Note: Descriptions are shown in the official language in which they were submitted.
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
AI203/SiC Nanocomposite Abrasive Grains, Method for Producing Them,
and Their Use
The present invention relates to sintered A1203/SiC nanocomposite abrasive
grains as defined in the preamble to Patent Claim 11, a method for
producing these as set out in the preamble to Patent Claim 1, as well as the
use thereof as grinding agents.
Because of their great hardness, chemical inertness, and resistance to high
temperatures, large quantities of abrasive grains that are based on AI203 are
processed industrially to form grinding abrasives. In addition to fused
corundum, which can be manufactured in a relatively cost-effective manner
in arc furnaces, more recently reinforced sintered corundums, which can be
obtained by way of ceramic or chemical methods, have been used for
specific applications. From the standpoint of abrasive engineering practice,
the advantages of sintered corundums are determined by their
microcrystalline structure that, in its turn leads, to a particular wear
mechanism of the abrasive during the grinding process. Removal
performance can be greatly enhanced by using sintered corundums, mainly
in applications that require high contact pressures, for example , when
1
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
machining special steels, hardened steels, or alloys that are difficult to
machine.
The grain of sintered corundum, which is of a microcrystalline structure, is
considerably more resistant to wear for these applications than the structure
of fused corundum, which is macrocrystalline. In addition, there is the fact
that during the grinding process, when using microcrystalline corundum,
smaller areas break away from the grains so that new cutting edges are
formed, and these-in their turn-take part in the grinding process. No such
self-sharpening of the grains takes place in the case of macrocrystalline
fused corundum because here the cracks that result during the grinding
process because of the mechanical stresses on the grains can no longer be
deflected, but continue along the crystal planes of the overall grains and
thus
lead to the destruction of the abrasive grain.
When microcrystalline sintered abrasive grains are used, in many
applications it is possible to see that given comparable hardness and density,
the finer the configuration of the structure, the better the abrasive grain
will
behave during the grinding process. Particularly fine structures can be
obtained using the sol-gel method, in which, for example, finely dispersed
2
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
aluminum oxide monohydrate of the boehmite type is used, this being
processed to a gel once it has been colloidally dissolved and then further
processed by drying, calcination, and sintering to form a compact and dense
a-A1203 sintered body. This is subsequently processed to form an abrasive
grain. The advantage of the sol-gel method for producing microcrystalline
corundum lies in the fact that very finely divided and reactive starting
substances can be used and the resulting green body can be consolidated at
relatively low sintering temperatures, which facilitates the formation of a
fine
structure.
EP-B-0 152 768 describes microcrystalline corundums that are produced by
the sol gel technique by the addition of specific crystallization nuclei at
sintering temperatures of 1400°C and whose primary crystallites have
diameters that are mostly or all smaller than 1 Nm.
The growth rate of the crystals during the sintering process can be greatly
limited because of the low sintering temperatures and by the addition of
crystallization nuclei. Even finer structures at higher densities and hardness
are described in EP-B-0 408 771. According to EP-B-0 408 771, corundum
abrasive grains with a mean crystallite size of < 0.2 Nm are obtained by
3
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
using the sol-gel technique with the addition of very finely divided
crystallization nuclei and whilst keeping to a special temperature and
sintering program, in which the temperature range varies between the 900
and 1100°C in less than 90 seconds; the material can be brought to its
maximal temperature, which should not exceed 1300°C, for a brief period
of
time and subsequently densely sintered beneath this maximal temperature
in the range between 1000 and 1300 °C. The temperature program is
selected so as to permit a high level of consolidation without the resulting
sintered bodies or their precursors being exposed for too long to
temperatures that would favor crystal growth.
If one wishes to obtain the finest possible crystalline structure, it is
recommended that sintering additives be used in addition to using
crystallization nuclei, since these additives inhibit crystal growth or
accelerate the sintering process and thus indirectly inhibit the formation of
larger crystals. The effect of individual additives on the sintering process
and on crystal growth when sintering A1203are described in the Journal of the
American Ceramic Society, Volume 39, No. 10, 1956. The following are
examples of the numerous patents that describe the use of sintering
additives or combinations of sintering additives with crystallization nuclei
for
4
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
the production of abrasive grains using the sol-gel method. EP-B-0 024 099
describes the addition of spinets or precursors there that are converted to
spinets during production process. EP-B-0 200 487 describes the use of a-
FE203 crystallization nuclei in combination with at least one of modifying
component from the group comprising the oxides of magnesium, zinc,
cobalt, nickel, zirconium, hafnium, chromium and/or titanium. EP-B-0 373
765 describes yttrium and neodymium compounds-also in combination with
a-Fe203 nuclei-in addition to the above-quoted oxides as additional
modifying components. When used for specific applications, the abrasive
grains produced using the above method have advantages as compared to
the prior art.
