Advanced Powder Technology 25 (2014) 154–162

Contents lists available at ScienceDirect

Advanced Powder Technology

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

Invited review paper

Processing defects in ceramic powders and powder compacts

http://dx.doi.org/10.1016/j.apt.2014.01.0090921-8831/� 2014 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.

⇑ Address: Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka,Niigata 940-2188, Japan. Tel.: +81 258350072.

E-mail addresses: uematsu@vos.nagaokaut.ac.jp, keizouematsusan@hotmail.co.jp

Keizo Uematsu ⇑Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan

a r t i c l e i n f o a b s t r a c t s

Article history:Received 31 August 2013Received in revised form 7 January 2014Accepted 11 January 2014Available online 28 January 2014

Keywords:PowderStructureAggregatesLarge particlesAnisotropyCeramicsCharacterization

Structure and defects in powders and powder compacts are examined in detail to determine their forma-tion mechanisms, and their relevance to the production of high quality ceramics. New characterizationtools that are indispensable in the characterization of defect structure are described with the aid of sche-matic illustrations. Defects such as aggregates and large particles are present in all powders examined,even after rigorous grinding. These short-range defects degrade the properties and the quality of a sin-tered ceramic. Long-range defects due to particle orientation and anisotropic particle packing are alsocommon in powder compacts, and are attributed to a shear stress field and/or a directional stress fieldduring forming. Particle packing anisotropy produces long-range defects that are responsible for aniso-tropic deformation and cracking during drying and sintering.� 2014 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder

Technology Japan. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1542. Liquid immersion method and related methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1553. Examples of structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

3.1. Aggregates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1563.2. Extremely large particle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

4. Particle orientation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1585. Anisotropic packing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1596. Particle packing and orientation structures and their significance in the production of ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1607. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

1. Introduction

To produce a ceramic, raw powder is formed into a compact of adesired shape and subsequently heated at a high temperature toform a dense cohesive body. The characteristics of the raw startingpowders as well as the processes used to shape the body affect theproperties and quality of the resultant ceramic. That is, there is acritical relationship between the properties of a sintered ceramic

and the structure of the powders and powder compacts fromwhich the ceramic is produced [1].

Schematic illustrations of possible structures in a powder orpowder compacts are summarized in Fig. 1. Powder packing struc-ture affects densification and microstructure development duringsintering that directly govern the properties of the ceramic. Whilemicrostructure–property relationships are very important in ceram-ics, our explicit understanding of powder and compact structure isvery limited and is often based on empirical data and/or intuition[1]. Consequently, traditional engineering ceramic developmentand optimization often involves a laborious and time consumingtrial-and error approach [2]. Explicit understanding and control onpacking structure and critical process–structure–property

Fig. 1. Representation of packing structures and defects in powder compact.

air 1.56 1.74 1.76 1.79

Fig. 2. Images of alumina powder compacts with different refractive index liquids.

K. Uematsu / Advanced Powder Technology 25 (2014) 154–162 155

relationships will significantly improve ceramic quality whiledecreasing development time.

The poor understanding of structure–property relations inceramics is mostly ascribed to a lack of adequate characterizationmethods to study particle packing structure. With the exceptionof more recently developed tomography, scanning electronmicroscopy (SEM) has historically been the primary tool used tocharacterize microscopic structure, while porosimetry is used tocharacterize macroscopic structure [3]. SEM is excellent to exam-ine packing structure in detail in a small region. However, the verysmall volume subjected to SEM examination severely limitd thedetection and characterization of the ‘‘unusual’’ or defect struc-tures responsible for reducing the quality of a ceramic. Likewise,the information acquired from porosimetry is often too generalto correlate a specific defect/structure with ceramic performanceor quality.

The simple fact is our knowledge of the local structural defectsthat govern the quality of a ceramic is very limited [1]. Themechanical strength of a ceramic is dictated by the largest defectthat behaves as the fracture origin [4]. Consequently, a clear under-standing of the source and location of large packing irregularities iskey to improving the strength of a ceramic. However, becausethese defects are extremely rare, their detection and characteriza-tion is almost impossible with conventional characterization toolsand methods.

