Hydrolysis-induced aqueous gelcasting of β-SiAlON-SiO2 ceramic composites: The effect of AlN additive

Article abstract of DOI:10.1111/j.1551-2916.2010.03885.x

Dense β-SiAlON-SiO₂ (SiO₂520, 40, 50, 60, and 80 wt%)ceramic composites have been prepared from β-Si₄Al ₂O₂N₆ and fused silica by sintering at 15001-1750°C for 3-4 h. For comparison purposes, a powder mixture consisting 60 wt% β-Si₄Al₂O₂N₆ and 40 wt% fused silica has been consolidated following a new near-net shape technique based on hydrolysisinduced aqueous gelcasting (GCHAS) and sintered for 3 h at 1750°C. In the GCHAS process, consolidation of suspensions containing 50 vol% solids was achieved by adding a polymerization initiator, a catalyst, and AlN powder equivalent to 1-5 wt%Al₂O₃. Thin-wall radomes consolidated by GCHAS (using AlN equivalent to 5 wt% Al₂O₃ in the suspension) have exhibited green strengths 420 MPa. The sintered materials were characterized for various properties including hardness, fracture toughness, mass loss, shrinkage, coefficient of thermal expansion, and dielectric constant. The Si2N2O formed from a powder mixture of 60 wt% β-Si₄Al₂O₂N₆ and 40 wt% fused silica at 1750°C for 3 h exhibited a flexural strength of B140 MPa, Young's modulus of 214 GPa, coefficient of thermal expansion of 3.5×10 _-6°C_-1, hardness of 1390 kg/mm2, fracture toughness of 4.2MPa .m1/2, and a dielectric constant of 5.896 and tan δ of 0.002 at 17 GHz.

Full text of DOI:10.1111/j.1551-2916.2010.03885.x

J. Am. Ceram. Soc., 93 [10] 3180–3189 (2010)
DOI: 10.1111/j.1551-2916.2010.03885.x
r 2010 The American Ceramic Society
ournal
J
Hydrolysis-Induced Aqueous Gelcasting of b-SiAlON–SiO2 Ceramic
Composites: The Effect of AlN Additive
Ã
Ibram Ganeshw and Govindan Sundararajan
Centre for Advanced Ceramics, International Advanced Research Centre for Powder Metallurgy and New Materials
(ARCI), Hyderabad 500 005, AP, India
Dense b-SiAlON–SiO2 (SiO2 5 20, 40, 50, 60, and 80 wt%)
ceramic composites have been prepared from b-Si4Al2O2N6 and
fused silica by sintering at 15001–17501C for 3–4 h. For com-
parison purposes, a powder mixture consisting 60 wt%
b-Si4Al2O2N6 and 40 wt% fused silica has been consolidated
following a new near-net shape technique based on hydrolysis-
induced aqueous gelcasting (GCHAS) and sintered for 3 h at
17501C. In the GCHAS process, consolidation of suspensions
containing 50 vol% solids was achieved by adding a polymer-
ization initiator, a catalyst, and AlN powder equivalent to 1–5
wt% Al2O3. Thin-wall radomes consolidated by GCHAS (using
AlN equivalent to 5 wt% Al2O3 in the suspension) have exhib-
ited green strengths 420 MPa. The sintered materials were
characterized for various properties including hardness, fracture
toughness, mass loss, shrinkage, coefficient of thermal expan-
sion, and dielectric constant. The Si2N2O formed from a powder
mixture of 60 wt% b-Si4Al2O2N6 and 40 wt% fused silica at
17501C for 3 h exhibited a flexural strength of B140 MPa,
Young’s modulus of 214 GPa, coefficient of thermal expansion
of 3.5 ꢀ 10ꢁ6 1Cꢁ1, hardness of 1390 kg/mm2, fracture tough-
Al2O2N4-based radomes have been successfully test fired for
certain standard missile applications.4,5 Despite their several use-
ful properties, there is no straight forward fabrication route for
dense components of b-SiAlON–SiO2 ceramic composites.2 This
is due to the fact that b-Si4Al2O2N4 with 49 wt% SiO2 cannot be
prepared following a conventional reaction sintering of the pre-
cursor mixture consisting Si3N4, AlN, and Al2O3 powders, be-
cause they require an additional SiO2 as one of the precursor
materials.4,5 As SiO2 concentration increases in the precursor
mixture, the diffusion paths between reactive Si41 and Al31 spe-
cies also increase, which restricts the formation of the desired
b-SiAlON phase during sintering. The presence of the required
b-SiAlON phase in these composites is very important to have
the desired flexural strength and fracture toughness properties.14
In a study, Gilde et al.,1 have prepared a silicon oxy-nitride
(SiON) nanocomposite with a dielectric constant of 4.78 at 251C
and 5.0 at 10001C, and a loss tangent of 0.0014 at 251C by
calcining a-Si3N4 powder in an open-air atmosphere at 17001C
followed by fine grinding, recompaction, and resintering at
418001C in N2 atmosphere. However, this process does not yield
the final product with consistent properties as its final chemical
composition does not depend on the starting raw material com-
position; instead, it depends on the extent of oxidation that
occurred upon calcination of a-Si3N4 powder in the air atmo-
sphere. The oxidation of this powder is very sensitive to several
processing parameters such as atmospheric humidity, the degree
of agglomeration of powder particles, temperature fluctuations in
furnace, etc. Despite these problems, in a recent study, Ganesh
et al.,2 have synthesized a ceramic composite with b-Si4Al2
O2N619.0 wt% SiO2 following a simple reaction sintering (for
4 h at 17501C) of a precursor mixture, Si3N4/Al2O3 (51.5) using
7 wt% Y2O3 as a sintering aid (i.e., 1.5 Si3 N41Al2O3-
b-Si4Al2O2N610.5 SiO2). However, none of the combinations
of Si3N4 and Al2O3 yield b-Si4Al2O2N6 with 49 wt% SiO2. Of
late, Advanced Materials Organization Inc., New York has
started producing 25 cm height and 17.8 cm base diameter ra-
domes with a chemical composition of 70% b-SiAlON130%
fused silica for vehicles that travel at a speed of 4Mach 5. This
ceramic composite exhibits a dielectric constant of 4.9 at 8.6
GHz, a flexural strength of 532 MPa, and a coefficient of thermal
expansion (CTE) of 2.1 ꢀ 10ꢁ6 1Cꢁ1.3 However, this process in-
volves two very complicated, cumbersome, and expensive steps.
In this process, initially, a green radome is produced by the gel-
casting of a nonaqueous suspension containing precursor mixture
plus 30% organic fugitive material, followed by drying and sinte-
ring of the green radome for 2–4 h at 417501C under an N2
pressure of 50–100 bar to form b-SiAlON with a porosity of
30%. Then, this porosity is filled with fused silica following a
plasma spraying technique. Fused silica is added not only to re-
duce the dielectric constant from 7.3 to 4.2 but also to reduce the
density of the radome from 3.2 to 2.2 g/cm3, which is required
for these high-speed radomes.3 Nevertheless, introducing 430%
porosity in the gelcast sintered b-SiAlON radomes by utilizing
organic fugitive material is not an easy step particularly when the
product yield is aimed at a high rate.
ness of 4.2 MPa m1/2, and a dielectric constant of 5.896 and tan
.
d of 0.002 at 17 GHz.
