Precipitation coating of rare-earth orthophosphates on woven ceramic fibers-effect of rare-earth cation on coating morphology and coated fiber strength

Article abstract of DOI:10.1111/j.1551-2916.2008.02431.x

Monazite (La, Ce, Nd, and GdPO₄) and xenotime (Tb, Dy, and YPO₄) coatings were deposited on woven Nextel 610 and 720 fibers by heterogeneous precipitation from a rare-earth citrate/ phosphoric acid precursor. Coating phases and microstructure were characterized by SEMand TEM, and coated fiber strength was measured after heat treatment at 1200°C for 2 h. Coated fiber strength increased with decreasing ionic radius of the rareearth cation in the monazite and xenotime coatings, and correlates with the high-temperature weight loss and the densification rate of the coatings. Dense coatings with trapped porosity and high weight loss at a high temperature degrade fiber strength the most. The degradation is consistent with stress corrosion driven by thermal residual stress from coating precursor decomposition products trapped in the coating at a high temperature.

Full text of DOI:10.1111/j.1551-2916.2008.02431.x

J. Am. Ceram. Soc., 91 [7] 2117–2123 (2008)
DOI: 10.1111/j.1551-2916.2008.02431.x
r 2008 The American Ceramic Society
No claims to original US government works
ournal
J
Precipitation Coating of Rare-Earth Orthophosphates on Woven
Ceramic Fibers—Effect of Rare-Earth Cation on Coating Morphology
and Coated Fiber Strength
w
Geoff E. Fair and Randall S. Hay
Materials and Manufacturing Directorate, Air Force Research Laboratory, Wright-Patterson AFB, Ohio
Emmanuel E. Boakye
UES, Inc., Dayton, Ohio
2
5
Monazite (La, Ce, Nd, and GdPO ) and xenotime (Tb, Dy, and
4
YPO ) coatings were deposited on woven Nextelt 610 and 720
4
precipitation of rhabdophane, LaPO
4
Á xH
2
O. This method
was used to coat Nextelt 610 and 720 fibers with La, Ce, Nd,
and GdPO monazite and Tb, Dy, and YPO xenotime to com-
fibers by heterogeneous precipitation from a rare-earth citrate/
phosphoric acid precursor. Coating phases and microstructure
were characterized by SEM and TEM, and coated fiber strength
was measured after heat treatment at 12001C for 2 h. Coated
fiber strength increased with decreasing ionic radius of the rare-
earth cation in the monazite and xenotime coatings, and corre-
lates with the high-temperature weight loss and the densification
rate of the coatings. Dense coatings with trapped porosity and
high weight loss at a high temperature degrade fiber strength the
most. The degradation is consistent with stress corrosion driven
by thermal residual stress from coating precursor decomposition
products trapped in the coating at a high temperature.
4
4
pare coating weight loss, microstructure, morphology, and coated
fiber strength following a high-temperature heat treatment. Var-
ious factors that may affect coated fiber strength are discussed.
As CMC strength is proportional to fiber strength (for in-plane
on-axis loading), coated fibers with minimal strength degrada-
tion following exposure to composite processing temperatures
are desirable.
II. Experimental Procedure
(
1) Materials Preparation
Nextelt 610 and 720 oxide fiber 8 harness satin weave cloths
(1500 denier, 3M Co., St. Paul, MN) were used for all coating
experiments and were desized at 9001C for 5 min in air using a
tube furnace. Precursor solutions of phosphoric acid and RE-
citrate were prepared by dissolving concentrated phosphoric
acid (Fisher Scientific Co., Pittsburgh, PA) or a combination of
RE nitrate (RE 5 La, Ce, Nd, Gd, Tb, Dy, or Y, Aldrich Chem-
ical Co., Milwaukee, WI) and citric acid (Fisher Scientific Co.)
in deionized water. The precursor solutions were prepared to
I. Introduction
ARE-EARTH orthophosphates (REPO ) form two poly-
morphs: monoclinic monazite for lanthanides from La to
Gd and tetragonal xenotime for lanthanides from Tb to Lu, and
also Y. Orthophosphates exhibit a number of interesting
properties, and have been studied extensively as hosts for
actinide waste disposal due to their stability in water, and as
constituents of ceramic composites (CMCs).
LaPO , is used as an oxidation-resistant fiber–matrix interface
4
R
1,2
3
,4
5
–11
La-monazite,
yield 100 g REPO /L on mixing equal volumes of each solution.