The large number of different AI203 sintered abrasive grains can be
explained by the fact that grinding is itself an extremely varied process
during which both the material that is being processed as well as working
conditions such as contact pressure, cooling, and the like can be greatly
varied. This means that the most varied materials (various types of steel,
alloys, and metals, plastics, wood, stone, ceramic and the like) can be
processed under the most varied conditions, depending on the objective that
is to be achieved (surface quality, removal of material, and the like). The
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
demands that are placed on the abrasive grains that is to be used are
similarly varied, so that the usefulness and efficacy of an abrasive grain for
a
specific grinding process cannot be characterized by variables such as
hardness, density, and crystallite structure alone. Other criteria, such as
chemical inertness, thermal conductivity, resistance to oxidation and
temperature, toughness, etc., will play a major role, depending on the
particular application.
Other variables that affect the grinding process are bonding and the
specification of the grinding agent, which can also be further varied by the
addition of additives such as secondary grinding agents, pore forming
materials, and the like.
Thus, even in the case of abrasive grains produced by a the sol-gel method
attempts were made in the past to increase performance by varying the
degrees of fineness of the crystallite structure, and to obtain particularly
favorable characteristic for a specific application by way of doping. EP-A-0
228 856 describes the addition of yttrium that is added to the sol-gel process
of the a-aluminum monohydrate dispersion (for example, in the form of an
6
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
yttrium salt) with a slightly volatile anion (nitrate, acetate, or the like),
which
reacts with the AI203 to form yttrium-aluminum garnet. This material has
particular advantages when machining stainless steel, titanium, nickel alloys,
aluminum and other alloys that are difficult to machine as well as when
machining simple structural steel. Obviously, the incorporation of garnet
crystals in the abrasive grains imparts a special wear resistance for these
applications, and this is reflected in a high removal performance . In
addition to Y203 or its precursors, the addition of crystallization nuclei
and/or
other sintering additives is also described. EP-B-0 293 164 describes the
addition of rare earths from the groups that comprises praseodymium,
samarium, ytterbium, neodymium, lanthanum, gadolinium, cerium,
dysprosium, erbium and/or combinations of several from this group.
Together with the AI203, the rare earths form hexagonal aluminates that
obviously bring about an additional increase in the wear resistance of the
abrasive grains when incorporated into the AI203 matrix. EP-B-0 368 837
describes the abrasive grains whose toughness is increased by the formation
of whisker-like crystals that are obtained by adding cerium compounds.
Here, too, increased toughness is achieved by strengthening the structure.
By using the sol-gel method, one also obtains the composites that are
described in DE-A-196 07 709, which differ from the compounds described
7
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
above in that in addition to the A1203 matrix there are at least two
additional
discontinuous structural components that differ from each other with respect
to average particle size by a factor of at least 10. EP-B-0 4 91 184 describes
composites based on AI203 that has inclusions of isometric hard materials
that are greater than the primary crystals of which the matrix is build up by
a factor of at least 10.
All of the above methods and materials are based on the sol-gel technology
with which it is possible, given the simultaneous use of sintering additives,
to
obtain a very fine, preferably sub-micron, crystal structure. In addition,
there is the fact that the abrasive grains are frequently tailored and
optimized for specific areas of application by additional doping.
Generally speaking, the grinding agents or abrasive grains can simply be
divided into two major groups. In addition to SiC, corundum belongs to the
so-called conventional grinding agents that have been known for a
considerable time and which can be produced very economically and in large
quantities. More recently, ever more frequent use has been made of the so-
called super abrasives such as diamond and cubical boron nitride, the
production costs for which are between 1000 and 10,000 times the
8
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
production costs for conventional abrasive grains but which, because of their
enhanced performance and the associated reduction in machine down time
and lower use of the grinding agent itself, or because of the increase in the
unit rate per unit time and grinding body offer an extremely favorable cost-
performance ratio for many grinding operations.
The use of super abrasives requires special machinery which, in its turn ,
requires appropriate investment, which means that the range of applications
for high-performance grinding agents is restricted even further.
For this reason, one of the main objectives foreseen for the development of
new abrasive grains is to obtain abrasive grains that can be used on
conventional machinery but which lie between conventional grinding agents
and super abrasives as far as level of performance is concerned. This has
been achieved, in part, with the above-quoted sol-gel corundums, which can
be used for many grinding operations at a very favorable price/performance
ratio. However, the sol-gel corundums are to be classified more as
conventional types of the abrasive grains not only in view of their production
costs but also in view of their performance, and for this reason are better
9
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
suited to replace conventional corundurns for grinding operations that do not
justify the use of super abrasives.