I. Introduction
ECENTLY, b-SiAlON–SiO2 ceramic composites have re-
ceived a great deal of attention as materials of choice for
R
certain high-speed (4Mach 5) radome applications.1–3 Nor-
mally, high-speed radomes experience a transient temperature
of about 13701C together with high mechanical and aero-ther-
mal loads in service. To withstand these severe environments
while transmitting electromagnetic RADAR signals without any
disturbance, the materials used for radome construction should
have a low dielectric constant, a low loss tangent, a high flexural
strength, high thermal shock resistance, high rain erosion, and
particle impact resistance together with a high elastic modulus to
keep the thin walls of the radome from buckling.4,5 Although
several materials such as fused silica,6 Pyroceram 9606,4
Rayceram 8,4 barium aluminum silicate,7 aluminum phosphate
(commercially known as Cerablakt and manufactured by Ap-
plied Thin Films Inc., Evanston, IL),8,9 SiO2–AlN ceramics,10
SiO2–BN ceramics,11 reaction-bonded silicon nitride,12 liquid-
phase-sintered silicon nitride,13 silicon nitride nanocomposite,1
b-Si4Al2O2N4,4,5 etc., have been investigated for these applica-
tions, only the b-SiAlON–SiO2-based ceramic composites have
received much importance as these latter composites possess
the desired flexural strength and dielectric properties and can
withstand high temperatures (413001C).3 Recently, b-Si4
W.-C. Wei—contributing editor
Manuscript No. 27094. Received November 12, 2009; approved April 19, 2010.
Member, The American Ceramic Society.
Ã
In view of the above, in this investigation, a systematic
study was undertaken to prepare dense b-SiAlON–SiO2 ceramic
wAuthor to whom correspondence should be addressed. e-mail: ibram_ganesh@
yahoo.com
3180

October 2010
Hydrolysis-Induced Aqueous Gelcasting of b-SiAlON–SiO2 Ceramic Composites
3181
composites with a wide chemical composition following a con-
ventional powder-processing route. Initially, a b-Si4Al2O2N6
powder was prepared by the extrusion of the precursor mixture
paste followed by drying, reaction sintering, and fine grind-
ing.15–20 The fine-ground b-Si4Al2O2N6 powder was mixed
with different amounts of fused silica (0, 20, 40, 50, 60, 80,
and 100 wt%) and then die-pressed and sintered for 3–4 h at
15001–17501C.17 For comparison purposes, a powder mixture
consisting 60 wt% b-Si2Al2O2N6, and 40 wt% fused silica was
also consolidated following an hydrolysis-induced aqueous gel-
casting (GCHAS) process.18,20–23 In the GCHAS process, sus-
pensions with 50 vol% solids were utilized. To accomplish
consolidation, a polymerization initiator, a catalyst, and AlN
powder equivalent to 1–5 wt% Al2O3 on the powder weight
basis have been introduced into the particulate suspensions. The
fully dried green consolidates were then sintered for 3 h at
17501C in an N2 atmosphere to achieve the desired densification.
All the sintered materials were thoroughly characterized in order
to understand the effects of added fused silica and AlN into
particulate suspension on the physical, thermal, mechanical, and
dielectric properties of b-Si4Al2O2N6 and of sintered ceramic
composites consolidated by the GCHAS route.
(a)
10 cm
(b)
II. Experimental Section
(1) Synthesis of b-Si4Al2O2N6 Powder
The starting raw materials used in this study were a-Si3N4
(SicoNide-P95H, VESTA Ceramics AB, Sweden, BET surface
area—6.51 m2/g and average particle/agglomerate size, APS—
17.77 mm), a-Al2O3 (HP Grade, ACC India Limited, Thane,
India, BET SA—1.24 m2/g and APS—1.39 mm), AlN (Grade
AT, H.C. Stark, Goslar, Germany, BET SA—6.72 m2/g and
APS—7.62 mm), Y2O3 (Rhodia Inc., Phoenix, AZ, BET SA—
3.41 m2/g and APS—3.23 mm), and fused silica (Chettinad
Quartz Products Private Ltd., Chennai, India, APS—19.95
mm). Before use, the as-purchased AlN powder was treated in
an ethanol solution of ortho-phosphoric acid and aluminum
dihydrogen phosphate at 801C for 24 h in order to passivate it
against hydrolysis.16 It is known that AlN decomposes into am-
monia and aluminum hydroxide when it comes into contact with
water (AlN1 3H2O-Al(OH)31NH3).24,25 If AlN is utilized as
one of the starting precursor materials without passivating it
against hydrolysis in the processing of SiAlON following an
aqueous colloidal processing route, this AlN will be decomposed
in the suspension. This decomposition causes coagulation of the
suspension in the vessel used for powder deagglomeration pur-
pose, setting of the suspension within the vessel, and the overall
change in the targeted chemical composition.26,27
120 mm
In a typical experiment, a powder mixture of 64.33% a-Si3N4,
23.36% a-Al2O3, 9.37% AlN (surface passivated against hydro-
lysis),16 and 7% Y2O3 was kneaded in a conventional dough-
making machine (Sigma Kneader, Prigmayers India Private
Limited, Mumbai, India) with 3 wt% methyl cellulose, 2 wt%
polyethylene glycol (PEG 400) (both are GR grade procured
from Loba Chemie, Mumbai, India), and about 30 wt% double-
distilled water to obtain an easily extrudable dough.15 The
resultant dough was then extruded through an indigenously
designed and fabricated stainless steel die using a ram-type ex-
truder at a rate of 50 mm/min under a pressure of 10–20 kg/cm2
to form extrudates of about 5 mm thickness.15 These extrudates
were allowed to dry in an open-air atmosphere over night under
ambient conditions, which were then cut into 5–6 mm length
pieces. These cut pieces were then subjected to binder removal
for 2 h at 5001C and then sintered for 4 h at 16751C under an N2
atmosphere of about 800 torr to form b-Si4Al2O2N6 extrudates
(Fig. 1(a)).17–19 During sintering, these extrudates were covered
with a powder mixture of 50% Si3N4150% BN to protect them
against decomposition.17 The sintered extrudates were crushed
in an hammer mill followed by wet grinding in a stainless steel
jar (350 mm diameter and 400 mm height) using stainless steel
balls of about 12 mm as well as 25 mm diameters by maintaining
Fig. 1. (a) Extrudates of b-Si4Al2O2N6 formed by the reaction sintering
of a precursor mixture consisting a-Si3N4, a-Al2O3, surface passivated
AlN against hydrolysis, and Y2O3 (7 wt%) at 16751C for 4 h. (b) A
digital photograph of the green radome shape consolidated by hydro-
lysis-induced aqueous gelcasting of a suspension containing 50 vol%
solids loading and AlN equivalent to 5 wt% Al2O3 on the total powder
basis.
the 1:8 weight ratio of charge to balls. This milling operation
was continued till the average particle size of b-Si4Al2O2N6
reached about 3 mm.
(2) Synthesis of b-Si4Al2O2N6–SiO2 Ceramic Composites
A conventional dry-powder pressing technique was used to
make these ceramic composites. Before pressing into pellets of

3182
Journal of the American Ceramic Society—Ganesh and Sundararajan
Vol. 93, No. 10
30 or 90 mm diameterꢀ B10 mm height under a pressure of
200 MPa, B200 g of powder mixture containing the requisite
amounts of b-Si4Al2O2N6 and fused silica, B60 g of 5 wt%
aqueous PVA solution and B200 g of ZrO2 cylindrical pebbles
(10 mm diameter and 12 mm length) were suspended in
B200 mL of toluene in a 500 mL alumina bowl and were
ground for 30 min in a planetary ball mill (Retsch GmbH,
Haan, Germany). The resultant dough was separated from the
toluene solution, dried at B851C for 12 h in an electrically
heated hot-air oven. This dried mass was passed through a ꢁ30
to 1100 BSS mesh to form granules before pressing into pellets.
