V. Koshkin, O. Timofeev, J. Li, Influence of SHS precursor composition on the properties of yttria powders and optical ceramics, Materials., 16 (2023) 1, 260, doi:10.3390/ma16010260 3 A. Ratsimba, A. Zerrouki, N. Tessier-Doyen, B. Nait-Ali, D. André, P. Duport, A. Neveu, N. Tripathi, F. Francqu, G. Delaizir, Densification behaviour and three-dimensional printing of Y2O3 ce- ramic powder by selective laser sintering, Ceram. Int., 47 (2021)6, 7465–7474, doi:10.1016/j.ceramint.2020.11.087 4 L. Gan, Y. J. Park, M. J. Park, H. Kim, J. M. Kim, J. W. Ko, J. W. Lee, Facile fabrication of highly transparent yttria ceramics with fine microstructures by a hot-pressing method, J. Am. Ceram. Soc., 98 (2015) 7, 2002–2004, doi:10.1111/jace.13648 J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420 419 Figure 9: SEM micrographs of the fracture surfaces of the green bodies obtained from the suspension with various solids loading: a) 25 /%, b) 29 /%, c) 31 /% Figure 10: In-linear transmittance of 1.2-mm-thick Y 2 O 3 ceramic. In- set showing an image of polished Y 2 O 3 ceramics 5 X. R. Zhang, W. Z. Lu, G. F. Fan, X. H. Wang, Fabrication of well-dispersed Y2O3 nano-powders by poly (acrylic acid) low-tem- perature combustion, Adv. Powder Technol., 27 (2016) 1, 295–298, doi:10.1016/j.apt.2015.11.004 6 F. F. Lange, Powder processing science and technology for increased reliability, J. Am. Ceram. Soc., 72 (1989) 1, 3–15, doi:10.1111/ j.1151-2916.1989.tb05945.x 7 L. L. Zhu, Y. J. Park, L. Gan, S. I. Go, H. N. Kim, J. M. Kim, J. W. Ko, Effects of the Zr concentration on transparent Y2O3 Ceramics fabricated by vacuum pre-sintering and a subsequent HIP treatment, J. Mater. Sci. Mater. Electron., 28 (2017), 7854–7861, doi:10.1007/ s10854-017-6482-9 8 J. A. Lewis, Colloidal processing of ceramics, J. Am. Ceram. Soc., 83 (2000) 10, 2341–2359, doi:10.1111/j.1151-2916.2000.tb01560.x 9 X. Y. Zhang, W. L. Huo, S. Yan, Y. G. Chen, K. Gan, J. J. Liu, J. L. Yang, Innovative application of PVA hydrogel for the forming of po- rous Si3N4 ceramics via freeze-thaw technique, Ceram. Int., 44 (2018) 11, 13409–13413, doi:10.1016/j.ceramint.2018.03.071 10 A. Shafeiey, M. H. Enayati, A. Al-Haji, The effect of slip casting pa- rameters on the green density of MgAl2O4 spinel, Ceram. Int., 43 (2017) 8, 6069–6074, doi:10.1016/j.ceramint.2017.01.151 11 F. Mohammadi, O. Mirzaee, M. Tajally, Influence of solids loading on the rheological, porosity distribution, optical and the micro- structural properties of YAG transparent ceramic, Ceram. Int., 44 (2018) 11, 12098–12105, doi:10.1016/j.ceramint.2018.03.230 12 V. M. Candelario, M. I. Nieto, F. Guiberteau, R. Moreno, A. L. Ortiz, Aqueous colloidal processing of SiC with Y3Al5O12 liquid-phase sintering additives, J. Eur. Ceram. Soc., 33 (2013) 10, 1685–1694, doi:10.1016/j.jeurceramsoc.2013.01.030 13 G. V. Franks, C. Tallon, A. R. Studart, M. L. Sesso, S. Leo, Colloidal processing: enabling complex shaped ceramics with unique multiscale structures, J. Am. Ceram. Soc., 100 (2017) 2, 458–490, doi:10.1111/jace.14705 14 I. Santacruz, M. J. Zayas-Rey, J. M. Porras-Vázquez, D. Marrero-López, E. R. Losilla, Colloidal processing and characterisa- tion of lanthanum tungstate sheets, La5.5WO11.25, prepared by tape casting and reaction sintering, Ceram. Int., 41 (2015)9 , 11334–11340, doi:10.1016/j.ceramint.2015.05.091 15 J. Mouzon, E. Glowacki, M. Oden, Comparison between slip-casting and uniaxial pressing for the fabrication of translucent yttria ceram- ics, J. Mater. Sci., 43 (2008) 2849–2856, doi:10.1007/s10853- 007-2261-y 16 Y. Sun, S. Z. Shimai, X. Peng, G. H. Zhoua, H. Kamiyac, S. W. Wang, Fabrication of transparent Y2O3 ceramics via aqueous gelcasting, Ceram. Int., 40 (2014) 6, 8841–8845, doi:10.1016/ j.ceramint.2014.01.106 17 Z. C. Fu, X. D, Li, Y. Ren, M. Zhang, X. T. Geng, Q. Zhu, J. G. Li, X. D. Sun, Coating Y2O3 nano-particles with ZrO2-additive via pre- cipitation method for colloidal processing of highly transparent Y2O3 ceramics, J. Eur. Ceram. Soc., 39 (2019) 15, 4996–5004, doi:10.1016/ j.jeurceramsoc.2019.07.011 18 G. B. Granger, C. Guizard, Sintering behavior and optical properties of yttria, J. Am. Ceram. Soc., 90 (2007) 9, 2698–2702, doi:10.1111/ j.1551-2916.2007.01759.x 19 L. L. Jin, X. J. Mao, S. W. Wang, M. J. Dong, Optimization of the rheological properties of yttria suspensions, Ceram. Int., 35 (2009)2, 925–927, doi:10.1016/j.ceramint.2008.03.009 20 J. He, X. D. Li, Q. Zhu, C. Ma, M. Zhang, J. G. Li, X. Sun, Disper- sion of nano-sized yttria powder using triammonium citrate disper- sant for the fabrication of transparent ceramics, Ceram. Int., 42 (2016) 8, 9737–9743, doi:10.1016/j.ceramint.2016.03.064 21 C. Duran, Y. Jia, Y. J. Hotta, K. Sato, K. Watari, Colloidal process- ing, surface characterization, and sintering of nano ZrO2 powders, J. Mater. Res., 20 (2005), 1348–1355, doi:10.1557/JMR.2005.0168 22 C. H. Chin, A. Muchta, C. H. Azhari, M. Razali, M. Aboras, Optimi- zation of pH and dispersant amount of Y-TZP suspension for colloi- dal stability. Ceram. Int., 41 (2015) 8, 9939-9946, doi:10.1016/j.ceramint.2015.04.073 23 J. Sun, L. Gao, Dispersing SiC powder and improving its rheological behavior, J. Eur. Ceram. Soc., 21 (2001) 13, 2447–2451, doi:10.1016/S0955-2219(01)00196-0 24 X. W. Zhu, T. Uchikoshi, T. S. Suzuki, Y. Sakka, Effect of Polyethylenimine on Hydrolysis and Dispersion Properties of Aque- ous Si3N4 Suspensions, J. Am. Ceram. Soc., 90 (2007) 3, 797–804, doi:10.1111/j.1551-2916.2007.01491.x 25 Y. Y. Xu, X. J. Mao, J. T. Fan, X. K. Li, M. H. Feng, B. X. Jiang, F. Lei, L. Zhang, Fabrication of transparent yttria ceramics by alcoholic slip-casting, Ceram. Int., 43 (2017) 12, 8839–8844, doi:10.1016/ j.ceramint.2017.04.017 26 D. Marani, B. R. Sudireddy, J. J. Bentzen, P. S.Jørgensen, R. Kiebach, Colloidal stabilization of cerium-gadolinium oxide (CGO) suspensions via rheology, J. Eur. Ceram. Soc., 35 (2015) 10, 2823–2832, doi:10.1016/j.jeurceramsoc.2015.03.043 27 Z. Q. Sun, X. W. Zhu, M. S. Li, Y. C. Zhou, Y. Sakka, Hydrolysis and dispersion properties of aqueous Y2Si2O7 suspensions, J. Am. Ceram. Soc., 92 (2009) 1, 54–61, doi:10.1111/j.1551-2916.2008. 02860.x 28 A. Goswami, K. Ankit, N. Balashanmugam, A.M. Umarji, G. Ma- dras, Optimization of rheological properties of photopolymerizable alumina suspensions for ceramic microstereolithography, Ceram. Int., 40 (2014) 2, 3655–3665, doi:10.1016/j.ceramint.2013.09.059 J. HE et al.: DISPERSION OF AQUEOUS Y2O3 NANO-SUSPENSION AND FABRICATION OF TRANSPARENT CERAMICS ... 420 Materiali in tehnologije / Materials and technology 58 (2024) 3, 413–420