For this reason, it is the objective of the present invention to provide
abrasive grains with even better performance potential as compared to the
above described prior art, and a method for manufacturing these.
According to the present invention, this objective has been achieved with the
features as set out in Patent Claim 11 or Patent Claim 1. The secondary
claims refer to advantageous configurations of the present invention. Claim
20 applies to the use of the abrasive grains according to the present
invention.
The expression nanocomposite, which has been in use for some ten years in
the domain of ceramics, is used to describe systems that comprise at least
two different solid phases, of which at least one phase has particle sizes in
the manometer range.
A1203/SiC composites in which SiC particles have been incorporated in an
AI203 matrix for reinforcement, are described in EP-B-0 311 289 and are
intended, for example, as structural ceramic for use in engines and turbines.
The diameter of the SiC particles, which account for between 2 and 10 mol-
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
% , should be less than 0.5 Nm, whereas the A1203 particles should not be
larger than 5 pm. These materials, in which the SiC particles are dispersed
among the A1203 particles, are distinguished by great toughness and
extremely high transverse rupture strength, and because of their good high-
temperature properties can be used as structural ceramic in engines.
Similar AI203/SiC nanocomposites, which differ from the known whisker,
fibre, or platelet reinforced composite materials by their good high-
temperature properties and their resistance to oxidation, are described by
Niihara in the Journal of the Ceramic Society ofJapan, 99 (10) 974-982
(1991). The effect of finely divided SiC particles on the grain growth and
sintering behavior of the AI203 matrix is described by Stearns, Zhao, and
Harmer in the Journal of the European Ceramic Society, 10 (1992) 473-477.
The mechanical properties of AIz03/SiC nanocomposites are described by
Zhao, Stearns, Harmer, Chan, Miller, and Cook in the Journal of the
American Ceramic Society, 76, (2) 503-510, 1993. Nanocomposites
produced by the sol-gel method are described by Xu, Nakahira, and Niihara
in the Journal of the Ceramic Society of Japan, 1994, 102, 312-315.
11
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
Whereas the above cited references relate mostly to composites with SiC
fractions of > 2 mol-%, the mechanical properties of hot-pressed A1203/SiC
composites with smaller fractions of SiC are described in an article by
Wilhelm and Wruss in cfi/Ber. DGK 75, 40-44 (1988). AI203/SiC
nanocornposites are described in numerous other publications in addition to
the above cited references, and these have been combined in a survey by
Stirnizke in the Journal of the European Ceramic Society, 17 (1997), 1061-
1082. This article expresses the idea that, as abrasive grains, AI203/SiC
nanocomposites could fill the gap between conventional abrasive grains and
super abrasives. In contrast to this conjecture, however, almost all the
publications referred to in the article, and the material properties quoted
from this, relate expressly to use as structural ceramic. Thus, the
microstructures, thermodynamic stability, density, hardness, breaking
strength, fracture tougness, and creep rate are discussed. Certainly, all of
these variables play an important role in the grinding process without,
however, permitting a valid statement with respect to the usability of a
material as abrasive grain. For example, hardness is certainly a basic
requirement for using a material as an abrasive grain. However, as is shown
by the example of B4C, which is frequently referred to in professional
circles,
and which despite its great hardness, has never been used to any
12
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
noteworthy extent as a grinding agent because of it deficient chemical and
thermal resistance and its great brittleness, the sum of a number of
properties must be considered in order to identify suitability as a grinding
agent. Other hard materials that have hardness values that range between
conventional grinding agents and super abrasives have not found acceptance
as abrasive grains because they lack additional properties such as
toughness, thermal and chemical stability, or other prerequisites that are
important for the grinding process. The nanocomposite materials that are
described in the literature, and which have properties that are essential for
the grinding process, have not up to now been used successfully as abrasive
grains. They behave more like the cutting ceramics based on AI203 that
have been used with great success for milling or turning operations but
which, once processed to form grit, display only an unsatisfactory removal
performance which is at the level of conventional smelted corundums or
even below this during the grinding process.