Binder burnout was operated on all die-pressed samples for 2 h
at 5001C before sintering for 3–4 h at 15001–17501C in nitrogen
atmosphere (4900 torr) by covering them in a powder bed of
50 wt% Si3N4150 wt% BN.17–19
suspension from top of the mold along with air/gas bubbles
(if any leftover after vacuum pumping) while inserting the man-
drel into the suspension in the mold. Always, some rectangular
(60 mm ꢀ 30 mm ꢀ 30 mm) bars were also cast along the ra-
dome shapes to check the condition of gel before removing the
mandrel from the radome mold.18 After confirming the gelation
of suspension, the mandrel was removed from the mold and
then was filled with polyethylene glycol-400 (PEG400) (AR
grade, Loba Chemie).26 Owing to high chemical potential,
water moves from gel (i.e., from wet radome) into the PEG.
After about 2 h, the PEG was removed from mold, and then the
wet-green radome was removed from the mold by simply tilting
the mold up-side down. This radome was then placed in a hu-
midity-controlled oven (Model: LHL-113; Espec Corporation,
Osaka, Japan) that was preset at 401C and 90% RH.22 After
drying for 24 h at 401C, the temperature in the humidity oven
was raised to 701C and then to 901C. At each of these temper-
atures, the radome was held for B24 h. One of the thus-dried
green radome shapes is shown in Fig. 1(b). All the GCHAS
green samples were then subjected to binder removal for 2 h at
5001C and then sintered for 3 h at 17501C under about 900 torr
N2 pressure.
(3) Colloidal Processing of b-SiAlON–SiO2 Ceramic
Composites
In a typical experiment, a powder mixture consisting 60 wt%
b-Si4Al2O2N6140 wt% fused silica was dispersed in an aqueous
solution of 8.6 wt% methacrylamide, 2.8 wt% methylenebisa-
crylamide, and 8.6% n-vinylpyrrolidinone, with the help of
Dolapix CE64, an amino alcohol-based cationic dispersant
(Zchimmer & Schwarz, Berlin, Germany) used at a ratio of
10 mL/g of powder mixture to achieve a suspension with
50 vol% solids loading.27 This aqueous solution has been termed
as NVPNCPM (n-vinylpyrrolidone nonconvectional premix)
solution.17–23 In order to optimize the amount of Dolapix CE
64 (dispersant) required to obtain suitable particulate suspen-
sions, initially a separate study was conducted aiming at low
viscous (o450 MPa ꢂ s) suspensions with solids loading as high
as 50 vol%. In this study, Dolapix CE 64 was used at a ratio of
5–35 mL/g of powder mixture. Only the suspension containing
the ratio of 10 mL/g of powder mixture could accommodate
the solids loading up to 50 vol% and exhibited a viscosity of
o450 MPa ꢂ s. It has been reported that suspensions with a vis-
cosity of o450 MPa ꢂ s and solids loading of 448 vol% are
required to fabricate defect-free thin-wall/complex-shaped
ceramic components following advanced net-shaping colloidal
routes such as either aqueous gelcasting or hydrolysis-induced
aqueous gelcasting.17–23 The minimum solids loading in the sus-
pensions required to fabricate radome shapes without any diffi-
culty was found to be 48 vol% in the present study. When this
solids loading was o48 vol%, the suspensions took more time
to get gelled and the gelled parts were always found to be
cracked as the mold mandrel could not allow the part to shrink
to get enough handling strength. In contrast, when the solids
loadings were 450 vol%, the viscosity of the suspensions was
increased to 4500 MPa ꢂ s. It was found to be quite difficult to
remove the entrapped air/gas bubbles fully from a suspension
having 4500 MPa ꢂ s viscosity.17–23,27 The suspensions contain-
ing 50 vol% powders composition and alumina balls (with a
charge to balls weight ratio of 1:3) in polypropylene bottles were
rolled on a roller mill for about 18 h to accomplish the powder
deagglomeration process. To the resultant suspensions, the as
purchased pure AlN (equivalent to 1–5 wt% of Al2O3 on total
powder weight basis) powder without any surface modification
was introduced and the deagglomeration process was continued
for a further 2 h.18,20–23 Thus, the obtained suspensions were
added with a polymerization initiator (10 wt% aqueous solution
of ammonium persulfate, APS) and a catalyst (as purchased
tetramethylethylenediammine, TEMED) at a ratio of 4 and 2 mL/g
of particulate suspension, respectively, and then degassed for
B2 min by vacuum pumping. Both APS and TEMED are of
GR grade procured from Loba Chemie. These degassed sus-
pensions were then cast into an indigenously designed and fab-
ricated split-type Teflon-lined aluminum radome-shape mold,
and inserted aluminum solid mandrel into the suspension. After
mandrel insertion, the suspension was allowed to completely gel
under ambient conditions in the mold.18–27 Always, some excess
suspension was poured in the mold so as to overflow this excess
(4) Material Characterization
A Gemini Micromeritics BET surface area analyzer (Model 2360,
Micromeritics, Norcross, GA) was used for specific surface area
measurements of the powders. Surface area was measured by ni-
trogen physisorption at liquid nitrogen temperature (ꢁ1961C) by
taking 0.162 nm2 as the area of cross section of N2 molecule.28
Particle size analysis of the powders was performed using a light
scattering equipment (Coulter LS 230, London, U.K., Fraunho-
fer optical model). The viscosity of suspensions was measured
using a rotational Rheometer (Bohlin C-VOR Instruments,
Worcestershire, U.K.).20,21 The measuring configuration adopted
was a cone and plate (41, 40 mm, and gap of 150 mm), and the
flow measurements were conducted between 0.1 and 800 sꢁ1
.
Before carrying out the viscosity measurements, bubbles were
removed by slow speed agitation from the suspensions.
The XRD patterns were recorded using a Bruker (Karlsruhe,
Germany) D8 advanced system with a diffracted beam mono-
chromated CuKa (0.15418 nm) radiation source.29 Crystalline
phases were identified by comparison with PDF-4 reference data
from the International Centre for Diffraction Data (ICDD).29
Bulk density, apparent porosity, and water absorption capacity
of sintered samples were measured according to the Archimedes
principle (ASTM C372) using a Mettler balance and the attach-
ment (AG 245, Mettler Toledo, Greifensee, Switzerland). On an
average, three density measurements were performed on each
sample in this study (70.01 SD).18 The microstructure of sintered
ceramic composites was examined using a scanning electron mi-
croscope (S-3400N, SEM, Hitachi, Tokyo, Japan). The flexural
strength of 60ꢀ 4 ꢀ 3 mm size samples was measured using the
four-point bend test (JIS-R1601).20,21 These samples were cut
from sintered rectangular bars of B90 mm length and B30 mm
width and B10 mm thickness. Six to eight samples were exam-
ined per case and all the readings were averaged out (73 SD).