Additionally, the overall structures of a powder compact hastraditionally been very difficult to characterize including structuralheterogeneities such as those due to particle orientation and aniso-tropic packing that may be responsible for anisotropic deformationand the formation of cracks upon heating [5].

The development of the liquid immersion method (LIM) hascreated a route to better understanding packing structure at vari-ous levels [6]. Using the LIM and the related unique methodsdeveloped thereafter, it is now possible to characterize a varietyof packing structures explicitly for the first time. This paper dis-cusses powder packing structure at various levels, including howprocess affects packing structure, and how packing structure influ-ences the mechanical properties of a ceramic and the troubles inproduction such as cracking and warping.

2. Liquid immersion method and related methods

The LIM was developed over 20 years ago and has been appliedin the fundamental research of ceramics ever since. Details of themethod have been documented elsewhere [3]. An outline of themethod is as follows.

A powder compact is made transparent by immersing it anappropriate index of refraction liquid to observe the internal struc-ture with a proper microscopy. Fig. 2 shows alumina powder com-pacts immersed in liquids with various refractive indices [8]. Thebest transparency is noted for the liquid with the same refractiveindex as alumina. The liquid transforms the alumina/air interfaceinto an alumina/liquid interface to minimize the reflection/refraction.

The thickness of the powder compact affects the quality of theimage significantly. Practically, a 0.2 mm thick specimen can beprepared by grinding a small piece of powder compact with a sandpaper. The sample can be held with the tip of finger and rubbed onthe sand paper. Alternatively, a jig with a successively shallowergroove can be used to hold the specimen. Heating at an adequatetemperature below the densification temperature gives the com-pact some strength, which can ease the preparation of a thinspecimen.

A variety of microscopes can be used to observe many kinds ofstructures [6–10]. Table 1 summarizes the structures that can beobserved. Characteristics including crystallographic propertiescan be examined with a polarized light microscope. The high trans-mittance of infrared light through powder compacts makes IRmicroscopy the best for the bulk observations in thick specimenas well as in specimens with a high refractive index for which anadequate immersion liquid is not available. Confocal fluorescentscanning laser microscopy (LFSLM) provides high resolution de-tailed structures. Structures and features that have successfullybeen identified with these tools include agglomerates, large parti-cles, packing heterogeneities, local particle orientation of particles,and the segregation of additives.

The structure of a ceramic after sintering also can be examinedwith a variety of transmission microscopy. These methods are

Table 1Microscopy used to examine green compact structure.

Type of microscope Features examined

Transmissionmicroscope

Pores, cracks, additives (binder, etc.), foreignmaterials

Polarized lightmicroscope

Large particles, particle orientation, aggregates

Infrared microscope Pores, cracks, additives (binder, etc.), foreignmaterials

CFSLM Large particle, aggregates, pores, cracks, additives,foreign materials

156 K. Uematsu / Advanced Powder Technology 25 (2014) 154–162

based on those commonly applied in optical minerallorogy [11].The key is to prepare a thin specimen by grinding, so that the bulkcan be examined using transmission microscopy [12]. The LIM isapplicable for a wide variety of ceramic materials, including alu-mina, zirconia, silicon nitrides, and barium titanate.

(a)

3. Examples of structures

3.1. Aggregates

Aggregates are present in essentially all commercial aluminapowders. Fig. 3 shows a few examples of aggregates separatedfrom the bulk powder (nominal particle size about 0.4 lm) bywet sieving [13]. They are hard and cannot be destroyed even withrigorous grinding. Their volume fraction in the powder is onlyabout 0.005, or 0.5 vol%. There are various types of structures.Some aggregates are composed of tightly bound large particles ofcharacteristic shape, while loosely-bound smaller, nearly sphericalparticles form other aggregate.

Fig. 4 shows CFSLM image of a compact prepared from a slurryof a commercial low-soda alumina powder [14]. The slurry was

Fig. 3. SEM micrograph of aggregates in a commercial alumina powder (averageparticle size under 1 lm).

passed through a bead mill up to 15 times. The aggregates in thecompacts appear as dark spots in the micrographs. The size andnumber is large in the compact prepared with the slurry thatwas not milled, i.e., the as-received powder. Aggregate size andnumber decrease markedly after 1 pass through the bead mill,but does not change significantly with additional milling up to15 passes through the mill. Clearly, the removal of aggregates fromthe slurry is very difficult. It is interesting to add, that the particlesize analysis with commercial equipment shows the slight changein average particle size with grinding, but does not detect thechange in the extremely large size range.