In practice, it is been found extremely difficult to characterize the
usability or
the grinding behavior that can be expected from abrasive grains simply on
the basis of specific material properties which are known to have a favorable
effect on the grinding process. Up to now , theories covering the
13
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
mechanisms that actually take place during the removal of material by a
grinding tool could be developed only subsequently to the process itself, on
the basis of the work piece that is machined and on the basis of changes
that take place in the grinding tool. Naturally, in addition to all the
material
properties of the abrasive grains, the composition of the grinding tool
(bonding, porosity, additives, and the like) and the work piece itself affect
grinding behavior, so that it is frequently very difficult to look back and
correlate specific grinding results that have been achieved with specific
material properties of the abrasive grains. One can arrive at a firm
conclusion only by way of application tests of abrasive grints or even by way
of practical and field testing, which entail considerable expenditures of time
and money.
For this reason it is worthwhile attempting to find an independent measuring
method and measurement factors that make a direct statement with regard
to the usefulness of a material as an abrasive grain. In practice, in recent
years increasing use has been made of the so-called individual grit test
(Figure 1: Individual grit scratch test), in which an individual abrasive
grain
is examined under conditions that as far as possible are realistic and
modeled on the grinding process. The test apparatus is a converted surface
14
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
grinding machine in which a scratch disk is mounted on the grinding spindle
in place of a grinding disk. The scratch disc that, for practical reasons, is
manufactured from a relatively light material that is easy to machine (e.g.,
aluminum) has on its periphery a holder into which a carrier, which has a
grain of abrasive grains soldered onto it, is inserted. During the scratch
process itself, the table with the work piece clamped to it is moved in the x-
direction against the direction of rotation beneath the rotating scratch disk.
Because of a predetermined setting in the y-direction the abrasive grit,
which extends beyond the periphery of the disc, leaves a scratch track in the
work piece during each rotation. As the length of the scratch or scratch
time increases, the depth of the scratch and the cross section of the scratch
decrease because of wear on the grain, and this continues until such time as
the tip of the grain is worn down by the amount of the setting in the y-
direction and it no longer leaves a track. These scratch tracks can be
sampled with a surface measuring apparatus and then evaluated. The
underlying principle for these measurements in shown in Figure 1 and Figure
2, and is described in detail below using the relevant reference numbers.
Figure 1 shows the principles underlying the construction of the test
apparatus, with the scratch disc (1) and the scratch grain (2), with the axes
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
(3 ,4, 5) in the x, y, z directions, and with the work piece (6), the machine
table (7) and the grinding spindle head piece (8). In order to make these
measurements, it is necessary to define standard conditions for the cut rate
v~, the work piece speed vW, the setting ae, all of which should as far as
possible be matched to the grinding operation for which the abrasive grain
will subsequently be used. In addition, the work piece material and the use
of cooling lubricants (9) must be defined.
The evaluation principle can be seen from the curves for different types of
abrasive grains (Figure 2) that are shown here as examples, in which the
change in the scratch cross section AR~/ARO is plotted against the length IR
of
the scratch. ARO is the scratch cross section during the first test, and AR"
is
the cross section of a scratch that is n mm long.
The performance factor LF25 for the individual grains is located at the
intersection point of the characteristic curve for the individual grains types
with the ordinates after a scratch length of 25 mm, and corresponds to the
change of the scratch cross section ARO/AR25. The performance factor is
expressed as a percentage relative to the theoretical case that there is no
wear on the grain and ARZS = ARO. The evaluation after a scratch that is 25
16
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
mm long is selected because, in the case of the typical curve , the decisive
first, steep area of the curve, when the grains that is under the greatest
load, is closed. This area, which is a relatively close approximation to the
actual grinding process relative to the setting ae,makes it possible to arrive
at the very good assessment of the capability of an abrasive grain. As they
continue, the curves flatten out since the grains are not as greatly stressed
because of the reduced setting, and they also wear at a slower rate. In
order to arrive at a representative result for an abrasive grains, at least
twenty grains of one type should be measured and the wear curve should be
formed from the average value of the individual measurement points.
In agreement with the results achieved in practice, the single grain scratch
test permits an assessment of the suitability of abrasive grains in which all
the variables in the grinding process, such as hardness, toughness, density,
strength, creep rate, thermal and chemical resistance, crystallite structure,
and the like all contribute indirectly to the overall sum, without specific
properties or combinations of properties having to be known or identified
explicitly, and accordingly taken into account. Nevertheless, specific
minimum prerequisites must be satisfied for all the properties in order that a
material can be considered as an abrasive grain at all. Thus, for example, a
17
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
material which is of a hardness that is clearly below the usual hardness for
grinding agents will never be suitable for the grinding process, even if all
its
other properties are outstanding.