The fracture toughness value (KIc) was calculated on the basis of
the indentation method [KIc 5 Ha1/2 ꢀ 0.203 (C/a)ꢁ3/2].30 Here,
2a represents Vickers indent diagonal length, 2C the resulted
crack length, and H is a Vickers hardness (HV 5kg/mm2 5 10
MPa). Five to six samples were examined per case; in order to
check the reproducibility of fracture toughness, results and all the
readings were averaged out (70.2 SD). The required hardness,
crack, and diagonal length data were collected using a micro-
hardness tester (Leitz Wetzler, Siegmund-Hiepe-Strasse, Ger-
many) by holding the indenter tip (having 1371) under the load
of 2 kg for 20 s on the surface of the sample polished to a mirror
finish. More than 20 indentations were performed to estimate a
hardness value. Dielectric properties were measured using a
Hewlett Packard HP 8510 Network Analyzer (Santa Rosa,
CA) using 15.8 mm ꢀ 7.8 mm ꢀ 1.7 mm size specimen fixed in

October 2010
Hydrolysis-Induced Aqueous Gelcasting of b-SiAlON–SiO2 Ceramic Composites
3183
mild steel flanges (32.5 mm ꢀ 32.5 mm ꢀ 1.7 mm) according to
the Nicolson and Ross technique.17–19,31
III. Results and Discussion
Table I lists the powder compositions used to prepare various
ceramic composites following the conventional dry-powder
pressing route, the values of green density of dry-powder pressed
pellets, and of bulk density, apparent porosity, and linear
shrinkage of ceramic composites sintered for 4 h at 15001–
16751C and for 3 h at 17501C. Different codes are given to the
powder compositions for easy identification purposes. SS in the
sample codes indicates the ceramic composites made from a
powder mixture of b-Si4Al2O2N6 and fused silica, and the num-
bers 0–100 represent the fused silica content in the starting pow-
der compositions. The following conclusions can be drawn from
the data of Table I. (1) There is a gradual increase in the value of
green density with the increase of b-Si4Al2O2N6 content in the
composite. This is due to the fact that fused silica possesses
a lower density (2.2 g/cm3) than that of b-Si4Al2O2N6 (3.10
g/cm3). (2) In general, the compositions with fused silica
ꢃ50 wt% (i.e., SS-50, SS-60, and SS-80) underwent early dens-
ification even at a temperature of 15001C when compared with
b-Si4Al2O2N6-rich compositions (SS-20 and SS-40) and with the
fused silica (SS-100). The latter compositions underwent desired
densifications only when sintered at 17501 and 16751C, respec-
tively. The early densification of fused silica-rich compositions
when compared with pure fused silica could be attributed to the
eutectic melts formed among Y2O3, Al2O3, and SiO2 at 13501C
during sintering.14 (3) The b-Si4Al2O2N6 densified from its pow-
der (i.e., SS-0) exhibited considerably low apparent porosity
values (o0.5%) after sintering at 16751 and 17501C. The re-
maining composites exhibited significantly higher apparent po-
rosity values (up to 41.2%) even after sintering at the desired
temperature. (4) Among all the sintered samples, pure fused sil-
ica (SS-100) underwent the highest shrinkage (12.41%) after
sintering for 4 h at 16751C. The remaining sintered composites
exhibited linear shrinkage values comparable with those values
reported for similar materials prepared by dry-powder pressing
followed by sintering.17,18
XRD patterns of sintered SS-0 to SS-100 are presented in
Fig. 2. It can be seen that SS-0 shows XRD lines primarily due
to b-Si4Al2O2N6 (ICDD File No.: 00-048-1616) together with
some minor peaks that are attributed to Y2SiAlO5N (ICDD File
No.: 00-048-1627) and other un-known phases.17,18 The phase
Y2SiAlO5N forms due to the reactions that occur among Y2O3,
a-Al2O3, and surface SiO2 of a-Si3N4 during sintering.14 Altho-
ugh, SS-20 also exhibited XRD lines primarily due to the
b-Si4Al2O2N6 phase with a slight shift in their 2y values, it has
also exhibited some minor peaks due to a new phase, Si2N2O
(ICDD File No.: 00-047-1627). On the other hand, SS-40 ex-
hibited XRD lines exclusively due to this new phase Si2N2O.
The sintered fused silica revealed XRD lines exclusively due to
the cristobalite phase indicating that it stabilizes into the cristob-
alite phase during sintering. Although SS-50 was prepared from
a powder mixture consisting 50 wt% b-Si4Al2O2N6 and 50 wt%
fused silica, it exhibited XRD lines primarily due to fused silica-
derived cristobalite together with some minor lines due to start-
ing b-Si4Al2O2N6 and due to newly formed phase, Si2N2O. On
the other hand, both SS-60 and SS-80 exhibit strong XRD lines
due to cristobalite phase together with some minor lines due to
b-Si4Al2O2N6 and Si2N2O phases. In these composites, interest-
ingly, the starting material b-Si4Al2O2N6 exhibited stronger
peaks than the new phase, Si2N2O. The considerable amount
of viscous liquid melt formed in these two composites during
sintering normally does not allow moving relatively largely iso-
lated b-Si4Al2O2N6 particles with enough speed to react with
and form the Si2N2O phase.2,3
As mentioned in the introduction section, certain high-speed
(4Mach 5) radomes see a transient temperature of about
13001C in service. Therefore, the construction materials of these

3184
Journal of the American Ceramic Society—Ganesh and Sundararajan
Vol. 93, No. 10
curred upon casting and drying, and green flexural (four-point
bend) strength of consolidates. In the given sample codes, SS40
indicates the ceramic composites made from a powder mixture
of 60 wt% b-Si4Al2O2N6 and 40 wt% fused silica, and the num-
bers 1–5 represent the AlN added to suspensions equivalent to
the weight percent of Al2O3. The viscosity of the suspensions
considerably decreased with increasing concentrations of AlN in
the suspension, reaching the lowest value of 406 MPa ꢂ s for
SS40-5. Usually, when AlN is introduced into aqueous suspen-
sions, the following three reactions occur:
− β-Si Al O N
2 2 6
•
♦
4
− Si2N2O
− Cristobalite
SS-100: 1675 °C/4 h
SS-80: 1500 °C/4 h
•
•
•
•
•
•
•
•
•
AlN þ 2H2O ! AlOðOHÞ þ NH3
NH3 þ H2O ! NH4OH
(1)
(2)
SS-60: 1500 °C/4 h
♦
♦
•
•
•
•
•
♦
SS-50: 1500 °C/4 h
♦
•
♦
AlOðOHÞ þ H2O ! AlðOHÞ3
(3)
•
♦
♦
♦
It consumes water, releases ammonia, and forms aluminum
hydroxides. This released ammonia changes the pH and rheo-
logical characteristics of aqueous suspensions.24,25 In the present
case, the generated ammonia gas increases the pH of the sus-
pension away from the iso-electric-point of the dispersed pow-
ders, thereby increasing the stability and fluidity of suspensions
by decreasing their viscosities.18,25 Further, the pH of the sus-
pension was found to be slightly increased from about 8.5 to 10
with the increase of AlN concentration in the suspension equiv-
alent to 1–5 wt% of Al2O3. The suspension, SS40-1, took about
11 min to completely convert into a gel after introducing a po-
lymerization initiator, a catalyst, and the AlN equivalent to
1 wt% of Al2O3. On the other hand, the suspension, SS40-5,
took only about 5 min after introducing the polymerization ini-
tiator, the catalyst, and the AlN equivalent to 5 wt% of Al2O3.
The suspension introduced with APS and TEMED at a ratio of
4 and 2 mL/g of particulate suspension, respectively, and with
AlN equivalent to 5 wt% Al2O3 was found to be best for fab-
ricating thin-wall radome shapes (Fig. 1(b)) without any diffi-
culty. These results indicate that AlN concentration in the
suspension drastically reduces the setting time of suspensions
under ambient conditions.18,25 In all the three events (Eqs. (1)–
(3)), water is consumed and solids content is increased. These
events support the early setting of suspensions when higher
amounts of AlN are introduced into them.24 The flexural
strength of green bodies also increased with the AlN content.