4
4
The RE:citrate:P ratio of the solutions was fixed at 1:2:1. The
RE-citrate and phosphoric acid solutions were chilled to B51C
before mixing during the coating procedure.
material for CMCs due to its low hardness and weak bonding to
alumina and other structural ceramics; it also has a good CTE
5–7,10,11
match with Nextelt 610 and 650 fibers.
Coatings of
xenotime (REPO where RE5 Tb–Lu and Y) and mixed-phase
4
monazite–xenotime solid solutions have been proposed for
Nextelt 720 fibers due to a better CTE match with this fiber
(
2) Coating Oxide Fiber Cloths
12–15
Coating was performed by submerging a cloth previously satu-
rated with the chilled, mixed precursor solutions into warm water
to precipitate the coating onto the fibers. Five milliliter portions
of both chilled precursor solutions (RE-citrate and phosphoric
(B6 ppm/1K) and lower hardness.
DyPO - and LaPO -
4 4
coated single-crystal alumina fibers have similar sliding stresses
in fiber push-out experiments.
1
6
During the past decade, methods to apply La-monazite coat-
ings to fibers tows using a sol and solution-based continuous
These coated fibers of-
4
acid) prepared to yield 100 g REPO /L were mixed in a large
17–21
dip-coating process were developed.
beaker, to which a 7.6 cm  7.6 cm piece of cloth was added. The
beaker containing the mixed solutions and cloth was placed in
an ultrasonic bath for 15 s. The beaker was removed from the
bath and the cloth was removed from the solution. Excess so-
lution was drained from the cloth against the side of the beaker.
The cloth, saturated with the precursor solution, was submerged
in a vessel containing deionized water at 901C to precipitate hy-
2
2
ten have to be woven into fiber preforms, and the coating may
debond during weaving. Recently, methods to coat woven ce-
ramic fibers with La-monazite using heterogeneous precipitation
2
3–25
from solution precursors have been developed.
One method
uses the temperature dependence of the reaction between La-
citrate and phosphoric acid to coat woven fibers by first satu-
rating the woven fibers with the solution precursors and then
submerging the saturated fibers in warm water to cause rapid
drated REPO rapidly. After 5 min in the warm water bath, the
4
cloth was removed and rinsed for 20 s in 400 mL of deionized
water to rinse away any remaining dissolved chemicals and
loosely bound solids. The cloth was subsequently dried in air
at 1201C for 15 min and finally placed in a tube furnace for
R. Bordia—contributing editor
5
min in air; furnace temperatures of 6001 and 9001C were used
for this intercoat firing. The cloth was quickly moved into and
out of the tube furnace, which was equilibrated at the desired
temperature. The coating process (saturate, submerge, dry, fire)
Manuscript No. 23801. Received September 28, 2007; approved February 26, 2008.
Author to whom correspondence should be addressed. e-mail: geoff.fair@wpafb.af.mil
w
2
117

2
118
Journal of the American Ceramic Society—Fair et al.
Vol. 91, No. 7
was repeated to build up thicker coatings comprised of either
five or 10 coats.
1
.9
.8
Monazites
Xenotimes
0
0
6
7
10
20
(
3) Powder Preparation for Densification Studies
0.7
0.6
26,27
Rhabdophane and churchite
(hydrated xenotime, RE-
PO O, RE5 Tb–Lu and Y) powders for densification
4
 H
2
0
0
.5
.4
studies were prepared by mixing 10 mL portions of chilled pre-
cursor solutions (RE-citrate and phosphoric acid) in a small
beaker and pouring the liquid into B600 mL of stirring deion-
ized water at 901C, resulting in rapid precipitation. The powder
was stirred in hot water for 5 min. The vessel containing the
powder and hot water was then removed from the hot plate and
cooled to near room temperature before being evenly divided
into centrifuge jars. The slurry was then centrifuged (Sorvall
Super T21, Kendro Laboratory Products, Newton, CT) at
2h 1200°C strength :
N610 ~2.6 GPa
N720 ~2.1 GPa
0.3
0
0
.2
.1
0
La Ce Pr Nd Sm Eu Gd Tb Dy
Y
Ho Er Tm Yb Lu
REPO4 (~Ionic radius)
1
3000 rpm for 25 min to collect the precipitated solids. After
Fig. 1. Normalized heat-treated strength as a function of rare-earth
cation.