Most surprisingly, using the method described above, it has been possible to
find performance factors that are clearly superior to the performance factors
found up to now for AIZ03/SiC nanocomposites with SiC fractions below 5
mol-%, which have been produced directly by the sol-gel method with the
addition of crystallization nuclei. The performance factors for the
nanocomposites according to the present invention lie above the values for
the known pure or doped sol-gel corundums and lie in the desired range
between conventional abrasive grains and the super abrasives.
In contrast to known AI203/SiC nanocomposites that are manufactured by
powder technology, by mixing the starting substances, compressing (for
example, by hot pressing, non-pressure sintering, or hot isostatic pressing)
and sintering, the abrasive grains according to the present invention are
produced by wet-chemical methods by a direct sol-gel method using
crystallization nuclei. Xu, Nakahira, and Niihara in an article that appeared
in the Journal of the Ceramic Society of Japan, 1994, 102, 312-315 describe
18
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
the use of the sol-gel technique for manufacturing AI203/SiC
nanocomposites. However, they only use this technique in order to arrive at
the most homogenous possible mixture of the nano-powder by way of a
preceding colloidal solution of the particles. This sol is subsequently
processed to form an homogenous mixture of ultra-fine AI203 and SiC
powders by drying and calcining, and this is then hot pressed in a nitrogen
atmosphere at a pressure of 30 MPa, at a temperature of 1600°C, in the
same way as when using conventional powder technology.
Specific advantages of the sol-gel method that are important for
manufacturing abrasive grains are lost by isolating the powder as an
intermediate product, with subsequent conventional powder technique
processing. The grinding properties of the composite produced by the
method described heretofore corresponds to those of the previously
discussed nanocomposite. In addition, there are economic aspects, since
cost effective large-scale mass production of abrasive grains is not possible
using a hot press method.
In contrast to this, in the case of the direct sol-gel method according to the
present invention, which is used to produce a AI203/SiC nanocomposite,
19
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
AI203 sol is first produced in the usual manner. When this is done, very
finely dispersed aluminum oxide monohydrate of the Boehmite type,
aluminum alkoxides, aluminum halogenides and/or aluminum nitrate can be
used as the solid component that contains the aluminum oxide. These are
dispersed with a dispergator, a powerful agitator, or by using ultrasound. It
is preferred that the solids content of the suspension lie between 5 and
60%-wt. Then, nano-scale SiC, preferably in the form of a suspension-so as
to arrive at the most homogenous possible distribution, between 0.1 and < 5
mol-%, preferably in the range between the 0.3 and 2.5 mol-% relative to
the aluminum content of the mixture, calculated as AI203 -is added to this
suspension. It is, of course, also possible to stir SiC into the suspension as
a
solid. As the examples set out in Table 3 show, particularly good results are
achieved with comparatively small quantities of SiC. Finely ground SiC
powder obtained by the Acheson process, or even nano powder
manufactured in the gas phase by thermal or laser assisted gas phase
reactions or by various plasma and methods, can be used as the SiC basis.
Additional sintering additives in the form of crystallization nuclei, crystal
growth inhibitors, and/or other modifying components can be added,
preferably before gelling, in order to enhance the subsequent sintering
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
process. All known sintering additives for AIz03, for example, spinet-forming
oxides of Co, Mg, Ni, and Zn, the oxides of Ce, Cu, B, Ba, Hf, K, t_i, Nb, Si,
Sr, Ti, Y, Zr or the rare earths or their precursors, and the oxides with a
corundum-like structure such as Fez03, Cr203, AI203, and others, that act as
crystallization nuclei. Combinations of these can be used in order to impart
specific properties to the abrasive grain.
It is preferred that an aqueous suspension of very finely ground AI20s be
added to the AI203 sot prior to the gelatinization of the SiC. The maximum
size of the «-A1203 particles, which serve as a crystallization nuclei, is
less
than 1 pm, and preferably less than 0.2 arm. The quantity of nucleation
material that is to be used depends on the particle size, and lies between 0.5
and 10%-wt, relative to the AI203 of the end product. Since the number of
nuclei is important in addition to fineness, at very high levels of fineness
very small quantities of nuclei by weight are sufficient to further the
sintering
process.
The prepared suspension is then heated to boiling point and gelled by the
addition of acid. Here, too, it is again possible to use all of the known
types
of gelatinization (aging, addition of electrolyte, temperature increase, and
21
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
reduction of the suspension, amongst others). The gel is dried (after
cooling) in a temperature range between 50°C and 120°C.
Calcining then
takes place in the temperature range between 500°C and 800°C, in
order to
vaporize residual water and the acid. After calcining, the composites are in
the form of green bodies with diameters of up to several millimeters, and
these are then sintered. The advantages of direct densification lie in the
particularly high sinter activity of the dried and calcined green bodies, in
which the starting materials are already bonded to each of chemically, so
that densification and consolidation to form a solid composite proceed far
more effectively and efficiently.