In GCHAS process, strength not only comes from polymeriza-
tion of organic monomers present in the suspension but also by
the cementing action of boehmite formed by the hydrolysis of
AlN added to the suspension.18,20–23 The generated boehmite
connects the ceramic particles into a stiff body and the content
of this boehmite increases with the concentration of AlN in the
suspension.24 Because of the involvement of both the strength-
ening mechanisms, SS40-5 exhibited the highest green strength
value of 20.2572.66 MPa. Green ceramics with a flexural
strength of even 18 MPa have been successfully green machined
to obtain the desired shape following usual machining opera-
tions.27 Green machining can in turn greatly reduces the expen-
♦
♦
SS-40: 1750 °C/3 h
♦
♦
♦
♦
♦
•
♦
♦
♣
•
•
•
•
•
•
SS-20: 1750 °C/3 h
•
♦
♦
•
•
•
•
•
•
•
•
•
•
•
•
•
SS-0: 1750 °C/3 h
•
•
•
•
•
•
•
•
•
10
20
30
40
2θ°
50
60
70
Fig. 2. XRD patterns of SS-0, SS-20 to SS-80 ceramic composites, and
SS-100 sintered at different conditions (Table I).
radomes should be able to withstand a temperature of at least
13001C and also should have dielectric constants preferably
o6.1–15 Among the various ceramic composites prepared in
this study (listed in Table I), only SS-40 is expected to meet both
these requirements, i.e., the high-temperature-withstanding ca-
pability and the dielectric constant as it needed the highest sinte-
ring temperature (17501C) used in this study for achieving the
maximum densification and it contains the maximum amount of
fused silica, a low dielectric constant material among composites
sintered at 17501C. Considering these facts, SS-40 was selected
to consolidate following the advanced new net-shaping tech-
nique, GCHAS as this technique allows the fabrication of thin-
wall radomes with near-net shapes and with very high green
strengths (i.e., 420 MPa).18,20–23
Table II lists the precursor mixture compositions used in the
GCHAS process, the values of viscosity of suspensions (at a
shear rare, g, of 141 sꢁ1), the setting time of suspensions, as well
as the values of green density, percentage linear shrinkage oc-
Table II. Properties of Suspensions and of Green Consolidates Obtained by the GCHAS Route
Precursor mixture
a-Al2O3 (wt%) to be
formed upon
Suspension
Linear shrinkage
from casting to
drying (%)
Ceramic
compositez
b-Si4Al2O2N6 Fused silica
hydrolysis and firing of Suspension viscosity at setting time
g 5 141 sꢁ1 (MPa ꢂ s)
Green
Green
(%)
(%)
A-AlN (%)
added A-AlN
(min)
strength (MPa) density (g/cm3)
SS40-1
SS40-2
SS40-3
SS40-4
SS40-5
59.51
59.03
58.52
58.06
57.56
39.67
39.35
39.01
38.70
38.37
0.80
1.61
2.47
3.24
4.06
1
2
3
4
5
503
467
433
416
409
B11
B10
B8
14.1371.87 1.8370.06 2.3270.23
15.3272.17 1.8270.04 2.3370.11
16.7272.32 1.8270.05 2.3370.32
18.4272.48 1.8170.03 2.3470.27
20.2572.66 1.8070.05 2.3470.21
B6
B5
wValues were obtained as explained in ‘‘Section I’’. Suspensions were prepared by dispersing powder mixtures of 60 wt% b-Si4Al2O2N6140 wt% fused silica in
NVPNCPM (n-vinylpyrrolidone nonconvectional premix solution) using 10 mL Dolapix CE 64 dispersing agent/g of powder mixture. zIn sample codes, SS-40 indicates the
ceramic composite formed from a powder mixture of 60 wt% b-Si4Al2O2N6140 wt% fused silica; the numbers 1–5 represent the AlN content equivalent to 1–5 wt% Al2O3 on
total powder basis added to the suspensions; A-AlN stands for as purchased AlN powder, which was not passivated against hydrolysis. GCHAS, hydrolysis-induced aqueous
gelcasting.

October 2010
Hydrolysis-Induced Aqueous Gelcasting of b-SiAlON–SiO2 Ceramic Composites
3185
sive post sintering machining operation, and thereby the man-
ufacturing cost.18,26
A gradual decrease in the values of green density (from 1.83
to 1.80 g/cm3) and a gradual increase in the values of percentage
linear shrinkage (from 2.32% to 2.34%) with the increasing AlN
concentration in suspension can be observed from the data pre-
sented in Table II. The decrease of green density with the in-
creasing AlN concentration in the suspension could be
attributed to the amount of NH3 gas released in the suspension
due to AlN hydrolysis (Eq. (1)).24 Normally, some of the am-
monia gas remains entrapped in the suspension having a viscos-
ity of 4400 MPa ꢂ s, even after vacuum pumping. This
entrapped gas leads to the formation of fine pores in the green
bodies, which in turn is responsible for lower densities of
consolidates.24 Nevertheless, these green density values are
well-comparable with the one (1.82 g/cm3) measured for
a dry-powder-pressed sample having the same composition of
b-Si4Al2O2N6 and fused silica (SS-40).
XRD patterns of SS40-1 to SS40-5 sintered for 3 h at 17501C
are presented in Fig. 3. XRD lines only due to Si2N2O (ICDD
File No.: 00-047-1627) are seen in all the spectra indicating that
the available 40 wt% fused silica in the reaction mixture could
not be converted into the cristobalite phase; instead, it reacted
with b-Si4Al2O2N6 and led to the formation of a new phase,
Si2N2O. The boehmite and/or Al(OH)3 formed by the hydrolysis
of AlN in the suspension appear to have little effect on the phase
evolution in these ceramic composites. In fact, this boehmite
and/or Al(OH)3 are supposed to react with fused silica and/or
the surface SiO2 of b-Si4Al2O2N6 during sintering and a led to
the formation of certain alumino-silicate phases such as, mul-
lite.32 However, no XRD lines due to any of the alumino-silicate
phases are noted in these XRD spectra. The reason for the ab-
sence of these phases could be that the phases formed and/or
their crystallite sizes are below the detection limit of the XRD
instrument.28
Fig. 4. (a) Bulk density, apparent porosity, and water absorption
capacity, and (b) linear shrinkage and mass loss of SS40-1 to SS40-5
sintered for 3 h at 17501C.
♦
♦
The SS-40 consolidated by the dry-powder pressing route
followed by sintering for 3 h at 17501C exhibited a bulk density
of 2.51 g/cm3, apparent porosity of 5.445%, water absorption
capacity of 2.168%, linear shrinkage of 10.68%, hardness of
1319 kg/mm2, fracture toughness of 3.22 MPaꢂ m1/2, and flex-
ural strength of 113 MPa. On the other hand, the pure stoic-
hiometric b-Si4Al2O2N6 formed by the reaction sintering (at
16751C for 4 h) of extrudates consisting a-Si3N4, a-Al2O3, sur-
face passivated AlN against hydrolysis, and Y2O3 (7 wt %) ex-
hibited a bulk density of 3.08 g/cm3, apparent porosity of
0.19%, water absorption capacity of 0.12%, linear shrinkage
of 14%, hardness of 1317 kg/mm2, fracture toughness of
3.30 MPa ꢂ m1/2, and flexural strength of 216 MPa. Although
the hardness and fracture toughness values of these two different
sintered materials are comparable, there is a large gap between
their flexural (four-point bend) strength values (b-Si4Al2O2N6
had a strength of 226 MPa, whereas the SS-40 (i.e., Si2N2O) had
only 113 MPa). The kinds of bonds and the amounts of porosity
present in these sintered materials could be responsible for the
difference in the flexural strength values measured.14
The values of bulk density, apparent porosity, and water ab-
sorption capacity, and the values of linear shrinkage and mass
loss of SS40-1 to SS40-5 sintered for 3 h at 17501C are plotted in
Figs. 4(a) and (b), respectively. Their corresponding hardness
and fracture toughness values, and the flexural strength and
Young’s modulus values are plotted in Figs. 5(a) and (b),
respectively, as a function of the amount of AlN added to their
corresponding suspensions. In general, all these properties
appear to be very sensitive to the AlN concentration in the sus-
pensions. There is a significant difference between the apparent
porosity and water absorption capacity values measured for
SS-40, and SS40-1 to SS40-5, although all these samples were
sintered under identical conditions. Furthermore, there is a
− Si N O
♦
2
2
♦
♦
♦
♦
♦
SS40-5
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦♦
♦
♦
♦
♦
♦
♦
♦
♦
SS40-4
♦
♦
♦
♦
♦
♦
♦
♦
♦ ♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
SS40-3
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
SS40-2
♦
♦
♦
♦
♦
♥
♦
♦ ♦
♦
♦♦
♦
♦
♦
♦
♦
♦
SS40-1
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦ ♦
10
20
30
40
2θ°
50
60
70
Fig. 3. XRD patterns of SS40-1 to SS40-5 ceramic composites sintered
for 3 h at 17501C (Table II).