centrifugation, the clear supernatant was poured off and 100 mL
deionized water was added to each jar. The settled solids were
redispersed into the liquid by scraping and stirring with a spat-
ula. The redispersed slurries were then centrifuged as before and
the supernatant was poured off. The solids were allowed to dry
in laboratory air and then ground with a mortar and pestle to a
fine powder. To prepare pellets for densification studies, pow-
ders were mixed with ethanol and 4 wt% (based on powder
weight) PEG-8000 (Carbowax Sentry, Dow Chemical Co., Mid-
land, MI) was added. The mixture was then ground in a mortar
and pestle and the ethanol was allowed to evaporate. Dried
powder mixtures were then pressed in a 12.7 mm die at 200 MPa
for 5 min. The resulting pellets were then fired at 9001C for
to assess the level of strength degradation by fiber coatings for
the given processing conditions.
III. Results and Discussion
Figure 1 shows the heat-treated strengths of Nextelt 610 and
720 fibers coated with different rare-earth orthophosphates un-
der fixed processing conditions (five coats, 9001C intercoat firing
temperature). The strengths are normalized by the heat-treated
strengths of uncoated fibers. The heat-treated strengths of both
fibers increase with decreasing ionic radius of the rare-earth
3
0 min to provide mechanical integrity for density measure-
ments. Both Archimedes and geometrical densities of the pellets
were measured following heat treatment at 12001C for 2 h.
cation in REPO across the lanthanide series of the periodic ta-
4
ble. The strengths of coated Nextelt 720 fibers are lower than
those of Nextelt 610. This is consistent with a strength degra-
1
9–21
dation mechanism driven by residual stress.
Higher residual
(
4) Characterization and Testing
stresses are present in Nextelt 720 due to the CTE mismatch
between the alumina and mullite phases and the CTE anisotropy
of these phases. In Nextelt 610, the residual stresses are due
solely to CTE anisotropy of the alumina grains. The difference
in average polycrystalline CTE between mullite and alumina is
B4 ppm/1K, whereas the difference in CTE along the a- and c-
axis of alumina is B1 ppm/1K. In polycrystalline alumina, re-
sidual stresses on the order of several hundred megapascals are
Coatings were characterized using both polished cross sections
of coated cloth and fiber tows extracted from the coated
2
8,29
cloth.
Cross sections and surfaces of extracted tows were
examined using a scanning electron microscope (SEM) (Leica 360
FE, Ernst Leitz, Westler, Germany). Transmission electron mi-
croscopy (TEM) (Phillips CM200FEG, Eindhoven, the Nether-
lands) was performed to characterize the coating microstructure
and identify any intergranular phases. Thermogravimetric anal-
ysis of powders was performed using a combined DSC/TGA
instrument (SDT Q600, TA Instruments, New Castle, DE); ex-
periments were performed at a fixed heating rate of 101C/min to
a final temperature of 14001C with the sample held in a small
alumina crucible.
3
3,34
obtained on cooling from a high temperature;
these stresses
relax with increasing temperature but are nonzero at the tem-
peratures where high-temperature outgassing of the coatings
occurs, i.e., 9001–12001C. The large CTE mismatch in Nextelt
720, coupled with the elastic modulii reported for the fibers (380
GPa for Nextelt 610 vs 260 GPa for Nextelt 720), leads to re-
sidual stresses that are more than twice as large as those
in polycrystalline alumina (Nextelt 610) cooled from a high
temperature.
Tensile strengths of coated fibers were measured on an MTS
machine (Synergie 400, MTS Systems Corp., Eden Prairie,
MN). Tows were extracted from the coated cloth and 20–50 fil-
aments from each coating condition were tested using a 25.4 mm
3
0,31
gauge length
very low. Fibers were tested following a 2-h exposure at 12001C
in air within a box furnace with MoSi elements. This heat
; fewer fibers were tested if the strengths were
1
DyPO /N610, 900°C
4
0
.9
2
DyPO /N720, 600°C
4
treatment was chosen to simulate matrix processing tempera-
tures for an alumina matrix that would likely be used with these
fibers. For control experiments, the strength of uncoated
Nextelt 610 and 720 fibers was measured following exposure
to the same 2-h 12001C heat treatment performed an coated fi-
0.8
DyPO /N720, 900°C
4
LaPO /N610,
0
0
0
.7
.6
.5
4
600°C
LaPO /N610,
4
900°C
3
2
bers. The results of this testing appear in a companion paper.