The process, and thus the quality of the product, can be further improved by
the additional use of sintering additives or crystallization nuclei. It is
preferred that the calcined gel be sintered at temperatures between
1300°C
and 1600°C, preferably under inert conditions (e.g., in an nitrogen
atmosphere) and, particularly advantageously, in a gas tight rotary tube
furnace in order that the product is heated as rapidly as possible and the
sintering time is kept as short as possible, since this has a particularly
favorable effect on the structure and thus on the efficacy of the abrasive
grains. Alternatively, any other known type of furnace can be used,
22
CA 02407736 2002-10-29
WO 01/21547 PtrT/EP00/09230
providing it permits a rapid heating rate and high temperatures. Since the
sintering takes place very quickly, it is possible for processing to take
place
in a vacuum or in an oxidizing atmosphere, since the greatest part of the
SiC nano particles are imbedded in the matrix and thus protected against
oxidation.
Reduction to the desired grains size can take place either before or after
sintering, using the usual reducing apparatuses. It is advantageous to
prepare the calcined gels in the green state, since once sintering has taken
place, considerably more energy has to be expended for reduction of the
then dense and hard composite material.
During sintering, the nano-scale SiC acts as a crystal growth inhibitor for
the AI203 matrix and, at the same time, delays densification of the green
body so that, comparatively speaking, higher sintering temperatures have to
be used than is the case of sol-gel material that is based on pure aluminum
oxide, in order to achieve sufficient densification of the material, when not
inconsiderable crystal growth can take place. Even at 1400°C, much
larger
crystallites occur more frequently. This phenomenon has already been
described in U.S. Patent 4,623,364 wherein the undesirable occurrence of
23
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
coarse crystals in an otherwise fine matrix is attributed to impurities. In
this
specification, an attempt is made to find the finest crystalline matrix
possible, with the fewest possible coarse fractions, as have been described in
the patents cited in the introduction hereto, and which correspond to the
prior art.
Most surprisingly, it has now been found that the abrasive performance of
the nanocomposite abrasive grains according to the present invention is
particularly high if a specific fraction of coarse crystals with lengths of up
to
20 Nm and with an average diameter of greater than 2 pm, preferably
greater than 5pm, is present in the matrix. In this case, the removal
performance is clearly above that of the finely structured pure sol gel AI203
grinding grains, the average crystal size of which is usually 0.2-0.3 pm and
in which all crystals are in the sub-micron range, preferably in the range
below 0.4 Nm. This is all the more surprising since it is generally known in
professional circles that the abrasive performance of sintered corundums is
increase dramatically as the structures become finer, particularly in the d5o
range beneath 0.5 pm.
24
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
As is shown in Examples 1-6 and in the comparative Examples 7-11, which
describe the effect of sintering conditions on the structure and the
performance strength of sinter corundums, the performance curve for the
AI203/SiC nanocomposite is a non-linear curve with a maximum at the
sintering temperature between 1400-1450°C. The first coarse crystalline
and
columnar crystals appear within the matrix in this temperature range, at a
holding time of 30 minutes. It is preferred that the coarse AI203 crystals be
of an elongated form with a length to width ratio between 2:1 and 10:1,
especially between 4:1 and 6:1. Typical images of the matrix with the
coarse crystalline incorporations are shown as electron and optical
microscope images in Figures 3 and 4 on Page 19. Below 1400°C, a purely
sub-micron structure occurs, in which all the particles are in the range of <
1
pm, preferably < 0.5 Nm. The efficiency of these materials also lies above
the efficiency of pure sol-gel corundums as based on the prior art, but
surprisingly lies beneath that of materials with coarse crystallite
incorporations obtained in the above-quoted temperature range. At even
higher sintering temperatures, at which there is an increased occurrence of
coarse crystals, the performance curve falls off once again.
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
However, even at sintering temperatures of 1500°C with high
proportions of
quartz crystals, abrasive performances that are at the level of the best pure
sol-gel corundums obtained in the above-quoted temperature range are
achieved. In contrast to this, in the case of the pure sol-gel corundums ,
one can see an almost linear curve of the performance potential with the
fineness of the structure, and good performance is achieved first in the sub-
micron range at an average crystal size d5o < 0.4 Nm.
Obviously, in the case of the nanocomposite, the coarse crystallites act as a
kind of structural reinforcement that has a positive effect on the grinding
behaviour of the grit and compensate not only for the reduction in
performance, anticipated because of the grains growth; in combination with
the incorporated nano-SiC particles, they help the abrasive grains to achieve
a clear increase in performance.