3186
Journal of the American Ceramic Society—Ganesh and Sundararajan
Vol. 93, No. 10
alumino-silicate chemical bonds formed by the reactions oc-
curred between aluminum hydroxides derived from AlN added
to the suspensions, and the Si2N2O (Fig. 2) could be responsible
for the above observed changes in the sintered properties. Nev-
ertheless, further work is under active progress to fully establish
the underlying mechanisms in improving the above properties
due to AlN addition into the suspensions.
The fractured SEM micrographs of sintered SS-40 and
SS40-1 to SS40-5 are presented in Figs. 6(a)–(f), respectively.
Purposefully, micrographs were recorded at pores in order to
understand the features of grains that were not broken upon
fracture. A considerable amount of porosity together with elon-
gated grains and fused mass is observed in all these micrographs.
The observed porosity justifies the mass loss that occurred upon
sintering. Relatively more elongated grains are seen in SS40-1 to
SS40-5 in comparison with SS-40. The EDAX study revealed
that the elongated grains are made of Si2N2O, and the fused
mass is made of cristobalite and/or fused silica. Furthermore, as
none of the SS40-1 to SS40-5 sintered ceramics revealed the ap-
parent porosity values 41%, the porosity appearing in these
micrographs could be a closed one, where water could not
be entered while estimating apparent porosity values following
the water displacement method. In line to its bulk density and
apparent porosity properties, the micrograph of SS-40 reveals
the presence of relatively large size pores when compared with
those present in SS40-1 to SS40-5. Furthermore, an increased
connectivity among the elongated grains can be seen with the
increasing amount of AlN in suspension. These improved con-
nections could be responsible for superior properties measured
for SS40-5 in comparison with those measured for SS-40, and
SS40-1 to SS40-4.
Dielectric constant profiles of SS40-1 to SS40-5 ceramic com-
posites sintered for 3 h at 17501C are presented in Fig. 7 in the
frequency range of 16.8–17.2 GHz. A quite stable and low di-
electric constant values (in the range of 5.896–6.313) can be seen
at 17 GHz frequency. The dielectric constant value of 6.03 was
measured at 17 GHz frequency for sintered SS-40, whereas the
values 7.3 and 3.9 were reported for the stoichiometric
b-Si4Al2O2N6 formed by the reaction sintering of Si3N4,
Al2O3, AlN, and Y2O3 precursor mixture and for sintered fused
silica, respectively.17,33,34 The dielectric constant of any ceramic
material is not only influenced by its chemical composition but
also by the several other properties such as, frequency of the
applied field, temperature of the dielectric, humidity, crystal
structure, and other external factors including porosity.35,36
Furthermore, as the applied frequency is varied through 16.8
to 17.2 GHz at an ambient temperature in this study, the mea-
sured dielectric constant values could be mainly contributed by
Fig. 5. (a) Hardness and fracture toughness, and (b) flexural strength
and Young’s modulus of SS40-1 to SS40-5 sintered for 3 h at 17501C.
gradual increase in the values of bulk density, hardness, fracture
toughness, flexural strength, and Young’s modulus, and a con-
current decrease in the values of apparent porosity, water
absorption capacity, linear shrinkage, and mass loss with the
increasing concentration of AlN in the suspension. Hardness is
increased from 1270 to 1390 kg/mm2, fracture toughness from
3.35 to 4.2 MPa ꢂ m1/2, flexural strength from 90 to 140 MPa,
and Young’s modulus from 184 to 213 GPa, and on the other
hand, the linear shrinkage is decreased from 22.5% to 18% and
mass loss from 18% to 11%. The increased amounts of new
(a)
(b)
(c)
(d)
(e)
(f)
Fig. 6. The sintered SEM micrographs of SS-40 (a), SS40-1 (b), SS40-2 (c), SS40-3 (d), SS40-4, (e), and SS40-5 (f).

October 2010
Hydrolysis-Induced Aqueous Gelcasting of b-SiAlON–SiO2 Ceramic Composites
3187
Fig. 8. Coefficient of thermal expansion profiles SS40-1 to SS40-5
sintered for 3 h at 17501C.
Fig. 7. Dielectric constant profiles of SS40-1 to SS40-5 sintered for 3 h
at 17501C.
electronic and ionic polarization mechanisms only.36 In addition
to these, a gradual decrease in the dielectric constant value with
the increasing amount of AlN can also be seen. Although, po-
rosity decreases the dielectric constant of a sintered material, in
the present case, the dielectric constant values were decreased
with the decreasing porosity, which indicates that the changes
noted in the dielectric constant values with the increasing con-
centration of AlN are mainly due to the changes that occurred in
the chemical composition but not due to the changes that oc-
curred due to their physical properties such as porosity. The
decreasing trend (Fig. 4(b)) in degree of mass loss occurred in
SS40-1 to SS40-5 sinters with the increasing amount of AlN in
the suspension. The decreasing of degree of mass loss means an
increasing of the degree of SiO2 in SS40-1 to SS40-5 sinters as
mass loss in SiAlON-based ceramics occurs mainly due to the
evaporation of SiO2 at elevated temperatures.27 As the dielectric
constant of SiO2 is lower than that of b-Si4Al2O2N6 and Si2N2O,
the increased amount of SiO2 with AlN amount has caused the
reduction in the dielectric constant of SS40-1 to SS40-5 ceram-
ics. In addition to this, the dielectric constant (5.5) of alumino-
silicate compounds such as mullite to be formed due to reactions
occurring between the alumina derived from AlN in the sus-
pension and fused silica and/or b-Si4Al2O2N6 is lower than that
of b-Si4Al2O2N6 (7.4).9,33,34 From these dielectric constant re-
sults, it can be concluded that by mixing 40 wt% fused silica to
the b-Si4Al2O2N6 powder, the dielectric constant of their sin-
tered composite product could be reduced from 7.313 (measured
for b-Si4Al2O2N6) to 6.313 (measured for SS-40), which can be
further reduced to 5.896 (measured for SS40-5) by the addition
of AlN equivalent to 5 wt% Al2O3 in its starting suspension.
Dielectric constant values in the range of 6.84–7.46 were re-
ported for gelcast b-Si4Al2O4N6 ceramics.2–7,17–19 A dielectric
constant value of 4.78 at 251C and 5.0 at 10001C, and a loss
tangent value of 0.0014 at 251C were reported for SiON nano-
composite.1 A dielectric constant value of 6.3 was reported for
b-Si4Al2O2N6-9 wt% SiO2 composite formed at 17501C.2 A di-
electric constant value of 4.9 at 8.6 GHz was reported for a ce-
ramic composite having a chemical composition of 70%
Table III. Sintered Properties of SS40-5 and of Various Commercial Radome Materials
Coefficient of
Materials trade and
manufacturer’s name
Basic chemical
composition
Bulk density
(g/cm3)
Elastic
Flexural strength thermal expansion,
CTE (10ꢁ6 1Cꢁ1
Dielectric
permittivity (e0) Tan d
modulus (GPa)
(MPa)
)
References
Pyoceram 9606
(Corning Inc.,
Corning, NY, USA)
Fused silica
Cordierite
SiO2
2.6
2
121
37
240
4.7
5.5
0.0005
Kirby et al.4
43
0.7
3.3
0.003
Nei et al.6
(Ceradyne Inc.,
Costa Mesa, CA)
IRBAS (Lockheed Si3N4
Martin Inc.)
3.18
280
550
800
180
260
—
3.2
—
7.6
8
0.002
0.002
Mangels and
Mikijelj37
Mangels and
Ceralloy 147-31N
(Ceradyne Inc.)