Fig. 5(a) shows LIM image of commercial granules made of finealumina powder [15]. Elongated near-rectangular shapes struc-tures about 15–20 lm long are present in essentially all granules.Their volume fraction is low, and is estimated to be under0.1 vol%. This makes their detection extremely difficult with com-mercial particle size analyzers. However, once the presence of thelarge abnormal structure was noted using LIM, SEM was applied tostudy the structure in detail. Fig. 5(b) shows a low magnificationSEM image of the structure taken after laborious observation onhundreds of polished surface of the granules embedded in resin.

(b)

(c)

Fig. 4. Micrograph showing aggregates (dark regions) in alumina powder madewith slurries that were bead milled: (a) 0 pass; (b) 1 pass; or (c) 15 passes.

(a)

(b)

K. Uematsu / Advanced Powder Technology 25 (2014) 154–162 157

The high magnification SEM image in Fig. 5(c) shows that thestructure is an aggregate consisting of very fine particles.

Fig. 6 shows polarized light micrographs of granules takenthrough a Nicol prism. The rectangular-shaped structure inFig. 6(a) changes the brightness with the rotation of specimenstage in Fig. 6(b) and (c), indicating it is optically anisotropic.Fig. 6(d) is a schematic of the structure of the granule, whichconsists of fine alumina particles with their c-axis oriented inone direction in the aggregate. The fine alumina powderparticles in the aggregate are randomly oriented except in theaggregate.

The structure of these aggregates is understandable consider-ing how the alumina powder granule is formed. According tothe manufacturer, the powder is produced through the thermaldecomposition of a mother-salt, followed by grinding. Topotacticreaction, which often occurs in the thermal decomposition pro-cess, result in a specific crystallographic relationship betweenthe mother salt and the fine alumina particle reaction product.Fine alumina particles produced from a large single crystal ofmother salt will have the same crystallographic direction as thesalt crystal.

Fig. 5. Aggregates in alumina granules revealed by: (a) LIM; (b) low magnificationSEM; and (c) high magnification SEM.

(c)

(d)

Fig. 6. Alumina granule structure: (a) polarized light microscopy; (b) and (c) underclosed Nicols; and (d) illustrated schematially.

3.2. Extremely large particle

Extremely large particles are also present in essentially all pow-ders. The total number is often extremely small, but a single largeparticle can have a very detrimental effect on the quality of powderand/or the ceramic produced from that powder. For example, a sin-gle large particle in a polishing media may reset all efforts in oper-ation by forming scratches on a polished surface. Likewise, a largeparticle can become the strength limiting flaw in a sintered cera-mic that markedly degrades performance and quality.

Grinding is supposed to remove these undesirable large parti-cles. However, the lack of an accurate evaluation method to detectthe rare large particles has made the selection of a proper grindingprocess very difficult. The optical imaging techniques introduced inthis paper can readily be applied to characterize large particles inceramic powder or powder compacts.

158 K. Uematsu / Advanced Powder Technology 25 (2014) 154–162

Fig. 7 shows a polarized light micrograph of polishing mediapowder showing the presence of coarse particles (bright spots). Be-cause the volume of coarse particles present in the micrograph isproportional to the third power of the representative length, thevolume fraction of coarse particle is estimated to be under theppm level. With conventional analysis, the coarse particles cannotbe detected.

Fig. 8. A SEM micrograph showing the characteristic shape of the particles in atypical low-soda alumina powder.

4. Particle orientation

A shear stress field is present in almost all of the forming pro-cesses used to form a ceramics, which tends to orient elongatedparticles in one direction. Particle orientation and its significancein processing is well recognized in traditional ceramics such as claybodies comprised of platelet-shape particles [15]. However, parti-cle orientation and the effects on ceramic microstructure are lessrecognized in advanced engineering ceramics where the raw pow-ders have nearly spherical shapes. Considering the problems ofdeformation that can be encountered in commercial production,particle orientation needs to be characterized and understood.Again the optical characterization methods described in this papercan be applied to evaluate particle orientation quantitatively [16].

