Note: Descriptions are shown in the official language in which they were submitted.
~ 2~2~323
O.Z. 4425
Zirconium dioxide powder, ~roce~ses for it8 preparation,
it~ use and sintered articles produced therefrom
Structural materials which are sub~ected to high mechan-
ical and~or thermal stres~es are increasingly being
S produced from ceramic materials. ~ecause of it~ hard-
ness, its toughne~s, it~ low thermal conductivity and its
chemical-resista~ce, zirconium dioxide $8 such a ceramic.
The ZrO2 powder used for the production of sintered
zirconium dioxide articles i8 generally used in a form
which~ parti-al~y or completely stabilized with other
oxides, such as, for example Y203, CeO2 (or mixtures of
rare earth~), CaO and~or MgO. In order for the green
compacts produced from doped ZrO2, for example by compres-
sion or ~lip cagting, to have the desired sinter proper-
ties and the sintered moulding6 to have the required goodmechanical and thermal properties, it is necessary for
the oxided metered in to be distributed as uniformly as
possible in the 8rO2 lattice. Furthermore, the powder
should be free-flowing and the powder particles should
con~i~t of loose agglomerate~.
For the preparation of ZrO2 powder~, the starting mater-
ials used are baddeleyite ~impure ZrO2) or zircon sands
(ZrO2.SiO2), which are worked up on a lar~e industrial
scale by two routes.
In the first method, the ~tarting materials are d~gested
in alkali at high temperatures and the compounds obtained
are hydrolysed, the resulting hydrated zirconium hydrox-
ide (or hydrated oxide) i~ dissolved again in ~ulphuric
acid and precipitated with ammonia as the basic sulphate
or hydroxide (or hydrated oxide) and finally calcined.
These processes are expensive, and the resulting powder~
have relatively large crystallites which have combined to
form hard agglomerates, 80 that these powders are dif-
ficult to process into high-density sintered articles.
In the second route, the zircon sand i8 reacted with
.. . ..
2 0 2 6 ~ 7 3
- 2 - O.Z. 4425
chlorine in the presence of carbon to qiv~ zirconlum
tetrachloride, which i5 hydrolysed to zlrconyl chloride
(zrOCl2.8H20). Zirconium hydroxide (or hydrated zircon-
ium oxide) is precipitated from the aqueous zirconyl
S chloride ~olutions with the aid of ammonia or ammonia
donors and is washed several times and then calcined.
The di~advantage of this process i8 that the resulting
hydroxide precipitates are difficult to filter, the
stabilizer is not uniformly di~tributed and the calcined
products are hard agglomerates.
~etter products are obtained if the precipitated zir-
conium hydroxide i~ partially dehydrated by azeotropic
distillation prior to calcination, and the precipitate i8
additionally rendered chloride-free by repeated dissolu-
tion in sulphuric acid and further precipitation withammonia. The hydroxide is converted with citric acid
into the corresponding complex and the latter i8
dehydrated by azeotropic distillation and finally
calcined. These proce~se~ are labour-intensive and cause
environmental pollution owing to the oxides of nitrogen.
European Patent 0,251,538 describes a process in which an
aqueous solution of zirconyl chloride is heatod for a
prolonged period at temperatures below the boiling point
of water. The zirconiwm hydroxide formed i8 separated
off from the solution, washed and calcined. Isolation
and washing of the ~ery fine hydroxide precipitate are
very difficult, and the ZrOz obtained is not stabilized.
In order to ~tabilize it, the particles must be resuspen-
i ded, laden with the hydroxide of the stabi1$~er by30 alkaline precipitation, washed, filtered and calcined
again. These additional proce~s steps make the process
even more difficult.