From the examples shown in Table 4 it can be seen that improvement of the
product by the incorporation of SiC particles is confined not only to nano-SiC
powder; outstanding abrasive performance can also be achieved with grains
that have relatively coarse SiC incorporations. However, it is quite clear to
see that the finer the SiC powders that are used, the better the abrasive
26
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
performance will be. For commercial reasons, and for reasons of availability,
initially only the powders in the examples were used to produce the abrasive
grains according to the present invention; these were obtained by fine
grinding industrial SiC that had been obtained by using the Acheson method.
One can, however, proceed in that the above-quoted trend is continued by
using even finer powders.
In the nanocomposites according to the present invention, the SiC particles
can be both intragranular, within the AI203 matrix particles, as well as inter-
granular, on the grain borders between the AIz03 particles; it should be
observed that the smaller particles are preferably incorporated so as to be
intragranular. The effect that the type of incorporation of the Sic particles
has on abrasive performance is the subject of ongoing investigations and, for
the time being, can only be regarded on a speculative basis.
Some theories have been discussed in the publications referred to above;
however, these discuss only the individual properties of composite materials
and do not discuss the effect of the totality of the properties that are
decisive for abrasive performance. At all events, Examples 14-17 clearly
illustrate the trend that abrasive performance is enhanced with decreasing
27
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
particle sizes of the incorporations. From this, it can be concluded that it
is
mainly intergranular SiC that is responsible for the improvement of abrasive
performance .
The present invention creates nanocomposite abrasive grains that are based
on AIzO3 and which have predominantly intragranular SiC nanoparticles
incorporated in them. They have a hardness (HVo,2) that is greater than 18
GPa, a density that is 95% of the theoretical density, and a performance
factor LF25 °> 75% (° = measured as the mean value of 20
individual
measurements on 100Cr6 material (HRH = 62) with a cutting speed of 30
m/s, a feed rate of 20 pm, a work piece speed of 0.5 mm/s, and using a 3-
emulsion as a cooling agent.
The present invention will be described below on the basis of the examples
that follow, without necessarily being restricted thereto.
Examples 1 - 6
Using a Megatron MT 1-90 dispergator (Kinematica), 10 kg pseudo-boehmite
(Disperal, Condea) is dispersed in SO liters of distilled water, the pH value
of
which had been adjusted to 2.4 by the addition of approximately 300 ml of
28
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
concentrated nitric acid. Approximately 300 ml of 50% nucleation slurry
that contains a-AI203 within maximum particle size of dmaX = 0.4 Nm and
which was obtained by wet grinding and subsequent centrifuging of a fine
particle a-A1203 powder (CS40M, Martinswerk) was then added to the
dispersion, also by using a dispergator. After the addition of the nucleation
slurry, approximately 2%-wt of AI203 crystallization nuclei was present in the
sol.
Suspension B (SiC suspension)
1.5 g of a 50-% aqueous polyethyleneimine suspension (Fluka) was added to
600 ml of distilled water during vigorous stirring. Next, 30 g nano-scale SiC
(UF 45, H.C. Starck) was stirred into the diluted suspension.
Suspension B was added to the boehmite sol (Suspension A) during stirring
and the pH value of the mixture was adjusted to 1.8 using nitric acid. Next,
the mixture was heated to 95°C whilst being constantly stirred, and
gelling
was initiated by drop-by-drop addition of additional nitric acid. After
cooling,
the gel was dried in a drying cabinet at 85°C. The dry gel is reduced
to a
particle size of less than 5 mm and then calcined at 500°C.
29
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
In the Examples 1-6, only the sintering temperature is varied. In Table 1,
the measured hardness values, performance factors, and crystal structures
are set out as a function of the sintering conditions.
Table 1, Examples 1-6
Example Sintering Program Hardness Crystallite LFZSoo
(HVo,z) Structure
(dso)
1 1300/N2/60/30 11.3 GPa < 0.4 ~.m 23
2 1350/N2/60/30 13.3 GPa < 0.4 ~cm 29
3 1380/N2/60/30 19.8 GPa < 0.4 wm 73
4 1400/N2/60/30 22.9 GPa l~.m 85
1450/N2/60/30 20.7 GPa 5 - 10 ~,m 83
6 1500/N2/60/30 20.1 GPa 10 - 20 ~m 70
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
Sintering program =
Sintering temperature (°C)/ furnace atmosphere/ heating rate
(°C/minute)/holding time (min).