Si3N4
3.21
310
Mikijelj37
Ceralloy 147-01EXP Reaction bonded
(Ceradyne Inc.) Si3N4
b-SiAlON (ORNL) b-Si4Al2O2N6
1.8–2.5
3.02
50–200
230
3.1
4.1
5.0
4–6
7.4
3.3–5
0.002– Mangels and
0.005
0.003
Mikijelj37
Kirby and
colleagues4,6
Mora and
Cerablakt, Applied Al2O3–P2O5
2.0–2.5
—
—
Thin Films Inc.,
Lockheed Martin
Corporation,
composite
colleagues8,9
Bethesda, MD
Invisicone (AMO) b-SiAlON–SiO2
SS40-5 (material of Si2N2O
this study, Table II)
2.2
2.81
—
214
532
140
2.0
3.50
4.9
0.002
Paquette3
—
5.896 0.002–
0.003
ORNL, Oak Ridge National Laboratory, Oak Ridge, TN, USA; AMO, Advanced Materials Organization; CTE, coefficient of thermal expansion.

3188
Journal of the American Ceramic Society—Ganesh and Sundararajan
Vol. 93, No. 10
SiAlON130% SiO2.3 Thus, the dielectric constant values
(5.896–6.313) measured in this study for SS40-1 to SS40-5 pre-
pared by simple GCHAS are well comparable with those values
reported in the literature for similar kind of materials that were
prepared following quite complicated procedures involving
quite capital-intensive equipments.3
40 wt% fused silica exhibits flexural strength values more than
three times of sintered bulk fused silica, and a dielectric constant
value less than that of stoichiometric b-Si4Al2O2N6.
Acknowledgment
The CTE profiles recorded between 301 and 10001C for SS40-
1 to SS40-5 sintered for 3 h at 17501C are presented in Fig. 8.
These ceramics revealed CTE values in the range of
3.50ꢁ4.016 ꢀ 10ꢁ6 1Cꢁ1. There is a gradual increase of CTE
from 3.766 to 4.016 ꢀ 10ꢁ6 1Cꢁ1 with the increase of AlN con-
centration in the suspension from 1 to 3 equivalent wt% of
Al2O3, and then there is a decreasing trend in the value reaching
the lowest value of 3.50 ꢀ 10ꢁ6 1Cꢁ1 when the concentration of
AlN increased the equivalent to 5 wt% Al2O3. The variations in
the CTE values measured for these ceramics can be mainly at-
tributed to the changes that occurred to their chemical compo-
sition by the incorporation of AlN into their starting
suspensions. The CTE value of 4.1ꢀ 10ꢁ6 1Cꢁ1 for a gelcast
b-Si4Al2O2N6 and of 2.1 ꢀ 10ꢁ6 1Cꢁ1 for a ceramic composite of
b-SiAlON130% SiO2 between 251 and 10001C are reported.3–5
The CTE values in the range of 3.532ꢁ4.657 ꢀ 10ꢁ6 1Cꢁ1 be-
tween 301 and 7001C have been reported for b-Si6ꢁzAlzOzN8ꢁz
in which z varied from 1.5 to 4.17 Thus, the CTE values mea-
sured for SS40-1 to SS40-5 are quite comparable with the values
reported for similar kind of materials in the literature.4,5,17,26
Table III presents the basic chemical composition, bulk den-
sity, elastic modulus, flexural strength, CTE, dielectric permit-
tivity, and tan d of SS40-5 and of several other commercial
radome materials. Material trade names as well as the supplier’s
and their corresponding country names are also presented in this
table. In general, all the radome materials exhibited reasonably
low bulk density values (o3.2 g/cm3), which are recommended
for high-speed radome materials. It can be seen that among
various materials, only the Si3N4-and b-SiAlON-based materials
exhibited relatively higher elastic modules and flexural strength
values.37 Except Cerablakt (5.0ꢀ 10ꢁ6 1Cꢁ1), Pyroceram 9606
(4.7ꢀ 10ꢁ6 1Cꢁ1), and b-SiAlON (4.1 ꢀ 10ꢁ6 1Cꢁ1), all other
materials exhibited relatively low CTE values (o3.5 ꢀ 10ꢁ6
1Cꢁ1). As far as the important dielectric permittivity property
is concerned, the fused silica appears to be the best as it has the
value of 3.3. The highest permittivity values are reported for b-
SiAlON (7.4), and the Si3N4-based materials, IRBAS (7.6) and
Ceralloy 147-31N (8).37 Nevertheless, all these materials exhib-
ited considerably low tan d values (i.e., o0.003). When com-
pared with other materials, the SS40-5 possesses elastic modulus
and flexural strength values that are almost three times higher
than those of fused silica. The dielectric permittivity value of
SS40-5 was measured to be 5.896, which is less than that of
b-Si4Al2O2N6 (7.4), IRBAS (7.6), and Ceralloy 147-31N (8).
Based on these permittivity and strength properties together
with the high temperature (413001C) withstanding and radome
shape-forming capability, this SS40-5 can be considered for cer-
tain high-speed radome applications.1–15
Authors express their gratitude to their colleagues at ARCI for their contri-
butions to this study.
References
1G. Gilde, P. Patel, C. Hubbard, B. Pothier, T. Hynes, W. Croft, and J. Wells,
‘‘SiON Low Dielectric Constant Ceramic Nanocomposite’’; U.S. Patent No.:
5,677,252, October 14, 1997.
2I. Ganesh, N. Thiyagarajan, D. C. Jana, P. Barik, and G. Sundararajan, ‘‘An
Aqueous Gelcasting Route to Dense b-Si4Al2O2N6–0.5SiO2 Ceramics,’’ J. Am.
Ceram. Soc., 91 [5] 1566–71 (2008).
3D. G. Paquette, ‘‘Method of Making a Radar Transparent Window Material
Operable Above 20001C’’; U.S. Patent No.: 5,627,542, May 6, 1997, and references
therein.
4K. W. Kirby, A. T. Jankiewicz, M. A. Janney, C. Walls, and D. Kupp, ‘‘Gel-
casting of GD-1 Ceramic Radomes’’; pp. 287–95 Proceedings of the 8th DoD Elec-
tromagnetic Windows Symposium, St. Louis, MO, 2000.
5M. A. Janney, C. A. Walls, D. M. Kupp, and K. W. Kirby, ‘‘Gelcasting
SiAlON Radomes,’’ Am. Ceram. Soc. Bull., 83, 9201–6 (2004).
6J. T. Neil, L. J. Bowen, and B. E. Michaud, ‘‘Fused Silica Radome’’; U.S.
Patent No.: 4,949,095, August 14, 1990.
7D. W. Freitag and K. K. Richardson, ‘‘BAS Reinforced In-Situ with Silicon
Nitride’’; U.S. Patent No.: 5,358,912, October 25, 1994.
8F. G. Mora, K. C. Goretta, D. Singh, J. L. Routbort, S. Sambasivan, K. A.
Steiner, J. Adabie, and K. K. Rangan, ‘‘High-Temperature Deformation of Amor-
phous AlPO4-Based Nano-Composites,’’ J. Eur. Ceram. Soc., 26, 1179–83 (2006).
9Y. Wang and J. Liu, ‘‘Aluminum Phosphate-Mullite Composites for High-
Temperature Radome Applications,’’ Int. J. Appl. Ceram. Technol., 6 [2] 190–4
(2009).
10J. Wu, J. Guo, and B. Li, ‘‘Preparation and Properties of SiO2 Matrix Com-
posites Doped with AlN Particles,’’ J. Mater. Sci., 35 [19] 4895–900 (2000).
11G. Wen, G. L. Wu, T. Q. Lei, Y. Zhou, and Z. X. Guo, ‘‘Co-Enhanced SiO2–
BN Ceramics for High-Temperature Dielectric Applications,’’ J. Eur. Ceram. Soc.,
20 [12] 1923–8 (2000).