Particle orientation is observed macroscopically even in powdercompacts made by die pressing, where the shear stress is quite lowrelative to other ceramic forming process [17,18]. Particle orienta-tion is readily apparent in slip cast alumina compacts, where SEMshows elongated particles aligning in the direction of their longestaxis parallel to the casting mold surface [19].

Fig. 8 shows a SEM micrograph of representative low-soda alu-mina powder. While the individual particles appear to have a near-spherical shape, careful examination reveals that particles areactually elongated with an aspect ratio of approximately 1.5. Thehexagonal alumina crystal structure is responsible for this aniso-tropic particle shape.

Fig. 9 shows LIM polarized light micrographs of an aluminapowder compact prepared by die pressing at low pressure, fol-lowed by isostatic pressing at high pressure. The side view imagetaken perpendicular to the direction of die pressing shows achange in brightness with the rotation of the specimen [6]. Thisanisotropic optical behavior is similar to that observed in aluminacrystal observed in Fig. 6. By contrast, no change in brightness isnoted in the top view image taken from the direction of die press-ing, indicating optical isotropy.

A schematic of the structure of the compact viewed from thetop and side is shown in Fig. 10. This is the structure expected fromthe stress field present during die pressing.

The degree of particle orientation was determined through themeasurement of retardation of polarized light [16]. Fig. 11 showsthe effect of pressing pressure on particle orientation for uniaxially

Fig. 7. Polarized light micrograph of alumina grinding powder.

pressed alumina compact with and without subsequent isostaticpressing [18]. The particle orientation increases with increasingpressure, and surprisingly, the effect is even enhanced after postisostatic pressing.

Particle orientation is noted also in the powder granules used toform ceramics by powder compaction. Fig. 12 shows LIM-polarizedlight micrograph of granules prepared with dispersed and floccu-lated slurry. A dark cross spreading from the center of granule isobserved in Fig. 12 of the granule prepared with a well dispersedslurry. The direction of the cross does not change even with therotation of the sample stage. No specific pattern is noted in thegranules prepared from a flocculated slurry, and the image ismostly monochromatic except for the small bright spots due tothe presence of large alumina particles.

For comparison, Fig. 13 shows a schematic illustration of thestructure of granules in Fig. 12. The development of these struc-tures is explained as follows. In a droplet of well dispersed slurry,the particles in the slurry are free to move, at least for a short per-iod of time immediately after spraying. With the evaporation ofwater at the surface of droplet, the particles start to rearrangethemselves for stable packing near the surface. Particles of elon-gated shape align with their longest axis parallel to the surface inthis stage. As the evaporation of water proceeds, the internal par-ticles in the slurry droplet are carried to the surface, and form sub-sequent layers of aligned alumina particles inward to establish thefinal granule structure. The formation of dimple in the latter stageof drying complicates the structure, but viewed from the dimpleside, the structure in Fig. 13(a) is apparent. The image shown inFig. 12 is consistent to this structure. By contrast, the weaklybound particle in a droplet of flocculated slurry cannot movefreely. Consequently, the random orientation of the particles inthe flocculated slurry is preserved in the granules after drying.The image shown in Fig. 12 is again consistent to this structure.

Particle orientation can be examined in more detail at micro-scopic level [19]. Fig. 14 shows a method that can be used to deter-mine the orientation of each particle in a SEM micrograph. Theangle between the mold surface and the longest axis of each parti-cle in a slip cast alumina powder compact was measured from SEMmicrographs taken parallel to the mold surface in the x and y direc-tions. Measurements in z direction were also taken perpendicularto the mold surface using the angle between an arbitrary linedrawn in a fixed direction and the longest axis of particle.

Fig. 15 shows the result of analysis. A wide distribution of angleis noted, which is represented well by the March–Dollase distribu-tion [20]. The largest fraction of particles are oriented in the x and ydirections parallel to the mold surface. There is no specific direc-tion of particle orientation in the z direction parallel to the moldsurface. The fitting with the above function is only for illustration.