0.4
0.3
The lot-to-lot variation in the average single filament strength of
Nextelt 610 and 720 is estimated by the manufacturer to vary
by as much as 77%. In our work, the average single filament
strengths for Nextelt 610 fibers extracted from woven cloths
and heat treated for 2 h at 12001C ranged from 2.3 to 2.8 GPa
with an average of 2.6 GPa. For uncoated Nextelt 720, heat-
treated strengths ranged from 2.0 to 2.2 GPa, with an average of
LaPO /N720, 600°C
4
0
0
.2
.1
0
RE:cit:P = 1:2:1
LaPO4/N720, 600°C
100g/L, 90°C dunk
0
2
4
6
8
10
12
Number of coats
3
2
2
.1 GPa. This is consistent with the variations reported by the
Fig. 2. Effect of intercoat firing temperature and number of coats on
the normalized heat-treated strength of Nextelt 610 and 720 fibers with
LaPO and DyPO coatings.
4 4
manufacturer. The results of single-filament testing of heat-
treated monazite-coated fibers were compared with these values

July 2008
Precipitation Coating of Rare Earth Orthophosphates on Woven Ceramic Fibers
2119
1
x 900°C
2 x 900°C
6µm
600nm
6µm
600nm
6
µm
600nm
and DyPO
6µm
600nm
Fig. 3. Coating evolution during first two coats of LaPO
4
4
coatings on Nextelt 720 fibers using an intercoat firing temperature of 9001C.
Figure 2 shows the variation in heat-treated strength with
increasing number of coats and different intercoat firing tem-
peratures for Nextelt 610 and 720 fibers coated with LaPO and
along the long axis of the particle. In the case of DyPO , the
4
nuclei are round and have a surface texture that suggests a poly-
crystalline aggregate. After the second coat, the nuclei in both
cases grow while simultaneously a new generation of particles
nucleate. This new generation of particles has an appearance
similar to the first generation but a smaller size, B100 nm.
Figure 4 shows the nuclei morphology on Nextelt 720 fibers for
different rare-earth orthophosphates after one coat. The differ-
ence in nuclei morphology may be due to differences in the
solubility product constant for rhapdophane (La–Gd) and
churchite (Tb–Lu) and variation in growth rate along different
4
3
2
DyPO . As observed in previous work, the heat-treated
4
strengths decrease with increasing number of coats, and for a
fixed number of coats, coated fibers with a lower intercoat firing
temperature have higher strength. The heat-treated strengths of
DyPO -coated Nextelt 610 and Nextelt 720 fibers are much
4
higher than those of the same fibers coated with LaPO under
4
the same conditions.
4 4
Figure 3 shows the evolution of LaPO and DyPO coatings
on Nextelt 720 fibers. In the first coat, nuclei B200 nm in di-
ameter are deposited on the fiber surface. In the case of LaPO4,
these nuclei appear as barrel-shaped polycrystalline aggregates.
The nuclei exhibit stepped and nodular surfaces that are aligned
3
5,36
crystallographic axes under the conditions of precipitation.
Additional studies are necessary to understand the effect of a
rare-earth cation on nuclei morphology. Similar coating depo-
sition behavior was observed for all rare-earth orthophosphates:
6
00nm
La
600nm
Ce
600nm
Nd
600nm
Gd
600nm
Dy
600nm
Y
Fig. 4. Effect of rear earth cation on the nuclei morphology after one coat using and intercoat firing temperature at 9001C.

2
120
Journal of the American Ceramic Society—Fair et al.
Vol. 91, No. 7
1st dunk, rinse:
Precipitation
of particles on
and near fiber;
loose particles
rinsed away
1st drying,
firing:
Particles
adhere to
fiber surface
2nd dunk, rinse:
Particles from 1st
dunk grow; new
particles form on
and near fiber;
loose particles
rinsed away
2nd drying,
firing:
Particles
adhere to
fiber surface,
each other
3rd dunk, rinse:
Particles from
previous dunks
grow; new
particles form on
and near fiber;
loose particles
rinsed away, etc.
Fig. 5. Proposed mechanism by which coatings are deposited on fibers using the coating procedure described herein.
the initial nuclei grow during the second coat and a new gener-
ation of smaller nuclei appears. Based on these observations, it is
proposed that the mechanism by which uniform coatings form is
through simultaneous nucleation on bare fiber surfaces and on
existing rare-earth orthophosphate particles during each sequen-
tial coating, as shown schematically in Fig. 5.
Figure 6 shows representative TEM micrographs of heat-
treated LaPO - and DyPO -coated Nextelt 720 fibers applied
4
4
using an intercoat firing temperature of 6001C; samples coated
five times with DyPO and 10 times with LaPO are shown.