Comparative examples 7-il (without SiC incorporations)
Using a Megatron MT 1-90 dispergator (Kinematics), 10 kg pseudo-boehmite
(Disperal, Condea) is dispersed in 50 liters of distilled water, the pH value
of
which had been adjusted to 2.4 by the addition of approximately 300 ml of
concentrated nitric acid. Approximately 300 ml of 50% nucleation slurry
that contains a-AI203 within maximum particle size of dmaX = 0.4 Nm and
which was obtained by wet grinding and subsequent centrifuging of a fine
particle «-AI203 powder (CS400M, Martinswerk) was then added to the
dispersion, also by using a dispergator. After the addition of the nucleation
slurry, approximately 2%-wt of AI203 crystallization nuclei was present in the
sol.
The pH value of the mixture was adjusted to 1.8 using nitric acid. Next, the
mixture was heated to 95°C while being constantly stirred and gelling
was
31
- CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
initiated by drop-by-drop addition of additional nitric acid. After cooling,
the
gel was dried in a drying cabinet at 85°C. The dry gel is reduced to a
particle size of less than 5 mm and then calcined at 500°C.
In the Examples 7-11, only the sintering temperature is varied. In Table 2,
the measured hardness values, performance factors, and crystal structures
are set out as a function of the sintering conditions.
Table 2: Comparative Examples 7-11
Example Sintering Program Hardness Crystallite LFzsaa
(HVo,2) Structure
(d5o)
7 1240/NZ/60/30 19.7 GPa 0.2-0.3 ~,m 75
8 1300/N2/60/30 22.4 GPa 1 ~m 63
9 1350/N2/60/30 23.1 GPa 1 - 5 ~.m 60
1400/Nz/60/30 21.6 GPa 3 - 7 ~,m 49
11 1450/N2/60/30 20.6 GPa S - 10 ~m 40
*Sintering program = Sintering temperature (°C)/ furnace atmosphere/
heating rate (°C/minute)/holding time (min).
32
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
Example 12
Production of Example 12 is effected analogously to the Examples 1-6.
However, 75 g of a nano-scale SiC UF45 was used.
Example 13
Production is effected as for Example 12. 150 g of nano-scale SiC UF45 was
used instead of 75 g. Table 3 shows the performance factors as function of
the SiC concentration.
Table 3: Examples 4, 12, 13
Example Sintering Program Hardness Crystallite LFZS~,o
(HVo.2) Structure
(dso)
4 1400/Nz/60/30 22.9 GPa 1.0 ~,m 85
12 1400/Nz/60/30 22.4 GPa 2.5 ~cm 59
14 1400/N2/60/30 23.1 GPa 5.0 ~m 37
33
~
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
Example 14
The production of Example 14 was effected analogously to Example 4. The
somewhat coarser SiC OF 25 (H.C. Starck) was used in place of the SiC
UF45. Sintering was carried out at a temperature of 1400°C in a
nitrogen
atmosphere. The heating rate was 60°C per minute and the holding time
was 30 minutes.
Example 15
The production of Example 15 was effected analogously to Example 14. The
somewhat coarser SiC OF 15 (H.C. Starck) was used in place of the SiC
UF25.
Comparitive Example 16
The production of Example 16 was effected analogously to Example 15. SiC
P1000 (Elektroschmelzwerk Kempten) was used in place of the SiC UF15.
Com~aritive Example 17
34
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
The production of Example 17 was effected analogously to Example 16. SiC
P600 (Elektroschmelzwerk Kempten) was used in place of the SiC P1000.
Table 4 shows the performance factor of the nanocomposite as a function of
the particle size of the incorporated SiC's .
Table 4: Examples 4, 15-18
Example SiC Average particleHardness LF25 (%)
size d5o (HVo,z)
4 UF45 300 nm 19.7 GPa 85
14 UF25 500 nm 22.4 GPa 82
15 UF15 600 nm 23.1 GPa 77
16 P1000 18 ~,m 21.6 GPa 73
17 P600 26 ~,m 23.3 Gpa 58
CA 02407736 2002-10-29
WO 01/21547 PCT/EP00/09230
Grinding tests
In addition to the scratch test, several selected examples were subjected to
a grinding test in grinding belts. The results of the test are set out in
Table
5.
Table 5 Grinding test (belt grinding)
Steel Types
Abasive grainTurbine Steel Titanium alloy
Removal (g) Performance Removal (g) Performance
(%) (%)
Example 4 1096 145 127 176
Example 5 994 131 109 151
Example 14 1023 135 112 155
Example 15 843 111 85 118
Example 7 781 103 68 94
Commercial 757 100 72 100
sol-
gel corundum
Fused corundum320 42 23 32
36