Top view Side view

Fig. 9. Polarized light micrographs of a die pressed alumina powder compact taken from the top, pressing direction, and from the side, perpendicular to the pressingdirection.

Fig. 10. A schematic of the granule structure and particle alignment in a die pressed powder compact viewed from the top, pressing direction and from the side,perpendicular to the pressing direction.

Fig. 11. Effect of uniaxial pressure on the orientation degree in die pressed aluminapowder compact.

K. Uematsu / Advanced Powder Technology 25 (2014) 154–162 159

This distribution is empirical and often noted in a variety ofsystems.

5. Anisotropic packing

Anisotropic particle packing has long been recognized qualita-tively, and is often believed to be the reason for anisotropic sinter-ing behavior. LIM-confocal scanning fluorescent light microscopyhas been applied to analyze anisotropic structure in detail [21].

Fig. 16 shows LIM-CSFLM micrographs taken parallel to andnormal to the pressing direction of a die pressed compact preparedwith mono-sized and nearly spherical alumina particles. To com-plete a detailed analysis of particle packing in the compact, firstthe contrast of the macrograph was enhanced. Next, lines weredrawn to connect the center of each neighbor pair. Then, the ratiosof the line length drawn parallel and normal to the die pressingdirection were measured to determine the average value for 800particle pairs. The average value should be unity if the particlepacking is random.

The values determined for the cross sections parallel and nor-mal to the pressing directions are 1.15 and 1.0, respectively, whichis consistent to what is expected in a die pressed powder compact.

(a)

(b)

Fig. 12. Polarized light micrograph of alumina powder granules prepared from a:(a) dispersed; and (b) flocculated slurry.

(a)

(b)

Fig. 13. Schematic for powder packing structures in granules prepared with a: (a)dispersed slurry; and (b) flocculated slurry.

Fig. 14. A schematic showing how particle orientation is determined from the angleformed by a line through the longest axis of a particle relative to the mold surface.

Fig. 15. The distribution of particle orientation angles in the x (large squares) and y(small squares) direction measured between the mold surface and the longest axisof the particle in a slip cast alumina compact, and in the z direction (mediumsquares).

160 K. Uematsu / Advanced Powder Technology 25 (2014) 154–162

The forces applied to the particles during die pressing are largerin the pressing direction, and thus the particles pack to a higherdensity in the direction parallel to the pressing direction. In thedirection normal to the pressing direction the packing density is

lower. The dimensional changes measured during the subsequentisotropic pressing are consistent with this description. The com-pact tends to shrink little in the direction parallel to the directionof die pressing that is already compacted to a high density, whilelarge shrinkage is observed in the direction normal to the directionof die pressing, indicating the presence of more empty space be-tween the particles in this direction.

6. Particle packing and orientation structures and theirsignificance in the production of ceramics

This paper provides some new insight on the particle packingand orientation structure in powder and compact. Although someof the structures are shown explicitly for the first time, they havebeen often imagined for some time as shown in Fig. 1, and arenot surprising. Nevertheless, the accurate understanding of particleorientation and structure can markedly accelerate advances inceramics.

The powder structures in Fig. 1 are divided into two categories,short range structure and long range structure. Short range struc-ture is commonly under 1 mm, and include large pores, agglomer-ate and extremely large particles. Long range structure can extendthe length of the compact, and include particle orientation, large

Fig. 16. Confocal scanning fluorescent light micrograph of a die pressed powdercompact showing particle orientation and packing in the direction: (a) normal to;and (b) parallel to the pressing direction.

Fig. 17. Structures of alumina compact and ceramics (upper right) made by diepressing.

Fig. 18. Measured size and distribution of flaws in a sintered alumina ceramicpowder compact determined by observation of the internal structure.

1

2

5

10

20

50

90

99

ailu

re p

roba

bilit

y / %

Measured strengthN = 25

ave = 428 MPam = 12.4

Predicted strengthN = 500

K. Uematsu / Advanced Powder Technology 25 (2014) 154–162 161

scale density distribution, particle size segregation and anisotropicpacking.