Earlier literature, for ex~mple the works cited in
Gmelins Handbuch der anorganischen Chemie [Gmelins
Handbook of Inorganic Chemlstry], 8th edition, Volume 42
(zirconium)~ pages 303 to 306, discloses that ~ol~d
' :. ' ' " ' . " ' '
.:,: j.: . , " . : . :
`.. ,~ : . .' : ' . ~ '. - :
202~!~23
,
- 3 - O.Z. 442S
ZrOCl2.8H2O can be converted into ZrO2 by vigorous heating
with liberation of H2O, HCl and possibly ZrCl~. This
method has not been developed into an industrial process
to date because the resulting product~ are not high
quality powders and subsequent satisfactory stabilization
of the zro2 is not possible.
It was therefore the ob~ect of the present invention to
provide an unstabilized or stabilized microcry~talline
zirconium dioxide powder which i8 present in the form of
soft agqlomerates, contains any stabilizers pr~sent in
homogeneous distribution and has good flow, compre~sion,
spray and ~inter propertie~, and to develop a process for
the preparation of such a powder, avoiding aqueow ~olu-
tions and the removal of precipitate3.
This ob~ect is achieved, according to the invention, by
a method in which hydrated zirconyl chloride (ZrOCl2.8N2O)
and optionally a stabilizer or its precursor ar~ melted,
gaseous ammonia is passed into the melt, the a~ ionium
chloride formed is su~l~med at elevated temperature from
the melt evaporated to dryne~s, and the residue i8 cal-
c$ned at high temperatures and optionally milled.
In the re~ction of the ammonia with ~he zirconyl chlor-
ide, ammonium chloride and hydrated oxides of zirconium,
wh~ch are not i3yve~tigated further, are formed. By
incorporating the ammonium chloride in the hydrated zir-
conium oxide, pronounced agglomeration of the hydrated
zirconium oxide i~ prevented during the subsequent
dehydration ~tep. The chlorine originally pre~ent in the
zirconyl chloride is removed from the reaction product in
the form of the ammonium chloride formed, by subliming
the ammonium chloride at elevated temperature. According
to the literature, the sublimation temperature of ammon-
ium chloride is 611 K.
The ammonia ga~ can be combined with the- melt by the
method~ known in industry, such a~, for example, pa~sing
ii :
. i............................... . . .
~`` 2 Q ~
- 4 - O.Z. 4425
it into the melt by mean~ of an inlet tubeJ forcing the
ammonia onto the moving melt or spraying the melt into an
ammonia atmosphere with the aid of a nozzle.
The molar ratio of zirconyl chloride to ammonia may vary
S between 10 : 1 and 1 5 2. When the molar ratio of zir-
conyl chloride to ammonia i~ 1 s 2, the total chloride of
the zirconyl chloride i~ b~und stoichiometrically as
ammonium chloride. If the lar ratio i~ le~s than 1 s
2, only ~ome of the chloride i8 bound as ammonium chlor-
ide and the remainder must be-removed from the reaction
product together with th~ water, as hydrogen chloride,
during evaporation of the meLt. The partial conversion
of the chloride to---ammonium chloride has the advantage
that the melt remains stirrable while the hmmonia is
being pas3ed in. In this case, molar ratios of zirconyl
chloride to ammonia of between 5 5 1 and 1 5 1 ~ par-
ticularly those of 2 5 1 to 1 : 1, are preferred. The
other preferred operating range i8 complet~ conversion of
the chloride present in the zirconyl chloride into
ammon~um chloride. In this-process variant, stirring of
the melt is effected as a function of the viwo~ity of
the melt, or the melt is agitated by rotation of the
reactor--containing the melt. ~ven when the melt is
sprayed into -an ammonium atmo~phere, as a rule complete
conversion of chloride into a~monium chloride i8 desir-
abl~. The reaction of the zirconyl chloride with ammonia
in a ratio of about 1 s 2 ha~ the advantage that, during
evaporation of the reaction product, chloride-free water
distils off and there are fewer corrosion problems. The
disadvantage is that relatively large amounts of ammonium
chloride have to be sublimed. The proces~ variant used
also depends on, inter alia, the possibilities for dis~
posing of hydrochloric acid and ammonium chloride. In
principle, the ammonia can also be circulated by reacting
the ammonium chloride with, for example, sodium hydroxide
solution with formation of ammonia and sodium chloride
and recycling the resulting ammonia to the zirconyl
chloride melt.