12M. Y. Hsieh, ‘‘Silicon Nitride Having Low Dielectric Loss’’; U.S. Patent No.:
4,642,299, February 10, 1987.
13M. Y. Hsieh, H. Mizuhara, and P. W. Smith, ‘‘Pressureless Sintered Silicon
Nitride as a Promising Candidate for Radome Materials’’; 17th Symposium on
Electromagnetic Windows, Atlanta, GA., July 25–27, 1984.
14S. Hampshire, H. K. Park, D. P. Thompson, and H. K. Jack, ‘‘a-SiAlON,’’
Nature (London), 274, 880–2 (1978).
15B. P. Saha, Roy Johnson, I. Ganesh, G. V. N. Rao, S. Bhattacharjee, and Y.
R. Mahajan, ‘‘Thermal Anisotropy in Sintered Cordierite Monoliths,’’ Mater.
Chem. Phys., 67 [1–3] 140–5 (2001).
16I. Ganesh, S. M. Olhero, A. A. Branca, M. R. Correia, G. Sundararajan, and
J. M. F. Ferreira, ‘‘Chemisorption of Phosphoric Acid and Surface Characteriza-
tion of as Passivated AlN Powder Against Hydrolysis,’’ Langmuir, 24 [10] 5359–65
(2008).
17I. Ganesh, N. Thiyagarajan, D. C. Jana, Y. R. Mahajan, and G. Sundarara-
jan, ‘‘Influence of Chemical Composition and Y2O3 on Sinterability, Dielectric
Constant and CTE of b-SiAlON,’’ J. Am. Ceram. Soc., 91 [1] 115–20 (2008).
18I. Ganesh, ‘‘Near-Net Shape b-Si4Al2O2N6 Parts by Hydrolysis Induced
Aqueous Gelcasting Process,’’ Int. J. Appl. Ceram. Technol., 6 [1] 89–101 (2009).
19I. Ganesh, N. Thiyagarajan, D. C. Jana, Y. R. Mahajan, and G. Sundarara-
jan, ‘‘An Aqueous Gelcasting Process for b-Si4Al2O2N6 Ceramics,’’ J. Am. Ceram.
Soc., 91 [9] 3121–4 (2008).
20I. Ganesh, S. M. Olhero, P. M. C. Torres, F. J. Alves, and J. M. F. Ferreira,
‘‘Hydrolysis Induced Aqueous Gelcasting for Near-Net Shaping of ZTA Ceramic
Composites,’’ J. Eur. Ceram. Soc., 29, 1393–401 (2009).
21I. Ganesh, G. J. Reddy, G. Sundararajan, S. M. Olhero, P. M. C. Torres, and
J. M. F. Ferreira, ‘‘Hydrolysis Induced Aqueous Gelcasting of MgAl2O4 Spinel,’’
Int. J. Appl. Ceram. Technol., doi:10.1111/j.1744-7402.2010.02520.x (2010).
22I. Ganesh, G. Sundararajan, S. M. Olhero, P. M. C. Torres, and J. M. F.
Ferreira, ‘‘A Novel Colloidal Processing Route to Alumina Ceramics,’’ Ceram.
Int., 36, 1357–64.
IV. Conclusions
The following conclusions can be drawn from the above study:
dense stoichiometric b-Si4Al2O2N6 extrudates with 490% b-
SiAlON phase can be prepared by reaction sintering of a-Si3N4,
a-Al2O3, surface passivated AlN against hydrolysis, and Y2O3
(7 wt%) precursor mixture for 4 h at 16751C. The dry-powder-
pressed composites of fine b-Si4Al2O2N6 and fused silica pow-
ders can be sintered for 3–4 h at 15001–17501C to form dense
ceramic composites with SiO2 in the range of 20–80 wt%. De-
fect-free green radome shapes with green strengths 420 MPa
can be consolidated following a net-shape technique based on
GCHAS process by involving an AlN equivalent 5 wt% Al2O3
in the starting particulate suspension together with a polymer-
ization initiator and a catalyst. The Si2N2O formed at 17501C
for 3 h from a powder mixture of 60 wt% b-Si4Al2O2N6 and
23I. Ganesh, G. J. Reddy, S. M. Olhero, P. M. C. Torres, and J. M. F. Ferreira,
‘‘Influence of Processing Route on Micro-Structure and Mechanical Properties of
MgAl2O4 Spinel,’’ Ceram. Int., 36, 473–82 (2010).
24T. Kosmac, S. Novak, and M. Sajko, ‘‘Hydrolysis-Assisted Solidification
(HAS): A New Setting Concept for Ceramic Net-Shaping,’’ J. Eur. Ceram. Soc.,
17, 427–32 (1997).
25I. Ganesh, N. Thiyagarajan, D. C. Jana, P. Barik, J. M. F. Ferreira, and G.
Sundararajan, ‘‘Dense b-SiAlONs Consolidated by
a Modified Hydrolysis
Assisted Solidification Route,’’ J. Eur. Ceram. Soc., 28 [4] 879–85 (2008).
26M. A. Janney, C. A. Walls, D. M. Kupp, and K. W. Kirby, ‘‘Gelcasting
SiAlON Radomes,’’ Am. Ceram. Soc. Bull., 7, 9201–6 (2004).
27M. A. Janney, S. D. Nunn, C. A. Walls, O. O. Omatete, R. B. Ogle, G. H.
Kirby, and A. D. McMillan, ‘‘Gelcasting’’; pp. 1–15 in The Handbook of Ceramic
Engineering, Edited by M. N. Rahman. Marcel Dekker, New York, 1998.
28S. Brunauer, P. H. Emmett, and E. Teller, ‘‘Adsorption of Gases in Multi-
molecular Layers,’’ J. Am. Chem. Soc., 60, 309–19 (1938).

October 2010
Hydrolysis-Induced Aqueous Gelcasting of b-SiAlON–SiO2 Ceramic Composites
3189
29H. P. Klug and L. E. Alexander, ‘‘X-Ray Diffraction Procedure for Polycrys-
talline and Amorphous Materials,’’ J. Appl. Crystallogr., 8, 573–4 (1975).
30G. R. Anstis, P. Chantikul, B. R. Lawn, and D. B. Marshall, ‘‘A Critical
Evaluation of Indentation Techniques for Measuring Fracture Toughness. I.
Direct Crack Measurements,’’ J. Am. Ceram. Soc., 64, 533–8 (1981).
31A. M. Nicolson and G. F. Ross, ‘‘Measurement of the Intrinsic Properties of
Materials by Time Domain Techniques,’’ IEEE Trans. Instrum. Meas., IM-19,
377–82 (1970).
34D. R. Lide (ed.), CRC Hand Book of Chemistry and Physics, 73rd edition,
CRC Press, London, 1993.
35R. C. Buchanan, ‘‘Properties of Ceramic Insulators’’; pp. 1–64 in Ceramic
Materials for Electronics, Processing, Properties and Applications, Chapter 1, 2nd
edition, Edited by R. C. Buchanan. Marcel Decker Inc., New York, 1991.
36V. Raghavan, ‘‘Dielectric Materials’’; pp. 380–92 in Materials Science and
Engineering—A First Course, Chapter 17, 4th edition, Prentice-Hall of India (P)
Ltd., New Delhi, 1998.
32W. D. Kingery, H. K. Bowen, and D. R. Uhlman, Introduction to Ceramics,
2nd edition, Wiley, New York, 1976.
37J. Mangels and B. Mikijelj, ‘‘Ceramic Radomes for Tactical Missile Systems
Available at http://www.cera dyne.com/uploads/Radome%20White%20Paper
33J. A. Dean (ed.), Lange’s Handbook of Chemistry, 12th edition, McGraw-Hill,
New York, 1979.
%20PDF.pdf (accessed November 11, 2009)
&