Short- and long-range structure are compared and contrasted inTable 2. Short range structure is responsible for the formation ofdefects that govern the properties of a ceramics. Fig. 17 shows anexample of the typical short-range structures in powder compactand in a ceramics after sintering. The two micrographs are clearlysimilar, and show that the origin of defect in the sintered ceramic isthe original defect in the powder compact. In this specific example,large pores create large pores in the sintered ceramics [3].

Fig. 18 shows the size and distribution of flaws in a ceramicsdetermined by observing internal structure with an optical micro-scope [22]. It is interesting to note that a linear relationship is pres-ent in the log–log plot. Fig. 19 shows the measured compared tothe fracture mechanics-based simulated strength distribution ofthe sintered alumina ceramic determined using the defect sizedistribution in Fig. 18. Excellent agreement is noted between themeasured and simulated strengths. Clearly, a compact with homo-geneous packing structure of particles is very important to producehigh performance and quality ceramics.

Table 2Grouping of structures in powder compacts, their characteristics and significance onceramic production.

Structures Size Problems Occurrence

Short Small Defect AccidentalLong Large Deformation Systematic

200 300 400 500 600

F

Flexural strength / MPa

0.5

0.2

ave = 441 MPam = 12.7

Fig. 19. Measured alumina ceramic strength compared to the strength calculatedusing the defect size distribution in Fig. 18.

Sint

erin

g sh

rink

age

[%]

Temperature [ C]

1000 1200 1400 1600

0

10

5

20

15

10K/min

Elongated Particles

Sintering: 2h

o

Fig. 20. Sintering shrinkage in a die pressed alumina powder compact in thedirections parallel to and normal to the pressing direction.

162 K. Uematsu / Advanced Powder Technology 25 (2014) 154–162

Long-range structures can result in large deformation andcracking of products, troubles probably the most important prob-lem in the commercial production of ceramics. Fig. 20 shows thesintering shrinkage in an alumina powder compact formed by diepressing [17]. There is a clear difference in shrinkage at all sinter-ing temperatures and times between the directions parallel andnormal to the pressing direction. Recalling the structure shownin Fig. 10, the particle orientation in the die pressed powder com-pact is clearly responsible to the deformation during sintering.Local difference of deformation in a powder compact can create alarge stress during sintering, and may develop cracks in theceramic products. Control of long range structure is crucial in thesuccessful production of ceramics.

7. Conclusion

A powder compact can have various types of packing structureas summarized in Fig. 1. The particle packing structure depends onthe characteristics of the raw powder and how it is processed.Short and long range defect structures are possible, and their rele-vance to the processing and properties of a ceramics is very differ-ent and must be carefully distinguished in solving problemsencountered in the processing of ceramics. For example, SEMobservation of local structure provides little helps in solving defor-mation and cracking in production, which are governed by thelong-range structure. The structures discussed are mostly consis-tent to the imagination. Recalling our poor characterization toolsto examine the structure, sound imagination is still very useful inthe research and development of ceramics, and must be kept train-ing. Always bear in mind that the absence of clear cause does notmean true absence. Judgment depends on the capability of toolused for analysis. Likewise, the measurement of average particle

size or average density provides little value in determining theproperties and quality of a ceramic with larger pores or grains. Thispaper presents a suite of characterization techniques that can beapplied to quantitatively characterize short and long-range defectstructures in ceramics to better understand processing–micro-structure–property relations.

References

[1] F.F. Lange, Powder processing science and technology for increased reliability,Journal of American Ceramic Society 72 (1) (1989) 3–15.

[2] W.D. Kingery, Firing – the proof testing for ceramic processing, in: G.Y. Onoda,L.L. Hench (Eds.), Ceramic Processing Before Firing, Wiley, New York, 1978, pp.291–305.

[3] K. Uematsu, Process defects, in: R. Riedel, I.-Wei Chen (Eds.), Ceramic Scienceand Technology, Wiley, New York, 2012, pp. 369–394.

[4] A.A. Griffith, Phenomena of rupture and flaw in solids, Phil. Trans. Royal Soc.Lond. A221 (1920) 163.