,~," ,.. , ~.,., .. .. , .. ; ~ ... . . . . . . . . .. . .
,?,,,6,~,
-
_ 5 _ o.z. 4425
A~i a rule, the zirconium dioxide according to the inven-
tion or prepared according to the invention i8 more or
lessi ~itabilized (partial or complete stabilization).
Yttrium oxide has proved to be a -~uitable 3tabilizer.
The yttrium can be usied in the form of it~ oxide or of
another compound which is converted into the oxide form
(precursor) under the calcination conditions. Examples
of suitable compounds are hydroxides, halides, organic
6alts and organic complexes of yttrium. Yttrium chloride
0 i8 preferred since the starting compound for the ZrO2
- is also used in the chloride form.
For partially stabilized ZrO2, the amount of stabilizer
(yttrium oxide) is as a rule up to 7 percent by weiqht,
in particular between 0.1 and 6 percent by weight and
very particularly between 2 and 5 percent by weight,
based on the prepared powder. For completely stabilized
ZrO2, on the other hand, the amounts are between 7 and 15
percent by weight, preferably ~etween 8 and 10 percent by
weight, the boun~aries between complete and partial
stabilization being fluid and depending on the other pow-
der properties and the sinter conditions.
After melting of the salt~i and reaction with the ammonia
ga~, which as a rule i8 carried out with refluxing of the
w~ter of crystallization, the water of crystallization
and, depending on the extent of reaction with ammonia,
siome of the chloride present in the zirconyl chloride are
distllled off in the form of hydrogen chloride at temp-
erature~ between 410 g and 480 g. The ammonium chloride
is then sublimed from the rema~ning powder at elevated
temperature (sublimation temperature of ammoniumi chlor-
ides 611 R). Furthermore, sublimation of the ammonium
- chloride may be accelerated by a qentle ga~ stream, air
beinq preferred. The amounts of gasi ~ihould be kept as
low a~i poeisiible since it is difficult to separate off the
ammonium chloride from relatively large amounts of gas.
Flow velocities of 0.001 to 0.1 m/i3 are preferred.
: ~. . . ;
.: '; ~ , : .
~ . .
~, , .
.:
2 ~ 3
- 6 - O.Z. 4425
After sublimation, the remaining product i9 calclned at
high temperatures. The calcination temperature may vary
within wide limit~, depending on the de~ired crystallite
size of the ZrO2 and the hardne~s of the agglomerates.
Since the ammonium chloride must be sublimed as complete-
ly as possible, the lower temperature limit ~hould be
> 611 K. The upper calcination temperature is determined
by the sinter proces~ which takes place. In general,
calcination is carried out at temperatures of 780 to
1,280 R. Temperature~ between 880 and 1,080 R are
preferred. The calcination time depends on the calcina-
tion temperature and the des~red powder properties. As
a rule, it i~ 0.7 to 90 ks, times between 1.5 and 10 k~
~ein~ preferred. The powder obtained after calcination
is brought to the desired particle size, if neces3ary by
milling, which may be effected in the dry state or in
liquid media, such a~, for example, water or alcohols,
and optionally sieving.
The process according to the invention can be carried out
a~ a sinqle-sta~e or two-stage process. In the ~ngle~
stage process, meltinq of the starting salts, reaction
with ammonia, di~tilling off the water and, if approp-
riate, the hydrochloride, sublimation of the ammonium
chloride and calc~nation of the remaininq product are
carried out in one apparatus, and the temperature iB con-
trolled by a corresponding temperature programme. The
two-stage process is carried out in two different reac-
tors. In the first reactor, the salts are melted, the
reaction with ammonia is carried out and the reaction
product is evaporated. In the second stage, the residue,
if appropriate after an intermediate milling step, is
freed from the ammonium chloride and calcined in B fur-
nace, for example a rotary kiln. Sublimation of the
ammonium chloride can, however, also be effected in B
~eparate step. -
The powders according to the invention or prepared
accordinq to the invention have a loose agglomerate
2~?~ 3?,3
- 7 - O.Z. 4425
structure with a bimodal pore di~tribution. The crystal-
lite sizes are ~etween 5 and 40 nm, and some crystallites
may even have combined to form larger crystallites. The
powders can be readily compressed to give high-density
green compacts and lead to pore-free sintered articles
which have the desired mechanical and thermal properties.