4
4
Despite the differing number of coats applied to the samples,
the differences in coating microstructure are clear. The DyPO4
LaPO4
Fiber
DyPO4
Fiber
200nm
LaPO4
DyPO4
Fiber
Fiber
20nm
4 4
Fig. 6. Representative transmission electron spectroscopy micrographs of LaPO and DyPO coatings on Nextelt 720 fibers.

July 2008
Precipitation Coating of Rare Earth Orthophosphates on Woven Ceramic Fibers
2121
a
b
c
d
6
µm
6 µm
6 µm
6 µm
1
µm
1 µm
1 µm
1 µm
Fig. 7. Representative scanning electron microscopy micrographs of LaPO
c) and (d) YPO
4
, DyPO
4
, and YPO coatings on Nextelt 610 fibers. (a) LaPO , (b) DyPO ,
4 4
4
(
4
.
coatings are porous with both inter- and intragranular pores. In
contrast, the LaPO coatings are hermetic and almost fully
an RE:citrate:P ratio of the solutions set at 1:5:5. The hydrated
REPO powders were collected by centrifugation and washed
4
4
dense, with occasional intergranular pores. In both cases, the
coating fiber interface is free of second phases, indicative of
coating nonstoichiometry such as AlPO . Figure 7 shows SEM
4
once with deionized water before drying and TGA analysis.
High-temperature weight loss, i.e., above 9001C, was observed
for all powders; the amount decreases as follows: La4
Dy4Y4Gd4Tb.
micrographs of the coated fiber surfaces following heat treat-
ment. For LaPO , the coating is dense with very few intergran-
The densification behavior of citrate-derived REPO powders
4
4
ular pores. For DyPO
the grains on the surface of the coating are rounded. YPO
coatings had an appearance similar to DyPO coatings, but
4
, many intergranular pores are visible and
has been studied previously for pellets of REPO
tered for 2–3 h at different temperatures. The data at 12001C
appear in Fig. 9. It shows the percentage of theoretical density
4
powders sin-
3
8,39
4
4
some parts of the fiber surface were uncoated, which may
contribute to the high strengths of these coated fibers.
The coating microstructure observations suggest that coated
fiber strength following heat treatment may correlate with dens-
ification of the coating. To investigate this correlation, the
for different REPO powders and is compared with our mea-
4
surements for a 2-h heat treatment at 12001C. The results are
consistent with previous work and show a pronounced decrease
in sintered density for xenotimes near the middle of the lan-
thanide series in comparison with monazite. The xenotime coat-
ings studied here are more porous than monazite coatings due to
intrinsically slower densification rates. The increased porosity
may contribute to higher coated fiber strengths by allowing
high-temperature precursor decomposition products that cause
weight loss and densification of citrate-derived REPO
4
powders
was studied. Figure 8 shows TGA data for citrate-derived
REPO powders where RE5 La, Gd, Tb, Dy, and Y ; the
3
7
4
powders were precipitated by warming precursors to 301C using
1
5
0
5
0
100
RE:Citrate:P = 1:5:5
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
La > Dy > Y > Gd > Tb
1
La
Gd
Tb
Dy
Y
La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu
REPO (~Ionic radius)
0
200
400
600
800
1000
1200
1400
4
Temperature (°C)
Fig. 9. Measured density following 12001C heat treatment for citrate-
34,35
powders. Blue bars 5 data from Hikichi et al.,
Fig. 8. Thermogravimetric analysis curves illustrating the effect of rare
earth cation on outgassing behavior of citrate-derived REPO powders.
derived REPO
4
red
4
bars 5 measured in this study.

2
122
Journal of the American Ceramic Society—Fair et al.
Vol. 91, No. 7
2
U. Kolitsch and D. Holtstam, ‘‘Crystal Chemistry of REEXO
4
Compounds
stress corrosion to escape without being trapped near the fiber
surface.
(X5 P, As, V). II. Review of REEXO
Eur. J. Mineral., 16 [1] 117–26 (2004).
4
Compounds and Their Stability Fields,’’
The collective observations suggest that coated fiber strength
depends on both the high-temperature weight loss from the
3
L. A. Boatner, ‘‘Monazite’’; pp. 495–564, in Radioactive Waste Forms for the
Future, Edited by W. Lutze and R. C. Ewing. Elsevier North Holland, Amsterdam,
The Netherlands, 1988.