[5] K.G. Ewsuk, J.G. Arguello, D.H. Zeuch, B. Farber, L. Carinci, J. Kaniuk, J. Keller, C.Cloutier, B. Gold, R.B. Cass, J.K. French, B. Dinger, W. Blumenthal, CRADAdevelops model for powder pressing and die design: Part two, Am. Ceram. Soc.Bull. 80 (2) (2001) 41–46.

[6] Keizo. Uematsu, Immersion microscopy for detailed characterization of defectsin ceramic powders and green bodies, Powder Technol. 88 (1996) 291–298.

[7] K. Uematsu, J.Y. Kim, Z. Kato, N. Uchida, K. Saito, Direct observation method forgreen body of ceramics – with alumina body as an example, J. Ceram. Soc. Jpn.98 (5) (1990) 515–516.

[8] K. Uematsu, M. Saito, Liquid immersion technique coupled with IR-microscopyfor direct observation of internal structure of ceramic powder compact – withalumina as an example, J. Mater. Res. 14 (12) (1999) 4463–4465.

[9] Z. Kato, T. Sato, S. Tanaka, N. Uchida, K. Uematsu, Mid-Infrared Microscopy,Am. Ceram. Soc. Bull. 81 (12) (2002) 42–44.

[10] Y. Saito, S. Tanaka, N. Uchida, K. Uematsu, CSLM for ceramic greenmicrostructure, Am. Ceram. Soc. Bull. 81 (1) (2002) 35–38.

[11] D. Shelly, Optical Mineralogy, 2nd ed., Elsevier, 1985.[12] JIS R1642, Testing method for macro-heterogeneity in microstructure of fine

ceramics.[13] Y.-I. Cho, S. Katori, H. Emoto, T. Yamada, T. Takahashi, M. Okumiya, C. Kumagai,

K. Uematsu, Characterization of coarse particles in alumina powders by a wetsieving method, J. Euro. Ceram. Soc. 23 (2003) 1661–1666.

[14] S. Tanaka, Z. Kato, N. Uchida, K. Uematsu, Grinding behavior of aluminapowder, Am. Ceram. Soc. Bull. 82 (8) (2003) 9301–9303. <http://www.ceramicbulletin.org>.

[15] G.C. Robinson, Extrusion defects, in: G.Y. Onoda Jr., L.L. Hench (Eds.), CeramicProcessing Before Firing, Wiley, New York, 1978, pp. 391–407.

[16] A. Makiya, S. Tanaka, D. Shouji, T. Ishikawa, N. Uchida, K. Uematsu, Aquantitative evaluation method for particle orientation structure in aluminapowder compacts, J. Euro. Ceram. Soc. 27 (2007) 3399–3406.

[17] A. Shui, Zenji Kato, Satoshi Tanaka, Nozomu Uchida, Keizo Uematsu, Sinteringdeformation caused by particle orientation in uniaxially and isostaticallypressed alumina compacts, J. Euro. Ceram. Soc. 22 (3) (2002) 311–316.

[18] Anze Shui, Nozomu Uchida, Keizo Uematsu, Origin of shrinkage anisotropyduring sintering of uniaxially pressed alumina compacts, Powder Technol. 127(1) (2002) 9–18.

[19] S. Tanaka, A. Makiya, S. Watanabe, Z. Kato, N. Uchida, K. Uematsu, Particleorientation distribution in alumina compact body prepared by the slip castingmethod, J. Ceram. Soc. Jpn. 112 (5) (2004) 276–279.

[20] E. Zolotoyabko, Fast quantitative analysis of strong uniaxial texture using aMarch–Dollase approach, J. Appl. Cryst. 46 (6) (2013) 1877–1879.

[21] S. Tanaka, Y. Kuwano, K. Uematsu, Packing structure of particles in greencompact and its influence on sintering deformation, J. Am. Ceram. Soc. 90 (11)(2007) 3717–3719.

[22] S. Nakamura, S. Tanaka, Z. Kato, K. Uematsu, Strength-processing defectsrelationship based on micrographic analysis and fracture mechanics inalumina ceramics, J. Am. Ceram. Soc. 92 (3) (2009) 688–693.