The invention i8 illustrated in more detail by the
Examples below. ~he abbreviations and measuring and test
methods used in the Examples ares
EDX
The distribution of elements in the samples was deter-
mined using a commercial EDAX-apparatus ( Type: EDAX
9900), connected to a commercial scanning electron micro-
scope, by the method of energy-dispersive X-ray analysi~
(EDX). The resolution was about 25 nm.
SEM
Commercial scanning electron microscope
STEM
Commercial scanning transmission electron microscope
Pore st Ncture distribution
The pore structure distributlon was investigated using a
commercial high-pressure Hg porosimeter from Carlo Erba.
Surface area
The surface area of the powders was determined using a
co ercial app~ratus based on the BET (~runauer-Emmett-
Teller) method (N2) and a commercial Hg porosimeter from
Carlo Erba.
Crystal phase
The crygtal structure was determined by X-ray diffraction
analysis using a commercial apparatus.
.
... .
., -
; - . .
.
, . ~,, ,,, , ~
2~2~23
- 8 - O.Z. 4425
Crystal diameter
The diameter of the crystallite~ wa~ obtained by measur-
ing the crystallites in the scanning transmission elec-
tron micrographs and from the individual peaks of the
X-ray diffraction patterns. The diffractometer used was
a commercial apparatus from Philips (Types PW 1800).
Sinter behaviour
The kinetic~ of sintering of the s2mples (change in
length as a function of the temperature) were monitored
using a commercial dilatometer from Baehr.
Chlorine content
The chlorine content of the samples was determined by the
X-ray fluorescence method usin~ a commercial apparatus.
Hardne~s of the aa~lomerates
lS Since there is no generally customary method for deter-
mining the hardness of the agglomerates, it was deter-
mined qualitatively - and only for the products in
relation to one another - by grinding the powders between
two glass discs with the fingers.
Vicker's X~rdness -~
The hardness of the sintered samples was determined by
the ~icker's method (~IN 50,351).
Example 1
200 g of ZrOCl2.8H20 (0.62 mole) and 8.2 g (0.027 mole) of
~ YC13- 6H20 were homogenized in a mortar and melted in a
glass flask at 413 R. About lS dm3 (S.T.P.) of NH3 were
passed into the melt while stirring and refluxinq the -~
evaporating water. While a~monia wa~ be~ng passed in,
the ~iscosity of the melt initially decreased slightly
and then increased. After about lS dm3 ~S.T.P.) of NH3
stirring was stopped and a further 15 dm3 ~S-T.P-) of NH3 -
(total amount of NH3~ > 1.3 moles) were passed in. After
the end of the introduction of NH3, the reaction product
; ' i :' . . . ' .. : ; - ~ .
` 2a2~23
_ g _ O.Z. 4425
was evaporated to dryness by increasing the temperature
of the external heating bath to 473 R and evaporating off
the water. After the temperature of 473 K had been
reached, the evaporation process was accelerated by
applying a vacuum of 50 hPa. The pH of the water
distilled off was 9.3. The evaporation residue was
milled for 1.8 ks in a ball mill and then calcined in a
tubular furnace in a gentle stream of air (about 35
cm3/~ ) . During this procedure, the temperature was in-
creased from 473 R to 1,073 R at 10 R/60 s and was kept
at 1,073 K for 3.6 ks. During heating, the ammonium
chloride sublimed. The calcined powder, in which no
chlorine was detectable, had no hard agglomerates, and
furthermors no relatively large agglomerate~ were detect-
able in the scanning electron micrographs and scanning
transmission electronic micrograph~. It had a bimodal
pore distribution and consisted of 88% of the tetragonal
phase and 12~ of th~ monoclinic phase. The crystallite
size was about 40 nm and the surface area was found to be
17 m2/g both by the BET method and by the Hg porosimetry
method. No inhomogeneities in the yttrium di~tribution
were found by the ST~M and EDX methods. In the dilat-
ometer, tablets produced ~rom the powder showed maximum
sintering at 1,500 ~ and a change in length of 15.5~.