1
9–21
coatings
and the tendency of the coating to form closed
4
P. E. D. Morgan, R. M. Housley, J. B. Davis, M. L. DeHaan, and D. B.
pores near the fiber surface, which is related to densification
rates. More high-temperature weight loss and faster densificat-
ion (e.g., LaPO ) lead to severely degraded fiber strength after
4
Marshall, ‘‘Chemical and Ceramic Methods Towards Safe Storage of Ac-
tinides’’, Final Technical Report for Department of Energy Contract
DE-FG07-96ER45617, http://www.osti.gov/bridge/servlets/purl/850364-pV1esX/
8
50364.PDF
P. E. D. Morgan and D. B. Marshall, ‘‘Functional Interfaces for Oxide/Oxide
heat treatment at matrix processing temperatures. Less high-
temperature weight loss, coupled with poor densification, yields
a more porous coating (e.g., DyPO and YPO ) and higher
5
Composites,’’ Mat. Sci. Eng., A126, 15–26 (1993).
P. E. D. Morgan and D. B. Marshall, ‘‘Ceramic Composites of Monazite and
6
4
4
coated fiber strengths; here, the precursor decomposition prod-
ucts that do remain at a high temperature can escape and are not
trapped near the fiber surface. GdPO4 and TbPO4 powders
showed the smallest high-temperature weight loss, but these
Alumina,’’ J. Am. Ceram. Soc., 78 [6] 1553–63 (1995).
J. B. Davis, D. B. Marshall, and P. E. D. Morgan, ‘‘Oxide Composites of Al
7
2 3
O
and LaPO ,’’ J. Eur. Ceram. Soc., 19, 2421–6 (1999).
4
8
M. H. Lewis, A. Tye, E. Butler, and I. Al-Dawery, ‘‘Development of Interfaces
in Oxide Matrix Composites,’’ Key Eng. Mat., 164–165, 351–6 (1999).
9
powders densified faster than DyPO
4
higher weight loss than GdPO and TbPO
4
and YPO
4
. Despite a
- and
M. H. Lewis, A. Tye, E. G. Butler, and P. A. Doleman, ‘‘Oxide CMCs: In-
terphase Synthesis and Novel Fibre Development,’’ J. Eur. Ceram. Soc., 20,
4
, DyPO
4
6
39–44 (2000).
J. B. Davis, D. B. Marshall, and P. E. D. Morgan, ‘‘Monazite-Containing
YPO -coated fibers are suggested to have higher coated fiber
4
strength because their poor densification allows surface-active
10
Oxide/Oxide Composites,’’ J. Eur. Ceram. Soc., 20, 583–7 (2000).
K. A. Keller, T. Mah, T. A. Parthasarathy, E. E. Boakye, P. Mogilevsky, and
11
species that cause stress corrosion to escape. GdPO
4
showed
densification behavior similar to LaPO , but has less high-
M. K. Cinibulk, ‘‘Effectiveness of Monazite Coatings in Oxide/Oxide Composites
after Long-Term Exposure at High Temperature,’’ J. Am. Ceram. Soc., 86 [2] 325–
4
temperature weight loss, and so GdPO -coated fibers were
4
3
2 (2003).
12
stronger than LaPO -coated fibers. Similar results were found
4
P. Mogilevsky, ‘‘On the Miscibility Gap in Monazite-Xenotime Systems,’’
Phys. Chem. Minerals, 34, 201–14 (2007).
4
in a recent work by Boakye et al. , which showed inhibition of
0
13
P. Mogilevsky, E. E. Boakye, and R. S. Hay, ‘‘Solid Solubility and Thermal
Expansion in a LaPO –YPO System,’’ J. Am. Ceram. Soc., 90 [6] 1899–907
2007).
densification in mixed (Y,La)PO coatings on Nextelt 720 and
4
4
4
correspondingly higher coated fiber strengths following heat
treatment despite having a high-temperature weight loss com-
parable to LaPO coatings.
(
14
P. Mogilevsky, E. B. Zaretsky, T. A. Parthasarathy, and F. Meisenkothen,
‘‘Composition, Lattice Parameters, and Room Temperature Elastic Constants of
4
Natural Single Crystal Xenotime from Novo Horizonte,’’ Phys. Chem. Minerals,
3
The results in this work are consistent with high-temperature
stress corrosion as a degradation mechanism—surface-active de-
composition products from the coating precursor attack
strength-limiting flaws along grain boundaries under a high re-
sidual stress. The mechanism requires (1) residual stresses in the
fiber, (2) high-temperature weight loss from the coating, and (3)
trapped porosity next to the fiber surface. Operation of this
strength degradation mechanism implies: (1) improved strength
with a lower residual stresses in the fiber (higher strengths with
Nextelt 610 vs Nextelt 720), (2) improved strength with less
3, 691–8 (2006).
D. Kou and W. M. Kriven, ‘‘A Stong Damage Tolerant Oxide Laminate,’’
J. Am. Ceram. Soc., 80 [9] 2421–4 (1997).
15
16
G. E. Fair, R. S. Hay, E. E. Boakye, T. A. Parthasarathy, and K. A. Keller,
‘Effect of Rare Earth Cation of Push-Out Behavior of REPO -Coated Sapphire
Fibers’’; Work in progress
‘
4
17
R. S. Hay, ‘‘Method for Coating Continuous Tows’’; U.S. Patent No.
,164,229, Nov. 17, 1992.
E. Boakye, R. S. Hay, and M. D. Petry, ‘‘Continuous Coating of Oxide Fiber
5
18
Tows Using Liquid Precursors: Monazite Coatings on Nextelt 720,’’ J. Am.
Ceram. Soc., 82 [9] 2321–31 (1999).
R. S. Hay, E. Boakye, and M. D. Petry, ‘‘Effect of Coating Deposition Tem-
19
4 4
high-temperature weight loss (GdPO vs LaPO ), and 3) im-
perature on Monazite Coated Fiber,’’ J. Eur. Ceram. Soc., 20, 589–97 (2000).
E. E. Boakye, R. S. Hay, P. Mogilevsky, and L. M. Douglas, ‘‘Monazite
20
proved strength in coatings with continuous porosity (DyPO vs
4
Coatings on Fibers: I, Effect of Temperature and Alumina Doping on Coated-
Fiber Strength,’’ J. Am. Ceram. Soc., 84 [12] 2783–92 (2001).
E. E. Boakye, R. S. Hay, P. Mogilevsky, and L. M. Douglas, ‘‘Monazite
4
LaPO ). Future work should concentrate on defining the surface-
active chemical species responsible for high-temperature stress
corrosion. With this knowledge, it may be possible to develop
coating precursors that do not degrade coated fiber strength.
21
Coatings on Fibers: II, Coating without Strength Degradation,’’ J. Am. Ceram.
Soc., 84 [12] 2793–801 (2001).
E. E. Boakye, T. Mah, C. M. Cooke, K. Keller, and R. J. Kerans, ‘‘Initial
22
Assessment of the Weavability of Monazite-Coated Oxide Fibers,’’ J. Am. Ceram.
Soc., 87 [9] 1775–8 (2004).
T. Yano, P. Lee, and M. Imai, ‘‘In-Situ Reaction Deposition Coating of
IV. Conclusions
23
LaPO
4 2 3 2 3 2 3
on Al O Fabric Cloth for Al O /Al O Composites,’’ Ceram. Sci. Eng.
Various rare-earth orthophosphates, both monazite and xeno-
time, were applied as coatings to Nextelt 610 and 720 fibers by
heterogeneous precipitation. The extent of strength degradation
during fiber coating was assessed following a high-temperature
heat treatment. Higher coated fiber strengths were obtained for
Nextelt 610 fibers due to lower thermal residual stresses. Higher
coated fiber strengths were obtained for both types of fibers when
coatings had lower high-temperature weight loss and slower
densification rates. These results are consistent with a stress
corrosion mechanism for strength degradation, where the sur-
face-active species responsible for stress corrosion arises from
high-temperature decomposition products of the coating pre-
cursor, which are trapped at the fiber–coating interface in coat-
ings that densify rapidly.
Proc., 25 [4] 93–8 (2004).
24
J. H. Weaver, J. Yang, C. G. Levi, F. W. Zok, and J. B. Davis, ‘‘A Method
for Coating Fibers in Oxide Composites,’’ J. Am. Ceram. Soc., 90 [4] 1331–3
(2007).
25
G. E. Fair, R. S. Hay, and E. E. Boakye, ‘‘Precipitation Coating of Monazite
on Woven Ceramic Fibers: I. Feasibility,’’ J. Am. Ceram. Soc., 90 [2] 448–55
2007).
(
26
S. Lucas, E. Champion, D. Bregiroux, D. Bernache-Assollant, and F. Audu-
bert, ‘‘Rare Erath Phosphate Powders RePO O (Re5 La, Ce, or Y)—Part I.
Á nH
Synthesis and Characterization,’’ J. Solid State Chem., 177, 1302–11 (2004).