; 25 Tablets sintered at 1,830 K showed no pores in the
~icrograph of a ground section (diamond powder) and had
a Vicker's hardness of 12.0 GPa.
~xamples 2 and ~
The experiment of Example 1 was repeated in two further
experiments with different amounts of ammonias 1.4 dm3
(S.T.P.) and 9.7 dm3 (S.T.P.). The calcination tempera-
ture was 927 K. The calcined and milled powders had a
bimodal pore distribution and had a surface area of 50
and 54 m2/g, respectively, according to the BET method
and a surface area of 20 and 32 m2/g, respectively,
according to the Hg poroeimetry method. The crystallite
size, determined by X-ray diffraction analy~is, was 14
. .~ . .,. ~, .
, .. ~ . . . , ;
2~2~23
- 10 - O.z. 4425
and 16 nm, respectively, for the tetragonal phase, and
the ratio of the tetragonal phase to the monoclinic phase
was found to be 94 s 6 and 88 : 12, respectively. In the
dilatogram, tablets produced from the powder showed
maximum sintering at 1,492 and 1,517 K, respectively. In
the micrographs of ground sections of the tablets sinter-
ed at 1,800 K for 3 and 7 k~, no pores were detectable,
and the Vicker'3 hardne~s was 10.5 and 11.1 to 12.0 GPa,
respectively.
Example 4
The experiment described in Example 3 was repeated,
except that, after the addition of ammonia, the mixture
was kept at the temperature of 413 R for a further 1.2 ks
with refluxing of the hydrochloric acid distilled off.
In the dilatometer, tablets compre~sed from the calcined
and milled powder under a pressure of 760 MPa showed ~-
maximum sintering at 1,470 R. Tablets which had been
sintered for 1.2 ks at 1,830 X were pore-free and had a
Vicker'~ hardness of 10.8 GPa.
Examples 5 and 6
: .~
58.1 q of N~Cl were dissolved 5 n a melt of 500 g of
ZrOCl2.8H20 ~nd 20.5 g of YCl3.6H20 at a bath temperature
of 413 R, and the melt was dried by increasing the
temperature to 470 R and, tow~rds the end of the process,
by applying a vacuum of 54 hPa. Some of the residue w~s
comminuted in the flask and treated with ammonia gas for
1.8 ks at 470 R and then for S0 ks at room temperature
and finally calcined at 920 g (3.6 ks). The other part
of the evaporation res~due was directly calcined at
920 R (likewise for 3.~ ks) and then treated with ammonia
gas for 3.6 k~i at thi~i temperature. In the dilatogram,
tablets produced from the two powder~ i3howed maximum
sintering at 1,520 and 1,550 R, respectively, and tablets
which had been sintered at 1,840 R (8 ks) had Vicker~s
hardnesses of 10.2 and 11.2 GPa, re~ipectively.
: ", . . . ~ .
2 ~ 2 ~
` - 11 - O.Z. 4425
Comparative Example
100 g of ZrOCl2.8H20 and 2.4 g of YCl3.6~0 were melted
without further additive3 and were proce~sed to a powder
as in Example 1. The product consisted of hard agg-
S lomerates having a diameter of 0.5 to 1 ~m, which couldnot be compressed and sintered to give high-density
mouldings.
,.'., "' , " - . ' '
: ., ~ .; .
., .. , ~:~ . , .