4
2
27
G. F. Claringbull and M. H. Hey, ‘‘A Re-Examination of Churchite,’’ Min-
eral. Mag., 30, 211–7 (1953).
M. K. Cinibulk, J. R. Welch, and R. S. Hay, ‘‘Preparation of Thin Sections
28
of Coated Fibers for Characterization by Transmission Electron Microscopy,’’
J. Am. Ceram. Soc., 79 [9] 2481–4 (1996).
M. K. Cinibulk, J. R. Welch, and R. S. Hay, ‘‘Transmission Electron Mi-
29
croscopy Specimen Preparation of Ceramic Coatings on Ceramic Fibers,’’ Mater.
Res. Soc. Symp. Proc., 480, 3–17 (1997).
M. D. Petry, T. Mah, and R. J. Kerans, ‘‘Validity of Using Average Diameter
30
Acknowledgments
for Determination of Tensile Strength and Weibull Modulus of Ceramic Fila-
ments,’’ J. Am. Ceram. Soc., 80 [10] 2741–4 (1997).
31
The authors would like to thank Kristen Davis and Eric Urban for assistance
with coating and mechanical testing of the fibers used in this work and Marlin
Cook for his assistance with the preparation of microscopy samples.
M. D. Petry and T.-I. Mah, ‘‘Effect of Thermal Exposures on the
Strength of Nextel 550 and 720 Filaments,’’ J. Am. Ceram. Soc., 82 [10] 2801–7
(1999).
32
G. E. Fair, R. S. Hay, and E. E. Boakye, ‘‘Precipitation Coating of Monazite
on Woven Ceramic Fibers: II. Effect of Processing Conditions on Coating Mor-
phology and Strength Retention of Nextelt 610 and 720 Fibers,’’ J. Am. Ceram.
Soc. in press (2008).
A. Zimmerman, E. R. Fuller, and J. Rodel, ‘‘Residual Stress Distributions in
Ceramics,’’ J. Am. Ceram. Soc., 82 [11] 3155–60 (1999).
References
1
33
Y. Ni, J. M. Hughes, and A. N. Mariano, ‘‘Crystal Chemistry of the Monazite
and Xenotime Sturctures,’’ Am. Mineralogist, 80, 21–6 (1995).

July 2008
Precipitation Coating of Rare Earth Orthophosphates on Woven Ceramic Fibers
2123
3
4
V. R. Vedula, S. J. Glass, D. M. Saylor, G. S. Rohrer, W. C. Carter, S. A.
Langer, and E. R. Fuller, ‘‘Residual-Stress Predictions in Polycrystalline
Alumina,’’ J. Am. Ceram. Soc., 84 [12] 2947–54 (2001).
Tb, Dy, Y) and Solid Solutions,’’ J. Am. Ceram. Soc., in preparation for
submission.
Y. Hikichi, T. Ota, K. Daimon, T. Hattori, and M. Mizuno, ‘‘Thermal, Me-
38
35
E. E. Boakye, P. Mogilevsky, and R. S. Hay, ‘‘Synthesis of Nanosized Spher-
ical Rhabdophane Particles,’’ J. Am Ceram. Soc., 88 [10] 2740–6 (2005).
chanical, and Chemical Properties of Sintered Xenotime-Type RPO4 (R5 Y, Er,
Yb, or Lu),’’ J. Am. Ceram. Soc., 81 [8] 2216–8 (1998).
Y. Hikichi, T. Ota, and T. Hattori, ‘‘Thermal, Mechanical, and Chemical Prop-
36
39
X. Liu and R. H. Byrne, ‘‘Rare Earth and Yttrium Phosphate Solubilities
in Aqueous Solutions,’’ Geochimica et Cosmochimica Acta, 61 [8] 1625–33
1997).
erties of Sintered Monazite—(La, Ce, Nd, or Sm),’’ Miner. J., 19, 123–30 (1997).
E. E. Boakye, R. S. Hay, P. Mogilevsky, and M. K. Cinibulk, ‘‘Two Phase
40
(
37
E. E. Boakye, P. Mogilevsky, G. E. Fair, and R. S. Hay, ‘‘Synthesis and
Phase Composition of Lanthanide Phosphate Particles LnPO (Ln 5 La, Gd,
Monazite/Xenotime 30LaPO –70YPO
4
Ceram. Soc., 91 [1] 17–25 (2008).
4
Coating of Ceramic Fiber Tows,’’ J. Am.
&
4